Energy saving and environmental protection are two main themes for automobile development in the world. Fuel cell hybrid electric vehicle (FCHEV) aims to achieve these two goals. Since multiple power sources including fuel cell and energy storage sources are used to power the vehicle, an energy management strategy (EMS) plays an important role to improve the vehicle''s performance. Objectives should be pre-set in the process of designing EMS. According to the analysis of characteristics of FCHEV, classification of objectives and methods to achieve these objectives for different control strategies are presented and concluded.
In a fuel cell electric vehicle (FCEV) powered by a metal hydride tank, the performance of the tank is an indicator of the overall health status, which is used to predict its behaviour and make appropriate energy management decisions. The aim of this paper is to investigate how to evaluate the effects of charge/discharge cycles on the performance of a commercial automotive metal hydride hydrogen storage system applied to a real FCEV. For this purpose, a mathematical model is proposed based on uncertain physical parameters that are identified using the stochastic particle swarm optimisation (PSO) algorithm combined with experimental measurements. The variation of these parameters allows an assessment of the degradation level of the tank’s performance on both the quantitative and qualitative aspects. Simulated results derived from the proposed model and experimental measurements were in good agreement, with a maximum relative error of less than . The validated model was used to establish the correlations between the observed degradations in a hydride tank recovered from a real FCEV. The results obtained show that it is possible to predict tank degradations by developing laws of variation of these parameters as a function of the real conditions of the use of the FCEV (number of charging/discharging cycles, pressures, mass flow rates, temperatures).
This paper aims to propose an accurate method for estimating the state of charge (SoC) in metal hydride tanks (MHT) to enhance the energy management of hydrogen-powered fuel cell systems. Two data-driven prediction methods, Long Short-Term Memory (LSTM) networks and Support Vector Regression (SVR), are developed and tested on experimental charge/discharge data from a dedicated MHT test bench. Three distinct LSTM architectures are evaluated alongside an SVR model to compare both generalization performance and computational overhead. Results demonstrate that the SVR approach achieves the lowest root mean square error (RMSE) of 0.0233% during discharge and 0.0283% during charge, while also requiring only 164 ms per inference step for both cycles. However, LSTM variants have a higher RMSE and significantly higher computational cost, which highlights the superiority of the SVR method.