传统电机参数辨识采用固定脉冲宽度调制(PWM)波占空比。该方法导致电流大小不可控,而辨识电流大小又直接影响辨识准确性。为此,提出新的参数离线辨识方法,采用恒电流辨识原理并检测电机相电流,控制辨识电流在给定值附近波动,解决了电机辨识电流不合适问题。设计了续流二极管和IGBT电压模型,d轴和q轴电感辨识都采用在270°而非0°进行辨识,以此提高了辨识精度。最后,在变频冰箱上进行验证并与原来数据进行对比。该方法很好地解决了冰箱电机带背压条件下,电机d、q轴电感辨识不准确、辨识时间长等问题,试验结果证明了该方法的可行性和有效性。

The parameters identification method of traditional motor often used fixed pulse width modulation (PWM) wave duty cycle, which led to uncontrollable identification current, and the accuracy of identification and was directly affected by the size of identified current. A new parameters offline identification method was proposed, which was based on the principle of constant current identification. It solved the problem of inappropriate identification current of motor by detecting the phase current of motor and controlling the fluctuation of identification current near the given value. The model of the flywheel diode and IGBT voltage was designed, which required that both daxis and qaxis inductance identification should be carried out at 270° instead of 0°, thus it could improve the identification accuracy. Finally, the method was applied to the frequency conversion refrigerator, and the experimental results were compared with the original data. This method could effectively solve problems such as inaccurate identification of the d,qaxis inductance and long identification time of the refrigerator motor with back pressure. The experimental results showed that the new method was feasible and effective.

TM 301.2