People Education Research Industrial Agenda  
Current Research Archive Publications PhD theses Software      

Previous      Up      Next

Dynamic modeling, optimization and control of intensified production of fine chemicals in continuous reactors

Project members:  N. Nikacevic, A.E.M. Huesman
Keywords:  Modeling of process systems, Dynamic optimization, Actuation, Process intensification, Process control, Process technology, Old - project will be removed
Sponsored by:  Institute for Sustainable Process Technology ISPT (CoRIAC project)
Fine chemicals and pharmaceuticals are still dominantly produced in batch operated stirred-tank reactors due to their simple design and flexibility for multi-purpose production. Yet, continuous production in novel intensified reactors offers much more advantages than a traditional batch process. Intensified reactors radically improve transport properties and thus allow a reaction to approach its intrinsic kinetic limits. A precise control of the conditions directly relates to improved selectivity and productivity and to reduction in energy use, waste generation and time-to-market. In this project we will analyze operational, design and control characteristics of intensified reactors by using dynamic optimization and first principles models (experimentally validated) for several industrially relevant applications.

This work is a part of the CoRIAC project, (coordinated by the institute TNO) which scope is a demonstration of continuous reactors with in-line analytics for fine chemical production. Several production companies (Procter & Gamble, DSM and Janssen Pharma) provide broad variety of commercially relevant reaction cases. Most of these are complex and multiphase reactions. Several equipment manufacturing companies provide intensified reactors such as glass and ceramic microreactors, oscillatory flow (Figure 1), TNO Helix and static mixer reactor. The goal is to prove on a lab-scale and to gain knowledge for a selection of suitable reactors for a specific reaction. Furthermore, the aim is to develop and test robust process analytical tools for on-line monitoring of continuous reactions. The next phase is a pilot plant demonstration of selected reactors with in-line analytics.

In this project we are focusing on analysis of operational behavior for few selected reaction cases, which would provide optimal operation properties. This will be addressed through five stages:

  • Development of a dynamic model. First principles modeling will be used. Reaction kinetics and flow properties of reactors will be experimentally determined in the lab-scale setups (model verificationa and validation).
  • Analysis of the operational degrees-of-freedom. Actuation has a large impact on optimal performance and controllability of continuous reactors for fine chemicals. In this stage, actuation will be systematically analyzed and designed.
  • Dynamic optimization. Using developed models, a dynamic optimization program is going to be constructed and solved. The results will facilitate preliminary scale-up and design (pilot-plant) and will be used for improvement of operation and actuation and design of experiments for a larger system. Furthermore, optimization will provide useful data for sensors development and guidelines for optimal control.
  • Controllability analysis. Using robust optimization, a controllability study will be performed in order to address the stability and flexibility issues of the optimal solutions suggested in the previous stage.
  • Conceptual control design. Depending of the outcome of the previous stages, a control concept for a particular reactor case will be suggested. Techniques which will be examined are: base layer control, model predictive control and optimal control.

a) b)

Figure 1. Oscillatory flow reactor (OFR): a) A typical design of OFR with inner baffles and external oscillator (© University of Cambridge, UK) b) Intensive mixing in the reactor tube due to flow oscillations in the presence of baffles.

Previous      Up      Next

Back to top

Last modified: 15 November 2013, 16:30 UTC
Search   Site map