转轴扭转变形动态测量噪声消除方法研究

转轴的扭转变形信号中包含了大量与转轴运行状态相关的信息,但往往被噪声所淹没,合理的去噪处理能够得到更有效的信号成分,有助于对转轴的工作状态做出正确的判断。相比于传统滤波方法,基于EMD的去噪方法有较强的自适应性,更适合动态信号的处理。从转轴相互正交的两个方向采集的相位差信号,依据两路信号EMD分解后各阶本征模态函数的相关性去除噪声并恢复出有效信号。仿真验证了该方法的有效性,并将其运用到实际转轴扭转变形信号的处理中,得到了更为准确的扭转变形信息。

Reduction of Dynamic Noise in the Measurement of Shaft Torsional Deformation Based on EMD

A large number of information related to the running state of the rotating shaft is contained in the torsional deformation signal of the shaft. Reasonable method can get more effective signal components, it can help to make correct judgement on the working state of the shaft. Compared with the traditional denoising method, the method based on EMD has a strong adaptability. Effective signal is recovered from the original signals collected from two mutually orthogonal directions, after being filtered outnoise according to the correlation of EMD decomposition. The simulation results show that the method is effective and can be used to the actual rotating shaft to reduce the distortion of the signal.