Distribution Network Planning Method Based on Hybrid Genetic Algorithm
Yunfeng Shao,Yuanming Sun,Yajing Wang,Zhongjing Ma,Yongqiang Liu,Yang Zhao
DOI: https://doi.org/10.1088/1742-6596/1673/1/012032
2020-01-01
Journal of Physics Conference Series
Abstract:Facing the discrete, multi-constrained, non-linear, multi-objective combination optimization problem of distribution network grid planning, some traditional heuristic algorithms such as genetic algorithms sometimes fall into local optimum. This paper proposes a distribution network planning method based on hybrid genetic algorithm The algorithm consists of two stages. In the first stage, the genetic algorithm is used to obtain the initial planning scheme. In the second stage, the initial planning scheme obtained in the first stage is used to form the planned route set. The improved minimum spanning tree method is used to obtain the final planning scheme. In order to make full use of the effective information obtained in the first stage, this paper proposes a transmission line classification method to assess the importance of the transmission line, provide guidance for the second stage, and improve the search efficiency and accuracy. The algorithm solves the problem that heuristic algorithms such as genetic algorithm often fall into local optimization to a certain extent, and the problem of slow convergence when the minimum spanning tree algorithm has a large number of lines to be planned.