An Optimization Method-Assisted Ensemble Deep Reinforcement Learning Algorithm to Solve Unit Commitment Problems

TL;DR

This paper introduces an ensemble deep reinforcement learning algorithm enhanced with optimization methods to efficiently solve large-scale unit commitment problems in electricity markets, outperforming traditional methods.

Contribution

It proposes a novel ensemble deep RL approach that integrates optimization techniques to handle high-dimensional UC problems more effectively.

Findings

Outperforms baseline RL and MIQP in numerical tests

Shows strong generalization to unforeseen operational conditions

Reduces calculation time compared to traditional methods

Abstract

Unit commitment (UC) is a fundamental problem in the day-ahead electricity market, and it is critical to solve UC problems efficiently. Mathematical optimization techniques like dynamic programming, Lagrangian relaxation, and mixed-integer quadratic programming (MIQP) are commonly adopted for UC problems. However, the calculation time of these methods increases at an exponential rate with the amount of generators and energy resources, which is still the main bottleneck in industry. Recent advances in artificial intelligence have demonstrated the capability of reinforcement learning (RL) to solve UC problems. Unfortunately, the existing research on solving UC problems with RL suffers from the curse of dimensionality when the size of UC problems grows. To deal with these problems, we propose an optimization method-assisted ensemble deep reinforcement learning algorithm, where UC problems are formulated as a Markov Decision Process (MDP) and solved by multi-step deep Q-learning in an ensemble framework. The proposed algorithm establishes a candidate action set by solving tailored optimization problems to ensure a relatively high performance and the satisfaction of operational constraints. Numerical studies on IEEE 118 and 300-bus systems show that our algorithm outperforms the baseline RL algorithm and MIQP. Furthermore, the proposed algorithm shows strong generalization capacity under unforeseen operational conditions.