No Detail Found!

Ensemble modeling of mill load based on empirical mode decomposition and partial least squares.

Author(s)
Lijie Zhao College of Information Engineering, Shenyang University of Chemical Technology, Shenyang 110042, Liaoning, China; State Key Laboratory of Synthetical Automation for Process Industries, Northeastern University, Shenyang 110819, Liaoning, China
Jian Tang Unit 92941, PLA, Huludao, China
Wengrong Zheng Naval University of Engineering, Wuhan 430032, Hubei, China
Abstract
Reliable measurements of ball mill load parameters and reorganization of the operating statuses are the key factors for saving energy and optimization control. Empirical mode decomposition (EMD) and partial least squares (PLS) are used to analyze shell vibration signal and monitor mill load parameters of ball mill. The shell vibration signal is decomposed into several intrinsic mode functions (IMFs) adaptively. The power spectral density (PSD) of each IMF is analyzed under different grinding conditions. A new index is defined to measure the relative change of each IMF to original signal, which is also used to select more informational IMFs. The ensemble PLS method is used to build soft-sensor models based on frequency spectrum of the selected IMFs. Experimental results show that ensemble soft-sensor model based on EMD and PLS can extract effective features of shell vibration signal and monitor mill load effectively.
Publication Details
Page(s) 179-191
DOI DOI not available
Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 45, Issue: 1, Year: 2012

Request the full text from PASTIC

More from Journal of Theoretical and Applied Information Technology
0

Citations

0

Downloads

126

Views

Copyright ©  PASTIC National Center, Islamabad www.pastic.gov.pk