Abstract
Extended Kalman filter algorithm for direct thrust force sensorless control of traditional permanent magnet synchronous motor has time delay, which affects the estimation accuracy. To solve this problem, an improved Extended Kalman Filter algorithm is proposed, which is connected by two Extended Kalman Filters in parallel. The Taylor first-order linearization of the system state transition matrix is carried out with the estimated value of the current time to solve the delay characteristic of the traditional Extended Kalman Filter. Firstly, the mathematical model of PMLSM in two-phase static coordinate system is established, and the state equation of traditional Extended Kalman Filter is established and linearized. Then, the Dual Extended Kalman Filter algorithm is designed. The estimated state variables in EKFI are taken as the input values of EKFII state variables, and the obtained EKFII state variables are input into EKFI to calculate the Taylor first-order linear approximation of the current EKFI state transition matrix. Finally, the improved algorithm is applied to PMLSM direct thrust force sensorless control. The experimental results show that the dual Extended Kalman Filter has better estimation accuracy than the traditional Extended Kalman Filter. (Abstract)
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have