摘要
针对换相换流器(LCC)-模块化多电平换流器(MMC)混合三端特高压直流输电线路因结构复杂度提高而造成的故障定位困难问题,提出一种基于多分辨奇异值分解(MRSVD)-门控循环单元(GRU)神经网络的特高压直流输电线路单极接地故障定位方法。首先,对故障线路进行选线,判断故障发生线路区段。然后,通过MRSVD对双端故障电压波形进行逐层分解和波形重构。最终,搭建GRU神经网络模型进行故障定位,模型参数通过粒子群优化(PSO)算法进行设定,提升其故障定位准确性。利用PSCAD/EMTDC软件搭建±800kV LCC-MMC混合三端特高压直流输电系统,对不同过渡电阻值、不同故障距离进行仿真。仿真结果证明,所提出的故障定位方法准确度高,能够为混合三端特高压直流输电线路单极接地故障定位提供新的解决方案。
Aiming at the problem of fault location difficulty caused by the increase of structural complexity of line-commutated converter(LCC)-modular multilevel converter(MMC)hybrid threeterminal ultra high voltage direct current(UHVDC)transmission line,a single-pole grounding fault location method for UHVDC transmission line based on multi-resolution singular value decomposition(MRSVD)-gated current unit(GRU)neural network is proposed.First,select the fault line and judge the fault line section.Then,the double-terminal voltage fault waveform is decomposed layer by layer and reconstructed by MRSVD.Finally,the GRU neural network model is built to locate the fault,and the model parameters are set by particle swarm optimization(PSO)algorithm to improve the accuracy of fault location.The±800kV LCC-MMC hybrid three-terminal UHVDC transmission system is built using PSCAD/EMTDC software,and the simulation of different transition resistance values and different fault distances is carried out.The simulation results show that the proposed fault location method has high accuracy and can provide a new solution for single-pole grounding fault location of hybrid threeterminal UHVDC transmission lines.
作者
李志川
兰生
魏柯
LI Zhichuan;LAN Sheng;WEI Ke(School of Electrical Engineering and Automation,Fuzhou University,Fuzhou 350108)
出处
《电气技术》
2023年第3期1-8,63,共9页
Electrical Engineering