HEV modelling in GT-Suite environment and control strategy design using Dynamic Programming and ECMS

Today’s automotive industry is moving toward electrification to comply with the increasingly stringent emission limitations but also to succeed in a market more fuel saving-oriented, in a scenario like the depicted one, hybrid electric vehicles (HEVs) represent a suitable solution to satisfy all these new needs. HEVs allow to exploit benefits of electric vehicles without losing the advantages of conventional internal combustion engine ones, but electrification introduces a major level of complexity in first step design and calibration of the hybrid systems. Numerical simulation becomes then fundamental for designers to be able to estimate which could be the best solution, avoiding time-demanding and expensive experiments; the aim of this thesis work was the design and analysis of vehicle models for simulation, both backward and forward. The work was divided mainly into two parts, in the first one a backward simulation campaign was carried out using an Optimal design tool, previously developed by other colleagues, based on Matlab code which run over a selected driving cycle different hybrid electric vehicles trying different configurations in terms of geometry and components size; data collected has been then postprocessed and a ranking in terms of emissions and fuel consumption was done for each configuration. The second step was the creation of vehicle models in GT-Drive environment to run forward dynamic simulations so to include the variability brought by the presence of a driver and obtain results more in line with real driving conditions. The design process followed successive steps, at first a conventional ICE-based vehicle model was designed, moving then to a battery electric vehicle (BEV) model and finally to a P2 parallel hybrid electric vehicle. A last step, carried out in collaboration with Michele Giulio, was the programming in Matlab of a ECMS (Equivalent consumption minimization strategy) controller for the hybrid electric vehicle that allows a real time control of the torque split between thermal and electrical propellers so to achieve a charge-sustaining based control policy The models designed have been used to run dynamic simulations over the same driving cycle already used in the backward simulations, results in terms of fuel consumption or battery charge consumption have been collected and analysed. Moreover a comparison between the ICE-based and BEV models designed in GT-Drive and homologous models designed in Simulink/Simscape environment by Michele Giulio in his thesis work has been done to check the validity of the models, achieving almost converging results of fuel consumption or battery state of charge for BEV.