Fast Stackelberg equilibrium learning for real-time coordinated energy control of a multi-area integrated energy system
Fast Stackelberg equilibrium learning for real-time coordinated energy control of a multi-area integrated energy system
复制标题
多区域综合能源系统实时协调能量控制的快速Stackelberg平衡学习
DOI：
10.1016/j.applthermaleng.2019.02.053
复制
发表时间：
2019-05
影响因子：
6.4
通讯作者：
Yu Tao
中科院分区：
文献类型：
--
作者：
Zhang Xiaoshun;Yu Tao
在传统互联电网自动发电控制（AGC）的基础上，充分考虑电、热、气等多种能源之间的紧密耦合特性，提出了一种多区域综合能源系统（IES）的实时协调能源控制（CEC）方法。为了提高不同电能扰动下的动态响应性能，在不打破热、气能量平衡的前提下，采用热电联产（CHP）和电-气（P2G）机组参与AGC。从整个多区域IES的角度来看，其目的是尽可能快地平衡电能扰动，这与每个子区域IES的目标不一致（即最小化总运行成本）。为了实现二者之间的有效协调，将实时CEC设计为一个领导者和多个追随者的分层多智能体框架，其中每个智能体只关注自己的收益。为此，提出了一种新的快速Stackelberg平衡学习（FSEL）方法，用于快速获得实时CEC的高质量Stackelberg平衡。为了加快优化速度，引入了实数编码的联想记忆和知识迁移，因此FSEL足以满足实时CEC的在线分布式优化。在海南电网修正模型上，对FSEL的实时CEC性能进行了全面评价。
Based on the traditional automatic generation control (AGC) of interconnected power grids, this paper develops a novel real-time coordinated energy control (CEC) of a multi-area integrated energy system (IES) by fully considering the tight coupling features among various energies (electricity, heat, and gas). In order to improve the dynamic response performance under different electricity energy disturbances, the combined heat and power (CHP) plants and power-to-gas (P2G) units are employed for participating AGC without breaking the heat energy and gas energy balances. From the perspective of the whole multi-area IES, it aims to balance the electricity energy disturbance as fast as possible, which is not consistent with that of each sub-area IES (i.e., minimizing the total operating cost). To achieve an effective coordination among them, the real-time CEC is designed as a hierarchical multi-agent framework with a leader and multiple followers, in which each agent only focuses on its own payoff. Consequently, a novel fast Stackelberg equilibrium learning (FSEL) is proposed for rapidly obtaining a high-quality Stackelberg equilibrium of real-time CEC. Both the real-coded associative memory and knowledge transfer are introduced for accelerating the optimization speed, thus FSEL is adequate to satisfy the online distributed optimization of real-time CEC. The performance of FSEL has been thoroughly evaluated for real-time CEC on the modified model of Hainan power grid of southern China.
登录
查看更多内容
影响因子：
11.2
作者：
Venkatachalam Lakshmanan;M. Marinelli;Junjie Hu;H. Bindner
通讯作者：
Venkatachalam Lakshmanan;M. Marinelli;Junjie Hu;H. Bindner
基于虚拟发电部落的去中心化发电命令调度共识迁移Q-Learning
影响因子：
9.6
作者：
Zhang Xiao Shun;Li Qing;Yu Tao;Yang Bo;Zhang XS
通讯作者：
Zhang XS
DOI：
10.2307/136062
发表时间：
1995-08
期刊：
--
影响因子：
--
作者：
M. Osborne;Ariel Rubinstein
通讯作者：
M. Osborne;Ariel Rubinstein
基于虚拟发电部落的智能电网动态发电指挥调度优化鲁棒协同共识算法
影响因子：
9
作者：
Zhang Xiaoshun;Yu Tao;Yang Bo;Yang Bo;Li Li;Yu T
通讯作者：
Yu T
DOI：
10.1109/pica.2001.932329
发表时间：
2001
期刊：
PICA 2001. Innovative Computing for Power - Electric Energy Meets the Market. 22nd IEEE Power Engineering Society. International Conference on Power Industry Computer Applications (Cat. No.01CH37195)
影响因子：
--
作者：
V. Donde;M. Pai;I. Hiskens
通讯作者：
V. Donde;M. Pai;I. Hiskens