Setup of a latest generation automotive HIL rig for a BEV controller application

In the context of an increasingly competing automotive market, the possibility to reduce powertrain developing time and reducing cost is of paramount importance. To this extent, virtualization of controller development by means of Hardware-in-the-Loop (HIL) systems can consistently contribute in reducing time to market by frontloading and parallelizing verification and validation of electronic control unit at a fraction of the cost of carrying it out on physical vehicle prototypes. For these reasons, Powertech Engineering (PWT) is investing resources in developing methodologies and tools for HIL testing of state-of-the art automotive Electronic Control Units. In this framework, this thesis aim is to setup the first HiL application for the Verification and Validation of an Electric Vehicle Control Unit (EVCU) on the first full-HIL test bench in PWT. The work was mainly divided into two work-packages: in the first one (referred to as Dyno-mode) only the vehicle’s powertrain has been simulated on the HIL, making the EVCU operating as on a powertrain test bench; In the second (referred as Vehicle mode) the powertrain model was co-simulating with a vehicle model and a rest-of-the-bus emulation model to make the EVCU operating as in the real vehicle. For the Dyno-mode, the electric powertrain model was deployed on the FPGA board of the PWT’s HIL system and all the needed I/O (resolver, voltage and current probes, temperature sensor, etc) was configured to read and generate the correct control signals to the EVCU to close the loop. Then, it was possible to analyse and compare the EVCU control parameters estimates regarding motor torque, direct current (id) and quadrature current (iq) with the actual results of the HIL plant model on steady state operating points. As the first work-package results have been considered more than acceptable, the activity moved to the second, integrating the vehicle model in order to simulate driving cycles. To do this, it was necessary to continue the implementation of the wiring in order to transmit all the signals required by the EVCU in this condition. For this purpose, it was necessary to configure also few CAN lines to emulate the rest-of-the bus signals that the EVCU would have exchanged with other control nodes in the real vehicles. To make the EVCU operate as in the real vehicle, all the interactions between the powertrain and the rest of the vehicle components must be modelled. To this extent, an already-available full vehicle model (including vehicle dynamics, brakes, thermal circuits, etc) was integrated with the powertrain one. The validation was carried out on three driving cycles: the city part of WLTP, the UDDS and the NYCC. These driving cycles were used to verify: how the EVCU is able to satisfy the speed request; the control parameters were aligned in transient and the EVCU was able to operate in real driving conditions without errors. For a current configuration problem of the HIL it wasn’t possible to implement the LIN and this meant that for safety reasons the EVCU limited the speed to 70km/h. Despite this, the results obtained were very consistent with what was expected. As future approaches, other nodes could be integrated into the loop to have an even closer setup with respect to the real vehicle, as well as completing the HIL configuration by adding the LIN.