论文部分内容阅读
利用瞬变电磁法在煤矿采空区实际探测中,接收到衰减的二次感应电动势,其中中晚期的信号受噪声的干扰较大,信噪比低,不能有效真实地反映采空区的情况.本文通过对快速独立分量分析(FastICA)和小波阈值算法的改进,提出了FIW算法步骤,进而对TEM中晚期信号降噪计算,整个FI-W运算过程利用MATLAB程序.FI-W计算后的数据与其它降噪算法处理后的数据对比,分析得出FI-W算法提高了信噪比,处理后的曲线更接近于理想的衰减曲线;将测线上每个测点中晚期信号选取一部分数据用FI-W算法处理后,用GRAPHER软件形成多测道曲线,根据异常点确定采空区的位置,并钻孔验证,证实:通过多测道曲线确定的异常点位置与实际钻孔确定的采空区位置一致.经过FI-W算法降噪计算后的TEM中晚期数据能真实地反映采空区的位置.
Transient electromagnetic method is used to detect the secondary induced electromotive force of the mined-out area in coal mines. Among them, the signal in the middle and late period is greatly disturbed by the noise and the signal-to-noise ratio is low, which can not effectively and truly reflect the situation of the goaf In this paper, we propose the steps of FIW algorithm to improve the fast independent component analysis (FastICA) and wavelet threshold algorithm, and then use the MATLAB program to reduce the noise of the TEM signal in the late stage of the TEM .FI-W Data and other noise reduction algorithm processed data contrast analysis FI-W algorithm to improve the signal-to-noise ratio, the processed curve is closer to the ideal attenuation curve; the measurement line at each measurement point in the late signal selection part After the data were processed by FI-W algorithm, multi-trace curves were formed by GRAPHER software, and the location of goaf was determined according to the anomalous points. The borehole verification proved that the location of abnormal points and the actual drilling determined by multi-track curve The location of the goafs is consistent.The late and middle TEM data after the FI-W algorithm noise reduction can truly reflect the location of the goafs.