TY - JOUR
AU - Li,, Jing
AB - Abstract To date, both micro-seismic (MS) and electromagnetic radiation (EMR) techniques are used as normal, daily safety monitoring tools for coal or rock dynamic disasters in China. In previous studies, these two non-destructive techniques are usually analyzed independently; few works have been done to characterize the correlation or difference between them. This paper aims to analyze the correlated features of the MS and EMR signals obtained from a field site test on a deep blasting workface in Pingdingshan 10# coal colliery. The de-noised signals are firstly compared for their associated features, both in time synchronization and energy correlation, and then the mechanism for the correlated response is also investigated. The results show that: (1) MS and EMR signals have a higher time-synchronization and energy correlation. (2) The EMR signal in a blasting operation is a local signal, near to the location of the detectors. (3) The two orthogonal layout magnetic antennas (along the roadway and vertical to the coal wall) can detect a single pulse signal and group-occurring cluster signals. These two kinds of EMR signals result from coal crack evolution and resistance–capacitance (RC) oscillation circuits respectively, which are triggered by seismic longitudinal waves. (4) The seismic transverse wave, especially for the low frequency component of it, makes a rubbing friction effect on coal, producing a low-frequency electromagnetic oscillation signal. Affected by the power and propagation direction of the energy, the signal can only be captured by the antenna in the vertical direction of the coal wall. coal and rock dynamic disaster, micro-seismic, electromagnetic radiation, synchronism, correlated characteristics 1. Introduction Even now, blasting operations, such as tunneling blasting, blast-mining and other processes, are still widely applied in China’s coal mines. During these processes, the stress distribution will be changed, along with crushing of the coal and rock body, thus causing the quick transfer and concentration of the load (Iannacchione and Zelanko 1995, Yang et al2011). At the same time, the blasting seismic waves can propagate and reflect, leading to tensile fractures, together with numerous fracture growths and evolvution (Yasitli and Unver 2005, Yang et al2013). The factors above might cause structural instability and eventually coal or rock dynamic disaster (Milev et al2001). Coal and gas outbursts and rockbursts are two common types of coal or rock dynamic disasters caused by blasting operations, which often result in enormous casualties and property loss. Therefore, an accurate coal or rock dynamic disaster forecast is the key to prevent them. These prediction methods can be roughly divided into two types: one is traditional forecasting methods, such as the hole-drilling method (Dou et al2009), separation monitoring method (Jiang et al2012), or the statistical analysis method (Arjang and Herget 1997); the other kind involves geophysical methods, including acoustic emission (AE) (Hardy 1981), micro-seismic (MS) (Fujii et al1997, Ge 2005, Xu et al2010, Lu et al2012) and electromagnetic radiation (EMR) (Frid 1997a, 1997b, 1999, He et al1999, Rabinovitch et al2002, Frid and Vozoff 2005). Compared with the traditional methods, geophysical methods have been recognized by the majority of scholars for their advantages of non-contact, continuity, high timeliness, wide space range, abundant spectrum, broadband character, relative security and not affecting the production, and are considered to be the most promising methods for forecasting coal or rock dynamic disaster (Abuov et al1988, Frid 1992, 1997b, Lu and Dou 2014). The principle between AE and MS technology is basically the same, while the difference lies in the higher frequency range of the former (Hardy 1986). In fact, the word ‘rock talk’ was regarded as a synonym of ‘AE/MS activity’ right up to the late 1960s. As to the MS, since it was discovered by Obert and Duvall (1957) from the United States Bureau of Mines (USBM), numerous studies have been conducted around this interesting phenomenon caused by blasting operations in coal mines. Dou et al discussed the relationship of roadway dynamic destruction and hypocenter parameters by simulation (Gao et al2008) and theoretical analysis (Cao et al2008). Lu suggested that the MS vibrations in high-stress rock masses induced by blasting excavation were attributed to the combined actions of explosions and the transient release of in situ stress (Lu et al2011, 2012). Xu analyzed the frequency spectrum of the rockburst’s MS signals induced by dynamic disturbance (Xu et al2010). Milev and Spottiswoode, adopting MS and other monitoring systems, simulated a rockburst by means of a large underground explosion (Milev et al2001). On the basis of elastic and anelastic wave theory, Sambuelli proposed a scaled distance (SD) law for predicting the maximum particle velocity vibration (PPV) from blasting operations (Sambuelli 2009). A similar study was carried out by Mesec (2010). In addition, there are some scholars focused on the uncertain elements of MS from blasting (Hargather et al2007, Segarra et al2012). Compared with the MS monitoring technology, the research of coal or rock EMR technology began slightly later, but it has been very active in recent years. Many attempts have been made to forecast and prevent the occurrence of coal or rock dynamic disasters caused by blasting operations in underground coal mines. Frid, a famous scholar of EMR study, has done a great deal of exploration in this field (Frid 1997a, 1997b, 1999, Frid and Vozoff 2005). As mentioned earlier, coal and rock would be deformed and ruptured in the process of a blasting operation, producing a great deal of EMR signals (Nitsan 1977, Enomoto and Hashimoto 1990, He et al1999, Butler et al2001, Wang et al2011b). Sobolev et al (1984) noted that the high-frequency EMR waves were emitted along quartz and sulfides’ grain boundaries when the explosive charge was detonated. O’Keefe and Thiel (1991) pointed out that the generation mechanism of EMR during quarry blasting should include three parts: rock fracture at the time of the explosion, charged rocks discharging on impact with the pit floor and micro-fracture of the remaining rock wall due to the pressure adjustment behind the blast. Tomizawa et al (1994) carried out the blasting experiment at 70 meters underground so as to study the characteristics of the electromagnetic field and blasting site. Nesbitt and Austin (1988) showed that EMR accompanied blasting activity in a deep mine. Guo et al (1999) analyzed the EMR signals at different distances caused by quarry blasting, and concluded that the EMR signals were synchronized with, or even lagged behind, the shock waves. This experiment revealed that the rock micro-fracture can release abnormal EMR signals. Rabinovitch et al (2002) formulated that numerous cracks in the blasting process was the origin of strings of EMR pulses. Scott et al (2004) assumed the EMR signals were generated by seismicity or rock breaking in the process of blasting. With the continual development of theoretical study and test equipment, MS and EMR monitoring technology has become ripe day after day. However, due to the complexity of the mechanisms and the uncertainty of locations, it is difficult to make accurate predictions about coal or rock dynamic disasters through a single approach. In fact, both MS and EMR are the releasing energy formed during the fracture of coal or rock. In some way, the two things are actually homologous (Xu and Wu 1991). Some seismologists have carried out this project going back a long time, for instance, Martner and Sparks (1959), Gokhberg et al (1979), Migunov et al (1985) and Serebryakova et al (1992). The mechanisms of rockbursts and earthquakes are similar, i.e. both