Abstract
Abstract:
Hybrid Electric Vehicles (HEVs) improvements in fuel economy and emissions strongly depend on the energy management strategy. The global solution to the HEV control problem can be found with Dynamic Programming if the driving cycle is known a priori. It is shown that a local approach gives similar results if a pair of parameters related to the cost of using the electric motor is determined. It is also shown how the optimization on the whole cycle can be reduced to an optimization over shorter missions composed of past and predicted data. Since the value of the parameters is deeply connected to the nature of the driving cycle, the paper presents an adaptive algorithm to determine the best choice of the pair according to the current driving conditions. Initial simulation results show that better fuel economy can be achieved compared to instantaneous minimization techniques. These results are now being implemented on the Ohio State University BuckHybrid 2004, a hybrid-electric version of a production Ford Explorer SUV developed as part of the Ford-U.S. DOE FutureTruck student competition. Experimental results will be presented at the conference.