Data Fusion Based Power Benchmarking for The Cement Manufacturing Process

摘要

In order to get the actual power consumption of the cement manufacturing process, the clustering algorithm and linear minimum variance fusion algorithm are proposed to establish the actual power benchmark of the cement manufacturing process. To manipulate the change of clustering number that is due to the variation of operation conditions and noise data, the DBSCAN algorithm is applied to cluster the power consumption data of each process unit in the cement manufacturing process. Classifications, clustering centers and their variances of power consumption data are obtained. The actual power consumption benchmark of each process unit in the cement manufacturing process is obtained by using the clustering centers and their variances with the linear minimum variance fusion algorithm. The actual power consumption benchmarking obtained by data fusion can reasonably reflect the actual level of power consumption, and reveal power saving potentialities.