Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy

Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Citation: Juliyana SJ, Prakash JU,Čep R, Rubi CS, LM5 alloy is suitable for metal castings for marine and aesthetic uses due to its admirable Salunkhe S, Sadhana AD, et al. (2024) Taguchi resistance to corrosion. In order to make intricate shapes in the LM5 alloy, this study intends optimization of Wire EDM process parameters for machining LM5 aluminium alloy. PLoS ONE to assess the impact of Wire Electric Discharge Machining process variables, like Pulse on 19(10): e0308203. https://doi.org/10.1371/journal. Time (T ), Pulse off Time (T ), Gap Voltage (GV) and Wire Feed (WF) on responses like on off pone.0308203 Material Removal Rate (MRR), Surface Roughness (SR), and Kerf Width (K ). The LM5 Editor: Siddhartha Kar, Ramaiah Institute of aluminium alloy plate was produced through stir casting process. SEM, EDAX and XRD Technology, INDIA images confirm the LM5 Al alloy’s microstructure and crystal structure. WEDM studies were Received: November 25, 2023 conducted using design of experiments approach based on L orthogonal array and ana- Accepted: July 19, 2024 lysed using Taguchi’s Signal to Noise Ratio (S/N) analysis. Pulse on Time has the greatest statistical effects on MRR (68.25%), SR (79.46%) and kerf (81.97%). In order to assess the Published: October 28, 2024 surface integrity of the WEDM machined surfaces, the SEM study on the topography was Copyright:© 2024 Juliyana et al. This is an open conducted using the optimum surface roughness process variables: T 110μs, T 50μs, on off access article distributed under the terms of the Creative Commons Attribution License, which GV 40 V, and WF 9 m/min. SEM images show the recast layer and its thickness. The aver- permits unrestricted use, distribution, and age absolute error for MRR is 1.69%, SR is 3.89% and kerf is 0.88%, based on mathemati- reproduction in any medium, provided the original cal (linear regression) models. The Taguchi’s Signal to Noise ratio analysis is the most author and source are credited. appropriate for single objective optimization of responses. Data Availability Statement: All relevant data are within the manuscript. Funding: The authors extend their appreciation to King Saud University for funding this work through Researchers Supporting Project number Introduction (RSP2023R164), King Saud University, Riyadh, Saudi Arabia. The role of funder in the revised Wrought alloys and cast alloys are the two main divisions of aluminium alloys. In the first paper is as follows. 1) Edit and revised the case, alloys undergo treatment in a solid state but in the later; they are liquefied in a furnace comments from reviewers 2) Providing the testing and then poured into moulds. Aluminium alloys may be categorized as heat-treatable or non- facility 3) The funder is involved in data collection, design a Taguchi model. 4) Preparation of heat-treatable depending on the strengthening mechanisms used. Due to their beneficial PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 1 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy manuscript 5) The funder is designed a qualities, such as their high strength-to-weight ratio, affluence of fabrication, high degree of methodology. workability, noteworthy ductility, owing thermal conductivity, strong resistance to corrosion, and appealing physical appearance at their inherent finish, aluminium alloys have become Competing interests: The authors have declared that no competing interests exist. popular as structural materials over the past few years [1]. Because of this, the marine industry currently uses 25% of the world’s aluminium manufacturing. The thin, malleable metal aluminium has great plasticity, acceptable weldability, sufficient tensile and compressive strength, and extremely significant thermal and electrical conductiv- ity. The electronics industries use it frequently. In the automotive, marine, and aerospace industries, where weight and mechanical qualities are prioritized, aluminium alloys are pri- marily employed. Aluminium alloy’s machinability is far worse than pure aluminium. The degree of strain hardening, soft particles, and precipitates has each a positive impact on an alloy’s capacity to be machined [2]. Nevertheless, aluminium alloys are regarded as tough to machine materials, especially for dry machining, not withstanding their mechanical qualities. Its high heat conductivity, low melting point and propensity to stick to the cutting edges of tool materials are difficulties. In addition to the high thermal conductivity of aluminium, which removes a significant amount of heat from the cutting edge into the work piece throughout the machining process, the material is thermally deformed. Because of the low melting point of aluminium alloys, there are issues with chip development, chip elimination, and material clinging to the cutting tool [3]. So it would appear to be quite beneficial to utilize an innovative way while cutting aluminium. Due to aluminium’s strong electrical conductivity, WEDM technique should be appropriate [4]. In order to remove material, WEDM technology uses thermoelectric energy amongst the work piece and a wire electrode. The pulse discharging takes away the material from the work piece by melting and evaporating it in a tiny space separating the work piece and the electrode. This technique is typically employed to produce intricate shapes and to manufacture materials that are challenging for regular equipment to work with [5]. EDM is superior to traditional machining in many ways. Electrically conductive materials can be cut using EDM, and this technique has been used to machine work pieces that have been heat-treated and hardened. Intricate and complex profiles can be cut more quickly, precisely, and affordably. Burrs have been avoided and thin, delicate parts have been created with ease. The mechanical properties of the material have no bearing on the WEDM cutting process; the only need is to achieve the bare minimum conductivity of the processed material. Because of its great accuracy and mini- mum surface roughness, this technology is suited for cutting very hard conductive materials, composites, ceramics, or sandwiches. Considering the significance of WEDM manufacturing, it must be less expensive than traditional machining in order to be successful. In general, the use of electrical discharges represents a trade-off between productivity and machining quality. Numerous factors might have an impact on the WEDM cutting process [6]. Each one of them has a different impact on the price of production and the work piece’s final grade finish. Thus, applying DoE should be beneficial. Schedule of test and statistical analysis of specified plan make up the DoE (Design of Experiments), very effective. The conclusion of a planned experi- ment’s evaluation is whether the variables under observation were affected by the tested ele- ments. An experiment’s output is a particular value of the observable variable, also known as the dependent variable or response, which describes the test’s quality. The ultimate quality is affected by a wide range of factors. Under the perspective of experimental design, they can be classified into specific and randomized. WEDM machine parameters are the inputs to the pro- cedure of machining. Just those factors which have a statistically noteworthy outcome on the degree of quality should be ultimately chosen after thoroughly examining all components and their common interactions. It enables the variables that matter most to be set at their ideal PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 2 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy values while also identifying the unnecessary variables, lowering their tolerance and thereby lowering manufacturing costs [7, 8]. Research on WEDM on Al6061 alloy show that EDM, a nonconventional machining method, is superior in terms of dimensional accuracy and precision; when it comes to cutting hard, conductive materials. In EDM, sparks repeatedly discharge when voltage is applied after both electrodes have been dipped in dielectric. These sparks flush away the material from the surface as melted and degraded particles that are removed by dielectric. Researchers have taken into consideration advancements and modifications to the EDM process over time. A recent invention to improve the EDM process’s competency involves agglomerating powder with both an external magnetic field and a dielectric. This procedure demonstrates the ability to machine complex and sophisticated 3D profiles with reduced tool wear rate (TWR), increased productivity, and better surface quality and precision. The MFAPM-EDM technique is used in the industrial, aerospace, automotive, defense, and surgical industries for the machining of different components. In the PMEDM process, the powder particles’ accumula- tion of charges causes several sparks to occur during machining. More charges are produced when ions from powder particles collide with dielectric molecules due to the accumulated charge in the machining area. Multiple sparks cause the surface materials to be removed more quickly and create shallow craters, which improved MRR and SR [9]. The feasibility of utilizing the Maglev EDM for machining aluminium 6062 alloy is assessed and examined. The Maglev EDM achieves tool location by means of a methodical combination of magnetic repulsive forces. Brass tools were used in the experiments, and the surrounding air served as a dielectric for the Al-6062 alloy. The data from the previously published literature was further contrasted with the experimental results. Field-emitting SEM investigation of the machined surface revealed the presence of recast layers, globules, lumps of debris, melted debris, micro-pores, micro-voids, micro-cracks, and craters [10]. WEDM’s essential indicators of performance are MRR, SR, and kerf. The MRR in WEDM processes determines the cost of machining and the rate of output. The primary purpose when establishing the machining parameters is to maximize MRR and minimize SR. The Taguchi technique, a powerful experimental design tool, takes a simple, effective, and systematic approach to determining the ideal machining parameters. Furthermore, this strategy has a low experimental cost and effectively reduces the effect of the source of variation. An economical and simple technology for modifying machined surfaces while maintaining accuracy must be developed [11]. The main objective of this study is to optimize the WEDM process parameters for machin- ing LM5 aluminium alloy using Taguchi’s Signal to Noise ratio (S/N) analysis. Other objective is to study the effect of machining parameters on the MRR, SR and kerf. Another objective is to construct Mathematical models using regression equations and to find the deviation % between the experimental and predicted values of MRR, SR and kerf. The machined surfaces were analyzed under SEM to find the recast layer. Literature survey Numerous researchers adjusted the settings of several machines to produce high-quality prod- ucts [12]. Various techniques have been used in past to measure how control parameters affect part surface characteristics, MRR and K . The unstable wire portion is the main cause of diffi- culties [13]. To ascertain the relationship between machining performance and machining parameters, an experimental analysis of the wire breakage phenomenon using a thermal model was conducted. With a reduction in T , the MRR initially rises [14]. However, the gap off becomes unstable after a relatively brief period of time, which lowers the MRR. As the MRR PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 3 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy rises, surface quality declines. K and SR have been found to be significantly influenced by the T and the D . Additionally, it has been discovered that using a medium D can improve on ww ww SR; modifying T and D can regulate final cutting. on ww It was discovered that by reducing I, T and T , surface roughness can be improved. on off Shorter and long pulses will create a comparable SR but differing surface morphologies and MRR. In comparison to long pulse duration, the clearance rate is significantly higher under short pulse duration. Furthermore, compared to a long pulse, a short pulse can produce better SR [15]. Numerous criteria and difficult experimental effort are needed for the study of the WEDM process, which takes a greater amount of money and time. Therefore, it is vital to optimize the control parameters to decrease the money and time required to assess the WEDM process. There is research on experiment design that enables a method to show the effect of control fac- tors with the least amount of experiments. Some investigations use fractional factorial designs, while others use full-factorial designs to evaluate variables with impact [16]. Coated wires are chosen so as to achieve homogeneous surface qualities. Most delicate char- acteristic that affects the creation of a layer made up of a combination of oxides is the T and on T . The development of oxides can be significantly reduced by reducing the duration between off two pulses [17]. Research of Kansal et al [18] shows in PMEDM, the electrically conductive powder is mixed in the dielectric of EDM, which reduces the insulating strength of the dielectric fluid and increases the spark gap between the tool and work piece. As a result, the process becomes more stable, thereby, improving the material removal rate (MRR) and surface finish. The pow- der particles simplify the igniting process by increasing the spark gap and reducing the insulat- ing strength of the dielectric fluid. The WEDM process’s performance for titanium alloys was the focus of investigation of Debnath and Patowar [19]. Based on Taguchi DoEs, the results show that the flushing pressure of dielectric, wire tension, and T are important process parameters that have a considerable on impact on the machined hole’s circularity, cylindricity, and diametral errors, since wire ten- sion affects both stability and wire electrode rigidity. This study does not include the MRR and SR, which are the most vital process responses. In the case of MMCs, T is the most important component for K [20]. Taguchi’s DoE on w based orthogonal arrays should be used instead of full factorial experiments. Sahoo et al. [21] used EDM technology to analyze titanium diamond machining perfor- mance measures. I and T are the chosen control variables, and during experimental runs, a p on duty factor of 50% to 75% is maintained. R , TWR and MRR are the responses taken into account for this operation. The analysis has been carried out using L OA. In order to find the ideal input parametric combination, analysis of variance is also used in conjunction with over- all evaluation criteria (OEC). The findings of the analysis indicate that T has a greater influ- on ence on TWR than does current on MRR and R . Additionally, the identification of various material phases on a machined work surface is aided by EDX, SEM and XRD. For tool positioning, Maglev EDM uses a novel bipolar linear self-servo technique that pro- duces consistent machining stability using a special magnetic repulsive force balance action. The performance characteristics of each dielectric have been investigated. The evaluation of surface morphology shows that the use of bio-dielectrics has the potential to significantly improve surface uniformity and reduce deformities [22, 23]. For the processing of nanosecond pulsed lasers, a unique and straightforward model based on geometric mathematics and heat transport was proposed. Experiments verify that the mod- els are feasible. The single-pulse laser ablation craters had errors in both diameter and depth of 2.56% - 7.14% and 6.82% - 18.91%, respectively. Between 3.47% and 12.47% is the recast layer PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 4 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Table 1. Chemical composition of LM5 aluminium alloy. Cu Mg Si Mn Fe Pb Zn Al 0.032 3.299 0.212 0.022 0.268 0.02 0.01 Balance https://doi.org/10.1371/journal.pone.0308203.t001 depth error for consecutive stacked ablation pulses. On metal surfaces ablated using pulsed laser, it forecasts the shape of the recast layer. The model serves as a guide for the preparation of surface functionality [24]. Based on the extensive literature survey, the authors of this article were persuaded to carry out a research study to recommend the most optimal machining parameters for efficiently machining stir-casted LM5 Aluminium alloy in order to achieve maximum MRR, minimum SR and minimum Kerf using Pulse on Time (T ), Pulse off Time (T ), Gap Voltage (GV) on off and Wire Feed (WF) at three levels, which have not been reported before by any researcher. This study also provides a general overview of a thorough process to determine the best machining parameter settings based on the design of experiments approach. The research paper indeed aims the mathematical models that are constructed to associate the machining response characteristics with machining control parameters, in addition to revealing the results of signal/noise (S/N) ratio analysis and ANOVA. To better understand the occurrence of recast layer generation during machining, energy dispersive spectroscopy (EDS) analysis and scanning electron microscopy (SEM) analysis were used to examine the machined surface textures. Materials and methods LM5 aluminium alloy The LM5 aluminium alloy has excellent casting properties, sturdy structure, and great durabil- ity against corrosion. The