Completed Report ENGG815
Completed Report ENGG815
Completed Report ENGG815
Present to
By
Faculty of Engineering
WINTER 2006
TABLE OF CONTENTS
CHAPTER PAGE
I ABSTRACT 1
II INTRODUCTION 2
III LITERATURE REVIEW 3
IV DESCRIPTION OF THE DEMETHANIZING SYSTEM 4
V METHODOLOGY 5
VI NUMERICAL METHOD 6
VII RESULT AND DISCUSSION 7
VIII CONCLUSION 8
IX REFERENCE 9
I. ABSTRACT
The demethanizing process has been used widely in the industrials to separate methane
out of any other hydrocarbon components. Basically, two phases, vapor and liquid phases, of
feed stream, which previously are separated by flash drum, flow to an upper and a lower tray of
demethanizer column, respectively. The vapor phase which mostly composed of methane,
nitrogen and some ethane are heated by the reboiler at the bottom of the demethanizer column.
Methane and nitrogen, which have the lowest boiling point, readily flow upward through the top
of the column, meanwhile the ethane which has a higher boiling point as well as the rest of
hydrocarbon components at a lower tray flow downward and so far they are withdrawn out of the
bottom of demethanizer column. This paper is focused on the behavior of temperature profile
along the demethanizer column. The calculated boundary conditions at the steady state using
PRO/II 7.0 are demonstrated. Further, for temperature profile calculation, the transformation
from partial differential equation to the algebraic difference equation based on the central
difference approximation is applied on this particular problem. As the results, the calculated
solutions are in a good agreement with the analytical solutions using PRO/II 7.0.
II. INTRODUCTION
The purpose of this term project is to employ the algebraic numerical method to solve the
scientific problems instead of using calculus method to directly solve the partial differential
equation. In addition, the content of this course outlines is to bring the computer aided software,
for example MATLAB, FEMLAB and so on, to solve for the particular engineering problems
(Mahinpey, 2006). To appropriately fit with the course objectives, this term project has been
employed the PRO/II 7.0 (SIMSCI), and MATLAB as the software aided tools to calculate the
boundary conditions at the steady state of the demethanizer column and so far compute the
temperature profile along the demethanizer column based on the partial differential equation but
using algebraic difference equation by an approach of central difference approximation which,
further, it is solved by an approach of the Gauss-Seidel method.
The advantage of this studying helps to predict the behavior of the temperature profile at
the steady state by the degree of temperature inside the demethanizer column. In addition, The
advantage of this study is to help reduce the capital cost of the temperature sensors.
On the other hands, the model simulations have helped the scientists and researchers to
minimize error from the measurement and help to approximately predict the temperature profiles
along the column instead of installation of temperature sensors in various locations in column
(Luyben, 2006). There are several researches studying about the temperature profile model such
as using the Newton-Raphson methods to predict the temperature in the distillation column by
Nelson, 1971; Komatsu and Holland, 1977; Carra et al., 1979, using modified tri-diagonal
methods by Susuki et al., 1971; Izarraraz et al., 1980; Tierney and Riquelme, 1982; Xu and
For the demethanizing system which is one type of distillation processes as illustrated in
Figure 1, firstly hydrocarbon stream which mainly is composed of methane, ethane, propane, iso-
butane, n-butane, iso-pentane, n-pentane, n-hexane, n-heptane and N2 flows through the flash
drum to separate stream into two phases, the vapor phase and liquid phase. The vapor phase will
flow through the expander and then to an upper tray of the demethanizer column. Meanwhile, the
liquid phase will flow through the liquid valve and so far to a lower tray of column. The
demethanizer column in this study is no reflux which has no condenser unit at the top of the
column because of obviously different boiling points between methane and the rest of
hydrocarbon components. In the demethanizer column, the hydrocarbon liquid flowing down to
the bottom will be heated by reboiler and then move through the top of the column. Methane and
nitrogen which have the significant lowest boiling point will flow out of the top of the column,
but the rest of hydrocarbon stream such as ethane, and propane will flow back to the column and
consequently are withdrawn out of the bottom of the column by reboiler.