are the sudden failure of coal and rock mass due to the action of an external force (Li et al2007). And, of course, gas should also be considered when it comes to a coal and gas outburst. Under the conditions of laboratory tests, partial AE detected were associated with EMR (Dou et al2009). Their onset times were chaotic (Nie et al2009) or synchronous (Yamada et al1989, Rabinovitch et al2002, Pralat and Wojtowicz 2004, Mori and Obata 2008, Mori et al2009). As to MS and EMR, similar phenomena are reported by He et al (2003) and Yang (2014). In the field, the larger the vertical stress gradient value is, the stronger the intensity of EME signals and the lower the frequency band of MS signals are (Lu and Dou 2014). There are obvious correlations between the MS and EMR signals (Nesbitt and Austin 1988). The studies on the relationship between the MS and EMR signals in a coal mine are still limited, and the correlated characteristics caused by blasting operations are seldom reported on. The blasting is the result of the interaction of multiple factors, such as the explosive charge, transmission forms and mesostructured etc, so the single detection method usually cannot reflect all available information in blasting driving. In particular, the complexity and probability of dynamic disaster caused by blasting are ever-increasing with the increase of mining depth (Wang et al2009, Brady and Brown 2013). Therefore, this paper will focus on collecting the exact and complete MS and EMR signals in the process of blasting, concurrently exploring the correlated characteristics between them. On this research theme, the text details different aspects across four sections. Firstly, we review the achievements related to this research in retrospect, then introduce the data acquisition and processing method, before finally analyzing the relationship between the MS and EMR signals, as well as their correlation mechanisms. 2. Data acquisition and processing method 2.1. Monitoring system installation and parameters setting The field site data collection work was performed on the Ding 5, 6-21180 excavating workface in Pingdingshan 10# Coal Colliery, where the blasting driving method was adopted for tunneling (0.8 m in advance per attack). Due to the complexity of the geologic environment, dynamics phenomena such as coal bursts and roof scaling-off occur frequently, which result in great difficulties for steady driving, but provide excellent site positions for field experiment observations (Wang et al2014). As a consequence, a set of coal mine daily monitoring facilities named ZDKT-1 was installed and applied near the workface. ZDKT-1 is a type of coal and rock dynamic disaster monitoring system developed by China University of Mining and Technology, with both EMR and MS monitoring functions. Taking the characteristics of signal propagation and normal coal mining production into consideration, the system was placed in the 21180 excavation roadway, 300 m away from the east transportation tunnel. The detector position of the system was maintained constantly throughout the whole monitoring period (figure 1). The monitoring system mainly consists of a MS geophone and two orthogonal EMR receiving antennas. Each EMR antenna has a group of Galvanized solenoids, and the two solenoids are placed vertically. In the meantime, the MS geophone is embedded in this system (figure 2). It is worth mentioning that one of the two orthogonal antennas, defined as 1#, was set as pointing to the working face along the roadway, while the other, defined as 2#, was arranged in a perpendicular direction to the other, pointing to the coal wall. After receiving the origin signals, both the MS and EMR signals were first amplified and converted, then transmitted to the ground by the communication cable, so as to realize the synchronous data acquisition purpose. Figure 1. Open in new tabDownload slide The arrangement of the monitoring system. Figure 1. Open in new tabDownload slide The arrangement of the monitoring system. Figure 2. Open in new tabDownload slide The schematic diagram and actual equipment of MS geophone and two orthogonal EMR antennas. (a) The schematic diagram. (b) The actual equipment. Figure 2. Open in new tabDownload slide The schematic diagram and actual equipment of MS geophone and two orthogonal EMR antennas. (a) The schematic diagram. (b) The actual equipment. The field site experiment test was conducted on the 21180 excavating workface, where 20 effective driving blasting tests were conducted, monitored and recorded. The experimental parameters are illustrated in table 1. Table 1. Blasting test parameters on 21180 excavating face. Test ID . Explosive charge/roll . Monitor distance/m . 1 42 188.2 2 49 189.8 3 54 190.6 4 54 192.2 5 45 194.6 6 25 195.4 7 44 196.2 8 22.5 197 9 25 197.8 10 22.5 198.6 11 24 199.4 12 22 201 13 45 201.8 14 41 202.6 15 41 203.4 16 37 204.2 17 36 205 18 20 205.8 19 35 206.6 20 17.5 207.4 Test ID . Explosive charge/roll . Monitor distance/m . 1 42 188.2 2 49 189.8 3 54 190.6 4 54 192.2 5 45 194.6 6 25 195.4 7 44 196.2 8 22.5 197 9 25 197.8 10 22.5 198.6 11 24 199.4 12 22 201 13 45 201.8 14 41 202.6 15 41 203.4 16 37 204.2 17 36 205 18 20 205.8 19 35 206.6 20 17.5 207.4 Note: As the exact blasting charge is too complicated to be evaluated, cartridge count filled in tunneling blasting is used to represent it in the table. Open in new tab Table 1. Blasting test parameters on 21180 excavating face. Test ID . Explosive charge/roll . Monitor distance/m . 1 42 188.2 2 49 189.8 3 54 190.6 4 54 192.2 5 45 194.6 6 25 195.4 7 44 196.2 8 22.5 197 9 25 197.8 10 22.5 198.6 11 24 199.4 12 22 201 13 45 201.8 14 41 202.6 15 41 203.4 16 37 204.2 17 36 205 18 20 205.8 19 35 206.6 20 17.5 207.4 Test ID . Explosive charge/roll . Monitor distance/m . 1 42 188.2 2 49 189.8 3 54 190.6 4 54 192.2 5 45 194.6 6 25 195.4 7 44 196.2 8 22.5 197 9 25 197.8 10 22.5 198.6 11 24 199.4 12 22 201 13 45 201.8 14 41 202.6 15 41 203.4 16 37 204.2 17 36 205 18 20 205.8 19 35 206.6 20 17.5 207.4 Note: As the exact blasting charge is too complicated to be evaluated, cartridge count filled in tunneling blasting is used to represent it in the table. Open in new tab 2.2. The method of data processing and signal de-noising The signal is the carrier and physical manifestation of information, as well as a function of time and space and other independent variables (Kamen and Heck 2000). The signals can be divided into two major categories: stationary signals and non-stationary signals. The usual methods of analysis include Fourier transform (FT), fast Fourier transform (FFT), wavelet transform (WT), Hilbert–Huang transform (HHT), fractal analysis, chaos analysis and neural network analysis, within the first four methods are more commonly used. Fourier analysis is a common mathematical tool in signal processing, whose function is to realize the signal conversion between the time and frequency domains. Basically, there are two common kinds of Fourier analysis: Fourier series and FT. As the MS and EMR signals have typical non-cyclical characteristics, FT is more suitable for signal analysis. For the record, according to Nyquist sampling theory, the sampling frequency is at least two times faster than the signal bandwidth. In other words, when the sampling frequency fs is determined, the bandwidth of the signal frequency is fs/2 ⁠. As described in the previous section, the essence of signal analysis is to convert original signal (continuous signal) into a digital signal (discrete signal). No matter how high the sampling frequency is, the information carried by the signals would inevitably be lost. Therefore, the Fourier transform discussed here is actually a discrete Fourier transform (DFT). The disadvantage of DFT is it often allows for plentiful time in engineering. Many attempts are made to overcome this problem, and FFT is