Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Stall Margin Improvement in an Axial Compressor by Continuous and Pulsed Tip Injection

Stall Margin Improvement in an Axial Compressor by Continuous and Pulsed Tip Injection International Journal of Turbomachinery Propulsion and Power Article Stall Margin Improvement in an Axial Compressor by Continuous and Pulsed Tip Injection Joseph Moubogha Moubogha, Gabriel Margalida, Pierric Joseph, Olivier Roussette and Antoine Dazin * UMR 9014—LMFL—Laboratoire de Mécanique des Fluides de Lille—Kampé de Fériet, Arts et Métiers Institute of Technology, Centrale Lille, University Lille, CNRS, ONERA, F-59000 Lille, France; joseph.moubogha-moubogha@ensam.eu (J.M.M.); gabriel.margalida@ensam.eu (G.M.); pierric.joseph@ensam.eu (P.J.); olivier.roussette@ensam.eu (O.R.) * Correspondence: antoine.dazin@ensam.eu † This manuscript is an extended version of our meeting paper published in the Proceedings of the 14th European Turbomachinery Conference, Gdansk, Poland, 12–16 April 2021. Abstract: Stall and surge are strong limitations in the operating range of compressors and thus one of the major limits of jet engine performance. A promising way to push the stability limit of compression machines is to inject a small amount of flow at the blade tip to alter the physical mechanism responsible for stall onset. This study focuses on the experimental performance of such a system. To do so, an axial compressor test bench was equipped with 40 actuators connected to an auxiliary pressurised air supply system. They were able to generate high-speed jet blowing just at the tip of the rotor blades. The opening of each actuator was controlled by an electromagnetic valve. This allowed generating continuous or pulsed jets with frequencies up to 500 Hz at different duty cycles. The performance of the control system was investigated for various control strategies, where the injected flow rate, the injection angle, the number of injectors, the jet frequency and the duty cycle were systematically varied. This paper is concluded by a study of the energy balance of the system for various configurations. To the best of the authors’ knowledge, this constitutes a rarely seen analysis in the literature. Citation: Moubogha, J.M.; Margalida, Keywords: axial compressor; active flow control; stall margin improvement; tip blowing; energy cost G.; Joseph, P.; Roussette, O.; Dazin, A. Stall Margin Improvement in an Axial Compressor by Continuous and Pulsed Tip Injection. Int. J. 1. Introduction Turbomach. Propuls. Power 2022, 7, 10. https://doi.org/10.3390/ijtpp7010010 The problem of stall and surge instabilities developing in axial compressors is nearly as old as the first gas turbine [1,2], and the loss of stability of axial compressors at a Received: 14 October 2021 high-pressure ratio is still nowadays an important limitation of the global performance of Accepted: 6 March 2022 modern aero engines, as the occurrence of these phenomena can cause dramatic events. It Published: 16 March 2022 thus leads engine manufacturers to apply a consequent security margin (the so-called stall Publisher’s Note: MDPI stays neutral margin (SM)) to keep the compressor far from its stability limit. This deprives the machines with regard to jurisdictional claims in of their higher-pressure ratio and higher-efficiency operating ranges, which penalises the published maps and institutional affil- global performance of the engine. Stall margin improvements (SMI) can be achieved by iations. various methods: a large number of works in the literature are devoted to passive control techniques, such as casing treatments [3]. These methods involve permanent modifications of the casing that cannot cope with changes in flow conditions. Using active flow control systems, which can easily be turned off and on, is thus an attractive alternative, and Copyright: © 2022 by the authors. many research works have been carried out concerning this kind of techniques and were Licensee MDPI, Basel, Switzerland. summarised recently by Li et al. [4]. One should keep in mind, however, that such methods This article is an open access article come with some technical drawbacks in terms of complexity, added weight and, potentially, distributed under the terms and conditions of the Creative Commons reliability issues. According to Day’s review [2], the most common stall inception in the Attribution (CC BY-NC-ND) license modern engines is the spike, which originates from phenomena occurring at the blade (https://creativecommons.org/ tip [5,6]. The most efficient way to control these phenomena is, thus, to blow a high- licenses/by-nc-nd/4.0/). momentum jet at the blade tip, as demonstrated by several authors [7–9], to decrease the Int. J. Turbomach. Propuls. Power 2022, 7, 10. https://doi.org/10.3390/ijtpp7010010 https://www.mdpi.com/journal/ijtpp Int. J. Turbomach. Propuls. Power 2022, 7, 10 2 of 15 loading and/or act on the dynamic of the tip gap vortex. Among all the literature dedicated to this subject, there is a consensus about some specific characteristics of the control system that must be used to obtain a significant effect in terms of SMI: The jet has to be as close as possible to the casing wall; consequently, many authors have successfully used the Coandă effect to get a wall-attached jet [10,11]. A high momentum (or high velocity) and good angular coverage [9] are needed. Nevertheless, some points remain unclear. As stated by Li et al. [4], there is no consensus about the effect of the yaw angle (positive angle values are given by the rotor rotation direction). Whereas the first paper [12] originally claimed that a positive yaw angle is better, other authors have demonstrated the opposite [13,14]. In addition, to the best of the authors’ knowledge, the balance between the energy cost of blowing and the positive effect on machine performance has never been clearly established. It is therefore difficult to answer the question of whether such a system “pays its place” in a real engine. Finally, the potential interest of pulsed blowing has barely been addressed [7], whereas it has shown its efficiency in some other flow control applications, such as separation control [15]. In previous studies conducted in the laboratory by the same research team, the flow mechanisms involved during the onset of stall with and without active flow control have notably been analysed. The control system effectively succeeded in increasing the operating range of the compressor by neutralising the spike mechanisms and by moving the last stability point close to the maximum of the performance curve. For some control config- urations, a low-frequency phenomenon appears in pressure measurements, suggesting a transition from spike-type stall to modal-type stall inception. In addition, some important effects, such as the injected momentum, have been investigated [16,17]. This paper aims to complete this previous study by addressing the above-mentioned questions concerning the yaw angle and energy balance of the system using pulsed ac- tuation. This study relies on an experimental parametric study conducted on the same single-stage axial compressor test bench equipped with a modular flow injection sys- tem [17]. After a description of the experimental set-up, the paper focuses on the effect of the blowing angle at several rotation speeds to clearly point out the effect of blowing on the rotating frame. The paper is then dedicated to an estimation of the energy costs and savings of the control system to try to evidence the most interesting blowing strategies in pulsed or continuous blowing. This work is the first part of the EU-funded Horizon 2020 research project ACONIT [18], which aims at designing, manufacturing and testing actuators for flow control for implantation in an aircraft engine. 2. Experimental Set-Up The support of the experiments performed in this research work is the low-speed, single-stage axial compressor CME2 located at the Arts et Métiers Institute of Technology in Lille (France). This compressor is a subsonic machine comparable to a stage of a high- pressure component of an aero engine [2]. Initially designed as a convenient tool to study rotating stall [19,20], this specific test bench has been equipped in the recent years with an active flow control (AFC) system relying on magnetic valves to produce pulsed jets [16]. The compressor itself, depicted in Figure 1a and whose characteristics are listed in Table 1, is operated at rotational speeds ranging from 3200 rpm to 4500 rpm. A typical per- formance curve is provided in Figure 1b, plotting the total-to-static pressure rise coefficient Y = DP /0.5rU as a function of the flow coefficient F = V /U . During the tests, ts ts x tip tip the mass flow is varied using a throttling valve located downstream of the compressor stage (see Figure 1a). In this paper, stall is triggered by continuously closing the throttling valve up to the unstable part of the performance curve. Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 3 of 16 Int. J. Turbomach. Propuls. Power 2022, 7, 10 3 of 15 (a) (b) Figure 1. CME2 test rig: (a) schematic description; (b) typical performance curve of the compressor Figure 1. CME2 test rig: (a) schematic description; (b) typical performance curve of the compressor at 3200 rpm. at 3200 rpm. Table 1. Compressor parameters at 3200 rpm. Table 1. Compressor parameters at 3200 rpm. Parameters Parameters Value Value UUni nits ts 1 −1 Design mass flow rate 5.3 kg· s Design mass flow rate 5.3 kgs Design axial velocity, LE * 43 ms −1 Design axial velocity, LE * 43 m· s Rotor blade number 30 Rotor blade number 30 Stator blade number 40 Stator bl Casing diameter ade number 550 40 mm Hub-tip ratio, LE 0.75 Casing diameter 550 mm Rotor tip chord 84 mm Hub-tip ratio, LE 0.75 Rotor tip stagger angle 54 Rotor Rotor t tipip cho gap rd 0.5 84 mm mm Rotor tip speed 94 ms Rotor tip stagger angle 54 ° * LE, leading edge. Rotor tip gap 0.5 mm −1 Rotor tip speed 94 m· s The performance of the compressor is evaluated using two differential pressure sensors * LE, leading edge. located on the test rig. The first sensor measures the difference between the total pressure recorded in the plenum chamber and a mean static pressure measured at the end of the converging pipe located just downstream of the plenum chamber. This value allows The performance of the compressor is evaluated using two differential pressure sen- capturing the dynamic pressure at the compressor inlet, and then the flow rate. The second sors located on the test rig. The first sensor measures the difference between the total pres- sensor is used to evaluate the stage performance by measuring the static pressure in front sure recorded in the plenum chamber and a mean static pressure measured at the end of of the rotor and downward of the stator. The precision of these measurements has been the converging pipe located just downstream of the plenum chamber. This value allows evaluated to 0.012 kgs and 1.5 Pa for the flow rate and the total-to-static pressure capturing the dynamic pressure at the compressor inlet, and then the flow rate. The sec- rise, respectively [16]. ond sensor is used to evaluate the stage performance by measuring the static pressure in The control system [17] (see the overall description in Figure 2) consists of 20 injection front of the rotor and downward of the stator. The precision of these measurements has blocks, each one counting two injectors. This configuration was selected because of the −1 been evaluated to ±0.012 kg· s and ±1.5 Pa for the flow rate and the total-to-static pressure space constraint caused by the curvature of the casing. A solenoid valve (Matrix MX821), whose driving frequency can be set between 0 (continuous) and 500 Hz, with a supply rise, respectively [16]. pressure of up to 8 bar, feeds each injector. Each injector can be then operated independently The control system [17] (see the overall description in Figure 2) consists of 20 injection 1 2 and can produce a jet speed of up to 200 ms through a 10  0.5 mm slot. Accordingly, blocks, each one counting two injectors. This configuration was selected because of the the injected mass flow can be set from 0 to approximately 2.5% of the main flow rate of the space constraint caused by the curvature of the casing. A solenoid valve (Matrix MX821), stage either by changing the supply pressure or by changing the duty cycle (DC) of the whose driving frequency can be set between 0 (continuous) and 500 Hz, with a supply solenoid valves when pulsed injection is used. The duty cycle is defined as the blowing pressure of up to 8 bar, feeds each injector. Each injector can be then operated inde- time duration divided by the total period duration (i.e., the sum of blowing and no blowing −1 pendently and can produce a jet speed of up to 200 m· s through a 10 × 0.5 mm² slot. time). All solenoid valves are driven by the same command signal, and manufacturer Accordingly, the injected mass flow can be set from 0 to approximately 2.5% of the main data indicate a response time lower than 1 ms. As stated in the literature, blowing is most effective in front of the rotor leading edge [14], and the critical area is located at the tip, flow rate of the stage either by changing the supply pressure or by changing the duty cycle close to the casing [4]. Consequently, actuators are located 10 mm upstream of the rotor (DC) of the solenoid valves when pulsed injection is used. The duty cycle is defined as the (x = 20%  Cx), and injectors are shaped using the Coandă effect to blow along the casing blowing time duration divided by the total period duration (i.e., the sum of blowing and no blowing time). All solenoid valves are driven by the same command signal, and man- ufacturer data indicate a response time lower than 1 ms. As stated in the literature, blow- ing is most effective in front of the rotor leading edge [14], and the critical area is located at the tip, close to the casing [4]. Consequently, actuators are located 10 mm upstream of the rotor (x = −20% ∙ Cx), and injectors are shaped using the Coandă effect to blow along Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 4 of 16 Int. J. Turbomach. Propuls. Power 2022, 7, 10 4 of 15 the casing right in the tip gap [11]. Injectors can also be rotated along their axis to vary the right in the tip gap [11]. Injectors can also be rotated along their axis to vary the yaw angle yaw angle of injection with an angular step of 15°. of injection with an angular step of 15 . Figure 2. Active flow control system description. Figure 2. Active flow control system description. The experimental protocol is extensively described in a previous paper [17], but key The experimental protocol is extensively described in a previous paper [17], but key information is reported here for the sake of completeness. In a typical stall test, the working information is reported here for the sake of completeness. In a typical stall test, the work- point of the compressor is moved along the performance curve using the controlled closing ing point of the compressor is moved along the performance curve using the controlled of a throttling valve. Stall is easily perceptible through an abrupt drop in the