基本信息:
- 专利标题： 一种基于深度强化学习调整电力系统潮流的方法及系统
- 专利标题（英）：Method and system for adjusting power flow of power system based on deep reinforcement learning
- 申请号：CN201910585367.7 申请日：2019-07-01
- 公开(公告)号：CN110443447A 公开(公告)日：2019-11-12
- 发明人： 徐华廷 , 侯金秀 , 郑清平 , 于之虹 , 李淑芳 , 郑惠萍 , 吕颖 , 鲁广明 , 刘新元 , 史东宇 , 马东娟 , 戴红阳 , 李蒙赞 , 王兵 , 杨尉薇 , 曲莹 , 张璐路
- 申请人： 中国电力科学研究院有限公司 , 国网山西省电力公司电力科学研究院 , 北京邮电大学
- 申请人地址： 北京市海淀区清河小营东路15号
- 专利权人： 中国电力科学研究院有限公司,国网山西省电力公司电力科学研究院,北京邮电大学
- 当前专利权人： 中国电力科学研究院有限公司,国网山西省电力公司电力科学研究院,北京邮电大学
- 当前专利权人地址： 北京市海淀区清河小营东路15号
- 代理机构： 北京工信联合知识产权代理有限公司
- 代理人： 姜丽辉
- 主分类号： G06Q10/06
- IPC分类号： G06Q10/06 ; G06Q50/06 ; G06N3/04 ; G06N3/08
The invention discloses a method and a system for adjusting power flow of a power system based on deep reinforcement learning, and relates to the field of automatic adjustment of large power grid power flow. The method comprises the following steps: taking a to-be-adjusted active power range of each target section as input information of a training target; carrying out preliminary screening on anadjustable generator in the power system; randomly obtaining the number of the target section m and the transmission power of the target section m; further determining a fine screening strategy of theadjustable generator based on the training target of each round, and compensating the change of the active power in real time; generating an adjustment strategy by using a deep reinforcement learningalgorithm; and executing an adjustment strategy to adjust the power flow state of the power system until the transmission power is adjusted to a target value. According to the method, automatic calculation of the operation mode of the power system becomes possible, and the method has great engineering application value and popularization prospect.
公开/授权文献:
- CN110443447B 一种基于深度强化学习调整电力系统潮流的方法及系统 公开/授权日：2022-12-09