基于适应性权重遗传算法的配电网多目标综合优化

针对配电网多目标综合优化问题的复杂性,以及网络重构、电容器投切的离散控制变量优化问题,基于遗传算法从全局角度来求解此优化问题,研究了网络重构、电容器投切的编码方案,以及两者的综合编码方案,并研究给出合理的选择、交叉、变异以及保留操作策略。针对线损、电压质量以及线路负载多目标的各个目标权重难以确定问题,进一步引入适应性权重遗传算法,随着遗传代数的进化,算法自适应地给出各个目标权重。仿真算例验证了所提算法用于配电网多目标综合优化的合理性。

Study on Multi-Objective Comprehensive Optimization Based on Genetic Algorithm with Adaptive Weight in Distribution System

Network reconfiguration and capacitor switching are two important measures of distribution system optimization,synthetically using them to decrease network loss,increase voltage quality and balance line loading,such multi-objective optimization is comprehensive optimization in distribution system.In view of the optimization of network reconfiguration and discrete controlled variables of capacitor switching as well as the consideration on the complexity of multi-objective synthesis optimization,genetic algorithm is presented for finding an overall optimal solution.The network reconfiguration coding,capacitor switching coding and their synthetic coding are studied and suitable strategies for the selection,cross,mutation and retention operation are confirmed.Due to the difficulty in identifying the line loss,voltage quality and each objective weight of multi-objective line load,the adaptive weight genetic algorithm is introduced further.With the evolution of hereditary algebra,each objective weight is given adaptively by the genetic algorithm.Case studies show that the proposed algorithm for distribution network comprehensive optimization is reasonable.