基于IACA-BP神经网络的GIS局部放电故障诊断方法

GIS partial discharge fault diagnosis method based on IACA-BP neural network

针对封闭式气体绝缘开关装置(gas insulated switchgear,GIS)局部放电造成的停电问题,提出了一种利用改进蚁群算法(improved ant colony algorithm,IACA)优化BP神经网络的GIS局部放电故障诊断方法。首先,提取GIS局部放电特征参量,并采用主成分分析法进行降维处理,旨在降低诊断方法的计算量;然后,针对BP神经网络性能容易受权重与偏置的影响,采用改进蚁群算法进行参数优化,并将降维数据作为样本输入优化后的BP神经网络,进行故障分类诊断;最后,通过实验验证发现,相比原算法,本文所提方法故障识别准确率明显得到改善,且其平均准确率达到了96.5%。

Aiming at the problem of power outage caused by partial discharge of gas insulated switchgear(GIS),a fault diagnosis method of GIS partial discharge based on BP neural network optimized by improved ant colony algorithm(IACA)is proposed.Firstly,the characteristic parameters of GIS partial discharge are extracted,and the principal component analysis method is used to reduce the dimension,so as to reduce the calculation amount of the diagnosis method.Secondly,since the performance of BP neural network is easily affected by weight and bias,the improved ant colony algorithm is used to optimize the parameters,and the dimensionality reduction data is used as the sample input to the optimized BP neural network for fault classification and diagnosis.Finally,through experimental verification,it is found that compared with the original algorithm,the fault recognition accuracy of the proposed method is significantly improved,and its average accuracy rate reaches 96.5%.