Univ. Franche-Comté (UFC), FEMTO-ST Institute, France.
American University of Beirut, Electrical and Computer Engineering Department, Beirut, Lebanon.
J Environ Manage. 2024 Mar;354:120392. doi: 10.1016/j.jenvman.2024.120392. Epub 2024 Feb 21.
The Paris Agreement, a landmark international treaty signed in 2016 to limit global warming to 2°C, has urged researchers to explore various strategies for achieving its ambitious goals. While Renewable Energy (RE) innovation holds promise, it alone may not be sufficient as critical deadlines approach. This field of research presents numerous challenges, foremost among them being the costliness of materials involved. However, emerging advancements in Machine Learning (ML) technologies provide a glimmer of hope; these sophisticated algorithms can accurately predict the output of energy systems without relying on physical resources and instead leverage available data from diverse energy platforms that have emerged over recent decades. The primary objective of this paper is to comprehensively explore various ML techniques and algorithms in the context of Renewable Energy Systems (RES). The investigation will address several vital inquiries, including identifying and evaluating existing RE technologies, assessing their potential for further advancement, and thoroughly analyzing the challenges and limitations associated with their deployment and testing. Furthermore, this research examines how ML can effectively overcome these obstacles by enhancing RES performance. By identifying future research opportunities and outlining potential directions for improvement, this work seeks to contribute to developing environmentally sustainable energy systems.
《巴黎协定》是 2016 年签署的一项具有里程碑意义的国际条约，旨在将全球变暖限制在 2°C 以内，该协定敦促研究人员探索各种策略来实现其雄心勃勃的目标。虽然可再生能源（RE）创新具有潜力，但随着关键截止日期的临近，它可能本身还不够。该研究领域存在许多挑战，其中最重要的是所涉及材料的昂贵性。然而，机器学习（ML）技术的新兴进展提供了一线希望；这些复杂的算法可以在不依赖物理资源的情况下准确预测能源系统的输出，而是利用近几十年来出现的各种能源平台上的可用数据。本文的主要目的是全面探讨可再生能源系统（RES）背景下的各种 ML 技术和算法。该研究将涉及几个重要的问题，包括确定和评估现有的 RE 技术，评估它们进一步发展的潜力，并彻底分析与部署和测试相关的挑战和限制。此外，本研究还探讨了 ML 如何通过提高 RES 性能来有效地克服这些障碍。通过确定未来的研究机会并概述潜在的改进方向，这项工作旨在为开发环境可持续的能源系统做出贡献。
