Design of optimal rule-based controller for plug-in series hybrid electric vehicle

Abstract

Energy consumption of Hybrid Electric Vehicles (HEV) strongly depends on the adopted energy management strategy (EMS). Rule-Based (RB) controllers are the most commonly used for their ability of integration in real-time applications. Unlike global optimization routines, RB controllers do not ensure optimal energy savings. This study presents a methodology to design a close-to-optimal RB controller derived from global optimization strategies. First, dynamic programming (DP) optimization is used to derive the optimal behaviour of the powertrain components on the Worldwide Harmonized Light Vehicles Test Cycle (WLTC), and then, the resulting performance of the powertrain components is used to design an optimized RB energy management strategy. Furthermore, the strategy is developed to cope with the variations in trip length and traffic conditions. The plug-in series hybrid electric vehicle is modelled using the energetic macroscopic representation (EMR). Results show that the proposed optimal RB controller is only consuming 1-2% more fuel compared to DP controllers and is resulting in a 13-16% less fuel consumption compared to basic RB controllers.