微网能源系统的滚动优化管理  

Receding-Horizon Optimization for Microgrid Energy Management

作　　者：何舜[1] 郑毅[1] 蔡旭[1] 吴小东[2] 时珊珊[3] 

机构地区：[1]风力发电研究中心(上海交通大学),上海市闵行区200240 [2]上海电器科学研究所(集团)有限公司,上海市普陀区200063 [3]国网上海市电力公司电力科学研究院,上海市杨浦区200437

出　　处：《电网技术》2014年第9期2349-2355,共7页Power System Technology

基　　金：国家科技支撑计划项目(2013BAA01B04);上海市科委项目(12dz1200203)~~

摘　　要：以上海某园区微网为例,提出一种包含分布式能源、储能、可平移负荷3类电力资源的能源管理优化方法。该方法首先将可再生能源进行完全消纳,然后运用可控分布式能源、储能和可平移负荷对削减后的微网负荷进行第二轮优化。考虑到优化问题存在大量的非线性规划,提出分解迭代算法,将"源–荷–储"3类可控资源利用粒子群算法进行独立求解,并通过迭代使得整体解逼近全局最优解。同时,本系统针对微网预测困难,提出滚动优化方法,提高整体优化的准确性和实时性。算例结果验证了该方法的有效性。Taking a certain Sci-tech Park＆#39;s microgrid in Shanghai as example, an energy source management optimization method, in which three electric power resources such as distributed generation, energy storage system and shiftable load are included, is proposed. In the proposed method firstly the output of renewable energy sources are fully consumed by the load in the park; then the controllable DGs, energy storage system and shiftable load are utilized to perform the second round optimization for the load that has been reduced. Considering the fact that there are a lot of non-linear programmings in the optimization problem, a decomposition iteration algorithm, which independently solves three kinds of controllable resources such as DG, load and energy storage by particle swarm optimization （PSO） to make the solutions of the three kinds of controllable resources closed to globally optimal solution through iterations, is put forward. Besides, in allusion to the difficulty in load prediction due to the randomness of the load in microgrid, a receding horizon optimization method is given to improve the accuracy and real-time of global optimization. The effectiveness of the given method is validated by the results of case calculation.

关 键 词：微网 能源管理系统 分布式能源 储能 可平移负荷 滚动优化 粒子群算法 

分 类 号：TM721[电气工程—电力系统及自动化]