摘要
采用遗传算法求解配电网重构问题时,由于交叉和变异操作的随机性使得产生的个体可能无法满足配电网辐射运行的约束,因此需对每个个体进行校验和修正。文中利用配电网的回支关联矩阵动态切割及合并回路以保证辐射状拓扑结构,避免了频繁的拓扑搜索,提高了算法的计算速度。此外,针对降低系统网损的网络重构优化目标,文中在最优流算法"开断电流最小支路"的基础上进一步提炼启发规则,将迭代过程分解为确定电流分点周围开关最优断开组合和电流分点转移2个步骤,有效提高了收敛速度。算例表明所述算法具有染色体长度短、初始种群规模小的优势,能够有效提高计算效率。
Owing to the randomness of crossover and mutation, when genetic algorithm （GA） is used to solve the reconfiguration problem of distribution systems, solutions that do not satisfy the system＇s topological constraints may be generated, and must be fixed. In order to generate a radial network, the basic circuit matrix is introduced to open and combine the loops dynamically. The method keeps the whole calculation from frequent topological search while improving its speed. In addition, a new heuristic rule refined from the one of optimal flow pattern （OFP） is adopted. And the entire reiteration is divided into two steps： searching for the optimal set of switches connected with the current convergence points （CPs） to open and the CPs＂ transfer. Calculation results show that the proposed algorithm has shortened the length of chromosome, reduced the population size, and improved computational efficiency.
出处
《电力系统自动化》
EI
CSCD
北大核心
2013年第19期74-79,共6页
Automation of Electric Power Systems
基金
国家自然科学基金青年基金资助项目(51107037)~~
关键词
配电网重构
最优流
遗传算法
回支关联矩阵
电流分点
distribution system reconfiguration
optimal flow pattern （OFP）
genetic algorithm （GA）
basic circuit matrixcurrent point
