Optimal energy management in a dual-storage fuel-cell hybrid vehicle using multi-dimensional dynamic programming

Abstract

Hybrid storage systems consisting of battery and ultra-capacitor have recently emerged as an alternative to the conventional single buffer layout in hybrid vehicles. Their high power and energy density could improve the performance indices of the vehicle, provided that an optimal energy management strategy is employed that could handle systems with multiple degrees of freedom (DOF). The majority of existing energy management strategies is limited to a single DOF and the small body of work on multi-DOF systems is mainly heuristic-based.We propose an optimal solution to the energy management problem in fuel-cell hybrid vehicles with dual storage buffer for fuel economy in a standard driving cycle using multi-dimensional dynamic programming (MDDP). An efficient MDDP code is developed using MATLAB™'s vectorization feature that helps reduce the inherently high computational cost of MDDP. Results of multiple simulated experiments are presented to demonstrate the applicability and performance of the proposed strategy. A comparison is also made between a single and a double buffer fuel-cell hybrid vehicle in various driving cycles to determine the maximum reduction in fuel consumption that can be achieved by the addition of an ultra-capacitor.