摘要
本文在对电力系统故障诊断方法进行分析的基础上,利用径向基函数(RadialBasisFunction RBF)网络适合于求解模式识别问题的优势,提出了应用RBF神经网络(简称RBFNN)来实现高压输电线路的故障诊断,建造了基于RBFNN的高压输电线路故障诊断模型结构,并且给出了确定RBF网络最佳聚类数的标准。仿真分析及容错性测试结果表明,本文所提方法能有效地实现高压输电线路系统的故障诊断,而且在网络的训练速度以及对因干扰而畸变的输入信息情况的容错能力方面都优于传统的BP神经网络,对实时信息处理系统具有一定的适用性。
On the basis of analyzing the methods of power system fault diagnosis, this paper presents RBFNN based high voltage transmission line fault diagnosis method by using radial basis function network's capability which is adaptable to solving pattern recognition problems. A RBFNN model structure used for transmission line fault diagnosis is built, and the standard of evaluating RBF optimal clustering number is proposed. By simulation analysis and test for fault tolerance of network, the results show that the method presented in this paper can accomplish correct fault diagnosis for high voltage transmission line but also has high fault tolerance performance for distorted input signal, and therefore has practical application value in real time information processing system.
出处
《电工电能新技术》
CSCD
2004年第4期13-17,共5页
Advanced Technology of Electrical Engineering and Energy
基金
天津大学留学回国人员基金资助项目
