Varying-Domain Optimal Management Strategy for Parallel Hybrid Electric Vehicles

In this paper, a management strategy is developed to realize the real-time optimal torque distribution between the internal combustion (IC) engine and the electric motor of parallel hybrid electric vehicles (HEVs). Without depending on future information, a set of instantaneous cost functions is defined as the objective of a multiobjective problem, which includes vehicle energy consumption, selected emission species, and an evaluation criterion for the battery state of charge (SOC). The varying-domain method is then utilized to introduce a flexible priority among objectives and to transform the multiobjective problem into a nonlinear programming problem, the optimal solution of which is subsequently found by a genetic algorithm, i.e., GENOCOPIII. A comparison of the simulation results demonstrates the flexibility of the proposed varying-domain optimal management strategy (VOMS) under different driving conditions. Compared with the rule-based management strategy (RBMS) and the weighted sum management strategy (WSMS), the VOMS potentially improves the fuel economy, emission reduction, and stability of the SOC.