摘要
针对现行模拟电路故障诊断方法大多是针对电路中的硬故障而难以发现软故障的问题,利用小波信号的局部特征表征能力,对模拟电路故障诊断中的软故障信号进行了预处理和特征提取,获取信号中突变点的位置判定其奇异性;利用模拟电路软故障信号的局部突变特性,将正常信号和软故障信号的小波系数进行了比较分析。以带通滤波器作为电路仿真对象,对基于小波分析的软故障信号诊断技术进行了试验验证。验证结果表明,与频率相关的元器件对电路的影响不同,电容相较于电阻对电路的影响更大,与频率无关的元器件在不发生短路的情况下对电路的影响很小。
Most of the current fault diagnosis methods of analog circuits are for hard faults,and it is hard to find the soft faults.Using the capability of local characteristics representation of wavelet signal,the soft fault signals of analog circuit fault diagnosis are preprocessed,and their features extracted,to obtain the signal mutation point position to determine its singularity.Using the local mutation of analog circuit soft fault signal,the wavelet coefficients of the normal signal and the soft fault signal are compared.With the band-pass filter used as the object of circuit simulation,the soft fault signal diagnosis technology based on wavelet analysis is tested and verified.It is found that the components related to frequency have different effects on the circuit.The capacitance has a greater effect on the circuit than the resistance,and the components independent of frequency have little effect on the circuit when short circuit doesn’t occur.
作者
董江涛
朱琳
巨小微
DONG Jiangtao;ZHU Lin;JU Xiaowei(The 54th Research Institute of CETC,Shijiazhuang 050081,China;Taiyuan Satellite Launch Center,Yinchuan 750004,China)
出处
《无线电工程》
北大核心
2021年第7期622-627,共6页
Radio Engineering
基金
国家自然科学基金资助项目(20310702D)。
关键词
软故障
小波分析
奇异性
信号诊断
soft fault
wavelet analysis
singularity
signal diagnosis