material is commonly utilized in the automotive, aerospace, and marine sectors, whenever a combination of properties is required [25, 26]. Since it is easy to grind, weld, and cast into complex shapes, Machinery require extremely strong resistance to corrosion from sea water or marine atmospheres, along with castings which need to display and maintain a high polish Gestalt, all have uses for LM5 aluminium [27, 28]. The chemical composition of LM5 Aluminium Alloy using Optical Emission Spectrometry (ASTM E 1251– 07) is shown in Table 1. Fabrication Stir casting process was used to manufacture plates of the LM5 Al alloy, measuring 120*120*10 mm. Easy use, inexpensive manufacturing, a uniform dispersion of reinforcing elements, and improved mechanical qualities are only a few benefits of stir casting. Fig 1 depicts the stircasting setup. A graphite-coated container served to melt LM5 alloy ingots in a furnace powered by elec- tricity. To 850˚C, the ambient temperature was raised steadily. The liquid state of the melt at 800˚C was degassed using hexachloroethane. The molten metal was subsequently fed into pre- heated (650˚C) cast-iron moulds after being agitated at 600 rpm for 10 minutes [29, 30]. Micro-structural analysis Fig 2A shows the Optical Micro graph of LM5 aluminium alloy Fig 2B shows the SEM image of LM5 Fig 2C and 2D shows the SEM of selected area and EDAX respectively. Primary aluminum grain interdendritic pattern is visible in the microstructure. MgAl eutectic particles that were not dissolved during solidification precipitate near the grain PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 5 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 1. Stir casting setup. https://doi.org/10.1371/journal.pone.0308203.g001 Fig 2. LM5 Aluminium alloy a) Optical Micro graphs b) SEM c) SEM of selected area d) EDAX. https://doi.org/10.1371/journal.pone.0308203.g002 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 6 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy boundaries [31]. The grain size of the primary aluminium phase is 40–50 microns. The magni- fication is 100x. Scanning electron microscopy (SEM) was utilized to collect data from a mix of backscattered electron (BSE) and secondary electron (SE) signals. The BSE signal emphasizes compositional difference, whereas the SE signal depicts sample topography, including cracks and voids. Thermomechanical processing and alloying influence the development of iron-rich intermetallic particles and can be tuned to reduce the likelihood of micro-cracking [32]. Figure depicts micro-cracks in an iron-rich intermetallic particle. These endogenous micro cracks can enlarge as the material is bent into its final shape, resulting in bigger material frac- tures. The scale bar is 100 μm. EDAX test results confirm that the elements like aluminium, iron, copper, magnesium, zinc, silicon, carbon, oxygen, lead and other elements are present in the aluminium alloy as shown in all the specimens, aluminium shows the peak value followed by magnesium as LM5 is magnesium based alloy. The results confirm the presence of Al (high-intensity peaks). Experimental analysis on machinability of LM5 Al alloy by WEDM Aluminium alloy was used to prepare the test samples. The experimentation was done using the ECOCUT WEDM, which let users to select input settings based on the sample’s material and thickness. The manufacturer’s instructions should be used to select the tool material. Brass wire of 0.25 mm in diameter was used in this instance. The dielectric medium utilized was demineralized water. The worktable, the servo regulating system, the power supply, the dielec- tric supply system, and the wire are the fundamental components of the WEDM machine. Using common grips, the material samples were secured to the machine’s worktable. WEDM removes material using a sequence of recurrent spark discharges across the tool (wire electrode) and work piece, which are submerged in a liquid dielectric and isolated by a distance known as the spark gap. Whenever a suitable voltage is given during pulse-on time, the dielectric breaks down, causing an electrical spark to form between the tool and workpiece. Thermal conduction converts electrical energy into heat energy by means of the creation of a discharge column. The tool and workpiece begin to melt as a result of high-energy plasma pro- duction. While the discharge begins, the tool, work piece, and dielectric begin to vaporize, resulting in the development of a compressed vapor bubble which increases till the pulse-on time. At the start of the pulse-off period, the discharge stops, resulting in a dramatic implosion of the plasma channel and squeezed vapor bubble, allowing the superheated and molten liquid to explode into the dielectric. The ejected materials re-solidify into small spheres, which are washed away by the dielectric. This caused the creation of a small cavity or crater on the work piece surface. With each discharge, the needed amount of material is removed from the work piece surface [33]. The WEDM process is chosen for studying the machinability of the LM5 aluminium alloy. The photograph of CNC WEDM is presented in Fig 3. Design of Experiments (DoE) The Design of Experiments (DoE) technique is used to specify what information, in what amount, and under what conditions needs to be gathered during an experiment in order to meet two primary objectives: a lower cost and greater statistical precision for the response parameters. Four distinct process parameters at three levels were chosen for the current inves- tigation: wire feed (WF), gap voltage (GV), pulse on time (T ), and pulse off time (T ). Mate- on off rial removal rate (MRR), surface roughness (SR), and cutting width (kerf) are the responses. For this research, the L orthogonal array is chosen based on the parameters that were chosen. For every experimental condition, three repetitions of the experiment have been conducted. Table 2 shows the machining variables together with their respective levels. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 7 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 3. Photograph of ECOCUT–CNC wire EDM. https://doi.org/10.1371/journal.pone.0308203.g003 Machining performance variables (Responses) The basic objective is to establish the machining parameters to achieve the maximum MRR, the smallest possible K , and the lowest possible SR. Fig 4 shows the wire Electrical Discharge machined samples. Material Removal Rate (MRR). The material removal rate (MRR) is calculated as the vol- ume of material eliminated from the specimen (mm ) divided by the time taken (min). Surface roughness. According to Kousik-Kumaar et al. [34] surface roughness assessment is crucial for a number of basic problems, such as friction, surface deformation, transfer of heat, current flow, stiffness of joints, and spatial precision. Three distinct sites on the surface that was machined were used to get the data, and the average of the three readings was used to Table 2. Process parameters and levels. Level Pulse on Time (μs) T Pulse off Time (μs) T Gap Voltage (V) GV Wire Feed (m/min) WF on off 1 110 30 20 3 2 115 40 30 6 3 120 50 40 9 https://doi.org/10.1371/journal.pone.0308203.t002 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 8 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 4. Machined specimens. https://doi.org/10.1371/journal.pone.0308203.g004 determine the SR. The orientation of an SR computation was orthogonal to the surface being machined. The Surfcorder SE 3500 surface roughness tester is displayed in Fig 5. Ra (roughness average) is a commonly used metric for quantifying a material’s surface roughness. It is the arithmetic average of the absolute values of profile height deviations from the centerline measured within a given evaluation length. The work piece was positioned verti- cally, with the WEDMed specimens’ axes oriented horizontally. The roughness tester was Fig 5. Surface roughness tester. https://doi.org/10.1371/journal.pone.0308203.g005 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 9 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 6. Output of surface roughness measurement. https://doi.org/10.1371/journal.pone.0308203.g006 placed in front of each machined surface, allowing the stylus arm with the probe tip to be placed on the WEDMed surface. Fig 6 depicts the output shown during the measurement of surface roughness. Cutting width (kerf). The width of material which is machined by a cutting procedure is referred to as Cutting Width (Kerf). When it comes to CNC Wire EDM form cutting with tra- ditional cutting techniques, kerf refers to the amount of material removed as the process cuts through the plate. It is measured by the help of Vision Measuring Machine. The most crucial sign of WEDM technology success is the width of the kerf in conjunction with the rate of material removal. Results and discussion To ascertain the impact of machining input parameters on performance measures, experi- ments on wire EDM were conducted based on the Taguchi DOE approach and analyzed using S/N ratio analysis [35]. The results and pertinent factors are presented. This section converses the experimental findings related to the Wire EDM of the LM5 Aluminium alloy. The analysis and discussion focus on MRR, SR and k . The ideal machining parameters were established using Taguchi’s S/N analysis and confirmation experiments to validate the results. Experimental results Wire EDM experiments were conducted using Taguchi’s DoE and analyzed by S/N analysis. Table 3 displays the Wire EDM experimental results of the effect of process variables such as T , T , GV, WFon the responses like MRR, SR, and k . on off w Analysis and discussion of results of MRR S/N ratio is a metric for evaluating quality attributes. Smaller is better, nominal the best, and larger is better are three elements of performance qualities that Taguchi identified. The bigger amount of MRR is found to be more advantageous for surplus product quality, hence the "larger is better" class is chosen for S/N calculation. The ideal machining variables are esti- mated at the level where each variable has the highest S/N value. This section discusses the PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 10 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Table 3. WEDM experimental results of LM5 aluminium alloy. Ex A B C D MRR (mm / S/N of Surface Roughness S/N of Kerf S/N No min) MRR (μm) SR (mm) Kerf Pulse on Time Pulse off Time Gap Voltage Wire Feed (m/ (μs) (μs) (V) min) 1 110 30 20 3 3.61 11.16 3.10 -9.8 0.286 10.87 2 110 40 30 6 2.98 9.50 3.27 -10.3 0.286 10.87 3 110 50 40 9 2.26 7.07 2.76 -8.8 0.286 10.87 4 115 30 30 9 4.50 13.07 3.63 -11.2 0.304 10.34 5 115 40 40 3 3.40 10.64 3.45 -10.8 0.295 10.60 6 115 50 20 6 5.52 14.84 3.66 -11.3 0.313 10.09 7 120 30 40 6 4.56 13.18 3.83 -11.7 0.304 10.34 8 120 40 20 9 7.09 17.01 3.63 -11.2 0.313 10.09 9 120 50 30 3 5.78 15.24 3.96 -12.0 0.313 10.09 https://doi.org/10.1371/journal.pone.0308203.t003 implications of the Wire EDM process settings on MRR. The S/N ratio of response characteristics for each variable at different phases is quantified using experimental results. The parametric influ- ences on response characteristics were examined using the main effects plot (response graphs). Analysis of variance (ANOVA) was applied to S/N data in order to categorize pertinent variables and assess the effects of those variables on response characteristics. The optimum process variables are obtained by analyzing the response graphs and ANOVA table. As shown in Fig 7, the MRR rises with increasing T and decreases with rising T and GV. This occurs as a result of the dis- on off charge energy increases brought on by the increase in T , which leads to a greater MRR. As the on T decreases, more discharges occur in a given period of time, increasing the MRR. Lower MRR off is caused by the normal discharge gap widening as the gap voltage increases [36]. Selection of optimal levels for MRR The delta values and ranks indicate that T has the greatest effect on MRR, followed by GV. on Fig 7 and Table 4 demonstrates that the third level of T , first level of T , first level of GV on off Fig 7. Response graphs for MRR. https://doi.org/10.1371/journal.pone.0308203.g007 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 11 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Table 4. Response table for MRR. Level Pulse on Time Pulse off Time Gap Voltage Wire Feed 1 9.24 12.47 14.34 12.34 2 12.85 12.38 12.60 12.51 3 15.14 12.39 10.30 12.39 Delta 5.90 0.09 4.04 0.16 Rank 1 4 2 3 https://doi.org/10.1371/journal.pone.0308203.t004 and second level of WF produce the maximum MRR. R value is 99.93%, which is most desir- able. The p-value for T and GV is below 0.05, demonstrating the significance of the factors, on but the p-value for T and WF is greater than 0.05, demonstrating the meager impact of these off parameters on MRR. According to the F-test, the effect is determined to be considerable if the computed value of the F-ratio is higher than the tabulated F-value. The table’s F-value at the 5% level of signifi- cance is F = 6.944. So, T and GV are important process parameters for obtaining (0.05, 2,4) on larger MRR as seen in ANOVA Table 5. T and WF are two variables that have very minimal off contribution on the MRR, so they are pooled up with error. Analysis and discussion of results of SR The "smaller is better" class is used for S/N analysis since it is found that a lesser amount of SR is better for excellent product quality. As shown in Fig 8, the SR increases with the increase in T and decreases with the increase in GV, T and WF. The cause is the discharge energy on off changes with T and that a higher T produces a larger crater, which raises the SR on the on on work piece. As T rises, the number of discharges decreases, leading to greater surface preci- off sion from steady machining. As the GV increases, the average discharge gap widens, improv- ing SR. T has no discernible impact [37]. Following wire EDM, the surface exhibits an off irregular mix of overlapping craters, micro-globules, and melted debris. During the WEMD process, the generated heat ranges between 8000 and 12,000˚C which produce local melting and evaporation of the work piece material. The heat produces a high level pressure, but it is insufficient for removing all of the molten material. The balance of the molten material re- solidifies on the surface of the machined sample, resulting in an undulating topography. A larger amount of molten material re-solidifies on the machined surface, resulting in a thicker recast layer and a higher Ra. Selection of optimal levels for SR For the researcher to select the ideal parameters, the experiment results are assessed. Since SR is an output characteristic where "lower is better" applies, Fig 8 and Table 6 demonstrates that the lowest SR values are found in the T first level, third level of T , third level of WF, and on off third level of GV. The process factors significance was examined using an ANOVA. R for SR Table 5. ANOVA for MRR. Source of Variation DF SS MS F P C % Pulse on Time 2 53.07 26.53 1836.77 0 68.25 Gap Voltage 2 24.63 12.32 852.57 0 31.68 Pooled Error 4 0.06 0.01 0.07 Total 8 77.76 100 https://doi.org/10.1371/journal.pone.0308203.t005 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 12 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 8. Response graphs for SR. https://doi.org/10.1371/journal.pone.0308203.g008 is 99.09%, which is a desirable value. The p-value for T is less than 0.05, indicating that the on factor is significant, whereas the p-values for T and WF are greater than 0.05, indicating that off there is no noteworthy effect on SR. The F-test shows Table’s F-value at the 5% level of signifi- cance is F = 19. So, T is the most important parameter, as shown in Table 7. Even (0.05, 2, 2) on though GV and WF are included, there is no visible impact on SR. The T is pooled up with off error. Analysis and discussion of results of kerf Lesser cutting width (Kerf) is found to be more advantageous for better product quality, the "smaller is better" class is selected for S/N calculation. Fig 9 illustrates how the T , T , and on off WF drop as the K increases. It gets smaller as the GV gets bigger. The cause is that the dis- charge energy changes with T , and higher discharge energies produce considerably larger on craters, which raise the K on the work piece. As the GV increases, the average discharge gap widens, resulting in reduced K . Lower K is also provided by the lower WF [38]. w w Selection of optimal levels for kerf For the purpose of selecting the ideal parameters, the experiment results are assessed. In the Wire EDM phase, Fig 9 and Table 8 demonstrates that the first level of T , second level of on T , third level of GV, and first level of WF have the smallest K . R coefficient is 98.44%, off w which is an excellent value. The p-value for T , WF, and GV is greater than 0.05, indicating off that there was no significant effect on K , whereas the p-value for Pulse on Time is less than Table 6. Response table for SR. Level Pulse on Time Pulse off Time Gap Voltage Wire Feed 1 -9.65 -10.89 -10.76 -10.85 2 -11.07 -10.74 -11.14 -11.07 3 -11.60 -10.68 -10.41 -10.40 Delta 1.95 0.21 0.74 0.67 Rank 1 4 2 3 https://doi.org/10.1371/journal.pone.0308203.t006 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 13 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Table 7. ANOVA for SR. Source of Variation DF SS MS F P C % Pulse on Time 2 6.11 3.06 86.87 0.01 79.46 Gap Voltage 2 0.81 0.41 11.51 0.08 10.53 Wire Feed 2 0.70 0.35 9.95 