ABSTRACT
Hybridized powertrains with multiple onboard energy converters offer an additional degree of freedom to fulfill the propulsion power demand. This degree of freedom coupled with increased vehicle connectivity, provide opportunities to improve powertrain efficiency along with reduction in pollutant emissions. Powertrain Energy Management (EM) coordinates these different objectives and decides on coordination between the power sources to meet the motive power demand. This paper presents an online optimizing EM controller, which solves the optimal power split among the energy converters. The controller thereby uses acausal data from predicted driving conditions in a receding horizon fashion. The proposed framework can operate with information resulting from vehicle connectivity under autonomous vehicle guidance. In case of driver-aware scenario, it operates using non-deterministic/stochastic predictions resulting from stochastic modeling of driving characteristics using finite Markov Chain. The receding horizon control problem is transcribed as finite dimensional constrained optimization problem, which can be solved using nonlinear optimization methods. Simulation analysis of the proposed approach is used to demonstrate the functionality of the controller.
