A Denoising Method Based on Improved VMD-Wavelet Threshold and its Application to Rolling Mill Vibration Monitoring

Steel is the lifeblood of the nation, and rolling mill are crucial for steel production. Therefore, it is necessary to monitor rolling mill. The aim of this study is to achieve accurate monitoring of rolling mill vibrations in the presence of noise. This paper utilizes an accelerometer sensor to design a rolling mill vibration signal acquisition device and proposes an improved denoising method that combines variational mode decomposition (VMD) with enhanced wavelet thresholding. The improved coati optimization algorithm is used to optimize the optimal number of decomposition layers and penalty factor for the VMD, enhancing the decomposition performance. Enhanced wavelet threshold function solves the shortcomings of discontinuity and distortion of wavelet denoising methods. The improved joint denoising method is compared with other methods. The experimental results show that the proposed method can effectively reduce the noise of the rolling mill vibration signal and improve the accuracy of the vibration monitoring device, which has good practical application value.