Prog. EnergyCombust.Sci. 1993,Vol.19,pp. 227 257 0360-1285/93 $24.

00
PrintedinGreatBritain.Allrightsreserved. ©1993PergamonPressLtd

T H E R M O E C O N O M I C A N A L Y S I S A N D OPTIMIZATION OF E N E R G Y
SYSTEMS

GEORGE TSATSARONIS
Center for Electric Power, Tennessee Technological University, P.O. Box 5032, Cookeville, T N 38505, U.S.A.

Received 1 June 1993

Abstract--This article discusses the development, state-of-the-art and applications of exergy analysis and
thermoeconomics (exergoeconomics). The latter is a relatively new method that combines exergy with
conventional concepts from engineering economics to evaluate and optimize the design and performance
of energy systems. This paper is written mainly for the person interested in applying tbermoeconomics to
an energy system. It reviews the history of exergy analysis and thermoeconomics, the performance
evaluation of an energy system from the viewpoints of the second law of thermodynamics and
thermoeconomics as well as applications of thermoeconomic optimization techniques.

CONTENTS

Nomenclature 227
1. Introduction 228
2. Historical Background 229
3. Exergy Analysis 231
3.1. Exergy balance and exergy destruction 231
3.2. Calculation of exergy values 232
3.3. Exergetic efficiency 234
3.3.1. Compressor, pump or fan 235
3.3.2. Heat exchanger without mixing 235
3.3.3. Heat exchanger with mixing 236
3.3.4. Turbine or expander 236
3.3.5. Combusion chamber 236
3.3.6. Boiler 237
3.3.7. Other components 237
3.4. Performance criteria 237
3.5. Example 238
4. Economic Analysis 239
5. Exergy Costing 240
5.1. Cost balance 241
5.2. Average cost of fuel and product 242
5.3. Costing of material and energy-transfer streams 242
5.4. Costing ofexergy-loss streams 247
5.5. Cost ofexergy destruction 247
5.6. Relative cost difference 247
6. Thermoeconomic Evaluation 248
7. Thermoeconomic Optimization of an Energy System 249
7.1. Cost optimal exergetic efficiency 249
7.2. Design optimization 251
7.2.1. Optimization of major plant components 251
7.2.2. Other optimization approaches 252
7.3. Guidelines for designing energy systems 252
7.3.1. Generalguidelines 252
7.3.2. Process specific guidelines 253
8. Closure 253
Acknowledgements 254
References 254

NOMENCLATURE e specific exergy [MJ/kg]

B constant in cost equations E exergy IM J]
c cost per exergy unit IS/G J] f exergoeconomic factor denoting the contribution
cost flow rate [$/hr] of capital costs, ,~., to the cost increase, Ac,
Ac difference between cost per exergy unit of prod- between fuel and product in a plant component
uct and cost per exergy unit of fuel in a plant F dimensionless variable expressing the criterion of
component [S/G J] thermoeconomic similarity, Eq. 102
d cost per unit of mass [$/kg] h specific enthalpy l'MJ/kg]

