Ball Mill Load Measurement Using Self-adaptive Feature Extraction Method and LS-SVM Model

  • Authors:

  • Gangquan Si;Hui Cao;Yanbin Zhang;Lixin Jia

  • Affiliations:

  • School of Electrical Engineering, Xi'an JiaoTong University, Xi'an, P.R. China 710049;School of Electrical Engineering, Xi'an JiaoTong University, Xi'an, P.R. China 710049;School of Electrical Engineering, Xi'an JiaoTong University, Xi'an, P.R. China 710049;School of Electrical Engineering, Xi'an JiaoTong University, Xi'an, P.R. China 710049

  • Venue:
  • ICIC '08 Proceedings of the 4th international conference on Intelligent Computing: Advanced Intelligent Computing Theories and Applications - with Aspects of Theoretical and Methodological Issues
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Ball mill load is the most important parameter optimized in the Ball Mill Pulverizing System (BMPS) in Thermal Power Plant. The accurate measurement of ball mill load is imperative and difficult. The approach based on self-adaptive feature extraction algorithm for noise signal and LS-SVM model is proposed to achieve this purpose. By analyzing the sensitivity distributions of working condition transitions, the characteristic power spectrum (CPS) is obtained. Then the self-adaptive weights can be calculated based on the CPS and the centroid frequency. The extracted features of the mill noise using this method show good adaptability to working condition transitions. Further more, softsensing models are built by combining a LS-SVM model and various feature extraction methods so as to estimate the mill load. Experimental results show that the performance of the model combined with the proposed extraction method is better than that with other methods.