V. METHODOLOGY
The steps to compute the temperature profile have been demonstrated in Figure 2. Firstly
the process flow diagram and the operating conditions as shown in Figure 1 and Table 1 are
drawn and put into PRO/II 7.0 to calculate for the temperatures at the boundary conditions in the
steady state. This study focuses on two cases of process flows, the first case by feeding the liquid
stream flowing out of liquid valve to the 3rd tray of column and the second case by feeding the
liquid stream flowing out of liquid valve to the 5th tray of column. The calculated boundary
conditions are put into the grids in Figure 3 which represent the dimension of demethanizer
column. Next, the partial differential equation is applied to solve for the unknown temperatures
in various locations inside the column. This employs algebraic difference approximation by an
approach of Gauss-Seidel method. The results are reported and so far they are graphically
plotted. Further, these calculated results will be compared with the analytical solutions using
PRO/II.
Stream Name
EXP_TO_DIST 2
EXPANDER
Temperature -174.277 F 3
Pressure 139.696 PSIA
Flowrate 1818.585 LB-MOL/HR 4
Phase Mixed DEMETHANIZER
Stream Name VAPOR
5
Temperature -84.000 F
Pressure 587.700 PSIA For Case 1
6
Flowrate 1818.585 LB-MOL/HR
Phase Vapor For Case 2 7
Stream Name
VALVE_TO_DIS 9
Figure 1 A scheme of demethanizing system with its operating conditions and calculated results
for case 1 and case 2
Operating
conditions PRO/II 7.0 Graphical plot
Boundary
BoundaryConditions
Conditionsatatthe
surface of the column
the steady-state End
End
Gauss-Seidel Method
No Yes
εa ≤ εs
Table 2 The categories of second order PDEs: Elliptic, Parabolic and Hyperbolic PDEs
Criterion Type of PDE Example Properties
B2-4AC < 0 Elliptic Laplace's equation Boundary value problem
∂ 2u ∂ 2u
+ =0
∂x 2 ∂y 2
2
B -4AC = 0 Parabolic Fourier's equation Mixed boundary value problem
∂ u2
∂u and initial value problem
α =C
∂x 2 ∂t
B2-4AC > 0 Hyperbolic Wave equation Mixed boundary value problem
∂ u 2
∂ u2 and initial value problem
c2 =
∂x 2 ∂t 2 or initial value problem
Case 1 Case 2
j=1 j=5 54 unknown temperatures j=1 j=5
i=1 i=1
Tray 1 (Ti, j = -169.912°F)i=1, 2; j=1, 5 except T2, 1 Tray 1 (Ti, j = -170.590°F)i=1, 2; j=1, 5 except T2, 1
-174.277°F -174.277°F
Tray 3 (Ti, j = -134.015°F)i=5, 6; j=1, 5 except T6, 1 Tray 3 (Ti, j = -148.612°F)i=5, 6; j=1, 5
-131.381°F
Tray 5 (Ti, j = -132.899°F)i=9,10; j=1,5 Tray 5 (Ti, j = -132.624°F)i=9,10; j=1,5 except T10, 1
-131.381°F
Tray 6 (Ti, j = -130.833°F)i=11, 12; j=1, 5 Tray 6 (Ti, j = -129.823°F)i=11, 12; j=1, 5
Tray 7 (Ti, j = -117.836°F)i=13, 14; j=1, 5 Tray 7 (Ti, j = -116.765°F)i=13, 14; j=1, 5
Tray 8 (Ti, j = -71.699°F)i=15, 16; j=1, 5 Tray 8 (Ti, j = -71.069°F)i=15, 16; j=1, 5
Tray 9 (Ti, j = -16.170°F)i=17, 18; j=1, 5 Tray 9 (Ti, j = -16.131°F)i=17, 18; j=1, 5
Figure 3 The grids for case 1 and case 2 with their boundary conditions and 54 unknown
temperatures in each column
Thus,
Ti+1, j − 2Ti, j + Ti−1, j Ti,J+1 − 2Ti, j + Ti, j−1
2
+ =0 (5)
∆x ∆y 2
By simplifying the equation by defining ∆x = ∆y , hence
Ti +1, j + Ti −1, j + Ti, j +1 + Ti, j −1 − 4Ti, j = 0 (6)
Therefore, the Eq.(6) can be solved by rearrange into the matrix form by
[ A ]{T} = {b}
Then, the transformation of boundary condition values is substituted into Eq.(6). The result has
been written as shown in Eq.(7).
