基于线调频小波路径追踪阶比跟踪算法的齿轮箱故障诊断研究

A New Order Tracking Method Based on Chirplet Path Pursuit and Its Application to Gearbox Fault Diagnosis

针对阶比跟踪转速获取硬件方法需要额外安装转速测量设备,软件方法精度不高、抗噪能力弱的问题,提出基于线调频小波路径追踪瞬时频率估计的齿轮箱阶比跟踪故障诊断方法。该方法利用基于线调频小波路径追踪瞬时频率估计算法适于分解频率呈曲线变化的非平稳信号的特点,采用其对齿轮箱的啮合频率分量进行估计以获取转速信号,依据转速信号对等时间间隔采样信号进行等角度重采样,将非平稳信号转化为角域平稳信号,得到振动信号的阶次谱,判断齿轮箱故障。仿真算例与应用实例表明上述方法在瞬时频率估计方面具有精度高和抗噪能力强的优点,可以根据信号自身的特点自适应的选择基函数,准确地对转速进行估计,其与阶比跟踪算法的结合能有效诊断齿轮箱故障。

Aiming at the problems of order tracking method, such as extra installation of speed measuring devices, low precision of speed acquisition compute methods, and weak antinoise ability, a new order tracking gearbox fault diagnosis method based on Chirplet path pursuit in instantaneous frequency estimation is proposed. Taking the advantage of the said estimation which is suitable to decompose nonstationary signals with frequency variation curve, the method takes speed signal by estimating the gearbox＇s meshing frequency component. With the speed signal, it carries out even-angle resampling on even interval sampling signals, transfer non-stationary signals to angular domain stationary signals, and acquire the order spectrum of vibration signals, so as to judge the fault of gearbox. Simulation and application examples prove that the method is of high precision and antinoise ability, which, based on its own features, can adaptively choose base function, and accurately estimate the speed, whose combination with order tracking method can be effectively used in gearbox fault diagnosis.