Development and Optimization of Energy Management Strategies for Hybrid Heavy Duty Vehicle: From Model to Hardware in the Loop Validation

This paper presents a methodology for energy management strategies (EMS) development and integration into hardware in the loop testing for a heavy duty fuel cell powered vehicle use case. Hardware in the loop methodology allows for safe and fast development of EMS and insure the real-time capability of the controller. They also allow to work at critic operating points that could lead to degradation on real systems. In this paper, the EMS are implemented on a real world use case and are applied to a multi-physical dynamic model of the hybrid fuel cell and battery system. The EMS is developed and improved off-line by a model in the loop process using mixed integer linear programming and global optimization using Bayesian methods. Then, the model and the EMS are tested in a hardware in the loop platform. The on-line power splitting between the fuel cell and the battery is performed by a lookup table. Then by a model in the loop optimization, an improved version of the lookup table allows to reduce the equivalent hydrogen consumption by 25 % compared to the original expert-tuned lookup table.