Mathematical Modeling and Genetic Algorithm-Based Energy Management in Hydrogen PEM Fuel Cell Electric Vehicles

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Provides a detailed PEM FCEV model and GA-based EMS that yields 9-12% hydrogen consumption improvement, useful for further optimization studies.

🏢実務担当者:Offers a validated EMS strategy that can be adapted for commercial FCEV control systems to enhance fuel economy.

Hydrogen Fuel Cell Electric Vehicles (FCEVs) represent a significant trajectory in vehicular decarbonization, harnessing the inherently high energy density of diatomic hydrogen within electrochemical conversion systems. When sourced via renewable pathways, such hydrogen facilitates propulsion architectures characterized by zero tailpipe emissions, enhanced energy efficiency, and extended operational range profiles. Realizing peak systemic efficacy necessitates the synergistic orchestration of high-fidelity fuel cell stack design, resilient compressed gas storage modalities, and nuanced energy governance protocols. To reduce transient stressors and guarantee long-term electrochemical stability, employing multi-scale modeling and predictive simulation, combined with constraint-aware architectural synthesis, is crucial in handling stochastic driving conditions spectra.This study develops a high-fidelity mathematical plant model of a hydrogen Proton Exchange Membrane (PEM) fuel cell vehicle and implements advanced Energy Management Strategies (EMS). The FCEV plant model is developed with the forward approach method, taking into account the power limitations of the power plant. A PEM fuel cell system is accurately and in detail modeled, representing voltage loss mechanisms. The performance of the mathematical model was calibrated with the experimental results with an error margin of 8-10%. Whereas, a permanent magnet synchronous motor is modeled mathematically along with a Field-Oriented Controller (FoC) for ensuring precise torque regulation.Energy Management Strategies (EMS) optimize fuel cell and battery coordination to boost vehicle performance and efficiency. Online EMS adapts control using real-time data, while offline EMS applies machine learning to past driving patterns for predictive energy allocation. In this study, a Genetic Algorithm (GA)-based EMS, which is one of the types of offline EMS, is implemented to enhance fuel economy, dynamic performance, and component-level energy usage. Compared to non-optimized operation, the GA approach offers improved power split efficiency, 9-12% improvement in hydrogen consumption, resulting in lower energy consumption and enhanced overall vehicle performance.This work improves PEM FCEV technology through better design, simulation, and optimization methods, laying a solid foundation for future advancements in sustainable and efficient transportation.

• semanticscholar https://doi.org/10.4271/2026-26-0257first seen 2026-05-15 19:58:24