A survey on AI-based scheduling models, optimization and prediction for hydropower generation: Variants, challenges, and future directions

, et

référence BibTeX

The most common form of renewable energy production around the world is hydropower. As a result of the growing demand for robust and environmentally friendly methods of energy generation around the world, it is imperative to develop and improve the current energy production processes. Machine learning has significantly contributed to numerous academic domains in the past decade, and hydropower is no exception. All three horizons of hydropower models, short-term, medium-term, and long-term, could benefit from machine learning. Currently, the majority of hydropower scheduling models employ dynamic programming. As a result of machine learning's use of a new evolution of pre-existing methodologies, unconstrained optimization also enables improvement. In this research paper, we review the current state of the hydropower scheduling problem and the development of machine learning as a type of optimization problem and prediction tool. In addition, the paper investigates the other conceivable roles that machine learning has taken on in recent years. To the best of our knowledge, this is the first survey article that provides a comprehensive overview of machine learning and artificial intelligence application in the hydroelectric power industry for Scheduling, Optimization, and Prediction.

, 28 pages

Axe de recherche

Application de recherche



G2252.pdf (600 Ko)