mass flow and closing of a throttling valve. Stall is easily perceptible through an abrupt drop in the mass pressure ratio. The baseline curve is compared with a controlled curve, which is obtained flow with and contr pres ol sur activated e ratio. for Ththe e bas entir eline e working curve is range compare of the d with compr a con essor troll . Ther ed cu erve, is thus which no is ob issue tained with with activation control delay activated of the fo active r the flow entire contr work ol in system. g range of the compressor. There is A typical example of results of the flow control system is proposed in Figure 3, where thus no issue with activation delay of the active flow control system. the total-to-static pressure rise coefficient is plotted as a function of the flow coefficient. A typical example of results of the flow control system is proposed in Figure 3, where For each control case, the actual value of the global mass flow rate injected is specified the total-to-static pressure rise coefficient is plotted as a function of the flow coefficient. along with the duty cycle (in pulsed blowing). Here and subsequently, the baseline results For each control case, the actual value of the global mass flow rate injected is specified correspond to the performance of the compressor without control and the injected mass along with the duty cycle (in pulsed blowing). Here and subsequently, the baseline results flow rate Q is expressed in scaled form, Q , as a percentage of the compressor flow rate inj inj correspond to the performance of the compressor without control and the injected mass at the last stable operating point before stall without control, q . It is obvious from these flow rate 𝑄 is expressed in scaled form, 𝑄 , as a percentage of the compressor flow results that the blowing acts positively on the stable operating range of the compressor. rate at the last stable operating point before stall without control, 𝑞 . It is obvious from To evaluate the effect of the control system on the performance curve of the machine, these results that the blowing acts positively on the stable operating range of the compres- the definition of the stall margin improvement given by Weigl et al. [21] is used and sor. calculated using the following equations: To evaluate the effect of the control system on the performance curve of the machine, q P N S the definition of the stall margin improvement given by Weigl et al. [21] is used and cal- SM = (  1)  100 and (1) q P culated using the following equationsS: N 𝑞 𝛱 𝑁 𝑆 SM SM C B = ( × − 1) × 100 and (1) SMI =  100, (2) 𝑞 𝛱 𝑆 𝑁 SM 𝑆𝑀 − with q and P, respectively, being the flow rate 𝐶 and 𝐵 the pressure ratio. Please note that in 𝑆𝑀𝐼 = × 100, (2) 𝑆𝑀 Equations (1) and (2), along with Figure 4, the subscripts N and S refer to quantities at the nominal operating point and at the last stable operating point, respectively, before stall with 𝑞 and 𝛱 , respectively, being the flow rate and the pressure ratio. Please note that in (i.e., the operating point with the lowest flow rate before stall onset). Similarly, subscript B Equations (1) and (2), along with Figure 4, the subscripts N and S refer to quantities at the refers to the baseline case, without control, and C to the controlled case. nominal operating point and at the last stable operating point, respectively, before stall (i.e., the operating point with the lowest flow rate before stall onset). Similarly, subscript B refers to the baseline case, without control, and C to the controlled case. 𝑆𝑀 𝑆𝑀 𝑖𝑛𝑗 𝑖𝑛𝑗 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 5 of 16 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 5 of 16 Int. J. Turbomach. Propuls. Power 2022, 7, 10 5 of 15 Figure 3. Performance curves obtained using the pulsed jet control system with various duty cycle and global injected mass flow rate values (actuation frequency 𝑓 = 100 Hz, 20 injectors activated, absolute blowing flow angle 𝛼 = 0° and rotation speed 𝛺 = 3200 rpm). The duty cycle 𝐷𝐶 = 𝑗𝑒𝑡 Figure 3. Performance curves obtained using the pulsed jet control system with various duty cycle Figure 3. Performance curves obtained using the pulsed jet control system with various duty cycle 1 corresponds to continuous blowing. and global injected mass flow rate values (actuation frequency f = 100 Hz, 20 injectors activated, and global injected mass flow rate values (actuation frequency 𝑓 = 100 Hz, 20 injectors activated, absolute blowing flow angle a = 0 and rotation speed W = 3200 rpm). The duty cycle DC = 1 jet absolute blowing flow angle 𝛼 = 0° and rotation speed 𝛺 = 3200 rpm). The duty cycle 𝐷𝐶 = 𝑗𝑒𝑡 corresponds to continuous blowing. 1 corresponds to continuous blowing. Figure 4. Schematic description of the SMI (adapted from [21]). Figure 4. Schematic description of the SMI (adapted from [21]). 3. Effect of the Injection Yaw Angle 3. Effect of the Injection Yaw Angle To highlight the effect of the injection yaw angle, a first series of experiments were Figure 4. Schematic description of the SMI (adapted from [21]). To highlight the effect of the injection yaw angle, a first series of experiments were conducted at 3200 rpm. The absolute blowing flow angle , the velocity (or the injected jet conducted at 3200 rpm. The absolute blowing flow angle jet, the velocity (or the injected mass flow rate) and the number of injectors activated were varied. Absolute and relative 3. Effect of the Injection Yaw Angle mass flow rate) and the number of injectors activated were varied. Absolute and relative blowing flow angles are defined in Figure 5. Positive absolute blowing angle values are To highlight the effect of the injection yaw angle, a first series of experiments were blowing flow angles are defined in Figure 5. Positive absolute blowing angle values are given by the rotor rotation direction in the study; therefore, the absolute blowing angle is conducted at 3200 rpm. The absolute blowing flow angle jet, the velocity (or the injected given by the rotor rotation direction in the study; therefore, the absolute blowing angle is considered negative when the blowing direction is opposite to the rotor rotation direction. mass flow rate) and the number of injectors activated were varied. Absolute and relative considered negative when the blowing direction is opposite to the rotor rotation direction. Several basic test results of the active flow control in continuous blowing at 3200 rpm blowing flow angles are defined in Figure 5. Positive absolute blowing angle values are with 40 injectors activated and various absolute blowing flow angle and injected mass given by the rotor rotation direction in the study; therefore, the absolute blowing angle is flow rate values are detailed and presented in Figure 6. The total-to-static pressure rise considered negative when the blowing direction is opposite to the rotor rotation direction. coefficient is plotted as a function of the flow coefficient for the six absolute blowing flow angles investigated: 30 , 15 , 0 , 15 , 30 and 45 . It can be seen that as already observed in Figure 3, for each absolute injection angle, the blowing extends the operating range of the compressor and simultaneously increases the compressor performance. It Int. J. Turbomach. Propuls. Power 2022, 7, 10 6 of 15 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 6 of 16 is also rather obvious that higher benefits are obtained for negative absolute blowing flow angles. Figure 5. Definition and illustration of the absolute and relative blowing flow angles. Positive angle Figure 5. Definition and illustration of the absolute and relative blowing flow angles. Positive angle values are given by the rotor rotation direction. values are given by the rotor rotation direction. To investigate the effect of the blowing flow angle, the stall margin improvement was Several basic test results of the active flow control in continuous blowing at 3200 rpm calculated for each experiment reported in Figure 6, and the results are reported in Figure 7. with 40 injectors activated and various absolute blowing flow angle and injected mass To make the comparison of the different sets of parameters easier, the absolute blowing flow rate values are detailed and presented in Figure 6. The total-to-static pressure rise velocity V was scaled by the rotor tip speed. The velocity of the jet, V , was obtained jet jet coefficient is plotted as a function of the flow coefficient for the six absolute blowing flow by hot-wire measurements conducted on the injectors in a dedicated fluidic actuator test angles investigated: 30°, 15°, 0°, −15°, −30° and −45°. It can be seen that as already observed bench [16]. In addition, the results are presented as a function of the relative blowing flow in Figure 3, for each absolute injection angle, the blowing extends the operating range of angle b , i.e., the jet flow angle seen by the blade, which was derived from the absolute jet the compressor and simultaneously increases the compressor performance. It is also ra- velocity and angle of the jet, and the rotor tip speed using the velocity triangle (see Figure 5). For the majority of the tested jet velocities, the stall margin improvement presents ther obvious that higher benefits are obtained for negative absolute blowing flow angles. a clear and monotonic increase as the relative blowing angle decreases and reaches in most cases a maximum value for relative flow angles in the range between 60 and 70 . It then appears to remain relatively constant or to decrease slightly at lower relative flow angles. The effect of the rotation velocity was investigated and is presented in Figure 8. The stall margin improvement is plotted as a function of the relative blowing angle for two main rotor rotational velocities: 3200 and 4500 rpm. After scaling the absolute blowing velocity by the appropriate rotor tip speed, results were fairly close for the two different rotational speeds. The monotonic increase in the stall margin improvement with the decreasing of the relative blowing angle and the plateau reached around 60 and 70 is once again clearly apparent. It has to be pointed out that the inlet blade angle at the tip (depicted in Figure 8 by the vertical blue dashed line) of this compressor is 65 . It thus appears that the highest effect of the stall margin improvement is obtained for relative blowing angles close to the inlet blade angle, which is consistent with some previous experimental observations [7]. This can also be easily explained as this blowing angle is certainly most suitable for decreasing the blade loading at the tip and thus preventing the mechanism occurring at the tip and leading to rotating stall. Int. J. Turbomach. Propuls. Power 2022, 7, 10 7 of 15 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 7 of 16 (𝐚 ) 𝛼 = 30° (𝐛 ) 𝛼 = 15° 𝑗𝑒𝑡 𝑗𝑒𝑡 ( ) ( ) 𝐜 𝛼 = 00° 𝐝 𝛼 = −15° 𝑗𝑒𝑡 𝑗𝑒𝑡 (𝐞 ) 𝛼 = −30° (𝐟 ) 𝛼 = −45° 𝑗𝑒𝑡 𝑗𝑒𝑡 Figure Figure 6. 6. Performance Performance curves curves obtained obtained using using continuous continuous blowing blowing with with various various absolute absoluteblowing blowing angles of: (a) 30°, (b) 15°, (c) 0°, (d) −15°, (e) −30° and (f) −45° and various injected mass flow rate angles of: (a) 30 , (b) 15 , (c) 0 , (d) 15 , (e) 30 and (f) 45 and various injected mass flow rate values (40 injectors activated and rotation speed 𝛺 = 3200 rpm). values (40 injectors activated and rotation speed W = 3200 rpm). Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 8 of 16 To investigate the effect of the blowing flow angle, the stall margin improvement was calculated for each experiment reported in Figure 6, and the results are reported in Figure 7. To make the comparison of the different sets of parameters easier, the absolute blowing velocity 𝑉 was scaled by the rotor tip speed. The velocity of the jet, 𝑉 , was obtained 𝑗𝑒𝑡 𝑗𝑒𝑡 by hot-wire measurements conducted on the injectors in a dedicated fluidic actuator test bench [16]. In addition, the results are presented as a function of the relative blowing flow angle jet, i.e., the jet flow angle seen by the blade, which was derived from the absolute velocity and angle of the jet, and the rotor tip speed using the velocity triangle (see Figure Int. J. Turbomach. Propuls. Power 2022, 7, 10 8 of 15 5). Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 9 of 16 Figure 7. Effect of the relative blowing flow angle with 40 injectors activated for a rotation speed of Figure 7. Effect of the relative blowing flow angle with 40 injectors activated for a rotation speed of 3200 rpm. 3200 rpm. For the majority of the tested jet velocities, the stall margin improvement presents a clear and monotonic increase as the relative blowing angle decreases and reaches in most cases a maximum value for relative flow angles in the range between −60° and −70°. It then appears to remain relatively constant or to decrease slightly at lower relative flow angles. The effect of the rotation velocity was investigated and is presented in Figure 8. The stall margin improvement is plotted as a function of the relative blowing angle for two main rotor rotational velocities: 3200 and 4500 rpm. After scaling the absolute blowing velocity by the appropriate rotor tip speed, results were fairly close for the two different rotational speeds. The monotonic increase in the stall margin improvement with the de- creasing of the relative blowing angle and the plateau reached around −60° and −70° is once again clearly apparent. It has to be pointed out that the inlet blade angle at the tip (depicted in Figure 8 by the vertical blue dashed line) of this compressor is −65°. It thus appears that the highest effect of the stall margin improvement is obtained for relative blowing angles close to the inlet blade angle, which is consistent with some previous ex- perimental observations [7]. Figure 8. Effect of the relative blowing flow angle with 40 injectors activated at different rotating speeds. Figure 8. Effect of the relative blowing flow angle with 40 injectors activated at different rotating speeds. 