Improvement of Overall Efficiency in the Gas Transmission Networks: Employing Energy Recovery Systems

Authors

Department of Chemical Engineering, Reservoir Engineering, Bahonar University, Kerman, Iran

Abstract

This study mainly focuses on enhancing the overall efficiency of gas transmission networks. The authors developed a model with detailed characteristics of compressor and pressure reduction stations. Following this, they suggested three different systems with gas turbine including: organic rankine cycle (ORC), air bottoming cycle (ABC), and ABC along with steam injection (SI-ABC). In addition, use of turbo-expander as a good alternative to the expansion valves was also studied. Performance of proposed cycles was investigated on a real case study for natural gas transmission network. Results showed that the highest efficiency would be obtained with ORC system where n-pentane was used as the working fluid and turbo-expander was considered in pressure reduction stations Moreover, the efficiency improvement was estimated to be 22% averagely in comparison with the existing network where the flow was in the span of 50 up to 90 MMSCMD.

Keywords

Main Subjects


 

1. Introduction
Natural  gas  is  one  of  the  most  widely  used
sources  of  energy  in  the  world  due  to  its
environmental  issues  and  global  warming
concerns. As such, the demand for natural gas,
as  a  primary  energy,  are  increasing  more  and
more.  It  is  predicted  that  the  natural  gas
demands  increase  at  an  average  rate  of  2.4
percent  annually  until  2030  in  the  world
(Najibi  &  Taghavi,  2011).  Normally,  the
locations  of  natural  gas  resources  and  end
points for various applications are far apart. As
a  result,  it  must  be  transformed  from  the
deposit  and  production  sites  to  the  consumers
through  the  pipeline  networks.  Increasing  the
natural  gas  demands  and  pipeline  networks
requires  focus  on  the  overall  efficiency  of
transmission. Currently, it is low because much
of  energy  is  wasted  in  the  transmission
network;  these  energies  are  recoverable.
Consequently,  the  development  of  high
efficiency  gas  transmission  network  is  a  key

issue  in  order  to  satisfy  the  growing  demand
from the various customers (Guo & Ghalambor,
2014).
In the transmission network, gas flows through
pipes  and  various  devices  such  as  regulators,
valves,  and  compressors.  The  gas  pressure  is
reduced  mainly  due  to  wall  friction  and  also
heat  transfer  between  the  gas  and  the
surroundings.  Hence,  compressor  stations
should  be  installed  to  boost  the  gas  pressure
and keep it moving to the desired destinations.
Gas  compression  is  usually  performed  using
the  centrifugal  or  reciprocating  compressors
driven  by  the  gas  turbines  or  the  electrical
motors.  Gas  turbines  are  more  common
because  of  several  reasons.  First  of  all,  their
operating  cost  is  lower  than  the  electrical
motors.  Secondly,  installation  cost  of  a  new
interconnecting  electric  power  transmission
line may be high. Finally, it may be difficult to
obtain  the  necessary  regulatory  approvals  for
the  electrical  motors.  However,  it  is  estimated