Tray 1 Tray 1
Tray 2 Tray 2
Tray 3 Tray 3
Tray 4 Tray 4
Temperature (oF)
Temperature (oF)
Tray 5 Tray 5
Tray 6 Tray 6
Tray 7 Tray 7
Tray 8 Tray 8
Tray 9 Tray 9
Tray 10 Tray 10
Figure 4 The temperature profiles of demethanizer column for case 1 and case 2, respectively
Table 4 and Figure 7 summarize the comparisons between the calculated solutions and
the analytical solutions using PRO/II 7.0. As the results, the calculated solutions are in a good
agreement with the analytical solutions.
Table 6 The calculated solutions comparing with the analytical solutions using PRO/II 7.0
Tray Case 1 (oF) Case 2 (oF)
Position Analytical Calculated Analytical Calculated
|εt| (%) |εt| (%)
solution results solution results
1 -169.912 -164.997 2.892% -170.590 -166.158 2.598%
2 -152.956 -150.394 1.675% -156.362 -155.419 0.603%
3 -134.015 -134.909 0.667% -148.612 -148.111 0.337%
4 -133.258 -133.290 0.024% -144.446 -142.367 1.439%
5 -132.899 -131.514 1.042% -132.624 -131.768 0.646%
6 -130.833 -126.008 3.688% -129.823 -125.282 3.498%
7 -117.836 -107.474 8.794% -116.765 -106.580 8.722%
8 -71.699 -63.452 11.502% -71.069 -62.923 11.461%
9 -16.170 -12.724 21.313% -16.131 -12.690 21.330%
10 27.085 27.085 0.000% 26.834 26.834 0.000%
Avarege of true error 5.160% 5.063%
Note the calculated results are based on the average temperature at the same row but different column.
Temperature ( F)
o
-50
-100
-150
-200
1 2 3 4 5 6 7 8 9 10
Tray number in the column
Figure 5 The comparisons between calculated results and analytical solution using PRO/II 7.0
VIII. CONCLUSION
By means of partial differential equation (PDE) and algebraic difference approximation
by an approach of Gauss-Seidel method, the mathematical model provides the results that have a
good agreement with the results obtained from analytical solutions using PRO/II 7.0. The results
obtained from the mathematical model have an approximate average true error of five percents.
IX. REFERENCES
1. Graduate course materials for ENGG 815, Dr. Nader Mahinpey, 2006.
2. S. C. Chapra and R. P. Canale, Numerical methods for engineers with programming and
software application, 1998, 3rd Edition, McGraw-Hill International Editions.
3. W. L. McCabe, J. C. Smith, P. Harriott, Unit operations of chemical engineering, 1993, 5th
Edition, McGraw-Hill International Editions.
4. J.M. Smith, H.C. Van Ness, H.M. Abbott, Introduction to chemical engineering
thermodynamics, 1999, 5th ed., McGraw-Hill International Editions.
5. PRO/II 7.0, SimSci-Esscor, http://www.simsci-esscor.com
6. MATLAB, The MathWorks, http://www.mathworks.com
7. W.L. Luyben, Evaluation of criteria for selecting temperature control trays in distillation
columns, 16 (2006) 115–134.