Linearized model of the variable flux induction motor drive

The advantages of the squirrel cage three-phase induction motors used in servo positioning applications are severely compromised due to the strong nonlinear coupling between the flux and the speed, which implies sophisticated control algorithms and structures. In the rotor field oriented control only an asymptotic decoupling between flux and speed / position is obtained and only if the flux is maintained at the constant value. By using the input-output linearization by state feedback, the outputs are complete decoupled. Therefore, torque optimization is possible without damaging the adjusted mechanical output variables (speed/position). In order to treats the highly nonlinear regime (working at variable flux) a powerfull tool, nonlinear multiple input - multiple output (MIMO) feedback linearization is used. In order to improve the performances of the induction machine, the obtained linearized mathematical models allow configuration of the control structures similar to the classical ones. In addition, these models can be used for the development of the optimal control of the induction machine based on the energetic criteria in the flux weakening region. The comparative numerical simulation results of the mathematical models (conventional and linearized ones) are presented in this paper. The mathematical models implemented in MATLAB-Simulink are validated during dynamic regime under the appropriate load torque step.