Predictive Energy Management for Fuel Cell Hybrid Vehicles Considering Efficiency and Safety
Y. Liu, Mince Li, Yujie Wang, Zhendong Sun, Zonghai Chen
Abstract
An effective energy coordination strategy is pivotal for fuel cell hybrid vehicles (FCHVs), aiming to optimize the energy synergy between lithium-ion batteries and fuel cells. This serves as the foundation for the sustainable progress in clean energy transportation. This article introduces an energy management strategy designed for real-time applications, with a focus on ensuring the security, efficiency, and prolonged operation of FCHVs. Precisely, leveraging precise system modeling, this article integrates considerations for efficiency awareness, system health, and battery thermal penalties into a structured model predictive control framework. The designed framework utilizes sequential quadratic programming to derive optimal control sequences while operating within multiphysical constraints. Moreover, a deep learning algorithm is employed for velocity prediction, and the article explores the effects of network structure and input–output feature size on prediction accuracy. Ultimately, the proposed strategy is implemented under diverse operating conditions to validate its robustness and to demonstrate its effectiveness in comparison with existing real-time optimization method. In addition, contrasting experiments are conducted to examine the impact of different weights in the multiobjective function.