RBF网络用于变压器油击穿电压的多参数关联预测

The prediction method of transformer oil breakdown voltage via multi-parameter correlation based on RBF neural network

基于变压器油性能参数之间的关联性,在Matlab平台下研究建立变压器油关键参数——击穿电压的多参数关联预测方法。利用RBF网络建立击穿电压与4个强关联性指标的关系模型,并与BP网络模型进行对比;采用模糊C聚类方法处理大量的训练样本,以聚类结果来训练网络,从而解决大样本训练网络时网络结构复杂、性能不佳等问题;仿真结果表明,RBF网络较BP网络的建模性能大为改善,预测值与实际值的相对误差均在10%以内,能够满足实际应用要求,具有重要的理论意义及应用价值。

Based on the fact that performance parameters of transformer oil correlation,this paper studied to establish a prediction method of breakdown voltage which is the most important parameter of transformer oil via multi-parameter correlation under the development environment of Matlab.The relational model of breakdown voltage and some parameters which have strong correlation with breakdown voltage was constructed using RBF neural network.The clustering centers used to train network were acquired through clustering the original monitoring data samples with fuzzy C-means clustering algorithm.This method can resolve natural problems of neural networks caused by large sample capacity,such as complication of net construction,inferior astringency,poor generalization ability,and so on.Test results show that the prediction performances of the RBF network are better and that the relative errors between predicted values and real values are all less than 10%,which can meet the requirements of practical application and indicates the significant practical value of these models.