Premium
Sub‐stage PCA modelling and monitoring method for uneven‐length batch processes
Author(s) -
Chang YuQing,
Lu YunSong,
Wang FuLi,
Wang Shu,
Feng ShuMin
Publication year - 2012
Publication title -
the canadian journal of chemical engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.404
H-Index - 67
eISSN - 1939-019X
pISSN - 0008-4034
DOI - 10.1002/cjce.20524
Subject(s) - principal component analysis , stage (stratigraphy) , division (mathematics) , process (computing) , batch processing , computer science , biological system , component (thermodynamics) , similarity (geometry) , algorithm , data mining , mathematics , pattern recognition (psychology) , artificial intelligence , physics , paleontology , thermodynamics , arithmetic , image (mathematics) , biology , programming language , operating system
For the characteristic of multistage of batch processes, a new PCA‐based sub‐stage division algorithm is proposed. This algorithm is based on the fact that production transition can be detected by analysing the loading matrixes and principal component matrixes, which reveal objectively evolvement of underlying process behaviours. Sub‐stage PCA models for each stage are built, and then extended to monitor the batch processes with uneven‐length durations by choosing the right sub‐stage model according to the principle of minimum similarity distance of principal component matrixes. With the proposed method, multi‐stage batch processes with durations of uneven‐length can be monitored effectively.