Hydrocarbon Processing - March 2022 - 88

Water Management
TABLE 6. Summary of results
Case description
Base Case hydraulic balance (TABLE 2)
Modified Base Case hydraulic balance with pressure control (FIG. 4)
Modified Base Case hydraulic balance with auxiliary pump (FIG. 5)
Cascaded cooling water network (FIG. 6)
Cascaded cooling water network with auxiliary pump (FIG. 8)
Cascaded cooling water network with pressure control (FIG. 9)
Therefore, it is recommended that a prudent design should
cater for a nominal excess (~5%-10%) of the total flow in the
cascaded header.
In the final analysis, only the thermal design and exchanger
geometry for heat exchanger E-6 were changed for a lower cooling
water rate. Also, at the lower cooling water rate, the hightemperature
rise (11.6°C) required the cooling water from this
exchanger to flow directly into the CWR header. To compensate
for this change, heat exchangers E-19 and E-20 were reassigned
as primary users, as these would benefit from a slightly
higher LMTD. The final configuration of the network model
was reconciled with the software used for the thermal design
(HTRI). The workflow was an iterative process until the pressure
drops and flowrates estimated in the network model were
within 1% of the performance estimated using the thermal design
softwarea
. A schematic of the workflow is shown in FIG. 7.
An optimized version of the cascaded cooling water balance
(TABLE 4) based on the use of an auxiliary pump for supply
of cooling water to the two elevated heat exchangers E-5
and E-6, along with an accompanying process schematic are
presented in FIG. 8.
For purposes of comparing the energy consumption, a
similar balance based on using an auxiliary pressure controlled
header (TABLE 5) and its accompanying process schematic
(FIG. 9) are also presented.
Summary of results. The analyses presented here bring into
sharp focus the need for considering cascaded cooling water
networks as a means for lowering the cooling water circulation
rate and reducing energy consumption. However, while this
may have a marginally positive impact on the initial CAPEX, it
is important to note that cascading the cooling water may not be
suitable for all networks. In the case study presented here, the
cooling water return temperature of 39.9°C was well within the
specified constraint of 46°C. Some systems may result in return
temperatures that could have an increase in fouling and corrosion,
and the decreased LMTD could increase the required heat
transfer surface area.
TABLE 6 shows a summary of results, clearly illustrating that
a detailed hydraulic analysis of the cooling water network could
result in significant water and energy savings if it is considered
as part of basic engineering. Omitting this analysis may result
in using more cooling water and energy than necessary or, in
the worst cases, supplying excess cooling water to a few users at
the expense of failing to deliver sufficient cooling water to many
others. The hydraulic analysis should be integral with the process
design of the unit and, as such, should not be relegated to
88 MARCH 2022 | HydrocarbonProcessing.com
Cooling water required, kg/hr
kW
27,284,921
23,982,939
23,982,939
12,823.73
11,358,460
11,383,333
4,100
3,600
3,100
2,230
1,745
1,980
Energy usage
% of Base Case
100
87.8
75.6
54.4
42.6
48.3
the detailed engineering phase when it may not be cost-effective
to make configurational changes.
The advantages of utilizing an auxiliary network have been
demonstrated. In the Base Case comparisons, the installation
with the auxiliary pump uses 1000 kW less energy (a 24% reduction)
with an estimated annual savings of $500,000 in OPEX.
For the cascaded network comparisons, the installation with
the auxiliary pump uses 485 kW less energy (a 22% reduction)
with an estimated annual savings of $242,500. In either case, the
CAPEX for the auxiliary pump can be recovered within a fairly
short operating period (< 2 yr).
Comparison of the two auxiliary networks indicates that a
pressure-controlled network requires more energy than a network
without it: 500 kW for the Base Case networks (a 16%
increase) and 235 kW for the cascaded networks (a 13% increase).
For comparison purposes, the cost of energy usage
for 8,400 hr of operation/yr is conservatively set at a nominal
$500/kW ($0.06/kWh). Since energy costs vary from user to
user, readers can substitute energy costs specific to their locations
and cooling water balances to determine which of the auxiliary
networks is more cost effective.
NOTES
a
Applied Flow Technologies: Fathom Version 11
LITERATURE CITED
1
2
Jeppson, R. W., Analysis of flow in pipe networks, Ann Arbor Science, Ann Arbor,
Michigan, 1976.
Sun, J., X. Feng, Y. Wang, C. Den and K. H. Chu, Pump network optimization for a
cooling water system, Elsevier Ltd., January 2014.
HEMANT L. BHATIA is a Technical Director for process
engineering with Fluor Enterprises in Aliso Viejo, California. He
has more than 40 yr of experience in conceptual studies, frontend
process engineering and detailed process design of
refineries, chemicals and petrochemical plants. Mr. Bhatia holds
BS and MS degrees in chemical engineering from Montana State
University.
CHRIS GARCIA is a Senior Process Engineer with Fluor
Enterprises in Aliso Viejo, California. He has more than 15 yr of
experience in the oil, gas and chemicals industries, and is
experienced with hydrotreating, delayed coking, gas treating and
methanol synthesis technologies. Mr. Garcia graduated from the
University of Michigan-Ann Arbor with a B.S.E. degree in chemical
engineering.
KRISH K. NANGIA is a Process Technology Director/Senior Fellow
with Fluor Enterprises in Aliso Viejo, California. He has more than
40 yr of experience in petroleum refining, gas processing and
petrochemical projects. Dr. Nangia holds a BS degree in chemical
engineering from Delhi University and a PhD from McGill
University. He is a registered PE in California.
http://www.HydrocarbonProcessing.com
Hydrocarbon Processing - March 2022

Table of Contents for the Digital Edition of Hydrocarbon Processing - March 2022