0.09 9.10 Pooled Error 2 0.07 0.04 0.91 Total 8 7.69 100.00 https://doi.org/10.1371/journal.pone.0308203.t007 Fig 9. Response graphs for kerf. https://doi.org/10.1371/journal.pone.0308203.g009 Table 8. Response table for kerf. Level Pulse on Time Pulse off Time Gap Voltage Wire Feed 1 10.87 10.52 10.35 10.52 2 10.35 10.52 10.43 10.43 3 10.17 10.35 10.61 10.43 Delta 0.7 0.17 0.26 0.09 Rank 1 3 2 4 https://doi.org/10.1371/journal.pone.0308203.t008 0.05, demonstrating the significance of the factor. The Fisher’s F-test shows at a 95% confi- dence level, the F-table value is F = 19.37. Therefore, the most important parameter, as (0.05, 2, 8) shown in ANOVA Table 9, is T , even if GV and T are additionally accounted for in the on off contribution but have no noticeable impact on K . The WF is a pooled up with the error. Confirmation experiments The confirmation experiments, according to Taguchi, is an essential stage in validating the experimental results. Based on the ideal confluence of variables impacting MRR, SR and Kerf, the confirmation experiments were effectively carried out. The tests were run three times to obtain an average value, and the correlation between the actual value and the predicted value PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 14 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Table 9. ANOVA for kerf. Sources of variation DOF SS MS F P C % Pulse on Time 2 0.80 0.40 52.6 0.02 81.97 Pulse off Time 2 0.06 0.03 3.83 0.21 5.96 Gap Voltage 2 0.10 0.05 6.74 0.13 10.50 Pooled Error 2 0.02 0.01 1.56 Total 8 0.97 100 https://doi.org/10.1371/journal.pone.0308203.t009 was then performed. Table 10 is a summary of the outcomes of the confirmation tests. The best parameters are used to predict the MRR, SR, and K , in the confirmation trials. The experiments’ results are assessed to determine the most important parameters. The best set- tings for increasing MRR are those at level A , B , C , and D (T 120 μs, T 30 μs, GV 20 V, 3 1 1 2 on off and WF 6 m/min). While the MRR predicted value is 6.56 mm /min, the MRR experimental result is 6.73mm /min. The predicted and experimental MRR values exhibit remarkable agree- ment, and the error is 2.53 percent. In order to achieve the lowest SR, the optimum process variables are T 110 μs, T 50 μs, GV 40 V, and WF 9 m/min. The experimental SR is on off 2.76μm, whereas the predicted SR is 2.76 μm. The predicted and experimental SR values exhibit a high degree of agreement. The variables at levels A , B , C , and D are T 110 μs, 1 2 3 1 on T 40 μs, GV 40 V, and WF 3 m/min are the optimum machining variables in order to off achieve the lowest Kerf. The experimental K is 0.285 mm and the predicted K is 0.270 mm. w w The predicted and experimental K values are in excellent agreement, and the error is only 2.81%. SEM analysis of WEDMed surfaces In order to assess the surface integrity, the SEM study on the topography of the WEDM machined LM5 aluminium alloy surfaces was conducted using the optimum surface roughness process variables: T 110 μs, T 50 μs, GV 40 V, and WF 9 m/min. on off The SEM micrograph of WEDM machined surface at machining parameters A B C D in 1 3 3 3 the orthogonal array of experiment is shown in Fig 10A, it is clear that the size of the crater depend on the discharge heat energy or in other words, on the gap voltage and pulse on time values. Higher Gap voltage causes an increase in discharge heat energy at the point where the discharge takes place. At this point, a pool of molten metal is formed and is overheated. The overheated molten metal evaporates forming gas bubbles that explode when the discharge ceases, taking molten material away. The result is the formation of crater. Successive dis- charges that have a random nature will result in the formation of globules of debris, shallow craters, pockmarks and cracks [39, 40]. The surface morphology of WEDMed LM5 alloy was studied by field emission scanning electron microscopy (FESEM). To determine the elemental composition, an EDX analysis was performed. Fig 10B shows the EDX image of machined LM5 Al alloy and the atomic weight. Table 10. Results of confirmation experiments. Response Optimum levels Experimental value Average Experimental Value Predicted Value Error % Trial 1 Trial 2 Trial 3 MRR (mm /min) A B C D 6.63 6.87 6.69 6.73 6.56 2.53 3 1 1 2 Surface Roughness (μm) A B C D 2.76 2.77 2.75 2.76 2.76 0 1 3 3 3 Kerf Width (mm) A B C D 0.278 0.290 0.287 0.285 0.277 2.81 1 2 3 1 https://doi.org/10.1371/journal.pone.0308203.t010 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 15 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 10. WEDM–Machined Surface a) SEM image b) EDAX c) EDX mapping d) Al e)Mg f) O g) C. https://doi.org/10.1371/journal.pone.0308203.g010 The formation of carbon is represented by distinct peaks following aluminium. The XRD anal- ysis revealed the presence of all the alloying elements, which was further confirmed by the EDX analysis. The EDAX of machined surface (recast layer) shows oxygen, carbon, copper and zinc. These are formed due to the dielectric fluid and tool material (brass wire). As shown in Fig 10C–10G, EDX mapping shows the quantity of each element inside the reaction product. The fundamental concept of the WEDM technique is the generation of electric sparks between the work piece and the wire electrode. These electrical discharges release a large quan- tity of heat at temperatures ranging from 8000 to 12,000˚C, resulting in melting and evapora- tion of work piece material at the nearby surface layers. The heat also melts the dielectric medium (de-ionized water) and creates high-pressure waves that wash away the melted and/or evaporated metal from the work piece. Throughout the WEDM process, dielectric fluid is con- tinuously supplied to transport the eroded metal apart. As a result of water’s strong thermal conductivity, the top surface cools and un-expelled material re-solidifies at a rapid pace. This re-solidified layer, known as a recast layer, is often highly fine-grained, hard, brittle, and struc- turally distinct from its parent material. The creation of these layers is determined by the pro- cess parameters as well as the work piece’s chemical composition and heat conductivity. The recast layer as shown in Fig 11 is generated at a slower cooling rate from the outermost layer, allowing the melted material to re-solidify fast and without grain boundaries. The heat- affected zone appears somewhat distinct in colour since it does not melt but is heated through- out the machining process. Zhang et al. [41] investigated different types of di-electrics and determined the recast layer created by water-in-oil emulsion dielectric, which has larger sur- face roughness and thickness than the kerosene and de-ionized water dielectrics. The results have revealed that the thickness of the recast layer increases with increasing peak current and decreases by using de-ionized water as the dielectric. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 16 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 11. Recast layer. https://doi.org/10.1371/journal.pone.0308203.g011 XRD analysis of WEDMed surface It is observed that the peaks corresponding to the sample LM5 alloy have good agreement with the JCPDS card 04–0787. So, it is concluded that the crystal structure and elemental composi- tion of the LM5 alloy have not disturbed even after the WEDM machining process. The aver- age crystallite size is 27 nm and the phase is Fm-3m [2 2 5]. Further, it is observed that the lattice parameters are equal which reflects the cubic structure and the primitives are a = b = c = 4.054A. The XRD image of Wire Electro Discharge machined specimen is shown in Fig 12. Effect of process parameters on responses (MRR, SR, and K ) Effect of T on responses. Fig 13 shows the effect of T on responses. Longer T emits on on on higher discharge energy, which causes stronger explosions, deeper craters on the surface of the work piece, and higher MRR. Deep craters suggest high rates of MRR and subpar surface qual- ity. Larger values of T should be used to get better MRR. The T that produces the highest on on MRR is 120 μs [42]. As gap voltage (GV) decreases and pulse-on time (T ) increases, surface on roughness increases. The size and shape of surface craters, which are influenced by discharge energy and the re-deposition of melted material on the work surface, are the main factors that define SR in WEDM. Surface roughness increases when the servo voltage is decreased and the pulse-on-time is increased because this increases the discharge energy across the electrodes and creates a deep erosion crater on the work piece’s surface. There is a significant chance that molten material will re-deposit on the work surface at high discharge energies [43]. According to the findings, raising the T causes a greater thermal energy transfer from the on wire to the work piece, which increases cutting velocity. As the T falls, SR lowers. SR has an on impact on the wire electrical discharge machining finish cut. Experiments have shown that lowering the T and the discharge current together can reduce SR. A large number of perti- on nent investigations found that as discharge energy increased the Wire EDMed surface rough- ness because of more craters were created, which led to higher SR values on the work piece. Increasing the pulse on time, the single pulse discharge electric energy increases thickness of cutting surface discharge, however the electrical erosion products’ duration of discharge PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 17 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 12. XRD of WEDMed aluminium alloy LM5. https://doi.org/10.1371/journal.pone.0308203.g012 reduces proportionately, leading to burns on the cutting surface and the generation of adhesive substance there. These factors collectively influence the increase in surface roughness [44]. Kerf reduces when T declines. K can be used to gauge the material’s dimensional correct- on w ness. Experiments show that lowering the discharge current and pulse duration can lower Kerf. Furthermore, since K is based on the size of the spark crater, related research discovered that T is the main variable affecting K . The discharge energy must be kept at a low level by on w employing short pulse duration in order to produce flat craters [45]. Effect of T on responses. Fig 14 shows the effect of T on responses. Results demon- off off strate that MRR declines as T rises. Due to longer non-cutting times, greater T causes a off off drop in MRR. A wider gap is produced by a longer T , but it also offers a longer flushing time off to remove the debris from the gap. Usually an extended T was used to stop wire rupturing or off to stop the abnormal process. It may be concluded that higher T causes lower SR since the off non-cutting time increases [46]. Increasing the T value extends the duration between 2 suc- off cessive sparks, resulting complete flushing of carbide debris out of the spark gap, low re-depo- sition of degraded material, and low SR. Surface roughness has a slight tendency to decrease with increased WF. Increasing wire feed allows carbide debris to easily escape from the spark gap, resulting in re-cast layer. To get an excellent surface finish, keep the electrical discharge energy to a minimum by choosing minimal T and an excessive T [47]. The quantity of sin- on off gle pulse discharge energy is unaffected by an increase in pulse interval; rather, it only affects the length of discharge time per unit of time. Consequently, the discharge duration per unit time doubles and the cutting speed decreases linearly as the pulse interval grows. It has been found that raising the T causes the MRR to decrease. This action enhances the procedure by off PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 18 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 13. Effect of Pulse on Time on responses. https://doi.org/10.1371/journal.pone.0308203.g013 enabling a more effective flush of debris into the gap. 50 μs is the ideal pulse off time for lower- ing SR and K [48]. Effect of GV on responses. Fig 15 shows the effect of GV on responses. The findings indi- cate that the MRR rises as the GV falls. 20 V is the ideal gap voltage for increasing MRR. It’s important to note that 20 V is the base alloy’s ideal voltage. Due to a greater electric field, spark discharge actually occurs beneath the same gap more frequently when voltage increases. Less voltage can provide enough energy to melt the dielectric particles in the vicinity. The SR reduces as the GV rises. This occurrence is remarkable. Lower GV may have sufficient energy to melt the re-solidified particles in the vicinity, which remain on the machined surface and produce a large number of projecting peaks. On the other side, a smoother surface is produced by high voltage [49]. Melting of the component’s surface is determined by the thermal conduc- tivity of the work piece material and the quantity of energy used per spark that is assumed to be proportional to T and GV. Enhancing the pulse on time (T ) generates more heat at the on on work surface, improving the cutting speed. Lowering Gap voltage decreases the spark gap, resulting in fast and significant ionization of the dielectric fluid, causing greater melting of the work material and hence increases cutting speed [50]. Increasing electrostatic force brought on the increasing GV causes wire wrapping during the discharge process. The SR lowers as the gap voltage rises. 40 V is the ideal GV for getting lower K . Sparks that form at the conducting phase and produce melting or possible evaporation also contribute to the craters on the machined surface. It goes without saying that huge K are caused by high crater diameters [51, 52]. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 19 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 14. Effect of Pulse off Time on responses. https://doi.org/10.1371/journal.pone.0308203.g014 Effect of WF on responses. Fig 16 shows the effect of WF on responses. The WF ought to be selected in a way that prevents wire breaking. As the wire feed accelerates, the MRR ascend. For maximizing MRR, a wire feed setting of 6 m/min is ideal. In a word, this investigation’s findings supported those in the literature [53]. At high wire feed values, the increase in cutting speed is particularly noticeable. Higher discharge energy causes increased melting and evapo- ration of the work material, resulting in the release of a significant quantity of carbide debris, which coagulates in the spark gap and so influences the machining process through the genera- tion of arcs. Raising the wire feed rate allows the eroded material to escape the spark gap more easily and quickly [54]. Even though the research was done on a variety of materials, the out- comes were reliable. It is important to choose a WF that wire won’t break. As the WF grows, so does the cutting speed. For a minimum SR, a wire feed setting of 6 m/min is ideal. Briefly stated, the findings of this investigation were consistent with those discovered in the litera- tures. The investigation was conducted on a variety of materials, but the outcomes were con- stant. A non-critical metric is wire feed. The ideal WF for the lowest K is 3 m/min. Increasing the WF rate might make the wire less rigid during discharge, which would reduce the amount of wire that would wind back during sparks and reduce the destructive power of the sparks on the work piece surface, leading to a higher K [55, 56]. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 20 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 15. Effect of Gap Voltage on responses. https://doi.org/10.1371/journal.pone.0308203.g015 Mathematical models of LM5 aluminium alloy Multiple linear regression (MLR) models frequently provide a suitable illustration of a more complex structure within particular ranges of the independent variables. The generalization model of MLR analysis describes the relationship among the response and the independent variables, and the estimated response is calculated using the generalized simple regression equation. The constants were calculated using the linear regression analysis method using MINITAB software. Mathematical models for MRR, SR and K for LM5 alloy are developed using Linear Regression are shown in Eqs 1–3. MRR ¼ 26:38þ 0:2857 T þ 0:01473 T 0:10002 GVþ 0:0585 WF ð1Þ on off SR ¼ 4:81þ 0:0760 T 0:00292 T 0:00575 GV 0:0278 WF ð2Þ on off Kerf ¼ 0:0225þ 0:002400 T þ 0:000300 T 0:000450 GVþ 0:000500 WF ð3Þ on off 2 2 From the ANOVA of the Regression, the R value of MRR is 99.7, the R value of SR is 82.6 and the R value of Kerf is 91.4. The aforementioned mathematical model for MRR, SR and kerf is critical for selecting machining settings while machining LM5 Aluminium alloy using WEDM. According to Eqs 1 and 3, MRR and Kerf are directly correlated to T , T and WF, but inversely correlated to on off gap voltage. Eq 2 demonstrates that SR is directly proportional to T and inversely propor- on tional to T , GV and WF. off PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 21 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 16. Effect of Gap Voltage on responses. https://doi.org/10.1371/journal.pone.0308203.g016 From Eq (1), it is seen that T , T and WF are directly proportional and GV is inversely on off proportional to MRR. The percentage deviation of MRR values between the experimental and predicted values are reported in Table 11. Referring to Table 11 on MRR model, it can be seen lesser deviation occurs between experimental and predicted values. The average absolute error Table 11. Experimental and predicted values of MRR, SR and kerf. Ex A B C D Experimental Predicted % Experimental Predicted % Experimental Predicted % No T T GV WF (m/ MRR (mm3/ MRR Deviation SR SR Deviation kerf kerf Deviation on off (μs) (μs) (V) min) min) 1 110 30 20 3 3.61 3.66 -1.5 3.1 3.26 -5.29 0.286 0.288 -0.7 2 110 40 30 6 2.98 2.99 -0.22 3.27 3.09 5.39 0.286 0.288 -0.7 3 110 50 40 9 2.26 2.31 -2.18 2.76 2.92 -5.93 0.286 0.288 -0.7 4 115 30 30 9 4.5 4.44 1.26 3.63 3.42 5.79 0.304 0.299 1.81 5 115 40 40 3 3.4 3.24 4.72 3.45 3.5 -1.44 0.295 0.294 0.34 6 115 50 20 6 5.52 5.56 -0.77 3.66 3.5 4.31 0.313 0.308 1.76 7 120 30 40 6 4.56 4.7 -2.98 3.83 3.83 0.11 0.304 0.305 -0.16 8 120 40 20 9 7.09 7.02 1 3.63 3.83 -5.45 0.313 0.318 -1.6 9 120 50 30 3 5.78 5.82 -0.61 3.96 3.91 1.31 0.313 0.314 -0.16 https://doi.org/10.1371/journal.pone.0308203.t011 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 22 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy for MRR is 1.69%. These deviations could be attributed to limitation of the modeling not accounting for interactive influences [57, 58]. From Eq (2), it is seen that GV, T and WF are directly proportional and T is inversely off on proportional to SR. The percentage deviations of SR values between the experimental and pre- dicted values are reported in Table 11. Referring to Table 11 on SR model, it can be seen lower order deviation between experimental and predicted values occurs. The average absolute error for SR is 3.89%. These deviations could be attributed to limitation of the modeling not accounting for interactive influences. From Eq (3), it is seen that T , T and WF are directly proportional and GV is inversely on off proportional to kerf. The percentage deviation of kerf values between the experimental and predicted values are reported in Table 11. Referring to Table 11 on kerf model, it can be seen minimum deviation between experimental and predicted values occurs. The average absolute error for kerf is 0.88%. Conclusions The LM5 aluminium alloy was successfully produced via the stir casting technique. The subsequent outcomes were attained i. The microscopic and SEM images show the microstructure of the LM5 Al alloy and EDX shows the constituents. ii. The Pulse on Time (68.25%) and Gap Voltage (31.68%) has the greatest statistical effects on MRR. The error associated with the ANOVA is 0.07%, indicating a 95% confidence level. iii. The most significant statistical impact on SR is possessed by the Pulse on Time (79.46%). The ANOVA error for SR is 0.91%, indicating a 95% confidence level. iv. The greatest statistical influence on K is by the Pulse on Time (81.97%). The 95% confi- dence level is shown by the error associated with the ANOVA for K , is 1.56%. v. The error % associated with the predicted and experimented values of the MRR is 2.53%, SR is 0 and K is 2.81%, based on the confirmation experiments. vi. The EDS of the machined surface (recast layer) shows oxygen, carbon, copper and zinc. These are formed due to the dielectric fluid and tool material (brass wire). vii. The average absolute error for MRR is 1.69%, for SR is 3.89% and for kerf is 0.88%, based on mathematical (linear regression) models. viii. The predicted and experimental values based on the mathematical models are very close to each other confirms that the Taguchi’s Signal to Noise ratio analysis is the most appro- priate for single objective optimization of responses. The findings of this investigation promises to enhance precision and surface quality in wire electro-discharge machining of LM5 aluminium alloy in today’s production processes. Author Contributions Conceptualization: Sunder Jebarose Juliyana, Charles Sarla Rubi. Formal analysis: Sunder Jebarose Juliyana, Charles Sarla Rubi, Arasumugam Divya Sadhana. Funding acquisition: Emad Abouel Nasr. Investigation: Sunder Jebarose Juliyana. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 23 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Methodology: Sunder Jebarose Juliyana, Charles Sarla Rubi, Arasumugam Divya Sadhana. Project administration: Charles Sarla Rubi, Arasumugam Divya Sadhana. Resources: Robert Čep, Charles Sarla Rubi, Sachin Salunkhe, Arasumugam Divya Sadhana. Software: Sunder Jebarose Juliyana, Charles Sarla Rubi. Supervision: Jayavelu Udaya Prakash, Sachin Salunkhe. Validation: Jayavelu Udaya Prakash, Robert Čep, Sachin Salunkhe, Emad Abouel Nasr. Visualization: Emad Abouel Nasr. Writing – original draft: Jayavelu Udaya Prakash. Writing – review & editing: Robert Čep. References 1. Wang L., Makhlouf M. and Apelian D., 1995. Aluminium die casting alloys: alloy composition, micro- structure, and properties-performance relationships. International materials reviews, 40(6): 221–238. 2. Rajan R., Kah P., Mvola B. and Martikainen J., 2016. Trends in aluminium alloy development and their joining methods. Reviews on Advanced Materials Science, 44(4) ´ ´ 3. Liptakova T., Zatkalıkova V., Alaskari A.M. and Malcho M., 2019. Dominant factors affecting erosion- corrosion resistance of aluminium–brass pipelines. Journal of Engineering Research, 7(4). 4. Boggarapu V., Sreekanth P.R. and Peddakondigalla V.B., 2023. Microstructure, Mechanical and Tribo- logical Properties of Al/Cu Functionally Graded Material Fabricated through Powder Metallurgy. Journal of Engineering Research, p.100119. 5. Ananth S., Prakash J.U., Juliyana S.J., Rubi C.S. and Sadhana A.D., 2021. Effect of process parame- ters on WEDM of Al–Fly ash composites using Taguchi Technique. Materials Today: Proceedings, 39, pp.1786–1790. 6. Ho K.H., Newman S.T., Rahimifard S. and Allen R.D., 2004. State of the art in wire electrical discharge machining (WEDM). International Journal of Machine Tools and Manufacture, 44(12–13), pp.1247– 7. Saravanan R., Anbuchezhiyan G., Mamidi V.K. and Kumaran P., 2022. Optimizing WEDM parameters on nano-SiC-Gr reinforced aluminium composites using RSM. Advances in Materials Science and Engi- neering, 2022. https://doi.org/10.1155/2022/1612539 8. Natarajan K., Ramakrishnan H., Gacem A., Vijayan V., Karthiga K., Ali H.E., et al. 2022. Study on opti- mization of WEDM process parameters on stainless steel. Journal of Nanomaterials, 2022. 9. Rouniyar A.K. and Shandilya P., 2021. Semi-empirical modeling and optimization of process parame- ters on overcut during MFAPM-EDM of Al6061 alloy. Proceedings of the Institution of Mechanical Engi- neers, Part E: Journal of Process Mechanical Engineering, 235(6), pp.1784–1796. https://doi.org/10. 1177/09544089211015890 10. Kar S. and Sahoo R., 2023. Viability and performance assessment of bipolar linear self-servo mecha- nism-based Maglev electrical discharge machining by processing Al-6062 alloy. Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering, p.09544089231207685. 11. Mahapatra S.S. and Patnaik A., 2006. Parametric optimization of wire electrical discharge machining (WEDM) process using Taguchi method. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 28, pp.422–429. 12. Verma R.K., Patel D. and Chopkar M.K., 2023. Wear behavior investigation of Al-B C functionally graded composite through Taguchi’s design of experiment. Journal of Engineering Research, p.100095. 13. Sarkar S., Sekh M., Mitra S. and Bhattacharyya B., 2011.A novel method of determination of wire lag for enhanced profile accuracy in WEDM. Precision Engineering, 35(2), pp.339–347. https://doi.org/10. 1016/j.precisioneng.2011.01.001 14. Arunnath A., Madhu S. and Tufa M., 2022. Experimental investigation and optimization of material removal rate and tool wear in the machining of aluminium-boron carbide (Al-B4C) nanocomposite using EDM process. Advances in Materials Science and Engineering, 2022. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 24 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy 15. Salavaravu L. and Dumpala L., 2022. Effects of process parameters on mechanical and metallurgical properties of AA5083 weld bead and optimization by using Taguchi based Grey Relational Analysis and ANOVA of submerged friction stir welding. Journal of Engineering Research (2307–1877), 10. 16. Zafar M.M. and Toor Z.S., 2023. Experimental and numerical investigation of electrochemical and tribo- logical behavior of aluminium alloys. Journal of Engineering Research, p.100134. 17. Kumar N., Kumari S., Abhishek K., Nandi G. and Ghosh N., 2022. Study on various parameters of WEDM using different optimization techniques: A review. Materials Today: Proceedings, 62, pp.4018– 18. Kansal H.K., Singh S. and Kumar P., 2007. Technology and research developments in powder mixed electric discharge machining (PMEDM). Journal of materials processing technology, 184(1–3), pp.32– 19. Debnath T. and Patowari P.K., 2019. Fabrication of an array of micro-fins using Wire-EDM and its parametric analysis. Materials and Manufacturing Processes, 34(5), pp.580–589. 20. Sarmah P. and Patowari P.K., 2023. Machinability study of Al 6063-based MMCs with SiC reinforce- ment particles using WEDM process. Materials and Manufacturing Processes, 38(7), pp.783–796. 21. Sahoo R., Kar S. and Patowari P.K., 2020. Experimental study on Titanium diamond by fabricating micro-holes using Micro-EDM with Taguchi’s method and response surface method. Engineering Research Express, 2(4), p.045026. 22. Sahoo R., Singh N.K. and Bajpai V., 2023. A novel approach for modeling MRR in EDM process using utilized discharge energy. Mechanical Systems and Signal Processing, 185, p.109811. 23. Sahoo R., Singh N.K. and Bajpai V., 2023. Approach towards green manufacturing in Maglev EDM using different biodegradable dielectrics at variable discharge conditions. Journal of Cleaner Produc- tion, 430, p.139623. 24. Ye L., Qian J., Haitjema H. and Reynaerts D., 2022. On-machine chromatic confocal measurement for micro-EDM drilling and milling. Precision Engineering, 76, pp.110–123. 25. Juliyana S.J., Prakash J.U., Salunkhe S., Hussein H.M.A. and Gawade S.R., 2022. Mechanical Charac- terization and Microstructural Analysis of Hybrid Composites (LM5/ZrO /Gr). Crystals, 12(9), p.1207. 26. Jebarose Juliyana S. and Prakash J.U., 2022. Optimization of machining parameters for wire EDM of AMCs (LM5/ZrO ) using Taguchi technique. INCAS Bulletin, 14(1), pp.57–68. 27. Jebarose Juliyana S., and Prakash J.U., 2020. Drilling parameter optimization of metal matrix compos- ites (LM5/ZrO2) using Taguchi Technique. Materials Today: Proceedings, 33, pp.3046–3050. 28. Jebarose Juliyana S., Udaya Prakash J.,Čep R. and Karthik K., 2023. Multi-Objective Optimization of Machining Parameters for Drilling LM5/ZrO2 Composites Using Grey Relational Analysis. Materials, 16 (10), p.3615. https://doi.org/10.3390/ma16103615 PMID: 37241242 29. Jebarose Juliyana S., Prakash J.U., Sadhana A.D. and Rubi C.S., 2022. Multi-objective optimization of process parameters of wire EDM for machining of AMCs (LM5/ZrO2) using grey relational analysis. Materials Today: Proceedings, 52, pp.1494–1498. 30. Juliyana S.J., Prakash J.U. and Salunkhe S., 2022.Optimisation of wire EDM process parameters using Taguchi technique for machining of hybrid composites. International Journal of Materials Engineering Innovation, 13(3), pp.257–271. 31. Udaya Prakash J; Jebarose Juliyana S.; Salunkhe S.; Gawade S.R.; Nasr E.S.A.; Kamrani A.K., 2023. Mechanical Characterization and Microstructural Analysis of Stir-Cast Aluminum Matrix Composites (LM5/ZrO2). Crystals, 13, 1220. 32. Wang B., Liu X., Wang J., Li Q., Liu K., & Zhang M., 2022. Uncovering the effects of Ce and superheat temperature on Fe-rich intermetallic and microporosity formation in aluminum alloy. Materials Charac- terization, 193, 112226. 33. Balan A.S.S. and Giridharan A., 2017. A progress review in wire electrical discharge machining pro- cess. International Journal of Automotive and Mechanical Engineering, 14, pp.4097–4124. 34. Kousik Kumaar R., Somasundara Vinoth K. and Kavitha M., 2022. Dry sliding wear performance of AA7075/MoS composite materials. Journal of Engineering Research (2307–1877), 10 35. Paulson D.M., Saif M. and Zishan M., 2023. Optimization of wire-EDM process of titanium alloy—Grade 5 using Taguchi’s method and grey relational analysis. Materials Today: Proceedings, 72, pp.144–153. 36. George J., Philip J.T., Mathew J. and Manu R., 2022. Prediction and analysis of material removal rate and white layer thickness during wire electrical discharge turning (WEDT) process. CIRP Journal of Manufacturing Science and Technology, 39, pp.210–222 37. Ram H.S., Uthayakumar M., Kumar S.S., Kumaran S.T. and Korniejenko K., 2022.Modelling Approach for the Prediction of Machinability in Al6061 Composites by Electrical Discharge Machining. Applied Sci- ences, 12(5), p.2673. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 25 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy 38. Sahu S.K., Anand M.V., Kumar T.C.A., Kumar A., Prasad G.S. and Niraj V.V., 2023. Kerf width analysis of wire electrical discharge machining of titanium alloy. Materials Today: Proceedings. 39. Kumar A., Kumar V. and Kumar J., 2016. Surface crack density and recast layer thickness analysis in WEDM process through response surface methodology. Machining Science and Technology, 20 (2), pp.201–230. 40. Azam M., Jahanzaib M., Abbasi J.A., Abbas M., Wasim A. and Hussain S., 2016. Parametric analysis of recast layer formation in wire-cut EDM of HSLA steel. The International Journal of Advanced Manufacturing Technology, 87, pp.713–722. 41. Zhang Y., Liu Y., Ji R. and Cai B., 2011. Study of the recast layer of a surface machined by sinking elec- trical discharge machining using water-in-oil emulsion as dielectric. Applied surface science, 257 (14), pp.5989–5997. 42. Chaudhari R., Shah H., Ayesta I., de Lacalle L.L. and Vora J., 2022. Experimental investigations and optimization of WEDM parameters using Taguchi analysis of pure titanium. In Recent Advances in Mechanical Infrastructure: Proceedings of ICRAM 2021 (pp. 349–358). Springer Singapore. 43. Sharmila B. and Selvakumar G., 2022. Investigations on the effect of dielectric medium and WEDM parameters on surface characteristics of Al 7068 (ordnance aluminium) alloy. Surface Topography: Metrology and Properties, 10(3), p.035031. 44. Han F., Jiang J. and Yu D., 2007. Influence of machining parameters on surface roughness in finish cut of WEDM. The International Journal of Advanced Manufacturing Technology, 34, pp.538–546. 45. Li W., Zhao L., Wang G. and Zhou C., 2023. Manufacturing of Mold Cavity of Micro-double Gear by WEDM. In Journal of Physics: Conference Series (Vol. 2459, No. 1, p. 012065). IOP Publishing. 46. Bisaria H. and Shandilya P., 2023. Surface integrity of NiTi-SMA during WEDM: effect of spark parame- ters. Materials and Manufacturing Processes, pp.1–9. 47. Sen R., Choudhuri B., Barma J.D. and Chakraborti P., 2020. Surface integrity study of WEDM with vari- ous wire electrodes: Experiments and analysis. Machining Science and Technology, 24(4), pp.569– 48. Sharma N., Gupta R.D., Khanna R., Sharma R.C. and Sharma Y.K., 2023. Machining of Ti-6Al-4V bio- medical alloy by WEDM: Investigation and optimization of MRR and Rz using grey-harmony search. World Journal of Engineering, 20(2), pp.221–234. 49. Chaubey S.K. and Gupta K., 2023. A review on Wire-EDM of bio titanium. Reports in Mechanical Engi- neering, 4(1), pp.141–152. 50. Sarala Rubi C., Udaya Prakash J., Jebarose Juliyana S., RobertČep, Sachin Salunkhe, Kouril K., et al. 2024. Comprehensive review on wire electrical discharge machining: a nontraditional material removal process. Frontiers in Mechanical Engineering, 10:1322605. 51. Raj A., Misra J.P., Khanduja D., Saxena K.K. and Malik V., 2022. Design, modeling and parametric opti- mization of WEDM of Inconel 690 using RSM-GRA approach. International Journal on Interactive Design and Manufacturing (IJIDeM), pp.1–11. 52. Babu R.R., Subhair S. and Siva S., 2018. Optimization of process parameter in WEDM for Stainless Steel 316 by Using Taguchi Method. International Journal of Engineering Research & Technology (IJERT). 53. Prakash J.U., Sivaprakasam P., Garip I., Juliyana S.J., Elias G., Kalusuraman G., et al. 2021. Wire elec- trical discharge machining (WEDM) of hybrid composites (Al-Si12/B4C/fly ash). Journal of Nanomater- ials, 2021, pp.1–10. 54. Kumar A., Kumar D.V. and Kumar D.J., 2011. A review on the state of the art in wire electric discharge machining (WEDM) process. International Journal of Mechanical Engineering Research and Develop- ment (IJMERD), 1(1). 55. Patra D.R., Rout I.S. and Sahoo M., 2015. Optimization of WEDM parameters using Taguchi method for higher material removal rate on EN31 steel. International Journal of Engineering Research and Applications, 5(6), pp.57–62. 56. Ehsan Asgar M. and Singh Singholi A.K., 2018. September. Parameter study and optimization of WEDM process: A Review. In Iop conference series: Materials science and engineering (Vol. 404, p. 012007). IOP Publishing. 57. Muthukumar V., Rajesh N., Venkatasamy R., Sureshbabu A. and Senthilkumar N., 2014. Mathematical modeling for radial overcut on electrical discharge machining of Incoloy 800 by response surface meth- odology. Procedia Materials Science, 6, pp.1674–1682. 58. Muthukumar V., Suresh Babu A., Venkatasamy R. and Senthil Kumar N., 2015. An accelerated particle swarm optimization algorithm on parametric optimization of WEDM of die-steel. Journal of The Institu- tion of Engineers (India): Series C, 96, pp.49–56. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 26 / 26 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png PLoS ONE Public Library of Science (PLoS) Journal

Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy

Download PDF
26 pages

Loading next page...