227
228 G. TSATSARONIS
/ / enthalpy flow rate [MW]
Ah ° higher heating value [MJ/kg]
K capital cost associated with a plant component
[$]
m mass [kg] or exponent in cost equations
n exponent in cost equations
ND number of components (devices) in a plant
NI number of material flow streams at the inlet of a
component, or of the total plant
NO number of material flow streams at the outlet of
a component, or of the total plant
P pressure [bar]
Q heat transfer [MJ]
r relative cost difference, Eq. 92
R remaining costs [$], or gas constant [kJ/(kg K)]
s specific entropy [kJ/(kg K)]
entropy flow rate [MW/K]
T temperature [K]
w specific work transfer [MJ/kg]
W work transfer [M J]
x mole fraction
y exergy destruction ratio
Z capital costs associated with a plant component
and considered in cost equations [$]
fl capital-recovery factor
7 coefficient expressing the part of the fixed operat-
ing and maintenance costs which depends on the
net investment expenditure for a plant compo-
nent, Eq. 96
A difference
e exergetic efficiency
v stoichiometric coefficient in a reaction
r annual number of hours of plant operation at
the nomial capacity [hr]
to constant used in calculating the variable levelized
operating and maintenance costs, Eq. 96 [S/M J]
Subscripts
a air
CM control mass
D exergy destruction
1. INTRODUCTION
Classical thermodynamics provides, among others,
the concepts o f energy, energy transfer by heat and
work, energy balance, entropy, and entropy balance,
as well as mathematical formulas for calculating
thermodynamic properties at equilibrium. The
second law of thermodynamics complements and
enhances an energy balance by enabling calculation
of both the true thermodynamic value of an energy
carrier, and the real thermodynamic inefficiencies
and losses from processes or systems. The concept of
exergy is extremely useful for this purpose.
Exergy is the maximum useful work attainable
from an energy carrier under the conditions imposed
by a given environment. The exergy of an energy
carrier is a thermodynamic property that depends on
both the state of the carrier being considered and the
state of the environment. It expresses the maximum
capability of the energy carrier to cause changes}
Thus, exergy is not only an objective measure of the
thermodynamic value of an energy carrier, but is
also very closely related to the economic value of the
carrier because users pay for the potential of energy
f fuel to the total plant
F fuel (according to the definition of exergetic
ciency)
gen entropy production
i input stream
I investment
j output stream
k plant component
L exergy loss
M mixture
MS material stream
n stream
NE non-exergetic effect
OM operating and maintenance costs
P product (according to the definition of exergetic
efficiency)
rev reversible
R heat reservoir
S system
tot total plant
0 environment
Superscripts
CH chemical exergy of a material stream
CI capital investment
M mechanical exergy of a material stream
OPT optimum
PH physical exergy of a material stream
Q heat transfer
T thermal exergy of a material stream
TOT total exergy of a material stream
X depending on the case being considered, refers to
some of the following superscripts: TOT, CH,
PH, M, T, Q, w
W work transfer
• time rate of the corresponding variable
- refers to cost values obtained when all invest-
ment, operating (excluding fuel) and maintenance
costs are neglected (i.e. when only fuel costs are
considered) in the thermoeconomic analysis
~ molar
to cause changes. When costs are assigned to energy
carriers, exergy, rather than energy or mass, should
serve as a basis in the costing process.
For energy systems operating above the ambient
temperature, we can think of exergy as the useful
part of energy, i.e. the part o f energy that can be
transformed into any other form of energy. The
concept of exergy is simple, understandable and
completely consistent with our intuition and every-
day perceptions. Indeed, exergy is what the layman
calls 'energy'.1 Unlike total energy, a part of the
total exergy supplied to a system is irreversibly de-
stroyed in all real processes. This exergy destruction
usually represents the largest part o f what the lay-
man calls 'energy waste'. The other part of 'energy
waste' is the exergy loss, i.e. the exergy associated
with a material or energy stream rejected to the en-
vironment (e.g. flue gas, cooling water and heat
loss). Both exergy destruction and exergy loss
are identified through an exergy analysis (second-law
analysis).
The exergy destruction is the direct result of the
irreversibilites in a system. The exergy loss is associ-
ated with the design engineer's decision to not further
 Energy systems
use the exergy of a stream in a given system. Malfunc-
tions and off-design performance of a component
usually increase both the exergy destruction and the
exergy losses in a system. At least a part of the
exergy destruction and exergy loss is dictated by
considerations involving costs, materials, environ-
mental impact, reliability, availability, maintainability
and operability of the system being considered. In a
truly optimized system, the entire amount of exergy
destruction and loss is justifiable through these consid-
erations.
In the past, the term 'thermoeconomics' has been
used to indicate an appropriate combination of exer-
getic and economic analysis. The most characteristic
element of this analysis is the assignment of costs to
the exergy (not the energy) content of an energy
carrier (exergy costing). In parallel, however, the
expression 'thermoeconomic analysis' has been used
by others to report conventional thermodynamic
analyses based only on the first law of thermo-
dynamics) and economic analyses conducted sepa-
rately from the thermodynamic ones and without
consideration of exergy or exergy costing. The last
case cannot be excluded from 'thermoeconomics'
because the first constituent of this expression,
'thermo', is a derivative of the Greek word for heat
and is used in most major languages. Thus, thermo-
economics does not in any way imply exergy costing
or exergy economics, but, taken literally, means a
combination of heat and economics. It could, there-
fore, be easily confused with the expression 'heat
economy' in some languages.
In 1983 the author coined the term 'exergoeconom-
ics'2 to give a more precise and unambiguous charac-
terization of the combination of an exergy analysis
with an economic one using exergy costing. In this
case, the term thermoeconomic analysis should be
used in a more general sense to indicate any thermo-
dynamic analysis reported next to any economic
analysis of the same system. The thermodynamic
and economic analyses do not have to be combined
in the more general field of thermoeconomics,
whereas in exergoeconomics they are integrated
through exergy costing. Consequently, exergoeco-
nomics is a part, the most significant one, of thermo-
economics. The term exergoeconomics is already
being used in several countries.
This article reviews the concepts of the relatively
new field of exergoeconomics. Since the readers, how-
ever, might be more familiar with the term thermo-
economic rather than exergoeconomic analysis, an
exception is made in this paper, and the former term
is retained in the title and throughout the text. Thus,
in this paper, the expressions thermoeconomics and
exergoeconomics are synonymous. The author be-
lieves that the distinction between thermoeconomics
and exergoeconomics is important and should be
emphasized in future publications.
A complete thermoeconomic analysis consists of
(a) a detailed exergy analysis, (b) an economic analy-
229
sis conducted at the compoment level of the energy
system being analyzed, (c) exergy costing and (d) an
exergoeconomic evaluation of each system compo-
nent. Following a brief presentation of the history of
exergy and thermoeconomics, the four components
of a thermoeconomic analysis are discussed. Finally,
the optimization methods used in thermoeconomics
are reviewed.
The objectives of a thermoeconomic analysis are:
• To identify the location, magnitude and source
of the real thermodynamic losses ('energy
