一种混合算法在配电网故障定位中的研究

A Hybrid Algorithm in Distribution Network Fault Location

对于复杂配电网中不止一处发生故障的现象,提出采用两种定位方法相结合的方式进行故障定位,首先根据配电网的拓扑结构运用分区理论将复杂的配电网进行分区,并采用遗传算法对分区后的简化模型进行一次定位,然后再用Bayes公式分别对每个故障区域进行二次定位,综合两次定位的结果最终确定故障区域.由于采用分区思想大大缩小了遗传算法可行解的维数,提高了定位速度;而运用Bayes公式对故障区域进行定位,使得定位的结果更加精确.通过仿真证明该方法在配电网故障定位中的可行性,结果也证实了该方法的准确性.

For multiple faults that happen to complex distribution network,this paper proposes a method combining the genetic algorithm with the Bayes formula for fault location.Firstly,according to the distribution network topology,the theory of partition is used to partition the complex distribution network,and the genetic algorithm is used to locate the simplified model after partition.Secondly,the Bayes formula is used to re-locate each fault zone respectively.The fault zone is eventually determined by synthesizing the result of the two locations.As the idea of partitioning is adopted,the dimension of feasible solution of the genetic algorithm is reduced greatly and the location speed is improved.When the Bayes formula is used to locate the fault zone,the location result is more accurate.The simulation result proves the feasibility and accuracy of this method in distribution network fault locationd.