that  3-5%  of  the  transported  gases  are
consumed  by  the  gas  turbines  as  fuel  (Borraz-Sánchez  &  Ríos-Mercado,  2005;  Wu,  Rios-Mercado, Boyd, & Scott, 2000) and the exhaust
gases from turbines with high temperature are
released  into  the  environment  which  leads  to
the serious environmental pollution(Dai, Wang,
& Gao, 2009).
After  passing  all  compression  stations,  the  gas
pressure  must  be  reduced  at  distribution
points.  Currently,  most  pressure  reduction
stations  use  expansion  valves,  which  may
waste  lots  of  energy  associated  with  high
pressure  gas.  Hence,  energy  recovery  from
pressure  compression  and  reduction  stations
can  be  regarded  as  one  of  the  focal  points  to
improve  the  overall  efficiency  of  gas
transmission network.
Edgar  et  al.  (Edgar,  Himmelblau,  &  Bickel,
1978),  Cobos-Zaleta  et  al.  (Zaleta  &  Ríos-Mercado,  2002),  Ríos-Mercado  et  al.  (Ríos-Mercado,  Kim,  &  Boyd,  2006),  Kabirian  et  al.
(Kabirian  &  Hemmati,  2007)  and  Safarian  et
al.  (Safarian,  Saboohi,  &  Kateb,  2013)
presented  different  models  or  various
procedures  to  optimize  the  gas  transmission
network.  The  objective  of  their  models  was  to
minimize  the  cost  or  energy  consumption  of
transmission  system.  The  effect  of  pressure
reduction  station  and  the  function  of  energy
recovery  from  current  gas  transmission
network were not considered in their work.
In  the  current  study,  minimizing  the  energy
consumption  and  maximizing  the  flow  rate
through  the  pipes  were  the  basic  issues of  this
study  and  the  main  objective  of  this  research
was  to  maximize  energy  recovery  and  excess
power by utilizing different technologies on the
compression and reduction stations. Evaluation
of various technologies such as turbo-expander,
organic  rankine  cycle,  air  bottoming  cycle,  and
steam  air  bottoming  were  conducted  in  the
literature  (Cho,  Cho,  &  Kim,  2008;  Maddaloni
&  Rowe,  2007);  It  is  worth  mention  that,  this
research  study,  investigating  all  these systems
together,  presents  a  comprehensive  model  to
compare  them.  Following  the  most  efficient
system can be identified.
In 2007, Maddaloni and Rowe investigated the
application  of  turbo  expander  in  the  gas
pressure  reduction  station  to  produce
electricity.  The  electricity  could  either  be
routed back into the electric distribution grid or
used  to  produce  small  amounts  of  hydrogen.
They  found  that  at  their  assumed  peak
efficiencies,  electricity  could  be  extracted  from
the  pressure  reduction  system  with  75%
exegetic  efficiency,  and  hydrogen  could  be
produced with 45% energetic efficiency.
In  the  compression  stations,  one  common
solution for increasing the performance of a gas
turbine  is  to  combine  it  with  a  steam  cycle  or
with an organic rankine cycle. This can be used
either  to  generate  electricity  alone,  or  to  co-generate  both  electric  power  and  heat  for
industrial  and  home  purposes  (Safarian  &
Bararzadeh, 2012).
Due to lower vaporization heat, organic fluids
are preferred to water; this is critical especially
when  the  available  power  and  the  heat  source
temperature  are  low.  In  this  situation,  these
fluids  can  better  follow  the  heat  source  to  be
cooled;  this  reduces  the  temperature
differences  and  therefore,  irreversibilities  at
the  evaporator.  Furthermore,  the  turbines  for
organic cycles could provide higher efficiencies
at partial loads. In addition their complexity is
usually  less  due  to  the  lower  enthalpy  drop  of
the  fluid(Drescher  &  Brüggemann,  2007;
Larjola, 1995).
In 2009, Desai et al. (Desai & Bandyopadhyay,
2009)  proposed  a  methodology  for  accurate
optimization  of  an  ORC  as  a  co-generation
process  to  generate  shaft-work,  with16
different  organic  fluids.  In  addition,  they
investigated  the  benefits  of  integrating  ORC
with  the  background  process  and  reported  on
the  applicability  of  proposed  methodology  with
illustrative examples.
In  2010,  Roy  et  al.  (Roy,  Mishra,  &  Misra,
2010)  performed  parametric  optimization  and
performance  analysis  of  a  waste  heat  recovery
system  along  with  Organic  Rankine  Cycle  to
generate  the  power.  This  analysis  was
performed  for  R-12,  R-123,  and  R-134a  as
working fluids. They also found that R-123 had
the  maximum  work  output  and  efficiency
among all the selected fluids.
Combining  the  gas  turbine  cycle  with  an  air
bottoming  cycle  (ABC)  is  another  method  that
has been found to increase the performance of a
gas  turbine  (Korobitsyn,  1998).  In1995,
Kambanis  (Kambanis,  1995)  and  in  1996,
Bolland  (Bolland,  Forde,  &  Hande,  1996),
reported  that  by  using  the  exhaust  gas  of  a
simple  gas  turbine  in  the    air  bottoming  cycle
the  efficiency  of  the  combined  cycle  improved
about 47% and 46.6% respectively.
In  the  recent  study,  Ghazikhani  et  al.
(Ghazikhani,  Passandideh-Fard,  &  Mousavi,
2011) developed a model for the steam injection
in  the  gas  turbine  with  air  bottoming  cycle.
They  also  found  two  new  cycles  with  ABC.
These cycles were: the Evaporating Gas turbine
with  Air  Bottoming  Cycle  (EGT-ABC),  and
Steam  Injection  Gas  turbine  with  Air
Bottoming  Cycle  (STIG-ABC).  Their  findings
indicated  that  EGT-ABC  had  a  lower

