Meta-heuristics and deep learning for energy applications: Review and open research challenges (2018–2023)

The synergy between deep learning and meta-heuristic algorithms presents a promising avenue for tackling the complexities of energy-related modeling and forecasting tasks. While deep learning excels in capturing intricate patterns in data, it may falter in achieving optimality due to the nonlinear nature of energy data. Conversely, meta-heuristic algorithms offer optimization capabilities but suffer from computational burdens, especially with high-dimensional data. This paper provides a comprehensive review spanning 2018 to 2023, examining the integration of meta-heuristic algorithms within deep learning frameworks for energy applications. We analyze state-of-the-art techniques, innovations, and recent advancements, identifying open research challenges. Additionally, we propose a novel framework that seamlessly merges meta-heuristic algorithms into deep learning paradigms, aiming to enhance performance and efficiency in addressing energy-related problems. The contributions of the paper include: 1. Overview of recent advancements in MHs, DL, and integration. 2. Coverage of trends from 2018 to 2023. 3. Introduction of Alpha metric for performance evaluation. 4. Innovative framework harmonizing MHs with DL for energy problems.