Real-time control system for a permanent magnet synchronous machine

Abstract

In the last decades there has been an always increasing demand for reducing the dependence on fossil fuel. The transportation sector is very crucial since it produces roughly more than one third of the overall pollution. The biggest change that is happening in this sector is the increasing number of electric and hybrid vehicles that circulate on the streets. Considering all these aspects is easy to forecast that electric motors will play an even more important role in the future. In this context, a key role is played by the electronic systems, also known as controllers, devoted to convert the electrical energy from the fixed form available into the specific form that is required by the motors. Nowadays, to achieve gather flexibility more and more systems are replacing analog control with digital control. Anyway, the problem with digital control is the high level of inefficiency in the computational effort, linked to the fact they are very demanding considering the amount of data to be processed and stored. The computational efficiency of the control algorithm is very important because directly affects the efficiency of the whole system. In this context of digital control constantly evolving to achieve the best compromise between flexibility and efficiency is inserted this thesis work. The goal is to develop a new control strategy for the IPM (Internal Permanent Magnets) motor. To achieve the goal the starting point was the implementation of the model for the controller and the motor in Matlab-Simulink environment. The simulation session investigates if the results obtained by the Matlab-Simulink model are in agreement with the theoretically estimated values. In the next phase, the model developed was adapted to work properly in the dSPACE environment, used to control the motor in real-time. Are reported also the tests in the laboratory to measure all the fundamental variables as current, speed, flux, directly on the motor and compare these measures with the results obtained in the simulation step. The results confirm the improvements expected from this new technique. Therefore, its use allows to improve the overall efficiency, it guarantees also an increase in the speed of the control and reduces the number of variables that have to be stored. The achievement of these goals leads also to a reduction of the useless rotor movements linked to delays along the feedback path. As a result, losses are minimized and output energy is maximized.