Interpretable Machine Learning for Unit Commitment Solutions

This research article presents a methodology utilizing interpretable machine learning techniques to enhance the understanding of solutions for the unit commitment problem in power systems. The study emphasizes the importance of explainability in mathematical models, particularly as the energy sector transitions towards sustainable sources. The authors apply their methodology to a case study based on the IEEE 118N system, aiming to provide clear and concise descriptions of optimal commitment solutions that can be easily understood by stakeholders. The paper outlines how traditional machine learning models often operate as black boxes, lacking transparency in their decision-making processes. By employing model trees and node clustering, the proposed approach seeks to extract insights that were previously reliant on human experience. The results indicate that the methodology successfully simplifies the explanation of operational modes, thereby facilitating better decision-making in the context of unit commitment. This work addresses a significant gap in the existing literature by focusing on the interpretability of machine learning applications in energy systems.