基于AVA优化RBF神经网络的变电站通信网络故障诊断方法  

Fault Diagnosis Method of Substation Communication Network Based on AVA-Optimized RBF Neural Network

作　　者：李超 LI Chao(Shenzhen Shenpengda Power Grid Technology Co.,Ltd.,Shenzhen 518034,China)

机构地区：[1]深圳市深鹏达电网科技有限公司,广东深圳518034

出　　处：《安徽电气工程职业技术学院学报》2025年第1期94-101,共8页Journal of Anhui Electrical Engineering Professional Technique College

摘　　要：为了保障变电站通信网络的可靠运行,文章提出了一种基于非洲秃鹫算法(African vulture algorithm,AVA)优化径向基(radial basis function,RBF)神经网络参数的变电站通信网络故障诊断方法。首先通过AVA的寻优搜索确定了RBF神经网络的最优网络参数,构建了AVA-RBF模型。其次通过变电站通信网络测试系统进行仿真实验。结果表明,AVA-RBF模型诊断结果的平均准确率高达97.5%,相比对比模型具有更高的诊断精度,最后根据仿真结果验证了所提方法能够显著提升变电站通信网络故障诊断的准确性。In order to ensure the reliable operation of substation communication networks,this paper proposes a fault diagnosis method for substation communication networks based on the African vulture algorithm(AVA)optimizing the parameters of radial basis function(RBF)neural networks.Firstly,the optimal network parameters of RBF are determined through the optimization search of AVA,and an AVA-RBF model is constructed.Secondly,simulation experiments are conducted on the substation communication network test system.The results show that the average accuracy of the fault diagnosis by the AVA-RBF model is as high as 97.5%,which is higher than the comparative model in terms of diagnostic accuracy.Finally,the simulation results verify that the proposed method can significantly improve the accuracy of fault diagnosis in substation communication network.

关 键 词：变电站 通信网络 故障诊断 径向基神经网络 非洲秃鹫算法 

分 类 号：TM63[电气工程—电力系统及自动化] TP183[自动化与计算机技术—控制理论与控制工程]