基于信息熵和Haar小波变换的开关电流电路故障诊断新方法  

Switched-current circuit fault diagnosis based on entropy and Haar wavelet transform

作　　者：龙英[1,2] 何怡刚[2] 张镇[1] 袁莉芬[2] 

机构地区：[1]长沙学院电子信息与电气工程系,长沙410003 [2]合肥工业大学电气与自动化学院,合肥230009

出　　处：《仪器仪表学报》2015年第3期701-711,共11页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(61201108;61102035);中国博士后科学基金(2014M551797);湖南省自然科学基金(13JJ6083)资助项目

摘　　要：针对开关电流(SI)电路的故障诊断和定位问题,为进一步提高故障准确率,提出了基于信息熵和Haar小波变换的开关电流电路故障诊断新方法。该方法采用伪随机信号激励经蒙特卡罗分析、Haar小波正交滤波器分解和信息熵及模糊集的计算来实现故障特征的提取,以减少信号的冗余。最后构建故障字典,完成各故障模式的故障分类。对六阶切比雪夫低通滤波器进行了仿真实验验证,获得了100%的故障诊断准确率,与其它方法进行比较,实验结果显示了本文方法的优越性。Focusing on the issue of switched-current（ SI） circuit fault diagnosis and location,a new method of SI circuit fault diagnosis based on entropy and Haar wavelet transform（ HWT） is presented to enhance the accuracy of fault diagnosis.The proposed method uses Monte-Carla analysis,Haar wavelet transform and the calculating of entropy and ambiguity groups to deal with pseudo-random sequences as a stimulate signal of the circuit under test（ CUT） to extract fault features,where the overlap of the features can be minimized.Finally,the fault dictionaries were constructed,and the fault classification of each failure mode was completed.A 6th-rder Chebyshev low-pass filter circuit is used to verify the proposed method,and the fault diagnosis accuracy is 100%.Compared with other methods,the presented method is superior.

关 键 词：开关电流电路 HAAR小波变换 信息熵 故障诊断 

分 类 号：TH165.3[机械工程—机械制造及自动化] TP306.3[自动化与计算机技术—计算机系统结构]