Logo image
Two-dimensional dynamic principal component analysis with autodetermined support region
Journal article   Peer reviewed

Two-dimensional dynamic principal component analysis with autodetermined support region

Yuan Yao, Yinghu Diao, Ningyun Lu, Junde Lu and Furong Gao
Industrial and Engineering Chemistry Research, Vol.48(2), pp.837-843
21/01/2009

Abstract

Dynamics are inherent characteristics of batch processes. In some cases, such dynamics exist not only within a particular batch, but also from batch to batch. In previous work, two-dimensional dynamic principal component analysis (2-D-DPCA) has been developed to monitor 2-D dynamics. Support region determination is a key step in 2-D-DPCA modeling and monitoring of a batch process. A proper support region can ensure modeling accuracy, monitoring efficiency, and reasonable fault diagnosis. In this work, an automatic method for support region determination is developed. This data-based method can be applied on different batch processes without prior process knowledge. Simulation shows that the developed method has good application potentials for both monitoring and fault diagnosis. © 2009 American Chemical Society.

Metrics

1 Record Views

Details

Logo image