T. Fujiwara et al., FAULT-DETECTION BASED ON FUNCTIONAL-RELAT IONSHIP AMONG PROCESS VARIABLES BY AUTOASSOCIATIVE NEURAL NETWORKS, Kagaku kogaku ronbunshu, 22(4), 1996, pp. 846-853
Some process variables measured in a plant are strictly constrained by
the material and heat balance equations, rate equations and correlati
ons. In this study, we propose a method to judge whether the state of
plant operation is normal or not, by examining whether a set of proces
s variables maintains the functional relationship specified at normal
operation. The functional relationship at normal operation is identifi
ed by an autoassociative neural network (AANN) which approximates the
identity mapping for a set of measured values of process variables. An
effective method to search for an adequate configuration of the AANN
is also presented. Abnormal operation or fault is detected by the magn
itude of discrepancy between the input vector and the output vector of
the trained AANN. This fault detection method is applied to a continu
ous flow polymerization process and compared with the conventional 3 s
igma fault detection method for a single process variable.