irreversibility  and  higher  output  compared  to
the STIG-ABC.
In  this  article,  the  authors  developed  a  model
for  the  technical  analysis  of  transmission
network  considering  characteristics  for
pressure  compression  and  reduction  stations
and  the  energy  recovery  technologies.  The
developed  model  estimated  net  possible  output
power, overall efficiency and system energy loss
to  evaluate  the  performance  of  the  gas
transport  networks.  It  is  should  be  noted  that,
the  model  utilized  equations  of  real  gases  for
estimation  of  enthalpy  which  produced  accurate
results no longer needing Moulier graphs.
2. Methodology
2.1. Definition of Transportation Efficiency
Transportation  efficiency  can  be  regarded  as  a
function  of  the  overall  system  design,  the
efficiency  of  individual  components,  and  the
way that the system is operated. Transportation
efficiency  is  defined  as  the  amount  of  fuel
burned  or  electric  power  used  per  unit  of  the
throughput  (i.e.,  British  thermal  unit  (Btu)  or
kW/Mcf).  In addition to this general definition,
there are three other related measures.
1. Hydraulic efficiency is a measure of the loss
of  energy  (pressure  drop)  caused  by  the
friction  of  the  flowing  gas  in  the  pipeline
facilities.
2. Thermal efficiency applied to a prime mover
(engine,  turbine  or  motor),  t  measures  as
fraction  of  the  potential  energy  of  an  input
fuel  or  electric  power  which  is  converted
into  useful  energy;  this  energy  can  be  used
to drive a compressor. The amount of energy
that  is  not  converted  into  useful  energy  is
considered as “waste heat” in the exhaust.
3. Compressor  efficiency  measures  how  much
energy  is  expended  in  compressing  the  gas
compared in comparison with overall energy
used  by  the  compressor.  Inefficient
compressors  heat  the  gas  instead  of  raising
its  pressure  and  thus  have  lower  efficiency
values.
2.2. Proposed Model
The  model  was  developed  for  the  existing  gas
pipelines  networks  from  supply  to  demand
nodes. The demand nodes were major locations
of  natural  gas  consuming  in  the  study  area.
The  demand  nodes  were  either  consumption
regions in the study area or export terminals of
natural gas from the study area to the outside.
In  contrast,  supply  nodes  were  resources
locations  for  natural  gas  processing  in  the
study  area.  These  nodes  included  either
refineries, natural gas producing plants, or the
import  terminals  of  natural  gas  from  the
outside of the study area (Safarian et al., 2013).
In  addition  the  model  considered  all  network
units  i.e.  gas  compressor,  air  compressor,  gas
turbine, combustion chamber, expansion valves
and  heater  in  the  compressor  stations,  and
pressure reduction stations.
The  model  consisted  of  four  sub-models  that
are defined below:
1. Base scenario,
2. Reduction stations along with turbo-expander,
3. Compression stations affixed to ORC,
4. Compression  stations  affixed  to  ABC  and
SI-ABC
The  main  thermodynamic  assumptions  that
were used in the  present analysis are reported
in  Table  1.Second  Iran  gas  transmission
network  was  chosen  as  the  case  study  in  this
work.  This  network  with  7  major  compression
stations was one of the most important network
systems in Iran.

2.3. Base Scenario
In  this  scenario,  gas  transmission  network  is
considered  without  energy  recovery  and  it
includes  simple  pressure  compression  and
reduction  stations.  In  this  sub-model,  the  aim
is  to  minimize  the  energy  used  in  the  gas
compressors.  This  can  be  written  as  following
(O'Neill, Williard, Wilkins, & Pike, 1979):

 

Where  α  is  the  unitary  energy  price  ($/kW),
ηtherm  is  the  compressor  thermic  efficiency  and calculated by (Menon, 2005):
Wj  is  compressor  required  power  that  is

Where  T1,  P1  and  Z1  are  input  temperature,
pressure  and  compressibility  factor  andT2,  P2
and  Z2  are output ones, respectively. ηa  stands
for the compressor adiabatic efficiency, Q is gas
flow rate through the pipeline (MMSCMD) and
γ is the ratio of specific heats which assumed to
be constant.
According  to  Weymouth  equation  (equation  3),
based  on  the  input  conditions,  designed
maximum  and  minimum  pressure  and  type  of
pipe material, the maximum flow rate through
the  pipeline  must  be  calculated.  Considering
this,  the  demand  should  always  be  lower  than
maximum  flow  rate  to  satisfy  the  restrictions
for flow capacity.

Also,  at  each  end  point,  the  demand  must  be
guaranteed  at  a  minimal  pressure.  On  the
other  hand,  the  gas  transmission  company
cannot  take  gas  at  a  pressure  higher  than
predefined  value.  Mathematically  (O'Neill  et
al., 1979):

Fig.  1  shows  a  schematic  of  gas  compression
station.  The  inlet  air  enters  the  compressor  at
state1.  Considering  an  isentropic  efficiency  of
ηcomp  for  the  compressor  and  a  constant
pressure ratio of rc that can be calculated as:

Because  of  high  pressure  of  the  inlet  natural
gas,  the  reduction  valve  must  be  used.  At  the
next  step,  compressed  air  combusts  with
medium  pressure  fuel.  Consequently,  exhaust
gases exit from combustion chamber and enters
the turbine. At final state, turbine output work
is  obtained  which  depends  on  exhaust
temperature  (TIT)  and  mass  flow.  Turbine
output power could be estimated from equation
2  but,  the  outcome  of  this  equation  was  not
accurate.  To  have  accurate  value  of  outcome,
equation  of  states  for  real  gases  was  used  in
our  model  as  follows  (Abbott,  Smith,  &  Van
Ness, 2001)

Every  component  has  unique  value  of  ;  for
example this  is -74920 (kJ/kgmol) for methane.
Second part of equation 9, can be simplified to:

Table  2,  shows  all  constants  for  calculation  of
methane specific heat. Last part of equation 9,
is called residual enthalpy which appears when
natural gas is assumed to be real not ideal, and
also gas pressure is more than atmospheric.

Where  Z,  Tr  and  w  are  gas  compressibility
factor,  reduced  temperature  and  acentric
factor, respectively.

In  the  same  way,  the  power  for  other  units  in
compression station is shown below:

The  other  parameter  which  is  different  in  real
and  ideal  gasses  is  entropy.  Regarding  this,
entropy for real gas can be written as:

 
 
is  the  standard  entropy  of  ideal  gas  that  is
constant  for  each  component  and  it  is  183.48
(kJ/kgmol.K).for methane.

