基于组件式GIS网络分析与多目标遗传算法的城市中压配电网规划

Urban Medium-Voltage Distribution Network Planning Based on ComGIS Network Analysis and Multi-objective Genetic Algorithm

城市中压配电网规划是一个多目标组合优化问题。通过组件式地理信息系统(ComGIS)平台建立规划区域的网络数据集,在此基础上使用多目标遗传算法进行优化。利用ComGIS平台的网络分析功能优化单环最优路径,简化了遗传编码,并根据编码特点设计交叉、变异操作,提高了遗传算法的效率。利用基于Pareto秩的个体适应度函数引导多目标遗传算法的进化方向,最终得到接近多目标优化问题Pareto最优解集的一组两端供电网络方案。所得规划方案集便于决策人员选择,并且有效结合了规划区域的地理信息,有较强的实际意义。对某实际系统的应用结果表明了文中所提方法的有效性。

Urban medium-voltage distribution network planning is a multi-objective combinatorial optimization problem. The network dataset of the planning district is built through Component GIS （ComGIS） platform, based on which the multi-objective genetic algorithm is used. The network analysis function of ComGIS platform is used to solve single loop optimization, simplifying the codification strategy. And the crossover and mutation operator are designed according to the character of codification. The orientation of evolution is directed through fitting function based on Pareto order of multi-objective optimization problem. Finally, a set of candidate loop networks is gotten, which are convenient for engineers to choose. The algorithm has been applied to a real power distribution system, showing its potential in practice.