促进间歇性分布式电源高效利用的主动配电网双层场景规划方法  

Bi-level Scenario Programming of Active Distribution Network for Promoting Intermittent Distributed Generation Utilization

作　　者：曾博[1] 刘念[1] 张玉莹[1] 杨煦[1] 张建华[1] 刘文霞[1] 

机构地区：[1]华北电力大学新能源电力系统国家重点实验室,北京102206

出　　处：《电工技术学报》2013年第9期155-163,171,共10页Transactions of China Electrotechnical Society

基　　金：国家自然科学基金(51277067);美国能源基金会(G-1006-12630)资助项目

摘　　要：提高对间歇性可再生能源发电(DREG)的接纳以促进终端电能低碳化是主动配电网(AND)规划工作面临的新增重要目标。为此,基于主从逻辑结构,建立了面向促进间歇性分布式电源高效利用的ADN双层场景规划模型。该模型计及了间歇性DG规划布局、输出功率及负荷的不确定性,以单位用电量的综合经济代价最小及DREG年有效发电量最大分别作为上、下层优化目标,在考虑ADN主动网络管理(ANM)优化控制能力的基础上,求解电网最优规划方案。根据模型特点,设计了一种综合蒙特卡洛随机模拟、和声搜索与列文伯格-马夸尔特算法组成的混合求解算法。以51节点配网系统为算例验证了所提模型及算法的有效性以及ADN规划中考虑ANM潜力作用的必要性。Promoting the accommodation of distributed renewable energy generation（DREG） to reduce the carbon-print of electricity is an additional planning objective under active distribution network（ADN） frame. To respond this, a novel bi-level scenario programming model of ADN is proposed in this paper. With the consideration of uncertainties associated in the DREG planning, operation, and load fluctuation, the model aims to minimize the comprehensive environmental-economic burden for per kilowatt hour of electricity consumed by end-users, while maximizing DREG generation along all the operation states. The contribution of active network management（ANM） in ADN is highlighted in the study and corresponding controlling parameters are dealt as decision variables in the model formulation. A powerful heuristic approach, i.e. harmony search algorism, and the fast-converging Levenberg-Marquardt algorithm（LMA） are combined to provide accurate planning decisions. The proposed model is tested on the 51-node distribution system to illustrate its feasibility and validity in the ADN planning.

关 键 词：主动配电网 间歇性分布式电源 双层场景规划 主动网络管理 混合算法 

分 类 号：TM715[电气工程—电力系统及自动化]