Vibration Analysis Based on Empirical Mode Decomposition and Partial Least Square

Abstract

Accurate detection of the ball mill load is one of the key factors to realize the automatic control of the grinding process. The shell vibration acceleration signal contains lots of useful information about the parameters of mill load, especially pulp density (PD). A empirical mode decomposition (EMD), fast Fourier transform (FFT) and partial least square (PLS) based approach is used to analyze the shell vibration signal and construct the PD soft sensor model. The EMD technology is used to decompose the shell vibration signals into a number of intrinsic mode functions (IMFs). Then, the power spectral density (PSD) of each IMFs is obtained using FFT. At last, pulp density model based PLS algorithm is constructed, which also is used to analyze the PSD of every IMFs. The experimental results show that the proposed approach is effective to analyze the shell vibration and modeling the PD.