摘要
研究电路故障诊断问题,提高诊断效率。由于电路集成度提高,电路信号与故障相关,针对传统故障诊断因采用线性诊断方法与提取的电路特征信息不全面,导致诊断定位精度不高,为有效提高电路故障诊断的速度与精度,提出了一种根据小波包能量熵的支持向量机电路故障诊断方法(EE-SVM)。首先利用小波包对电路故障信号进行3层的小波包分解,并提取小波包能量熵,构建输入特征向量。对于支持向量机进行非线性特征向量汰选,去除冗余特征,以保留特征向量构建智能化诊断模型。进行实例仿真,结果显示,方法在所有参比模型中精度最高,能高效地对电路故障进行检测与定位。
The diagnostic efficiency should be improved in the study on fault diagnosis of circuit.For the problem of low efficiency of traditional linear diagnosis method with incomplete circuit features,a novel method of circuit fault diagnosis based on wavelet packet-energy entropy and SVM was proposed to improve the efficiency and accuracy of circuit fault diagnosis.Firstly,the three layers wavelet packet decomposition of the original fault signal is performed by using the theory of wavelet packet and the wavelet packet-energy entropy is extracted.Secondly,input feature vector was constructed with the wavelet packet-energy entropy.Thirdly,some redundant features was eliminated by the nonlinear screening based on SVM and the other features named saved features which were used to build intelligent diagnosis model.Simulation results show that the method has the most accurate in all reference models,which can effectively detect circuit faults.
出处
《计算机仿真》
CSCD
北大核心
2011年第4期199-202,共4页
Computer Simulation
