Reinforcement Learning for Reactive Power Control
- 2 August 2004
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Systems
- Vol. 19 (3), 1317-1325
- https://doi.org/10.1109/tpwrs.2004.831259
Abstract
This paper presents a Reinforcement Learning (RL) method for network constrained setting of control variables. The RL method formulates the constrained load flow problem as a multistage decision problem. More specifically, the model-free learning algorithm (Q-learning) learns by experience how to adjust a closed-loop control rule mapping states (load flow solutions) to control actions (offline control settings) by means of reward values. Rewards are chosen to express how well control actions cause satisfaction of operating constraints. The Q-learning algorithm is applied to the IEEE 14 busbar and to the IEEE 136 busbar system for constrained reactive power control. The results are compared with those given by the probabilistic constrained load flow based on sensitivity analysis demonstrating the advantages and flexibility of the Q-learning algorithm. Computing times with another heuristic method is also compared.Keywords
This publication has 15 references indexed in Scilit:
- Optimal power flow by enhanced genetic algorithmIEEE Transactions on Power Systems, 2002
- Stability-constrained optimal power flowIEEE Transactions on Power Systems, 2000
- A particle swarm optimization for reactive power and voltage control considering voltage security assessmentIEEE Transactions on Power Systems, 2000
- Homogeneous interior point method for constrained power schedulingIEE Proceedings - Generation, Transmission and Distribution, 2000
- Large scale voltage stability constrained optimal VAr planning and voltage stability applications using existing OPF/optimal VAr planning toolsIEEE Transactions on Power Systems, 1999
- Reinforcement Learning: An IntroductionIEEE Transactions on Neural Networks, 1998
- Contingency constrained optimal reactive power flow procedures for voltage control in planning and operationIEEE Transactions on Power Systems, 1995
- Probabilistic constrained load flow based on sensitivity analysisIEEE Transactions on Power Systems, 1994
- Optimization of fixed tap transformer setting in bulk electric systemsIEEE Transactions on Power Systems, 1991
- Probabilistic techniques in a.c. load-flow analysisProceedings of the Institution of Electrical Engineers, 1977