基于卡尔曼滤波算法的无刷直流电机直接转矩控制

The Direct Torque Control of Brushless DC Motor Based on Kalman Filter Algorithm

阐述了一种新的无刷直流电机转矩检测方法,可应用于直接转矩控制。传统的转矩观测方法有基于磁链观测转矩和基于反电势观测转矩,其分别存在计算复杂、误差大和难以得到准确的反电势的问题。根据无刷直流电机运行特性,提出了通过采集端电压和相电流的卡尔曼滤波转矩观测策略,仿真结果表明,能够准确观测到电机转矩。通过设计卡尔曼滤波算法,有效避免了微分运算产生干扰的问题,且不需要额外的滤波电路,节约了系统成本,并且利用转矩观测时产生的中间变量,完成无刷直流电机无位置传感器控制。

A novel torque estimation for brushless dc motor was presented,this technology could applied to brushless dc motor direct torque control. The traditional torque estimation strategy could be basely classified as calculated by flux and calculated by back-EMF,it have the troubles of computation too complex and back-EMF is hardly obtained,respectively. Considering the characteristics of the brushless dc motor,a novel Kalman filter torque estimation strategy was designed,through detecting phase current and terminal voltage of motor,the simulation results show that the proposed scheme can acquire precise torque. The Kalman filter algorithm could effectively eliminate the noise generated by differential operation,and no filter hardware circuit needed,the system cost could been saved by use this algorithm. Sensorless control of brushless dc motor was completed also based on the process of previous torque estimation.