4. Energy Balance The main goal of the control system is to improve the compressor stall margin. An This can also be easily explained as this blowing angle is certainly most suitable for additional benefit is the improvement in the pressure rise provided by the compressor decreasing the blade loading at the tip and thus preventing the mechanism occurring at stage (and thus a gain in the energy provided by the compressor to the flow). Nevertheless, the tip and leading to rotating stall. the generation of jets involves an energy cost. The power balance, introduced next, can be seen as the net benefit (energy gain – energy cost) of the control system. So, if it is negative, 4. Energy Balance then the use of the control system costs more than it brings in. The main goal of the control system is to improve the compressor stall margin. An The power balance (PB) of the control system is thus evaluated by subtracting the additional benefit is the improvement in the pressure rise provided by the compressor power cost (PC) of the blowing to the associated power gain (PG), as defined below: stage (and thus a gain in the energy provided by the compressor to the flow). Neverthe- less, the generation of jets involves an energy cost. The power balance, introduced next, PB = PG PC. (3) can be seen as the net benefit (energy gain – energy cost) of the control system. So, if it is negative, then the use of the control system costs more than it brings in. Regarding the cost, it is the power consumed by the blowing system. In this case, weTh use e po a wer screw balance compr (PB essor ) of to thpr e con essur tro ise l sys the tem air isand thus solenoid evaluated valves by subtract to carry ing out the the populsed wer cost blowing. (PC) of the blo It is ther wing to th efore possible e associ to ated estimate power g the ain cost (PG of ), as de the injection fined below: through the electrical power consumed by all these elements. However, this includes many other 𝑃𝐵 = 𝑃𝐺 − 𝑃𝐶 . (3) factors that are not of direct interest, such as the choice of the solenoid valve or the way in which compressed air is generated, which can be subsequently improved and which will Regarding the cost, it is the power consumed by the blowing system. In this case, we use a screw compressor to pressurise the air and solenoid valves to carry out the pulsed blowing. It is therefore possible to estimate the cost of the injection through the electrical power consumed by all these elements. However, this includes many other factors that are not of direct interest, such as the choice of the solenoid valve or the way in which compressed air is generated, which can be subsequently improved and which will un- doubtedly be different from the final solution embedded in an engine. Consequently, the power consumed by the injection system at the last level is estimated by evaluating the aeraulic power added by the jets. Please note that in this study, the temperature of the injected air was close to the ambient one, as several buffer tanks (a large 500 L tank fol- lowed by two smaller 15 L ones) are present in the pressured air supply system. The power cost (PC) of the blowing is defined as the kinetic power added to the flow by the jets: 𝑃𝐶 = 𝑄 𝑉 , (4) 𝑗𝑒𝑡 where 𝑄 is the global injected mass flow rate and 𝑉 the mean jet velocity at the ac- 𝑗𝑒𝑡 tuator nozzle. 𝑖𝑛𝑗 𝑖𝑛𝑗 Int. J. Turbomach. Propuls. Power 2022, 7, 10 9 of 15 undoubtedly be different from the final solution embedded in an engine. Consequently, the power consumed by the injection system at the last level is estimated by evaluating the aeraulic power added by the jets. Please note that in this study, the temperature of the injected air was close to the ambient one, as several buffer tanks (a large 500 L tank followed by two smaller 15 L ones) are present in the pressured air supply system. The power cost (PC) of the blowing is defined as the kinetic power added to the flow by the jets: PC = Q V , (4) inj jet Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 10 of 16 where Q is the global injected mass flow rate and V the mean jet velocity at the inj jet actuator nozzle. The power gain (PG) is evaluated by comparing the performance of the compressor The power gain (PG) is evaluated by comparing the performance of the compressor with and without control at the flow rate corresponding to the last stable operating point with and without control at the flow rate corresponding to the last stable operating point without control (see the representation given in Figure 9). More precisely, it is defined as without control (see the representation given in Figure 9). More precisely, it is defined as the difference between the net power available in the fluid downstream of the compressor the difference between the net power available in the fluid downstream of the compressor with and without control; with and without control; " ! !# 2 22 𝑃 V 𝑉 𝑃 V 𝑉 P 2𝐶 P 2𝐵 2𝐶 2𝐵 2C 2C 2B 2B 𝑃𝐺 = 𝑞 [( + ) − ( + )], (5) PG = q + + , (5) 𝜌 2 𝜌 2 2𝐶 2𝐵 r 2 r 2 2C 2B where 𝑞 is the flow rate at the last stable operating point before stall without control; where q is the flow rate at the last stable operating point before stall without control; P the 𝑃 the static pressure; 𝑉 the velocity; 𝜌 the density; and the indexes 2, C and B, respec- static pressure; V the velocity; r the density; and the indexes 2, C and B, respectively, the tively, the stage outlet, the controlled configuration and the baseline (configuration with- stage outlet, the controlled configuration and the baseline (configuration without control). out control). Figure 9. Energy gain due to the active flow control system. Figure 9. Energy gain due to the active flow control system. 4.1. Continuous Mode 4.1. Continuous Mode Figure 10 presents the results of a series of tests performed on the compressor operating Figure 10 presents the results of a series of tests performed on the compressor oper- at 3200 rpm. All the experiments reported on the graph correspond to continuous blowing, ating at 3200 rpm. All the experiments repo  rted on the graph correspond to continuous with an absolute blowing angle a = 30 . This absolute blowing angle was retained jet blowing, with an absolute blowing angle jet = −30°. This absolute blowing angle was re- for the rest of the study as it allowed achieving the best SMI, according to the reasons tained for the rest of the study as it allowed achieving the best SMI, according to the rea- developed, due to the results in Figure 7. The effect of the number of injectors used (N) and sons developed, due to the results in Figure 7. The effect of the number of injectors used the global blowing flow rate was examined. The figure shows, for each tested configuration, (N) and the global blowing flow rate was examined. The figure shows, for each tested the SMI achieved compared to the power balance (PB) of the considered control strategy. configuration, the SMI achieved compared to the power balance (PB) of the considered On this graph, the most interesting points are on the top and the right of the figure, as they control strategy. On this graph, the most interesting points are on the top and the right of correspond to control parameters achieving significant SMI with a positive power balance. the figure, as they correspond to control parameters achieving significant SMI with a pos- itive power balance. Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 11 of 16 Int. J. Turbomach. Propuls. Power 2022, 7, 10 10 of 15 Figure 10. SMI and power balance of the control system. Continuous blowing with the absolute injec- Figure 10. SMI and power balance of the control system. Continuous blowing with the absolute tion angle a = 30 and the rotation speed W = 3200 rpm. N is the number of injectors activated. injection jang et le 𝛼 = −30° and the rotation speed 𝛺 = 3200 rpm. N is the number of injectors ac- 𝑗𝑒𝑡 tivated. Some remarkable points in the graph are highlighted with letters, from A to F. Points A, B and C correspond to strategies that allow reaching the greatest SMI. Nevertheless, Some remarkable points in the graph are highlighted with letters, from A to F. Points these configurations correspond to the maximum number of injectors with the maximum A, B and C correspond to strategies that allow reaching the greatest SMI. Nevertheless, flow rate (and velocity) per injector. Consequently, the energy cost is high, and the energy balance for this specific case is unfavourable. Points D, E and F correspond to interesting these configurations correspond to the maximum number of injectors with the maximum applicative configurations as they allow to obtain a fairly good SMI (from 55% for point F flow rate (and velocity) per injector. Consequently, the energy cost is high, and the energy to 80% for point D) with a positive power balance, which can reach 1.8% of the compressor balance for this specific case is unfavourable. Points D, E and F correspond to interesting nominal power. applicative configurations as they allow to obtain a fairly good SMI (from 55% for point F These most interesting points are all obtained for configurations with 30 to 40 injectors to 80% for point D) with a positive power balance, which can reach 1.8% of the compressor activated, which means that good angular coverage is necessary to reach a good compro- mise nom between inal pow theer. SMI and the positive power balance. This need for a sufficient angular cov- erage was also highlighted by Suder et al. [8] and more recently by Margalida et al. [16,17]. These most interesting points are all obtained for configurations with 30 to 40 injec- This parameter is determining for the increase in the SMI. It seems that this is also the case tors activated, which means that good angular coverage is necessary to reach a good com- for the energy balance. promise between the SMI and the positive power balance. This need for a sufficient angu- The second observation coming from Figure 10 is that whatever the number of injectors lar coverage was also highlighted by Suder et al. [8] and more recently by Margalida et al. used, when the flow rate starts to increase, both the SMI and the power balance increase, [16,17]. This parameter is determining for the increase in the SMI. It seems that this is also leading to the best configurations, such as points D, E or F. When the blowing flow rate continues the case f to or incr the energ ease, the y bal SMIance. continues to increase [14], whereas the power balance deteriorates rapidly. The second observation coming from Figure 10 is that whatever the number of injec- Figure 11 shows the evolutions of the power gain (a), the power cost (b) and the power tors used, when the flow rate starts to increase, both the SMI and the power balance in- balance (c) as a function of the injected flow rate. What can be clearly observed is that the crease, leading to the best configurations, such as points D, E or F. When the blowing flow power gain is low for a low injected flow rate (less than 1%) and that the power gain grows rate continues to increase, the SMI continues to increase [14], whereas the power balance almost linearly and more rapidly than the power cost, which evolves approximatively as deteriorates rapidly. the jet speed cubed (or the injected flow rate cubed). This evolution leads to a rapid growth of the cost Figure for values 11 sh higher ows th than e evo 1–1.5%, lutions leading of to the rapid power degradation gain (a), of the the power powe balance r cost (b) and the power balance (c) as a function of the injected flow rate. What can be clearly observed is that the power gain is low for a low injected flow rate (less than 1%) and that the power gain grows almost linearly and more rapidly than the power cost, which evolves approx- imatively as the jet speed cubed (or the injected flow rate cubed). This evolution leads to a rapid growth of the cost for values higher than 1–1.5%, leading to rapid degradation of the power balance (Figure 11c). What can also be noticed in Figure 11b is that for a given Int. J. Turbomach. Propuls. Power 2022, 7, 10 11 of 15 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 12 of 16 (Figure 11c). What can also be noticed in Figure 11b is that for a given injected flow rate, injected flow rate, the power cost is lower for the configuration with more injectors acti- the power cost is lower for the configuration with more injectors activated: in this case, the vated: in this case, the flow rate is distributed between more injectors, which leads to a flow rate is distributed between more injectors, which leads to a lower velocity per injector lower velocity per injector and a lower cost, according to Equation (4). and a lower cost, according to Equation (4). Figure 11. Evolution of the: (a) power gain, (b) power cost and (c) power balance as a function of the Figure 11. Evolution of the: (a) power gain, (b) power cost and (c) power balance as a function of injected flow rate. Continuous blowing with the absolute injection angle a = 30 and the rotation jet the injected flow rate. Continuous blowing with the absolute injection angle 𝛼 = −30° and the 𝑗𝑒𝑡 speed W = 3200 rpm. rotation speed 𝛺 = 3200 rpm. The fact that the power balance is positive for a large range of Q indicates that the inj The fact that the power balance is positive for a large range of 𝑄 indicates that the blowing leads also to an increase in the blade work at the tip and/or a decrease in the losses blowing leads also to an increase in the blade work at the tip and/or a decrease in the (the profile losses close to the tip, as the flow is realigned with the blade inlet angle and losses (the profile losses close to the tip, as the flow is realigned with the blade inlet angle certainly also to the losses associated with the secondary gap flows, as demonstrated in the and certainly also to the losses associated with the secondary gap flows, as demonstrated case of an isolated blade, for some blowing configurations [22]). in the case of an isolated blade, for some blowing configurations [22]). 4.2. Pulsed Mode Figure 12 presents a typical evolution of the stall margin improvement with the ac- tuation frequency, in pulsed mode. Note that in Figure 12, no result is presented for fre- quencies above 200 Hz, as beyond this value, the response time of the valve becomes sig- nificant compared with the blowing duration. In this case, 40 injectors are activated, and the global injected mass flow rate is kept constant (here 0.03 kg/s) at different actuation frequencies with the same duty cycle (DC = 0.7). The absolute pulsed blowing angle is 𝑖𝑛𝑗 Int. J. Turbomach. Propuls. Power 2022, 7, 10 12 of 15 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 13 of 16 4.2. Pulsed Mode Figure 12 presents a typical evolution of the stall margin improvement with the −30°, and the rotor rotation speed is set at 3200 rpm. It can be seen that there is a strong actuation frequency, in pulsed mode. Note that in Figure 12, no result is presented for frequencies above 200 Hz, as beyond this value, the response time of the valve becomes dependency between the SMI and the actuation frequency. The stall margin improvement significant compared with the blowing duration. In this case, 40 injectors are activated, and grows almost monotonically with the frequency and reaches its maximum for a value the global injected mass flow rate is kept constant (here 0.03 kg/s) at different actuation close to the maximum frequency allowed by the system. However, this maximum value frequencies with the same duty cycle (DC = 0.7). The absolute pulsed blowing angle is 30 , and the rotor rotation speed is set at 3200 rpm. It can be seen that there is a strong is almost already reached around an actuation frequency of 200 Hz. Consequently, the dependency between the SMI and the actuation frequency. The stall margin improvement results in pulsed actuation that are presented correspond to the best results in terms of the grows almost monotonically with the frequency and reaches its maximum for a value SMI. Please note that in this study, all injectors pulsated simultaneously but that a vari- close to the maximum frequency allowed by the system. However, this maximum value is almost already reached around an actuation frequency of 200 Hz. Consequently, the ating actuation in the circumferential direction was also possible with this set-up (as pre- results in pulsed actuation that are presented correspond to the best results in terms of the viously performed a few times in the literature [2]). SMI. Please note that in this study, all injectors pulsated simultaneously but that a variating actuation in the circumferential direction was also possible with this set-up (as previously performed a few times in the literature [2]). Figure 12. Evolution of the SMI with the actuation frequency in pulsed blowing. The number of Figure 12. Evolution of the SMI with the actuation frequency in pulsed blowing. The number of activated injectors N = 40, the injected mass flow rate Q = 0.03 kg/s, the duty cycle DC = 0.7, inj the absolute blowing flow angle a = 30 and the rotation speed W = 3200 rpm. activated injectors 𝑁 = 40, the jet injected mass flow rate 𝑄 = 0.03 kg/s, the duty cycle 𝐷𝐶 = 0.7, the 𝑖𝑛𝑗 absolute blowing flow angle 𝛼 = −30° and the rotation speed 𝛺 = 3200 rpm. 