waste') in an energy system (exergy destruction
and exergy losses).
• To calculate the cost associated with the exergy
destruction and exergy losses.
• To assess the production costs of each product
(output) in an energy-conversion system that
has more than one product.
• To facilitate feasibility and optimization studies
during the design phase for an energy system,
as well as process improvement studies for an
existing system.
• To assist in decision-making procedures concern-
ing plant operation and maintenance and alloca-
tion of research funds.
• To compare technical alternatives.
The literature on the subject of thermoeconomics
is very extensive. The references cited here cannot be,
and are not intended to be, exhaustive; usually, they
refer either to the most representative publications
or to the most recent work of each author, so that
the reader may trace back to earlier publications.
This article is written mainly for the person inter-
ested in applying a thermoeconomic analysis to an
energy system. More emphasis is therefore placed on
how to apply principles than on the underlying theo-
retical aspects. The nomenclature used here is closer
to that used in German publications. It is character-
ized by a combination of three important advantages
compared to the numerous other suggestions con-
tained mainly in publications in the English language:
simplicity, consistency and lack of ambiguity. It is,
therefore, the preferred nomenclature in the author's
opinion.
2. HISTORICAL BACKGROUND
Although the method of exergy is often considered
to be a new method for analyzing energy systems,
the underlying fundamentals were introduced in the
last century following the mathematical formulation
of the second law of thermodynamics. As outlined in
the critical historical review by R. W. Haywood, 3
the earliest contributions to the exergy concept are
due to R. Clausius (1865), P. G. Tait (1868), W.
Thomson (Lord Kelvin) and mainly to J. W. Gibbs
(1873) and J. C. Maxwell (1875). This early work, as
well as the subsequent developments by G. Gouy
(1889), A. Stodola (1898), G. A. Goudenough (1911)
and G. Darrieus (1930), generated little interest.
230 G. TSATSARONIS
The modern development of exergy analysis was
initiated by F. Bosnjakovic in Europe and J. H.
Keenan in the United States. The classical slogan
"Fight the Irreversibilities" by F. Bosnjakovic4
marks the beginning of this development. In the
1950s and 1960s contributions to the exergy concept
were also made by Z. Rant, P. Grassmann, W. M.
Brodyansky, E. A. Bruges, J. H. Keenan, M. Tribus,
E. F. Obert, R. A. Gaggioli, R. B. Evans, H. D.
Baehr, W. Fratzscher, J. Szargut, R. Petela and K.
F. Knoche, among others. During this period, the
exergy balance and its graphical presentation, the
calculation of the exergy of fuels and the definition
of reference states for calculating the chemical exergy
were introduced. In addition, several definitions of
exergetic efficiencies and the first exergy-analysis ap-
plications to industrial processes and plants were
presented. At the same time, the method of exergy
analysis was introduced into the textbooks of thermo-
dynamics. In the last twenty years, the annual
number of worldwide exergy applications to various
systems and processes has increased exponentially.
Extensive compilations of the exergy-related litera-
ture are given in several references. 5 1o
The term exergy was coined by Z. Rant 11 as a new
word for 'work capability' used previously by Bosnja-
kovic. This term has gained general acceptance in all
countries except the United States where the parallel
use of the terms exergy and availability (or available
energy) in both textbooks and research articles contin-
ues to contribute to some misconceptions and con-
fusion surrounding the exergy method. However,
there is little doubt that the word exergy will finally
prevail in the United States too (see, e.g. the fore-
word of Ref. 12).
The idea of using exergy for costing purposes was
initiated by J. H. Keenan 13 in 1932. His suggestion
was not to apply exergy costing as discussed in
Section 5, but to use exergy for appropriately appor-
tioning costs to the electric power and steam that
were produced in a cogeneration plant. He pointed
out that the economic value of steam and electricity
lay in their exergy not energy.
In an unpublished lecture at the Massachusetts
Institute of Technology in 1949, M. Benedict presented
the costing of exergy destruction in an air-separation
plant and the use of these costs for 'optimal design'.
It is very unfortunate that the contents of this lecture
were not published until thirty years later.14
The modem development of thermoeconomics was
initiated in the late 1950s by M. Tribus and R. B.
Evans at the University of California, Los Angeles,
and by E. F. Obert and R. A. Gaggioli at the
University of Wisconsin, Madison. Tribus and Evans
were applying the exergy concept to desalination
processes when they introduced the word thermoeco-
nomics, developed the idea of assigning costs to the
exergy unit of streams, and formulated cost balances
at the component level of energy systems, ls'~6 E. F.
Obert and R. A. Gaggioli applied exergy costing to
the optimal selection of steam piping and its insula-
tion.17, la
In Europe, E. Bergmann and K. R. Schmidt as-
signed costs to the exergy destruction in each compo-
nent of a steam power plant in a study dealing with
optimization of feedwater heaters. 19 W. Fratzscher
and D. Kl6ditz referred to the early work of R. B.
Evans and M. Tribus and applied exergy costing to
the design of a regenerative heat exchanger. 2° J.
Szargut used exergy costing in a cogeneration plant. 21
The paper by Y. M. EI-Sayed and R. B. Evans 22
published in 1970 marks the introduction of rigorous
calculus methods of optimization in thermoeconom-
ics. This approach, though it has not yet yielded the
expected practical results, continues to show promise
for the future.
References 1, 2, 3, 6-9, 14 and 22-71 (given in
chronological order) represent the most important
publications in the area of thermoeconomics from
1970 through 1986. Most of the work published in
the years 1987 through 1993 is included in conference
proceedings volumes (Refs 12 and 72-91) published
mainly by the American Society of Mechanical
Engineers (ASME) and in two ASME Journals, the
Journal of Engineering for Gas Turbines and Power
and the Journal of Energy Resources Technology.
These publications and Refs 92-106 contain several
applications of an exergy analysis to energy systems
and processes. R. W. Haywood 3 reviewed the early
history of exergy while Y. M. EI-Sayed and R. A.
Gaggioli 1°7 and the author l°s outlined the develop-
ment of thermoeconomics and presented critical
reviews of thermoeconomic methods.
In the history of engineering thermodynamics, no
other concept has been given so many symbols (e.g.
A, B, E, Ex, e, E, ~ and W for exergy) and names
(for the exergy, for instance, usable energy in French,
availability, available energy and essergy in English,
and work capability in German; for the term exergo-
economics, the expressions include thermoeconom-
ics, second-law costing, exergy costing, exergy eco-
nomics, etc.) or has been neglected, misunderstood
or misrepresented as often as the concepts of exergy
and exergoeconomics. In the early years of the devel-
opment of thermoeconomics some researchers,
mainly in Europe, attempted to base the cost assign-
ment to material streams on their exergy content
only, regardless of the composition, use and the
other non-exergetic properties of the materials. As a
result, even prominent proponents of the exergy con-
cept in Europe expressed scepticism about the applica-
bility of exergy costing, 1°'21"4s whereas others criti-
cized the exergy concept. At the same time in the
United States, some misrepresentations of the devel-
opment of the exergy concept in Europe can be
found. Had there been sufficient communication be-
tween the researchers from different countries in the
early years, the acceptance of both the exergy concept
and the thermoeconomic approach to the analysis of
energy systems would be wider today.

