To minimize the positioning error and improve the accuracy of fault location for the broken-line faults in 10kV distribution networks,a novel fault location method based on machine learning algorithms is proposed.Firstly,real-time operational data of the distribution network is collected using data acquisition equipment.Secondly,features with high correlation to broken-line faults,sensitivity to them,and high discrimination are selected from the collected data.On this basis,the machine learning algorithms are employed to initially determine the faulty section and then precisely locate the fault within that section.Experimental results demonstrate that the proposed method achieves a maximum fault location error of no more than 0.09km,exhibiting excellent positioning accuracy and stability.
关键词
机器学习算法/10kV配网/断线故障定位/故障区段
Key words
machine learning algorithms/10kV distribution networks/broken-line fault location/faulty section
维普
万方数据