Model-based ethanol blend estimation in flexible-fuel vehicles

This paper presents a model-based strategy estimating the ethanol content of an ethanol-gasoline blended fuel in flexible fuel vehicles. A steady-state parametric model relating engine speed, throttle angle, and air-fuel ratio to the fuel injector pulse-width is developed from physics. The parameters of this model are adapted and linked to percentage of ethanol content via a suitably defined metric. The proposed steady-state model structure is experimentally validated on a 2005 5.4L V8 Ford engine. The developed ethanol content estimation methodology is justified based on the combustion chemistry and physics involved. The methodology developed has a distinct advantage over previously proposed methods as it uses only the existing sensor set on a production vehicle.