Research On Direct Torque Control And Its Improvement Strategy Of Brushless Doubly-fed Machine

Brushless Doubly-Fed Machine(BDFM)eliminates brushes and slip rings.It has low maintenance cost and high reliability,can run under different operating conditions and has the advantages of small inverter capacity and adjustable power factor.This kind of machine has huge applicable potentials in variable frequency speed regulation systems.However,BDFM is a high-order,multi-variable,and strongly coupled nonlinear system.Its structure,operation mechanism,mathematical model and control strategy are relatively complicated.Direct Torque Control(DTC),possessing the features of smaller dependence on machine parameters and stronger robustness,is very suitable for the speed control of BDFM.However,the electromagnetic torque of BDFM is jointly produced by two sets of stator windings.The DTC method used to induction machine cannot be applied directly to BDFM.Furthermore,DTC has the problem of large flux linkage and torque pulsation at low frequency.Therefore,DTC strategy and its improving method of BDFM are deeply studied by the method of combining theoretical analysis,MATLAB simulation and prototype experiment in this thesis.Firstly,according to the basic structure and operating principle of BDFM,the mathematical model of BDFM in the rotor speed dq coordinate system is derived,and the open-loop characteristics of BDFM are studied,and then the simulation results are analyzed.Secondly,the DTC system of BDFM is deduced and established,and the speed signal is introduced on the basis of the amplitude limited flux linkage observer,which effectively solves the problem of inaccurate flux linkage observation of the voltage model at low frequency.Thirdly,to weaken the large flux linkage and torque ripples of the traditional DTC system,the expected voltage vector is used to estimate the required control voltage according to the magnitudes of flux linkage error and torque error,and the expected voltage DTC strategy is then proposed,which suppresses the traditional DTC’s flux linkage and torque ripples.Fourthly,the expected voltage vector estimation module is improved,and the Adaptive Neuro-Fuzzy Inference System(ANFIS)is used to generate the expected voltage vector,thereby eliminating the dependence on the mathematical model,reducing the amount of calculation and improving the real-time calculation performance and control effect.Finally,on the BDFM control platform based on TMS320F28335,the DTC system is verified by experiments.Simulation and experimental results show that DTC has good speed regulation performance when applied to BDFM,and the improved flux linkage observer is more accurate on the flux linkage observation at low frequency.Compared with the traditional DTC,the expected voltage DTC based on ANFIS can greatly reduce the flux linkage and torque ripples,achieve the expected effect of the optimization algorithm,and improve the stability and robustness of the system.The research results of this thesis provide theoretical and technical support for BDFM to obtain excellent speed regulation performance.