1-D model for drivetrain losses evaluation in a BEV vehicle

AbstractIn the last years, the world is witnessing to a radical change in the development of the power, especially for what concerns industrial applications and, as a subsystem of these, the automotive one. The requests to reduce CO2 emissions due to the global warming, in part attributable to the usage of fossil fuels, are the main reasons for this improvement. Lower CO2 means reduction of consumptions and decreasing of environmental problems, and these aspects can be achieved working on the losses and pointing out alternative and green energy sources. Both these considerations lead to two challenging points that are at the base of contemporary industrial strategies: • to search for other sources of energy capable of helping and/or replacing fossil fuels;• to try to increase the efficiency of all the elements inside every mechanism.The thesis project here presented considers both these two points, focusing on the drivetrain, one of the most critical parts in terms of losses inside a vehicle. Therefore, the transmission of a BEV (battery electric vehicle) is here considered, pointing out a future concept where no fossil fuels are used to propel conventional cars.Inside this environment, the task of the work is to build a 1-D model of the drivetrain, and evaluate the efficiency of the elements involved, to understand how the losses are categorised and caused, to be able to look at the way they change towards speed, torque, temperature and finally reduce them, keeping an optimal operation of the transmission.The main objects taken into account are gears and bearings, divided in load dependent and load independent losses, faced with different approaches found in literature and used in the history of the specific element investigated. Different programs are available on the market to evaluate these losses: here inside the user can find some of the approaches abovementioned, normally thought to match general casuistries.Because of this need, some software can be unprecise, or not perfectly tuned for the object the user is analysing, and become important to understand alternative theories, another task of this thesis.The software used and so the reference for the reasonings, is LMS AMEsim, used in Toyota Motor Europe for different parts of the vehicle, and specifically for transmission patterns.On the other hand, the alternative approaches are evaluated through Matlab scripts, later sent into the AMEsim model for the efficiency calculation and finally compared with internal AMEsim approaches through Excel files, that are the containers and comparators of the project.Last aim of the thesis, conceived and written between Politecnico di Torino and Toyota Motor Europe, is to make converge the academic and industrial worlds as much as possible, in an activity that must be useful for both environments.The 1-D model is for this reason based on a mass production electric passenger car, object of Toyota benchmark interests, and so with specific components of this particular typology of vehicle. Nevertheless, the architecture adopted is suitable for most of the BEVs on the market, since nowadays the same typology of a single-speed and double-stepped layshaft transmission can be found. Together with these industrial conditions, the steps of the work have been set according to Politecnico standards in order to reach the theoretical depth required.An application of the model on a WLTP driving cycle has been run as last part of the experience, to show the model is finally useful for real simulations.