In the field weakening region of induction motors, it is essential to consider not only voltage and current constraints but also to ensure smooth transitions across different field weakening regions. Existing field weakening strategies typically rely on the steady-state equations of the induction motor to compute the stator flux command and torque limits for constant power and constant voltage regions. These approaches often have poor dynamic performance and require switching between different torque limit values, which can make control implementation more complex. This paper proposes a approach grounded in model predictive control, which dynamically calculates torque reference limit values by monitoring the stator flux, rotor flux, and stator current, thereby obviating the necessity for a current controller to adhere to maximum current constraints. This method achieves seamless transitions between constant torque, constant power region, and constant voltage region, allowing the induction motor to maintain maximum load-carrying capability in the field weakening region and improve operational efficiency. Furthermore, an improved discretization method for the full-order observer is introduced, which significantly reduces discretization errors, particularly at low carrier ratios. Both simulation and experimental results are presented to confirm the effectiveness of the proposed method.

predictive flux control,induction motor,Flux observer