摘要
为克服传统阈值法定位在岩石声发射定位中的局限,通过设置在岩石试件表面的压电传感器记录加载过程产生的弹性波信号,采用小波变换及互相关分析法对该信号进行处理,得到不同通道之间的时间延迟,从而实现声发射源的定位。首先将细砂岩加工成109 mm×200 mm×20 mm的板状试件,利用数字点荷载仪按恒定加载速率对其进行加载,采用4个压电传感器以3 MHz频率记录各通道的声发射信号全波形;然后通过sym5小波基对该信号进行分解和重构,再利用互相关原理计算得到各通道之间的时间延迟;最后利用自编的MATLAB代码追溯到声发射源坐标,并与传统阈值法定位结果进行对比。结果表明,小波变换互相关分析方法有效减小了声发射源定位的平均绝对误差和平均相对误差,提高了岩石声发射定位精度。
In order to overcome the limitations of rock acoustic emission localization methods,such as threshold method,this paper has adopted the wavelet transform and cross-correlation analysis method to process the signal,and obtain the time delay between all the channels to realize the localization of the acoustic emission source. For the said purpose,first of all,we have managed to process the fine sandstone into plate-form specimens of109 mm × 200 mm × 20 mm,which can be loaded at a constant loading place by a digital point loader,whereas the 4 piezoelectric sensors tends to be used for recording the full waveforms of the acoustic emission signals of each channel at a frequency of 3 MHz. And,then,the signals can be decomposed and reconstructed by the so-called sym5 wavelets,wherein the time delay between the channels can be worked out in accordance with the principle of cross-correlation. And,last of all,it would be possible to use our self-programmed MATLAB code to account for the acoustic emission source coordinates in comparison with the outcome of the traditional threshold. The results of our localization of the acoustic emission source can be stated as follows:( 1)Using the said sym5 wavelet,the signals collected by the sensor can be decomposed and reconstructed in 5 layers,whereas the noise of the reconstructed signals can be processed more appropriately to keep on the main features of the original waveforms;( 2) On the basis of the wavelet transform,it would be possible to conduct the cross-correlation analysis of the reconstructed signal more accurately. Thus,according to the hysteresis of the peak value deduced by the cross-correlation function,it would be very easy to work out the time delay for the signal to reach all the sensors. Thus,the acoustic emission source produced by the sandstone damage can be inverted and transformed through the self-programmed MATLAB code;( 3) As compared with the actual position of the loading point, the mean absolute error( MAE) of the threshold method in the X and Y axes ca
作者
吴鑫
晏巧
张婷婷
彭雅雯
李国辉
WU Xin;YAN Qiao;ZHANG Ting-ting;PENG Ya-wen;LI Guo-hui(Institute of Technology,Sichuan Normal University,Chengdu 610101,China;State Key Laboratory of Hydraulics and Mountain River Engineering,Sichuan University,Chengdu 610065,China)
出处
《安全与环境学报》
CAS
CSCD
北大核心
2019年第5期1611-1617,共7页
Journal of Safety and Environment
基金
国家安监总局安全生产重特大事故防治关键技术科技项目(Sichuan-0004-2017AQ,Sichuan-0011-2018AQ)
四川省教育厅重点项目(18ZA0407)
关键词
安全工程
声发射
小波变换
互相关分析
声源定位
safety engineering
acoustic emission(AE)
wavelet transform
cross-correlation analysis
source location
