Control Strategy for Hybrid Electric Vehicle Based on Online Driving Pattern Classification

Hybrid Electric Vehicles (HEVs) are gaining popularity these days mainly due to their high fuel economy. While conventional HEV controllers can be classified into rule-based control and optimization-based control, most of the production vehicles employ rule-based control due to their reliability. However, once the rule is optimized for a given driving pattern, it is not necessarily optimal for other driving patterns. In order to further improve fuel economy for HEVs, this article investigates the feasibility of optimizing control algorithm for different driving patterns so that the vehicle maintains a high level of optimality regardless of the driving patterns. For this purpose, a two-level supervisory control algorithm is developed where the top-level algorithm classifies the current driving pattern to select optimal control parameters, and the lower level algorithm controls the vehicle power flow using the selected control parameters in a similar way to conventional supervisory controllers. To study the effectiveness of the proposed algorithm, a HEV model with a rule-based control algorithm is modified such that the control parameters are optimized for different driving patterns, and the simulation results obtained by the driving pattern-optimized control algorithm is compared with those with the original control algorithm with fixed control parameters.