Optimization of Supervisory Control Strategy for Parallel Hybrid Vehicle with Provisional Load Estimate

The paper focuses on the simulation, analysis and control of the energy flows in a parallel hybrid electric vehicle (HEV). HEVs operation is concerned with the on board conversion of chemical, electrical and mechanical energy and its optimal control is essential in order to increase the global system efficiency. A dynamic model is used to describe the driver-vehicle interaction for a generic transient and to simulate the vehicle driveline, the internal combustion engine (ICE) and the electric motor/generator (EM). An estimate of future vehicle load is performed with a neural network to optimize the supervisory control strategy during the estimated future time window. A description of the whole model is presented and the results obtained from the simulation of for a real driving cycle are reported.