基于改进遗传算法的配电网分布式风电源选址定容

Locating and Sizing for Distributed Wind Generation in the Distribution Network Based on the Improved Genetic Algorithm

受安装地风速的影响,间歇性分布式风电源接入配电网给分布式风电源的选址定容问题带来一定的挑战并制约风力发电大规模发展。针对此问题,首先建立分布式风力发电出力数学模型,分析其内在机理;其次综合考虑配电网中全年内的风速情况以及负荷水平,根据每小时风机出力效率以及对应的节点小时负荷负载率,构建小时场景,利用改进K-means聚类法进行场景聚类;最后以供电公司最小年费用成本为目标函数,利用改进的遗传算法求解所建模型。测试结果表明,考虑风速及节点负荷的年时序性,规划结果更接近实际情况,具有很好的工程实用价值。

Due to influence of the wind speed at the installation location,intermittent distributed wind power access to the distribution network brings some challenges to the locating and sizing of distributed wind generation and restricts large-scale development of wind power generation. To solve this problem,this paper firstly established a mathematical model for distributed wind power generation and analyzed its internal mechanism. Secondly,considering the wind speed and load level of the distribution network over the whole year,according to the output efficiency per hour and corresponding node load rate per hour,hourly scenario was established and clustered in the improved K-means clustering method. Finally,taking the power company＇s minimum annual cost as objective function,the established model was solved in the improved genetic algorithm. The test results showed that,considering the annual time sequence of wind speed and node load,the planning results were closer to actual situation and had good engineering value.