作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2009, Vol. 35 ›› Issue (5): 9-11. doi: 10.3969/j.issn.1000-3428.2009.05.004

• 博士论文 • 上一篇    下一篇

基于自适应动态规划的系统边际电价预测

张志刚1,马光文1,叶伟宝2,张军良1   

  1. (1. 四川大学水电工程学院,成都 610065;2. 浙江省丽水市莲都供电局,丽水 323000)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-03-05 发布日期:2009-03-05

System Marginal Price Forecasting Based on Adaptive Dynamic Programming

ZHANG Zhi-gang1, MA Guang-wen1, YE Wei-bao2, ZHANG Jun-liang1   

  1. (1. Institute of Hydroelectric Engineering, Sichuan University, Chengdu 610065; 2. Zhejiang Province Lishui Liandu Electric Power Supply Bureau, Lishui 323000)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-03-05 Published:2009-03-05

摘要: 针对目前电价预测算法的局限性,提出一种基于自适应动态规划方法的自学习、自适应智能算法。按照Bellman最优化基本原理,使用Agent逐步与环境的交互作用来寻求预测电价和实际电价的误差最小值,得到系统边际电价的最优解。采用美国加州电力市场的数据进行电价预测仿真。与常规方法相比,该方法的拟合精度和平均绝对百分误差均有很大提高。

关键词: 自适应动态规划, 智能体, 系统边际电价, 电力市场

Abstract: Aimed at disadvantages of the normal electricity price forecasting algorithm, this paper proposes which a method based on Adaptive Dynamic Programming(ADP) that has self study and self adaptive ability. Based on the Bellman optimization theory, that is gradually interaction with the environment to seek error minimum of the actual price and the forecast price with the Agent. It can obtain the optimal solution of System Marginal Price(SMP). It uses the California electricity market price data as forecasting simulation. Compared with the conventional method, results show that fitting accuracy and MAPE are greatly improved.

Key words: Adaptive Dynamic Programming(ADP), Agent, System Marginal Price(SMP), power markets

中图分类号: