基于谱峭度的锚杆参数检测方法  

Method of rock bolt parameters detection based on spectral kurtosis

作　　者：孙晓云[1] 王志远[1] 程久龙[2] 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 [2]中国矿业大学(北京)煤炭资源与安全开采国家重点实验室,北京100083

出　　处：《中国矿业》2015年第12期147-151,共5页China Mining Magazine

基　　金：国家自然科学基金"锚杆受荷条件下声学特征及承载力预测方法研究"项目资助(编号:51274144));河北省自然科学基金"锚杆受荷条件下极限承载力智能预测方法研究"资助(编号:F2012210031);河北省自然科学基金"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：锚杆长度是评价锚杆质量的重要因素,但由于恶劣的锚杆检测环境,以及仪器自身所带来的干扰,采集到的锚杆信号含有大量噪声,使信号分析和锚杆参数精确预测困难,为此需要对信号进行滤波处理。针对锚杆信号的特点,本文利用谱峭度理论构造了一个以最大化信噪比滤波的峭度滤波器,对锚杆检测信号进行滤波,完成锚杆参数检测。与传统的小波理论方法相比较,该方法的检测精度有了较大提高。另外,峭度滤波后的信号具有良好的频域特性,在频域图中可方便读取相应频率,利用频率差法也可求得比较精确的锚杆长度。The length of the rock blot are one of the important parameters to evaluate the quality of the rock bolt, during rock bolt detection, because of the harsh environment and the disturbance from instrumentation, there are a lot of noises, which make the signal analysis and rock bolt parameters detection difficult, so, it＇s necessary to remove noises. According to the characteristics of the anchor signal distribution,this paper proposes a detection method based on spectral kurtosis in parameter detection. A filter on the basis of the maximum SNR filtering using spectral kurtosis theory is constructed, parameters testing are finished after filtered. Compared with the traditional method based on wavelet theory, the detection precision of this method has been greatly improved. The denoised signal also has a good frequency domain characteristic,some frequency values are easy to get, so the length of anchor could be got using a frequency differential method.

关 键 词：锚杆 谱峭度 滤波器 小波理论 

分 类 号：TD353[矿业工程—矿井建设]