首页 | 本学科首页   官方微博 | 高级检索  
     

基于改进分层强化学习的CPS指令多目标动态优化分配算法
引用本文:余涛,王宇名,叶文加,刘前进.基于改进分层强化学习的CPS指令多目标动态优化分配算法[J].中国电机工程学报,2011,31(19).
作者姓名:余涛  王宇名  叶文加  刘前进
作者单位:1. 华南理工大学电力学院,广东省广州市,510640
2. 广东电网公司中山供电局,广东省中山市,528400
基金项目:国家自然科学基金项目(50807016); 广东省自然科学基金项目(9151064101000049); 中央高校基本科研业务费专项资金(2009ZM0251)~~
摘    要:应用经典强化学习方法的控制性能标准(control performance standard,CPS)下自动发电控制(automatic generation control,AGC)指令(CPS指令)由调度端至电网各台机组的分配过程不可避免出现维数灾难问题。提出应用分层强化学习的方法,将全网机组按调频时延做初次分类,CPS指令逐层分配形成任务分层结构。在分层Q学习算法层与层之间引入一个时变协调因子,改进的分层Q学习算法有效提高原算法收敛速度。奖励函数中设计不同的权值线性组合,展示保守及乐观控制下系统CPS控制水平和调节成本的变化关系。南方电网统计性仿真分析表明,改进分层Q学习算法较分层Q学习算法平均收敛时间缩短47%,在复杂随机扰动的环境中改进算法能有效提高系统CPS考核合格率,并降低调节成本约5%。

关 键 词:分层强化学习  协调因子  随机优化  控制性能标准  自动发电控制

Multi-objective Dynamic Optimal Dispatch Method for CPS Order of Interconnected Power Grids Using Improved Hierarchical Reinforcement Learning
YU Tao , WANG Yuming , YE Wenjia , LIU Qianjin.Multi-objective Dynamic Optimal Dispatch Method for CPS Order of Interconnected Power Grids Using Improved Hierarchical Reinforcement Learning[J].Proceedings of the CSEE,2011,31(19).
Authors:YU Tao  WANG Yuming  YE Wenjia  LIU Qianjin
Affiliation:YU Tao1,WANG Yuming2,YE Wenjia1,LIU Qianjin1(1.College of Electric Power,South China University of Technology,Guangzhou 510640,Guangdong Province,China,2.Zhongshan Power Supply Bureau of Guangdong Power Grid Corporation,Zhongshan 528400,China)
Abstract:This paper presented an improved hierarchical reinforcement learning(HRL) algorithm to solve the curse of dimensionality problem in the multi-objective dynamic optimization of automatic generation control(AGC) order dispatch based on control performance standard(CPS).The CPS order dispatch task was decomposed into several subtasks by classifying the AGC committed units according to their response time delay of power regulating.A time-varying coordination factor was introduced between layers of HRL to speed ...
Keywords:hierarchical reinforcement learning(HRL)  coordination factor  stochastic optimization  control performance standard(CPS)  automatic generation control(AGC)  
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号