中长期发电计划的值函数近似动态规划方法.pdfVIP

第 35 卷 第 20 期 中 国 电 机 工 程 学 报 Vol.35 No.20 Oct. 20, 2015 2015 年 10 月 20 日 Proceedings of the CSEE ©2015 Chin.Soc.for Elec.Eng. 5199 DOI ：10.13334/j.0258-8013.pcsee.2015.20.010 文章编号：0258-8013 (2015) 20-5199-11 中图分类号：TM 73 中长期发电计划的值函数近似动态规划方法 1 1 2 2 梁振成 ，韦化 ，李凌 ，田君杨 (1．广西电力系统最优化与节能技术重点实验室(广西大学) ，广西壮族自治区 南宁市 530004 ； 2．广西电网电力调度控制中心，广西壮族自治区 南宁市 530023) A Method of Value Function Approximate Dynamic Programming for Medium/Long Term Generation Scheduling 1 1 2 2 LIANG Zhencheng , WEI Hua , LI Ling , TIAN Junyang (1. Guangxi Key Laboratory of Power System Optimization and Energy Technology (Guangxi University), Nanning 530004, Guangxi Zhuang Autonomous Region, China; 2. Guangxi Power Grid Dispatching Control Center, Nanning 530023, Guangxi Zhuang Autonomous Region, China) ABSTRACT: With respect to the features in medium/long- 摘要：针对中长期发电优化调度问题周期长、规模大、随机 term optimal scheduling such as long period, large scale and 性强等特点，提出一种基于近似动态规划(approximate strong randomness, a multi-stage optimized decision model dynamic programming，ADP) 的多阶段优化决策模型。以预 based on approximate dynamic programming (ADP) was 测场景表示日前市场煤价、流域来水、风速等随机变量，将 proposed. Prediction scenarios were used to represent 远期合约购煤与市场购煤、水库用水等视为 ADP 框架下的 uncertainties in day-ahead coal prices, water inflows and wind 阶段决策。分阶段决策降低了问题的求解规模和难度，提出 speeds. Decisions including forward contracts coal, day-ahead 的燃煤和水库蓄水的值函数近似策略解决了如何优化决策 coal and water usage were recognized