the one of most efficient methods. The purpose of FFT is to remove duplicate calculations, reduce multiplication and simplify the structure. FT is a global transformation in the whole time domain, where we could get the overall spectrum, rather than the local properties of time-frequency distributions. Obviously, this method has not been able to accurately grasp the total information of the signals. Windowed algorithms have emerged to solve this conundrum. It includes a short-time Fourier transform (STFT) and WT, and the latter could obtain varied time and frequency windows, more precise time-frequency resolutions and other advantages when compared with the former. However, WT depends on the choice of the wavelet function. Generally, the wavelet basis is not unique, and different wavelet functions will lead to different results. Unfortunately, selecting the rational and effective wavelet basis is often difficult and sometimes even controversial (Daubechies 1992). The non-linear, non-stationary and transient signals, such as the MS and EMR signals mentioned in this article, are more applicable to a HHT for it can avoid the trouble of wavelet base selection. The HHT method contains two parts; the empirical mode decomposition (EMD) and the Hilbert transform (HT) (Huang 2000). The EMD method is a fundamental part of HHT, which deconstructs any complicated data into a finite number of components, namely, the intrinsic mode function (IMF). What follows is the instantaneous frequency and related concepts of these IMF components obtained by HHT. By comparing and analyzing the time-frequency characteristics of these IMFs, the eligible IMFs with typical characteristics are reserved, while unqualified IMFs are abandoned (Wang et al2012). Reconstructed and retained IMFs obtain effective signals. The program flow chart of the EMD method and the HT are shown in figure 3. Figure 3. Open in new tabDownload slide The program flow chart of the EMD method and the HT. Figure 3. Open in new tabDownload slide The program flow chart of the EMD method and the HT. The advantage of the EMD method is self-adaptive, without choosing prior basis functions. All of the IMF components are arranged in accordance with the frequency from high to low automatically. What is worth mentioning is that the EMD might sometimes lead to the phenomena of mode mixture (Wang et al2012). In addition, as shown in figure 3, the cubic spline function is used in the process of EMD. When the filter number is too much or extremal points are too few, end wing problems will occur. In order to solve these problems, Huang proposes the ensemble empirical mode decomposition (EEMD) algorithm, whose essence is to add origin signals with Gauss white noise (Wu and Huang 2009). As Gauss white noise is uniform distribution with zero mean value, it can be cancelled out by averaging in the process of decomposition. 2.3. The example of data processing and signal de-noising Due to the complexity and uncertainty of the underground environment, the MS and EMR signals monitored would inevitably be interfered with various noises (Chen and Wang 2011). As such it is necessary for de-noising before analyzing them. To analyze the background noise components, 5s signals without blasting operations are selected randomly, then the signals are analyzed based on FFT. The oscillogram in the time domain and spectrogram in the frequency domain for background noise of both the MS and EMR signals are shown in figure 4. Figure 4. Open in new tabDownload slide The background noise oscillograms and spectrograms of the MS & EMR signals. (a) The background noise oscillogram of the MS signal. (b) The background noise spectrogram of the MS signal. (c) The background noise oscillogram of the EMR signal. (d) The background noise spectrogram of the EMR signal. Figure 4. Open in new tabDownload slide The background noise oscillograms and spectrograms of the MS & EMR signals. (a) The background noise oscillogram of the MS signal. (b) The background noise spectrogram of the MS signal. (c) The background noise oscillogram of the EMR signal. (d) The background noise spectrogram of the EMR signal. As shown in figure 4, the background noise of the MS and EMR signals vary in the range of  ±0.15mV or so, and amplitude variations are relatively stable, while frequency distributions are slightly different from each other. Specifically, the frequency bands of MS are mainly about 100–200, 500–800 and 1400 Hz, while frequencies for EMR are distributed in around 500, 600–800 and 1400–1450 Hz, in which both generally focus on 500–800 Hz. To be clear, as the monitoring system works without interruption for a long time, in order to avoid too large an amount of data, the sampling frequency of both the MS and EMR signals are set to 2962 Hz, which should be worth paying more attention to in the FFT transform. In terms of the MS signal, the signal itself has presented typical nonlinear characteristics in a wide frequency spectrum with ambient noise. In this work, by adopting the self-adaptive HHT method (Huang et al1998), the background or ambient noise in the composite signal is obviously reduced. Taking the MS signal of the test numbered as ID 6 in table 1 for example, we give the details of our whole signal de-noising process and a comparison before and after de-noising in the following figure 5. Figure 5. Open in new tabDownload slide The de-noising process and the effect of a typical MS event. (a) Original MS signals. (b) EEMD decomposition as a result of MS signals. (c) Effective reconstruction of MS signals. (d) Hilbert spectrum of the MS signals before de-noising. (e) Hilbert spectrum of the MS signals after de-noising. Figure 5. Open in new tabDownload slide The de-noising process and the effect of a typical MS event. (a) Original MS signals. (b) EEMD decomposition as a result of MS signals. (c) Effective reconstruction of MS signals. (d) Hilbert spectrum of the MS signals before de-noising. (e) Hilbert spectrum of the MS signals after de-noising. Figure 5(a) shows the oscillogram of the original MS signal. By the application of the EEMD on the original signal, 14 IMFs and one residual function are obtained. It can be concluded from figure 5(b) that IMF 1–5 have typical MS characteristics and sizeable fraction energy, so all of them should be reserved. IMF 6–15, in contrast, not showing demonstrable change, can be abandoned. The effective MS signal is reconstructed on the basis of effective IMF selection and shown in figure 5(c). Comparing the Hilbert energy spectrum before and after de-noising (figures 5(d) and (e)), it can be found that the background noise in 500–800 Hz has been removed efficiently, and, in the meantime, the valid signals are well-reserved. Although the background noise of the MS signals could be removed effectively by the EEMD filtering method, it is found that the direct utility of this method lacks applicability when applied to the EMR signals, which is possibly caused by the special features of field site EMR signals. Thereinafter, the EMR signals, recorded simultaneously as the MS signal in figure 5, are selected to illustrate our reasonable method for the EMR data de-noising process. Figure 6 shows the processing results of the EMR signals in detail. By comparing figure 6(a) with 6(b), the results show that the collected two channel EMR signals in blasting operations both own obvious pulse characteristics, with a short duration and a relatively large amplitude. To explore its features adequately, the pulse parts from the original signals are extracted and separated. Then the residual parts between 1# and 2# antenna are proofed to vary markedly (Li et al2014b). It is shown that figure 6(d) is basically the same as figure 4(c), illustrating that there are only background noise signals, besides the pulse parts, for 1# antenna. Figure 6. Open in new tabDownload