基于BSA与BP神经网络的矿用变压器故障诊断

Fault Diagnosis of Mining Transformer Based on BSA and BP Neural Network

针对矿用变压器故障问题,提出了一种基于鸟群与BP神经网络算法的矿用变压器故障诊断方法。利用鸟群算法对BP神经网络进行优化,利用传统的粒子群算法和此算法分别对BP神经网络进行优化,并对矿用变压器诊断结果进行对比,有效地解决了BP神经网络训练时间长且易陷入局部最优的缺陷。

Aiming at the problem of mining transformer faults,a fault diagnosis method for mining transformers based on bird swarm and BP neural network algorithm is proposed.The bird swarm algorithm is used to optimize the BP neural network.The traditional particle swarm algorithm and this algorithm are used to optimize the BP neural network,and the diagnostic results of mining transformers are compared.Effectively solving the defects of long training time and easy falling into local optima in the BP neural network.