Process faults may be detected on-line using existing measurements based up
on modelling that is entirely data driven. A multivariate statistical model
is developed and used for fault diagnosis of an industrial fed-batch ferme
ntation process. Data from several (25) batches are used to develop a model
for cultivation behaviour. This model is validated against 13 data sets an
d demonstrated to explain a significant amount of variation in the data. Th
e multivariate model may directly be used for process monitoring. With this
method faults are detected in real time and the responsible measurements a
re directly identified. The fault detection and identification is enabled t
hrough inspection of a few simple plots. Thus, the presented methodology al
lows the process operator to actively monitor data from several cultivation
s simultaneously. (C) 1999 Elsevier Science S.A. All rights reserved.