 
/lp/public-library-of-science-plos-journal/taguchi-optimization-of-wire-edm-process-parameters-for-machining-lm5-bI0crQfynk

References (65)

Publisher
Public Library of Science (PLoS) Journal
Copyright
Copyright: © 2024 Juliyana et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: All relevant data are within the manuscript. Funding: The authors extend their appreciation to King Saud University for funding this work through Researchers Supporting Project number (RSP2023R164), King Saud University, Riyadh, Saudi Arabia. The role of funder in the revised paper is as follows. 1) Edit and revised the comments from reviewers 2) Providing the testing facility 3) The funder is involved in data collection, design a Taguchi model. 4) Preparation of manuscript 5) The funder is designed a methodology. Competing interests: The authors have declared that no competing interests exist.
eISSN
1932-6203
DOI
10.1371/journal.pone.0308203
Publisher site
See Article on Publisher Site

Abstract

Citation: Juliyana SJ, Prakash JU,Čep R, Rubi CS, LM5 alloy is suitable for metal castings for marine and aesthetic uses due to its admirable Salunkhe S, Sadhana AD, et al. (2024) Taguchi resistance to corrosion. In order to make intricate shapes in the LM5 alloy, this study intends optimization of Wire EDM process parameters for machining LM5 aluminium alloy. PLoS ONE to assess the impact of Wire Electric Discharge Machining process variables, like Pulse on 19(10): e0308203. https://doi.org/10.1371/journal. Time (T ), Pulse off Time (T ), Gap Voltage (GV) and Wire Feed (WF) on responses like on off pone.0308203 Material Removal Rate (MRR), Surface Roughness (SR), and Kerf Width (K ). The LM5 Editor: Siddhartha Kar, Ramaiah Institute of aluminium alloy plate was produced through stir casting process. SEM, EDAX and XRD Technology, INDIA images confirm the LM5 Al alloy’s microstructure and crystal structure. WEDM studies were Received: November 25, 2023 conducted using design of experiments approach based on L orthogonal array and ana- Accepted: July 19, 2024 lysed using Taguchi’s Signal to Noise Ratio (S/N) analysis. Pulse on Time has the greatest statistical effects on MRR (68.25%), SR (79.46%) and kerf (81.97%). In order to assess the Published: October 28, 2024 surface integrity of the WEDM machined surfaces, the SEM study on the topography was Copyright:© 2024 Juliyana et al. This is an open conducted using the optimum surface roughness process variables: T 110μs, T 50μs, on off access article distributed under the terms of the Creative Commons Attribution License, which GV 40 V, and WF 9 m/min. SEM images show the recast layer and its thickness. The aver- permits unrestricted use, distribution, and age absolute error for MRR is 1.69%, SR is 3.89% and kerf is 0.88%, based on mathemati- reproduction in any medium, provided the original cal (linear regression) models. The Taguchi’s Signal to Noise ratio analysis is the most author and source are credited. appropriate for single objective optimization of responses. Data Availability Statement: All relevant data are within the manuscript. Funding: The authors extend their appreciation to King Saud University for funding this work through Researchers Supporting Project number Introduction (RSP2023R164), King Saud University, Riyadh, Saudi Arabia. The role of funder in the revised Wrought alloys and cast alloys are the two main divisions of aluminium alloys. In the first paper is as follows. 1) Edit and revised the case, alloys undergo treatment in a solid state but in the later; they are liquefied in a furnace comments from reviewers 2) Providing the testing and then poured into moulds. Aluminium alloys may be categorized as heat-treatable or non- facility 3) The funder is involved in data collection, design a Taguchi model. 4) Preparation of heat-treatable depending on the strengthening mechanisms used. Due to their beneficial PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 1 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy manuscript 5) The funder is designed a qualities, such as their high strength-to-weight ratio, affluence of fabrication, high degree of methodology. workability, noteworthy ductility, owing thermal conductivity, strong resistance to corrosion, and appealing physical appearance at their inherent finish, aluminium alloys have become Competing interests: The authors have declared that no competing interests exist. popular as structural materials over the past few years [1]. Because of this, the marine industry currently uses 25% of the world’s aluminium manufacturing. The thin, malleable metal aluminium has great plasticity, acceptable weldability, sufficient tensile and compressive strength, and extremely significant thermal and electrical conductiv- ity. The electronics industries use it frequently. In the automotive, marine, and aerospace industries, where weight and mechanical qualities are prioritized, aluminium alloys are pri- marily employed. Aluminium alloy’s machinability is far worse than pure aluminium. The degree of strain hardening, soft particles, and precipitates has each a positive impact on an alloy’s capacity to be machined [2]. Nevertheless, aluminium alloys are regarded as tough to machine materials, especially for dry machining, not withstanding their mechanical qualities. Its high heat conductivity, low melting point and propensity to stick to the cutting edges of tool materials are difficulties. In addition to the high thermal conductivity of aluminium, which removes a significant amount of heat from the cutting edge into the work piece throughout the machining process, the material is thermally deformed. Because of the low melting point of aluminium alloys, there are issues with chip development, chip elimination, and material clinging to the cutting tool [3]. So it would appear to be quite beneficial to utilize an innovative way while cutting aluminium. Due to aluminium’s strong electrical conductivity, WEDM technique should be appropriate [4]. In order to remove material, WEDM technology uses thermoelectric energy amongst the work piece and a wire electrode. The pulse discharging takes away the material from the work piece by melting and evaporating it in a tiny space separating the work piece and the electrode. This technique is typically employed to produce intricate shapes and to manufacture materials that are challenging for regular equipment to work with [5]. EDM is superior to traditional machining in many ways. Electrically conductive materials can be cut using EDM, and this technique has been used to machine work pieces that have been heat-treated and hardened. Intricate and complex profiles can be cut more quickly, precisely, and affordably. Burrs have been avoided and thin, delicate parts have been created with ease. The mechanical properties of the material have no bearing on the WEDM cutting process; the only need is to achieve the bare minimum conductivity of the processed material. Because of its great accuracy and mini- mum surface roughness, this technology is suited for cutting very hard conductive materials, composites, ceramics, or sandwiches. Considering the significance of WEDM manufacturing, it must be less expensive than traditional machining in order to be successful. In general, the use of electrical discharges represents a trade-off between productivity and machining quality. Numerous factors might have an impact on the WEDM cutting process [6]. Each one of them has a different impact on the price of production and the work piece’s final grade finish. Thus, applying DoE should be beneficial. Schedule of test and statistical analysis of specified plan make up the DoE (Design of Experiments), very effective. The conclusion of a planned experi- ment’s evaluation is whether the variables under observation were affected by the tested ele- ments. An experiment’s output is a particular value of the observable variable, also known as the dependent variable or response, which describes the test’s quality. The ultimate quality is affected by a wide range of factors. Under the perspective of experimental design, they can be classified into specific and randomized. WEDM machine parameters are the inputs to the pro- cedure of machining. Just those factors which have a statistically noteworthy outcome on the degree of quality should be ultimately chosen after thoroughly examining all components and their common interactions. It enables the variables that matter most to be set at their ideal PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 2 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy values while also identifying the unnecessary variables, lowering their tolerance and thereby lowering manufacturing costs [7, 8]. Research on WEDM on Al6061 alloy show that EDM, a nonconventional machining method, is superior in terms of dimensional accuracy and precision; when it comes to cutting hard, conductive materials. In EDM, sparks repeatedly discharge when voltage is applied after both electrodes have been dipped in dielectric. These sparks flush away the material from the surface as melted and degraded particles that are removed by dielectric. Researchers have taken into consideration advancements and modifications to the EDM process over time. A recent invention to improve the EDM process’s competency involves agglomerating powder with both an external magnetic field and a dielectric. This procedure demonstrates the ability to machine complex and sophisticated 3D profiles with reduced tool wear rate (TWR), increased productivity, and better surface quality and precision. The MFAPM-EDM technique is used in the industrial, aerospace, automotive, defense, and surgical industries for the machining of different components. In the PMEDM process, the powder particles’ accumula- tion of charges causes several sparks to occur during machining. More charges are produced when ions from powder particles collide with dielectric molecules due to the accumulated charge in the machining area. Multiple sparks cause the surface materials to be removed more quickly and create shallow craters, which improved MRR and SR [9]. The feasibility of utilizing the Maglev EDM for machining aluminium 6062 alloy is assessed and examined. The Maglev EDM achieves tool location by means of a methodical combination of magnetic repulsive forces. Brass tools were used in the experiments, and the surrounding air served as a dielectric for the Al-6062 alloy. The data from the previously published literature was further contrasted with the experimental results. Field-emitting SEM investigation of the machined surface revealed the presence of recast layers, globules, lumps of debris, melted debris, micro-pores, micro-voids, micro-cracks, and craters [10]. WEDM’s essential indicators of performance are MRR, SR, and kerf. The MRR in WEDM processes determines the cost of machining and the rate of output. The primary purpose when establishing the machining parameters is to maximize MRR and minimize SR. The Taguchi technique, a powerful experimental design tool, takes a simple, effective, and systematic approach to determining the ideal machining parameters. Furthermore, this strategy has a low experimental cost and effectively reduces the effect of the source of variation. An economical and simple technology for modifying machined surfaces while maintaining accuracy must be developed [11]. The main objective of this study is to optimize the WEDM process parameters for machin- ing LM5 aluminium alloy using Taguchi’s Signal to Noise ratio (S/N) analysis. Other objective is to study the effect of machining parameters on the MRR, SR and kerf. Another objective is to construct Mathematical models using regression equations and to find the deviation % between the experimental and predicted values of MRR, SR and kerf. The machined surfaces were analyzed under SEM to find the recast layer. Literature survey Numerous researchers adjusted the settings of several machines to produce high-quality prod- ucts [12]. Various techniques have been used in past to measure how control parameters affect part surface characteristics, MRR and K . The unstable wire portion is the main cause of diffi- culties [13]. To ascertain the relationship between machining performance and machining parameters, an experimental analysis of the wire breakage phenomenon using a thermal model was conducted. With a reduction in T , the MRR initially rises [14]. However, the gap off becomes unstable after a relatively brief period of time, which lowers the MRR. As the MRR PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 3 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy rises, surface quality declines. K and SR have been found to be significantly influenced by the T and the D . Additionally, it has been discovered that using a medium D can improve on ww ww SR; modifying T and D can regulate final cutting. on ww It was discovered that by reducing I, T and T , surface roughness can be improved. on off Shorter and long pulses will create a comparable SR but differing surface morphologies and MRR. In comparison to long pulse duration, the clearance rate is significantly higher under short pulse duration. Furthermore, compared to a long pulse, a short pulse can produce better SR [15]. Numerous criteria and difficult experimental effort are needed for the study of the WEDM process, which takes a greater amount of money and time. Therefore, it is vital to optimize the control parameters to decrease the money and time required to assess the WEDM process. There is research on experiment design that enables a method to show the effect of control fac- tors with the least amount of experiments. Some investigations use fractional factorial designs, while others use full-factorial designs to evaluate variables with impact [16]. Coated wires are chosen so as to achieve homogeneous surface qualities. Most delicate char- acteristic that affects the creation of a layer made up of a combination of oxides is the T and on T . The development of oxides can be significantly reduced by reducing the duration between off two pulses [17]. Research of Kansal et al [18] shows in PMEDM, the electrically conductive powder is mixed in the dielectric of EDM, which reduces the insulating strength of the dielectric fluid and increases the spark gap between the tool and work piece. As a result, the process becomes more stable, thereby, improving the material removal rate (MRR) and surface finish. The pow- der particles simplify the igniting process by increasing the spark gap and reducing the insulat- ing strength of the dielectric fluid. The WEDM process’s performance for titanium alloys was the focus of investigation of Debnath and Patowar [19]. Based on Taguchi DoEs, the results show that the flushing pressure of dielectric, wire tension, and T are important process parameters that have a considerable on impact on the machined hole’s circularity, cylindricity, and diametral errors, since wire ten- sion affects both stability and wire electrode rigidity. This study does not include the MRR and SR, which are the most vital process responses. In the case of MMCs, T is the most important component for K [20]. Taguchi’s DoE on w based orthogonal arrays should be used instead of full factorial experiments. Sahoo et al. [21] used EDM technology to analyze titanium diamond machining perfor- mance measures. I and T are the chosen control variables, and during experimental runs, a p on duty factor of 50% to 75% is maintained. R , TWR and MRR are the responses taken into account for this operation. The analysis has been carried out using L OA. In order to find the ideal input parametric combination, analysis of variance is also used in conjunction with over- all evaluation criteria (OEC). The findings of the analysis indicate that T has a greater influ- on ence on TWR than does current on MRR and R . Additionally, the identification of various material phases on a machined work surface is aided by EDX, SEM and XRD. For tool positioning, Maglev EDM uses a novel bipolar linear self-servo technique that pro- duces consistent machining stability using a special magnetic repulsive force balance action. The performance characteristics of each dielectric have been investigated. The evaluation of surface morphology shows that the use of bio-dielectrics has the potential to significantly improve surface uniformity and reduce deformities [22, 23]. For the processing of nanosecond pulsed lasers, a unique and straightforward model based on geometric mathematics and heat transport was proposed. Experiments verify that the mod- els are feasible. The single-pulse laser ablation craters had errors in both diameter and depth of 2.56% - 7.14% and 6.82% - 18.91%, respectively. Between 3.47% and 12.47% is the recast layer PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 4 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Table 1. Chemical composition of LM5 aluminium alloy. Cu Mg Si Mn Fe Pb Zn Al 0.032 3.299 0.212 0.022 0.268 0.02 0.01 Balance https://doi.org/10.1371/journal.pone.0308203.t001 depth error for consecutive stacked ablation pulses. On metal surfaces ablated using pulsed laser, it forecasts the shape of the recast layer. The model serves as a guide for the preparation of surface functionality [24]. Based on the extensive literature survey, the authors of this article were persuaded to carry out a research study to recommend the most optimal machining parameters for efficiently machining stir-casted LM5 Aluminium alloy in order to achieve maximum MRR, minimum SR and minimum Kerf using Pulse on Time (T ), Pulse off Time (T ), Gap Voltage (GV) on off and Wire Feed (WF) at three levels, which have not been reported before by any researcher. This study also provides a general overview of a thorough process to determine the best machining parameter settings based on