Sensorless Speed Estimation of Permanent Magnet Synchronous Motor

Xiao He,Yinghong Zhao
DOI: https://doi.org/10.1109/icmc.2014.7231731
2014-01-01
Abstract:A new approach to sensorless speed estimation is derived in the unbiased minimum variance sense by applying the Unknown Input Kalman Filter (UIKF) in the networked environment. The recursive Networked Unknown Sensor Input Extend Kalman Filter (NUSIEKF) is developed to deal with the sensorless speed estimation problem for permanent magnet synchronous motor (PMSM) in the existence of packet dropout and unknown sensor inputs. The PMSM considered is not equipped with a position sensor, but the velocity information should be obtained from the on-line estimation from the measurements of voltages and currents of the motor. The measurements from sensors are transmitted via a shared communication network with limited transmission capacity. Packet dropout and unknown sensor inputs are considered in the filter design process. Simulations of the sensorless speed estimation have been carried out on a PMSM driving system to test the feasibility and effectiveness of the proposed filtering techniques.
What problem does this paper attempt to address?