Advanced electrical and thermal models for battery electric vehicles multi-step design

This thesis presents two vehicle models developed for a Battery Electric Vehicles dimensioning tool based on a multi-step design approach. The final goal is to identify the main parameters of the powertrain components considering the mechanical, electrical and thermal constraints that limit the performance of the system, iteratively refining the analysis based on the outcomes of a multi-step simulation process. The work started on two pre-existing vehicle models implemented in MATLAB/Simulink environment. The first of them, called Level 0, estimates the performance of the vehicle based on the e-motor maximum power, speed and torque. Level 0 makes it is possible to individuate the range of acceptable e-motor macro parameters to meet the target performance. The other model, called Level 1, can be used to discriminate among different e-motors that are suitable for the application. It refines the vehicle performance simulation by introducing the actual e-motor steady-state and transient characteristic and its efficiency map. The major contribution of the thesis is the development of two additional models for the multi-step dimensioning tool, called Level 2 and Level 3 respectively. An additional contribution is the enlargement of the pre-existing components database of the tool by introducing new ones. Level 2 introduces the electrical models of the e-motor and battery pack, and simulates the behaviour of the inverter. At each time instant of the simulation, the vehicle controller generates the accelerator and brake pedal commands based on the vehicle speed error, which are converted into a torque reference signal that is provided as an input to the inverter subsystem. Based on two look-up tables calculated offline, the current references to operate the e-motor at MTPA are generated, weakening the flux if needed. Then, the corresponding inverter AC voltages and DC power request are calculated by considering the limitations imposed by the battery dynamics and the DC voltage available at its terminals. The 5-parameters e-motor model calculates the e-motor torque and finally the laws of longitudinal vehicle dynamics are used to update the vehicle speed. Since Level 2 calculates voltages and currents at any point in the powertrain, it allows to spot the presence of bottlenecks that limit the performance of the vehicle and to take measures against them. Level 3 is built on top of Level 2 and simulates the thermal behaviour of the system. It exploits the data generated by Level 2 at each timestep to estimate the heat generated by power dissipation in each of the powertrain macro-components. The temperature signal are then estimated for each part by means of a dedicated thermal balance equation, accounting for both heat generation phenomena and heat exchange between the powertrain components. Since the trend of temperatures in the powertrain is estimated, it is possible to have an insight on the vehicle cooling requirements and to assess the effects of non-optimal temperature levels on vehicle performance and integrity.