Design and implementation of different control strategies for an induction motor

This thesis presents a comprehensive approach to the modeling, simulation, and real-time control of a three-phase squirrel-cage induction motor. To achieve stable dynamic performance, an accurate numerical model is developed using the finite element software FEMM 4.2, from which the operational parameters required for the control design are extracted. Subsequently, two distinct control strategies are developed in MATLAB & Simulink: a scalar Volt/Hertz (V/Hz) control and a vector Field-Oriented Control (FOC). During this phase, the implementation of a Luenberger observer for speed and disturbance torque estimation is thoroughly analysed. After simulation validation, the algorithms are automatically converted into C-code and deployed onto a dSPACE MicroLabBox platform, which, alongside ControlDesk, enables the real-time execution of the control loops on the physical motor. The tracking capabilities are successfully verified across both simulation and experimental tests. A direct comparison confirms that the FOC ensures superior transient performance and disturbance rejection, whereas the V/Hz strategy guarantees a lower steady-state speed ripple. Finally, the main limitations of this study lie in the non-modelled physical phenomena, such as inverter dead-times, load torque ripple, and temperature variations, which cause slight deviations between the simulated environment and the real test-bench performance.