Fault Diagnosis Based on Electromagnetic Particle Swarm Optimization Clustering for Rotating Machine
Fault diagnosis is an important procedure to ensure the equipment efficiency and stability.The diagnosis process is actually a pattern recognition process,and usually,the fault samples are lack of tags of fault types.In this case,the nonsupervised learning method is more available,and kernel clustering is one of the most effective methods.In this paper,a novel electromagnetic particle swarm optimization clustering (EPSOC) was proposed.The weighted kernel clustering model was established by weighting the samples according to the effect of different characters.The clustering center,character weight and the model parameter were united as the optimal variable,and the Xie-Beni index was utilized as the objective function.To find the clustering center and model parameters effectively,the electromagnetism-like operator was introduce into the particle swarm optimization algorithm,and the modified mixing algorithm was utilized.The UCI data and fault samples were used to test the performance of the proposed method EPSOC,and comparing with the other common algorithms,the superiority of EPSOC was demonstrated.
fault diagnosis,mercer kernel,weighted fuzzy kernal clustering electromagnetic particle swarm optimization
Daquan Cai Han Xiao Chen Chen
Henan Electric Power Survey & Design Institute Zhengzhou, China
国际会议
哈尔滨
英文
224-227
2016-07-21（万方平台首次上网日期，不代表论文的发表时间）