Process modeling and control applied to real-time monitoring of distillation processes by near-infrared spectroscopy

Abstract

A distillation device that acquires continuous and synchronized measurements of temperature, percentage of distilled fraction and NIR spectra has been designed for real-time monitoring of distillation processes. As a process model, synthetic commercial gasoline batches produced in Brazil, which contain mixtures of pure gasoline blended with ethanol have been analyzed. The information provided by this device, i.e., distillation curves and NIR spectra, has served as initial information for the proposal of new strategies of process modeling and multivariate statistical process control (MSPC). Process modeling based on PCA batch analysis provided global distillation trajectories, whereas multiset MCR-ALS analysis is proposed to obtain a component-wise characterization of the distillation evolution and distilled fractions. Distillation curves, NIR spectra or compressed NIR information under the form of PCA scores and MCR-ALS concentration profiles were tested as the seed information to build MSPC models. New on-line PCA-based MSPC approaches, some inspired on local rank exploratory methods for process analysis, are proposed and work as follows: a) MSPC based on individual process observation models, where multiple local PCA models are built considering the sole information in each observation point; b) Fixed Size Moving Window - MSPC, in which local PCA models are built considering a moving window of the current and few past observation points; and c) Evolving MSPC, where local PCA models are built with an increasing window of observations covering all points since the beginning of the process until the current observation. Performance of different approaches has been assessed in terms of sensitivity to fault detection and number of false alarms. The outcome of this work will be of general use to define strategies for on-line process monitoring and control and, in a more specific way, to improve quality control of petroleum derived fuels and other substances submitted to automatic distillation processes monitored by NIRS.

Control and Monitoring of Milk Renneting Using FT-NIR Spectroscopy as a Process Analytical Technology Tool

“…By the combination of T 2 and Q control charts, it was possible to detect the in-control tested batch (NOC 13c ) and to distinguish failure batches just from the first minutes of the process. Similarly, de Oliveira et al [30] concluded that the combination of T 2 and Q charts gave specificity and sensitivity results more reliable than their single check when applying MSPC to synthetic commercial gasoline distillation.…”

“…Before performing MCR analysis, the selection of the proper number of components was performed through singular value decomposition (SVD algorithm on which PCA is based) by the ALS approach. Since two components were used to describe the process changes observed in PCA of mean-centered data, it is consistent that three components were found relevant by ALS applied to the not-centered data, because the rank decreases by one when mean-centering is performed [30].…”

Detection of Gasoline Adulteration Using Modified Distillation Curves and Artificial Neural Network

“…For example, Mendes and Barbeira 29 performed a study for detecting adulteration and the amount of solvent in gasoline using distillation curves combined with multivariate methods, including PCA and PLS discriminant analysis (PLS‐DA). Oliveira et al 30 designed a distillation‐based quality control procedure to analyze synthetic commercial gasoline, which contains mixtures of pure gasoline blended with ethanol. Their automatic device made it possible to obtain continuous and synchronized measurements of the distillated mass percentage, distillation temperature, and NIR spectra during the distillation process.…”

Data fusion strategies to combine sensor and multivariate model outputs for multivariate statistical process control

“…The outputs based on spectroscopic measurements come from the use of different multivariate analysis tools, e.g. multivariate calibration models provide prediction of product key properties [6], multivariate curve resolution (MCR) deliver concentration profiles associated with the evolution of compounds in a process [7] and multivariate statistical process control (MSPC) gives indicators that may tell whether the process is on-or off-specifications [8]. In addition to the spectroscopic sensors, most processes are also monitored with simpler devices providing other univariate measurements, such as temperature, pressure, pH or flow rates.…”

“…c) Distillation process. This process is monitored by NIR spectroscopy and vapour temperature measurements [7]. Here, the evolution of the distillation process was controlled via data fusion on-line MSPC models based on the combination of compressed NIR-based information, expressed by the concentration profiles derived from multivariate resolution analysis (MCR) of the process spectra and vapour temperature measurements (see Fig.…”