Reinforcement learning and mixed-integer programming for power plant scheduling in low carbon systems: Comparison and hybridisation

Article Computer Science, Theory & Methods

Tackling Climate Change with Machine Learning

David Rolnick et al.

Summary: This article discusses the importance of machine learning in reducing greenhouse gas emissions and helping society adapt to climate change. It identifies existing gaps and proposes solutions and opportunities.

ACM COMPUTING SURVEYS (2023)

Add to Collection

Article Engineering, Electrical & Electronic

Is learning for the unit commitment problem a low-hanging fruit?

S. Pineda et al.

Summary: This article explores the importance of simplicity in solving fundamental problems in power systems and proposes a method using a naive algorithm to solve the unit commitment problem. Through experiments on real power systems, the article demonstrates that this method can significantly improve speed while finding optimal or quasi-optimal solutions.

ELECTRIC POWER SYSTEMS RESEARCH (2022)

Add to Collection

Article Multidisciplinary Sciences

Discovering faster matrix multiplication algorithms with reinforcement learning

Alhussein Fawzi et al.

Summary: Improving the efficiency of fundamental computations, such as matrix multiplication, is crucial for various fields, and machine learning can automate the discovery of efficient algorithms that surpass human-designed ones. AlphaTensor, based on deep reinforcement learning and AlphaZero, successfully discovered algorithms that outperform state-of-the-art ones. Its flexibility in accelerating algorithm discovery and optimization is showcased through different use-cases.

NATURE (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Reinforcement learning and A* search for the unit commitment problem

Patrick de Mars et al.

Summary: This study combines model-free reinforcement learning with advanced search algorithms based on A* search to solve the unit commitment problem with stochastic demand and renewables generation. By applying problem-specific heuristic and anytime algorithm, the study successfully reduces run time, lowers operating costs, and addresses run time variability issue.

ENERGY AND AI (2022)

Add to Collection

Article Energy & Fuels

Applying reinforcement learning and tree search to the unit commitment problem

Patrick de Mars et al.

Summary: Recent advances in AI demonstrate that reinforcement learning methods are superior in decision-making under uncertainty. Day-ahead unit commitment (UC) has become more complex with increasing uncertainty, and researchers propose a novel RL algorithm, guided tree search, to address the issue of exponential growth in action space with increasing number of generators.

APPLIED ENERGY (2021)

Add to Collection

Article Engineering, Electrical & Electronic