渐进信号瞬时频率提取及其在电机性能测试中的应用

Instantaneous Frequency Extraction of Asymptotic Signal and Its Applications in Motor Performance Test

平稳相位原理表明,渐进信号在连续小波变换时频平面上的分布呈现脊线的形态,信号的主要参数可以从其对应脊上的信息中提取出来。本文根据分析母小波在特定中心频率条件下小波脊线与信号瞬时频率的关系,提出了渐进信号瞬时频率的提取方法。根据直流电机起动电枢电流信号中高频分量的渐进特性,将其应用到电机性能测试系统中,实验结果表明该方法能实现无转矩?转速传感器的电机性能快速准确测试。由于小波基函数的时频局部特性,该方法具有良好的去噪性能。

The stationary phase principle proves that for the asymptotic signal, the behavior of its continuous wavelet coefficients shows ridge characteristics on the time-frequency plane. Some main parameters of the asymptotic signal can be extracted based on the information of its corresponding wavelet ridge. This paper proposes the asymptotic signal instantaneous frequency extraction method based on the relationship between its wavelet ridge and instantaneous frequency when the center frequency of the mother wavelet is selected. Due to the asymptotic characteristic of the high frequency component in the DC motor startup current, this method can be used in the intelligent motor performance test system. Experimental results show that the torque-speed characteristic curve can be acquired rapidly and exactly. Furthermore, this method is proved to have excellent de-noising ability due to the local characteristic in time-frequency plane of the wavelet base functions.