Abstract
This paper addresses the problem of quality modeling in polymethyl methacrylate (PMMA) production. The key challenge is handling the large amounts of missing quality measurements in each batch due to the time and cost sensitive nature of the measurements. To this end, a missing data subspace algorithm that adapts nonlinear iterative partial least squares (NIPALS) algorithms from both partial least squares (PLS) and principal component analysis (PCA) is utilized to build a data driven dynamic model. The use of NIPALS algorithms allows for the correlation structure of the input–output data to minimize the impact of the large amounts of missing quality measurements. These techniques are utilized in a simulated case study to successfully model the PMMA process in particular, and demonstrate the efficacy of the algorithm to handle the quality prediction problem in general.
Highlights
IntroductionPolymethyl methacrylate (PMMA) is an important industrial polymer with widespread applications ranging from plastics to electronics
Quality Modeling with Missing DataPolymethyl methacrylate (PMMA) is an important industrial polymer with widespread applications ranging from plastics to electronics
The present manuscript provides a solution for instances where the Processes 2021, 9, 1691 low frequency of quality measurements in relation to other online measurements results in a missing data problem. To overcome these challenges and model the Polymethyl Methacrylate (PMMA) process, the current paper focuses on a recent modification to subspace identification [20] that treats nonuniform sampling rate data as a missing data problem
Summary
Polymethyl methacrylate (PMMA) is an important industrial polymer with widespread applications ranging from plastics to electronics. This leads to scenarios where inputs and outputs have some missing values due to differences in sampling rates, whereas quality measurements are available at extremely low frequencies This presents a challenge to traditional data driven modeling approaches that require complete data sets to identify a model. The approach identifies a time-varying model and, since PLS inherently does not distinguish between input and output variables, it has limited applicability in traditional model predictive control and scenarios where batch duration is a decision variable Another technique that is suitable for building process models is subspace identification [5,6,7,12], which has been appropriately modified for handling batch data using.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
