基于优化小波神经网络的输电线路行波故障测距

针对单端行波故障测距方法中故障点反射波与对端母线反射波的识别问题,提出了一种改进粒子群算法优化的小波神经网络的故障测距模型。提取保护安装处检测到的行波波头时间值与反向行波线模分量的李氏指数作为行波特征值,利用小波神经网络拟合行波特征值与输电线路故障距离之间的关系,构建小波神经网络故障测距模型,利用该模型可以直接得到输电线路的故障距离。在标准粒子群算法中引入遗传算法变异因子,利用改进后的粒子群算法作为小波神经网络的训练算法,优化小波神经网络的权值与阈值参数,加快了小波神经网络故障测距模型的收敛速度,并提高了输出结果的精度。仿真结果证明,该方法有效且可行。

Traveling Wave Fault Location of Transmission Line Based on Optimized Wavelet Neural Network

In order to identify the reflection wave at the fault point and the reflection wave of terminal bus when using the single-terminal traveling wave fault location method,a fault location model of wavelet neural network optimized by the enhanced particle swarm optimization(PSO)algorithm is proposed.The time value of traveling wave head detected at the protection installation and the Lipschitz index of the line mode component of reverse traveling wave are extracted as the traveling wave eigenvalue.By using the wavelet neural network to fit the relationship between the traveling wave eigenvalue and transmission line fault distance,the fault location model of wavelet neural network is constructed,which is used to obtain the fault distance of transmission line directly.By introducing the mutation factor of genetic algorithm into the standard PSO algorithm,the enhanced PSO algorithm is used as the training algorithm for wavelet neural network to optimize the weight and threshold parameters of wavelet neural network.As a result,the convergence speed of the fault location model based on wavelet neural network is accelerated,and the accuracy of output results is improved.Simulation results show that the proposed method is effective and feasible.