Speed Measurement Error Suppression for PMSM Control System Using Self-Adaption Kalman Observer

This paper proposes a self-adaption Kalman observer (SAKO) used in a permanent-magnet synchronous motor (PMSM) servo system. The proposed SAKO can make up measurement noise of the absolute encoder with limited resolution ratio and avoid differentiating process and filter delay of the traditional speed measuring methods. To be different from the traditional Kalman observer, the proposed observer updates the gain matrix by calculating the measurement noise at the current time. The variable gain matrix is used to estimate and correct the observed position, speed, and load torque to solve the problem that the motor speed calculated by the traditional methods is prone to large speed error and time delay when PMSM runs at low speeds. The state variables observed by the proposed observer are used as the speed feedback signals and compensation signal of the load torque disturbance in PMSM servo system. The simulations and experiments prove that the SAKO can observe speed and load torque precisely and timely and that the feedforward and feedback control system of PMSM can improve the speed tracking ability.