the design of experiments approach. The research paper indeed aims the mathematical models that are constructed to associate the machining response characteristics with machining control parameters, in addition to revealing the results of signal/noise (S/N) ratio analysis and ANOVA. To better understand the occurrence of recast layer generation during machining, energy dispersive spectroscopy (EDS) analysis and scanning electron microscopy (SEM) analysis were used to examine the machined surface textures. Materials and methods LM5 aluminium alloy The LM5 aluminium alloy has excellent casting properties, sturdy structure, and great durabil- ity against corrosion. The material is commonly utilized in the automotive, aerospace, and marine sectors, whenever a combination of properties is required [25, 26]. Since it is easy to grind, weld, and cast into complex shapes, Machinery require extremely strong resistance to corrosion from sea water or marine atmospheres, along with castings which need to display and maintain a high polish Gestalt, all have uses for LM5 aluminium [27, 28]. The chemical composition of LM5 Aluminium Alloy using Optical Emission Spectrometry (ASTM E 1251– 07) is shown in Table 1. Fabrication Stir casting process was used to manufacture plates of the LM5 Al alloy, measuring 120*120*10 mm. Easy use, inexpensive manufacturing, a uniform dispersion of reinforcing elements, and improved mechanical qualities are only a few benefits of stir casting. Fig 1 depicts the stircasting setup. A graphite-coated container served to melt LM5 alloy ingots in a furnace powered by elec- tricity. To 850˚C, the ambient temperature was raised steadily. The liquid state of the melt at 800˚C was degassed using hexachloroethane. The molten metal was subsequently fed into pre- heated (650˚C) cast-iron moulds after being agitated at 600 rpm for 10 minutes [29, 30]. Micro-structural analysis Fig 2A shows the Optical Micro graph of LM5 aluminium alloy Fig 2B shows the SEM image of LM5 Fig 2C and 2D shows the SEM of selected area and EDAX respectively. Primary aluminum grain interdendritic pattern is visible in the microstructure. MgAl eutectic particles that were not dissolved during solidification precipitate near the grain PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 5 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 1. Stir casting setup. https://doi.org/10.1371/journal.pone.0308203.g001 Fig 2. LM5 Aluminium alloy a) Optical Micro graphs b) SEM c) SEM of selected area d) EDAX. https://doi.org/10.1371/journal.pone.0308203.g002 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 6 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy boundaries [31]. The grain size of the primary aluminium phase is 40–50 microns. The magni- fication is 100x. Scanning electron microscopy (SEM) was utilized to collect data from a mix of backscattered electron (BSE) and secondary electron (SE) signals. The BSE signal emphasizes compositional difference, whereas the SE signal depicts sample topography, including cracks and voids. Thermomechanical processing and alloying influence the development of iron-rich intermetallic particles and can be tuned to reduce the likelihood of micro-cracking [32]. Figure depicts micro-cracks in an iron-rich intermetallic particle. These endogenous micro cracks can enlarge as the material is bent into its final shape, resulting in bigger material frac- tures. The scale bar is 100 μm. EDAX test results confirm that the elements like aluminium, iron, copper, magnesium, zinc, silicon, carbon, oxygen, lead and other elements are present in the aluminium alloy as shown in all the specimens, aluminium shows the peak value followed by magnesium as LM5 is magnesium based alloy. The results confirm the presence of Al (high-intensity peaks). Experimental analysis on machinability of LM5 Al alloy by WEDM Aluminium alloy was used to prepare the test samples. The experimentation was done using the ECOCUT WEDM, which let users to select input settings based on the sample’s material and thickness. The manufacturer’s instructions should be used to select the tool material. Brass wire of 0.25 mm in diameter was used in this instance. The dielectric medium utilized was demineralized water. The worktable, the servo regulating system, the power supply, the dielec- tric supply system, and the wire are the fundamental components of the WEDM machine. Using common grips, the material samples were secured to the machine’s worktable. WEDM removes material using a sequence of recurrent spark discharges across the tool (wire electrode) and work piece, which are submerged in a liquid dielectric and isolated by a distance known as the spark gap. Whenever a suitable voltage is given during pulse-on time, the dielectric breaks down, causing an electrical spark to form between the tool and workpiece. Thermal conduction converts electrical energy into heat energy by means of the creation of a discharge column. The tool and workpiece begin to melt as a result of high-energy plasma pro- duction. While the discharge begins, the tool, work piece, and dielectric begin to vaporize, resulting in the development of a compressed vapor bubble which increases till the pulse-on time. At the start of the pulse-off period, the discharge stops, resulting in a dramatic implosion of the plasma channel and squeezed vapor bubble, allowing the superheated and molten liquid to explode into the dielectric. The ejected materials re-solidify into small spheres, which are washed away by the dielectric. This caused the creation of a small cavity or crater on the work piece surface. With each discharge, the needed amount of material is removed from the work piece surface [33]. The WEDM process is chosen for studying the machinability of the LM5 aluminium alloy. The photograph of CNC WEDM is presented in Fig 3. Design of Experiments (DoE) The Design of Experiments (DoE) technique is used to specify what information, in what amount, and under what conditions needs to be gathered during an experiment in order to meet two primary objectives: a lower cost and greater statistical precision for the response parameters. Four distinct process parameters at three levels were chosen for the current inves- tigation: wire feed (WF), gap voltage (GV), pulse on time (T ), and pulse off time (T ). Mate- on off rial removal rate (MRR), surface roughness (SR), and cutting width (kerf) are the responses. For this research, the L orthogonal array is chosen based on the parameters that were chosen. For every experimental condition, three repetitions of the experiment have been conducted. Table 2 shows the machining variables together with their respective levels. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 7 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 3. Photograph of ECOCUT–CNC wire EDM. https://doi.org/10.1371/journal.pone.0308203.g003 Machining performance variables (Responses) The basic objective is to establish the machining parameters to achieve the maximum MRR, the smallest possible K , and the lowest possible SR. Fig 4 shows the wire Electrical Discharge machined samples. Material Removal Rate (MRR). The material removal rate (MRR) is calculated as the vol- ume of material eliminated from the specimen (mm ) divided by the time taken (min). Surface roughness. According to Kousik-Kumaar et al. [34] surface roughness assessment is crucial for a number of basic problems, such as friction, surface deformation, transfer of heat, current flow, stiffness of joints, and spatial precision. Three distinct sites on the surface that was machined were used to get the data, and the average of the three readings was used to Table 2. Process parameters and levels. Level Pulse on Time (μs) T Pulse off Time (μs) T Gap Voltage (V) GV Wire Feed (m/min) WF on off 1 110 30 20 3 2 115 40 30 6 3 120 50 40 9 https://doi.org/10.1371/journal.pone.0308203.t002 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 8 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 4. Machined specimens. https://doi.org/10.1371/journal.pone.0308203.g004 determine the SR. The orientation of an SR computation was orthogonal to the surface being machined. The Surfcorder SE 3500 surface roughness tester is displayed in Fig 5. Ra (roughness average) is a commonly used metric for quantifying a material’s surface roughness. It is the arithmetic average of the absolute values of profile height deviations from the centerline measured within a given evaluation length. The work piece was positioned verti- cally, with the WEDMed specimens’ axes oriented horizontally. The roughness tester was Fig 5. Surface roughness tester. https://doi.org/10.1371/journal.pone.0308203.g005 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 9 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 6. Output of surface roughness measurement. https://doi.org/10.1371/journal.pone.0308203.g006 placed in front of each machined surface, allowing the stylus arm with the probe tip to be placed on the WEDMed surface. Fig 6 depicts the output shown during the measurement of surface roughness. Cutting width (kerf). The width of material which is machined by a cutting procedure is referred to as Cutting Width (Kerf). When it comes to CNC Wire EDM form cutting with tra- ditional cutting techniques, kerf refers to the amount of material removed as the process cuts through the plate. It is measured by the help of Vision Measuring Machine. The most crucial sign of WEDM technology success is the width of the kerf in conjunction with the rate of material removal. Results and discussion To ascertain the impact of machining input parameters on performance measures, experi- ments on wire EDM were conducted based on the Taguchi DOE approach and analyzed using S/N ratio analysis [35]. The results and pertinent factors are presented. This section converses the experimental findings related to the Wire EDM of the LM5 Aluminium alloy. The analysis and discussion focus on MRR, SR and k . The ideal machining parameters were established using Taguchi’s S/N analysis and confirmation experiments to validate the results. Experimental results Wire EDM experiments were conducted using Taguchi’s DoE and analyzed by S/N analysis. Table 3 displays the Wire EDM experimental results of the effect of process variables such as T , T , GV, WFon the responses like MRR, SR, and k . on off w Analysis and discussion of results of MRR S/N ratio is a metric for evaluating quality attributes. Smaller is better, nominal the best, and larger is better are three elements of performance qualities that Taguchi identified. The bigger amount of MRR is found to be more advantageous for surplus product quality, hence the "larger is better" class is chosen for S/N calculation. The ideal machining variables are esti- mated at the level where each variable has the highest S/N value. This section discusses the PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 10 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Table 3. WEDM experimental results of LM5 aluminium alloy. Ex A B C D MRR (mm / S/N of Surface Roughness S/N of Kerf S/N No min) MRR (μm) SR (mm) Kerf Pulse on Time Pulse off Time Gap Voltage Wire Feed (m/ (μs) (μs) (V) min) 1 110 30 20 3 3.61 11.16 3.10 -9.8 0.286 10.87 2 110 40 30 6 2.98 9.50 3.27 -10.3 0.286 10.87 3 110 50 40 9 2.26 7.07 2.76 -8.8 0.286 10.87 4 115 30 30 9 4.50 13.07 3.63 -11.2 0.304 10.34 5 115 40 40 3 3.40 10.64 3.45 -10.8 0.295 10.60 6 115 50 20 6 5.52 14.84 3.66 -11.3 0.313 10.09 7 120 30 40 6 4.56 13.18 3.83 -11.7 0.304 10.34 8 120 40 20 9 7.09 17.01 3.63 -11.2 0.313 10.09 9 120 50 30 3 5.78 15.24 3.96 -12.0 0.313 10.09 https://doi.org/10.1371/journal.pone.0308203.t003 implications of the Wire EDM process settings on MRR. The S/N ratio of response characteristics for each variable at different phases is quantified using experimental results. The parametric influ- ences on response characteristics were examined using the main effects plot (response graphs). Analysis of variance (ANOVA) was applied to S/N data in order to categorize pertinent variables and assess the effects of those variables on response characteristics. The optimum process variables are obtained by analyzing the response graphs and ANOVA table. As shown in Fig 7, the MRR rises with increasing T and decreases with rising T and GV. This occurs as a result of the dis- on off charge energy increases brought on by the increase in T , which leads to a greater MRR. As the on T decreases, more discharges occur in a given period of time, increasing the MRR. Lower MRR off is caused by the normal discharge gap widening as the gap voltage increases [36]. Selection of optimal levels for MRR The delta values and ranks indicate that T has the greatest effect on MRR, followed by GV. on Fig 7 and Table 4 demonstrates that the third level of T , first level of T , first level of GV on off Fig 7. Response graphs for MRR. https://doi.org/10.1371/journal.pone.0308203.g007 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 11 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Table 4. Response table for MRR. Level Pulse on Time Pulse off Time Gap Voltage Wire Feed 1 9.24 12.47 14.34 12.34 2 12.85 12.38 12.60 12.51 3 15.14 12.39 10.30 12.39 Delta 5.90 0.09 4.04 0.16 Rank 1 4 2 3 https://doi.org/10.1371/journal.pone.0308203.t004 and second level of WF produce the maximum MRR. R value is 99.93%, which is most desir- able. The p-value for T and GV is below 0.05, demonstrating the significance of the factors, on but the p-value for T and WF is greater than 0.05, demonstrating the meager impact of these off parameters on MRR. According to the F-test, the effect is determined to be considerable if the computed value of the F-ratio is higher than the tabulated F-value. The table’s F-value at the 5% level of signifi- cance is F = 6.944. So, T and GV are important process parameters for obtaining (0.05, 2,4) on larger MRR as seen in ANOVA Table 5. T and WF are two variables that have very minimal off contribution on the MRR, so they are pooled up with error. Analysis and discussion of results of SR The "smaller is better" class is used for S/N analysis since it is found that a lesser amount of SR is better for excellent product quality. As shown in Fig 8, the SR increases with the increase in T and decreases with the increase in GV, T and WF. The cause is the discharge energy on off changes with T and that a higher T produces a larger crater, which raises the SR on the on on work piece. As T rises, the number of discharges decreases, leading to greater surface preci- off sion from steady machining. As the GV increases, the average discharge gap widens, improv- ing SR. T has no discernible impact [37]. Following wire EDM, the surface exhibits an off irregular mix of overlapping craters, micro-globules, and melted debris. During the WEMD process, the generated heat ranges between 8000 and 12,000˚C which produce local melting and evaporation of the work piece material. The heat produces a high level pressure, but it is insufficient for removing all of the molten material. The balance of the molten material re- solidifies on the surface of the machined sample, resulting in an undulating topography. A larger amount of molten material re-solidifies on the machined surface, resulting in a thicker recast layer and a higher Ra. Selection of optimal levels for SR For the researcher to select the ideal parameters, the experiment results are assessed. Since SR is an output characteristic where "lower is better" applies, Fig 8 and Table 6 demonstrates that the lowest SR values are found in the T first level, third level of T , third level of WF, and on off third level of GV. The process factors significance was examined using an ANOVA. R for SR Table 5. ANOVA for MRR. Source of Variation DF SS MS F P C % Pulse on Time 2 53.07 26.53 1836.77 0 68.25 Gap Voltage 2 24.63 12.32 852.57 0 31.68 Pooled Error 4 0.06 0.01 0.07 Total 8 77.76 100 https://doi.org/10.1371/journal.pone.0308203.t005 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 12 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 8. Response graphs for SR. https://doi.org/10.1371/journal.pone.0308203.g008 is 99.09%, which is a desirable value. The p-value for T is less than 0.05, indicating that the on factor is significant, whereas the p-values for T and WF are greater than 0.05, indicating that off there is no noteworthy effect on SR. The F-test shows Table’s F-value at the 5% level of signifi- cance is F = 19. So, T is the most important parameter, as shown in Table 7. Even (0.05, 2, 2) on though GV and WF are included, there is no visible impact on SR. The T is pooled up with off error. Analysis and discussion of results of kerf Lesser cutting width (Kerf) is found to be more advantageous for better product quality, the "smaller is better" class is selected for S/N calculation. Fig 9 illustrates how the T , T , and on off WF drop as the K increases. It gets smaller as the GV gets bigger. The cause is that the dis- charge energy changes with T , and higher discharge energies produce considerably larger on craters, which raise the K on the work piece. As the GV increases, the average discharge gap widens, resulting in reduced K . Lower K is also provided by the lower WF [38]. w w Selection of optimal levels for kerf For the purpose of selecting the ideal parameters, the experiment results are assessed. In the Wire EDM phase, Fig 9 and Table 8 demonstrates that the first level of T , second level of on T , third level of GV, and first level of WF have the smallest K . R coefficient is 98.44%, off w which is an excellent value. The p-value for T , WF, and GV is greater than 0.05, indicating off that there was no significant effect on K , whereas the p-value for Pulse on Time is less than Table 6. Response table for SR. Level Pulse on Time Pulse off Time Gap Voltage Wire Feed 1 -9.65 -10.89 -10.76 -10.85 2 -11.07 -10.74 -11.14 -11.07 3 -11.60 -10.68 -10.41 -10.40 Delta 1.95 0.21 0.74 0.67 Rank 1 4 2 3 https://doi.org/10.1371/journal.pone.0308203.t006 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 13 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Table 7. ANOVA for SR. Source of Variation DF SS MS F P C % Pulse on Time 2 6.11 3.06 86.87 0.01 79.46 Gap Voltage 2 0.81 0.41 11.51 0.08 10.53 Wire Feed 2 0.70 0.35 9.95 0.09 9.10 Pooled Error 2 0.07 0.04 0.91 Total 8 7.69 100.00 https://doi.org/10.1371/journal.pone.0308203.t007 Fig 9. Response graphs for kerf. https://doi.org/10.1371/journal.pone.0308203.g009 Table 8. Response table for kerf. Level Pulse on Time Pulse off Time Gap Voltage Wire Feed 1 10.87 10.52 10.35 10.52 2 10.35 10.52 10.43 10.43 3 10.17 10.35 10.61 10.43 Delta 0.7 0.17 0.26 0.09 Rank 1 3 2 4 https://doi.org/10.1371/journal.pone.0308203.t008 0.05, demonstrating the significance of the factor. The Fisher’s F-test shows at a 95% confi- dence level, the F-table value is F = 19.37. Therefore, the most important parameter, as (0.05, 2, 8) shown in ANOVA Table 9, is T , even if GV and T are additionally accounted for in the on off contribution but have no noticeable impact on K . The WF is a pooled up with the error. Confirmation experiments The confirmation experiments, according to Taguchi, is an essential stage in validating the experimental results. Based on the ideal confluence of variables impacting MRR, SR and Kerf, the confirmation experiments were effectively carried out. The tests were run three times to obtain an average value, and the correlation between the actual value and the predicted value PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 14 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Table 9. ANOVA for kerf. Sources of variation DOF SS MS F P C % Pulse on Time 2 0.80 0.40 52.6 0.02 81.97 Pulse off Time 2 0.06 0.03 3.83 0.21 5.96 Gap Voltage 2 0.10 0.05 6.74 0.13 10.50 Pooled Error 2 0.02 0.01 1.56 Total 8 0.97 100 https://doi.org/10.1371/journal.pone.0308203.t009 was then performed. Table 10 is a summary of the outcomes of the confirmation tests. The best parameters are used to predict the MRR, SR, and K , in the confirmation trials. The experiments’ results are assessed to determine the most important parameters. The best set- tings for increasing MRR are those at level A , B , C , and D (T 120 μs, T 30 μs, GV 20 V, 3 1 1 2 on off and WF 6 m/min). While the MRR predicted value is 6.56 mm /min, the MRR experimental result is 6.73mm /min. The predicted and experimental MRR values exhibit remarkable agree- ment, and the error is 2.53 percent. In order to achieve the lowest SR, the optimum process variables are T 110 μs, T 50 μs, GV 40 V, and WF 9 m/min. The experimental SR is on off 2.76μm, whereas the predicted SR is 2.76 μm. The predicted and experimental SR values exhibit a high degree of agreement. The variables at levels A , B , C , and D are T 110 μs, 1 2 3 1 on T 40 μs, GV 40 V, and WF 3 m/min are the optimum machining variables in order to off achieve the lowest Kerf. The experimental K is 0.285 mm and the predicted K is 0.270 mm. w w The predicted and experimental K values are in excellent agreement, and the error is only 2.81%. SEM analysis of WEDMed surfaces In order to assess the surface integrity, the SEM study on the topography of the WEDM machined LM5 aluminium alloy surfaces was conducted using the optimum surface roughness process variables: T 110 μs, T 50 μs, GV 40 V, and WF 9 m/min. on off The SEM micrograph of WEDM machined surface at machining parameters A B C D in 1 3 3 3 the orthogonal array of experiment is shown in Fig 10A, it is clear that the size of the crater depend on the discharge heat energy or in other words, on the gap voltage and pulse on time values. Higher Gap voltage causes an increase in discharge heat energy at the point where the discharge takes place. At this point, a pool of molten metal is formed and is overheated. The overheated molten metal evaporates forming gas bubbles that explode when the discharge ceases, taking molten material away. The result is the formation of crater. Successive dis- charges that have a random nature will result in the formation of globules of debris, shallow craters, pockmarks and cracks [39, 40]. The surface morphology of WEDMed LM5 alloy was studied by field emission scanning electron microscopy (FESEM). To determine the elemental composition, an EDX analysis was performed. Fig 10B shows the EDX image of machined LM5 Al alloy and the atomic weight. Table 10. Results of confirmation experiments. Response Optimum levels Experimental value Average Experimental Value Predicted Value Error % Trial 1 Trial 2 Trial 3 MRR (mm /min) A B C D 6.63 6.87 6.69 6.73 6.56 2.53 3 1 1 2 Surface Roughness (μm) A B C D 2.76 2.77 2.75 2.76 2.76 0 1 3 3 3 Kerf Width (mm) A B C D 0.278 0.290 0.287 0.285 0.277 2.81 1 2 3 1 https://doi.org/10.1371/journal.pone.0308203.t010 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 15 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 10. WEDM–Machined Surface a) SEM image b) EDAX c) EDX mapping d) Al e)Mg f) O g) C. https://doi.org/10.1371/journal.pone.0308203.g010 The formation of carbon is represented by distinct peaks following aluminium. The XRD anal- ysis revealed the presence of all the alloying elements, which was further confirmed by the EDX analysis. The EDAX of machined surface (recast layer) shows oxygen, carbon, copper and zinc. These are formed due to the dielectric fluid and tool material (brass wire). As shown in Fig 10C–10G, EDX mapping shows the quantity of each element inside the reaction product. The fundamental concept of the WEDM technique is the generation of electric sparks between the work piece and the wire electrode. These electrical discharges release a large quan- tity of heat at temperatures ranging from 8000 to 12,000˚C, resulting in melting and evapora- tion of work piece material at the nearby surface layers. The heat also melts the dielectric medium (de-ionized water) and creates high-pressure waves that wash away the melted and/or evaporated metal from the work piece. Throughout the WEDM process, dielectric fluid is con- tinuously supplied to transport the eroded metal apart. As a result of water’s strong thermal conductivity, the top surface cools and un-expelled material re-solidifies at a rapid pace. This re-solidified layer, known as a recast layer, is often highly fine-grained, hard, brittle, and struc- turally distinct from its parent material. The creation of these layers is determined by the pro- cess parameters as well as the work piece’s chemical composition and heat conductivity. The recast layer as shown in Fig 11 is generated at a slower cooling rate from the outermost layer, allowing the melted material to re-solidify fast and without grain boundaries. The heat- affected zone appears somewhat distinct in colour since it does not melt but is heated through- out the machining process. Zhang et al. [41] investigated different types of di-electrics and determined the recast layer created by water-in-oil emulsion dielectric, which has larger sur- face roughness and thickness than the kerosene and de-ionized water dielectrics. The results have revealed that the thickness of the recast layer increases with increasing peak current and decreases by using de-ionized water as the dielectric. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 16 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 11. Recast layer. https://doi.org/10.1371/journal.pone.0308203.g011 XRD analysis of WEDMed surface It is observed that the peaks corresponding to the sample LM5 alloy have good agreement with the JCPDS card 04–0787. So, it is concluded that the crystal structure and elemental composi- tion of the LM5 alloy have not disturbed even after the WEDM machining process. The aver- age crystallite size is 27 nm and the phase is Fm-3m [2 2 5]. Further, it is observed that the lattice parameters are equal which reflects the cubic structure and the primitives are a = b = c = 4.054A. The XRD image of Wire Electro Discharge machined specimen is shown in Fig 12. Effect of process parameters on responses (MRR, SR, and K ) Effect of T on responses. Fig 13 shows the effect of T on responses. Longer T emits on on on higher discharge energy, which causes stronger explosions, deeper craters on the surface of the work piece, and higher MRR. Deep craters suggest high rates of MRR and subpar surface qual- ity. Larger values of T should be used to get better MRR. The T that produces the highest on on MRR is 120 μs [42]. As gap voltage (GV) decreases and pulse-on time (T ) increases, surface on roughness increases. The size and shape of surface craters, which are influenced by discharge energy and the re-deposition of melted material on the work surface, are the main factors that define SR in WEDM. Surface roughness increases when the servo voltage is decreased and the pulse-on-time is increased because this increases the discharge energy across the electrodes and creates a deep erosion crater on the work piece’s surface. There is a significant chance that molten material will re-deposit on the work surface at high discharge energies [43]. According to the findings, raising the T causes a greater thermal energy transfer from the on wire to the work piece, which increases cutting velocity. As the T falls, SR lowers. SR has an on impact on the wire electrical discharge machining finish cut. Experiments have shown that lowering the T and the discharge current together can reduce SR. A large number of perti- on nent investigations found that as discharge energy increased the Wire EDMed surface rough- ness because of more craters were created, which led to higher SR values on the work piece. Increasing the pulse on time, the single pulse discharge electric energy increases thickness of cutting surface discharge, however the electrical erosion products’ duration of discharge PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 17 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 12. XRD of WEDMed aluminium alloy LM5. https://doi.org/10.1371/journal.pone.0308203.g012 reduces proportionately, leading to burns on the cutting surface and the generation of adhesive substance there. These factors collectively influence the increase in surface roughness [44]. Kerf reduces when T declines. K can be used to gauge the material’s dimensional correct- on w ness. Experiments show that lowering the discharge current and pulse duration can lower Kerf. Furthermore, since K is based on the size of the spark crater, related research discovered that T is the main variable affecting K . The discharge energy must be kept at a low level by on w employing short pulse duration in order to produce flat craters [45]. Effect of T on responses. Fig 14 shows the effect of T on responses. Results demon- off off strate that MRR declines as T rises. Due to longer non-cutting times, greater T causes a off off drop in MRR. A wider gap is produced by a longer T , but it also offers a longer flushing time off to remove the debris from the gap. Usually an extended T was used to stop wire rupturing or off to stop the abnormal process. It may be concluded that higher T causes lower SR since the off non-cutting time increases [46]. Increasing the T value extends the duration between 2 suc- off cessive sparks, resulting complete flushing of carbide debris out of the spark gap, low re-depo- sition of degraded material, and low SR. Surface roughness has a slight tendency to decrease with increased WF. Increasing wire feed allows carbide debris to easily escape from the spark gap, resulting in re-cast layer. To get an excellent surface finish, keep the electrical discharge energy to a minimum by choosing minimal T and an excessive T [47]. The quantity of sin- on off gle pulse discharge energy is unaffected by an increase in pulse interval; rather, it only affects the length of discharge time per unit of time. Consequently, the discharge duration per unit time doubles and the cutting speed decreases linearly as the pulse interval grows. It has been found that raising the T causes the MRR to decrease. This action enhances the procedure by off PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 18 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 13. Effect of Pulse on Time on responses. https://doi.org/10.1371/journal.pone.0308203.g013 enabling a more effective flush of debris into the gap. 50 μs is the ideal pulse off time for lower- ing SR and K [48]. Effect of GV on responses. Fig 15 shows the effect of GV on responses. The findings indi- cate that the MRR rises as the GV falls. 20 V is the ideal gap voltage for increasing MRR. It’s important to note that 20 V is the base alloy’s ideal voltage. Due to a greater electric field, spark discharge actually occurs beneath the same gap more frequently when voltage increases. Less voltage can provide enough energy to melt the dielectric particles in the vicinity. The SR reduces as the GV rises. This occurrence is remarkable. Lower GV may have sufficient energy to melt the re-solidified particles in the vicinity, which remain on the machined surface and produce a large number of projecting peaks. On the other side, a smoother surface is produced by high voltage [49]. Melting of the component’s surface is determined by the thermal conduc- tivity of the work piece material and the quantity of energy used per spark that is assumed to be proportional to T and GV. Enhancing the pulse on time (T ) generates more heat at the on on work surface, improving the cutting speed. Lowering Gap voltage decreases the spark gap, resulting in fast and significant ionization of the dielectric fluid, causing greater melting of the work material and hence increases cutting speed [50]. Increasing electrostatic force brought on the increasing GV causes wire wrapping during the discharge process. The SR lowers as the gap voltage rises. 40 V is the ideal GV for getting lower K . Sparks that form at the conducting phase and produce melting or possible evaporation also contribute to the craters on the machined surface. It goes without saying that huge K are caused by high crater diameters [51, 52]. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 19 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 14. Effect of Pulse off Time on responses. https://doi.org/10.1371/journal.pone.0308203.g014 Effect of WF on responses. Fig 16 shows the effect of WF on responses. The WF ought to be selected in a way that prevents wire breaking. As the wire feed accelerates, the MRR ascend. For maximizing MRR, a wire feed setting of 6 m/min is ideal. In a word, this investigation’s findings supported those in the literature [53]. At high wire feed values, the increase in cutting speed is particularly noticeable. Higher discharge energy causes increased melting and evapo- ration of the work material, resulting in the release of a significant quantity of carbide debris, which coagulates in the spark gap and so influences the machining process through the genera- tion of arcs. Raising the wire feed rate allows the eroded material to escape the spark gap more easily and quickly [54]. Even though the research was done on a variety of materials, the out- comes were reliable. It is important to choose a WF that wire won’t break. As the WF grows, so does the cutting speed. For a minimum SR, a wire feed setting of 6 m/min is ideal. Briefly stated, the findings of this investigation were consistent with those discovered in the litera- tures. The investigation was conducted on a variety of materials, but the outcomes were con- stant. A non-critical metric is wire feed. The ideal WF for the lowest K is 3 m/min. Increasing the WF rate might make the wire less rigid during discharge, which would reduce the amount of wire that would wind back during sparks and reduce the destructive power of the sparks on the work piece surface, leading to a higher K [55, 56]. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 20 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 15. Effect of Gap Voltage on responses. https://doi.org/10.1371/journal.pone.0308203.g015 Mathematical models of LM5 aluminium alloy Multiple linear regression (MLR) models frequently provide a suitable illustration of a more complex structure within particular ranges of the independent variables. The generalization model of MLR analysis describes the relationship among the response and the independent variables, and the estimated response is calculated using the generalized simple regression equation. The constants were calculated using the linear regression analysis method using MINITAB software. Mathematical models for MRR, SR and K for LM5 alloy are developed using Linear Regression are shown in Eqs 1–3. MRR ¼ 26:38þ 0:2857 T þ 0:01473 T 0:10002 GVþ 0:0585 WF ð1Þ on off SR ¼ 4:81þ 0:0760 T 0:00292 T 0:00575 GV 0:0278 WF ð2Þ on off Kerf ¼ 0:0225þ 0:002400 T þ 0:000300 T 0:000450 GVþ 0:000500 WF ð3Þ on off 2 2 From the ANOVA of the Regression, the R value of MRR is 99.7, the R value of SR is 82.6 and the R value of Kerf is 91.4. The aforementioned mathematical model for MRR, SR and kerf is critical for selecting machining settings while machining LM5 Aluminium alloy using WEDM. According to Eqs 1 and 3, MRR and Kerf are directly correlated to T , T and WF, but inversely correlated to on off gap voltage. Eq 2 demonstrates that SR is directly proportional to T and inversely propor- on tional to T , GV and WF. off PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 21 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Fig 16. Effect of Gap Voltage on responses. https://doi.org/10.1371/journal.pone.0308203.g016 From Eq (1), it is seen that T , T and WF are directly proportional and GV is inversely on off proportional to MRR. The percentage deviation of MRR values between the experimental and predicted values are reported in Table 11. Referring to Table 11 on MRR model, it can be seen lesser deviation occurs between experimental and predicted values. The average absolute error Table 11. Experimental and predicted values of MRR, SR and kerf. Ex A B C D Experimental Predicted % Experimental Predicted % Experimental Predicted % No T T GV WF (m/ MRR (mm3/ MRR Deviation SR SR Deviation kerf kerf Deviation on off (μs) (μs) (V) min) min) 1 110 30 20 3 3.61 3.66 -1.5 3.1 3.26 -5.29 0.286 0.288 -0.7 2 110 40 30 6 2.98 2.99 -0.22 3.27 3.09 5.39 0.286 0.288 -0.7 3 110 50 40 9 2.26 2.31 -2.18 2.76 2.92 -5.93 0.286 0.288 -0.7 4 115 30 30 9 4.5 4.44 1.26 3.63 3.42 5.79 0.304 0.299 1.81 5 115 40 40 3 3.4 3.24 4.72 3.45 3.5 -1.44 0.295 0.294 0.34 6 115 50 20 6 5.52 5.56 -0.77 3.66 3.5 4.31 0.313 0.308 1.76 7 120 30 40 6 4.56 4.7 -2.98 3.83 3.83 0.11 0.304 0.305 -0.16 8 120 40 20 9 7.09 7.02 1 3.63 3.83 -5.45 0.313 0.318 -1.6 9 120 50 30 3 5.78 5.82 -0.61 3.96 3.91 1.31 0.313 0.314 -0.16 https://doi.org/10.1371/journal.pone.0308203.t011 PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 22 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy for MRR is 1.69%. These deviations could be attributed to limitation of the modeling not accounting for interactive influences [57, 58]. From Eq (2), it is seen that GV, T and WF are directly proportional and T is inversely off on proportional to SR. The percentage deviations of SR values between the experimental and pre- dicted values are reported in Table 11. Referring to Table 11 on SR model, it can be seen lower order deviation between experimental and predicted values occurs. The average absolute error for SR is 3.89%. These deviations could be attributed to limitation of the modeling not accounting for interactive influences. From Eq (3), it is seen that T , T and WF are directly proportional and GV is inversely on off proportional to kerf. The percentage deviation of kerf values between the experimental and predicted values are reported in Table 11. Referring to Table 11 on kerf model, it can be seen minimum deviation between experimental and predicted values occurs. The average absolute error for kerf is 0.88%. Conclusions The LM5 aluminium alloy was successfully produced via the stir casting technique. The subsequent outcomes were attained i. The microscopic and SEM images show the microstructure of the LM5 Al alloy and EDX shows the constituents. ii. The Pulse on Time (68.25%) and Gap Voltage (31.68%) has the greatest statistical effects on MRR. The error associated with the ANOVA is 0.07%, indicating a 95% confidence level. iii. The most significant statistical impact on SR is possessed by the Pulse on Time (79.46%). The ANOVA error for SR is 0.91%, indicating a 95% confidence level. iv. The greatest statistical influence on K is by the Pulse on Time (81.97%). The 95% confi- dence level is shown by the error associated with the ANOVA for K , is 1.56%. v. The error % associated with the predicted and experimented values of the MRR is 2.53%, SR is 0 and K is 2.81%, based on the confirmation experiments. vi. The EDS of the machined surface (recast layer) shows oxygen, carbon, copper and zinc. These are formed due to the dielectric fluid and tool material (brass wire). vii. The average absolute error for MRR is 1.69%, for SR is 3.89% and for kerf is 0.88%, based on mathematical (linear regression) models. viii. The predicted and experimental values based on the mathematical models are very close to each other confirms that the Taguchi’s Signal to Noise ratio analysis is the most appro- priate for single objective optimization of responses. The findings of this investigation promises to enhance precision and surface quality in wire electro-discharge machining of LM5 aluminium alloy in today’s production processes. Author Contributions Conceptualization: Sunder Jebarose Juliyana, Charles Sarla Rubi. Formal analysis: Sunder Jebarose Juliyana, Charles Sarla Rubi, Arasumugam Divya Sadhana. Funding acquisition: Emad Abouel Nasr. Investigation: Sunder Jebarose Juliyana. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 23 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy Methodology: Sunder Jebarose Juliyana, Charles Sarla Rubi, Arasumugam Divya Sadhana. Project administration: Charles Sarla Rubi, Arasumugam Divya Sadhana. Resources: Robert Čep, Charles Sarla Rubi, Sachin Salunkhe, Arasumugam Divya Sadhana. Software: Sunder Jebarose Juliyana, Charles Sarla Rubi. Supervision: Jayavelu Udaya Prakash, Sachin Salunkhe. Validation: Jayavelu Udaya Prakash, Robert Čep, Sachin Salunkhe, Emad Abouel Nasr. Visualization: Emad Abouel Nasr. Writing – original draft: Jayavelu Udaya Prakash. Writing – review & editing: Robert Čep. References 1. Wang L., Makhlouf M. and Apelian D., 1995. Aluminium die casting alloys: alloy composition, micro- structure, and properties-performance relationships. International materials reviews, 40(6): 221–238. 2. Rajan R., Kah P., Mvola B. and Martikainen J., 2016. Trends in aluminium alloy development and their joining methods. Reviews on Advanced Materials Science, 44(4) ´ ´ 3. Liptakova T., Zatkalıkova V., Alaskari A.M. and Malcho M., 2019. Dominant factors affecting erosion- corrosion resistance of aluminium–brass pipelines. Journal of Engineering Research, 7(4). 4. Boggarapu V., Sreekanth P.R. and Peddakondigalla V.B., 2023. Microstructure, Mechanical and Tribo- logical Properties of Al/Cu Functionally Graded Material Fabricated through Powder Metallurgy. Journal of Engineering Research, p.100119. 5. Ananth S., Prakash J.U., Juliyana S.J., Rubi C.S. and Sadhana A.D., 2021. Effect of process parame- ters on WEDM of Al–Fly ash composites using Taguchi Technique. Materials Today: Proceedings, 39, pp.1786–1790. 6. Ho K.H., Newman S.T., Rahimifard S. and Allen R.D., 2004. State of the art in wire electrical discharge machining (WEDM). International Journal of Machine Tools and Manufacture, 44(12–13), pp.1247– 7. Saravanan R., Anbuchezhiyan G., Mamidi V.K. and Kumaran P., 2022. Optimizing WEDM parameters on nano-SiC-Gr reinforced aluminium composites using RSM. Advances in Materials Science and Engi- neering, 2022. https://doi.org/10.1155/2022/1612539 8. Natarajan K., Ramakrishnan H., Gacem A., Vijayan V., Karthiga K., Ali H.E., et al. 2022. Study on opti- mization of WEDM process parameters on stainless steel. Journal of Nanomaterials, 2022. 9. Rouniyar A.K. and Shandilya P., 2021. Semi-empirical modeling and optimization of process parame- ters on overcut during MFAPM-EDM of Al6061 alloy. Proceedings of the Institution of Mechanical Engi- neers, Part E: Journal of Process Mechanical Engineering, 235(6), pp.1784–1796. https://doi.org/10. 1177/09544089211015890 10. Kar S. and Sahoo R., 2023. Viability and performance assessment of bipolar linear self-servo mecha- nism-based Maglev electrical discharge machining by processing Al-6062 alloy. Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering, p.09544089231207685. 11. Mahapatra S.S. and Patnaik A., 2006. Parametric optimization of wire electrical discharge machining (WEDM) process using Taguchi method. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 28, pp.422–429. 12. Verma R.K., Patel D. and Chopkar M.K., 2023. Wear behavior investigation of Al-B C functionally graded composite through Taguchi’s design of experiment. Journal of Engineering Research, p.100095. 13. Sarkar S., Sekh M., Mitra S. and Bhattacharyya B., 2011.A novel method of determination of wire lag for enhanced profile accuracy in WEDM. Precision Engineering, 35(2), pp.339–347. https://doi.org/10. 1016/j.precisioneng.2011.01.001 14. Arunnath A., Madhu S. and Tufa M., 2022. Experimental investigation and optimization of material removal rate and tool wear in the machining of aluminium-boron carbide (Al-B4C) nanocomposite using EDM process. Advances in Materials Science and Engineering, 2022. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 24 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy 15. Salavaravu L. and Dumpala L., 2022. Effects of process parameters on mechanical and metallurgical properties of AA5083 weld bead and optimization by using Taguchi based Grey Relational Analysis and ANOVA of submerged friction stir welding. Journal of Engineering Research (2307–1877), 10. 16. Zafar M.M. and Toor Z.S., 2023. Experimental and numerical investigation of electrochemical and tribo- logical behavior of aluminium alloys. Journal of Engineering Research, p.100134. 17. Kumar N., Kumari S., Abhishek K., Nandi G. and Ghosh N., 2022. Study on various parameters of WEDM using different optimization techniques: A review. Materials Today: Proceedings, 62, pp.4018– 18. Kansal H.K., Singh S. and Kumar P., 2007. Technology and research developments in powder mixed electric discharge machining (PMEDM). Journal of materials processing technology, 184(1–3), pp.32– 19. Debnath T. and Patowari P.K., 2019. Fabrication of an array of micro-fins using Wire-EDM and its parametric analysis. Materials and Manufacturing Processes, 34(5), pp.580–589. 20. Sarmah P. and Patowari P.K., 2023. Machinability study of Al 6063-based MMCs with SiC reinforce- ment particles using WEDM process. Materials and Manufacturing Processes, 38(7), pp.783–796. 21. Sahoo R., Kar S. and Patowari P.K., 2020. Experimental study on Titanium diamond by fabricating micro-holes using Micro-EDM with Taguchi’s method and response surface method. Engineering Research Express, 2(4), p.045026. 22. Sahoo R., Singh N.K. and Bajpai V., 2023. A novel approach for modeling MRR in EDM process using utilized discharge energy. Mechanical Systems and Signal Processing, 185, p.109811. 23. Sahoo R., Singh N.K. and Bajpai V., 2023. Approach towards green manufacturing in Maglev EDM using different biodegradable dielectrics at variable discharge conditions. Journal of Cleaner Produc- tion, 430, p.139623. 24. Ye L., Qian J., Haitjema H. and Reynaerts D., 2022. On-machine chromatic confocal measurement for micro-EDM drilling and milling. Precision Engineering, 76, pp.110–123. 25. Juliyana S.J., Prakash J.U., Salunkhe S., Hussein H.M.A. and Gawade S.R., 2022. Mechanical Charac- terization and Microstructural Analysis of Hybrid Composites (LM5/ZrO /Gr). Crystals, 12(9), p.1207. 26. Jebarose Juliyana S. and Prakash J.U., 2022. Optimization of machining parameters for wire EDM of AMCs (LM5/ZrO ) using Taguchi technique. INCAS Bulletin, 14(1), pp.57–68. 27. Jebarose Juliyana S., and Prakash J.U., 2020. Drilling parameter optimization of metal matrix compos- ites (LM5/ZrO2) using Taguchi Technique. Materials Today: Proceedings, 33, pp.3046–3050. 28. Jebarose Juliyana S., Udaya Prakash J.,Čep R. and Karthik K., 2023. Multi-Objective Optimization of Machining Parameters for Drilling LM5/ZrO2 Composites Using Grey Relational Analysis. Materials, 16 (10), p.3615. https://doi.org/10.3390/ma16103615 PMID: 37241242 29. Jebarose Juliyana S., Prakash J.U., Sadhana A.D. and Rubi C.S., 2022. Multi-objective optimization of process parameters of wire EDM for machining of AMCs (LM5/ZrO2) using grey relational analysis. Materials Today: Proceedings, 52, pp.1494–1498. 30. Juliyana S.J., Prakash J.U. and Salunkhe S., 2022.Optimisation of wire EDM process parameters using Taguchi technique for machining of hybrid composites. International Journal of Materials Engineering Innovation, 13(3), pp.257–271. 31. Udaya Prakash J; Jebarose Juliyana S.; Salunkhe S.; Gawade S.R.; Nasr E.S.A.; Kamrani A.K., 2023. Mechanical Characterization and Microstructural Analysis of Stir-Cast Aluminum Matrix Composites (LM5/ZrO2). Crystals, 13, 1220. 32. Wang B., Liu X., Wang J., Li Q., Liu K., & Zhang M., 2022. Uncovering the effects of Ce and superheat temperature on Fe-rich intermetallic and microporosity formation in aluminum alloy. Materials Charac- terization, 193, 112226. 33. Balan A.S.S. and Giridharan A., 2017. A progress review in wire electrical discharge machining pro- cess. International Journal of Automotive and Mechanical Engineering, 14, pp.4097–4124. 34. Kousik Kumaar R., Somasundara Vinoth K. and Kavitha M., 2022. Dry sliding wear performance of AA7075/MoS composite materials. Journal of Engineering Research (2307–1877), 10 35. Paulson D.M., Saif M. and Zishan M., 2023. Optimization of wire-EDM process of titanium alloy—Grade 5 using Taguchi’s method and grey relational analysis. Materials Today: Proceedings, 72, pp.144–153. 36. George J., Philip J.T., Mathew J. and Manu R., 2022. Prediction and analysis of material removal rate and white layer thickness during wire electrical discharge turning (WEDT) process. CIRP Journal of Manufacturing Science and Technology, 39, pp.210–222 37. Ram H.S., Uthayakumar M., Kumar S.S., Kumaran S.T. and Korniejenko K., 2022.Modelling Approach for the Prediction of Machinability in Al6061 Composites by Electrical Discharge Machining. Applied Sci- ences, 12(5), p.2673. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 25 / 26 PLOS ONE Taguchi optimization of Wire EDM process parameters for machining LM5 aluminium alloy 38. Sahu S.K., Anand M.V., Kumar T.C.A., Kumar A., Prasad G.S. and Niraj V.V., 2023. Kerf width analysis of wire electrical discharge machining of titanium alloy. Materials Today: Proceedings. 39. Kumar A., Kumar V. and Kumar J., 2016. Surface crack density and recast layer thickness analysis in WEDM process through response surface methodology. Machining Science and Technology, 20 (2), pp.201–230. 40. Azam M., Jahanzaib M., Abbasi J.A., Abbas M., Wasim A. and Hussain S., 2016. Parametric analysis of recast layer formation in wire-cut EDM of HSLA steel. The International Journal of Advanced Manufacturing Technology, 87, pp.713–722. 41. Zhang Y., Liu Y., Ji R. and Cai B., 2011. Study of the recast layer of a surface machined by sinking elec- trical discharge machining using water-in-oil emulsion as dielectric. Applied surface science, 257 (14), pp.5989–5997. 42. Chaudhari R., Shah H., Ayesta I., de Lacalle L.L. and Vora J., 2022. Experimental investigations and optimization of WEDM parameters using Taguchi analysis of pure titanium. In Recent Advances in Mechanical Infrastructure: Proceedings of ICRAM 2021 (pp. 349–358). Springer Singapore. 43. Sharmila B. and Selvakumar G., 2022. Investigations on the effect of dielectric medium and WEDM parameters on surface characteristics of Al 7068 (ordnance aluminium) alloy. Surface Topography: Metrology and Properties, 10(3), p.035031. 44. Han F., Jiang J. and Yu D., 2007. Influence of machining parameters on surface roughness in finish cut of WEDM. The International Journal of Advanced Manufacturing Technology, 34, pp.538–546. 45. Li W., Zhao L., Wang G. and Zhou C., 2023. Manufacturing of Mold Cavity of Micro-double Gear by WEDM. In Journal of Physics: Conference Series (Vol. 2459, No. 1, p. 012065). IOP Publishing. 46. Bisaria H. and Shandilya P., 2023. Surface integrity of NiTi-SMA during WEDM: effect of spark parame- ters. Materials and Manufacturing Processes, pp.1–9. 47. Sen R., Choudhuri B., Barma J.D. and Chakraborti P., 2020. Surface integrity study of WEDM with vari- ous wire electrodes: Experiments and analysis. Machining Science and Technology, 24(4), pp.569– 48. Sharma N., Gupta R.D., Khanna R., Sharma R.C. and Sharma Y.K., 2023. Machining of Ti-6Al-4V bio- medical alloy by WEDM: Investigation and optimization of MRR and Rz using grey-harmony search. World Journal of Engineering, 20(2), pp.221–234. 49. Chaubey S.K. and Gupta K., 2023. A review on Wire-EDM of bio titanium. Reports in Mechanical Engi- neering, 4(1), pp.141–152. 50. Sarala Rubi C., Udaya Prakash J., Jebarose Juliyana S., RobertČep, Sachin Salunkhe, Kouril K., et al. 2024. Comprehensive review on wire electrical discharge machining: a nontraditional material removal process. Frontiers in Mechanical Engineering, 10:1322605. 51. Raj A., Misra J.P., Khanduja D., Saxena K.K. and Malik V., 2022. Design, modeling and parametric opti- mization of WEDM of Inconel 690 using RSM-GRA approach. International Journal on Interactive Design and Manufacturing (IJIDeM), pp.1–11. 52. Babu R.R., Subhair S. and Siva S., 2018. Optimization of process parameter in WEDM for Stainless Steel 316 by Using Taguchi Method. International Journal of Engineering Research & Technology (IJERT). 53. Prakash J.U., Sivaprakasam P., Garip I., Juliyana S.J., Elias G., Kalusuraman G., et al. 2021. Wire elec- trical discharge machining (WEDM) of hybrid composites (Al-Si12/B4C/fly ash). Journal of Nanomater- ials, 2021, pp.1–10. 54. Kumar A., Kumar D.V. and Kumar D.J., 2011. A review on the state of the art in wire electric discharge machining (WEDM) process. International Journal of Mechanical Engineering Research and Develop- ment (IJMERD), 1(1). 55. Patra D.R., Rout I.S. and Sahoo M., 2015. Optimization of WEDM parameters using Taguchi method for higher material removal rate on EN31 steel. International Journal of Engineering Research and Applications, 5(6), pp.57–62. 56. Ehsan Asgar M. and Singh Singholi A.K., 2018. September. Parameter study and optimization of WEDM process: A Review. In Iop conference series: Materials science and engineering (Vol. 404, p. 012007). IOP Publishing. 57. Muthukumar V., Rajesh N., Venkatasamy R., Sureshbabu A. and Senthilkumar N., 2014. Mathematical modeling for radial overcut on electrical discharge machining of Incoloy 800 by response surface meth- odology. Procedia Materials Science, 6, pp.1674–1682. 58. Muthukumar V., Suresh Babu A., Venkatasamy R. and Senthil Kumar N., 2015. An accelerated particle swarm optimization algorithm on parametric optimization of WEDM of die-steel. Journal of The Institu- tion of Engineers (India): Series C, 96, pp.49–56. PLOS ONE | https://doi.org/10.1371/journal.pone.0308203 October 28, 2024 26 / 26

Journal

PLoS ONE – Public Library of Science (PLoS) Journal

Published: Oct 28, 2024

There are no references for this article.