AI-Based Energy Management in Fuel-Cell Powertrains: A Review of Control Strategies and Implementation Challenges

In Fuel-Cell Hybrid Powertrains (FCHPs), optimizing energy management is essential for enhancing fuel efficiency and maximizing powertrain performance. Conventional control strategies, such as rule-based (RB) and optimization-based (OB) methods, have been widely utilized but have limitations in adaptability under real-time operation. Recently, AI-based (AIB) approaches leveraging artificial intelligence (AI) technologies, such as machine learning (ML) and deep learning (DL), present an alternative for improving predictive capabilities and adaptive energy distribution. Therefore, a literature review comparing traditional and AIB control strategies under real-time operation to date is highly essential, and this is the main topic of the manuscript. In theory, an AIB energy management system (EMS) potentially offers dynamic power allocation based on real-time data acquisition. However, challenges such as high computational demands and extensive data training must be fully addressed. The review emphasizes that future research should focus on integrating AI technologies with physics-based models and exploring both cloud-edge and low-cost local computing for enhanced adaptability and scalability in real-world applications.