典型文献
Active Power Correction Strategies Based on Deep Reinforcement Learning—Part Ⅱ:A Distributed Solution for Adaptability
文献摘要：
This article is the second part of Active Power Correction Strategies Based on Deep Reinforcement Learning.In Part Ⅱ,we consider the renewable energy scenarios plugged into the large-scale power grid and provide an adaptive algorithmic implementation to maintain power grid stability.Based on the robustness method in Part Ⅰ,a distributed deep reinforcement learning method is proposed to overcome the influence of the increasing renewable energy penetration.A multi-agent system is implemented in multiple control areas of the power system,which conducts a fully cooperative stochastic game.Based on the Monte Carlo tree search mentioned in Part Ⅰ,we select practical actions in each sub-control area to search the Nash equilibrium of the game.Based on the QMIX method,a structure of offline centralized training and online distributed execution is proposed to employ better practical actions in the active power correction control.Our proposed method is evaluated in the modified global competition scenario cases of"2020 Learning to Run a Power Network-Neurips Track 2".
文献关键词：
中图分类号：
作者姓名：
Siyuan Chen;Jiajun Duan;Yuyang Bai;Jun Zhang;Di Shi;Zhiwei Wang;Xuzhu Dong;Yuanzhang Sun
作者机构：
School of Electrical Engineering and Automation,Wuhan University,Wuhan 430072,China;GEIRI North America,San Jose,CA 95134,USA
文献出处：
引用格式：
[1]Siyuan Chen;Jiajun Duan;Yuyang Bai;Jun Zhang;Di Shi;Zhiwei Wang;Xuzhu Dong;Yuanzhang Sun-.Active Power Correction Strategies Based on Deep Reinforcement Learning—Part Ⅱ:A Distributed Solution for Adaptability)[J].中国电机工程学会电力与能源系统学报（英文版）,2022(04):1134-1144
A类：
Adaptability,Neurips
B类：
Active,Power,Correction,Strategies,Based,Deep,Reinforcement,Learning,Part,Distributed,Solution,This,article,second,part,In,consider,renewable,energy,scenarios,plugged,into,large,scale,power,grid,provide,adaptive,algorithmic,implementation,maintain,stability,robustness,method,distributed,deep,reinforcement,learning,proposed,overcome,influence,increasing,penetration,agent,system,implemented,multiple,control,areas,which,conducts,fully,cooperative,stochastic,game,Monte,Carlo,tree,search,mentioned,select,practical,actions,each,sub,Nash,equilibrium,QMIX,structure,offline,centralized,training,online,execution,employ,better,active,correction,Our,evaluated,modified,global,competition,cases,Run,Network,Track
AB值：
0.58345
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
