Model Predictive Control for the Thermal System of an Electric Vehicle- Analysis of a Model Predictive Control for WLTP Drive Cycle

Abstract

With the continually increase of the global temperature and the CO2 level in the atmosphere, stringent environmental policies for the automotive industry have been enacted to limit the growth and their consequences. 60.7% of the EU’s transport sector’s CO2 is caused by passenger vehicles, constituting themselves as major contributor; electric vehicles (EVs) could decrease these emissions. Nevertheless, the competition is high and every advantage could be decisive. The range of EVs is critical and improving the energy efficiency of the thermal management could be the key. New methods and control strategies have been developed over the last years and limiting factors like the processors available for vehicles have increased their performance massively. A theory which is common in slow processes like production plants has been redesigned in many academic investigations for the automotive usage and successfully tested. Predictive controllers have proven that they can defacto increase the energy efficiency and thus, increase the driving range of EVs. The objective of this study is the design and analysis of a model predictive control (MPC) against classic control laws. The usage case will be the WLTP drive cycle which has been introduced recently as a more realistic drive cycle. Additionally,the WLTP will be modified by high demanding constant speed scenarios and increased in duration in order to reach thermal critical points. The test environment will provided by an interface model with Simulink and GT-SUITE; the control is based on Simulink whereas a cooling system of a battery-electric vehicle is created in the GT-SUITE. Consequently, the results of the simulation will be of high quality as the GT-SUITE model is established on real-life test data. A method has been developed to connect multiple linear MPCs witht he nonlinear GT-SUITE model. The MPC indicates a very high potential and could reduce the energy consumption by 59.29% fort he modified WLTP, enhancements of this magnitude are not uncommon in research papers. However, tests with real-life applications need to be carried out to confirm these results.