Development of a predictive thermal management function for Plug-in Hybrid Electric Vehicles

The present thesis is focused on the development of a predictive control strategy oriented to battery thermal management for plug-in hybrid electric vehicles (PHEVs). The basic principle of the strategy is to reduce as much as possible battery energy usage related to power request from the respective cooling circuit actuators. At this end, a thermo-hydraulic model of the in-vehicle battery cooling circuit has been developed in AMESim environment. Then, it has been implemented in an already existing Simulink vehicle model, which includes components analytical models and control strategies. The predictive aspect of the novel strategy is related to the evaluation of battery temperature over the electronic horizon on the base of input signals such as vehicle speed and road slope profile. As a consequence of temperature prediction, the developed strategy is able to establish in an energy-efficient way if cooling power is either required or not. Results highlight the advantages of applying the predictive strategy instead of a rule-based one, which is on-board implemented in each vehicle. It is shown that major energetic benefits, related to the extension of the all-electric range and the reduction of fuel consumption, take place at middle environmental temperatures, at which battery cooling power request can seriously make the difference on its drain rate. Therefore, project goal has been reached and the results can be considered an interesting starting point for further development and enhancing of predictive control strategies.