𝑗𝑒𝑡 Figure 13 reports the comparison of the stall margin improvement and power balance for several pulsed actuation tests compared to the points obtained in continuous mode for 30 and 40 injectors activated. The absolute injection angle and the rotation speed are Figure 13 reports the comparison of the stall margin improvement and power balance kept constant and equal, respectively, to 30 and 3200 rpm. It is clear that the benefit of for several pulsed actuation tests compared to the points obtained in continuous mode for pulsed actuation is not obvious in terms of the SMI, as it allows reaching values up to 50% 30 and 40 injectors activated. The absolute injection angle and the rotation speed are kept maximum, sensibly lower than the higher ones obtained in continuous blowing. Nevertheless, if the power balance is considered, pulsed actuation presents a real constant and equal, respectively, to −30° and 3200 rpm. It is clear that the benefit of pulsed interest. Point H is able to reach, in pulsed actuation, an SMI performance close to the one actuation is not obvious in terms of the SMI, as it allows reaching values up to 50% maxi- obtained by point F (in continuous blowing) with a lower flow rate taken from the external mum, sensibly lower than the higher ones obtained in continuous blowing. system and a better power balance. Int. J. Turbomach. Propuls. Power 2022, 7, 10 13 of 15 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 14 of 16 Figure 13. SMI and power balance in pulsed and continuous blowing with the absolute injection Figure 13. SMI and power balance in pulsed and continuous blowing with the absolute injection angle 𝛼 = −30°  and the rotation speed 𝛺 = 3200 rpm. 𝑗𝑒𝑡 angle a = 30 and the rotation speed W = 3200 rpm. jet 5. Conclusions Nevertheless, if the power balance is considered, pulsed actuation presents a real in- terest. Point H is able to reach, in pulsed actuation, an SMI performance close to the one This study reports the results of flow control application for stall margin improvement obtained by point F (in continuous blowing) with a lower flow rate taken from the external (SMI) using an experimental parametric study conducted on a single-stage axial compressor system and a better power balance. test bench. The machine is equipped with fluidic actuators installed on the casing, upstream of the rotor. A series of experiments with various blowing conditions in pulsed and 5. Conclusions continuous modes were conducted. In particular, the aim of the study was, firstly, to shed some light on the controversial influence of the blowing yaw angle. Secondly, it was also This study reports the results of flow control application for stall margin improve- an opportunity to carry out a never-seen investigation of the energy budget of such control ment (SMI) using an experimental parametric study conducted on a single-stage axial methods applied to an axial compressor. compressor test bench. The machine is equipped with fluidic actuators installed on the Concerning the blowing yaw angle effect, it appears that blowing angles, in the casing, upstream of the rotor. A series of experiments with various blowing conditions in relative frame, close to the blade angle at the tip produce the best results in terms of the pulsed and continuous modes were conducted. In particular, the aim of the study was, SMI, as, at this blowing angle, the jet directly acts on the blade loading at the tip and firstly, to shed some light on the controversial influence of the blowing yaw angle. Sec- thus prevents the phenomena at the origin of stall. It is then not a matter of positive or ondly, it was also an opportunity to carry out a never-seen investigation of the energy negative absolute values, as stated often in the literature, and this confirms some findings budget of such control methods applied to an axial compressor. of Kefalakis et al. [7]. For real applications, the relative blowing angle is not the easiest Concerning the blowing yaw angle effect, it appears that blowing angles, in the rela- parameter to adjust, as it depends on the blowing velocity, the absolute blowing angle and tive frame, close to the blade angle at the tip produce the best results in terms of the SMI, the rotor speed. Fortunately, a high SMI value appears to be achieved for a quite large as, at this blowing angle, the jet directly acts on the blade loading at the tip and thus pre- range of relative blowing angles. It means that a single absolute blowing angle can cover vents the phenomena at the origin of stall. It is then not a matter of positive or negative several operating points. absolute values, as stated often in the literature, and this confirms some findings of Kefala- Concerning the energy budget, this study has shown that some of the blowing con- kis et al. [7]. For real applications, the relative blowing angle is not the easiest parameter figurations present a positive net gain on the energy balance for an SMI up to 110% and to adjust, as it depends on the blowing velocity, the absolute blowing angle and the rotor up to nearly 140% with a net energy consumption. Former configurations imply sufficient speed. Fortunately, a high SMI value appears to be achieved for a quite large range of angle coverage and are obtained when an advantageous balance is achieved between the relative blowing angles. It means that a single absolute blowing angle can cover several positive effect of the blowing (increase in the SMI, decrease in the losses) and the energy cost operating points. necessary to produce high-speed jets. The benefit of using pulsed blowing is not obvious in Concerning the energy budget, this study has shown that some of the blowing con- terms of the SMI but is clearly interesting for the power balance as some configurations figurations present a positive net gain on the energy balance for an SMI up to 110% and allow a positive power balance of 2% with a still interesting SMI of 50%. up to nearly 140% with a net energy consumption. Former configurations imply sufficient angle coverage and are obtained when an advantageous balance is achieved between the Int. J. Turbomach. Propuls. Power 2022, 7, 10 14 of 15 The analysis of the efficiency of any flow control system devoted to reducing the energy consumption is always an interesting step, as it allows to state whether such a system “pays its place” in a complex industrial machine, such as an aircraft, where every gram counts. This study is in that sense enlightening, as it allows imagining various uses of the different tested configurations. Firstly, the highest SMI points, despite their net energy cost, could be devoted to critical situations where the safety and level of engine performance need to be maintained. One thinks of take-off and landing cases and of combat situations at high angles of attack to cope with inlet distortion effects. On the contrary, even with a lesser SMI, a configuration exhibiting a net positive energy gain could be continuously used to improve the overall efficiency of the engines and thus reduce their environmental impact. This study constitutes an encouraging proof-of-concept that active flow control is viable from an energy point of view at the laboratory scale, that is, on a simplified, low- speed and single-stage test rig using low-TRL actuators. The next step is now to reproduce and validate the concept on a real engine using industrial-grade actuators. This constitutes the next step of the current project and will be published in the near future. Author Contributions: Conceptualisation, P.J. and A.D.; methodology, P.J., O.R. and A.D.; software, G.M. and P.J.; validation, J.M.M., P.J. and A.D.; formal analysis, J.M.M., P.J. and A.D.; investigation, J.M.M. and G.M.; resources, O.R. and G.M.; writing—original draft preparation, P.J., A.D. and J.M.M.; writing—review and editing, O.R., J.M.M., P.J. and A.D.; supervision, O.R. and A.D.; project administration, O.R. and A.D.; funding acquisition, A.D., P.J. and O.R. All authors have read and agreed to the published version of the manuscript. Funding: This project (ACONIT) has received funding from the Clean Sky 2 Joint Undertaking under the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 886352. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data available on request. Acknowledgments: We are grateful to the technical staff at the Arts et Métiers Institute of Technology in Lille, France, especially Patrick Olivier and Claude Lamacz, for technical support and expertise provided during the test campaign. Conflicts of Interest: The authors declare no conflict of interest. Nomenclature a Absolute blowing flow angle ( ) jet AFC Active flow control b Relative blowing flow angle ( ) jet SM Stall margin V Absolute jet speed (m/s) jet SMI Stall margin improvement Q Injected mass flow rate (kg/s) inj PG Power gain N Number of injectors PC Power cost DC Duty cycle PB Power balance f Driving frequency P Pressure (Pa) N Number of injectors activated V Velocity (m/s) Subscripts r Density (kg/m ) 1 2 Stage inlet outlet q Mass flow rate (kg/s) N Nominal operating point W Rotor rotational velocity (rpm) S Last stable operating point U Rotor tip speed (m/s) B Baseline or case without control P Pressure ratio C Controlled case F Flow coefficient x Axial quantity Y Pressure rise coefficient tip Quantity at blade tip x Axial position (mm) mid Quantity at mid-span Cx Axial chord length (mm) t s Total-to-static quantity Int. J. Turbomach. Propuls. Power 2022, 7, 10 15 of 15 References 1. Moubogha, J.M.; Margalida, G.; Joseph, P.; Roussette, O.; Dazin, A. Surge Margin Improvement by Continuous and Pulsed Tip injection. In Proceedings of the 14th European Conference on Turbomachinery Fluid Dynamics & Thermodynamics, ETC2020-639, Gdansk, Poland, 12–16 April 2021. 2. Day, I.J. Stall, surge, and 75 years of research. J. Turbomach. 2016, 138, 011001. [CrossRef] 3. Hathaway, M.D. Passive Endwall Treatments for Enhancing Stability; NASA Report No. TM-2007-214409; July 2007. Available online: ntrs.nasa.gov (accessed on 13 October 2021). 4. Li, J.; Du, J.; Nie, C.; Zhang, H. Review of Tip Air Injection to Improve Stall Margin in Axial Compressors. Prog. Aerosp. Sci. 2019, 106, 15–31. [CrossRef] 5. Hewkin-Smith, M.; Pullan, G.; Grimshaw, S.D.; Greitzer, E.M.; Spakovszky, Z.S. The Role of Tip Leakage Flow in Spike-Type Rotating Stall Inception. J. Turbomach. 2019, 141, 061010. [CrossRef] 6. Pullan, G.; Young, A.M.; Day, I.J.; Greitzer, E.M.; Spakovszky, Z.S. Origins and Structure of Spike-Type Rotating Stall. J. Turbomach. 2015, 137, 051007. [CrossRef] 7. Kefalakis, M.; Papailiou, K.D. Active Flow Control for Increasing the Surge Margin of an Axial Flow Compressor. Turbo Expo Power Land Sea Air 2006, 424, 101–111. 8. Stößel, M.; Bindl, S.; Niehuis, R. Ejector Tip Injection for Active Compressor Stabilization. Turbo Expo Power Land Sea Air Am. Soc. Mech. Eng. 2014, 45608, V02AT37A004. 9. Suder, K.L.; Hathaway, M.D.; Thorp, S.A.; Strazisar, A.J.; Bright, M.B. Compressor Stability Enhancement Using Discrete Tip Injection. J. Turbomach. 2001, 123, 14–23. [CrossRef] 10. Kern, F.; Brehm, S.; Niehuis, R. Ejector Tip Injection System for Active Aerodynamic Compressor Stabilization Part I: Design and Experiment. In Proceedings of the 12th European Conference on Turbomachinery Fluid Dynamics & Thermodynamics, ETC2017-244, Stockholm, Sweden, 3–7 April 2017. 11. Strazisar, A.J.; Bright, M.M.; Thorp, S.; Culley, D.E.; Suder, K.L. Compressor Stall Control through Endwall Recirculation. Turbo Expo Power Land Sea Air 2004, 41707, 655–667. 12. D’Andrea, R.; Behnken, R.L.; Murray, R.M. Rotating Stall Control of an Axial Flow Compressor Using Pulsed Air Injection. J. Turbomach. 1997, 119, 742–752. [CrossRef] 13. Khaleghi, H.; Teixeira, J.A.; Tousi, A.M.; Boroomand, M. Parametric Study of Injection Angle Effects on Stability Enhancement of Transonic Axial Compressors. J. Propuls. Power 2008, 24, 1100–1107. [CrossRef] 14. Nie, C.; Xu, G.; Cheng, X.; Chen, J. Micro Air Injection and Its Unsteady Response in a Low-Speed Axial Compressor. J. Turbomach. 2002, 124, 572–579. [CrossRef] 15. Greenblatt, D.; Wygnanski, I. The Control of Flow Separation by Periodic Excitation. Prog. Aerosp. Sci. 2000, 36, 487–545. [CrossRef] 16. Margalida, G. Analyse Expérimentale et Contrôle Actif de l’écoulement Dans Un Compresseur Axial Mono-Étagé Durant La Transition Vers Le Décrochage Tournant. Ph.D. Thesis, Ecole Nationale Supérieure d’Arts et Métiers, Lille, France, 2019. 17. Margalida, G.; Joseph, P.; Roussette, O.; Dazin, A. Active Flow Control in an Axial Compressor for Stability Improvement: On the Effect of Flow Control on Stall Inception. Exp. Fluids 2021, 62, 1–13. [CrossRef] 18. Dazin, A.; Joseph, P.; Romano, F.; Gallas, Q.; Marty, J.; Aigouy, G.; Stô el, M.; Niehuis, R. The ACONIT Project: An Innovative Design Approach of Active Flow Control for Surge Prevention in Gas Turbines. IOP Conf. Ser. Mater. Sci. Eng. 2021, 1024, 012068. [CrossRef] 19. Margalida, G.; Joseph, P.; Roussette, O.; Dazin, A. Comparison and Sensibility Analysis of Warning Parameters for Rotating Stall Detection in an Axial Compressor. Int. J. Turbomach. Propuls. Power 2020, 5, 16. [CrossRef] 20. Vegliò, M.; Dazin, A.; Bois, G.; Roussette, O. Unsteady Pressure Measurements of Spike Type Inception in Axial Compressor: Time Frequency Analysis and Averaging Procedure. In Proceedings of the 11th European Conference on Turbomachinery Fluid Dynamics and Thermodynamics, ETC2015-209, Madrid, Spain, 23–27 March 2015. 21. Weigl, H.J.; Paduano, J.D.; Frechette, L.G.; Epstein, A.H.; Greitzer, E.M.; Bright, M.M.; Strazisar, A.J. Active Stabilization of Rotating Stall and Surge in a Transonic Single Stage Axial Compressor. In Volume 4: Manufacturing Materials and Metallurgy; Ceramics; Structures and Dynamics; Controls, Diagnostics and Instrumentation; Education; IGTI Scholar Award; American Society of Mechanical Engineers: Orlando, FL, USA, 1997. 22. Deveaux, B.; Fournis, C.; Brion, V.; Marty, J.; Dazin, A. Experimental Analysis and Modeling of the Losses in the Tip Leakage Flow of an Isolated, Non-Rotating Blade Setup. Exp. Fluids 2020, 61, 126. [CrossRef] http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Turbomachinery, Propulsion and Power Multidisciplinary Digital Publishing Institute

Stall Margin Improvement in an Axial Compressor by Continuous and Pulsed Tip Injection

Download PDF
15 pages

Loading next page...
