基于模糊粗糙集和GA-BP神经网络的L配电网故障选线方法

针对BP神经网络故障选线当输入数据量大时,其结构复杂、收敛慢,并且易陷入局部最优的缺点,将模糊粗糙集和遗传算法优化神经网络的方法引入配电网单相接地故障选线中.通过MatLab仿真试验,得出大量的各线路零序电流信号,并将多种提取的特征量进行信息融合.利用粗糙集理论对条件属性进行约简,去掉冗余条件属性,将约简后的属性作为输入层的BP神经网络,然后通过遗传算法优化BP神经网络进行训练和测试.测试结果表明,该方法具有训练速度快、误判率低的优点,能够满足电力系统对选线精度和准确性的要求.

Faulty Line Detection Method for Distribution Network Based on Fuzzy Rough Sets and GA-BP Neural Network

This paper proposes particle swarm optimization method of BP neural network into a single distribution network in the phase to ground fault line. The MatLab simmulation tests show that it obtained the line of zero sequence current signals and a variety of extract characteristic information fusion. We use particle swarm optimization method of BP neural network for fault line. In order to verify the efficiency of the algorithm and the superiority, we respectively use two kinds method of BP neural network and GA-BP neural network simulated. The results show that this method has the more training speed and lower false positives than the traditional method, and the system can meet the power requirements for precision and accuracy.