Supervisory Control of Parallel Hybrid Electric Vehicles for Fuel and Emission Reduction

Abstract

Past research on hybrid electric vehicles (HEVs) focused primarily on improving their fuel economy. Emission reduction is another important performance attribute that needs to be addressed. When emissions are considered for hybrid vehicles with a gasoline engine, horizon-based optimization methodologies should be used because the light-off of the three-way catalytic converter heavily depends on the warming-up of catalyst temperature. In this paper, we propose a systematic design method for a cold-start supervisory control algorithm based on the dynamic programming (DP) methodology. First, a system-level parallel HEV model is developed to efficiently predict tailpipe emissions as well as fuel economy. The optimal control problem for minimization of cold-start emissions and fuel consumption is then solved via DP. Since DP solution cannot be directly implemented as a real-time controller, more useful control strategies are extracted from DP solutions over the entire state space via the comprehensive extraction method. The extracted DP results indicate that the engine on/off, gear-shift, and power-split strategies must be properly adjusted to achieve fast catalyst warm-up and low cold-start tailpipe emissions. Based on DP results, we proposed a rule-based control algorithm that is easy to implement and adjust while achieving near-optimal fuel economy and emission performance.