论文部分内容阅读
将EMD算法作为信号预处理的一种手段,将齿轮箱非平稳振动信号分解为一系列平稳过程的IMF分量,并将能量最大的有效分量做为系统模型的输出响应信号,输入轴转速转矩信号为系统输入信号,建立齿轮箱系统的时间序列ARX模型。实验结果表明:ARX模型将输入的波动计算在模型中,增强了系统模型的抗干扰能力,对于提高故障诊断的辨识精度具有重要意义。
The EMD algorithm is used as a signal preprocessing method to decompose the non-stationary vibration signal of the gearbox into a series of IMFs in a stationary process, and the effective component with the largest energy is taken as the output response signal of the system model. The input shaft speed torque The signal is the system input signal, and the time series ARX model of gearbox system is established. The experimental results show that the ARX model calculates the input fluctuation in the model and enhances the anti-interference ability of the system model, which is of great significance for improving the identification accuracy of fault diagnosis.