Multivariate statistical process monitors

摘要: An extended partial least squares (EPLS) approach for the condition monitoring of industrial processes is described. This EPLS provides two statistical charts to detect abnormal process behaviour as well contribution diagnose this behaviour. A theoretical analysis provided, together with application studies show that either more sensitive or easier interpretation than conventional PLS. Generalised scores are calculated by constructing an augmented matrix, form Z=[Y{dot over (:)}X], where X predictor matrix and Y response a score T n =T* −E* in which T* E* generally form: * = [ ⁢ ⋮ ] ⁡ B PLS ( ) : 1 R E F the columns providing generalised t-scores residual scores, where ℑ denotes M×M identity matrix, (n) regression matrix.