Modeling and performance monitoring of multivariate multimodal processes

Abstract

A multimodal modeling and monitoring approach based on maximum likelihood principal component analysis and a component‐wise identification of operating modes are presented. Analyzing each principal component individually allows separating components describing the variation within the individual modes from those capturing variation which the modes commonly share. On the basis of the former set, a Gaussian mixture model produces a statistical fingerprint that describes the production modes. The advantage of the component‐wise analysis is a simple identification of the mixture model parameters, which does not rely on the computationally cumbersome expectation maximization. The proposed method diagnoses abnormal process conditions by defining statistics relating to the components describing (1) between‐cluster variation, (2) within cluster variation, and (3) model residuals. The article demonstrates the benefits of this approach over existing work by an application to a continuous stirred tank reactor (CSTR) simulator and the analysis of recorded data from a furnace and a chemical reaction process. © 2012 American Institute of Chemical Engineers AIChE J, 59: 1557–1569, 2013

Nonlinear plant-wide process monitoring using MI-spectral clustering and Bayesian inference-based multiblock KPCA

“…In such condition, with all measured variables involved in one model can submerge or suppress the fault information. Building a global monitoring model, such as global PCA (GPCA, the conventionally used PCA) or global KPCA (GKPCA), for the entire process is not appropriate because the local behaviors are ignored [27,30,38]. To reduce the complexity and model the process more accurately, multiblock or distributed monitoring can be an efficient solution.…”

An aligned mixture probabilistic principal component analysis for fault detection of multimode chemical processes

“…The distance between the sample and the center of local models can be used as an indicator [5] and the minimum SPE value is also effective for choosing a proper local PCA model [6]. Feital et al [7] proposed a component-wise identification approach to estimate the current operating condition through optimizing a maximum likelihood objective function. If all sub-models are trustworthy, their monitoring results can be involved with respect to their posterior probabilities.…”

Just‐in‐time reorganized PCA integrated with SVDD for chemical process monitoring

“…In addition, PCA can explore the latent factors of the data, thereby providing better explanation and description of the process . Given the simplicity and efficiency of PCA, it has been extended to kernel PCA, dynamic PCA (DPCA), multiway PCA, multiscale PCA, and among others to solve various process monitoring problems . Although numerous successful applications of PCA for process monitoring have been reported, discussion of several problems, such as the following points, is still needed: (1) selection of retained principal components (PCs) is still an open question; (2) online process information are not considered when building a PCA model; and (3) the use of two statistics assumes Gaussian distribution of the process data …”