 
/lp/multidisciplinary-digital-publishing-institute/stall-margin-improvement-in-an-axial-compressor-by-continuous-and-1Y1VVI6QVh
Publisher
Multidisciplinary Digital Publishing Institute
Copyright
© 1996-2022 MDPI (Basel, Switzerland) unless otherwise stated Disclaimer The statements, opinions and data contained in the journals are solely those of the individual authors and contributors and not of the publisher and the editor(s). MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Terms and Conditions Privacy Policy
ISSN
2504-186X
DOI
10.3390/ijtpp7010010
Publisher site
See Article on Publisher Site

Abstract

International Journal of Turbomachinery Propulsion and Power Article Stall Margin Improvement in an Axial Compressor by Continuous and Pulsed Tip Injection Joseph Moubogha Moubogha, Gabriel Margalida, Pierric Joseph, Olivier Roussette and Antoine Dazin * UMR 9014—LMFL—Laboratoire de Mécanique des Fluides de Lille—Kampé de Fériet, Arts et Métiers Institute of Technology, Centrale Lille, University Lille, CNRS, ONERA, F-59000 Lille, France; joseph.moubogha-moubogha@ensam.eu (J.M.M.); gabriel.margalida@ensam.eu (G.M.); pierric.joseph@ensam.eu (P.J.); olivier.roussette@ensam.eu (O.R.) * Correspondence: antoine.dazin@ensam.eu † This manuscript is an extended version of our meeting paper published in the Proceedings of the 14th European Turbomachinery Conference, Gdansk, Poland, 12–16 April 2021. Abstract: Stall and surge are strong limitations in the operating range of compressors and thus one of the major limits of jet engine performance. A promising way to push the stability limit of compression machines is to inject a small amount of flow at the blade tip to alter the physical mechanism responsible for stall onset. This study focuses on the experimental performance of such a system. To do so, an axial compressor test bench was equipped with 40 actuators connected to an auxiliary pressurised air supply system. They were able to generate high-speed jet blowing just at the tip of the rotor blades. The opening of each actuator was controlled by an electromagnetic valve. This allowed generating continuous or pulsed jets with frequencies up to 500 Hz at different duty cycles. The performance of the control system was investigated for various control strategies, where the injected flow rate, the injection angle, the number of injectors, the jet frequency and the duty cycle were systematically varied. This paper is concluded by a study of the energy balance of the system for various configurations. To the best of the authors’ knowledge, this constitutes a rarely seen analysis in the literature. Citation: Moubogha, J.M.; Margalida, Keywords: axial compressor; active flow control; stall margin improvement; tip blowing; energy cost G.; Joseph, P.; Roussette, O.; Dazin, A. Stall Margin Improvement in an Axial Compressor by Continuous and Pulsed Tip Injection. Int. J. 1. Introduction Turbomach. Propuls. Power 2022, 7, 10. https://doi.org/10.3390/ijtpp7010010 The problem of stall and surge instabilities developing in axial compressors is nearly as old as the first gas turbine [1,2], and the loss of stability of axial compressors at a Received: 14 October 2021 high-pressure ratio is still nowadays an important limitation of the global performance of Accepted: 6 March 2022 modern aero engines, as the occurrence of these phenomena can cause dramatic events. It Published: 16 March 2022 thus leads engine manufacturers to apply a consequent security margin (the so-called stall Publisher’s Note: MDPI stays neutral margin (SM)) to keep the compressor far from its stability limit. This deprives the machines with regard to jurisdictional claims in of their higher-pressure ratio and higher-efficiency operating ranges, which penalises the published maps and institutional affil- global performance of the engine. Stall margin improvements (SMI) can be achieved by iations. various methods: a large number of works in the literature are devoted to passive control techniques, such as casing treatments [3]. These methods involve permanent modifications of the casing that cannot cope with changes in flow conditions. Using active flow control systems, which can easily be turned off and on, is thus an attractive alternative, and Copyright: © 2022 by the authors. many research works have been carried out concerning this kind of techniques and were Licensee MDPI, Basel, Switzerland. summarised recently by Li et al. [4]. One should keep in mind, however, that such methods This article is an open access article come with some technical drawbacks in terms of complexity, added weight and, potentially, distributed under the terms and conditions of the Creative Commons reliability issues. According to Day’s review [2], the most common stall inception in the Attribution (CC BY-NC-ND) license modern engines is the spike, which originates from phenomena occurring at the blade (https://creativecommons.org/ tip [5,6]. The most efficient way to control these phenomena is, thus, to blow a high- licenses/by-nc-nd/4.0/). momentum jet at the blade tip, as demonstrated by several authors [7–9], to decrease the Int. J. Turbomach. Propuls. Power 2022, 7, 10. https://doi.org/10.3390/ijtpp7010010 https://www.mdpi.com/journal/ijtpp Int. J. Turbomach. Propuls. Power 2022, 7, 10 2 of 15 loading and/or act on the dynamic of the tip gap vortex. Among all the literature dedicated to this subject, there is a consensus about some specific characteristics of the control system that must be used to obtain a significant effect in terms of SMI: The jet has to be as close as possible to the casing wall; consequently, many authors have successfully used the Coandă effect to get a wall-attached jet [10,11]. A high momentum (or high velocity) and good angular coverage [9] are needed. Nevertheless, some points remain unclear. As stated by Li et al. [4], there is no consensus about the effect of the yaw angle (positive angle values are given by the rotor rotation direction). Whereas the first paper [12] originally claimed that a positive yaw angle is better, other authors have demonstrated the opposite [13,14]. In addition, to the best of the authors’ knowledge, the balance between the energy cost of blowing and the positive effect on machine performance has never been clearly established. It is therefore difficult to answer the question of whether such a system “pays its place” in a real engine. Finally, the potential interest of pulsed blowing has barely been addressed [7], whereas it has shown its efficiency in some other flow control applications, such as separation control [15]. In previous studies conducted in the laboratory by the same research team, the flow mechanisms involved during the onset of stall with and without active flow control have notably been analysed. The control system effectively succeeded in increasing the operating range of the compressor by neutralising the spike mechanisms and by moving the last stability point close to the maximum of the performance curve. For some control config- urations, a low-frequency phenomenon appears in pressure measurements, suggesting a transition from spike-type stall to modal-type stall inception. In addition, some important effects, such as the injected momentum, have been investigated [16,17]. This paper aims to complete this previous study by addressing the above-mentioned questions concerning the yaw angle and energy balance of the system using pulsed ac- tuation. This study relies on an experimental parametric study conducted on the same single-stage axial compressor test bench equipped with a modular flow injection sys- tem [17]. After a description of the experimental set-up, the paper focuses on the effect of the blowing angle at several rotation speeds to clearly point out the effect of blowing on the rotating frame. The paper is then dedicated to an estimation of the energy costs and savings of the control system to try to evidence the most interesting blowing strategies in pulsed or continuous blowing. This work is the first part of the EU-funded Horizon 2020 research project ACONIT [18], which aims at designing, manufacturing and testing actuators for flow control for implantation in an aircraft engine. 2. Experimental Set-Up The support of the experiments performed in this research work is the low-speed, single-stage axial compressor CME2 located at the Arts et Métiers Institute of Technology in Lille (France). This compressor is a subsonic machine comparable to a stage of a high- pressure component of an aero engine [2]. Initially designed as a convenient tool to study rotating stall [19,20], this specific test bench has been equipped in the recent years with an active flow control (AFC) system relying on magnetic valves to produce pulsed jets [16]. The compressor itself, depicted in Figure 1a and whose characteristics are listed in Table 1, is operated at rotational speeds ranging from 3200 rpm to 4500 rpm. A typical per- formance curve is provided in Figure 1b, plotting the total-to-static pressure rise coefficient Y = DP /0.5rU as a function of the flow coefficient F = V /U . During the tests, ts ts x tip tip the mass flow is varied using a throttling valve located downstream of the compressor stage (see Figure 1a). In this paper, stall is triggered by continuously closing the throttling valve up to the unstable part of the performance curve. Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 3 of 16 Int. J. Turbomach. Propuls. Power 2022, 7, 10 3 of 15 (a) (b) Figure 1. CME2 test rig: (a) schematic description; (b) typical performance curve of the compressor Figure 1. CME2 test rig: (a) schematic description; (b) typical performance curve of the compressor at 3200 rpm. at 3200 rpm. Table 1. Compressor parameters at 3200 rpm. Table 1. Compressor parameters at 3200 rpm. Parameters Parameters Value Value UUni nits ts 1 −1 Design mass flow rate 5.3 kg· s Design mass flow rate 5.3 kgs Design axial velocity, LE * 43 ms −1 Design axial velocity, LE * 43 m· s Rotor blade number 30 Rotor blade number 30 Stator blade number 40 Stator bl Casing diameter ade number 550 40 mm Hub-tip ratio, LE 0.75 Casing diameter 550 mm Rotor tip chord 84 mm Hub-tip ratio, LE 0.75 Rotor tip stagger angle 54 Rotor Rotor t tipip cho gap rd 0.5 84 mm mm Rotor tip speed 94 ms Rotor tip stagger angle 54 ° * LE, leading edge. Rotor tip gap 0.5 mm −1 Rotor tip speed 94 m· s The performance of the compressor is evaluated using two differential pressure sensors * LE, leading edge. located on the test rig. The first sensor measures the difference between the total pressure recorded in the plenum chamber and a mean static pressure measured at the end of the converging pipe located just downstream of the plenum chamber. This value allows The performance of the compressor is evaluated using two differential pressure sen- capturing the dynamic pressure at the compressor inlet, and then the flow rate. The second sors located on the test rig. The first sensor measures the difference between the total pres- sensor is used to evaluate the stage performance by measuring the static pressure in front sure recorded in the plenum chamber and a mean static pressure measured at the end of of the rotor and downward of the stator. The precision of these measurements has been the converging pipe located just downstream of the plenum chamber. This value allows evaluated to 0.012 kgs and 1.5 Pa for the flow rate and the total-to-static pressure capturing the dynamic pressure at the compressor inlet, and then the flow rate. The sec- rise, respectively [16]. ond sensor is used to evaluate the stage performance by measuring the static pressure in The control system [17] (see the overall description in Figure 2) consists of 20 injection front of the rotor and downward of the stator. The precision of these measurements has blocks, each one counting two injectors. This configuration was selected because of the −1 been evaluated to ±0.012 kg· s and ±1.5 Pa for the flow rate and the total-to-static pressure space constraint caused by the curvature of the casing. A solenoid valve (Matrix MX821), whose driving frequency can be set between 0 (continuous) and 500 Hz, with a supply rise, respectively [16]. pressure of up to 8 bar, feeds each injector. Each injector can be then operated independently The control system [17] (see the overall description in Figure 2) consists of 20 injection 1 2 and can produce a jet speed of up to 200 ms through a 10  0.5 mm slot. Accordingly, blocks, each one counting two injectors. This configuration was selected because of the the injected mass flow can be set from 0 to approximately 2.5% of the main flow rate of the space constraint caused by the curvature of the casing. A solenoid valve (Matrix MX821), stage either by changing the supply pressure or by changing the duty cycle (DC) of the whose driving frequency can be set between 0 (continuous) and 500 Hz, with a supply solenoid valves when pulsed injection is used. The duty cycle is defined as the blowing pressure of up to 8 bar, feeds each injector. Each injector can be then operated inde- time duration divided by the total period duration (i.e., the sum of blowing and no blowing −1 pendently and can produce a jet speed of up to 200 m· s through a 10 × 0.5 mm² slot. time). All solenoid valves are driven by the same command signal, and manufacturer Accordingly, the injected mass flow can be set from 0 to approximately 2.5% of the main data indicate a response time lower than 1 ms. As stated in the literature, blowing is most effective in front of the rotor leading edge [14], and the critical area is located at the tip, flow rate of the stage either by changing the supply pressure or by changing the duty cycle close to the casing [4]. Consequently, actuators are located 10 mm upstream of the rotor (DC) of the solenoid valves when pulsed injection is used. The duty cycle is defined as the (x = 20%  Cx), and injectors are shaped using the Coandă effect to blow along the casing blowing time duration divided by the total period duration (i.e., the sum of blowing and no blowing time). All solenoid valves are driven by the same command signal, and man- ufacturer data indicate a response time lower than 1 ms. As stated in the literature, blow- ing is most effective in front of the rotor leading edge [14], and the critical area is located at the tip, close to the casing [4]. Consequently, actuators are located 10 mm upstream of the rotor (x = −20% ∙ Cx), and injectors are shaped using the Coandă effect to blow along Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 4 of 16 Int. J. Turbomach. Propuls. Power 2022, 7, 10 4 of 15 the casing right in the tip gap [11]. Injectors can also be rotated along their axis to vary the right in the tip gap [11]. Injectors can also be rotated along their axis to vary the yaw angle yaw angle of injection with an angular step of 15°. of injection with an angular step of 15 . Figure 2. Active flow control system description. Figure 2. Active flow control system description. The experimental protocol is extensively described in a previous paper [17], but key The experimental protocol is extensively described in a previous paper [17], but key information is reported here for the sake of completeness. In a typical stall test, the working information is reported here for the sake of completeness. In