应用多级低通滤波法增强深部采空区有效电性特征——以可控源音频大地电磁法在潍日高速公路测区应用为例

Enhancing the Effective Electrical Characteristics of Deep Doaf by Multistage Low-pass Filtering: a Case Study from Weiri Expressway Survey with the Controlled Source Audiomagnetotelluric

为了提高潍日高速公路下伏深部煤矿采空区探测精度,采用可控源音频大地电磁法(controlled source audio magnetotellurics,CSAMT)对测区进行探测,并应用“多级”低通滤波法对因多干扰因素影响所产生的不合理电性异常进行滤除,以提高测区采空区探测精度。针对不同异常特点,分步实施“多级”低通滤波。首先对在强干扰背景下获得的原始数据进行异常点剔除;然后,采用中值空间滤波法消除静态效应;最后,利用均值阈值分割法和幂函数法分别去除浅部和深部低阻异常。相比于原始电性特征图,经过“多级”低通滤波处理后的深部采空区有效电性特征得到了明显增强,所刻画的采空区边界更加清晰。依此布设钻孔并实施了钻探,结果表明,物探异常与实际采空区埋深等具有很高的吻合度。方案的有效实施为潍日高速公路施工建设的安全评价提供了可靠的物探依据。

In order to improve the detection accuracy of goaf in deep coal mine under Weiri Expressway,the"multi-stage"low-pass filtering,based on controlled source audiomagnetotellurics (CSAMT),was used to filter out the irrational electrical anomalies caused by multi-interference factors. According to different abnormal characteristics,multi-stage low-pass filtering was implemented step by step. Firstly,outliers from the original data obtained under strong interference background were removed. Secondly,the static effects were eliminated by using the median filter space method. Finally,the shallow and deep low resistivity anomalies respectively by means of mean threshold segmentation and power function were filtered. Compared with the original electrical characteristic map,the effective electrical characteristics of the deep goaf after "multi-stage"low-pass filtering were significantly enhanced,and the boundary of goaf was more clearly. The results show that the geophysical anomalies are in good agreement with the actual goaf depth. Therefore,the effective implementation of this scheme provides a reliable geophysical basis for the safety evaluation of Weiri Expressway construction.