|
The IUP Journal of Chemical Engineering
|
|
|
|
|
|
Abstract |
|
As a result of good modeling capabilities, neural networks have been
used extensively for a number of chemical engineering applications such as sensor
data analysis, fault detection and nonlinear process identification. However, only
in recent years, with the upsurge in the research on nonlinear control, its use in
process control has been widespread. But there is almost no work which has tested
the capacity of Artificial Neural Network (ANN) to model Reactive Distillation
(RD). The superiority of ANN was tested for the synthesis of 2-methylpropylacetate
from 2-methylpropanol and acetic acid by RD. The effects of various parameters like
feed, temperature, reflux ratio, and boil-up ratio, each varied at three levels, were
studied. At each combination, the composition of distillate and bottom was predicted
by Aspen simulation. An appropriate ANN model was fitted using generated data.
A three-layered feed-forward back-propagation model, with 3, 5 and 4 neurons in
the layers, was fitted. Log-sig function was used in each layer. The model was found
to have high R2 (= 0.993) value and good predicting capability. |
|
|
|
|
|
|
|
Description |
|
2-Methylpropylacetate is an industrial solvent widely used in the manufacture
of paints and coatings as well as in many other branches of the chemical
industry. Industrial synthesis of this ester is usually carried out in an equilibrium
reactor, with the reaction products (water and 2-methylpropylacetate) and the rest of
the reactants (acetic acid and 2-methylpropanol) being separated in a system of
several rectification columns. Reactive Distillation (RD) can be used for such type
of reactions (Hanika et al., 1999; and Quido et al., 2001).
Designing an RD column and maintaining an optimal operating condition
are complicated because of the nonlinear interactions between the operating
input and output variables. Methods such as fuzzy logic, Artificial Neural
Network (ANN), and neuro-fuzzy are generally used to construct the knowledge
database for crude distillation units (Leo et
al., 2004). However, almost no work has been found which tests the utility of ANN in RD. Hence, it was thought desirable
to work on ANN for RD. In this paper, attempts are made to test the utility of
ANN for the synthesis of 2-methylpropylacetate from 2-methylpropanol and acetic
acid by RD. |
|
|
|
|
|
|
|
Keywords |
|
Chemical Engineering Journal, Artificial Neural Network (ANN), Aspen, Reactive Distillation (RD),
Methylpropylacetate, Reflux Ratio, RR, Control
Display Unit, CDU, Iso-Butyl Alcohol, IBOH, Mean Square Error, MSE, Mean
Relative Error, MRE, Nonlinear Process Identification, Chemical Equilibrium.
|
|
|
|
|
|
|