Control design for high performance propulsion drive for electric vehicles
Global warming, limited reserves, vastly fluctuating prices, and emissions of greenhouse gas are the major problems regarding fossil fuels. One of the major consumers of fossil fuels is transportation sector and this inevitable consumption can be reduced by the electrification of vehicles, which results in the production of zero emission and eco-friendly electric vehicles. The most significant component in the powertrain of an electric vehicle is the traction machine drive which is solely responsible for converting the battery electrical energy into the vehicle mechanical energy in the form of traction force with appropriate torque and speed.
Among the available traction machines, the induction machine shows a promising performance in the automotive sector particularly in the electric and hybrid electric vehicles. This is mainly because of its wide speed operation, higher power density, greater starting torque, less maintenance, and de-excitation for inverter fault. Although, variations in the operating as well as ambient temperature cause unexpected uncertainties in induction machine parameters (stator and rotor resistance), which results in significant degradation in electric vehicle’s speed performance capability. This thermally degraded speed cannot be controlled by traditional control techniques. This problem poses the necessity of implementing a robust closed-loop control technique to enhance the dynamic performance of the electric vehicle. This thesis presents three innovative contributions towards the design and implementation of robust linear parameter varying (LPV) observer and controller schemes to enhance the speed performance of electrified powertrain of electric vehicle.
The first contribution of this research work is the design and implementation of a robust LPV observer to estimate the thermally degraded speed of an induction motor drive as well as an electric vehicle that uses the same propulsion drive at elevated temperatures during standard driving cycle. The proposed observer performance is compared with conventional sensorless field-oriented control (FOC) and sliding mode observer at elevated temperatures under various transient conditions to verify its better performance.
The second contribution of this dissertation is the development of a robust LPV control scheme to address the speed degradation problem at elevated temperatures in an electrified powertrain. The performance of the proposed control technique is compared with that of conventional FOC, sliding mode control (SMC) and higher order SMC to validate its efficacy and advantages. The robustness of the proposed control technique is investigated for an electric vehicle operation against the Worldwide Harmonised Light Vehicles Test Procedure Class 3 driving cycle through simulations as well as experiments.
The third contribution of this manuscript is the design and development of an optimized energy control (OEC) scheme using LPV control. The weighting functions of the control scheme are optimized by using Genetic Algorithms. The scheme addresses the two conflicting objectives which substantially affects the working of traction machine drive of an EV powertrain. These objectives are maximizing its speed performance and minimizing its energy consumption. The analysis of OEC scheme is conducted on the developed vehicle simulator as well as on an induction machine drive platform. The accuracy of the proposed OEC is verified against the New European Driving Cycle and Highway Fuel Economy Test driving cycle.
This research aims to design and implement a robust control and optimization technique to address the challenging issues in the speed of the propulsion drive (induction motor drive) used in electric vehicle application. These issues are the thermally degraded speed estimation and control of electric vehicle drive speed as well as the conflicting objectives of enhanced speed performance and reduced drive energy consumption. The control scheme is based on the linear parameter varying control, the weighting gains of which are optimized by using the Genetic Algorithms. The performance analysis of the proposed scheme is conducted through simulations and experiments.