Passing all compression stations, near demand
nodes, the gas enters to gas pressure reduction
stations. The process that currently performs in
reduction  stations  is  shown  in  Fig.  2.  At  the
first stage gas enters to the heater to make up
reducing temperature during expansion process
and then it passes under the constant enthalpy
process in Joule-Thomson valves.
The  model  estimates  dissipation  rate  of
pressure  energy  in  the  expansion  valves,
required  amount  of  heat  and  total  loss  of
energy  in  the  reduction  stations.  The  result  of
this  part  provides  us  a  general  overview  about
the amount of energy which is being wasted in
gas  pressure  reduction  stations.  This  can  be
written:

2.4.  Affix  of  turbo-expander  to
reduction stations
In  this  part,  the  authors  considered  using
turbo-expander  instead  of  J-T  valves.  For  this
purpose, they develop model to calculate output
power  and  required  heat  of  turbo-expander
system.  Fig.  3  shows  a  simple  schematic  of  an
improved station.

Equations  9-15,  19  and  20  are  the  base
equations  in  this  part.  Inlet  conditions  of  heat
exchanger  and  outlet  conditions  from  turbo
expander  are  key  inputs.  Considering  an
isentropic efficiency of ηexp for the expander and
pressure drop of 1.46(%) during heater.
The  most  important  operational  challenge  in
the turbo-expander is hydrate formation due to
the  slight  amount  of  water  in  the  gas.  Two
factors which are intensified hydrate formation
are the low temperature and high pressure. So
a proper temperature for outlet heat exchanger
should  be  found  out  considering  this
phenomenon. At the  first step, the expander is
considered as isentropic process and then using
trial and error for T2 and equation of expander
efficiency,  the  model  estimates  the  correct
answer for T2.

The  output  power  and  required  heat  can  be
obtained as follows: Having considered the heat
exchanger  pressure  drop,  turbo-expander

efficiency,  fuel  mass  flow  rate  and  generator
and gearbox efficiency, we can have:

The  thermodynamic  properties  of  natural  gas
are  used  in  the  model  and  the  procedure  of
model  solution  is  given  in  flow  chart1  in  the
Appendix.
2.4.1. Validation
To  validate  the  first  sub-model,  the  results  of
the  model  were  compared  to  those  of  the
experiments  performed  by  Pozivil  (Poživil,
2004) (Fig 4) and also to the simulation results
(Fig  5  and  6).  For  the  output  power  and  heat
duty  against  temperature  inlet  expander  and
inlet flow of natural gas, a good agreement was
observed.  The  discrepancy  between  the  two
results is less than 6%.

2.4.2. Sensitivity Analysis
A sensitivity analysis was conducted in order to
better  understand  the  effect  of  key-parameters
of  process  performance.  In  this  analysis,  the
effect of an additional percentage of parameters
P1,  T1  and  flow  rate  on  the  output  power  and
heat  duty  were  investigated  using  the  model.
The expander outlet pressure and temperature
were 1825 kpa and 18ºC, respectively.  
Fig. 7 and 8 show the details of the sensitivity
analysis.  Two  figures  are  plotted  for  the
capacities of 37, 139.73 and 371 MSCMH.
As  seen  in  Fig.  7,  although  output  power  and
required  heat  had  a  direct  relationship  with
inlet  pressure  and  flow  when  inlet  pressure
was  close  to  outlet  pressure,  the  expansion
system efficiency will be greater due to the low
difference between two graphs.

It is evident in Fig. 8 that expander power and
heat  duty  were  more  sensitive  to  inlet
temperature.  Following  this,  with  using  boiler
in  maximum  load,  more  power  would  be
obtained  without  the  extra  cost.  In  addition,
there  was  a  specific  temperature  for  each  inlet
pressure where output power and required heat
were  equal;  on  the  other  hand  expansion
system efficiency is 100%.
2.5.  Affix  of  ORC  to  Compression
Stations
2.5.1. Thermodynamic Analysis of ORC
The  ORC  system  consisted  of  an  evaporator,
turbine,  condenser  and  pump.  It  could  be
classified into two groups according to the level
of  turbine  inlet  pressure,  including  supercritical
ORCs  and  sub-critical  ORCs  (Safarian  &
Aramoun, 2015). In the present study, the sub-critical ORCs were investigated.
As is shown in Fig. 9, the working fluid left the
condenser as saturated liquid (point1). Then, it
was compressed by the liquid pump to the sub-critical  pressure  (point2).  The  working  fluid
was  heated  in  the  evaporator  until  it  became
superheated  vapor  (point3).  In  this  research,
heating process to working fluid was considered
to be indirect. In other words, an inductor fluid
such  as  oil  over  took  heat  transfer  to  working
fluid.  The  reason  of  this  was  to  increase  the
security and management level of process. The
superheated vapor flowed into the turbine and
expanded  to  the  condensing  pressure  (point4),
and  then,  the  low  pressure  vapor  led  to  the
condenser and condensed by air. The condensed
working  fluid  flew  in to  the  receiver and  was
pumped  back  to  the  evaporator,  and  a  new
cycle began.
In  the  mentioned  cycle,  if  the  temperature  T4
was  considerably  higher  than  the  temperature
T1, it might be useful to implement an internal
heat exchanger (IHE) into the cycle as shown in
Fig.10.This  heat  exchanger  is  also  depicted  in
Figs.9 by the additional state points 4a and 2a.
The turbine exhausts flowed in to the internal
heat exchanger and cool in the heat exchanger
in  the  process  (4–4a)  by  transferring  heat  to
the  compressed  liquid  that  was  heated  in  the
process (2–2a)(Vaja & Gambarotta, 2010).