slide The analysis of the EMR signals. (a) Original EMR signals of 1# antenna. (b) Original EMR signals of 2# antenna. (c) Pulse part of the original EMR signals of 1# antenna. (d) Remaining part of the original EMR signals of 1# antenna. (e) Pulse part of the original EMR signals of 2# antenna. (f) Remaining part of the original EMR signals of 2# antenna. (g) Remaining part of the original EMR signals of 2# antenna (partial view). Figure 6. Open in new tabDownload slide The analysis of the EMR signals. (a) Original EMR signals of 1# antenna. (b) Original EMR signals of 2# antenna. (c) Pulse part of the original EMR signals of 1# antenna. (d) Remaining part of the original EMR signals of 1# antenna. (e) Pulse part of the original EMR signals of 2# antenna. (f) Remaining part of the original EMR signals of 2# antenna. (g) Remaining part of the original EMR signals of 2# antenna (partial view). Unlike figure 6(d), (f) displays its small-scope fluctuation, which suggests it may contain effective information, so the remaining signals are analyzed by the EEMD method, as shown in figure 7. Through the EEMD analysis, it is easy to see that those IMFs in higher frequency bands, lacking the valid EMR characteristics, are noise signals. Consequently the IMFs with typical narrow turbulence characteristics are reserved and selected to be reconstructed, as shown in figure 7(b). By means of analyzing the frequency spectrum of the constructed signal, it is known that the frequency range is under 20Hz, and the dominant frequency is around 13.2Hz. Obviously, this frequency is much smaller than the dominant frequency of the pulse part (figure 7(d)). By refactoring the pulse part and effective reconstruct remainder of EMR signal, the valid EMR signals of 2# antenna could be obtained finally, as shown in figure 7(e). Figure 7. Open in new tabDownload slide The EEMD analysis result of the EMR signals for 2# antenna. (a) The analysis of the remaining EMR signals for 2# antenna. (b) Effective reconstruction of the remaining EMR signals for 2# antenna. (c) The spectrogram of the remaining EMR signals. (d) The spectrogram of the pulse parts of the EMR signals. (e) The whole reconstruction of the EMR signals for 2# antenna. Figure 7. Open in new tabDownload slide The EEMD analysis result of the EMR signals for 2# antenna. (a) The analysis of the remaining EMR signals for 2# antenna. (b) Effective reconstruction of the remaining EMR signals for 2# antenna. (c) The spectrogram of the remaining EMR signals. (d) The spectrogram of the pulse parts of the EMR signals. (e) The whole reconstruction of the EMR signals for 2# antenna. Besides the EMR signals recorded as test numbered ID 6, other monitored signals also show similar features. Specifically, the original EMR signals of 1# antenna (pointing to the working face along the tunnel) contain only a pulse EMR signal and background noise; but for the EMR signals of 2# antenna (pointing to coal wall), they still include low frequency oscillation compositions, separate from the pulse signal and ambient noise. 3. The correlation analysis of the MS and EMR signals The correlation analysis has been widely applied in the research of synchronization and correlation among different signals (Lee et al1950). There are two main categories: autocorrelation and cross correlation analysis. The former reflects the dependence of signals on time and other parameters, while the latter shows the interdependence of different signals (Barlow 1973). 3.1. The analysis of time synchronization To study the correlation characteristics between the MS and EMR signals during the fracture of coal and rock caused by blasting, the synchronization relationship in the time domain is analyzed. Again, the de-noised and processed data from the field test numbered as ID6 are chosen to analyze time synchronization for the MS and EMR events. Previous studies have shown that the impact pressure caused by blasting would spread out from its source in the form of a wave motion (Blair 1995, Blair and Jiang 1995). According to the difference of the direction between the point mass and wave, these waves can be classified into two categories: longitudinal wave (P-wave) and transverse wave (S-wave). The wave velocities of both the P- and S-wave in the same medium are usually constant, and the velocity of the P-wave is always higher than the S-wave. Obviously, as shown in figure 8, when the P-and S-wave of the MS signals spread from the source location to the position of the detectors, these two distinct types of waves are apparently separated for a long enough distance in our case. Due to the slightly lower propagation speed and more energy, the S-wave can be easily distinguished from the P-wave (Sheriff and Geldart 1995). Figure 8. Open in new tabDownload slide The synchronous analysis of the MS & EMR reconstruction signals. Figure 8. Open in new tabDownload slide The synchronous analysis of the MS & EMR reconstruction signals. Figure 8 also reflects the time domain features of the two EMR signals. It is found that the pulse part extracted from the signals actually contains two different forms: single pulse (I) and cluster signals (II). As we can see, the single pulse signal is detected earlier than the cluster ones. In addition, the single pulse is captured by 1# antenna a bit earlier than 2# antenna, while the receipt time of the two in the cluster signals is basically the same. Also, it is noticeable to conclude that besides the pulse part, the EMR signals of 2# antenna still contain narrow turbulence signal compositions. In the meantime, this figure also shows the relevancy characteristics of the MS and EMR signals. It is observed that there is great synchronicity between them. The single pulse signal occurs in the stage of the longitudinal wave of the MS signals, while the cluster signals and narrow turbulence signals are largely concentrated on the stage of transverse wave. Compared to the pulse signal, the turbulence part of the EMR signals have the characteristics of smaller amplitude, longer lasting duration and less frequency, as well as certain vibration attenuation characteristics. Outwardly, there is no certain time-domain relationship between the turbulence part of the EMR and MS signals. However, by deconstructing the effective MS signals (figure 5(c)) based on the EEMD method, then reconstructing its low frequency component and comparing it with the turbulence part of the EMR signals, it is evident that there exists a marked synchronization between them (figure 9). Figure 9. Open in new tabDownload slide The synchronous analysis of low frequency vibration components and EMR oscillation signals. Figure 9. Open in new tabDownload slide The synchronous analysis of low frequency vibration components and EMR oscillation signals. The same analysis method is applied to the acquired data of all 20 field tests, and similar patterns reveal that there is good synchronism between the MS and EMR signals caused by blasting within the time scale. 