Rotary Position Sensors Comparative study of different rotary position sensors for electrical machines used in an hybrid electric vehicle application

Abstract: Today, many projects about Electric Vehicles (EVs) and Hybrid Electric Vehicles (HEVs) are in progress within the automotive industry. Fuel-efficiency and reduction of carbon dioxide emissions from vehicles are the main targets. This thesis is within in one of these projects that is called electric All Wheel Drive(eAWD) at BorgWarner TorqTransfer Systems AB. A key parameter to perform an accurate and efficient control of an electric machine is the position sensor. The sensor measures the angular position of the rotor shaft and there are several ways and techniques to do this. This thesis aims to compare different common position sensors and identify ”new” sensor techniques by performing a literature study, model and simulate sensors and test an electric machine with different sensors implemented. Various enhancement methods to improve the position information and prediction are also evaluated. The electric motor prototype used in the eAWD project has different position sensors implemented and these are simulated in Matlab/Simulink together with the system model of the electric machine and control system. Tests are also performed and compared to the simulation results. The results show on best performance when using the resolver as position sensor. The Hall-effect sensor can be improved with an observer, but the observer is not suitable for this specific type of Torque Vectoring (TV) application. The Hall-effect sensor has a speed dependent torque ripple that leads to harmonics at frequencies that relates to the speed of the unit which may causes problems, such as mechanical resonances in the system. There are several ”new” sensor techniques based on the theory of eddy-currents that may be of interest since they are said to be more optimized for EV and HEV applications.