Non-parametric automatic microseismic data denoising via PD method and its application
PENG Ping?an1，2，WANG Liguan1，2，PEI Anlei2
(1. School of Resources and Safety Engineering，Central South University，Changsha，Hunan 410083，China；2. Changsha Digital Mine Co.，Ltd.，Changsha，Hunan 410083，China)
摘要工程微震监测中常包含较多低信噪比信号。现有去噪方法存在参数多且需要手工调节等问题，为此，提出一种针对微震信号的无参数自动去噪PD(pick and denoise)算法。算法首先通过改进AIC(akaike information criterion)方法初步拾取P波初至，得到信号的背景噪声段，通过傅里叶变换在频率域上提取噪声功率谱信息，在此基础上从微震信号的功率谱中减去噪声功率谱，最后应用傅里叶逆变换还原得到去噪后微震信号。利用Matlab人工合成不同类型、不同信噪比的含噪信号，应用PD算法进行去噪并与EEMD(ensemble empirical mode decomposition)、小波去噪方法进行比较，结果表明：PD算法去噪后的平均绝对误差和误差标准差均优于EEMD和小波去噪方法，并且对于低信噪比信号，PD算法仍具有良好的效果。最后将PD算法应用于陕西省引汉济渭工程秦岭4#支洞微震监测工程中，对2 730条微震信号进行去噪分析，得平均P波信噪比从滤波前的16.49提高到了35.62，表明PD算法对于提高工程微震信号质量具有良好的应用价值。
Abstract：Microseismic signals are often highly corrupted by unwanted noise in engineering. The performance of existing denoising methods depends on the accuracy of selected parameters that need to tune manually. Thus，we have proposed a non-parametric automatic denoising algorithm for microseismic data，named PD method. In this method，we use a modified AIC(akaike information criterion) algorithm to obtain the background noise of the signal，then the noise power spectrum is extracted by Fourier transform in the frequency domain. Next，the noise power spectrum is subtracted from the signal power spectrum，and then we can recover the microseismic signal by inverse Fourier transform. We have tested PD method by synthesized signals with different types and different signal-to-noise ratios using Matlab and compared the result with EEMD and wavelet denoising method. Results show that the mean absolute error and standard deviation of the denoised waveform after PD method are better than that after EEMD and wavelet denoising. For signals with low signal-to-noise ratios，PD method still has a good performance. We have denoised 2 730 microseismic signals recorded by a microseismic monitoring system in Qinling No.4 inclined shaft of Shaanxi Yinhanjiwei project，China. The average P-wave signal-to-noise ratio is increased from 16.49 to 35.62 after PD method denoising. The results show the effectiveness of the proposed method for improving microseismic data quality.
夏 森，王维波，李树荣，等. 微地震信号的参数辨识建模及其Kalman滤波[J]. 地球物理学进展，2016，31(5)：2 005–2 010.(XIA Sen，WANG Weibo，LI Shurong，et al. Parameter identification modeling of microseismic signals and Kalman filtering[J]. Progress in Geophysics，2016，31(3)：2 005–2 010.(in Chinese))
[2]
胡广书. 数字信号处理导论[M]. 北京：清华大学出版社，2005：233–238.(HU Guangshu. Introduction to digital signal processing[M]. Beijing：Tsinghua University Press，2005：233–238.(in Chinese))
[10]
刘劲松，王 赟，姚振兴. 微地震信号到时自动拾取方法[J]. 地球物理学报，2013，56(5)：1 660–1 666.(LIU Jinsong，WANG Yun，YAO Zhenxing. On micrseismic first arrival identification：a case study[J]. Chinese Journal of Geophysics，2013，56(5)：1 660–1 666.(in Chinese))
[12]
ZHANG H，THURBER C，ROWE C. Automatic P-wave arrival detection and picking with multiscale wavelet analysis for single-component recordings[J]. Bulletin of the Seismological Society of America，2003，93(5)：1 904–1 912.
[13]
MAEDA N. A method for reading and checking phase times in auto-processing system of seismic wave data[J]. Zisin，1985，38(3)：365–379.
[1]
徐奴文，唐春安，周 钟，等. 岩石边坡潜在失稳区域微震识别方法[J]. 岩石力学与工程学报，2011，30(5)：893–900.(XU Nuwen，TANG Chun?an，ZHOU Zhong，et al. Identification method of potential failure regions of rock slope using microseismic monitoring technique[J]. Chinese Journal of Rock Mechanics and Engineering. 2011，30(5)：893–900.(in Chinese))
[9]
MOUSAVI S M，LANGSTON C A，HORTON S P. Automatic microseismic denoising and onset detection using the synchrosqueezed continuous wavelet transform[J]. Geophysics，2016，81(4)：341–355.
[11]
赵大鹏，刘希强，李 红，等. 峰度和AIC方法在区域地震事件和直达P波初动自动识别方面的应用[J]. 地震研究，2012，35(2)：220–225.(ZHAO Dapeng，LIU Xiqiang，LI Hong，et al. Detection of regional seismic events by kurtosis method and automatic identification of direct P-wave first motion by Kurtosis-Aic method[J]. Journal of Seismological Research，2012，35(2)：220–225.(in Chinese))
[5]
HAN J，VAN DERBAAN M. Microseismic and seismic denoising via ensemble empirical mode decomposition and adaptive thresholding[J]. Geophysics，2015，80(6)：69–80.
[7]
徐宏斌，李庶林，陈际经. 基于小波变换的大尺度岩体结构微震监测信号去噪方法研究[J]. 地震学报，2012，34(1)：85–96.(XU Hongbin，LI Shulin，CHEN Jijing. A study on method of signal denoising based on wavelet transform for micro-seismicity monitoring in large-scale rockmass structures[J]. Acta Seismologica Sinica，2012，34(1)：85–96.(in Chinese))
[15]
WU Z，HUANG N E. Ensemble empirical mode decomposition：a noise-assisted data analysis method[J]. Advances in adaptive data analysis，2009，1(1)：1–41.
[4]
李夕兵，张义平，左宇军，等. 岩石爆破振动信号的EMD滤波与消噪[J]. 中南大学学报：自然科学版，2006，37(1)：150–154.(LI Xibing，ZHANG Yiping，ZUO Yujun，et al. Filtering and denoising of rock blasting vibration signal with EMD[J]. Journal of Central South University：Science and Technology，2006，37(1)：150–154. (in Chinese))
[6]
GACI S. A new ensemble empirical mode decomposition(EEMD) denoising method for seismic signals[J]. Energy Procedia，2016，97(1)：84–91.
[8]
李学龙，李忠辉，王恩元，等. 矿山微震信号干扰特征及去噪方法研究[J]. 中国矿业大学学报，2015，44(5)：788–792.(LI Xuelong，LI Zhonghui，WANG Enyuan，et al. Study of mine microseismic signals interference characteristic and its denoising method[J]. Journal of China University of Mining and Technology，2015，44(5)：788–792. (in Chinese))
[14]
RAO K R，KIM D N，HWANG J J. Fast Fourier transform-algorithms and applications[M]. New York：Springer Science and Business Media，2011：1–60.
2019, Vol. 38 
