Hits:

Indexed by:期刊论文

Date of Publication:2017-01-01

Journal:ENERGIES

Included Journals:SCIE

Volume:10

Issue:1

ISSN No.:1996-1073

Key Words:driving intention; mixed logic dynamical model; mixed integer linear programming (MILP); model predictive control (MPC); nonlinear auto-regressive (NAR) neural network

Abstract:Plug-in hybrid electric vehicles (PHEVs) can be considered as a hybrid system (HS) which includes the continuous state variable, discrete event, and operation constraint. Thus, a model predictive control (MPC) strategy for PHEVs based on the mixed logical dynamical (MLD) model and short-term vehicle speed prediction is proposed in this paper. Firstly, the mathematical model of the controlled PHEV is set-up to evaluate the energy consumption using the linearized models of core power components. Then, based on the recognition of driving intention and the past vehicle speed data, a nonlinear auto-regressive (NAR) neural network structure is designed to predict the vehicle speed for known driving profiles of city buses and the predicted vehicle speed is used to calculate the total required torque. Next, a MLD model is established with appropriate constraints for six possible driving modes. By solving the objective function with the Mixed Integer Linear Programming (MILP) algorithm, the optimal motor torque and the corresponding driving mode sequence within the speed prediction horizon can be obtained. Finally, the proposed energy control strategy shows substantial improvement in fuel economy in the simulation results.