a typical stall test, the work- point of the compressor is moved along the performance curve using the controlled closing ing point of the compressor is moved along the performance curve using the controlled of a throttling valve. Stall is easily perceptible through an abrupt drop in the mass flow and closing of a throttling valve. Stall is easily perceptible through an abrupt drop in the mass pressure ratio. The baseline curve is compared with a controlled curve, which is obtained flow with and contr pres ol sur activated e ratio. for Ththe e bas entir eline e working curve is range compare of the d with compr a con essor troll . Ther ed cu erve, is thus which no is ob issue tained with with activation control delay activated of the fo active r the flow entire contr work ol in system. g range of the compressor. There is A typical example of results of the flow control system is proposed in Figure 3, where thus no issue with activation delay of the active flow control system. the total-to-static pressure rise coefficient is plotted as a function of the flow coefficient. A typical example of results of the flow control system is proposed in Figure 3, where For each control case, the actual value of the global mass flow rate injected is specified the total-to-static pressure rise coefficient is plotted as a function of the flow coefficient. along with the duty cycle (in pulsed blowing). Here and subsequently, the baseline results For each control case, the actual value of the global mass flow rate injected is specified correspond to the performance of the compressor without control and the injected mass along with the duty cycle (in pulsed blowing). Here and subsequently, the baseline results flow rate Q is expressed in scaled form, Q , as a percentage of the compressor flow rate inj inj correspond to the performance of the compressor without control and the injected mass at the last stable operating point before stall without control, q . It is obvious from these flow rate 𝑄 is expressed in scaled form, 𝑄 , as a percentage of the compressor flow results that the blowing acts positively on the stable operating range of the compressor. rate at the last stable operating point before stall without control, 𝑞 . It is obvious from To evaluate the effect of the control system on the performance curve of the machine, these results that the blowing acts positively on the stable operating range of the compres- the definition of the stall margin improvement given by Weigl et al. [21] is used and sor. calculated using the following equations: To evaluate the effect of the control system on the performance curve of the machine, q P N S the definition of the stall margin improvement given by Weigl et al. [21] is used and cal- SM = (  1)  100 and (1) q P culated using the following equationsS: N 𝑞 𝛱 𝑁 𝑆 SM SM C B = ( × − 1) × 100 and (1) SMI =  100, (2) 𝑞 𝛱 𝑆 𝑁 SM 𝑆𝑀 − with q and P, respectively, being the flow rate 𝐶 and 𝐵 the pressure ratio. Please note that in 𝑆𝑀𝐼 = × 100, (2) 𝑆𝑀 Equations (1) and (2), along with Figure 4, the subscripts N and S refer to quantities at the nominal operating point and at the last stable operating point, respectively, before stall with 𝑞 and 𝛱 , respectively, being the flow rate and the pressure ratio. Please note that in (i.e., the operating point with the lowest flow rate before stall onset). Similarly, subscript B Equations (1) and (2), along with Figure 4, the subscripts N and S refer to quantities at the refers to the baseline case, without control, and C to the controlled case. nominal operating point and at the last stable operating point, respectively, before stall (i.e., the operating point with the lowest flow rate before stall onset). Similarly, subscript B refers to the baseline case, without control, and C to the controlled case. 𝑆𝑀 𝑆𝑀 𝑖𝑛𝑗 𝑖𝑛𝑗 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 5 of 16 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 5 of 16 Int. J. Turbomach. Propuls. Power 2022, 7, 10 5 of 15 Figure 3. Performance curves obtained using the pulsed jet control system with various duty cycle and global injected mass flow rate values (actuation frequency 𝑓 = 100 Hz, 20 injectors activated, absolute blowing flow angle 𝛼 = 0° and rotation speed 𝛺 = 3200 rpm). The duty cycle 𝐷𝐶 = 𝑗𝑒𝑡 Figure 3. Performance curves obtained using the pulsed jet control system with various duty cycle Figure 3. Performance curves obtained using the pulsed jet control system with various duty cycle 1 corresponds to continuous blowing. and global injected mass flow rate values (actuation frequency f = 100 Hz, 20 injectors activated, and global injected mass flow rate values (actuation frequency 𝑓 = 100 Hz, 20 injectors activated, absolute blowing flow angle a = 0 and rotation speed W = 3200 rpm). The duty cycle DC = 1 jet absolute blowing flow angle 𝛼 = 0° and rotation speed 𝛺 = 3200 rpm). The duty cycle 𝐷𝐶 = 𝑗𝑒𝑡 corresponds to continuous blowing. 1 corresponds to continuous blowing. Figure 4. Schematic description of the SMI (adapted from [21]). Figure 4. Schematic description of the SMI (adapted from [21]). 3. Effect of the Injection Yaw Angle 3. Effect of the Injection Yaw Angle To highlight the effect of the injection yaw angle, a first series of experiments were Figure 4. Schematic description of the SMI (adapted from [21]). To highlight the effect of the injection yaw angle, a first series of experiments were conducted at 3200 rpm. The absolute blowing flow angle , the velocity (or the injected jet conducted at 3200 rpm. The absolute blowing flow angle jet, the velocity (or the injected mass flow rate) and the number of injectors activated were varied. Absolute and relative 3. Effect of the Injection Yaw Angle mass flow rate) and the number of injectors activated were varied. Absolute and relative blowing flow angles are defined in Figure 5. Positive absolute blowing angle values are To highlight the effect of the injection yaw angle, a first series of experiments were blowing flow angles are defined in Figure 5. Positive absolute blowing angle values are given by the rotor rotation direction in the study; therefore, the absolute blowing angle is conducted at 3200 rpm. The absolute blowing flow angle jet, the velocity (or the injected given by the rotor rotation direction in the study; therefore, the absolute blowing angle is considered negative when the blowing direction is opposite to the rotor rotation direction. mass flow rate) and the number of injectors activated were varied. Absolute and relative considered negative when the blowing direction is opposite to the rotor rotation direction. Several basic test results of the active flow control in continuous blowing at 3200 rpm blowing flow angles are defined in Figure 5. Positive absolute blowing angle values are with 40 injectors activated and various absolute blowing flow angle and injected mass given by the rotor rotation direction in the study; therefore, the absolute blowing angle is flow rate values are detailed and presented in Figure 6. The total-to-static pressure rise considered negative when the blowing direction is opposite to the rotor rotation direction. coefficient is plotted as a function of the flow coefficient for the six absolute blowing flow angles investigated: 30 , 15 , 0 , 15 , 30 and 45 . It can be seen that as already observed in Figure 3, for each absolute injection angle, the blowing extends the operating range of the compressor and simultaneously increases the compressor performance. It Int. J. Turbomach. Propuls. Power 2022, 7, 10 6 of 15 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 6 of 16 is also rather obvious that higher benefits are obtained for negative absolute blowing flow angles. Figure 5. Definition and illustration of the absolute and relative blowing flow angles. Positive angle Figure 5. Definition and illustration of the absolute and relative blowing flow angles. Positive angle values are given by the rotor rotation direction. values are given by the rotor rotation direction. To investigate the effect of the blowing flow angle, the stall margin improvement was Several basic test results of the active flow control in continuous blowing at 3200 rpm calculated for each experiment reported in Figure 6, and the results are reported in Figure 7. with 40 injectors activated and various absolute blowing flow angle and injected mass To make the comparison of the different sets of parameters easier, the absolute blowing flow rate values are detailed and presented in Figure 6. The total-to-static pressure rise velocity V was scaled by the rotor tip speed. The velocity of the jet, V , was obtained jet jet coefficient is plotted as a function of the flow coefficient for the six absolute blowing flow by hot-wire measurements conducted on the injectors in a dedicated fluidic actuator test angles investigated: 30°, 15°, 0°, −15°, −30° and −45°. It can be seen that as already observed bench [16]. In addition, the results are presented as a function of the relative blowing flow in Figure 3, for each absolute injection angle, the blowing extends the operating range of angle b , i.e., the jet flow angle seen by the blade, which was derived from the absolute jet the compressor and simultaneously increases the compressor performance. It is also ra- velocity and angle of the jet, and the rotor tip speed using the velocity triangle (see Figure 5). For the majority of the tested jet velocities, the stall margin improvement presents ther obvious that higher benefits are obtained for negative absolute blowing flow angles. a clear and monotonic increase as the relative blowing angle decreases and reaches in most cases a maximum value for relative flow angles in the range between 60 and 70 . It then appears to remain relatively constant or to decrease slightly at lower relative flow angles. The effect of the rotation velocity was investigated and is presented in Figure 8. The stall margin improvement is plotted as a function of the relative blowing angle for two main rotor rotational velocities: 3200 and 4500 rpm. After scaling the absolute blowing velocity by the appropriate rotor tip speed, results were fairly close for the two different rotational speeds. The monotonic increase in the stall margin improvement with the decreasing of the relative blowing angle and the plateau reached around 60 and 70 is once again clearly apparent. It has to be pointed out that the inlet blade angle at the tip (depicted in Figure 8 by the vertical blue dashed line) of this compressor is 65 . It thus appears that the highest effect of the stall margin improvement is obtained for relative blowing angles close to the inlet blade angle, which is consistent with some previous experimental observations [7]. This can also be easily explained as this blowing angle is certainly most suitable for decreasing the blade loading at the tip and thus preventing the mechanism occurring at the tip and leading to rotating stall. Int. J. Turbomach. Propuls. Power 2022, 7, 10 7 of 15 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 7 of 16 (𝐚 ) 𝛼 = 30° (𝐛 ) 𝛼 = 15° 𝑗𝑒𝑡 𝑗𝑒𝑡 ( ) ( ) 𝐜 𝛼 = 00° 𝐝 𝛼 = −15° 𝑗𝑒𝑡 𝑗𝑒𝑡 (𝐞 ) 𝛼 = −30° (𝐟 ) 𝛼 = −45° 𝑗𝑒𝑡 𝑗𝑒𝑡 Figure Figure 6. 6. Performance Performance curves curves obtained obtained using using continuous continuous blowing blowing with with various various absolute absoluteblowing blowing angles of: (a) 30°, (b) 15°, (c) 0°, (d) −15°, (e) −30° and (f) −45° and various injected mass flow rate angles of: (a) 30 , (b) 15 , (c) 0 , (d) 15 , (e) 30 and (f) 45 and various injected mass flow rate values (40 injectors activated and rotation speed 𝛺 = 3200 rpm). values (40 injectors activated and rotation speed W = 3200 rpm). Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 8 of 16 To investigate the effect of the blowing flow angle, the stall margin improvement was calculated for each experiment reported in Figure 6, and the results are reported in Figure 7. To make the comparison of the different sets of parameters easier, the absolute blowing velocity 𝑉 was scaled by the rotor tip speed. The velocity of the jet, 𝑉 , was obtained 𝑗𝑒𝑡 𝑗𝑒𝑡 by hot-wire measurements conducted on the injectors in a dedicated fluidic actuator test bench [16]. In addition, the results are presented as a function of the relative blowing flow angle jet, i.e., the jet flow angle seen by the blade, which was derived from the absolute velocity and angle of the jet, and the rotor tip speed using the velocity triangle (see Figure Int. J. Turbomach. Propuls. Power 2022, 7, 10 8 of 15 5). Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 9 of 16 Figure 7. Effect of the relative blowing flow angle with 40 injectors activated for a rotation speed of Figure 7. Effect of the relative blowing flow angle with 40 injectors activated for a rotation speed of 3200 rpm. 3200 rpm. For the majority of the tested jet velocities, the stall margin improvement presents a clear and monotonic increase as the relative blowing angle decreases and reaches in most cases a maximum value for relative flow angles in the range between −60° and −70°. It then appears to remain relatively constant or to decrease slightly at lower relative flow angles. The effect of the rotation velocity was investigated and is presented in Figure 8. The stall margin improvement is plotted as a function of the relative blowing angle for two main rotor rotational velocities: 3200 and 4500 rpm. After scaling the absolute blowing velocity by the appropriate rotor tip speed, results were fairly close for the two different rotational speeds. The monotonic increase in the stall margin improvement with the de- creasing of the relative blowing angle and the plateau reached around −60° and −70° is once again clearly apparent. It has to be pointed out that the inlet blade angle at the tip (depicted in Figure 8 by the vertical blue dashed line) of this compressor is −65°. It thus appears that the highest effect of the stall margin improvement is obtained for relative blowing angles close to the inlet blade angle, which is consistent with some previous ex- perimental observations [7]. Figure 8. Effect of the relative blowing flow angle with 40 injectors activated at different rotating speeds. Figure 8. Effect of the relative blowing flow angle with 40 injectors activated at different rotating speeds. 