Each  process  in  the  ORC  can  be  described  as
follows:
Process  2  to  3:  This  was  the  heat  absorption
process  in  the  evaporator.  The  pressure  drop
due to evaporation was considered. The amount
of  heat  transferred  from  the  waste  heat  to  the
working fluid is (Wei, Lu, Lu, & Gu, 2007):

If  the  internal  heat  exchanger  is  added,  the
amount of heat transfer can be calculated by:

Process  3  to  4:  This  was  a  non-isentropic
expansion  process  in  the  turbine.  Ideally,  this
was  an  isentropic  process  3–4s.  However,  the
efficiency  of  the  energy  transformation  in  the
turbine  never  reached  100%,  and  the  state  of
the  working  fluid  at  the  turbine  outlet  is
indicated  by  state  point  4.  The  isentropic
efficiency of the turbine can be expressed as:

Process  4  to  1:  This  was  a  constant  pressure
exothermic process in the condenser.
Process  1  to  2:  This  was  a  non-isentropic
compression  process  in  the  liquid  pump.  The
isentropic  efficiency  of  the  pump  can  be
expressed as:

The thermal efficiency of the ORC is defined on
the basis of the first law of thermodynamics as
the  ratio  of  the  net  power  output  to  the  added
heat.

The  procedure  of  model  solution  is  given  in
Flow chart 2 in the Appendix.
2.5.2. Validation
To validate the model, the results of the model
were  compared  to  those  of  the  experiments
performed  by  Dai  et  al.  (Dai  et  al.,  2009).  Fig.
11  shows  a  comparison  of  model  results  with
these  experiments  for  3  different  working
fluids.  Fig.  11  shows  net  power  output  against
turbine  inlet  temperature  where  a  good
agreement  is  observed.  The  discrepancy
between  the  two  results  was  less  than  5%.
Although  as  the  turbine  inlet  temperature
increased,  the  net  output  power  for  ammonia
and  water  increased  correspondingly,  for  the
butane,  an  increase  in  turbine  inlet
temperature  led  to  a  reduction  in  net  output
power.  For  instance  when  TIT  varied  from  90
to  135  (C)  the  efficiency  of  ORC-butane
decreased  by  7.9  %  averagely,  although  it  was
increased by 7.2 % and 2.1 % for ammonia and
water, respectively. Consequently, for high TIT
ammonia  was  better  choice  among  these  three
working fluids.

2.5.3. Sensitivity Analysis
A  sensitivity  analysis  was  conducted  for  n-pentane  as  working  fluid  to  better  understand
the  effect  of  parameters  on  the  process
performance.  Fig.12  shows  the  effect  of  gas
turbine  outlet  temperature  on  output  power  of
n-pentane  turbine  and  ORCs  efficiency.  The
output  work  was  increased  by  increasing  flu
gas  temperature,  because  more  amount  of  n-pentane  could  be  evaporated  by  this  way.  But
the  ORCs  efficiency  was  decreased  due  to  the
increase in Q in equation 34.
As  seen  in  Fig.13  the  obtainable  work  was
grown  by  increment  of  flu  gas  rate,  because  of
increase  in  heat  absorption  process  in  the
evaporator.  The  net  output  power  was
enhanced  too.  In  addition,  Fig.13  shows  a
changeless  trend  for  ORCs  efficiency  because
the  ratio  of  increase  of  net  output  power  and
required heat were kept constant.

The  effect  of  n-pentane  turbine  inlet  pressure
on  net  output  power  and  ORCs  efficiency  at
constant inlet temperature (490C) is displayed
in  Fig.14.  The  net  output  power  and  ORCs
efficiency  were  augmented  by  increasing  of
inlet pressure.
 
2.6.  Affix  of  ABC  and  SI  -  ABC  to
Compression Stations
An  Air  Bottoming  Cycle  system  consists  of  an
air compressor, regenerator and turbine. Fig.15
shows  such  a  combined  cycle  in  which  the
exhaust of an existing, topping gas turbine was
sent  to  a  gas-air  heat  exchanger  which  heated
the air in the secondary gas turbine cycle.

