基于CPSO优化的RBF神经网络的电网故障诊断

Grid Fault Diagnosis for the RBF Neural Network Based on CPSO

针对电网故障诊断技术的优化,提出了基于混沌粒子群(CPSO)优化的径向基函数(RBF)神经网络的电网故障诊断方法。首先对径向基函数神经网络结构进行分析,然后利用混沌粒子群算法参与神经网络学习,调节径向基函数的权值和神经元宽度,在此基础上构建电网故障诊断策略,最后对方法进行算例验证。算例结果表明,方法不仅能够降低神经网络学习误差,也能提高电网故障诊断的精确度。

With respect to the optimization of grid fault diagnosis technology,a grid fault diagnosis method by means of the radial basis function(RBF)neural network based on the chaos particle swarm optimization(CPSO)was proposed.Firstly,the structure of the RBF neural network was analyzed.Then,the CPSO algorithm participated in the learning of the neural network to adjust the weight and neuron width of the RBF.Furthermore,a grid fault diagnosis strategy was designed.Finally,example verification was carried out for the presented scheme,and calculation results indicated that the proposed approach could not only reduce learning errors of the neural network,but also improve the accuracy of power grid fault diagnosis.