基于WPT-LMD预处理的震动信号有效辨识方法研究  

Study on effective seismic signals identification method based on WPT-LMD pre-processing method

作　　者：徐晓萌[1] 吕平洋[1] 王菲茵[1] 赵越超[1] 刘文彪[1] 

机构地区：[1]中国矿业大学(北京)资源与安全工程学院,北京市海淀区100083

出　　处：《中国煤炭》2016年第3期37-42,共6页China Coal

基　　金：国家自然科学基金资助项目(51274206);国家自然科学青年基金资助项目(51404277)

摘　　要：提出了一种联合小波包分解(WPT)和局域均值分解(LMD)的信号预处理方法,并在此基础上使用改进的STA/LTA方法实现对震动信号的有效辨识。研究结果表明WPT-LMD预处理方法是一种省时、高效、自适应强的信号处理方法,对类微震的非线性、非平稳、低信噪比信号有较好的应用效果,且能避免传统时频信号降噪处理方法应用时造成震相初至时刻偏移的弊端;联合改进的STA/LTA方法对超声波波速初至时刻拾取、实验室声发射测试、矿井现场爆破微震信号均表现了良好的应用效果;该种信号预处理方法特别适合于复杂煤层结构的微弱震动信号处理,开展此类研究将有助于提高煤矿现场微震监测的精度和效率。In this paper,a pre-processing method was proposed which was the combination of wavelet packet transform（WPT） and local mean decomposition（LMD） method,on this basis a modified method short/long time average ratio（STA/LTA） was applied to the effective identification of seismic signals.The results showed that WPT-LMD pre-processing method was a timesaving,efficient signal processing method that had good adaptability,it had preferable application effects on non-linear,non-stationary,low signal-noise ratio（SNR） signals of micro-seismic,it was better than traditional time-frequency signal noise reduction method to avoid first-arrival time shifting.Combined with modified STA/LTA method,it achieved good application effects on ultrasonic wave velocity test,laboratorial acoustic emission test and field site microseismic signals induced by blasting.The signal pre-processing method was especially fit for micro-seismic signal processing in complicated seam,and this research would help improving the accuracy and efficiency of micro-seismic monitoring work in coal mine field.

关 键 词：煤岩动力灾害 小波包变换 局域均值分解 STA/LTA方法 微震初相 

分 类 号：TD326[矿业工程—矿井建设] TD713