基于相量测量单元数据的电网故障诊断

Grid fault diagnosis based on phasor measurement unit data

传统电网故障诊断方法不能实现对电网各个节点工作状态的实时监控,导致电网故障诊断时间过长,准确率低。为了解决上述问题,提出基于相量测量单元数据的电网故障诊断方法。以相量测量单元数据为基础设计电网节点监测器,实现对电网系统的各个节点工作状态的实时监控。利用GPS进行电网故障定位,以已有故障样本数据为基础,利用果蝇算法优化支持向量机方法训练样本数据,提取电网故障特征,根据特征提取结果对电网故障进行分类,将分类结果与故障类型进行匹配,输出故障诊断结果。实验结果表明,研究方法的检测时间短,检测效果好,具有很高的实际应用价值。

The traditional power grid fault diagnosis method can not realize the real-time monitoring of the working state of each node of the power grid,resulting in the long time of power grid fault diagnosis and low accuracy.In order to solve the above problems,a fault diagnosis method based on phasor measurement unit data is proposed.Based on the data of phasor measurement unit,a network node monitor is designed to monitor the working state of each node in real time.Based on the existing fault sample data,the sample data are trained by using the fruit fly algorithm to optimize the support vector machine(SVM) method,and the fault characteristics of the power grid are extracted.The power grid faults are classified according to the feature extraction results.The classification results are matched with the fault types,and the fault diagnosis results are output.The experimental results show that the method has a short detection time,good detection effect and high practical value.