Observer-based deadbeat predictive speed controller for surface-mounted PM synchronous motor.

Area of Science:

• Electrical Engineering
• Control Systems
• Power Electronics

Background:

• Conventional deadbeat controllers are typically limited to inner-loop current control in cascaded structures.
• Surface-mounted permanent magnet synchronous motor (SPMSM) drives require advanced control strategies for improved performance and robustness.

Purpose of the Study:

• To develop and evaluate an observer-based deadbeat predictive speed controller (DBPSC) for SPMSM drives.
• To enhance speed control accuracy, reduce steady-state error (SSE), and minimize total harmonic distortion (THD).
• To improve robustness against parameter variations and load torque disturbances.

Main Methods:

• Design of a deadbeat speed control law with stabilizing feedback and compensation terms.
• Integration of a deadbeat load torque observer (DBLTO) for accurate load torque estimation.
• Application of Lyapunov theory for stability analysis of the observer-based system.
• Implementation and validation using MATLAB/Simulink simulations and experimental tests on a prototype SPMSM drive with a TI TMS320F28335 DSP.

Main Results:

• The proposed DBPSC achieves fast transient response and ensures closed-loop system stability.
• The deadbeat load torque observer (DBLTO) enables precise load torque estimation, leading to reduced SSE and THD.
• The controller demonstrates robustness against parameter variations.
• The simplified structure reduces computational burden for digital implementation.

Conclusions:

• The observer-based DBPSC offers superior performance compared to conventional deadbeat speed controllers (DBSC) for SPMSM drives.
• The proposed method effectively handles model nonlinearities, parameter uncertainties, and load torque disturbances.
• The controller is suitable for practical digital implementation due to its simple structure and reduced computational load.