基于MFO算法的含分布式电源配电网故障定位

Fault Location of Distribution Network with DG Based on MFO Algorithm

摘要: 针对分布式电源的接入，造成了故障电流的多向性，传统故障定位方法不再适用的问题。本文提出基于飞蛾扑火优化算法(MFO)的故障定位方法，采用馈线终端单元上传的电流状态信息作为判断依据。首先，将MFO算法中寻优主体的位置坐标进行二进制编码，使其与馈线终端单元上传的电流状态信息关联，将故障定位任务转化为二进制优化问题，其次将上报的过流告警与预期的过流状态进行判别，最后针对分布式电源的特点，构建新的开关函数，使其适用于分布式电源的动态投切，实现含分布式电源配电网的故障定位。通过对33节点配电网进行仿真实验，仿真结果表明该方法能够准确定位单一以及多重故障区段，在信息畸变情况下，仍能准确定位，具有良好的容错性，并且MFO算法与其他算法相比较收敛更快，全局寻优性更强。

Abstract: For the access of distributed power sources, the multi-directional fault overcurrent is caused, and the traditional fault location method is no longer applicable. In this paper, a fault location method based on the Moth-flame optimization algorithm (MFO) is proposed, and the current state infor-mation uploaded by the feeder terminal unit is used as the judgment basis. First, binary code the position coordinates of the optimization subject in the MFO to associate it with the current status information uploaded by the feeder terminal unit, and convert the fault location task into a binary optimization problem. The difference between the flow states is used as the objective function. Finally, according to the characteristics of the distributed power generation, a new switching function is constructed, which is suitable for the dynamic switching of the distributed power gen-eration, and realizes the fault location of the distribution network including the distributed power generation. Through the simulation experiment of the 33-node distribution network, the simulation results show that the method can accurately locate single and multiple fault sections, and it can still accurately locate in the case of information distortion, with good fault tolerance, and the MFO is comparable to other algorithms. The comparison converges faster and the global optimization is stronger.