改进阈值去噪法在锚杆参数检测中的应用  

Application of improved threshold denoising method in anchor bolt parameter detection

作　　者：曹威[1] 孙晓云[1] 程恩[1] 卞建鹏[1] 王明明[1] 

机构地区：[1]石家庄铁道大学电气与电子工程学院,河北石家庄050043

出　　处：《中国矿业》2016年第8期130-134,共5页China Mining Magazine

基　　金：国家自然科学基金项目"锚杆受荷条件下声学特征及承载力预测方法研究"资助(编号:51274144);国家自然科学基金项目"电力变压器检修策略制定的关键技术研究"资助(编号:51307112);河北省自然科学基金项目"基于导波的锚杆无损检测方法研究"资助(编号:13205002D);河北省自然科学基金项目"基于导纳频率响应函数的锚杆损伤检测与定位方法研究"资助(编号:E2014210075)

摘　　要：锚杆广泛应用在各种大型的岩土工程中,对锚杆锚固质量的检测显得尤为重要,其中锚杆参数检测是锚杆施工质量检测的一项重要内容。由于环境、设备等外在条件的影响,不可避免的出现信号被干扰的现象,使得检测结果与实际相差很大。针对传统的软、硬阈值法以及一些文献所提出的阈值法的不足,分别对阈值函数和阈值进行改进,提出了一种新的阈值去噪法。对阈值函数的改进是将小于阈值的小波系数分成两部分:小于1/2阈值的小波系数全部置零;大于1/2阈值的小波系数减小但不置零,能更好地保留高频中的有用信号。由于阈值选取的不合适直接影响去噪效果,提出了一种自适应阈值方法,随着分解尺度的变化自适应地选取不同的阈值。实验结果表明,改进的阈值去噪法比传统的软、硬阈值法及一些文献所提出的阈值法在锚杆参数检测方面更加准确。Anchor bolt has been used widely in large-scale geotechnical engineering, so, its quality detection is important, especially for parameter detection. Due to the influence of the environment, equipment and other conditions, detecting signal is corrupted greatly with additive noise, which causes great errors between the calculation results and the actual results. According to the shortcomings of the traditional soft , hard threshold methods and other threshold methods designed in some literature, a new wavelet threshold denoising method which improves the threshold functions and threshold is presented. The improved threshold function divides wavelet coefficients that less than the threshold into two parts：ones less than half of threshold is set to zero;the others reduce respectively, which could remain useful signal in high frequency. Since the selected threshold directly affects the denoising effect, an adaptive threshold method is proposed which selects adaptively different thresholds with the change of decomposition scale. The experimeny results show that, the improved threshold denoising method is more accurate than tranditional ones.

关 键 词：锚杆 小波阈值去噪法 自适应阈值方法 测试 

分 类 号：TD353[矿业工程—矿井建设]