Distributed multi-step Q(λ) learning for Optimal Power Flow of large-scale power grids

被引：41

作者：

Yu, T. ^{[2
]}

Liu, J. ^{[2
]}

Chan, K. W. ^{[1
]}

Wang, J. J. ^{[2
]}

机构：

[1] Hong Kong Polytech Univ, Hong Kong, Hong Kong, Peoples R China

[2] S China Univ Technol, Guangzhou 510640, Guangdong, Peoples R China

来源：

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS | 2012年 / 42卷 / 01期

基金：

中国国家自然科学基金;

关键词：

Optimal Power Flow (OPF); Q(lambda) learning; Multi-objective optimization; Distributed Reinforcement Learning (DRL); OPTIMIZATION; ALGORITHMS;

D O I：

10.1016/j.ijepes.2012.04.062

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

080906 [电磁信息功能材料与结构]; 082806 [农业信息与电气工程];

摘要：

This paper presents a novel distributed multi-step Q(lambda) learning algorithm (DQ(lambda)L) based on multi-agent system for solving large-scale multi-objective OPF problem. It does not require any manipulation to the conventional mathematical Optimal Power Flow (OPF) model. Large-scale power system is first partitioned to subsystems and each subsystem is managed by an agent. Each agent adopts the standard multi-step Q(lambda) learning algorithm to pursue its own objectives independently and approaches to the global optimal through cooperation and coordination among agents. The proposed DQ(lambda)L has been thoroughly studied and tested on the IEEE 9-bus and 118-bus systems. Case studies demonstrated that DQ(lambda)L is a feasible and effective for solving multi-objective OPF problem in large-scale complex power grid. (C) 2012 Elsevier Ltd. All rights reserved.

引用

页码：614 / 620

页数：7

共 25 条

[1]

Real and reactive power loss allocation in pool-based electricity markets [J].