文档介绍:基于SAAFSA优化加权模糊聚类算法的
变压器故障诊断
史丽萍,宋朝鹏,李明泽,陈苏黔,李加欣
(中国矿业大学 电气与动力工程学院,江苏 徐州 221008)
摘要:针对加权模糊聚类算法(WFCM)应用于变压器DGA分析时存在收敛速度慢、对初始值敏感的问题,提出了一种改进人工鱼群优化加权模糊聚类算法(SAAFSA-WFCM)的变压器故障诊断方法。该方法利用模拟退火算法(SA)来改进人工鱼群算法(AFSA)以求取最佳初始聚类中心,在发挥AFSA优异的全局寻优能力的同时,利用SA的概率性突跳搜索机制对AFSA实施局部优化,提高了AFSA的搜索精度。WFCM算法以得到的最佳初始聚类中心为初值进行迭代运算,最终求得更接近实际位置的聚类中心,克服了WFCM易受初值影响的缺陷,加快了收敛速度。仿真与实例分析表明,该方法可有效应用于变压器的故障诊断,并有着较高的诊断正确率和诊断效率。
关键词:加权模糊聚类;模拟退火;人工鱼群算法;聚类中心;故障诊断
中图分类号:TM41 文献标识码:A 文章编号:1001-1390(2018)00-0000-00
Transformer fault diagnosis based on weighted fuzzy clustering algorithm improved by SAAFSA
Shi Liping, Song Chaopeng, Li Mingze, Chen Suqian, Li Jiaxin
(School of Electrical and Power Engineering, China University of Mining and Technology,
Xuzhou 221008, Jiangsu, China)
Abstract:Aiming at the problem of the weighted fuzzy C-means (WFCM) clustering algorithm that the convergence speed is slow and sensitive to the initial value in transformer DGA analysis, a transformer fault diagnostic method based on the WFCM clustering algorithm optimized by improved artificial fish swarm algorithm (SAAFSA-WFCM) is proposed in this paper. This method uses adopts the artificial fish swarm algorithm (AFSA) improved by the simulated annealing (SA) to obtain the best initial clustering center, while taking advantage of the global optimization of AFSA, and the search accuracy of AFSA