Multi-rate Nonlinear Observers for Process Monitoring, with Application to Polymerization Reactors

Texas A&M Engineering Experiment Station, College Station TX

Investigators

Abstract

Very often, critical quantities related to safety, product quality and/or economic performance of a chemical process cannot be measured on line. However, on-line measurements of other physical variables are always available, and these must be used to estimate and monitor unmeasured critical quantities. In this project, a dynamic model, with additional appropriate feedback terms from all the on-line measurements, will be used to build a state observer to estimate the state of the system, from which all pertinent critical quantities can be estimated and monitored. The project aims at transforming current industrial practice on process monitoring, by providing general algorithmic tools in a unified framework, as well as practical guidelines for their application to chemical processes. The research aims at (a) advancing the theory of nonlinear observers and (b) enabling and expanding the applicability of advanced nonlinear observers for model based monitoring of chemical processes, including polymerization reactors. The first objective of the project is the development of as systematic general methodology for the design of multi-rate multi-delay nonlinear state observers for process monitoring. Meeting this objective will involve theoretical research efforts, to advance methods and results of nonlinear systems theory. The second objective involves testing and evaluation of the practical applicability of the theoretical algorithmic framework to chemical process monitoring applications. In particular, monitoring of polymerization reactions will be the principal application area. Meeting this objective will involve modeling, simulation and observer design by applying the general algorithms of the theory part of the proposed project, as well as experimental implementation and testing. The research hopes to advance the theory and algorithms of nonlinear observers, including (a) the proper handling of multi-rate measurements and (b) the proper handling of multi-delay measurements in nonlinear observer design. It also aims at advancing the practical application of nonlinear observers for monitoring product quality in the operation of chemical processes, including polymerization reactor monitoring, by enabling and improving the utilization of all available on-line measurements. The results of the work will be broadly disseminated to researchers in academia and industry through presentations at scientific conferences, scholarly refereed journal publications and through a dedicated web site for the project. In addition to graduate students, the PIs plan to recruit underrepresented undergraduate students to pursue research in their laboratories.