基于有限PMU配置与空域信号生成的配电网故障定位方法

Distribution Network Fault Location Method Based on Limited PMU Configuration and Airspace Signal Generation

随着配电网网络结构日益复杂,目前因成本限制有限安装的同步相量测量装置所采集到的数据难以保证配电网故障的全局可观性。为此,基于图卷积神经网络算法建立了一种可适用于含有限电源管配电网络的全局空域故障信号生成模型,提出了一种有限电源管安装下准确度较高的配电网故障线路定位方法。通过搭建20个节点的三相平衡配电网验证了所提模型与其他基准模型相比定位准确度更高。在电源管非全面配置时可达到98.8%的定位准确度,且模型对过渡电阻具有鲁棒性。

With the increasing complexity of distribution network structure,it is difficult to ensure the global observability of distribution network fault by the data collected by the synchronized phasor measurement device installed due to the limited cost.Therefore,based on the graph convolution neural network algorithm,a global spatial fault signal generation model suitable for distribution networks with limited power management unit(PMU)was established,and a high accuracy fault line location method for distribution networks with limited PMU was proposed.By building a three-phase balanced distribution network with 20 nodes,it is verified that the positioning accuracy of the proposed model is higher than that of other benchmark models.When the PMU is not fully configured,the positioning accuracy can reach 98.8%,and the model is robust to the transition resistance.