基于改进小生境遗传算法的Pareto多目标配电网重构  

Pareto multi-objective distribution network reconfiguration based on improved niche genetic algorithm

作　　者：李伟[1] 张振刚[1] 闫宁[1] 

机构地区：[1]华北电力大学经济与管理学院,河北保定071003

出　　处：《电力系统保护与控制》2011年第5期1-5,16,共6页Power System Protection and Control

摘　　要：配电网重构是一个多目标、多约束的复杂非线性组合优化问题,若采用传统的遗传算法处理此类问题,由于其易于陷入局部最优解和随着配电网规模的扩大搜索效率低的问题,难以得到理想结果。建立了Pareto多目标重构数学模型并提出一种改进小生境遗传算法来处理配电网重构问题。算法主要有以下几种特点:设置个体之间的距离判别标准L为动态函数,保持了种群的多样性;采用最优保存策略,提高了算法的收敛速度;交叉、变异采用自适应规则,避免了算法陷入局部最优的情况。另外,Pareto多目标数学模型的引入也使算法更具实际工程意义,采用国外一个实际的配电网络对算法进行了验证。理论分析和算例表明,该算法具有高收敛性、快实时性和强全局稳定性的优点。Distribution network reconfiguration is a complex non-linear combinatorial optimization problem with multi-objective and multi-constrained features.If the traditional genetic algorithm is used to deal with these problems,it is difficult to obtain the desired result because it is easy to fall into local optimal solution and it has the problem of low search efficiency with the expansion of distribution network.This paper establishes the Pareto multi-objective mathematical model of reconfiguration and presents an improved niche genetic algorithm to deal with the issue of distribution network reconfiguration.This algorithm is mainly characterized by the following：setting the distance criterion L between individuals as dynamic function to maintain the population diversity;using the elitist strategy to enhance the convergence speed of the algorithm,and applying the adaptive rules when crossover and mutation to avoid falling into local optimal.In addition,the introduction of the Pareto multi-objective mathematical model makes the algorithm possess more practical engineering significance,and an abroad real distribution network is used to prove the algorithm.Theoretical analysis and example show that this algorithm has high convergence,strong real-time and global stability.

关 键 词：配电网重构 Pareto多目标 小生境遗传算法 自适应 动态距离判别标准 最优保存策略 

分 类 号：TM715[电气工程—电力系统及自动化]