Very-low speed control of PMSM based on EKF estimation with closed loop optimized parameters

Area of Science:

• Electrical Engineering
• Control Systems

Background:

• Sensor precision limits real-time speed calculation for permanent magnet synchronous motors (PMSM).
• This limitation leads to crawling and jitter issues at very-low speeds.

Purpose of the Study:

• To develop an accurate and real-time speed estimation method for PMSM at low speeds.
• To address the challenges in tuning extended Kalman filter (EKF) parameters for speed control.

Main Methods:

• An extended Kalman filter (EKF) designed in dq-coordinate using position sensor data was employed.
• A closed-loop optimal (CLO) method was devised to optimize EKF measurement and system covariance matrices based on speed step response.

Main Results:

• The proposed EKF method significantly improved speed estimation accuracy at low speeds.
• The CLO method effectively optimized EKF parameters, overcoming tuning difficulties.
• Simulation and experimental results demonstrated the elimination of low-speed jitter.

Conclusions:

• The EKF with the CLO method provides an effective solution for accurate and jitter-free low-speed operation of PMSM.
• This approach enhances the performance and reliability of PMSM control systems.