I
2 TOTAL PLANT ~2
(i)
COMPONENT
NI
-I
T d Or)
ENVIRONMENT
To Po
FIG. 1. General representation of mass and energy balances
in an energy system operating under steady-state condi-
tions.
The value of exergy and thermoeconomic analyses
in the evaluation of an energy system is generally
recognized today. It is important to realize that these
analyses do not replace, but complement and enhance,
an energy analysis or a conventional economic analy-
sis.
3. EXERGY ANALYSIS
This section briefly discusses some general aspects
of the thermodynamic analysis of an energy system
using the exergy method. More emphasis is placed
on the analysis of systems operating under steady-
state conditions since the majority of important indus-
trial plants operates most of the time under steady-
state conditions. As shown in Fig. 1, the mass flow
streams thi (i = 1, ..., NI) are entering and the mass
flow streams rhj (.1'= 1, ..., NO) are leaving the
arbitrarily established control boundary of an energy
system (e.g., a device, a group of devices, or an
entire plant). The chemical composition and the state
variables of the entering streams can be changed
through chemical reactions, mixing, pressure drop,
heat transfer (Q at the temperature T) and work
exchange (W) between the system and its surround-
ings, etc. In the sign convention adopted here, heat
and work transfers are positive when they are sup-
plied to the system and negative when they are
removed from it; the arrows for Q and W in Fig. 1
indicate the positive direction for heat and work
transfers.
Neglecting the changes of kinetic and potential
energy between inlet and outlet, the energy balance
and the entropy balance for the system being consid-
ered are given by the following formulas:
NO NI
"de" ~V = E ]~jhj -- E fnihi (l)
j=l i=1
No
+ ~gen E I~IjSj fnis i
T s=l i=~
In these equations, m., h. and s. represent the
mass flow rate, specific enthalpy and specific entropy,
respectively, of the nth material stream. ~s.. is the
Energy systems 231
rate of entropy production in the system; it is a
measure of all irreversibilities which occur within the
control surface due to chemical reactions, heat ex-
(j)
change, mixing, friction, etc. Sge, is equal to zero
NO only in a completely reversible process.
The method of exergy provides:
(a) a measure to judge the magnitude of 'energy
waste' in relation to the 'energy' supplied or
transformed in the total plant and in the com-
ponent being analyzed,
(b) a measure for the quality (or usefulness) of
energy from the thermodynamic viewpoint, and
(c) a variable to define rational efficiencies for
energy systems.
Some examples of why a first-law analysis (an
analysis based only on energy balances, Eq. 1) fails
to identify the real energy waste or the effective use
of fuels and resources include the following:
(a) The first law does not recognize any waste in
an adiabatic chemical reactor (e.g. gasifier) or
an adiabatic throttling process.
(b) The first law does not show that a part of the
enthalpy of a stream or of the heat transfer
rate is thermodynamically useless.
(c) The first law does not detect any energy degra-
dation in an adiabatic heat exchanger.
(d) The first law identifies the condenser of a
power plant as the component responsible for
the relatively low overall plant efficiency. The
exergy method correctly reveals that the boiler
is mainly responsible for the low efficiency.
In other words the first law cannot conclude that
there are sometimes significant 'energy losses' which
are not recognizable as rejected heat and that rejec-
tion of heat does not necessarily need to be consid-
ered as 'energy loss' (in the Carnot process, for in-
stance).
The exergy of an energy carrier is the result of the
potential interaction between the carrier and the
common components of the environment. Exergy is,
therefore, a function of the state variables of the
energy carrier and of the state parameters (tempera-
ture, pressure and chemical composition) of the envi-
ronment. The exergy content of an energy carrier
is a measure of the thermodynamic value of the
carrier.
3.1. Exergy Balance and Exergy Destruction
An exergy balance states that the total exergy
increase or decrease within the system (subscript s)
boundary (dE~°T) plus the exergy destruction 6ED.s
within the same boundary equals the difference be-
tween the total exergy transfers in and out across the
(2)
boundary. The exergy transfer across the boundary
includes exergy transfer (JE) associated with the
transfer of heat (Q,), work (Ws) and mass (Jm~ enter-
ing and 6mj exiting the boundary) across the bound-
ary.
232
d E r ° r + JED.s = JE Qs + OEw~ - PodV
NI NO
+ ~ (~ETOT-- Z (~ETOT"
i=1 j=l
The term Ws in Eq. 3 denotes all work interactions
taking place at those portions of the boundary
through which there is no mass flow. The term
- P o d V represents the work done at the moving
boundary by or on the surroundings.
For the steady state process shown in Fig. 1 we
can write:
NI NO
ED = L"W ..~ ~'Q 2r_ ~. j~TOT -- ~ j~TOT
i=1 j=l
where E w, E °, E T°T and E T°T are the exergy flow
rates associated with the rates of work and heat
transfer, as well as with the mass flow rates rh~ at the
inlet and rni at the outlet, respectively. The exergy
destruction is always equal to the product of entropy
generation and the temperature of the surroundings.
Thus,
ED = ToSgen.
Hence, exergy destruction can be calculated either
from the entropy production using Eqs 2 and 5 or
directly from the exergy balance, Eq. 4.
For a system consisting of N D subsystems, the
total exergy destruction, ED.lot, is equal to the sum of
exergy destructions in all subsystems, ED.,, k = l,
.... N D
led
ED,tot = E ED,k.
k=l
A change in the exergy destruction in one subsys-
tem, in general, affects the exergy destruction in
other subsystems too. Thus, the change in the total
exergy input to a plant is usually different from the
change in the exergy destruction in one plant compo-
nent. The technique of the structural bond coeffi-
cients developed by J. Beyer25 ' 26 ' 33 ' 109 addresses this
point. This technique, however, has not yet been
fully developed.
Since the total energy is conserved in an energy
balance, there is no first law analog to exergy destruc-
tion. The exergy destruction represents the real
energy waste that cannot be identified through an
energy balance. More details about the exergy bal-
ance and exergy destruction are given in Refs 10, 96,
100 and 106.
3.2. Calculation o f Exergy Values
Energy calculation may be based on any assumed
reference state; by contrast, the state of reference for
exergy calculations is imposed by the environment.
The exergy E w associated with work transfer W over
the system boundary is equal to the work transfer:
EW = W.
G. TSATSARONIS
The exergy E Q associated with heat transfer Q is
given by:
(3)
))o
where T is the temperature at the system boundary
at which the heat transfer occurs, and To is the
temperature of the environment. A negative exergy
value calculated from Eq. 7 and Eq. 8 when T > To
means, according to the sign convention adopted
here, that exergy (associated with the corresponding
work or heat transfer) is supplied by the system to
(4)
the surroundings. When T < To in Eq. 8, the exergy
transfer occurs in the direction opposite heat trans-
fer.
When the heat transfer occurs at a continuously
varying temperature (e.g. fluid in a heat exchanger),
the thermodynamic average temperature Tm96' t t o
hour -- hin
Tm -- - - (9)
Soul -- Sin
(5) can be defined at constant pressure and used in Eq. 8
(T = Tm) to calculate the exergy associated with the
heat transfer (Q = m(houl - hin)).
Excluding nuclear, magnetic, electrical and interfa-
cial effects, the total exergy of a material stream,
E T°r, can be divided into four components: t° kinetic
exergy E KN, potential exergy E pT, physical exergy
E eu and chemical exergy E ca.
E TOT = E KN + E pT + E pH + E CH. (10)
(6)
Kinetic and potential exergy are equal to kinetic and
potential energy, respectively. The physical exergy
associated with a closed (control mass) thermody-
namic system is given by:
E~ = m[(u - uo) + Po(V -- Vo) - To(S -- so)].
0t)
Here, u, v and s represent the specific internal energy,
specific volume and specific entropy of the closed
system. The subscript 0 denotes the state of the same
system at the temperature To and pressure Po of the
environment.
In an open (control volume) system the physical
exergy (exergy of the enthalpy) associated with a
material stream (subscript MS), having a specific
enthalpy h and a specific entropy s, is due to the
differences in temperature and pressure between
stream and environment.*
EP~ = me pn = m [(h - ho) - To(s - so)]. (12)
Here, ho and so are the specific enthalpy and the
specific entropy, respectively, of the same material
stream at the temperature (To) and pressure (Po) of
*These differences, however, are not a sufficient condi-
tion for a stream to have non-zero exergy. This effect can
be illustrated by the environmental line'°4 drawn in an
(7) enthalpy-entropy diagram.
 Energy systems
co2 H~ s% N2
FUEL (f)
-~ RE^Cr~O. / -
02 ~ CHAMBER [ ~ Wr'v
T0. Po
To Po
Fro. 2. Reversible, isothermal and isobaric oxidation of a
fossil fuel at T = To and P = Po used to calculate the
chemical exergy of fuels.
the surroundings. Some authors call the sum of the
first three terms in Eq. I 0 thermomechanieal exergy.
Some publications in the English language contain
the incorrect and confusing statement that only the
