摘要
为了实现主动配电网分布式电源(distributed generation,DG)优化配置,提出了一种多阶段优化配置方法。该方法首先应用多场景聚类方式考虑DG与负荷的时序性与随机性,将场景空间划分为若干个大致均等的子空间,在每个子空间中分别取各维分量坐标的代数平均值对应的点为该场景的簇心;然后将求取目标函数的最优解问题转化为两阶段优化问题,第一阶段对各待选节点风电场的安装容量进行优化并筛选出极端场景,第二阶段在极端场景下优化需求侧控制变量;最后利用改进的进化策略算法对IEEE 33节点算例进行仿真计算。结果表明:与现有的直接法相比,所提出的多阶段优化方法简单有效,可以求出最优解,引起的误差可以忽略不计。
Amulti-stage optimal configuration method for distributed generation in active distribution network is proposed.Firstly,the multi scene clustering method is used to consider the time sequence and randomness of distributed generation and load,and the scene space is divided into several roughly equal subspaces.In each subspace,the points corresponding to the algebraic average value of the component coordinates of each dimension are taken as the cluster center of the scene.Then the optimal solution of the objective function is transformed into a two-stage optimization problem:firstly,the installation capacity of each wind farm is optimized and the extreme scenarios are screened out;secondly,the demand side control variables are optimized in the extreme scenarios.Finally,the improved evolutionary strategy algorithm is used to simulate the IEEE 33 bus system.The results show that compared with the existing direct methods,the multi-stage optimization method is simple and effective,and can obtain the optimal solution,and the errors caused can be ignored.
作者
李国梁
韩军峰
马平
LI Guoliang;HAN Junfeng;MA Ping(Shandong Zhenghan Survey and Design Institute Co.,Ltd.,Jinan,Shangdong 250100,China;Rizhao Power Supply Company of State Grid Shandong Electric Power Company,Rizhao,Shandong 276800,China;College of Electrical Engineering,Qingdao University,Qingdao,Shandong 266071,China)
出处
《广东电力》
2021年第4期53-58,共6页
Guangdong Electric Power
基金
山东省重点研发计划项目(2017GGX50114)。
关键词
场景聚类
多阶段优化
需求侧管理
分布式电源
scene clustering
multi-stage optimization
demand side management
distributed generation
