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

T. Yu; J. Liu; Ka Wing Chan; J. J. Wang

doi:10.1016/j.ijepes.2012.04.062

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

T. Yu, J. Liu, Ka Wing Chan, J. J. Wang

The Hong Kong Polytechnic University

Research output: Journal article publication › Journal article › Academic research › peer-review

36 Citations (Scopus)

Abstract

This paper presents a novel distributed multi-step Q(λ) learning algorithm (DQ(λ)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(λ) learning algorithm to pursue its own objectives independently and approaches to the global optimal through cooperation and coordination among agents. The proposed DQ(λ)L has been thoroughly studied and tested on the IEEE 9-bus and 118-bus systems. Case studies demonstrated that DQ(λ)L is a feasible and effective for solving multi-objective OPF problem in large-scale complex power grid.

Original language	English
Pages (from-to)	614-620
Number of pages	7
Journal	International Journal of Electrical Power and Energy Systems
Volume	42
Issue number	1
DOIs	https://doi.org/10.1016/j.ijepes.2012.04.062
Publication status	Published - 1 Nov 2012

Keywords

Distributed Reinforcement Learning (DRL)
Multi-objective optimization
Optimal Power Flow (OPF)
Q(λ) learning

ASJC Scopus subject areas

Energy Engineering and Power Technology
Electrical and Electronic Engineering

More information

10.1016/j.ijepes.2012.04.062

Cite this

@article{9718117a86f847b2ac55de0c1bbf2095,

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

abstract = "This paper presents a novel distributed multi-step Q(λ) learning algorithm (DQ(λ)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(λ) learning algorithm to pursue its own objectives independently and approaches to the global optimal through cooperation and coordination among agents. The proposed DQ(λ)L has been thoroughly studied and tested on the IEEE 9-bus and 118-bus systems. Case studies demonstrated that DQ(λ)L is a feasible and effective for solving multi-objective OPF problem in large-scale complex power grid.",

keywords = "Distributed Reinforcement Learning (DRL), Multi-objective optimization, Optimal Power Flow (OPF), Q(λ) learning",

author = "T. Yu and J. Liu and Chan, {Ka Wing} and Wang, {J. J.}",

year = "2012",

month = nov,

day = "1",

doi = "10.1016/j.ijepes.2012.04.062",

language = "English",

volume = "42",

pages = "614--620",

journal = "International Journal of Electrical Power and Energy Systems",

issn = "0142-0615",

publisher = "Elsevier Ltd",

number = "1",

}

TY - JOUR

T1 - Distributed multi-step Q(λ) learning for Optimal Power Flow of large-scale power grids

AU - Yu, T.

AU - Liu, J.

AU - Chan, Ka Wing

AU - Wang, J. J.

PY - 2012/11/1

Y1 - 2012/11/1

N2 - This paper presents a novel distributed multi-step Q(λ) learning algorithm (DQ(λ)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(λ) learning algorithm to pursue its own objectives independently and approaches to the global optimal through cooperation and coordination among agents. The proposed DQ(λ)L has been thoroughly studied and tested on the IEEE 9-bus and 118-bus systems. Case studies demonstrated that DQ(λ)L is a feasible and effective for solving multi-objective OPF problem in large-scale complex power grid.

AB - This paper presents a novel distributed multi-step Q(λ) learning algorithm (DQ(λ)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(λ) learning algorithm to pursue its own objectives independently and approaches to the global optimal through cooperation and coordination among agents. The proposed DQ(λ)L has been thoroughly studied and tested on the IEEE 9-bus and 118-bus systems. Case studies demonstrated that DQ(λ)L is a feasible and effective for solving multi-objective OPF problem in large-scale complex power grid.

KW - Distributed Reinforcement Learning (DRL)

KW - Multi-objective optimization

KW - Optimal Power Flow (OPF)

KW - Q(λ) learning

UR - http://www.scopus.com/inward/record.url?scp=84861802339&partnerID=8YFLogxK

U2 - 10.1016/j.ijepes.2012.04.062

DO - 10.1016/j.ijepes.2012.04.062

M3 - Journal article

SN - 0142-0615

VL - 42

SP - 614

EP - 620

JO - International Journal of Electrical Power and Energy Systems

JF - International Journal of Electrical Power and Energy Systems

IS - 1

ER -

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

Abstract

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this