20 November 2014 in IEEE/ASME Transactions on Mechatronics
This paper presents a novel disturbance compensation scheme to attenuate periodic disturbances on repetitive motion using permanent magnet linear synchronous motors (PMLSMs), and this scheme is called the periodical adaptive disturbance observer. The scheme is based on assumptions that all measured states and disturbances are periodic and repetitive when the tasks executed by PMLSM motion systems have periodic and repetitive characteristics. In the proposed control scheme, a lumped disturbance is estimated by the classical linear disturbance observer (DOB) for the initial time period and stored in memory storages. It consists of parametric errors multiplied by states, friction force, and force ripple, and then, it is updated for each time period by the periodic adaptation law. This scheme requires no mathematical models of disturbances and adaptation laws of model parameters such as the mass of the mover and viscous friction coefficient. Also, it is possible to compensate for disturbances above as well as below the bandwidth of the Q-filter (LPF) of DOB. The effectiveness of the proposed control scheme is verified by various experiments that take into account varying frequency components of disturbances along the operating speed of a mover of PMLSM such as force ripple and friction force.
18 July 2016 in IEEE Transactions on Industrial Electronics
This paper proposes a new scaleless position estimation method. From the industrial perspective, the Hall sensor has several advantages: tiny size, light weight, extremely low cost, and insensitivity to environmental contamination and external disturbance. Compared to the square wave Hall sensor, the linear Hall sensor provides more detailed information along the position. However, for combining more than two linear Hall sensors, the sensor offset, the difference in scale, and the unwanted phase shift between them and the harmonics would decrease the validity and reliability of sensor measurement. The fast Fourier transform and the fixed point iteration method are applied to compensate for those issues without any low-pass filter or additional signal conditioning. The effectiveness of the proposed scheme is verified through a prototype permanent magnet linear synchronous motor system.
27 July 2016 in International Journal of Control, Automation, and Systems