Batch process monitoring and its application to polymerization systems

Abstract

Slight changes in raw material properties or operating conditions during critical periods of operation of batch and semi‐batch polymerization reactors may have a strong influence on reaction mechanism and impact final product quality. Online process monitoring, fault detection, fault diagnosis, and product quality prediction in real‐time ensure safe reactor operation and warn operators about excursions from normal operation that may lead to deterioration in product properties. Multivariate statistical process monitoring and quality prediction using multiway principal components analysis and multiway partial least squares have been successful in detecting abnormalities in process operation and product quality. When abnormal process operation is detected, fault diagnosis tools are used to determine the source cause of the deviation. Illustrative case studies are presented via simulated polyvinyl acetate polymerization.

Diagnosis of Abnormal Situations in a Continuous Solution Polymerization Reactor

“…[9] There are very few studies focusing FD issues in polymerization reactors. Among these, Zhang et al [10] use neural networks, Kaboré et al [9] use non-linear high-gain observers, meanwhile Kumar et al, [11] Undey et al [12] and Lou et al [13] use multivariate statistical approaches. In some of these works it is considered that the process operates in open-loop while in other works it is considered that the FD system performs only the initial FD tasks.…”

“…For the design of the UIFDO we need to satisfy the following existence conditions. Theorem: Necessary and sufficient conditions for the existence of a reduced-order UIFDO for the system defined by (12) are: [30] (i) rankðCE 1 Þ ¼ rankðE 1 The decoupling procedure of f 1 in the system defined in Equation (12) and the structure of the reduced-order UIFDO is illustrated in Figure 3.…”

“…Also, the matrix E x is partitioned into E x ½E 1 E 2 . Here, the objective is to decouple f 1 from other faults in Equation (12). For this purpose, consider the non-singular transformation matrix T ¼ U T 1 , where U 1 is obtained from the SVD of E 1 , that is,…”