Scribd
Upload
15 views

Refinery Problem

RefineryProblem

Uploaded by

solo_sudhan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
15 views

Refinery Problem

RefineryProblem

Uploaded by

solo_sudhan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

13.

6 Refinery optimization
The petroleum industry is the major user of linear programming models. This
is a very small version of a typical application. Generally, the models used
will consist of thousands of constraints, linking together possibly more than
one oil refinery, giving a structured model as described in Section 4.1. The
application of linear programming in the petroleum industry is described by
Manne (1956).
13.6.1 Variables
In view of the many different sorts of variables in a model of this sort, it is
convenient to use mnemonic names in this description of the formulation. The
following variables are used to represent quantities of the materials (measured in
barrels):

CRA crude 1
CRB crude 2

LN light naphtha
MN medium naphtha
HN heavy naphtha
LO light oil
HO heavy oil
R residuum
LNRG light naphtha used to produce reformed gasoline
MNRG medium naphtha used to produce reformed gasoline
HNRG heavy naphtha used to produce reformed gasoline
RG reformed gasoline
LOCGO light oil used to produce cracked oil and cracked gasoline
HOCGO heavy oil used to produce cracked oil and cracked gasoline
CG cracked gasoline
CO cracked oil
LNPMF light naphtha used to produce premium motor fuel
LNRMF light naphtha used to produce regular motor fuel
MNPMF medium naphtha used to produce premium motor fuel
MNRMF medium naphtha used to produce regular motor fuel
HNPMF heavy naphtha used to produce premium motor fuel
HNRMF heavy naphtha used to produce regular motor fuel
RGPMF reformed gasoline used to produce premium motor fuel
RGRMF reformed gasoline used to produce regular motor fuel
CGPMF cracked gasoline used to produce premium motor fuel
CGRMF cracked gasoline used to produce regular motor fuel

LOJF light oil used to produce jet fuel


HOJF heavy oil used to produce jet fuel
RJF residuum used to produce jet fuel
COJF cracked oil used to produce jet fuel

RLBO residuum used to produce lube-oil


PMF premium motor fuel
RMF regular motor fuel
JF jet fuel
FO fuel oil
LBO lube-oil

There are 36 such variables.

13.6.2 Constraints

Availabilities
The limited availability of the crude oils gives simple upper bounding
constraints:
CRA ≤ 20 000,
CRB ≤ 30 000.

Capacities

The distillation capacity constraint is

CRA + CRB ≤ 45 000.

The reforming capacity constraint is

LNRG + MNRG + HNRG ≤ 10 000.

The cracking capacity constraint is

LOCGO + HOCGO ≤ 8000.

The stipulation concerning production of lube-oil gives the following lower


and upper bounding constraints:

LBO ≥ 500,
LBO ≤ 1000.
Continuities

The quantity of light naphtha produced depends on the quantities of the crude
oil used, taking into account the way in which each crude splits under
distillation. This gives

−0.1CRA − 0.15CRB + LN = 0.

Similar constraints exist for MN, HN, LO, HO and R.


The quantity of reformed gasoline produced depends on the quantities of the
naphthas used in the reforming process. This gives the constraint

−0.6 LNRG − 0.52 MNRG − 0.45 HNRG + RG = 0.


The quantities of cracked oil and cracked gasoline produced depend on the
quantities of light and heavy oil used. This gives the constraints

− 0.68 LOCGO − 0.75 HOCGO + CO = 0,


− 0.28 LOCGO − 0.2 HOCGO + CG = 0.

The quantity of lube-oil produced (and sold) is 0.5 times the quantity of
residuum used. This gives

−0.5RLBO + LBO = 0.

The quantities of light naphtha used for reforming and blending are equal to
the quantities available. This gives

−LN + LNRG + LNPMF + LNRMF = 0.

Similar constraints exist for MN and HN.


The quantities of light oil used for cracking and blending are equal to the
quantities available.
For the blending of fuel oil, the proportion of light oil is fixed at 10/18.
Therefore separate variables have not been introduced for this proportion as
it is determined by the variable LO. This gives

−LO + LOCGO + LOJF + 0.55 FO = 0.

Similar constraints exist for HO, CO and R, also involving fixed proportions
of FO, and for CG and RG.
The quantity of premium motor fuel produced is equal to the total quantity of
its ingredients. This gives

−LNPMF − MNPMF − HNPMF − RGPMF − CGPMF + PMF = 0.

Similar constraints exist for RMF and JF.


Premium motor fuel production must be at least 40% of regular motor fuel
production, giving
PMF − 0.4 RMF ≥ 0.
Qualities

It is necessary to stipulate that the octane number of premium motor fuel does
not drop below 94. This is done by the constraint

−90 LNPMF − 80 MNPMF − 70 HNPMF − 115 RGPMF


− 105 CGPMF + 94 PMF ≤ 0.

There is a similar constraint for RMF.


For jet fuel we have the constraint imposed by vapour pressure. This is

−LOJF − 0.6 HOJF − 1.5 COJF − 0.05 RJF + JF ≥ 0.

This model has 29 constraints together with simple bounds on three


variables.
Some comment should be made concerning the blending of fuel oil where
the ingredients (light, heavy, and cracked oil and residuum) are taken in fixed
proportions. Here it might be preferable to think of the production of FO as an
activity . It is common in the oil industry to think in terms of activities rather
than quantities. In Section 3.4, modal formulations are discussed where activities
represent the extreme modes of operation of a process. Here we have a special
case of a process with one mode of operation. The level of this activity then fixes
the proportions of the ingredients, in a case such as this, automatically.

13.6.3 Objective
The only variables involving a profit (or cost) are the final products. This gives
an objective (in pounds) to be maximized of

7 PMF + 6 RMF + 4 JF + 3.5 FO + 1.5 LBO.

You might also like