4. Energy Balance The main goal of the control system is to improve the compressor stall margin. An This can also be easily explained as this blowing angle is certainly most suitable for additional benefit is the improvement in the pressure rise provided by the compressor decreasing the blade loading at the tip and thus preventing the mechanism occurring at stage (and thus a gain in the energy provided by the compressor to the flow). Nevertheless, the tip and leading to rotating stall. the generation of jets involves an energy cost. The power balance, introduced next, can be seen as the net benefit (energy gain – energy cost) of the control system. So, if it is negative, 4. Energy Balance then the use of the control system costs more than it brings in. The main goal of the control system is to improve the compressor stall margin. An The power balance (PB) of the control system is thus evaluated by subtracting the additional benefit is the improvement in the pressure rise provided by the compressor power cost (PC) of the blowing to the associated power gain (PG), as defined below: stage (and thus a gain in the energy provided by the compressor to the flow). Neverthe- less, the generation of jets involves an energy cost. The power balance, introduced next, PB = PG PC. (3) can be seen as the net benefit (energy gain – energy cost) of the control system. So, if it is negative, then the use of the control system costs more than it brings in. Regarding the cost, it is the power consumed by the blowing system. In this case, weTh use e po a wer screw balance compr (PB essor ) of to thpr e con essur tro ise l sys the tem air isand thus solenoid evaluated valves by subtract to carry ing out the the populsed wer cost blowing. (PC) of the blo It is ther wing to th efore possible e associ to ated estimate power g the ain cost (PG of ), as de the injection fined below: through the electrical power consumed by all these elements. However, this includes many other 𝑃𝐵 = 𝑃𝐺 − 𝑃𝐶 . (3) factors that are not of direct interest, such as the choice of the solenoid valve or the way in which compressed air is generated, which can be subsequently improved and which will Regarding the cost, it is the power consumed by the blowing system. In this case, we use a screw compressor to pressurise the air and solenoid valves to carry out the pulsed blowing. It is therefore possible to estimate the cost of the injection through the electrical power consumed by all these elements. However, this includes many other factors that are not of direct interest, such as the choice of the solenoid valve or the way in which compressed air is generated, which can be subsequently improved and which will un- doubtedly be different from the final solution embedded in an engine. Consequently, the power consumed by the injection system at the last level is estimated by evaluating the aeraulic power added by the jets. Please note that in this study, the temperature of the injected air was close to the ambient one, as several buffer tanks (a large 500 L tank fol- lowed by two smaller 15 L ones) are present in the pressured air supply system. The power cost (PC) of the blowing is defined as the kinetic power added to the flow by the jets: 𝑃𝐶 = 𝑄 𝑉 , (4) 𝑗𝑒𝑡 where 𝑄 is the global injected mass flow rate and 𝑉 the mean jet velocity at the ac- 𝑗𝑒𝑡 tuator nozzle. 𝑖𝑛𝑗 𝑖𝑛𝑗 Int. J. Turbomach. Propuls. Power 2022, 7, 10 9 of 15 undoubtedly be different from the final solution embedded in an engine. Consequently, the power consumed by the injection system at the last level is estimated by evaluating the aeraulic power added by the jets. Please note that in this study, the temperature of the injected air was close to the ambient one, as several buffer tanks (a large 500 L tank followed by two smaller 15 L ones) are present in the pressured air supply system. The power cost (PC) of the blowing is defined as the kinetic power added to the flow by the jets: PC = Q V , (4) inj jet Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 10 of 16 where Q is the global injected mass flow rate and V the mean jet velocity at the inj jet actuator nozzle. The power gain (PG) is evaluated by comparing the performance of the compressor The power gain (PG) is evaluated by comparing the performance of the compressor with and without control at the flow rate corresponding to the last stable operating point with and without control at the flow rate corresponding to the last stable operating point without control (see the representation given in Figure 9). More precisely, it is defined as without control (see the representation given in Figure 9). More precisely, it is defined as the difference between the net power available in the fluid downstream of the compressor the difference between the net power available in the fluid downstream of the compressor with and without control; with and without control; " ! !# 2 22 𝑃 V 𝑉 𝑃 V 𝑉 P 2𝐶 P 2𝐵 2𝐶 2𝐵 2C 2C 2B 2B 𝑃𝐺 = 𝑞 [( + ) − ( + )], (5) PG = q + + , (5) 𝜌 2 𝜌 2 2𝐶 2𝐵 r 2 r 2 2C 2B where 𝑞 is the flow rate at the last stable operating point before stall without control; where q is the flow rate at the last stable operating point before stall without control; P the 𝑃 the static pressure; 𝑉 the velocity; 𝜌 the density; and the indexes 2, C and B, respec- static pressure; V the velocity; r the density; and the indexes 2, C and B, respectively, the tively, the stage outlet, the controlled configuration and the baseline (configuration with- stage outlet, the controlled configuration and the baseline (configuration without control). out control). Figure 9. Energy gain due to the active flow control system. Figure 9. Energy gain due to the active flow control system. 4.1. Continuous Mode 4.1. Continuous Mode Figure 10 presents the results of a series of tests performed on the compressor operating Figure 10 presents the results of a series of tests performed on the compressor oper- at 3200 rpm. All the experiments reported on the graph correspond to continuous blowing, ating at 3200 rpm. All the experiments repo  rted on the graph correspond to continuous with an absolute blowing angle a = 30 . This absolute blowing angle was retained jet blowing, with an absolute blowing angle jet = −30°. This absolute blowing angle was re- for the rest of the study as it allowed achieving the best SMI, according to the reasons tained for the rest of the study as it allowed achieving the best SMI, according to the rea- developed, due to the results in Figure 7. The effect of the number of injectors used (N) and sons developed, due to the results in Figure 7. The effect of the number of injectors used the global blowing flow rate was examined. The figure shows, for each tested configuration, (N) and the global blowing flow rate was examined. The figure shows, for each tested the SMI achieved compared to the power balance (PB) of the considered control strategy. configuration, the SMI achieved compared to the power balance (PB) of the considered On this graph, the most interesting points are on the top and the right of the figure, as they control strategy. On this graph, the most interesting points are on the top and the right of correspond to control parameters achieving significant SMI with a positive power balance. the figure, as they correspond to control parameters achieving significant SMI with a pos- itive power balance. Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 11 of 16 Int. J. Turbomach. Propuls. Power 2022, 7, 10 10 of 15 Figure 10. SMI and power balance of the control system. Continuous blowing with the absolute injec- Figure 10. SMI and power balance of the control system. Continuous blowing with the absolute tion angle a = 30 and the rotation speed W = 3200 rpm. N is the number of injectors activated. injection jang et le 𝛼 = −30° and the rotation speed 𝛺 = 3200 rpm. N is the number of injectors ac- 𝑗𝑒𝑡 tivated. Some remarkable points in the graph are highlighted with letters, from A to F. Points A, B and C correspond to strategies that allow reaching the greatest SMI. Nevertheless, Some remarkable points in the graph are highlighted with letters, from A to F. Points these configurations correspond to the maximum number of injectors with the maximum A, B and C correspond to strategies that allow reaching the greatest SMI. Nevertheless, flow rate (and velocity) per injector. Consequently, the energy cost is high, and the energy balance for this specific case is unfavourable. Points D, E and F correspond to interesting these configurations correspond to the maximum number of injectors with the maximum applicative configurations as they allow to obtain a fairly good SMI (from 55% for point F flow rate (and velocity) per injector. Consequently, the energy cost is high, and the energy to 80% for point D) with a positive power balance, which can reach 1.8% of the compressor balance for this specific case is unfavourable. Points D, E and F correspond to interesting nominal power. applicative configurations as they allow to obtain a fairly good SMI (from 55% for point F These most interesting points are all obtained for configurations with 30 to 40 injectors to 80% for point D) with a positive power balance, which can reach 1.8% of the compressor activated, which means that good angular coverage is necessary to reach a good compro- mise nom between inal pow theer. SMI and the positive power balance. This need for a sufficient angular cov- erage was also highlighted by Suder et al. [8] and more recently by Margalida et al. [16,17]. These most interesting points are all obtained for configurations with 30 to 40 injec- This parameter is determining for the increase in the SMI. It seems that this is also the case tors activated, which means that good angular coverage is necessary to reach a good com- for the energy balance. promise between the SMI and the positive power balance. This need for a sufficient angu- The second observation coming from Figure 10 is that whatever the number of injectors lar coverage was also highlighted by Suder et al. [8] and more recently by Margalida et al. used, when the flow rate starts to increase, both the SMI and the power balance increase, [16,17]. This parameter is determining for the increase in the SMI. It seems that this is also leading to the best configurations, such as points D, E or F. When the blowing flow rate continues the case f to or incr the energ ease, the y bal SMIance. continues to increase [14], whereas the power balance deteriorates rapidly. The second observation coming from Figure 10 is that whatever the number of injec- Figure 11 shows the evolutions of the power gain (a), the power cost (b) and the power tors used, when the flow rate starts to increase, both the SMI and the power balance in- balance (c) as a function of the injected flow rate. What can be clearly observed is that the crease, leading to the best configurations, such as points D, E or F. When the blowing flow power gain is low for a low injected flow rate (less than 1%) and that the power gain grows rate continues to increase, the SMI continues to increase [14], whereas the power balance almost linearly and more rapidly than the power cost, which evolves approximatively as deteriorates rapidly. the jet speed cubed (or the injected flow rate cubed). This evolution leads to a rapid growth of the cost Figure for values 11 sh higher ows th than e evo 1–1.5%, lutions leading of to the rapid power degradation gain (a), of the the power powe balance r cost (b) and the power balance (c) as a function of the injected flow rate. What can be clearly observed is that the power gain is low for a low injected flow rate (less than 1%) and that the power gain grows almost linearly and more rapidly than the power cost, which evolves approx- imatively as the jet speed cubed (or the injected flow rate cubed). This evolution leads to a rapid growth of the cost for values higher than 1–1.5%, leading to rapid degradation of the power balance (Figure 11c). What can also be noticed in Figure 11b is that for a given Int. J. Turbomach. Propuls. Power 2022, 7, 10 11 of 15 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 12 of 16 (Figure 11c). What can also be noticed in Figure 11b is that for a given injected flow rate, injected flow rate, the power cost is lower for the configuration with more injectors acti- the power cost is lower for the configuration with more injectors activated: in this case, the vated: in this case, the flow rate is distributed between more injectors, which leads to a flow rate is distributed between more injectors, which leads to a lower velocity per injector lower velocity per injector and a lower cost, according to Equation (4). and a lower cost, according to Equation (4). Figure 11. Evolution of the: (a) power gain, (b) power cost and (c) power balance as a function of the Figure 11. Evolution of the: (a) power gain, (b) power cost and (c) power balance as a function of injected flow rate. Continuous blowing with the absolute injection angle a = 30 and the rotation jet the injected flow rate. Continuous blowing with the absolute injection angle 𝛼 = −30° and the 𝑗𝑒𝑡 speed W = 3200 rpm. rotation speed 𝛺 = 3200 rpm. The fact that the power balance is positive for a large range of Q indicates that the inj The fact that the power balance is positive for a large range of 𝑄 indicates that the blowing leads also to an increase in the blade work at the tip and/or a decrease in the losses blowing leads also to an increase in the blade work at the tip and/or a decrease in the (the profile losses close to the tip, as the flow is realigned with the blade inlet angle and losses (the profile losses close to the tip, as the flow is realigned with the blade inlet angle certainly also to the losses associated with the secondary gap flows, as demonstrated in the and certainly also to the losses associated with the secondary gap flows, as demonstrated case of an isolated blade, for some blowing configurations [22]). in the case of an isolated blade, for some blowing configurations [22]). 4.2. Pulsed Mode Figure 12 presents a typical evolution of the stall margin improvement with the ac- tuation frequency, in pulsed mode. Note that in Figure 12, no result is presented for fre- quencies above 200 Hz, as beyond this value, the response time of the valve becomes sig- nificant compared with the blowing duration. In this case, 40 injectors are activated, and the global injected mass flow rate is kept constant (here 0.03 kg/s) at different actuation frequencies with the same duty cycle (DC = 0.7). The absolute pulsed blowing angle is 𝑖𝑛𝑗 Int. J. Turbomach. Propuls. Power 2022, 7, 10 12 of 15 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 13 of 16 4.2. Pulsed Mode Figure 12 presents a typical evolution of the stall margin improvement with the −30°, and the rotor rotation speed is set at 3200 rpm. It can be seen that there is a strong actuation frequency, in pulsed mode. Note that in Figure 12, no result is presented for frequencies above 200 Hz, as beyond this value, the response time of the valve becomes dependency between the SMI and the actuation frequency. The stall margin improvement significant compared with the blowing duration. In this case, 40 injectors are activated, and grows almost monotonically with the frequency and reaches its maximum for a value the global injected mass flow rate is kept constant (here 0.03 kg/s) at different actuation close to the maximum frequency allowed by the system. However, this maximum value frequencies with the same duty cycle (DC = 0.7). The absolute pulsed blowing angle is 30 , and the rotor rotation speed is set at 3200 rpm. It can be seen that there is a strong is almost already reached around an actuation frequency of 200 Hz. Consequently, the dependency between the SMI and the actuation frequency. The stall margin improvement results in pulsed actuation that are presented correspond to the best results in terms of the grows almost monotonically with the frequency and reaches its maximum for a value SMI. Please note that in this study, all injectors pulsated simultaneously but that a vari- close to the maximum frequency allowed by the system. However, this maximum value is almost already reached around an actuation frequency of 200 Hz. Consequently, the ating actuation in the circumferential direction was also possible with this set-up (as pre- results in pulsed actuation that are presented correspond to the best results in terms of the viously performed a few times in the literature [2]). SMI. Please note that in this study, all injectors pulsated simultaneously but that a variating actuation in the circumferential direction was also possible with this set-up (as previously performed a few times in the literature [2]). Figure 12. Evolution of the SMI with the actuation frequency in pulsed blowing. The number of Figure 12. Evolution of the SMI with the actuation frequency in pulsed blowing. The number of activated injectors N = 40, the injected mass flow rate Q = 0.03 kg/s, the duty cycle DC = 0.7, inj the absolute blowing flow angle a = 30 and the rotation speed W = 3200 rpm. activated injectors 𝑁 = 40, the jet injected mass flow rate 𝑄 = 0.03 kg/s, the duty cycle 𝐷𝐶 = 0.7, the 𝑖𝑛𝑗 absolute blowing flow angle 𝛼 = −30° and the rotation speed 𝛺 = 3200 rpm. 