Segmented ternary composite control method considering time delay for high-speed and high-precision linear motor

The conventional ternary composite control method utilized in linear motor control offers the advantages of high positioning accuracy and high dynamic response. However, its performance is constrained due to time delays in high-speed and high-precision servo systems. To tackle this problem, a segmented ternary composite control method considering time delay is developed in this work. Firstly, a nominal model is established with the consideration of sampling delay, current control loop delay, etc. A robust feedback controller is designed using the nominal model to achieve a high closed-loop bandwidth. Besides, a novel iterative feedforward tuning algorithm based on feedback control signal decomposition (IFFT-FCSD) is proposed to attain high dynamic performance. The algorithm decomposes the feedback control signal into several basis signals using the least squares algorithm. More precise tracking accuracy is obtained after compensating for the time delay between the basis signals and the feedback control signal. Furthermore, to increase the bandwidth of the disturbance observer (DOB), the observed control signal is compensated with the equivalent time delay of the nominal model. Finally, the segmented control strategy is implemented. The motion-profile-based feedforward controller is only engaged during the motion tracking period, and the DOB is employed exclusively during the settling period to attain precise positioning while minimizing the negative impact on dynamic performance. Comparative experiments conducted on a linear motor validate the effectiveness of the proposed method. The maximum tracking error is 1.28μm and the positioning error converges to within ±450nm in just 9.3 ms during the settling period for a point-to-point motion with a maximum speed of 0.55 m/s, a maximum acceleration of 150 m/s2, and a stroke of 4 mm.