Optimal operation of the Alfa Laval Plate Reactor using feedback control
Systematic methods and tools for managing the complexity
Process Control (T4-8P)
Keywords: process control, chemical reactor, model predictive control, optimization
This paper presents methods on how to operate and control the Alfa
Laval Plate Reactor for a challenging set of reactions. In a real
industrial process, there are disturbances and uncertainties that may
limit the benefits of a good process design. Here, feedback control is
employed to overcome disturbances and reach the desired specifications
despite uncertainties in the process. Online feedback optimization is
used to utilize the full potential of the new reactor concept. With
the large number of input variables it is non-trivial how to manually
adjust them to optimize performance, while maintaining safe operation.
The Alfa Laval Plate Reactor is a new type of chemical reactor, being
developed by Alfa Laval AB. It is a combination of a tubular reactor
and a plate heat exchanger, which combines good micro-mixing and high
heat transfer into one operation. Therefore the plate reactor appears
particularly suited to process intensification, as it allows at the
same time an increase of reactant concentration and a desired
reduction of the solvent consumption.
The design concept for the reactor allows for great flexibility in
adapting the process for new reaction schemes. The reactants can be
injected at multiple locations along the reactor. There may be several
independent cooling zones along the reactor to improve the temperature
control possibilities. Internal temperature sensors enable feedback control to
adapt the process operation to the current conditions.
In this paper, a second order exothermic reaction A + B \rightarrow C
+ D is studied. In addition, there is also a side reaction, where the
product C reacts to an unwanted byproduct, C \rightarrow E. There
is a trade off between high reactor temperature to accelerate the
primary reaction and low temperature that reduces the influence from
the side reaction. With the multiple injection points and the
independent cooling zones along the reactor, it is possible to design
the process so that the effect of the side reaction is minimized.
A Nonlinear Model Predictive Controller is designed to carry out
online dynamic optimization, to optimize the conversion or
productivity for the actual operating conditions.
The plate reactor configuration is evaluated in simulations. In
addition, simulations show that the optimization and the feedback
controller can lead to chemical production with high conversion and
productivity, despite feed disturbances and uncertain process parameters.
For more background information, see:
Haugwitz, Hagander and Norén: "Modeling and Control of
a Novel Heat Exchange Reactor, the Open Plate Reactor". Control
Engineering Practice, 2006. In press, available online through DOI
using doi:10.1016/j.conengprac.2006.02.019.
The Alfa Laval Plate Reactor - http://www.alfalaval.com/ecoreJava/WebObjects/ecoreJava.woa/wa/showNode?siteNodeID=9193&contentID=36740&languageID=1