exergy of a flow stream (or the flow availability, as it
is often called in these publications) is referred to as
exergy. Distinguishing in the nomenclature between
the exergy of a flow stream and the exergy of a
closed system is not only unnecessary but also mis-
leading.
The physical exergy E Pn of an energy carrier is the
sum of thermal exergy E x, which is due to the system
temperature, and mechanical exergy E u, which is due
to the system pressure.
E PH = E x + E u.
The distinction between thermal and mechanical
exergy, which is not always unique, is discussed in
Ref. 111. The thermal exergy cannot be negative.
However, the mechanical exergy (and, consequently,
the physical exergy) can take on negative values if
the pressure of the system is less than the pressure
of the surroundings. A negative value in this case
means that net work needs to be spent to bring the
carrier into mechanical equilibrium with the environ-
ment.
The chemical exergy of an energy carrier is the
maximum net work which can be attained if the
considered carrier having a temperature To and pres-
sure Po (i.e. zero physical exergy) is brought into
complete thermodynamic equilibrium with the envi-
ronment. Thus, for calculating the chemical exergy,
the chemical composition of the environment (refer-
ence state) has to be defined in addition to the
temperature To and pressure Po.
Thermodynamics considers the environment to be
a large equilibrium system in which the state vari-
ables (To, Po) and the chemical potential of the
environmental chemical components (reference sub-
stances or substances which have a non-zero
concentration in the environment) remain constant
when, in a thermodynamic process, heat and mated-
233
als are exchanged between a system and the environ-
ment. It is important that no chemical reactions can
take place between the reference substances. Only in
that case does a complete thermodynamic equilibrium
exist, and only in that case is the total exergy of the
environment equal to zero. For a given temperature
and pressure of the environment and for given
amounts of the chemical elements contained in it,
the chemical equilibrium for the environmentalchemi-
cal components can be calculated. It is, therefore,
possible to define many equilibrium 'environments'
by giving different temperatures and different
amounts of the chemical elements.
Exergy should represent the potential for work
extraction from a system not for any theoretically
correct environment but for one which takes into
account the conditions of our natural environment.
The latter, however, does not fulfil the requirements
of the thermodynamic theory because it is not in
chemical equilibrium and its state variables are not
constant with respect to location or time. In that
case, we have to make a compromise between the
requirements of the theory and the reality of the
natural environment.~~2
Models for defining the reference state for chemi-
cal exergy calculations were suggested by J.
Ahrendts, ~13'~14 H. D. Baehr and E. F. Schmidt, xls
F. Bosnjakovic, t16 R. A. Gaggioli and P. J. Petit, 1~7
L. Rodriguez, ~ls J. Szargut 1° and J. H. Shieh and
L. T. Fan. ~9 These models provide, as a minimum,
the standard chemical exergy of single reference sub-
stances, i.e. the chemical exergy of substances con-
tained in the environment at the standard state (T =
(13)
298.15 K, P = 1.013 bar). The effect of changes in
the temperature and pressure of the reference state
on the chemical exergy of the reference substances
might be neglected. The chemical exergy of non-refer-
ence substances (all other chemical components
which appear in the energy system being analyzed
but have a zero concentration in the reference envi-
ronment) might then be calculated as illustrated in
the following example for a fossil fuel.~ ~2
It is convenient to define the exergy of a fuel using
a hypothetical reversible, isothermal and isobaric
steady-state process as shown in Fig. 2. Fuel, oxidizer
and the reaction products (CO2, H20, SO2 and N2),
must be in thermal and mechanical equilibrium with
the actual thermodynamic environment which might
not be at the standard temperature and pressure. It
is assumed that the reaction products leave the reac-
tion chamber unmixed. Energy, entropy and exergy
balances for the reversible reaction process lead to
the molar chemical exergy of the fuel, ~t:
ef = A~f0 "]- ToAgrev+ A~ = ~rev + A~ (14)
with
A~ = t;, + vo2~o2- E v:~:, : = c o 2 , H2O, SO2, N~
(
(15)
234 G. TSATSARONIS
A~rev = ~ vei/ - if - vo2s'o~ (16)
t streams entering and leaving the system. It is impor-
and
A~ = ~ v , ~ / - vo2e2. (17)
Here, AgO is the molar higher heating value of the
fuel and Agrev is the molar entropy of the reversible
combustion reaction of the fuel with oxygen. This
reaction can be written as:
1 mol of fuel + v0202-----~ I~c02CO2 "1- VHzoH20
"1-PsozSO2 "~ vN2N2. (18)
In Eqs 14 through 17 ff~, is the molar reversible
work (or the negative molar Gibbs function) of the
chemical reaction (Eq. 18). The terms gf and if
represent the molar enthalpy and molar entropy of
the fuel, respectively, whereas v. is the stoichiometric
coefficient of the nth substance in the combustion
Eq. 18. The symbols /~':(h'o~), i:(go2) and ~: (~o~)
denote the molar enthaipy, molar entropy, and molar
chemical exergy, respectively, of the fth combustion
product (or oxygen). All the combustion products
and oxygen are reference substances.
The contribution of the work ff,~, to the value of
the fuel exergy (&), Eq. 14, is significantly higher
than that of the exergy difference A~. The latter
represents the work which theoretically can be gained
during the mixing of the unmixed reaction products
CO2, H20, SO2 and N2 with the chemical compo-
nents of the environment* minus the work required
for the production of pure oxygen from the chemical
components of the environment. Thus, the effect of
the chemical composition of the environment on the
chemical exergy value of a fuel is introduced through
the term A~. An example of how to calculate the
chemical exergy of coal using Eqs 14 through 18 is
given in Refs 8 and 120. For most fuels, the value of
chemical exergy is very close to their higher heating
value.
The molar chemical exergy o f an ideal mixture,
~cH, is calculated as a function of the molar chemical
exergy of the nth chemical constituent of the mixture,
~H, by
~cn = ~ x.~Ca + RTo ~ x,ln(x,) (19)
n n
where x, is the mole fraction of the nth substance in
the mixture and R is the universal gas constant.
The difficulties in selecting the appropriate refer-
ence state for a given energy system should not in
any way discourage application of exergy analysis.
Generally, 6° the effect of this selection on the final
conclusions from an exergy or a thermoeconomic
analysis is negligible.6a.112 This effect is shown
*That is practically the work obtained in the expansion
of the gaseous products from the pressure at which they
leave the reaction chamber (P0) to their respective partial
pressure in the environment.
mainly in the exergy accompanying the material
tant to realize that the exergy destruction, as its
calculation through Eq. 5 shows, is independent of
the chemical composition of the reference state used
to calculate the chemical exergy values, although the
individual values of the terms E:rOT and /~yOT on the
right side of Eq. 4 do depend on the chemical compo-
sition of the reference state. The author prefers to
use the modified model of Ahrendts tx3'x14 which
considers (a) restricted chemical equilibrium for nitric
acid and nitrates, and (b) a thickness of 1 m for the
layer of the solid earth crust. This model represents
the best compromise between requirements of the
thermodynamic theory (chemical equilibrium at the
reference state) and closeness to the natural environ-
ment.
3.3. Exergetic Efficiency
The exergetic efficiency evaluates the true perform-
ance of a process or energy system from the thermo-
dynamic viewpoint. Since the definition of an exer-
getic efficiency is subjective to some extent, there is
still some confusion with respect to its appropriate
definition for a given system. The following guide-
lines are useful in formulating exergetic efficiencies of
energy systems.
(1) It is helpful to define a (net) fuel and a (net)
product for the thermodynamic system being ana-
lyzed. The product represents the (net) desired result
produced in the system. The definition of the product
must be consistent with the purpose o f purchasing
and using the system. The fuel represents the net
resources which were spent to generate the product.
Then, the exergetic efficiency (e) of any energy system
can be defined as the ratio between the exergy value
of the product Ep and the exergy value of the fuel
EF2,8,6 7
Ep ED+EL
e =- = 1 (20)
EF EF
The term EL represents the exergy losses, i.e. the
exergy associated with a mass or energy stream re-
jected to the environment. After introducing the
exergy of the fuel and product, the overall exergy
balance becomes
Ev = Ee + ED + EL. (21)
Examples of the definition of the exergy of fuel and
product for various energy systems are given below.
(2) The definition of exergetic efficiency must be
meaningful from both the thermodynamic and the
economic viewpoints.
(3) Equating the exergy of fuel with the sum of
exergy inputs and the exergy of the product with the
sum of exergy outputs is, in general, not acceptable
for single plant components and devices because it
 Energy systems 235

