(488e) A Hybrid Simulation-Optimization Approach for the Design of Internally Heat-Integrated Distillation Columns

A Hybrid
Simulation-Optimization Approach for the Design of Internally Heat-Integrated Distillation
Columns

Juan A. Reyes-Labarta*, Miguel
A. Navarro, José A. Caballero

Department of
Chemical Engineering, University of Alicante, Ap. Correos 99, Alicante 03080, Spain (e-mail: ja.reyes@ua.es)

Abstract

This work introduces a systematic
method for the optimal and rigorous design of internally heat-integrated distillation
columns (HIDiC). The total number of intermediate heat exchanger and their optimal
location are determined. The number of trays of each section can be optimised
in an outer loop providing the method a great flexibility. The commercial
process simulator Hysys® has been used to implement the process and solve the
rigorous VLE using the available thermodynamic models such as NRTL, and MatLab®
to implement the optimization algorithm.

Keywords:
Distillation, hybrid simulation-optimization, optimal processes design, HIDiC.

1. Introduction

Distillation is still one of the most
important separation techniques, even though it is an expensive operation in
terms of capital and energy costs. This fact with the major current environmental
problems such as the global warming and depleting resources have led to growing
interest in more efficient and sustainable chemical processes. In these sense, since
the distillation columns normally present sections with different heat
requirements (to vaporize a liquid in the bottom and to condensate a vapour in
the top), heat-integrated distillation columns (HIDiC) can be an efficient way
of saving energy [1-3].

HIDiC columns are characterised by the
existence of internal heat integration between the rectifying section and the
stripping section. In order to allow the heat transfer from the rectifying to
the stripping section, the temperature of the rectifying has to be increased
with a compressor. Normally, the stripping-section pressure is the same as the
feed pressure, while the rectifying-section pressure should be elevated enough
that the different between the dew-point temperature of the distillate and the
bubble point of the bottoms product were significant (larger than 10 ºC).

In this kind of designs, once we have
defined the operating pressures, the main variables to get the best column
configuration for a specified separation are: the number of theoretical stages
of each section (n_RS, n_SS), reboiler duty (Q_R),
distillate flow rate (D), number and location of the intermediate heat
exchangers, overall internal heat transfer (Q_IHT) and compressor
shaft work (W_S).

Due to the inclusion of very efficient
simulation algorithms in commercial packages the design of distillation columns
has been usually performed by successive simulations [4,5]. However, this trial-and-error
procedure is very time consuming and of doubtful utility when trying to
evaluate complex configurations or sequences of interrelated columns.

On the other hand, the use of new
design methodologies based on superstructure models and mathematical
programming (GDP or MINLP techniques) present difficult convergence due to the
high non-linearity of the equations used (especially when rigorous equilibrium
calculations are modelled). Additionally, these methodologies normally have a strong
dependence of the initial values used to start the optimization calculations and
need the utilization of good initialization procedures and tight bounds on the
variables to convergence [6-8].

Thus, in the present paper, we present
a hybrid simulation-optimization approach for the optimal and rigorous design
of heat integrated distillation columns. This approach allows easily analyse
and quantify the effect of the presence of intermediate heat exchangers over
the overall heat duty, the total annualized cost or even over the environmental
impact, by combining the
capabilities of process simulation packages with optimization tools and life
cycle assessment (LCA) [9,10].

2. System description

Figure 1 represents the general
configuration of a HIDiC. The basic components are the rectifying section (RS),
the stripping section (SS), the compressor (Comp), the reboiler (R), the
condenser (cond), the expansion valve (EV) and the intermediate heat exchanger
(IHE). An additional pre-heater (p-H) is also show in Figure 1, just before the
compressor, to avoid the possibility of presence of a liquid phase in the
compressor. This pre-heated is controller by the hysys adjustment tool (A) that
introduce the minimum amount of heat in order to have a saturated vapour in the
stream leaving the compressor.

Figure 1. General configuration
of a internally heat-integrated column.

