基于混沌粒子群优化核极限学习机的变压器故障诊断方法

为提高电力变压器故障诊断的准确度,提出一种基于核极限学习机(KELM)的变压器故障诊断方法,利用混沌优化改善粒子群算法的全局寻优性能。该方法首先用KELM建立故障诊断模型,再利用改进后的混沌粒子群算法(CPSO)对KELM的参数进行优化。结合油中溶解气体分析法(DGA)获得样本数据,通过实例仿真结果对比分析表明,所用算法具有更高的诊断准确率,提高了变压器故障诊断的可靠性。

Transformer Fault Diagnosis Method Based on CPSO Kernel Extreme Learning Machine

In order to improve the accuracy of power transformer fault diagnosis,a transformer fault diagnosis method based on kernel extreme learning machine(KELM)is proposed.Chaos optimization is used to improve the global optimization performance of particle swarm optimization algorithm.In this method,firstly KELM is used to establish a fault diagnosis model,and then the improved chaotic particle swarm algorithm(CPSO)is used to optimize KELM parameters.Combined with the sample data obtained by dissolved gas analysis in oil(DGA),the comparative analysis of simulation results shows that the algo-rithm has higher diagnostic accuracy and improves the reliability of transformer fault diagnosis.