Hot Stream

W~ 3 21 Cold Stream

(a) COMPRESSOR, (b) HEAT EXCHANGER

PUMP, ORFAN WITHOUT MIXING

Hot
~~i~"~fl 4 ~ W
Cold .q 3
(c) MIXING UNIT (d) TURBINE

Flue Gas 14

I~5
BFW
I 6 ~ Main Steam
Fuel ~ Coal ~ (F)
I ~7 Cold Reheat
Oxidant 2 ~ Reaction
" ,i I , A,r '21 > I 8 ~ Hot Reheat
2 -I I
Ash ~ 3
(e) COMBUSTION
CHAMBER
(f) BOILER
FIG. 3. Selected components of energy systems.

usually results in misleading conclusions. This ap- inlet to outlet using mechanical or electrical power
proach, which is still followed by some researchers,
could be successfully applied only when the energy EP,A = ~,TOT __ ETOT (22)
system being evaluated consists of a large number of
devices. Ev,a = W. (23)
(4) Changes (differences) in exergy values should
be considered where appropriate in calculating the 3.3.2. Heat exchanger without mixing (Fig. 3b)
exergy of fuel and product. Thus, instead of operat-
ing with absolute exergy values, exergy increases If the purpose of the heat exchanger (component
(e.g., net exergy supply to a stream) should be consid- B in Fig. 3) is to increase the thermal exergy of the
ered in the definition of the exergy of product and cold stream (from state 1 to state 2 in Fig. 3b) at the
exergy decreases (e.g., net exergy supply by a stream) expense of the exergy of the hot stream (3 to 4),
in the definition of the exergy of fuel. Hence, for a then:
material stream that crosses the control boundary of E'p,B = L ' ] ° T -- g, TOT (24)
the system twice without a change in its chemical
composition, not the absolute exergy values at the /~v,B = / U °T - / U °T. (25)
inlet and outlet but their difference should be consid- When the distinction between thermal and me-
ered in the calculation of the exergy of fuel or prod- chanical exergy is made, the above definitions
uct. become:
The following not exhaustive presentation of the
exergy rate of fuel and product for selected energy EV.B = /?T _ / ? T (26)
systems at steady-state operation, Fig. 3, could serve /~,~ = (]r~ oT _ / ~ l o T ) + (/~ - E~)
as a model for developing exergetic efficiency expres- + (ECH _ L-ca). (27)
sions.
Equations 26 and 27 allow for the pressure drop
in the cold stream to be considered on the fuel side
3.3.1. Compressor, pump or fan (Fig. 3a)
where it correctly belongs. In the absence of leakages,
The purpose of these devices (component A in the last term in the parentheses in Eq. 27 is zero.
Fig. 3) is to increase the flow rate of exergy from When the purpose of a heat exchanger is to provide
236
FEEDWATER !
HEATER ] COOLER ,4 I " I N
Feedweter Cooling
Preheating L.....~.litei: .....................................................
FIG. 4. Schematic of a compression process with intercool-
ing.
cooling for a cold chamber as, for instance, in the
evaporator of a cooling plant, then:
/~a,a = E4r°T - L"T°T
or
,°.,El
Ev.B = j~TOT _ ]~OT.
The definition of the product according to Eq. 28b
considers the cold chamber as a constant temperature
heat reservoir at a temperature TR. QR is the heat
transfer rate from the heat reservoir to the cold
stream. The exergy rate accompanying this heat trans-
fer ( ~ ) , which represents the exergy rate of the
product for this process, flows in the direction oppo-
site heat transfer as indicated by the negative sign in
Eq. 28b.
Finally, if the heat exchanger being considered
uses cooling water to cool a stream (e.g., in a con-
denser or intercooler), no exergetic efficiency should
be defined for this cooler alone (e.g., cooler III in
Fig. 4) since the purpose of owning and operating
this cooler involves also at least one more compo-
nent. A meaningful exergetic efficiency can be de-
fined only for the system including the cooler and
the plant component(s) immediately following the
cooler (e.g., the compressor following the intercooler
III in Fig. 4).
Here, it is implied that the purpose of the inter-
cooler is to reduce the power to be supplied to the
compression process immediately following intercool-
ing. Thus, for the system shown in Fig. 4, an exergetic
efficiency is defined for the system including as a
minimum the components III and IV:
Ee,ill/lv = ~-ror _ L,aToT
L"F,IIi/IV = IVIV'
If the feedwater heater (II) is included in the defini-
tion, then:
~P,n-w = (L~ °T - Lcl°T) + (/~-~OT_ /~OT) (32)
/~v,.-w = IVw.
The exergy removal (L'a - ~74) in the cooler of Fig. 4
becomes part of the exergy destruction and the
exergy loss (L%) in the system being considered. This
is an example of the overall positive role that exergy
destruction and exergy loss play in a system: both
may contribute to the reduction of investment costs
G. TSATSARONIS
(e.g., smaller compressor IV) and of fuel required for
~ ) another component (e.g., electric power supplied to
]
compressor IV).
3.3.3. Heat exchanger with mixing (Fig. 3c)
The purpose of a direct contact heat exchanger or
a mixing unit (component C in Fig. 3) is to increase
the thermal exergy of the colder mass flow rate at
the expense of the thermal exergy of the hotter mass
flow rate and the mechanical exergy of both streams.
Thus:
(28a)
J~p.c = ml(e x - e x) (34a)
J~F.C = rhz(ex -- e x) + (/~rox _ EX)
+ ( / ~ o r _/~x) _ (/~OT _/~x). (35a)
=
The superscript X in the above equations repre-
(29) sents PH (physical exergy), when no distinction be-
tween thermal and mechanical exergy is made, or T
(thermal exergy), when the physical exergy is split
according to Eq. 13.
Equations 34a and 35a are meaningful from the
thermodynamic viewpoint, but not always from the
economic one because the specific cost (cost per unit
mass) of the physical or thermal exergy of stream 1 in
Fig. 3c might be higher than the corresponding speci-
fic cost of stream 2 (see Section 5 on exergy costing).
Therefore, the following expressions for the exergy
of product and fuel of a heat exchanger with mixing
might be preferable in a thermoeconomic analysis.
/~P,C = ~,x (34b)
E~.c = L'T°x +/~2r°x -- (]~TOT_ L-x) (35b)
where the superscript X represents again either physi-
cal exergy (PH) or thermal exergy (T) depending on
whether thermal and mechanical exergy are consid-
ered separately.
3.3.4. Turbine or expander (Fig. 3d)
The purpose of this device (component D in Fig.
3) is to generate power (IV) by decreasing the flow
rate of exergy from inlet to outlet. For a turbine with
one extraction (stream 2)
(30) Ep,D = IV (36)
(31) ~'v.D = L'T°r - E2T°T - E]OT. (37)
3.3.5. Combustion chamber (Fig. 3e)
The purpose of the combustion reaction in the
chamber (component E in Fig. 3) is, primarily, to
(33)
increase the thermal exergy difference between the
reaction products and the reactants at the expense of
chemical exergy of the fuel.
~'P,E = ~-x _ ~-x _ ~.x (38)
L'F,~ = (/~I °r -/~x) + (L-lOt _ ~:x) _
(/~]OT _ EX). (39)

The superscript X in Eqs 38 and 39 should be
equal to PH when the physical exergy is not split
into thermal and mechanical exergy, or when the
mechanical exergy of the reaction products is higher
than the sum of the mechanical exergies of fuel and
oxidizer. If this sum is higher than the mechanical
exergy of the reaction products, then the superscript
X should be equal to T (thermal exergy). In the
latter case, the difference E• ° x - E x for the ith
stream is equal to E c n + E~ while in the former
case, ET°T - E x means E ca. If the system in Fig. 3e
represents a simplified gasification reactor, then the
purpose of the reactor is to provide the total exergy
of stream 3.
J~P.E = ETOT __ ]~X __ Ex (40)
EF,E = ( E T OT -- EX) -Jr- (ETOT -- EX).
The superscript X may be again either PH or T.
The following definitions for the exergy of fuel
and product could be used as an alternative to Eqs
38 and 39 or 40 and 41:
]/P,E = /~OT
/~v.E = E~OT + E2TOT.
3.3.6. Boiler (Fig. 3f)
The purpose of the boiler (component F in Fig. 3)
is to increase the exergy rate between inlet and outlet
for the boiler-feedwater/main-steam stream and the
reheat stream.
Ep,F ~--- (ETOT _ ETOT) "-F (]~7T°T -- E7 "r°x)
EF.F = E I °T - E l °T - ~-~DT_ ~-IOT.
An alternative to Eq. 45 is to consider the exergy
flow rate of ash as exergy loss and to strike the term
/~]OT from the definition of the exergy rate of fuel.
Then /~]DT becomes a part of EL in the exergy
balance (Eq. 21). This treatment of ~'3 is appropriate
only when the cost associated with the ash stream is
zero (see Section 5).
3.3.7. Other components
As discussed previously for the intercooler (Fig.
4), for some system components, no meaningful,
from the exergy viewpoint, purpose can be defined
when they are considered just by themselves. In a
cost-effective design, a throttling valve, for instance,
serves another component and/or assists in total cost
minimization. A throttling valve and other compo-
nents with similar characteristics must be considered
together with the component they serve; an exergetic
efficiency must be defined only for this group of
components. The exergy destruction and exergy loss,
however, can and should be calculated for each
component separately. Various approaches to the
treatment of a condenser are discussed in Refs 8, 37,
52, 63, 67, 98, 107 and 121.
The purpose of a distillation system is to separate
Energy systems 237
chemical components at the expense of thermal
energy. Thus, the product of such a system is equal
to the total exergy difference between products and
feed. The fuel is the total exergy supplied in the re-
boiler.
The definitions of exergetic efficiency given above
are not unique. In specific cases, other expressions
might also be acceptable. The subject of exergetic
efficiency of energy systems is covered in Refs 8, 10,
67, 98, 100, 104, 105, 106 and 122.
3.4. Performance Criteria
The following variables are very useful in evaluat-
ing the performance of the kth component of an
(41) energy system from the thermodynamic viewpoint:
(a) Exergetic efficiency (ek) This efficiency (Eq.
20), which, in general, is superior to any other effÉ-
ciency based on the first or second law of thermody-
namics (including thermal efficiency and isentropic
or polytropic efficiency), shows what percentage of
(42)
the exergy of the fuel of the kth component can be
(43) found in the exergy of the product. The difference
between one hundred percent and the actual effi-
ciency value is the percentage of the component fuel
exergy that was wasted in this component as exergy
destruction or exergy loss.
(b) Exergy destruction (ED.k) The absolute value of
exergy destruction in a component provides a meas-
ure of the significance of the component irreversibili-
(44) ties from the thermodynamic viewpoint. For compari-
son purposes, however, the following ratios are used
(45) in addition to the absolute value of exergy destruc-
tion.
(c) Exergy destruction ratios The exergy destruc-
tion in the kth plant component can be related to the
• total exergy destruction in the plant
ED,k
y~.k -- . (46)
ED.tot
• exergy of the fuel to the total plant
ED,k
yD,k -- . (47)
EF.tot
• exergy of the total-plant product
J~D,k
yD.k -- (48)
Ep,tot "
The above exergy destruction fractions can be
used for comparisons among various components of
the same plant and, like exergetic efficiencies, for
comparisons among similar components of different
plants which use identical fuels or produce identical
products. 123.124 In the exergetic efficiency, the exergy
destruction in the kth component is compared (indi-
rectly) with the exergy of the fuel for the same
component whereas in the variable yD.k the same
exergy destruction is compared with the exergy of
238 G. TSATSARONIS

