含风电接入多区域电力系统的分散式随机动态经济调度方法  

Decentralized Stochastic Dynamic Economic Dispatch for Multi-area Power Systems With Wind Power Integrated

作　　者：赵文猛 刘明波[2] 周保荣 金小明 姚文峰 卢斯煜 王彤 

机构地区：[1]南方电网科学研究院有限责任公司,广东省广州市510663 [2]华南理工大学电力学院,广东省广州市510640

出　　处：《中国电机工程学报》2017年第24期7087-7098,共12页Proceedings of the CSEE

基　　金：国家重点基础研究发展计划项目(973计划)(2013CB228205);中国南方电网科技项目(CSGTRC-K163007);南方电网科学研究院青年基金项目(SEPRI-K176004)~~

摘　　要：该文提出一种含风电接入的多区域电力系统随机动态经济调度问题的分散求解方法。首先利用场景法对风电随机性进行建模,对每个区域的风电单独抽样,并构建预测场景和误差场景的边界节点相角一致性约束。其次通过复制边界节点变量和构建上层协调器,对多区域电力系统按区域进行分解。最后利用目标级联分析法对多区域随机动态经济调度模型进行分散优化,并采用部分聚合多切割方法高效求解各区域随机优化问题,构建上层协调器主问题、区域预测场景子问题和误差场景子问题交替求解。此外,为提高计算效率,先对预测场景下的确定性多区域动态经济调度模型进行优化,当算法收敛后再交替进行各区域的随机优化和上层协调器的分散协调。最后通过对三区域IEEE RTS系统和一个实际四区域2298节点系统进行测试,证明了该文方法的有效性。This paper proposed a decentralized optimization method for stochastic dynamic economic dispatch in multi-area power systems with wind power integrated. Firstly, the volatility of wind power was represented by sampling scenarios. The scenarios of each area were sampled respectively, and the consensus constraints of phase angles of boundary buses among forecast scenario and error scenarios were built. Secondly, the multi-area power system was decomposed into different areas by creating virtual boarder variables and upper coordinators. Finally analytical target cascading method was applied to solve the multi-area stochastic dynamic economic dispatch in a decentralized way and multi-cut method with partly aggregation was applied to solve the stochastic optimization problem of each area. The master problem of upper coordinator, the forecast scenario sub-problems of each area and the error scenarios sub-problems of each area were solved iteratively. Moreover, the decentralized deterministic optimization was implanted before the decentralized stochastic optimization to increase computation efficiency. The 3-area IEEE RTS system, and a real 4-area 2298-bus provincial power system were tested to demonstrate the effectiveness of the proposed method.

关 键 词：分散优化 随机动态经济调度 多区域电力系统 风力发电 多切割算法 目标级联分析法 

分 类 号：TM732[电气工程—电力系统及自动化]