摘要
目前基于行波法的配电网故障定位精度较低,该文提出了一种基于北斗和双阶段故障定位方法。首先,利用初始行波到达时间分别为所有路段构建最小化优化模型,并基于粒子群优化算法(PSO)进行初步的故障定位。为了减少时间误差对定位结果的影响,利用残差聚类算法对故障段进行二次校正,计算故障距离的修正解,以提高故障定位精度。随后,将确定的故障信息通过北斗短报文通信发送给电力巡检人员,从而快速实现定位结果的巡查维护。实验阶段,通过多组实验验证,所提故障定位方法的绝对误差均小于80 m。与多端行波故障定位方法相比,所提方法的平均定位精度提高了71.63%。实验结果验证了所提方法的有效性,表明所提方法对时间误差的容忍度较强,具有广阔的应用前景。
A method based on Beidou and two-stage fault localization is proposed to address the low accuracy of current fault localization in distribution networks using traveling wave methods.Build a minimum optimization model for all road sections using the initial arrival time of the traveling wave,and locate the initial fault once based on Particle Swarm Optimization(PSO)algorithm.In order to reduce the impact of time error on the positioning results,residual clustering algorithm is used to perform secondary correction on the fault segment,calculate the corrected solution of the fault distance,and improve the accuracy of fault positioning.Send the determined fault information to the power inspection personnel through Beidou short message communication,thereby quickly achieving inspection and maintenance of the positioning results.During the experimental stage,through multiple sets of experiments,the absolute error of the proposed fault location method was less than 80m.Compared with the multi terminal traveling wave fault location method,the average positioning accuracy of the proposed method has improved by 71.63%.The experimental results validate the effectiveness of the proposed two-stage positioning method,indicating that the proposed method has a strong tolerance for time errors and has broad application prospects.
作者
穆尔夏迪·艾买提
白小鹏
蒋德鹏
李小鹏
买振兴
MURSHADI Amat;BAI Xiaopeng;JIANG Depeng;LI Xiaopeng;MAI Zhenxing(State Grid Hami Power Supply Company Information and Communication Company(Data Center),Hami Xinjiang 839000,China)
出处
《人工智能科学与工程》
CAS
北大核心
2024年第3期75-83,共9页
ARTIFICIAL INTELLIGENCE SCIENCE AND ENGINEERING
关键词
配电网
故障定位
行波法
粒子群优化算法
聚类
北斗
distribution network
fault location
traveling wave method
particle swarm optimization
clustering
Beidou