引用本文:陈俊斌,余涛,殷林飞,唐建林.基于极限动态规划算法的微电网一体化调度与控制[J].控制理论与应用,2019,36(10):1698~1706.[点击复制]
CHEN Jun-bin,YU Tao,YIN Lin-fei,TANG Jian-lin.Integrated dispatch and control of microgrid based on extreme dynamic programming algorithm[J].Control Theory and Technology,2019,36(10):1698~1706.[点击复制]
基于极限动态规划算法的微电网一体化调度与控制
Integrated dispatch and control of microgrid based on extreme dynamic programming algorithm
摘要点击 2478  全文点击 1107  投稿时间:2018-08-22  修订日期:2019-01-17
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2019.80628
  2019,36(10):1698-1706
中文关键词  微电网  频率稳定  经济调度  极限动态规划
英文关键词  microgrid  frequency stability  economic dispatch  extreme dynamic programming(EDP)
基金项目  国家自然科学基金
作者单位E-mail
陈俊斌 华南理工大学电力学院 297206255@qq.com 
余涛* 华南理工大学电力学院 taoyu1@scut.edu.cn 
殷林飞 广西大学电气工程学院  
唐建林 华南理工大学电力学院  
中文摘要
      随着电力电子技术的发展,微电网已成为分布式发电的必然趋势。传统的多时间尺度控制策略之间的配合使用已经很难同时满足高品质频率稳定控制和经济调度的要求。为解决此问题,本文提出极限动态规划算法。所提算法以自适应动态规划算法为框架,以极限学习机作为其评价模块,模型模块,执行模块,预测模块的内核。基于所提算法的微电网一体化调控控制器能替代传统模式下“下垂控制+自动发电控制+经济调度”多时间尺度控制组合策略。最后,为验证所提算法的有效性,在5个节点的微电网模型进行仿真,结果验证了所提极限动态规划算法的可行性和有效性。
英文摘要
      With the development of power electronics technology, microgrid has become an inevitable trend of distributed generation. By means of traditional multi-time scale control strategies, it is difficult to meet the requirements of high quality frequency stability control and economic dispatch at the same time. To solve this problem, an extreme dynamic programming algorithm is proposed .The proposed algorithm takes the adaptive dynamic programming algorithm as the framework and the extreme learning machine as the kernels of its evaluation module, model module, execution module and prediction module. The integrated dispatch and control controller based on the proposed algorithm can replace the multi-time scale control combined strategy of "droop control + automatic generation control + economic dispatch" under the traditional mode .Finally, in order to verify the effectiveness of the proposed algorithm, a microgrid model of 5 nodes is simulated, and the results verify the feasibility and effectiveness of the proposed extreme dynamic programming algorithm.