Abstract: The modern automatic loading system for large caliber guns involves the collaborative control of multiple motors. The application of sensorless technology can significantly improve the reliability of driving systems. The rotor position identification of surface-mounted permanent-magnet synchronous motor at zero-/low-speeds is a difficult problem in sensorless control. To deal with it, an extended Kalman filter based on mechanical motion model is proposed to assist rotor position estimation. At the same time, the observer disturbance rejection technology is employed to eliminate the influence of disturbances such as model parameter uncertainty to rotor position identification, and to realize reliable start-stop control under zero/low speeds and heavy loads. The proposed method overcomes the limitation of the conventional rotor position estimation method based on the electrical model, greatly improves the load capacity of the sensorless technology at zero/low speeds, and realizes the performance comparable to that of the position sensor servo. This sensorless motor control technology is applied to the 155 mm gun loading system. Finally, the technology is verified by the operation simulation of the modular explosive loader and projectile loader as examples.

Key words: gun, automaticloadingsystem, sensorlesscontrol, motorcontrol, Kalmanfilter