TABLE 1. Thermodynamic variables of the material streams in the gas-turbine system shown in
Fig. 5

Mass flow rate Temperature Pressure Enthal.py rate Entropy rate

Stream (rn) (T) (P) (H) (S)
no. [kg s l] [K] [atm] [MW] [kW K -1]

1 5.0000 298.15 1.00 0.0006 34.4540

2 5.0000 509.25 5.51 1.0935 34.7538
3 5.0000 782.00 5.25 2.5603 37.1278
4 0.0595 298.15 1.00 -0.2776 0.6906
5 0.0595 448.50 5.25 -0.2554 0.6993
6 5.0595 770.51 5.09 2.3049 38.0448
7 5.0595 1273.04 4.99 2.3049 41.1827
8 5.0595 900.35 1.05 0.0809 41.4195
9 5.0595 643.00 1.02 -- 1.3860 39.5455

TABLE 2. Total, chemical, physical, thermal and mechanical exergy flow rates associated with
the streams in the gas-turbine system shown in Fig. 5. All values are given in MW

Total Chemical Physical Thermal Mech.

Stream exergy exergy exergy exergy exergy
no. (~.TOT) (,~-CH) (L,Pn) (~r) (L,M)
1 0.0110 0.0110 0.0000 0.0000 0.0000
2 1.0145 0.0110 1.0035 0.2679 0.7355
3 1.7735 0.0110 1.7625 1.0480 0.7145
4 3.0621 3.0621 0.0000 0.0000 0.0000
5 3.0817 3.0621 0.0196 1.0001 0.0195
6 4.7903 3.0285 1.7618 1.0455 0.7163
7 3.8545 0.0095 3.8450 3.1228 0.7222
8 1.5599 0.0095 1.5504 1.5285 0.0219
9 0.6517 0.0095 0.6422 0.6333 0.0089

I;V1 = 1.0928; Will = 0.0222; I~ZVI = 2.2240; l;gnet = 1.1311.

~ '~Fuel (CH 4 ) in Fig. 5. The total process in the combustion cham-

ber is split into two subprocesses: mixing (component
~/ . . ~ In I FUEL
nl ~ COMPRESSOR IV) and combustion (component V). Table 1 shows
REGENERATOR 5 the results obtained from mass, energy and entropy
balances, whereas Table 2 shows the total exergy
.......~..o..~.~.~.Lo.~ ......
flow rates o f each stream and the breakdown of
I I -I i ] , CHAMBER i these flows into chemical, physical, thermal and me-
chanical exergy. Table 3 presents, for each com-
ponent of the gas-turbine system, the (net) fuel
exergy, the (net) product exergy, the rate of exergy
- VI ~.
destruction, the ratio yD of exergy destruction to fuel
" ~ GAS exergy for the entire system, the exergetic efficiency
1 AIR TURBINE
COMPRESSOR and a cost flow rate which is discussed in Section
4.
FIG. 5. Schematic of a gas-turbine system.
Two cases are presented in Table 3; in case A the
the fuel to the totalplant. The variable yD,k expresses analysis is conducted using chemical and physical
the percentage contribution of the exergy destruction exergy alone, without distinguishing between thermal
in the kth component to the reduction in the overall and mechanical exergy in the definitions of fuel and
plant efficiency. product. In case B the distinction between thermal
The variables ek, ED.k, yD.k, yD.k, and y~.k allow a and mechanical exergy is made in the calculations.
detailed and objective evaluation of the overall per- As discussed in Section 3.3, cases A and B will, in
formance of the kth component. The information general, yield different results for the exergetic effi-
contained in the last four variables can be graphically ciencies. However, Table 3 indicates that the effect
represented as a pie chart or as a Grassmann dia- of this distinction on the results of the exergy analysis
gram.10L 125 are minor for this gas-turbine system. While the
values of/~p and/~v for the regenerator (II) and the
3.5. Example mixing process are affected significantly, their exer-
getic efficiencies are influenced only slightly by this
The approaches discussed here are illustrated with distinction.
the aid of a simplified gas turbine system as shown The T H E S I S computer program 8'126 was used to
 Energy systems 239

TABLE 3. Exergy of the fuel (~¥), exergy of the product (~Tp), exergy destruction (~'D), exergy destruction ratio (yo),
exergetic efficiency (e) and exergy dependent cost flow rate (2) for each component of the gas-turbine system shown in
Fig. 5. In case A the physical exergy was not split, while in case B the mechanical and thermal exergy values were
separated

