摘要
为提高变压器故障诊断精度,提出了一种改进粒子群优化RBF网络算法,用于优化RBF网络的中心参数。首先通过非线性递减权值策略改进粒子群算法,再利用改进粒子群优化RBF网络,最后建立用于变压器故障诊断的RBF网络模型,并在Matlab平台上进行了仿真测试。结果表明,优化后的RBF网络比单一RBF网络故障诊断率有大幅提高。通过诊断国网某公司的5组故障实例,验证了所提算法的可行性。
In order to improve the accuracy of transformer fault diagnosis,an improved particle swarm optimization RBF network algorithm is proposed to optimize the central parameters of RBF network.First,the particle swarm optimization algorithm is improved by nonlinear decrement weighting strategy.Then the improved particle swarm optimization is used to optimize the RBF network.Finally,the RBF network model for transformer fault diagnosis is established and simulated on the Matlab platform.The results show that the optimized RBF network has a significantly higher fault diagnosis rate than the single RBF network.The feasibility of the proposed algorithm is verified by diagnosing five sets of fault instances of a company in the State Grid.
作者
马松龄
郭小艳
张清敏
代一楠
MA Song-ling;GUO Xiao-yan;ZHANG Qing-min;DAI Yi-nan(School of Mechanical and Electrical Engineering,Xi’an University of Architecture and Technology,Xi’an 710055,China;Zhangye Power Supply Company,State Grid Gansu Electric Power Corporation,Zhangye 734000,China)
出处
《水电能源科学》
北大核心
2019年第4期184-186,191,共4页
Water Resources and Power
基金
陕西省教育厅自然科学研究项目(16JK1427)
关键词
变压器
故障诊断
改进粒子群算法
RBF网络
transformer
fault diagnosis
improved particle swarm optimization
RBF neural network