| Servo motors are a kind of motors that can realize accurate control of position,speed and acceleration.They are widely applied in the areas which have higher demand of fast dynamic response and control accuracy,such as robots and computer numerical control(CNC)machines.For a servo motor applied in such applications,a control method with good static/transient performance and robustness is essential.It is known that the control of position,speed and acceleration is realized by the control of current.Hence,current control methods are very critical for the control of servo motors.Nowadays,increasing servo systems adopt permanent magnet synchronous machines(PMSM)as actuators due to their merits,in terms of high power density,high efficiency,easy mainteinance and excellent control performance.Therefore,it is meaningful to research on current control methods for PMSM.Predictive current control(PCC)methods have been widely investigated in the area of servomotor control due to its fast dynamic response.As one of PCC methods,deadbeat predictive current control(DPCC)method has merits such as high dynamic performance,simple control algorithm,low computation cost and constant switching frequency.However,performance of DPCC will be influenced by the accuracy of mathematical model of motor,since it is deduced from mathematical model of motor.The aim of this thesis is to modify the conventional DPCC method,so that its dependency on accurate mathematical model of motor can be lowered.Main contributions of this thesis are listed as below:(1)The influence of parameter mismatches,nonlinearity of voltage source inverter(VSI)and phase delay introduced by the transformation on static/transient performance and robustness of conventional DPCC is analyzed.Control algorithm of conventional DPCC is deduced.The major factors in determining the control performance of conventional DPCC are analyzed.There will be steady-state error between stator current and reference current if parameter mismatches exist,and the dynamic performance of conventional DPCC will be deteriorated,too.The system will even be unstable if inductance mismatch is to large.Due to nonlinearity of VSI,there will be steadystate error between stator current and reference current,and current ripple will be larger.In order to reduce the influence of nonlinearity of VSI,a simple dead-time compensation method is introduced.Phase delay will also cause steady-state error between stator current and reference current.A simple compensation method is introduced to solve this problem.(2)An incremental model based deadbeat predictive current control(IDPCC)method is proposed.Incremental model of PMSM is given,and the influence of parameter mismatches,nonlinearity of VSI and phase delay on IDPCC is analyzed.The control algorithm of IDPCC is straightforward.And IDPCC is capable of elimating current errors in two control periods under ideal conditions.Moreover,IDPCC possesses excellent static performance,it can achieve zero steady-state error even when parameter mismatches,nonlinearity of VSI and phase delay exist.However,the narrow stable operation range of IDPCC limits its application.(3)In order to enhance the robustness of IDPCC,a robust incremental model based deadbeat predictive current control(RIDPCC)method is proposed.The influence of parameter mismatches on RIDPCC is analyzed.RIDPCC possesses simple control algorithm and excellent static performance,too.And RIDPCC can also elimate current errors in two control periods under ideal conditions.However,when voltage margin is insufficient,the dynamic response of RIDPCC will be slower than that of conventional DPCC and IDPCC.(4)Models of PI,conventional DPCC,IDPCC and RIDPCC are build using MATLAB/Simulink.And experiments are carried out on test bench based on TMDSHVMTRPFCKIT from Texas Instruments.The proposed methods are verified by simulation results and experimental results.Finally,characteristics of PI current control,conventional DPCC,IDPCC and RIDPCC are summarized. |
PDF Full Text Request