| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 712079 | IFAC Proceedings Volumes | 2007 | 6 Pages |
This paper gives an overview of multivariate methods for extracting information from large process databases. Process data matrices are almost always of less than full statistical rank, and therefore latent variable methods are shown to be well suited to obtaining useful subspace models for treating a variety of important industrial problems. An overview of the important concepts behind latent variable models is presented and the methods are illustrated with industrial examples in the following areas of material processing: (i) the analysis of historical databases and trouble-shooting process problems (copper leaching and nickel powder decomposition); (ii) process monitoring and fault detection (steel casting); (iii) extraction of information from images for monitoring and control (flotation monitoring). In each of these problems latent variable models provide the framework on which solutions are based.
