Real-time Power Distribution Strategy of Hybrid Tracked Vehicles Based on Nonlinear Model Predictive Control

A real-time power distribution strategy based on model predictive control is proposed for the difficulty of coordinated control of multiple power sources and poor multi-objective optimization of hybrid tracked vehicles. By analyzing the structure and characteristics of series hybrid tracked vehicles, the vehicle is modeled based under MATLAB/Simulink. A rule-based power tracking method of multi-speed switching of the engine is adopted to improve fuel economy in the pursuit of stable engine operation. Due to the difficulty of multi-objective optimization and lack of forward-looking in the rule-based strategy, a power distribution strategy based on nonlinear model predictive control is developed. In order to solve the nonlinear constraint optimization problem, two methods are proposed. Method 1: Through the Taylor expansion, the linear time-varying state space function is obtained, which is transformed into the linear time-varying model predictive control (LVT-MPC). Method 2: By considering MPC as a nonlinear programming problem, sequential quadratic programming (SQP) is used to obtain the descent directions of control variables and iteratively refine the results of the linear search. Simulation results demonstrate the fuel economy of SQP-MPC is improved by 11.39% and the fuel economy of LVT-MPC is improved by 10.85%, but LVT-MPC has advantages in real vehicle applications.