基于无向生成树的并行遗传算法在配电网重构中的应用

随着以风电、光伏为代表的不可控型分布式电源在配电网中的渗透率日益提高,分布式电源出力的不确定性成为配电网重构中必须考量的重要因素。因此建立了以系统网损最小为目标,计及潮流方程、节点电压、支路潮流和配电网开环运行约束的配电网重构随机优化模型。模型以机会约束描述节点电压和支路潮流约束,采用基于拉丁超立方采样的蒙特卡洛法随机潮流进行检验。提出了基于无向生成树的并行遗传算法以实现配电网重构模型的并行求解。IEEE 33节点系统的测试结果验证了模型的合理性,并将所提出的算法与基于无向生成树的遗传算法、粒子群优化算法、蚁群搜索算法和改进和声搜索算法进行比较,验证了其高效性。

Application of Undirected Spanning Tree-based Parallel Genetic Algorithm in Distributed Network Reconfiguration

With the increasing penetration of uncontrollable distributed generators such as distributed wind generator and photovoltaic generation, the uncertainty of distributed generator has become a key factor in distribution network reconfiguration. Hence, a stochastic network reconfiguration model was used to minimize the mathematical expectation of overall power loss, considering the constraints of power flow, voltage and topology. The stochastic power flow based on Latin Hypercube sampling-Monte Carlo simulation (LHS-MCS) was used to examine the chance constraints of nodal voltage and branch power flow. To improve computational efficiency, parallel undirected spanning tree-based genetic algorithm (PSTGA) was proposed to solve the network reconfiguration in parallel. The results of test in IEEE 33-bus distribution system have confirmed the rationality of model. The efficiency of PSTGA has been proved by comparing with undirected spanning tree-based genetic algorithm, particle swarm optimization, ant colony searching optimization and improved harmony search algorithm. This work is supported by National Key Technologies R&D Program (No. 2013BAA01B04) and State Grid Corporation of China (No. 520940120036).