Design of Speed Sensorless Control of Induction Motor Based on Dual-Nonlinear Control Technique

International audienceThis paper deals with performance improvement of direct flux and torque control of induction motor. The proposed algorithm consists of the combination of tow nonlinear control approaches. A decoupled control design is done by the exact feedback linearization control. Since, wastes the control stability and robustness while the presence of disturbance and modeling inaccuracy, it is recommended to be associated with a robust control approach like second-order sliding mode control (SOSMC). Therefore, the super twisting algorithm is integrated as auxiliary inputs to the feedback linearization control law to achieve robust feedback linearization control. On the other hand, the high-performance control design requires accurate knowledge of different control variables such as stator flux and rotor speed. instead of using costly and fragile sensors that may increase the volume and decrease the reliability of the control system, a proposed sliding super twisting observer and model reference adaptive system serves as sensorless algorithms for rotor speed and flux estimation in wide speed region. This conjunction is intended to enhance the overall control performances and speed/flux estimation, especially at low-speed operations. An experimental study has been done using MATLAB/Simulink with dSpace 1104 real-time interface for investigating the performance of the proposed algorithms