ABC  was  proposed  in  the  late  1980s  as  an
alternative  for  the  conventional  steam
bottoming  cycle.  Nowadays,  this  cycle  was
considered  as  a  compact  and  simple  bottoming
cycle  in  the  various  applications:  as  an
upgrading  option  for  simple-cycle  gas  turbines
in the offshore industry, a hot-air co-generation
plant, and a heat recovery installation at high-temperature furnaces.
Fig.16 also shows Steam Injection Gas Turbine
with  Air  Bottoming  Cycle  (SI-ABC).  The
topping  exhaust  gases  had  high  temperature
after passing through the regenerator. Thermal
energy  of  these  gases  could  be  used  for
evaporating  of  water.  The  steam  was  then
mixed with ABC compressor discharged air in a
mixer  (Ghazikhani  et  al.,  2011).  The
evaporating  process  was  performed  in  the
HRSG.  The  temperature  of  the  exhaust  gas
was  decreased  in  about  120  C  by  generating
steam.  The  amount  of  the  injected  steam  per
unit  fuel  flow  is  5-6  (kg/kg-fuel)  (Nishada,
Takagi, & Kinoshita, 2005)
The model developed in this study included the
calculation  of  three  cycles:  simple  gas  turbine,
ABC,  and  SI-ABC.  Each  process  in  the  ABC
can be described as follows:
Process  1  to  2:  This  was  the  heat  absorption
process  in  the  regenerator.  The  pressure  drop
due to regenerator was considered. Enthalpy of
flu  gas  was  calculated  by  equations  9-15.
Equation  35  is  utilized  for  estimation  of  air
outlet temperature from regenerator.

Process  3  to  4:  The  inlet  air  entered  the
compressor  at  state  3.  The  compressor  inlet
power was calculated as:

Air  outlet  temperature  from  regenerator  was
the  function  of  outlet  enthalpy  which  can  be
estimated  by  energy  balance  equation  around
regenerator.

Process  5  to  6:  This  was  a  non-isentropic
expansion  process  in  the  turbine.  The
isentropic  efficiency  of  the  turbine  can  be
expressed as:

The  power  generated  by  the  turbine  can  be
given by:

SI-ABC  process  was  similar  to  ABC;  the
differences  between  the  ABC  gas  turbine  and
SI-ABC  were  the  HRSG  and  a  mixer  which
provided  steam  for  the  bottoming  cycle.  The
procedure is given in Flowchart 3 and 4 in the
Appendix for ABC and SI-ABC.
2.6.1. Validation
To validate the model, the results of the model
for ABC and SI-ABC were compared with those
of the experiments performed by Ghazikhani et
al.  (Ghazikhani  et  al.,  2011).  Fig.17  displays
variations  of  compression  station  overall
efficiency with ABC or SI-ABC against TIT. SI-ABC system had more output power than ABC
at the same bottoming cycle pressure ratio and
TIT. This was due to more heat recovery in the
regenerator  in  the  SI-ABC  cycles.  It  could
produce exhaust with a lower temperature and
more  inlet  mass  to  bottoming  turbine.  In
addition, form Fig.17a good agreement between
the model results and experimental data can be
observed.  The  discrepancy  between  the  two
results was less than 4%.
2.6.2. Sensitivity Analysis
The  effect  of  key  parameters  of  process  on  its
performance  was  evaluated  by  a  sensitivity
analysis.  In  Fig.18,  the  thermal  efficiency  of
the  ABCs  and  SI-ABCs  varied  against
bottoming  pressure  ratio.  The  thermal
efficiencies  of  steam  injection  system  were
higher than ABCs. Fig.19 shows, the efficiency
reduces  as  the  ambient  temperature  is
increased. In addition, in the ABC, the reduction
rate of efficiency with ambient temperature was
steeper.  In  ABCs  the  effectiveness  of  the  heat
recovery in the bottom-cycle was also decreased
by  increasing  ambient  temperature  due  to  a
smaller  difference  between  the  two  stream
temperatures. Fig.19 also shows the superiority
of  the  SI-ABC to have the highest efficiency in
different ambient temperatures.

3. Results and Discussion
The  proposed  model  was  implemented  for
second  Iranian  gas  pipeline  network.  Table  3
shows  the  properties  of  this  network.  In  this
network,  the  demand  levels  were  different  in
different  quarters  of  year.  So  it  must  be
considered  as  an  important  variable  which
should  be  satisfied.  In  this  regards,  the  model
was  evaluated  for  different  demand  levels.
Optimization of the required power  for the gas
compressors  in  various  demands  was  the  first
step  in  the  minimizing  of  the  total  power
consumption.  Then  the  model  used  the  energy
recovery  technologies  within  the  pressure
compression and reduction stations to estimate
the  improved  overall  efficiency  and  net  output
power.  Table  4  displays  the  optimization
results  for  simple  network  in  the  case  study.
The  demand  and  the  pressure  ratio  for  all  gas
turbines  was  considered  50  MMSCMD  and  14,
respectively.

