【摘要】 针对二进制粒子群算法在复杂配电网故障定位时易出现早熟收敛情况,本文提出一种双态二进制粒子群优化算法。通过引入进化因子,把粒子群分成捕食状态和探索状态两个部分,让陷入或即将陷入局部极值的粒子跳出来进行全局搜索。构造故障定位的评价函数,以33节点配电网为例,在故障信息完整和部分畸变的情况下,用该算法与二进制粒子群算法分别对配电网中的单点故障定位和多点故障定位进行仿真分析,结果验证了该算法的高效性和高容错性。更多还原

【Abstract】 Considering that the binary particle swarm optimization algorithm is easy to converge for fault location in a complex distribution network,a bi-state binary particle swarm optimization algorithm is put forward. Through the introduction of evolution factors,the particle swarm is divided into two parts,i.e.,predation and exploration states,then a global search can be performed by the particles that have fallen or will fall into the local extremum. An evaluation function for fault location is constructed,and a 33-bus distribution network is taken as an example. Under the conditions of intact information and partial distortion,the proposed algorithm and the binary particle swarm optimization algorithm are used to simulate and analyze the single-and multi-point fault location in the distribution network,respectively. Results show that the novel algorithm is highly efficient and fault-tolerant.更多还原