Research On Generalized Predictive Control Strategy Of Brushless DC Motor

Brushless DC motors(BLDCMs)have been widely used in automotive electronics,aerospace industries and other fields owing to their advantages such as high operating efficiency and superior speed regulation performance.Speed regulation performance of BLDCMs is the key to successful applications,and the control scheme design is the focus of research.At present,model predictive control(MPC)is considered to be most likely the control algorithm to replace PI in the field of fast systems.Affected by factors such as measurement accuracy and sensor accuracy,there are errors in system modeling.Therefore,this thesis introduces generalized predictive control(GPC)with adaptive mechanism into the design of control schemes of BLDCM,improved strategies based on orthonormal basis functions(OBFs)are proposed.The main work and innovative points of this thesis are as follows:(1)Aiming at the problem of uncertain delay in the ARMAX predictive model of GPC,a control-oriented model called ARMAX-GOBF is proposed.This model uses generalized orthonormal basis functions(GOBFs)for parameterized design of the ARMAX model to make it have the advantages of both models.The simulation results show that the model has good approximation ability.(2)Aiming at the problem of large amount of calculation in moving horizon optimization of GPC,the improved GPC algorithm is proposed with the incremental input parameterization.The improved algorithm introduces the Laguerre orthonormal basis functions into the control design of the incremental GPC algorithm,at the same time,the cost function of PI structure is constructed to combine the advantages of traditional control algorithm.The simulation results show that the improved control algorithm improves response speed and dynamic performance of the system.(3)Based on the YX-EMBOX motor control hardware platform,the above improved control strategy is programmed and debugged to realize speed controlling of BLDCM under no-load condition.The experimental results show that the improved control algorithm has the characteristics of fast response speed,small overshoot and small steady-state error compared with traditional PI control.