The system operation at two pressure
levels is as follows. The feed steam (F) is introduced in the top of the
stripping section. The distillate (D) and residue (R) streams are the products
living its corresponding section. The vapour stream leaving the stripping
section (V_1,SS) is compressed before to be introduced in the
rectifying section. As commented before, the stream V_1,SS could be
preheated in order to avoid the presence of any liquid phase in the compressor.
The liquid stream leaving the rectifying section (L_n,RS) is expanded
before to be introduced in the stripping section. The overall internal heat
transfer (Q_IHT) is exchanged between the two rectifying and
stripping section by the IHE equipment.

3. Results

In order to show an example of the
advantages that the proposed approach offers, Figure 2 shows the results of a
sensitivity analysis where, for a defined configuration of a HIDiC (i.e.
concrete number of trays and pressures in both sections,) the overall internal
heat transfer (Q_IHT) has been varying from 0 to 2000 kW.

Figure 2. Condenser and reboiler duties, and compressor shaft work vs overall
internal heat transfer (Q_IHT).

As we can observe in Figure 2, the
existence of only one internal heat exchange between both sections can produce
interesting reduction in the total energy demand and therefore significant cost
saving, without being necessary heat exchange between each tray of both
sections. This results are consistent with those proposed by different authors
[2,11,12].

4. Conclusions

A hybrid simulation-optimization design
method has been suggested and applied to analyse the advantages and benefits of
using internal heat exchangers in the distillation columns. The effect of the
overall internal heat transfer on the energy requirements has studied for a
defined HIDiC configuration. This procedure is based on the robustness of the
commercial simulators.

The proposed approach can also be
applied to vapour recompression columns (VRC) and where

Acknowledgements: The authors would like to acknowledge financial support from the Spanish
Ministerio de Ciencias e Innovación (PPQ, CTQ2009-14420-C02-02).

References

1.       
Nakaiwa M, Huang MK, Endo A, Ohmori T, Akiya T,
Takamatsu T. Internally heat integrated distillation columns: A review. Chem. Eng. Res. Des. 2003;81:162-177.

2.       
Shenvi AA, Herron DM, Agrawal R. Energy
efficiency limitations of the conventional heat integrated distillation column
(HIDiC) configuration for binary distillation. Ind. Eng. Chem. Res.
2011;50:119-130.

3.       
Ho TJ, Huang CT, Lee LS, Chen CT. Extended
Ponchon-Savarit method for graphically analyzing and designing internally
heat-integrated distillation columns. Ind. Eng. Chem. Res. 2010;49:350-358.

4.       
Douglas J.M. Conceptual
Design of Chemical Processes. McGraw-Hill. 1988.

5.       
Doherty M.F.; Malone M.F. Conceptual Design of
Distillation Systems. McGraw-Hill. 2001.

6.       
Caballero JA, Milán-Yañez D, Grossmann IE.
Rigorous design of distillation columns: Integration of disjunctive programming
and process simulators. Ind. Eng. Chem. Res. 2005;44(17):6760-75.

7.       
Reyes-Labarta J.A., Grossmann I.E. Disjunctive
optimisation design models for complex liquid-liquid multistage extractors.
AIChE J. 2001; 47(10): 2243-2252.

8.       
Reyes-Labarta JA, Caballero JA, Marcilla A.
novel hybrid simulation-optimization approach for the optimal design of
multicomponent distillation columns. European Symposium of Computer Aided
Process Engineering ESCAPE22 (Computer Aided Chemical Engineering) 2012,
accepted (in press).

9.       
Guillén-Gosálbez G, Caballero JA, Jiménez L.
Application of life cycle assessment to the structural optimization of process
flowsheets. Ind. Eng. Chem. Res. 2008;47(3):777-89.

10.     Brunet R, Reyes-Labarta JA, Boer D, Guillén-Gosálbez G, Jiménez L.
Integrating Process Simulation, Multi-Objective Optimization and LCA for the
Design of Absorption Systems. Computers and Chemical Engineering 2012 (in
revision).

11.     Harwardt A, Kraemer K, Marquardt W. Identifying optimal mixture
properties for HIDiC application. Proceedings Distillation and Absorption 2010,
55-60.

12.     Suphanit B. Optimal heat distribution in the internally
heat-integrated distillation column (HIDiC). Energy 2011;36:4171-4181.

Nomenclature

GDP                        General
Disjunctive Programming

MINLP                   Mixed-integer
non-linear programming