Intelligent Equipment Fault Signal Denoising with Improved Perona-Malik Model Based on Wavelet Transform Module

Due to the defects of traditional denoising algorithms for equipment vibration signals, an improved Perona-Malik model based on wavelet transform module was put forward. This new model can be applied to equipment vibration signal denoising by considering partial differential equations in image denoising. At first, the interrelation between wavelet threshold denoising and nonlinear diffusion filtering was studied. Secondly, an improved Perona-Malik model based on wavelet transform module was proposed by structuring a new diffusion coefficient function with wavelet transform module instead of gradient module. At last, the discretization numerical format of this improved Perona-Malik model was provided. The simulation experiments indicate that the improved Perona-Malik model can not only realize signal denoising effectively but also well reserve detail features without signal distortions.