配电系统的快速故障检测与识别方法

Method for Fast Fault Detection and Identification in Power Distribution Systems

为了实现对配电系统中不同类型开路故障的高效检测和准确识别,提出了一种改进的多分类支持向量机方法。以电网的电压有效值为输入信号,可在配电系统不同节点电压变化影响下对开路故障进行检测和识别。利用MATLAB软件对IEEE 13节点测试系统进行暂态开路故障的仿真分析。结果表明了方法的准确性、有效性、鲁棒性和实时性。

In order to realize efficient detection and accurate identification of different types of open-circuit faults in the distribution system,an improved multi-classification support vector machine(SVM)method was proposed in this paper.Taking the effective voltage value of the power grid as input signal,open-circuit faults could be detected and identified under the influence of different node voltage changes in the distribution system.Simulation analysis was completed on transient open-circuit faults in IEEE 13 node test system by means of MATLAB software.The results indicat the accuracy,effectiveness,robustness and real-time performance of the proposed approach.