基于状态估计的含分布式电源树状配电网故障测距算法

多电源配电网的故障定位和测距对快速排除故障、提高供电可靠性具有重要意义。以含分布式电源(DG)树状配电网的区间定位为基础,按区间特性将配电网分为两电源区段和单电源区段,建立故障等值回路与其相应的微分方程和最优估计离散模型。以电源端口电压电流、各FTU电流为输出量,以故障距离为状态量,利用强跟踪的卡尔曼滤波器算法求解最优估计离散模型,跟踪相间短路故障测距。通过搭建含DG的树状配电网,分析负载、故障点过渡电阻对故障测距结果的影响。算例分析表明,该算法跟踪速度快,受初值的影响小,不受DG接入位置和数量的影响,不受故障点过渡电阻的影响,受负载和树状分支结构影响较少,测距精度满足工程精度要求。

Fault location algorithm based on state estimation in tree distribution network with distributed generation

Fault location of distribution network with multi-generation is of great significance to exclude the fault and improve the power supply reliability. The research in this paper is based on interval location of tree distribution network with distributed generation. The distribution network is divided into two power sections and single power sections according to the interval characteristics, the fault equivalent circuit with its corresponding differential equation and optimal estimation of discrete model are established. The method takes voltage and current of the power supply port and FTU current as output quantity, and the fault distance as state quantity. A strongly tracking Calman filter algorithm is used to solve the optimal estimation of discrete model and track the fault location of interphase short circuit. By building a tree distribution network with DG, the effect of load and transition resistance on fault location is analyzed. The simulation results show that the tracking speed of algorithm is fast and less affected by the initial value. It is not affected by the location or quantity of DG and transition resistance. It is also less affected by the load and tree topology. The location accuracy can meet the engineering accuracy requirements. This work is supported by Natural Science Foundation of Shandong Province (No. ZR2012EEL20), National Natural Science Foundation of China (No. 51477184) and National Key Research and Development Program of China (No. 2018YFB0904800).