Using multivariate regression techniques to analyse the performance of a steam heated drying process

This paper explores the use of a multivariate statistics approach to monitor and diagnose the performance of a steam heated drying process. The results presented in this paper are a first stage investigation into the capabilities and limitations of these methods when applied to this particular process. The results show that, despite the non-linear behaviour of the drier process, the model predictions correspond well with the process measurements. The results also indicate ability to detect process failures, however, accurate diagnosis of the failures remains a subject for future studies