3.2. The analysis of energy correlation 3.2.1. The correlation analysis between two EMR channels. The energy correlation of the EMR reconstructing signals monitored by two antennas is shown in figure 10. It displays a good linear relationship between the channels. Thereinto, the energy of 2# antenna is slightly higher than 1# antenna, which could be attributed to the low frequency oscillation of the EMR signals besides the pulse parts. In order to study the relationship of the pulse parts between the two antennas, they are compared in figure 10(b). The result shows that the slope of the fitting line is approximately equal to 1, the intercept is almost equal to 0, and the regression equations are available. Thus, the total energy of the two pulse parts are nearly the same. In addition, the higher correlation between them can be verified by comparing the related parameters, such as the pulse number and average power, etc. Figure 10. Open in new tabDownload slide The energy relational analysis of the EMR reconstruction signals. (a) The total energy of the two EMR signals. (b) The energy of the two pulse EMR signals. (c) The relationship among 1# EMR signals, distance and cartridge count. (d) The relationship among 2# EMR signals, distance and cartridge count. (e) The relationship among 2# pulse EMR signals, distance and cartridge count. (f) The relationship among 2# oscillation EMR signals, distance and cartridge count. Figure 10. Open in new tabDownload slide The energy relational analysis of the EMR reconstruction signals. (a) The total energy of the two EMR signals. (b) The energy of the two pulse EMR signals. (c) The relationship among 1# EMR signals, distance and cartridge count. (d) The relationship among 2# EMR signals, distance and cartridge count. (e) The relationship among 2# pulse EMR signals, distance and cartridge count. (f) The relationship among 2# oscillation EMR signals, distance and cartridge count. The relationship among refactoring EMR signals, monitoring distance and cartridge count is shown in figures 10(c) and (d). As can be seen, although the signals have a certain discreteness, the 3D plot of the energy of the EMR signals after interpolation reflects that the EMR energy attenuates gradually with increasing distance and decreasing cartridge count. As the pulse and low frequency oscillation signals are monitored by 2# antenna simultaneously, the energy relations among those signals, monitoring distance and cartridge count are shown in figures 10(e) and (f), respectively. It can be seen from the chart that the variation tendency displayed between the pulse part and the two global EMR signals are basically the same, and the attenuation of energy is largely driven by cartridge count, but the effect from those factors on the energy of low frequency oscillation is not obvious. 3.2.2. The correlation analysis between the MS and EMR signals. Figures 11(a) and (b) illustrate the relationship between the MS and two EMR signals. As can be seen, there are good linear dependences among them (R2  =  0.93 and 0.94), and the energy of the MS signals is a bit stronger than that of the EMR. Figure 11. Open in new tabDownload slide The energy relational analysis of the MS and EMR reconstruction signals. (a) The energy of the MS and 1# EMR signals. (b) The energy of the MS and 2# EMR signals. (c) The energy comparison of the P-waves and two EMR signals during the same period. (d) The energy comparison of the S-waves and two EMR signals during the same period. (e) The energy comparison of the P-waves and two pulse EMR signals. (f) The energy comparison of the S-waves and oscillation EMR signals of 2# antenna. Figure 11. Open in new tabDownload slide The energy relational analysis of the MS and EMR reconstruction signals. (a) The energy of the MS and 1# EMR signals. (b) The energy of the MS and 2# EMR signals. (c) The energy comparison of the P-waves and two EMR signals during the same period. (d) The energy comparison of the S-waves and two EMR signals during the same period. (e) The energy comparison of the P-waves and two pulse EMR signals. (f) The energy comparison of the S-waves and oscillation EMR signals of 2# antenna. According to the synchronization of the MS and EMR signals, as shown in figure 8, the EMR signals are divided into two consecutive stages by time; the P-waves action stage and S-waves action stage. Through calculating the energy of the EMR signals of these two stages and comparing them with the MS signals (figures 11(c) and (d)), it can be seen from the picture that the energy of the EMR signals is positively associated with the MS signals during the period of P- and S-waves. That is to say, the energy of the EMR signals boost with the increase of the MS signals over the same time period. Furthermore, the signals of MS and EMR during the period of the S-waves have a larger magnitude than those during the period of the P-waves. Analogously, if counting the energy of the pulse and oscillating component, and comparing them with the energy of the P-waves and S-waves, it could be concluded that there is a positive correlation in the relationship between them. 4. Discussion 4.1. The monitoring mechanism of the MS and EMR signals Generally speaking, when the explosive blasting procedure is performed on the excavation working face, it would produce external effects, such as forming an explosive blast wave on the coal and rock medium. Along with an increase in the transmitting distance, the blast wave with a steep wavefront gradually transforms into the stress and vibration waves. Between them, the stress wave mainly shows destructiveness in tensile failure effects, thus forming the radial direction fissure. But regarding the vibration wave, it mainly causes an elastic oscillation and leads to the destruction of coal and rock to some extent. For the monitor distance of the 21180 excavation working face, the coal seam nearby monitoring sites mainly suffer the effects from both the stress and vibration waves. After experiencing repeated actions of blasting operations, there must be micro-fractures initiation and damage evolution occurring, accompanying the medium vibrations and frictions phenomenon inside the coal seam. As mentioned before, the P-waves and S-waves of the MS signals are apparently separated, which means that they can be recognized easily. Since the particle motion and wave propagation directions are either the same or mutual perpendicularly, it is easy to conclude that P-waves could speed up the evolution process of a radial fissure, and cause the rapid opening and closing of an existing fissure; while S-waves, who are perpendicular to the particle, may produce repeatedly rubbing roles, and lead to the effect of sliding friction. It is important to note that the EMR signals would gradually decay with the increase of transmission distance, so the effective monitoring range of the antenna is limited (Rabinovitch et al2002). Take the KBD5 electromagnetic radiation monitoring and measuring instrument, for example; the effective monitoring range of the antenna is 7 ~ 22 m, and the maximum axial angle is 30° (Wang et al2011b), so the EMR signals monitored by the orthogonal antennas in this paper are actually local signals (Gershenzon and Bambakidis 2001). The coverage areas are shown in figure 12. When the width of the tunnel, 4.6m, is known, and if the valid monitoring range is 10m, then the effective monitoring distance of the coal wall of 1# antenna can be described by L=L12−(D2)2−(D2)ctgθ.1 Figure 12. Open in new tabDownload slide The mechanism analysis of effective MS and EMR signals on the blasting face. 1-1#EMR antenna; 2-2# EMR antenna; 3-MS transducer. Figure 12. Open in new tabDownload slide The mechanism analysis of effective MS and EMR signals on the blasting face. 1-1#EMR antenna; 2-2# EMR antenna; 3-MS transducer. In this formula, L is the effective monitoring distance of the coal wall, L1 is the valid monitoring distance of 1# antenna, 10m, D is the width of tunnel, 4.6m, and θ is the maximum axial angle, 30°. By this calculation, the effective monitoring distance of the coal wall is 5.75m. 4.2. The formation of three different forms of EMR signals It is well-known that the charge separation is a prerequisite and foundation for EMR. The common methods of charge separation in coal and rock mass are the piezoelectric effect (Nitsan 1977), the sliding and friction effect (Zhao et al2014, Wang 2015), the fracture effect (Brady and Rowell 1986, Frid 1997b) and the hydrogen bonding damage effect (Li et al2014a), etc. Obviously, when the roadway is not influenced by mining and driving, the coal and rock mass is subject to the weight of the overlying rock layers, and in a state of flux, which