Model Predictive Control-Based Energy Management Strategy for A Series Hybrid Electric Tracked Vehicle

Hong Wang,Yanjun Huang,Amir Khajepour,Qiang Song
DOI: https://doi.org/10.1016/j.apenergy.2016.08.085
IF: 11.2
2016-01-01
Applied Energy
Abstract:The series hybrid electric tracked bulldozer (HETB)'s fuel economy heavily depends on its energy management strategy. This paper presents a model predictive controller (MPC) to solve the energy management problem in an HETB for the first time. A real typical working condition of the HETB is utilized to develop the MPC. The results are compared to two other strategies: a rule-based strategy and a dynamic programming (DP) based one. The latter is a global optimization approach used as a benchmark. The effect of the MPC's parameters (e.g. length of prediction horizon) is also studied. The comparison results demonstrate that the proposed approach has approximately a 6% improvement in fuel economy over the rule-based one, and it can achieve over 98% of the fuel optimality of DP in typical working conditions. To show the advantage of the proposed MPC and its robustness under large disturbances, 40% white noise has been added to the typical working condition. Simulation results show that an 8% improvement in fuel economy is obtained by the proposed approach compared to the rule-based one. (C) 2016 Elsevier Ltd. All rights reserved.
What problem does this paper attempt to address?