首页|期刊导航|东北电力技术|基于VMD-GJO-DBN的有载分接开关故障诊断方法

基于VMD-GJO-DBN的有载分接开关故障诊断方法OA

Fault Diagnosis Method for on Load Tap Changer Based on VMD-GJO-DBN

中文摘要英文摘要

为了提升变压器有载分接开关(on-load tap changer,OLTC)的故障诊断精度,本文提出了一种 OLTC 故障诊断方法.首先,利用变分模态分解法提取到了 OLTC 运行状态的 6 个特征量;其次,以 6 个特征为输入量,OLTC 故障类型为输出量,采用金豺优化(golden jackal optimization,GJO)算法对深度信念网络(deep belief networks,DBN)的网络参数进行了寻优搜索;最后,构建 GJO-DBN 模型,利用 GJO-DBN 模型对 OLTC 进行了故障诊断.算例分析结果表明,GJO-DBN 模型的诊断精度高达98.88%,平均计算时间仅为12.33 s,该模型不仅能够提升 OLTC 故障诊断精度,同时也能缩短计算时间,验证了所提 OLTC 故障诊断方法的有效性.

In order to improve the fault diagnosis accuracy of on load tap changer(OLTC)for transformers,it proposes an OLTC fault diagnosis method.Firstly,it extracts six feature quantities of OLTC operating stateusing variational mode decomposition(VMD)method.Secondly,with the six features as input and OLTC fault type as output,it uses the golden jackal optimization(GJO)algorithmto optimize the network parameters of deep belief networks(DBN).Finally,it constructs a GJO-DBN model and carries out the fault diagnosis of OLTC using the GJO-DBN model.The case analysis results show that the diagnostic accuracy of the GJO-DBN model is as high as 98.88%,and the average computation time is only 12.33 s.This model can not only improve the fault diagnosis accuracy of OLTC,but also shorten the calculation time,verifying the effectiveness of the proposed OLTC fault diagnosis method.

罗锋

国网湖北省电力有限公司黄冈供电公司,湖北 黄冈 438000

信息技术与安全科学

有载分接开关故障诊断变分模态分解深度信念网络金豺优化算法

OLTCfault diagnosisVMDDBNGJO algorithm

《东北电力技术》 2026 (8)

22-27,6

评论