results in the EMR background signals (Wang et al2011a). During the blasting operation, the P-waves of the MS signals spread to the coal mass near to the monitoring point for the first time. Because of the existence of the micro-crack, the compression wave may reflect as a tensile one, causing the evolution of fissures (figure 13(a)). Then the fissures’ evolution form single pulse EMR signals (I). When contrasting figures 8, 10 and 12, it is known that the monitoring areas of the two antennas are relatively close, so the pulse EMR signals are basically synchronous, and the energetic relevant degree is extremely high. As the monitoring area A is nearer to the driving face, the single pulse EMR signals are monitored earlier by 1# antenna. Figure 13. Open in new tabDownload slide The formation mechanism of three different forms of EMR signals. (a) Fracture effect of the P-waves. (b) RC oscillating-loop effect of the P-waves. (c) Sliding and friction effect of the S-waves. The direction of wave propagation;   The direction of particle motion. Figure 13. Open in new tabDownload slide The formation mechanism of three different forms of EMR signals. (a) Fracture effect of the P-waves. (b) RC oscillating-loop effect of the P-waves. (c) Sliding and friction effect of the S-waves. The direction of wave propagation;   The direction of particle motion. Besides producing a fracture effect, the P-waves of the MS signals also may lead to the rapid opening and closing of cracks. As the electric charge in the crack tip end may not have enough time to transfer, as well as the fact that coal is not a good conductor of electricity, both crack surfaces might have opposite charge. The two surfaces constitute a capacitor plate, together with the surrounding medium, generating a resistance and capacitance (RC) oscillating-loop (figure 13(b)). It is known that the relative motion of the two electrode plates of the capacitor is the rapid charge-discharge process, in which EMR occurs. The relatively everlasting oscillation effect might primarily cause clusters of EMR signals (II). Because of the existing of relaxation phenomenon, the formation of an RC oscillating-loop is always relatively lagging. Besides that, since the EMR signals monitored are mainly local signals, and the energy that attenuates both the stress and seismic waves are relatively slow, obviously the fissures evolution and RC oscillating-loop are largely driven by the explosion charge, which would explain why the energy of the EMR signals mainly depends on the blasting cartridge. The S-wave, especially the low frequency components, play a rubbing and friction role on the coal mass (figure 13(c)). These acts often have the characteristics of long duration and low-frequency, and, as a result, the EMR signals also have the similar properties (figure 8). Moreover, the energy of these low frequency oscillation EMR signals is always weak, and is affected by the propagation direction, so the signals can only be captured by 2# antenna, which points to the coal wall. Considering the above-mentioned factors, it is possible for good temporal synchronization and energy-correlation between the MS and EMR signals. The pulse EMR signals with similar characteristics can be monitored by both antennas pointing to the coal wall and to the blasting workface. Besides that, the EMR signals captured by 2# antenna also contain low frequency oscillation components, and these components would only occur during the period of S-waves of MS signals. 5. Conclusions The MS and EMR signals caused by blasting on an excavated working face are actually homologous in some way. They both represent good time-synchronization and energy-correlation. The single pulse (I) and clusters of EMR signals (II) can be monitored by both antennas pointing to the coal wall and to the blasting workface. These signals have similar characteristics and energy accumulation, and are greatly affected by the cartridge count. The EMR signals captured by 2#antenna also contain low frequency oscillation components, which show the longer duration features, and these lower frequency components also show better synchronicity with the low frequency components of the MS signals. The EMR monitoring signals caused by blasting are supposed to be local signals. The P-waves of the MS signals firstly spread to the coal mass near to the monitoring point, making fissures evolution and a single pulse EMR signal. The rapid and continuous opening and closing of cracks inside the RC oscillating circuits, like the structure in the coal seam, finally result in the generation of clustered EMR signals. The S-waves, especially the low frequency components, play a rubbing and friction role on the coal mass. These relatively long duration and low-frequency actions finally result in the oscillations components of the EMR signals. Affected by the energy and propagation direction, these signals can only be captured by 2# antenna, which points to the coal wall. Acknowledgment This research is supported by National Natural Science Foundation project (51274206, 51404277), the Program of Central Universities Fundamental Research Funds from the Ministry of Education of China (2014QZ05). These support are greatly acknowledged and appreciated. References Abuov M G , Aitaliev S M , Ermekov T M , Zhanbyrbaev N B , Kayupov M A . , 1988 Studies of the effect of dynamic processes during explosive break-out upon the roof of mining excavations , Sov. Min. Sci. , vol. 24 (pg. 581 - 590 ) 10.1007/BF02498618 Google Scholar Crossref Search ADS WorldCat Crossref Arjang B , Herget G . , 1997 In situ ground stresses in the Canadian hardrock mines: an update , Int. J. Rock Mech. Min. Sci. , vol. 34 (pg. 15.e1 - 15.e.6 ) 10.1016/S1365-1609(97)00039-7 OpenURL Placeholder Text WorldCat Crossref Barlow J S . , 1973 , Autocorrelation and Crosscorrelation Analysis Frequency and Correlation Analysis , vol. vol 5 Amsterdam Elsevier Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Blair D P . , 1995 Soil-embedded detector mounts for seismic monitoring , Geophysics , vol. 60 (pg. 120 - 133 ) 10.1190/1.1443739 Google Scholar Crossref Search ADS WorldCat Crossref Blair D P , Jiang J J . , 1995 Surface vibrations due to a vertical column of explosive , Int. J. Rock Mech. Min. Sci. Geomech. Abst. , vol. 32 (pg. 149 - 154 ) 10.1016/0148-9062(94)00036-3 Google Scholar Crossref Search ADS WorldCat Crossref Brady B H G , Brown E T . , 2013 , Rock Mechanics: for Underground Mining Berlin Springer Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Brady B T , Rowell G A . , 1986 Laboratory investigation of the electrodynamics of rock fracture , Nature , vol. 321 (pg. 488 - 492 ) 10.1038/321488a0 Google Scholar Crossref Search ADS WorldCat Crossref Butler K E , Simser B , Bowes R . , 2001 , An Experimental Field Trial of Passive Seismoelectric Monitoring Ottawa, ON Canadian Geophysical Union Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Cao A Y , Dou L M , Chen G X , Gong S Y , Wang Y G , Li Z H . , 2008 Focal mechanism caused by fracture or burst of a coal pillar , J. Chin. Univ. Min. Technol. , vol. 18 (pg. 153 - 158 ) 10.1016/S1006-1266(08)60033-4 Google Scholar Crossref Search ADS WorldCat Crossref Chen S H , Wang E Y . , 2011 Electromagnetic radiation signals of coal or rock denoising based on morphological filter , Proc. Eng. , vol. 26 (pg. 588 - 594 ) 10.1016/j.proeng.2011.11.2210 Google Scholar Crossref Search ADS WorldCat Crossref Daubechies I . , 1992 , Ten Lectures on Wavelets Madison, WI Capital City Press Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dou L M , Lu C P , Mu Z L , Gao M S . , 2009 Prevention and forecasting of rock burst hazards in coal mines , Min. Sci. Technol. , vol. 19 (pg. 585 - 591 ) 10.1016/S1674-5264(09)60109-5 OpenURL Placeholder Text WorldCat Crossref Enomoto Y , Hashimoto H . , 1990 Emission of charged particles