Fault Diagnosis of Power Circuits Based on SVM Ensemble with Quantum Particles Swarm Optimization
Based on least squares wavelet support vector machines (LS-WSVM) ensemble with quantum particle swarm optimization algorithm (QPSO), a systematic method for fault diagnosis of power circuits is presented. Firstly, wavelet coefficients of output voltage signals of power circuits under faulty conditions are obtained with wavelet lifting decomposition, and then faulty feature vectors are extracted from the disposed wavelet coefficients. Secondly, a boosting strategy is adopted to select faulty feature vectors automatically for LS-WSVM-based multi-class classifiers, QPSO is applied to select the optimal values of the regularization and kernel parameters of multi-class LS-WSVM. So the multi-class LS-WSVM ensemble model with boosting for the power circuits fault diagnosis system is built. The simulation result of push-pull circuits shows that the fault diagnosis method of the power circuits using LS-WSVM ensemble with QPSO is effective.
Zhiyong Luo Binyuan Ye Linqing Cai Wenfeng Zhang
School of Automation,Chongqing University of Posts and Telecommunications,Chongqing,400065,China Department of Mechanical & Electrical Engineering,Guangdong Vocational College of Mechanical & Elect School of Automation,Northwestern Polytechnical University,Xian,710072,China
国际会议
深圳
英文
587-592
2008-12-10（万方平台首次上网日期，不代表论文的发表时间）