Improved Model Predictive Torque Control For Three-phase Asynchronous Motor With Weight Factor Elimination

In the field of electrical drives,asynchronous motors have obvious advantage of high performance-price ratio,and they are popular in the market.Model predictive control has the advantages of principle's simplicity,convenient modeling and easy handling of non-linear constraints,it is widely used in the field of motor control.In this paper,model predictive control of asynchronous motor is studied.Firstly,the coordinate transformation theory of three-phase AC system and the mathematical models of asynchronous motors in different coordinates are described.The traditional model predictive torque control strategy and the no-weight-factor model predictive torque control scheme are introduced.Aiming at the difficulty of weight coefficient's selection in model predictive torque control of induction motors,This paper studies a control strategy of model predictive torque control based on virtual torque without weight coefficient.By defining the way of virtual torque,the control of torque error and flux linkage error in traditional model predictive torque control can be transformed into the control of torque.By forcing the control variables in the objective function to be consistent,the objective of eliminating the weight factor is achieved,and the parameter tuning of the controller is simplified.The mathematical model of no-weight-factor model predictive torque control and traditional model predictive torque control of asynchronous motor are built.The feasibility and validity of the control scheme in this paper are verified by comparing the two schemes.Finally,an experimental prototype based on TMS320F28335 DSP is built to verify the no-weight-factor model predictive torque control strategy.The results show that the control scheme in this paper can still satisfy the speed regulation requirements of the system without the weight factor.