Melt index prediction with a mixture of Gaussian process regression with embedded clustering and variable selections

Abstract

In this study, a penalized mixture of the Gaussian process regression model was proposed for the prediction of melt index (MI) in industrial polymer production. MI plays an important role in detecting the grade of a product. It is difficult to measure directly and is characterized by a large number of variables and multigrades. Because of multigrade products, in the development of soft sensors for MI prediction, it is not valid to assume unimodal Gaussian distribution of the data. To this end, the proposed method is capable of the simultaneous identification of significant variables and determination of important clusters of multigrade products. It is based on the shrinkage methods that have been shown to provide stable models that are more interpretable. Case studies are presented to show the features of the proposed method and its applicability to industrial MI prediction. © 2017 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2017, 134, 45237.

Hybrid modeling approach for polymer melt index prediction

“…Additionally, two mechanistic MI prediction models were developed in order to compare the accuracy of their predictions. First, mechanistic model is developed based on the relationship between MI and weight-average molecular weight as in (22). Due to the lack of molecular weight measurement data for the SAN polymerization process, the weight-average molecular weight derived from the white-box submodel was used for modeling.…”

“…Rather than that, nonlinear machine learning modeling methods such as support vector machines (SVM), 12–13 least squares support vector machines (LSSVM), 15–17 relevance vector machines (RVM), 18 and Gaussian process regression (GPR) have been used 19–23 . SVM is a machine learning algorithm developed based on Vapnik–Chervonenkis (VC) theory and has been demonstrated good modeling performance in both classification and regression problems 24–26 .…”

Comprehensive Quality Assessment Algorithm for Smart Meters

“…By this data analysis technology, data are projected onto multiple Gaussian models and the probability of the data on each Gaussian model is obtained. Then, the model with the highest probability is selected to decompose a large data set into several small data sets that conform to the Gaussian distribution [20,21]. Therefore, the Gaussian mixture model can be used for data clustering to extract typical features from a data set.…”

“…Therefore, based on the statistical characteristics of the quality data of smart meters, this paper attempts to consider the probability of various fault alarm events produced by smart meters as the superposition of multiple Gaussian distributions. The Gaussian mixture model (GMM) [20,21] clustering algorithm is applied to discover the internal correlation among the fault alarm data and to extract the typical indexes that characterize the operation quality of the smart meters, thereby to reduce the storage scale and calculation complexity of the comprehensive assessment model of smart meters.…”

Soft Sensor Development Based on Quality‐Relevant Slow Feature Analysis and Bayesian Regression with Application to Propylene Polymerization

“…Unfortunately, as the model is linear, it is inadequate when the soft sensing variables are nonlinear. To this end, nonlinear modeling methods are used in MI prediction, such as artificial neural networks (ANNs) [17], support vector machines (SVMs) [18], Gaussian process regression (GPR) [19,20], and relevance vector machine (RVM) [21] [ [22][23][24][25]. Recently, Liu et al proposed an adversarial transfer learning-(ATL-) based soft sensor [26] and a domain adaptation transfer learning soft sensor for product quality prediction [7].…”