from indentation fracture of rocks , Nature , vol. 346 (pg. 641 - 643 ) 10.1038/346641a0 Google Scholar Crossref Search ADS WorldCat Crossref Frid V . , 1997a Electromagnetic radiation method for rock and gas outburst forecast , J. Appl. Geophys. , vol. 38 (pg. 97 - 104 ) 10.1016/S0926-9851(97)00017-7 Google Scholar Crossref Search ADS WorldCat Crossref Frid V . , 1997b Rockburst hazard forecast by electromagnetic radiation excited by rock fracture , Rock Mech. Rock Eng. , vol. 30 (pg. 229 - 236 ) 10.1007/BF01045719 Google Scholar Crossref Search ADS WorldCat Crossref Frid V . , 1999 Electromagnetic radiation associated with induced triaxial fracture in granite , Phil. Mag. Lett. , vol. 79 (pg. 79 - 86 ) 10.1080/095008399177570 Google Scholar Crossref Search ADS WorldCat Crossref Frid V , Vozoff K . , 2005 Electromagnetic radiation induced by mining rock failure , Int. J. Coal Geol. , vol. 64 (pg. 57 - 65 ) 10.1016/j.coal.2005.03.005 Google Scholar Crossref Search ADS WorldCat Crossref Frid V I , Shabarov A N , Proskuryakov V M , Baranov V A . , 1992 Formation of electromagnetic radiation in coal stratum , J. Min. Sci. , vol. 28 (pg. 139 - 145 ) 10.1007/BF00710732 Google Scholar Crossref Search ADS WorldCat Crossref Fujii Y , Ishijima Y , Deguchi G . , 1997 Prediction of coal face rockbursts and microseismicity in deep longwall coal mining , Int. J. Rock Mech. Min. Sci. , vol. 34 (pg. 85 - 96 ) 10.1016/S1365-1609(97)80035-4 Google Scholar Crossref Search ADS WorldCat Crossref Gao M S , Dou L M , Zhang N , Xie Y S , Zheng B S , Gao M S . , 2008 Simulation of the relationship between roadway dynamic destruction and hypocenter parameters , J. Chin. Univ. Min. Technol. , vol. 18 (pg. 93 - 97 ) 10.1016/S1006-1266(08)60020-6 Google Scholar Crossref Search ADS WorldCat Crossref Ge M C . , 2005 Efficient mine microseismic monitoring , J. Coal Geol. , vol. 64 (pg. 44 - 56 ) 10.1016/j.coal.2005.03.004 Google Scholar Crossref Search ADS WorldCat Crossref Gershenzon N , Bambakidis G . , 2001 Modeling of seismo-electromagnetic phenomena , Russ. J. Earth Sci. , vol. 3 (pg. 247 - 275 ) 10.2205/2001ES000058 Google Scholar Crossref Search ADS WorldCat Crossref Gokhberg M B , Morgunov V A , Aronov E L . , 1979 High-frequency electromagnetic-radiation during the seismic activity , Dokl. Akad. Nauk SSSR , vol. 248 (pg. 1077 - 1081 ) OpenURL Placeholder Text WorldCat Guo Z Q , Luo X L , Qian S Q , Zhao G F , Guo Z Q , Chen G H . , 1999 Electromagnetic emissions during quarry blasting , Chin. J. Geophys. , vol. 42 (pg. 834 - 840 ) 10.3321/j.issn:0001-5733.1999.06.014 OpenURL Placeholder Text WorldCat Crossref Hardy H R Jr . , 1981 , Applications of Acoustic Emission Techniques to Rock and Rock Structures: a State-of-the-Art Review Philadelphia, PA Astm Special Technical Publications pg. p 89 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hardy H R Jr . , 1986 , Acoustic Emission/Microseismic Activity, Volume 1: Principles, Techniques, and Geotechnical Applications Acoustic Emission/Microseismic Activity Rotterdam A A Balkema Publishers Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hargather M J , Settles G S . , 2007 Optical measurement and scaling of blasts from gram-range explosive charges , Shock Waves , vol. 17 (pg. 215 - 223 ) 10.1007/s00193-007-0108-8 Google Scholar Crossref Search ADS WorldCat Crossref He X Q , Wang E Y , Liu Z T . , 1999 The general charastics of electromagnetic radiation during coal fracture and its application in outburst prediction Proc. of the 8th US Mine Ventilation Symp. Rolla, Missouri 11–17 June Tien J C . (pg. 81 - 84 ) He X Q , Wang E Y , Nie B S , Liu M J , Zhang L . , 2003 , Electromagnetic Dynamics of Coal or Rock Rheology Beijing Science Press Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Huang N E . , 2000 New method for nonlinear and nonstationary time series analysis: empirical mode decomposition and Hilbert spectral analysis AeroSense 2000. Int. Society for Optics and Photonics (pg. 197 - 209 ) 10.1117/12.381681 Huang N E , Shen Z , Long S R , Wu M C , Shih H H , Zheng Q , Liu H H . , 1998 The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , Proc. R. Soc. , vol. 454 (pg. 903 - 995 ) 10.1098/rspa.1998.0193 Google Scholar Crossref Search ADS WorldCat Crossref Iannacchione A T , Zelanko J C . , 1995 Occurrence and remediation of coal mine bumps: a historical review Paper in Proc.: Mechanics and Mitigation of Violent Failure in Coal and Hard-Rock Mines US Bureau of Mines Spec. Publ,: 01-95 Jiang Y D , Wang H W , Xue S , Zhao Y X , Zhu J , Pang X F . , 2012 Assessment and mitigation of coal bump risk during extraction of an island longwall panel , Int. J. Coal Geol. , vol. 95 (pg. 20 - 33 ) 10.1016/j.coal.2012.02.003 Google Scholar Crossref Search ADS WorldCat Crossref Kamen E W , Heck B S . , 2000 , Fundamentals of Signals and Systems: Using the WEB and MATLAB Englewood Cliffs, NJ Prentice Hall Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lee Y W , Cheatham T P , Wiesner J B . , 1950 Application of correlation analysis to the detection of periodic signals in noise , Proc. IRE , vol. 38 (pg. 1165 - 1171 ) 10.1109/JRPROC.1950.233423 Google Scholar Crossref Search ADS WorldCat Crossref Li C W , Wang J G , Xie B J , Dong L H , Sun Y F , Cao X . , 2014a FTIR study of hydrogen bonds in coal under drop weight impact testing , Spectrosc. Spect. Anal. , vol. 34 (pg. 2961 - 2967 ) 10.3964/j.issn.1000-0593(2014)11-2961-07 OpenURL Placeholder Text WorldCat Crossref Li C W , Wang C , Xie B J , Sun X Y , Xu X M . , 2014b Study on low-frequency TEM effect of coal during dynamic rupture , Shock Vib. , vol. 2015 189569 10.1155/2015/189569 OpenURL Placeholder Text WorldCat Crossref Li T , Cai M F , Cai M . , 2007 Earthquake-induced unusual gas emission in coalmines—A km-scale in situ experimental investigation at Laohutai mine , Int. J. Coal Geol. , vol. 71 (pg. 209 - 224 ) 10.1016/j.coal.2006.08.004 Google Scholar Crossref Search ADS WorldCat Crossref Lu C P , Dou L M . , 2014 The relationship between vertical stress gradient, seismic, and electromagnetic emission signals at sanhejian coal mine, China , Int. J. Rock Mech. Min. Sci. , vol. 70 (pg. 90 - 100 ) 10.1016/j.ijrmms.2014.04.010 Google Scholar Crossref Search ADS WorldCat Crossref Lu C P , Dou L M , Liu H , Liu H S , Liu B , Du B B . , 2012 Case study on microseismic effect of coal and gas outburst process , Int. J. Rock Mech. Min. Sci. , vol. 53 (pg. 101 - 110 ) 10.1016/j.ijrmms.2012.05.009 Google Scholar Crossref Search ADS WorldCat Crossref Lu W B , Li P , Chen M , Zhou C B , Shu D Q . , 2011 Comparison of vibrations induced by excavation of deep-buried cavern and open pit with method of bench blasting , J. Cent. South Univ. Technol. , vol. 18 (pg. 1709 - 1718 ) 10.1007/s11771-011-0892-2 Google Scholar Crossref Search ADS WorldCat Crossref Martner S T , Sparks N R . , 1959 The electroseismic effect , Geophysics , vol. 24 (pg. 297 - 308 ) 10.1190/1.1438585 Google Scholar Crossref Search ADS WorldCat Crossref Mesec J , Kovac I , Soldo B . , 2010 Estimation of particle velocity based on blast event measurements at different rock units , Soil Dynam. Earthquake Eng. , vol. 30 (pg. 1004 - 1009 ) 10.1016/j.soildyn.2010.04.011 Google Scholar Crossref Search ADS WorldCat Crossref Migunov N I , Sobolev G A , Khromov A A . , 1985 Natural electromagnetic radiation in seismically active regions , USSR Report Earth Sciences JPRS UES 1 (pg. 63 - 64 ) Milev A M , Spottiswoode S M , Rorke A J , Finnie G J . , 2001 Seismic monitoring of a simulated rockburst on a wall of an underground tunnel , J. South Afr. Inst. Min. Metall. , vol. 101 (pg. 253 - 260 ) OpenURL Placeholder Text WorldCat Mori Y , Obata Y . , 2008 Electromagnetic emission and AE Kaiser Effect for estimating rock in situ stress Nihon University www.cit.nihon-u.ac.jp/laboratorydata/kenkyu/publication/report/93.pdf Mori Y , Obata Y , Sikula J . , 2009 Acoustic and electromagnetic emission from crack created in rock sample under deformation , J. Acoust. Emiss. , vol. 27 (pg. 157 - 167 ) OpenURL Placeholder Text WorldCat Nesbitt A C , Austin B A . , 1988 The emission and propagation of electromagnetic energy from stressed quartzite rock underground , Trans. SA Inst. Elect. Eng. , vol. 79 (pg. 53 - 57 ) OpenURL Placeholder Text WorldCat Nie B S , He X Q , Liu F B , Ping W . , 2009 Chaotic characteristics of electromagnetic emission signals during deformation and fracture of coal , Min. Sci. Technol. , vol. 19 (pg. 189 - 193 ) 10.1016/S1674-5264(09)60036-3 OpenURL Placeholder Text WorldCat Crossref Nitsan U . , 1977 Electromagnetic emission accompanying fracture of quartz-bearing rocks , Geophys. Res. Lett. , vol. 4 (pg. 333 - 336 ) 10.1029/GL004i008p00333 Google Scholar Crossref Search ADS WorldCat Crossref Obert L , Duvall W I . , 1957 , Micro-Seismic Method of Determining the Stability of Underground Openings Washington, DC US Government Printing Office Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC O’Keefe S G , Thiel D V . , 1991 Electromagnetic emissions during rock blasting , Geophys. Res. Lett. , vol. 18 (pg. 889 - 892 ) 10.1029/91GL01076 Google Scholar Crossref Search ADS WorldCat Crossref Pralat A , Wojtowicz S . , 2004 Electromagnetic & acoustic emission from the rock-experimental measurements , Acta. Geodyn. et. Geomater. , vol. 1 (pg. 111 - 119 ) OpenURL Placeholder Text WorldCat Rabinovitch A , Bahat D , Frid V . , 2002 Similarity and dissimilarity of electromagnetic radiation from carbonate rocks under compression, drilling and blasting , Int. J. Rock Mech. Min. Sci. , vol. 39 (pg. 125 - 129 ) 10.1016/S1365-1609(02)00012-6 Google Scholar Crossref Search ADS WorldCat Crossref Sambuelli L . , 2009 Theoretical derivation of a peak particle velocity–distance law for the prediction of vibrations from blasting , Rock Mech. Rock Eng. , vol. 42 (pg. 547 - 556 ) 10.1007/s00603-008-0014-0 Google Scholar Crossref Search ADS WorldCat Crossref Scott D F , Williams T J , Knoll S J . , 2004 Investigation of electromagnetic emissions in a deep underground mine Proc. 23rd Int. Conf. on Ground Control in Mining. (pg. 125 - 132 ) Segarra P , López L M , Sanchidrián J A . , 2012 Uncertainty in measurements of vibrations from blasting , Rock Mech. Rock Eng. , vol. 2012 (pg. 1 - 8 ) 10.1007/s00603-012-0229-y OpenURL Placeholder Text WorldCat Crossref Serebryakova O N , Bilichenko S V , Chmyrev V M , Parrot M , Rauch J L , Lefeuvre F , Pokhotelov O A . , 1992 Electromagnetic ELF radiation from earthquake regions as observed by low-altitude satellites , Geophys. Res. Lett. , vol. 19 (pg. 91 - 94 ) 10.1029/91GL02775 Google Scholar Crossref Search ADS WorldCat Crossref Sheriff R E , Geldart L P . , 1995 , Exploration Seismology Cambridge Cambridge University Press Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sobolev G A , Demin V M , Narod B B , Whaite P . , 1984 Tests of piezoelectric and pulsed-radio methods for quartz vein and base-metal sulfides prospecting at Giant Yellowknife Mine, NWT, and Sullivan Mine, Kimberley, Canada , Geophysics , vol. 49 (pg. 2178 - 2185 ) 10.1190/1.1441633 Google Scholar Crossref Search ADS WorldCat Crossref Tomizawa I , Hayakawa M , Yoshino T , Ohta K , Okada T , Sakai H . , 1994 Observation of ELF/VLF electromagnetic variations associated with a seismic experimental explosion , Electromagnetic Phenomena Related to Earthquake Prediction Tokyo Terra Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wang C . , 2015 Experimental study on the characteristics of transient electromagnetic signal emerged during the local destruction of coal by hammer impact , Chin. Univ. Min. Technol. OpenURL Placeholder Text WorldCat Wang E Y , He X Q , Liu X F , Li Z H , Wang C , Xiao D . , 2011a A non-contact mine pressure evaluation method by electromagnetic radiation , J. Appl. Geophys. , vol. 75 (pg. 338 - 344 ) 10.1016/j.jappgeo.2011.06.028 Google Scholar Crossref Search ADS WorldCat Crossref Wang E Y , He X Q , Wei J , Nie B S , Song D . , 2011b Electromagnetic emission graded warning model and its applications against coal rock dynamic collapses. , Int. J. Rock Mech. Min. Sci. , vol. 48 (pg. 556 - 564 ) 10.1016/j.ijrmms.2011.02.006 Google Scholar Crossref Search ADS WorldCat Crossref Wang C , Li C W , Wang J G , Hu P , Cui Y G , Xu X M . , 2014 Characteristics of electromagnetic signals obtained during blasting rupture of coal in excavating workface , Progress in Mine Safety Science and Engineering II Leiden CRC Press (pg. 63 - 70 ) 10.1201/b16606-14 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Crossref Wang E Y , Liu X F , Zhao E L , Liu Z T . , 2009 Prediction and control of rock burst of coal seam contacting gas in deep mining , J. Coal Sci. Eng. , vol. 15 (pg. 152 - 156 ) 10.1007/s12404-009-0208-4 Google Scholar Crossref Search ADS WorldCat Crossref Wang T , Zhang M C , Yu Q H , Zhang H Y . , 2012 Comparing the applications of EMD and EEMD on time–frequency analysis of seismic signal , J. Appl. Geophys. , vol. 83 (pg. 29 - 34 ) 10.1016/j.jappgeo.2012.05.002 Google Scholar Crossref Search ADS WorldCat Crossref Wu Z H , Huang N E . , 2009 Ensemble empirical mode decomposition: a noise-assisted data analysis method , Adv. Adapt. Data Anal. , vol. 1 (pg. 1 - 41 ) 10.1142/S1793536909000047 Google Scholar Crossref Search ADS WorldCat Crossref Xu W M , Wu P Z . , 1991 Experimental study of electromagnetic emission during rock fracture , Adv. Geophys. Res. , vol. 2 (pg. 73 - 83 ) 10.1016/b978-0-08-036390-5.50011-7 OpenURL Placeholder Text WorldCat Crossref Xu X F , Dou L M , Lu C P , Zhang Y . , 2010 Frequency spectrum analysis on micro-seismic signal of rock bursts induced by dynamic disturbance , Min. Sci. Technol. , vol. 20 (pg. 682 - 685 ) 10.1016/S1674-5264(09)60262-3 OpenURL Placeholder Text WorldCat Crossref Yamada I , Masuda K , Mizutani H . , 1989 Electromagnetic and acoustic emission associated with rock fracture , Phys. Earth Planet. Inter. , vol. 57 (pg. 157 - 168 ) 10.1016/0031-9201(89)90225-2 Google Scholar Crossref Search ADS WorldCat Crossref Yang J H , Lu W B , Chen M , Yan P , Zhou C B . , 2013 Microseism induced by transient release of in situ stress during deep rock mass excavation by blasting , Rock Mech. Rock Eng. , vol. 46 (pg. 859 - 875 ) 10.1007/s00603-012-0308-0 Google Scholar Crossref Search ADS WorldCat Crossref Yang W . , 2014 Study on the association response mechanism and characteristics of electromagnetic and microseismic signals during the deformation and fracture of coal or rock , Chin. Univ. Min. Technol. (In Chinese) OpenURL Placeholder Text WorldCat Yang W , Lin B Q , Qu Y A , Li Z W , Zhai C , Jia L L , Zhao W Q . , 2011 Stress evolution with time and space during mining of a coal seam , Int. J. Rock Mech. Min. Sci. , vol. 48 (pg. 1145 - 1152 ) 10.1016/j.ijrmms.2011.07.006 Google Scholar Crossref Search ADS WorldCat Crossref Yasitli N E , Unver B . , 2005 3D numerical modeling of longwall mining with top-coal caving , Int. J. Rock Mech. Min. Sci. , vol. 42 (pg. 219 - 235 ) 10.1016/j.ijrmms.2004.08.007 Google Scholar Crossref Search ADS WorldCat Crossref Zhao T B , Yin Y C , Xiao F K , Tan Y L , Zou J C . , 2014 Rockburst disaster prediction of isolated coal pillar by electromagnetic radiation based on frictional effect , Shock Vib. , vol. 2014 814050 10.1155/2014/814050 OpenURL Placeholder Text WorldCat Crossref © 2016 Sinopec Geophysical Research Institute
TI - The correlated characteristics of micro-seismic and electromagnetic radiation signals on a deep blasting workface
JF - Journal of Geophysics and Engineering
DO - 10.1088/1742-2132/13/6/1020
DA - 2016-12-01
UR - https://www.deepdyve.com/lp/oxford-university-press/the-correlated-characteristics-of-micro-seismic-and-electromagnetic-sWOTZ57WyM
SP - 1020
VL - 13
IS - 6
DP - DeepDyve
ER -