𝑗𝑒𝑡 Figure 13 reports the comparison of the stall margin improvement and power balance for several pulsed actuation tests compared to the points obtained in continuous mode for 30 and 40 injectors activated. The absolute injection angle and the rotation speed are Figure 13 reports the comparison of the stall margin improvement and power balance kept constant and equal, respectively, to 30 and 3200 rpm. It is clear that the benefit of for several pulsed actuation tests compared to the points obtained in continuous mode for pulsed actuation is not obvious in terms of the SMI, as it allows reaching values up to 50% 30 and 40 injectors activated. The absolute injection angle and the rotation speed are kept maximum, sensibly lower than the higher ones obtained in continuous blowing. Nevertheless, if the power balance is considered, pulsed actuation presents a real constant and equal, respectively, to −30° and 3200 rpm. It is clear that the benefit of pulsed interest. Point H is able to reach, in pulsed actuation, an SMI performance close to the one actuation is not obvious in terms of the SMI, as it allows reaching values up to 50% maxi- obtained by point F (in continuous blowing) with a lower flow rate taken from the external mum, sensibly lower than the higher ones obtained in continuous blowing. system and a better power balance. Int. J. Turbomach. Propuls. Power 2022, 7, 10 13 of 15 Int. J. Turbomach. Propuls. Power 2022, 7, x FOR PEER REVIEW 14 of 16 Figure 13. SMI and power balance in pulsed and continuous blowing with the absolute injection Figure 13. SMI and power balance in pulsed and continuous blowing with the absolute injection angle 𝛼 = −30°  and the rotation speed 𝛺 = 3200 rpm. 𝑗𝑒𝑡 angle a = 30 and the rotation speed W = 3200 rpm. jet 5. Conclusions Nevertheless, if the power balance is considered, pulsed actuation presents a real in- terest. Point H is able to reach, in pulsed actuation, an SMI performance close to the one This study reports the results of flow control application for stall margin improvement obtained by point F (in continuous blowing) with a lower flow rate taken from the external (SMI) using an experimental parametric study conducted on a single-stage axial compressor system and a better power balance. test bench. The machine is equipped with fluidic actuators installed on the casing, upstream of the rotor. A series of experiments with various blowing conditions in pulsed and 5. Conclusions continuous modes were conducted. In particular, the aim of the study was, firstly, to shed some light on the controversial influence of the blowing yaw angle. Secondly, it was also This study reports the results of flow control application for stall margin improve- an opportunity to carry out a never-seen investigation of the energy budget of such control ment (SMI) using an experimental parametric study conducted on a single-stage axial methods applied to an axial compressor. compressor test bench. The machine is equipped with fluidic actuators installed on the Concerning the blowing yaw angle effect, it appears that blowing angles, in the casing, upstream of the rotor. A series of experiments with various blowing conditions in relative frame, close to the blade angle at the tip produce the best results in terms of the pulsed and continuous modes were conducted. In particular, the aim of the study was, SMI, as, at this blowing angle, the jet directly acts on the blade loading at the tip and firstly, to shed some light on the controversial influence of the blowing yaw angle. Sec- thus prevents the phenomena at the origin of stall. It is then not a matter of positive or ondly, it was also an opportunity to carry out a never-seen investigation of the energy negative absolute values, as stated often in the literature, and this confirms some findings budget of such control methods applied to an axial compressor. of Kefalakis et al. [7]. For real applications, the relative blowing angle is not the easiest Concerning the blowing yaw angle effect, it appears that blowing angles, in the rela- parameter to adjust, as it depends on the blowing velocity, the absolute blowing angle and tive frame, close to the blade angle at the tip produce the best results in terms of the SMI, the rotor speed. Fortunately, a high SMI value appears to be achieved for a quite large as, at this blowing angle, the jet directly acts on the blade loading at the tip and thus pre- range of relative blowing angles. It means that a single absolute blowing angle can cover vents the phenomena at the origin of stall. It is then not a matter of positive or negative several operating points. absolute values, as stated often in the literature, and this confirms some findings of Kefala- Concerning the energy budget, this study has shown that some of the blowing con- kis et al. [7]. For real applications, the relative blowing angle is not the easiest parameter figurations present a positive net gain on the energy balance for an SMI up to 110% and to adjust, as it depends on the blowing velocity, the absolute blowing angle and the rotor up to nearly 140% with a net energy consumption. Former configurations imply sufficient speed. Fortunately, a high SMI value appears to be achieved for a quite large range of angle coverage and are obtained when an advantageous balance is achieved between the relative blowing angles. It means that a single absolute blowing angle can cover several positive effect of the blowing (increase in the SMI, decrease in the losses) and the energy cost operating points. necessary to produce high-speed jets. The benefit of using pulsed blowing is not obvious in Concerning the energy budget, this study has shown that some of the blowing con- terms of the SMI but is clearly interesting for the power balance as some configurations figurations present a positive net gain on the energy balance for an SMI up to 110% and allow a positive power balance of 2% with a still interesting SMI of 50%. up to nearly 140% with a net energy consumption. Former configurations imply sufficient angle coverage and are obtained when an advantageous balance is achieved between the Int. J. Turbomach. Propuls. Power 2022, 7, 10 14 of 15 The analysis of the efficiency of any flow control system devoted to reducing the energy consumption is always an interesting step, as it allows to state whether such a system “pays its place” in a complex industrial machine, such as an aircraft, where every gram counts. This study is in that sense enlightening, as it allows imagining various uses of the different tested configurations. Firstly, the highest SMI points, despite their net energy cost, could be devoted to critical situations where the safety and level of engine performance need to be maintained. One thinks of take-off and landing cases and of combat situations at high angles of attack to cope with inlet distortion effects. On the contrary, even with a lesser SMI, a configuration exhibiting a net positive energy gain could be continuously used to improve the overall efficiency of the engines and thus reduce their environmental impact. This study constitutes an encouraging proof-of-concept that active flow control is viable from an energy point of view at the laboratory scale, that is, on a simplified, low- speed and single-stage test rig using low-TRL actuators. The next step is now to reproduce and validate the concept on a real engine using industrial-grade actuators. This constitutes the next step of the current project and will be published in the near future. Author Contributions: Conceptualisation, P.J. and A.D.; methodology, P.J., O.R. and A.D.; software, G.M. and P.J.; validation, J.M.M., P.J. and A.D.; formal analysis, J.M.M., P.J. and A.D.; investigation, J.M.M. and G.M.; resources, O.R. and G.M.; writing—original draft preparation, P.J., A.D. and J.M.M.; writing—review and editing, O.R., J.M.M., P.J. and A.D.; supervision, O.R. and A.D.; project administration, O.R. and A.D.; funding acquisition, A.D., P.J. and O.R. All authors have read and agreed to the published version of the manuscript. Funding: This project (ACONIT) has received funding from the Clean Sky 2 Joint Undertaking under the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 886352. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data available on request. Acknowledgments: We are grateful to the technical staff at the Arts et Métiers Institute of Technology in Lille, France, especially Patrick Olivier and Claude Lamacz, for technical support and expertise provided during the test campaign. Conflicts of Interest: The authors declare no conflict of interest. Nomenclature a Absolute blowing flow angle ( ) jet AFC Active flow control b Relative blowing flow angle ( ) jet SM Stall margin V Absolute jet speed (m/s) jet SMI Stall margin improvement Q Injected mass flow rate (kg/s) inj PG Power gain N Number of injectors PC Power cost DC Duty cycle PB Power balance f Driving frequency P Pressure (Pa) N Number of injectors activated V Velocity (m/s) Subscripts r Density (kg/m ) 1 2 Stage inlet outlet q Mass flow rate (kg/s) N Nominal operating point W Rotor rotational velocity (rpm) S Last stable operating point U Rotor tip speed (m/s) B Baseline or case without control P Pressure ratio C Controlled case F Flow coefficient x Axial quantity Y Pressure rise coefficient tip Quantity at blade tip x Axial position (mm) mid Quantity at mid-span Cx Axial chord length (mm) t s Total-to-static quantity Int. J. Turbomach. Propuls. Power 2022, 7, 10 15 of 15 References 1. Moubogha, J.M.; Margalida, G.; Joseph, P.; Roussette, O.; Dazin, A. Surge Margin Improvement by Continuous and Pulsed Tip injection. In Proceedings of the 14th European Conference on Turbomachinery Fluid Dynamics & Thermodynamics, ETC2020-639, Gdansk, Poland, 12–16 April 2021. 2. Day, I.J. Stall, surge, and 75 years of research. J. Turbomach. 2016, 138, 011001. [CrossRef] 3. Hathaway, M.D. Passive Endwall Treatments for Enhancing Stability; NASA Report No. TM-2007-214409; July 2007. Available online: ntrs.nasa.gov (accessed on 13 October 2021). 4. Li, J.; Du, J.; Nie, C.; Zhang, H. Review of Tip Air Injection to Improve Stall Margin in Axial Compressors. Prog. Aerosp. Sci. 2019, 106, 15–31. [CrossRef] 5. Hewkin-Smith, M.; Pullan, G.; Grimshaw, S.D.; Greitzer, E.M.; Spakovszky, Z.S. The Role of Tip Leakage Flow in Spike-Type Rotating Stall Inception. J. Turbomach. 2019, 141, 061010. [CrossRef] 6. Pullan, G.; Young, A.M.; Day, I.J.; Greitzer, E.M.; Spakovszky, Z.S. Origins and Structure of Spike-Type Rotating Stall. J. Turbomach. 2015, 137, 051007. [CrossRef] 7. Kefalakis, M.; Papailiou, K.D. Active Flow Control for Increasing the Surge Margin of an Axial Flow Compressor. Turbo Expo Power Land Sea Air 2006, 424, 101–111. 8. Stößel, M.; Bindl, S.; Niehuis, R. Ejector Tip Injection for Active Compressor Stabilization. Turbo Expo Power Land Sea Air Am. Soc. Mech. Eng. 2014, 45608, V02AT37A004. 9. Suder, K.L.; Hathaway, M.D.; Thorp, S.A.; Strazisar, A.J.; Bright, M.B. Compressor Stability Enhancement Using Discrete Tip Injection. J. Turbomach. 2001, 123, 14–23. [CrossRef] 10. Kern, F.; Brehm, S.; Niehuis, R. Ejector Tip Injection System for Active Aerodynamic Compressor Stabilization Part I: Design and Experiment. In Proceedings of the 12th European Conference on Turbomachinery Fluid Dynamics & Thermodynamics, ETC2017-244, Stockholm, Sweden, 3–7 April 2017. 11. Strazisar, A.J.; Bright, M.M.; Thorp, S.; Culley, D.E.; Suder, K.L. Compressor Stall Control through Endwall Recirculation. Turbo Expo Power Land Sea Air 2004, 41707, 655–667. 12. D’Andrea, R.; Behnken, R.L.; Murray, R.M. Rotating Stall Control of an Axial Flow Compressor Using Pulsed Air Injection. J. Turbomach. 1997, 119, 742–752. [CrossRef] 13. Khaleghi, H.; Teixeira, J.A.; Tousi, A.M.; Boroomand, M. Parametric Study of Injection Angle Effects on Stability Enhancement of Transonic Axial Compressors. J. Propuls. Power 2008, 24, 1100–1107. [CrossRef] 14. Nie, C.; Xu, G.; Cheng, X.; Chen, J. Micro Air Injection and Its Unsteady Response in a Low-Speed Axial Compressor. J. Turbomach. 2002, 124, 572–579. [CrossRef] 15. Greenblatt, D.; Wygnanski, I. The Control of Flow Separation by Periodic Excitation. Prog. Aerosp. Sci. 2000, 36, 487–545. [CrossRef] 16. Margalida, G. Analyse Expérimentale et Contrôle Actif de l’écoulement Dans Un Compresseur Axial Mono-Étagé Durant La Transition Vers Le Décrochage Tournant. Ph.D. Thesis, Ecole Nationale Supérieure d’Arts et Métiers, Lille, France, 2019. 17. Margalida, G.; Joseph, P.; Roussette, O.; Dazin, A. Active Flow Control in an Axial Compressor for Stability Improvement: On the Effect of Flow Control on Stall Inception. Exp. Fluids 2021, 62, 1–13. [CrossRef] 18. Dazin, A.; Joseph, P.; Romano, F.; Gallas, Q.; Marty, J.; Aigouy, G.; Stô el, M.; Niehuis, R. The ACONIT Project: An Innovative Design Approach of Active Flow Control for Surge Prevention in Gas Turbines. IOP Conf. Ser. Mater. Sci. Eng. 2021, 1024, 012068. [CrossRef] 19. Margalida, G.; Joseph, P.; Roussette, O.; Dazin, A. Comparison and Sensibility Analysis of Warning Parameters for Rotating Stall Detection in an Axial Compressor. Int. J. Turbomach. Propuls. Power 2020, 5, 16. [CrossRef] 20. Vegliò, M.; Dazin, A.; Bois, G.; Roussette, O. Unsteady Pressure Measurements of Spike Type Inception in Axial Compressor: Time Frequency Analysis and Averaging Procedure. In Proceedings of the 11th European Conference on Turbomachinery Fluid Dynamics and Thermodynamics, ETC2015-209, Madrid, Spain, 23–27 March 2015. 21. Weigl, H.J.; Paduano, J.D.; Frechette, L.G.; Epstein, A.H.; Greitzer, E.M.; Bright, M.M.; Strazisar, A.J. Active Stabilization of Rotating Stall and Surge in a Transonic Single Stage Axial Compressor. In Volume 4: Manufacturing Materials and Metallurgy; Ceramics; Structures and Dynamics; Controls, Diagnostics and Instrumentation; Education; IGTI Scholar Award; American Society of Mechanical Engineers: Orlando, FL, USA, 1997. 22. Deveaux, B.; Fournis, C.; Brion, V.; Marty, J.; Dazin, A. Experimental Analysis and Modeling of the Losses in the Tip Leakage Flow of an Isolated, Non-Rotating Blade Setup. Exp. Fluids 2020, 61, 126. [CrossRef]

Journal

International Journal of Turbomachinery, Propulsion and Power – Multidisciplinary Digital Publishing Institute

Published: Mar 16, 2022

Keywords: axial compressor; active flow control; stall margin improvement; tip blowing; energy cost

There are no references for this article.