Having  a  clear  comparison  between  the
mentioned  systems,  the  variation  of  gas
transmission network overall efficiency against
the demand, is shown in Fig.20. The comparison
included six systems:
1. Simple pipeline
2. Simple  pipeline  with  the  turbo-expander  in
reduction stations
3. Simple  pipeline  with  ABCs  in  compression
stations  and  turbo-expander  in  reduction
stations
4. Simple pipeline with SI-ABCs in compression
stations  and  turbo-expander  in  reduction
stations
5. Simple  pipeline  with  ORC-butane  in
compression stations and turbo-expander in
reduction stations
6. Simple  pipeline  with  ORC-pentane  in
compression stations and turbo-expander in
reduction stations
Fig.20  shows  that  the  overall  efficiency  of  the
equipped  transmission  network  by  ORC-pentane  and  turbo-expander  was  higher  than
other  systems  at  the  same  flow  rate.  These
technologies  increase  the  network  overall
efficiency 13-28 % in span of 50- 90 MMSCMD.
As  seen  in  the  Fig.20,  the  overall  efficiency  of
SI-ABCs was higher than ORC-butane at lower
65  MMSCMD.  Although  the  effect  of  turbo
expander  on  the  overall  efficiency  was  low  at
low  flow  rates,  it  had  great  impact  on  the
efficiency at cold season when the demand level
is high.
Fig.21  displays  the  increase  of  the  overall
efficiency  with bottoming pressure ratio (rb) in
the range of 5-15 for all systems. The flow rate
was  considered  70  MMSCMD.  As  seen  in
Fig.21,  the  overall  efficiency  of  the  ORC-pentane  was  more  sensitive  to  bottoming
pressure ratio than the other systems.

4. Conclusions
In this study, the effect of three technologies on
the  overall  efficiency  of  gas  transmission
network  was  investigated.  Based  on  an  energy
analysis a computer program was developed to
survey  improving  the  performance  of  gas
transmission system. The examined technologies
were organic rankine cycle, air bottoming cycle,
and  steam  injection  air  bottoming  cycle  which
were  used  in  the  pressure  compression  station
and  turbo-expander  which  was  utilized  within
the pressure reduction stations.
The main conclusions were:
1)  The  turbo-expander  outlet  powers  and
required  heat  had  a  direct  relationship  with
expander inlet pressure, temperature and flow-rate.  In  addition,  when  the  inlet  pressure
ranged  from  450  to  750  pisa  and  gas  flow  was
maximum  i.e.  866  MSCMH,  the  efficiency  of
the expander system was82-44%, respectively.
2)  The  SI-ABC  was  found  to  have  maximum
output  power  at  the  same  bottoming  cycle
pressure  ratio  and  turbine  inlet  temperature
(TIT) in comparison with ABS. This was due to
more  heat  recovery  in  the  regenerator  in  the
SI-ABC  cycle  that  resulted  a  lower  exhaust
temperature; and more inlet mass to bottoming
turbine causes a higher output work. Moreover,
the results displayed that the overall efficiency
was decreased as the ambient temperature had
been increased for ABC and SI-ABC.
3) In this study, an organic rankine cycle using
working  fluid  such  as  ammonia,  butane,  and
water  was  analyzed  and  the  results  were
compared  together.  When  turbine  inlet
temperature  varied  from  90  to  135  (C)  the
efficiency of ORCs with butane as working fluid
decreased  by  7.9  %  averagely,  although  it
increased by 7.2 % and 2.1 % for ammonia and
water, respectively.
4)  The  ORC-pentane  was  more  sensitive  to
variation of bottoming pressure ratio.
The  model  was  tested  for  Iran  second  gas
transmission network. The results showed that
the  highest  efficiency  is  obtained  from
implementation  of  ORC  with  n-pentane  as
working  fluid  and  turbo-expander  in  pressure
compression  and  reduction  stations,
respectively.  The  demand  or  gas  flow  rate
through  pipeline  was  the  most  effective
parameter that needed to satisfy. So, variation
of  the  transmission  system  overall  efficiency
was investigated based on the flow rate.  
When  the  study  network  was  equipped  by
ORC-pentane  and  turbo-expander,  the  overall
efficiency  grew  by  22  %  averagely,  in  the  span
of 50 to 90 MMSCMD.

 

