Adaptive Pontryagin's Minimum Principle supervisory controller design for the plug-in hybrid GM Chevrolet Volt

This paper presents an adaptive supervisory controller, based on Pontryagin's Minimum Principle (PMP), for on-line energy management optimization of a plug-in hybrid electric vehicle. Using minimum driving information, such as the total trip length and the average cycle speed, the proposed algorithm relies on adaptation of the control parameter from state of charge feedback. The proposed strategy is referred in the paper to as Adaptive-PMP (A-PMP). The new controller is applied to a detailed forward vehicle simulator of the plug-in hybrid Chevrolet Volt manufactured by General Motors, where an experimentally validated LG Chem battery model is used. The strategy we propose aims at achieving a blended trajectory of the state of charge to minimize the consumed fuel, resulting in an overall better performance than the actual Charge Depleting/Charge Sustaining (CD/CS) strategy currently used on-board of the vehicle. A comparative analysis of three strategies, i.e., the optimal one (PMP), the proposed one (A-PMP) and the in-vehicle one (CD/CS), is conducted in simulation which shows that improvement above 20% in fuel consumption may be achieved when the proposed algorithm is used instead of the current on-board strategy.