Semester

Spring

Date of Graduation

2019

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Chemical and Biomedical Engineering

Committee Chair

Fernando V. Lima

Committee Co-Chair

Debangsu Bhattacharyya

Committee Member

Debangsu Bhattacharyya

Committee Member

Stephen Zitney

Abstract

Online Model Predictive Control of a Nonisothermal and Nonisobaric Membrane Reactor for Water-Gas Shift Reaction Applications Jacob M. Douglas Modern hydrogen production units are tasked with producing the most hydrogen possible while dealing with flow variations caused by changing power demands. Classical methods for hydrogen production employing the water-gas shift reaction are governed by equilibrium limitations that take effect at high temperatures and high concentrations of H2 (Georgis, et al., 2014). The implementation of a membrane reactor with temperature control enables the hydrogen concentration and temperature to reach an equilibrium at a higher concentration of H2. Another challenge that is prevalent in this process is the cyclical hydrogen demand from changing downstream reforming process conditions. These challenges can be addressed by the implementation of advanced controllers that can cope with dynamic changes associated with different conditions, such as temperature oscillations and mitigation of hot spots. In this thesis, linear and nonlinear model predictive control (MPC) methods are implemented on a designed water-gas shift membrane reactor model in Aspen Custom Modeler. The implementation aim is to increase the production of hydrogen by considering the temperature control performed by manipulating the flow rates of the coolant entering the cooling jacket at different reactor zones as well as the reactor sweep flowrate. The control strategies considered for this application are: Quadratic iii Dynamic Matrix Control (QDMC), Nonlinear MPC (NMPC), and a Biomimetic-based controller cast as MPC (BIO-CS as MPC) (Mirlekar, et al., 2018) . The coolant usage is constrained by the use of quadratic programing (QP), sequential quadratic programing (SQP), or dynamic operations toolbox (DYNOP) solvers, depending on the employed MPC type, to match industrial standards. To mimic industrial conditions, the flowrate of hydrogen in the sweep stream is changed by +15% from its operating steady state. The MPC results that will be discussed show a successful increase in the production

1. Georgis, D., and Lima, F. V. (2014). Thermal Management of a Water-Gas Shift Membrane Reactor for High-Purity Hydrogen Production and Carbon Capture. Industrial & Engineering Chemistry Research, 7461-7469. 2. Mirlekar, G. V., Li, S., and Lima, F. V. (2017). Design and Implementation of a Biologically-inspired Optimal Control Strategy (BIO-CS) for Chemical

Download

Share

COinS