References
Abbott, M. M., Smith, J. M., & Van Ness, H. C.
(2001).  Introduction  to  chemical
engineering  thermodynamics.  McGraw-Hill.  
Bolland, O., Forde, M., & Hande, B. (1996). Air
bottoming  cycle:  use  of  gas  turbine
waste  heat  for  power  generation.
Journal of engineering for  gas turbines
and power, 118(2), 359-368.  
Borraz-Sánchez,  C.,  &  Ríos-Mercado,  R.  Z.
(2005).  A  hybrid  meta-heuristic
approach  for  natural  gas  pipeline
network  optimization  Hybrid
Metaheuristics (pp. 54-65): Springer.
Cho,  S.-Y.,  Cho,  C.-H.,  &  Kim,  C.  (2008).
Performance  characteristics  of  a  turbo
expander  substituted  for  expansion
valve  on  air-conditioner.  Experimental
Thermal  and  Fluid  Science,  32(8),
1655-1665.  
Dai, Y., Wang, J., & Gao, L. (2009). Parametric
optimization  and  comparative  study  of
organic  Rankine  cycle  (ORC)  for  low
grade  waste  heat  recovery.  Energy
Conversion  and  Management,  50(3),
576-582.  
Desai,  N.  B.,  &  Bandyopadhyay,  S.  (2009).
Process  integration  of  organic  Rankine
cycle. Energy, 34(10), 1674-1686.  
Drescher,  U.,  &  Brüggemann,  D.  (2007).  Fluid
selection for the Organic Rankine Cycle
(ORC)  in  biomass  power  and  heat
plants.  Applied  Thermal  Engineering,
27(1), 223-228.  
Edgar, T., Himmelblau, D., & Bickel, T. (1978).
Optimal  design  of  gas  transmission
networks.  Society  of  Petroleum
Engineers Journal, 18(02), 96-104.  
Ghazikhani,  M.,  Passandideh-Fard,  M.,  &
Mousavi,  M.  (2011).  Two  new  high-performance cycles for gas turbine with
air bottoming. Energy, 36(1), 294-304.  
Guo,  B.,  &  Ghalambor,  A.  (2014).  Natural  gas
engineering handbook: Elsevier.
Kabirian,  A.,  &  Hemmati,  M.  R.  (2007).  A
strategic  planning  model  for  natural
gas  transmission  networks.  Energy
policy, 35(11), 5656-5670.  
Kambanis,  L.  M.  T.  (1995).  Analysis  and
modeling of power transmitting systems
for  advanced  marine  vehicles.
Massachusetts Institute of Technology.  
Korobitsyn,  M.  A.  (1998).  New  and  Advanced
Conversion  Technologies:  Analysis  of
Cogeneration,  Combined  and

Integrated  Cycles:  University  of
Twente.
Larjola,  J.  (1995).  Electricity  from  industrial
waste  heat  using  high-speed  organic
Rankine  cycle  (ORC).  International
journal  of  production  economics,  41(1),
227-235.  
Maddaloni,  J.  D.,  &  Rowe,  A.  M.  (2007).
Natural  gas  exergy  recovery  powering
distributed  hydrogen  production.
International  journal  of  hydrogen
energy, 32(5), 557-566.  
Menon,  E.  S.  (2005).  Gas  pipeline  hydraulics:
CRC Press.
Najibi,  H.,  &  Taghavi,  N.  (2011).  Effect  of
different  parameters  on  optimum
design  for  high  pressure  natural  gas
trunk-lines.  Journal  of  Natural  Gas
Science and Engineering, 3(4), 547-554.  
Nishada, K., Takagi, T., & Kinoshita, S. (2005).
Regenerative  steam  injection  gas
turbine  system.  Applied  Energy,  81,
231-246.  
O'Neill, R. P., Williard, M., Wilkins, B., & Pike,
R.  (1979).  A  mathematical
programming  model  for  allocation  of
natural  gas.  Operations  Research,
27(5), 857-873.  
Poživil, J. (2004). Use of expansion turbines in
natural  gas  pressure  reduction
stations.  Acta  Montanistica  Slovaca,
3(9), 258-260.  
Ríos-Mercado,  R.  Z.,  Kim,  S.,  &  Boyd,  E.  A.
(2006).  Efficient  operation  of  natural
gas  transmission  systems:  A  network-based  heuristic  for  cyclic  structures.
Computers  &  Operations  Research,
33(8), 2323-2351.  
Roy,  J.,  Mishra,  M.,  &  Misra,  A.  (2010).
Parametric  optimization  and
performance  analysis  of  a  waste  heat
recovery  system  using  Organic
Rankine  Cycle.  Energy,  35(12),  5049-5062.  
Safarian, S., & Aramoun, F. (2015). Energy and
exergy  assessments  of  modified
Organic  Rankine  Cycles  (ORCs).
Energy Reports, 1, 1-7.  
Safarian,  S.,  &  Bararzadeh,  M.  (2012).  Exergy
analysis  of  high-performance  cycles  for
gas  turbine  with  air-bottoming.
Journal  of  Mechanical  Engineering
Research, 5(2), 38-49.  
Safarian,  S.,  Saboohi,  Y.,  &  Kateb,  M.  (2013).
Evaluation  of  energy  recovery  and
potential  of  hydrogen  production  in
Iranian  natural  gas  transmission
network. Energy policy, 61, 65-77.  
Vaja,  I.,  &  Gambarotta,  A.  (2010).  Internal
combustion  engine  (ICE)  bottoming
with  organic  Rankine  cycles  (ORCs).
Energy, 35(2), 1084-1093.  
Wei,  D.,  Lu,  X.,  Lu,  Z.,  &  Gu,  J.  (2007).
Performance analysis and optimization
of  organic  Rankine  cycle  (ORC)  for
waste  heat  recovery.  Energy
Conversion  and  Management,  48(4),
1113-1119.  
Wu,  S.,  Rios-Mercado,  R.  Z.,  Boyd,  E.  A.,  &
Scott,  L.  R.  (2000).  Model  relaxations
for  the  fuel  cost  minimization  of
steady-state  gas  pipeline  networks.
Mathematical  and  Computer
Modelling, 31(2), 197-220.  
Zaleta,  D.,  &  Ríos-Mercado,  R.  (2002).  A
MINLP  Model  for  a  Minimizing  Fuel
Consumption  on  Natural  Gas  Pipeline
Networks: October.