Statistical Analysis and Monitoring For Handling Different Modes Batch Processes

摘要

In the paper, a new online monitoring approach is proposed for handling the multimode problem in the industrial batch processes. The contributions are as follows: (1) Extracting method of the common characteristics from different modes is proposed; (2) A new model analysis method is proposed. There are both similarity and dissimilarity in the underlying correlations of different modes. By an adequate decomposition, one part of the underlying variations in each subspace stay invariable over subspaces while the other part changes with the alternation of subspaces. After that two different subspaces are separated, modeling the common and specific subspaces respectively. The common subspace and specific subspace are analyzed respectively, and the monitoring process is carried out in each subspace. The corresponding confidence regions are constructed according to their models respectively. Then the online monitoring system is set up, which can track different types of variations closely.