Sliding Mode Variable Structure Control Of BLDCM

Brushless DC motors (BLDCM) are reliable to control with simple structure and convenience to maintenance which is owned by the AC machines, and have high efficiency, good performance of speed regulation as DC machines. So it is widely used in the industry. The performance of the motor is influenced by the control strategy mostly. Some scholars have designed many control systems and gotten the good performance of the motor.It is difficult to design an accurate model because BLDCM is a nonlinear and multivariable system. The classical PID control needs accurate model, so can not have precise control to BLDCM. Modern control theory can increase the performance of BLDCM, such as self-adaptive control, fuzzy control and neural network control. But, self-adaptive control is sensitive to alteration of load; in the fuzzy control, the algorithm has nonlinear performance by reason of the interactivity of complicated fuzzy regulation, which makes fuzzy control imperfect; NN controller regulates parameters by learning, which needs microprocessors with high speed, so it is difficult to design the NN controller for BLDCM. By reason of that, the variable structure control and the extended state observer are employed.Sliding mode control is characterized with order reduction, disturbance depression, and insensitivity to parameter alterations. It has been used in motor speed control, position control and observers. The extended state observer (ESO) has good performance which can evaluate the states of uncertain object and real-time quantities of the object's disturbance. In this paper, a model is designed based on the characters of the model of the motor and the variable structure control. Based on the model of motor the variable structure controller is designed which is insensitive to the disturbance. The load torque can be evaluated by the extended state observer which increases the control precision of the system. The simulation illustrates that variable structure controller makes BLDCM less overshoot and quick velocity response which is insensitive to parameters of BLDCM, and has good depression of many disturbances. Finally, a TMS320LF2407A digital signal processor (DSP) and a field programmable gate array (FPGA) are used to set a control system of BLDCM which is studied in this paper.