/~F L~p ~TD yn e 2

Component [MWJ [MW] [MW] [%] [%] [$/hr]
No. Name A B A B A&B A&B A B A&B

I Air compressor 1.0928 1.0928 1.0035 1.0035 0.0894 2.91 91.82 91.82 1.00
II Regenerator 0.9082 0.9292 0.7591 0.7801 0.1491 4.85 83.58 83.95 0.60
III Fuel compressor 0.0222 0.0222 0.0196 0.0196 0.0026 0.08 88.29 88.29 0.30
IV Mixing 1.8267 1.1104 1.7618 1.0455 0.0649 2.11 96.44 94.15 0.00
V Combustion 3.0190 3.0190 2.0832 2.0832 0.9358 30.45 69.00 69.00 0.30
VI Gas turbine 2.2946 2.2946 2.2240 2.2240 0.0706 2.30 96.92 96.92 1.40

conduct the calculations. The calculation of thermo-
dynamic properties was based on the relationships
given in Ref. 127 and the calculation of exergy values
on the restricted equilibrium model of Ref. 113.
It is apparent that 71.309/o of the total exergy
destruction (L"D.tot = 1.312 MW) occurs in the adi-
abatic combustion process (component V). As the
variable yu indicates, this exergy destruction alone
reduces the overall plant efficiency by 30.459/0. The
regenerator (II), air compressor (I), and gas turbine
(VI) contribute 11.36, 6.81 and 5.38% to the total
exergy destruction, respectively, whereas 4.95% of
the total exergy destruction is caused by mixing the
reactants (IV). The overall exergetic efficiency is:
Ep,tot ~Vnet- Wlll 1.1089
. . . . 36.08%.
Etot EF,tot j~TOT + L'TOT 3.0731
In the above calculation, the fuel exergy for the
entire system was defined as the sum of the total
exergies of the methane and air streams. This defini-
tion is recommended in all fossil-fuel combustion
systems because a fossil fuel has no potential to
generate work without the presence of an oxidizer.
When the sum of exergy of fossil fuel and exergy of
oxidizer is used, the deviations in the exergy values
calculated using different environmental models are
considerably lower than when only the fossil-fuel
exergy is considered. 6°
The exergy analysis clearly identifies the combus-
tion chamber as the major source of thermodynamic
inefficiency. An energy analysis cannot identify any
'energy waste' in the adiabatic combustion chamber.
The second largest source of thermodynamic ineffi-
ciency is the exergy loss associated with stream 9;
this exergy loss reduces the overall plant efficiency
by (0.6517)(100)/3.0731 = 21.21~. Thus, the overall
plant efficiency will increase when the inefficiencies
associated with these two sources are reduced. This
can be achieved, for instance, by increasing the air
preheating temperature 7"3, decreasing the gas ex-
haust temperature Tg, and decreasing the air-fuel
ratio in the combustion process. This is just an
example of how an exergy analysis identifies the true
thermodynamic inefficiencies and assists in develop-
ing process modifications to reduce these ineffi-
ciencies. The object of thermoeconomics is to identify
the cost-effective modifications among all changes
suggested by an exergy analysis. In a detailed exergy
analysis, the analyst should calculate separately the
exergy destruction caused by friction, heat transfer,
heat loss to the surroundings, mixing, chemical reac-
tions, vaporization of moisture contained in reac-
tants, etc. 123'124"
4. ECONOMIC ANALYSIS
The economic analysis, conducted as part of the
thermoeconomic analysis, provides the appropriate
monetary (cost) values associated with the invest-
ment, operating (excluding fuel), maintenance and
fuel costs of the system being analyzed. These values
are used in the cost balances discussed in the next
section. The cost values to be used in a thermoeco-
nomic analysis depend on the objectives of this analy-
sis as stated in Section I.
For abbreviation purposes, the sum of capital
investment and operating (excluding fuel) and main-
tenance costs is called capital costs (K) in the follow-
ing. If the objectives of a thermoeconomic analysis
include calculation of product costs or comparison
of technical alternatives, the total amount of capital
costs must be considered in the cost balance formu-
lated for a new system or component. The investment
costs of an existing plant are sunk costs.l When the
cost of exergy destruction is calculated or when the
appropriate trade-off between capital costs and fuel
costs is sought, only that part of the capital costs
(K,x) that depends on the size and efficiency of the
component being analyzed should be considered in
the cost balances
K = Kex + Ko. (49)
The variable Ko in Eq. 49 represents the part of
the capital costs (K) which is independent of the
efficiency and the size of the component, i.e. inde-
pendent of the exergy flow rate through and the
exergy destruction in the component.
The contribution of capital costs to the cost bal-
ances formulated in Section 5 is denoted here by Z.
Thus, depending on the analysis objectives, Z could
240 G. TSATSARONIS
be equal to Kex, K, or any value between zero and K
which is deemed appropriate by the person conduct-
ing the analysis.
For most applications of thermoeconomics, it is
sufficient to use average costs. There are economic
decisions, however, which require marginal costs. In
general, it is very expensive or impossible to obtain
marginal costs. 1'8'22'1°7 Then it is still better to use
incremental or even average costs than not at all. In
the following, when the costs are not specified, aver-
age costs will be implied.
The major variables used as input to an economic
analysis depend on the analysis objective. In general,
they include the following: expenditures and time
schedule associated with purchasing, installing, oper-
ating and maintaining each plant component; general
inflation rate; investment escalation rate; financing
sources and required return on investment for each
source; tax rates; fuel cost and corresponding escala-
tion rate; book life and tax life of the investment;
plant capacity factors assumed in the production
schedule; insurance expenditures; and revenues from
the sale of by-products. Several assumptions are usu-
ally required to provide all the necessary input vari-
ables. The effect of these assumptions on the final
conclusions of the thermoeconomic analysis should
be studied in a subsequent sensitivity analysis.
The major difference between a conventional econ-
omic analysis and an economic analysis conducted
as part of a thermoeconomic analysis is that the
latter is done at the plant component level and
provides the value Z k ( k = 1 . . . . . N D ) to be used in
the cost balance written for each plant component.
When the total plant generates more than one prod-
uct, a conventional economic analysis would use
energy costing to apportion the total costs to the
various products. This, however, is inappropri-
ate.l'2s'4°'49 A rational cost apportioning to the vari-
ous products and calculation of the cost of 'energy
waste' must be based on exergy costing.
In the evaluation and cost optimization of the
design of an energy system it is not advisable to
conduct a separate evaluation for each year of plant
operation. Thus, we need a representative year to
compare the annual values of investment costs, fuel
costs and operating (excluding fuel) and maintenance
expenditures. These annual cost components, how-
ever, may vary significantly within the plant's econ-
omic life. Therefore, the representative year cannot
be an actual year of plant operation. A representative
year is obtained through the use of levelled 12s
annual values for all cost components.
References 8, 67 and 129 contain examples of
detailed economic analyses used in thermoeconomic
evaluations. For the example discussed in Section
3.5, it was assumed that the purpose of the thermoeco-
nomic analysis is to calculate the cost of exergy
destruction. Thus, the values of the variable Z to be
used in cost balances contain only that part of capital
costs that depends on exergy destruction. The values
assumed in this example are given in the last column
of Table 3.
5. EXERGY COSTING
The exergy analysis yields the desired information
for a complete evaluation of the design and perform-
ance of an energy system from the thermodynamic
viewpoint. However, we still need to know how
much the exergy destruction in a plant component
costs the plant operator. Knowledge of this cost is
very useful in improving ('optimizing') the cost effec-
tiveness of the plant.
Exergy not only is an objective measure of the
thermodynamic value of an energy carrier but also is
closely related to the economic value of the energy
carrier, because users pay only for the useful part of
energy. 1,15,22 Consider the following two examples.
An energy engineer wants to 'buy' thermal energy
to accomplish a heating task. He can choose between
two energy carriers that consist of the same material
(e.g., water) and have the same pressure and the
same total energy (rh ['hcarrier - ho] = constant). One
energy carrier (option 1) is available at a high tem-
perature (superheated steam), and the other (option
2) at a low temperature (saturated liquid water). The
second carrier must have a larger mass.
The engineer would, in general, be willing to pay
more for the first carrier than for the second one
because the cost of the heat exchanger, where the
thermal energy would be used, would be lower if the
high-temperature energy carrier were used than if the
low-temperature energy carrier provided the needed
thermal energy. Compared with option 2, the cost of
the heat exchanger is expected to be lower in the first
option because of the (a) smaller mass flow rate on
the hot side, and (b) the higher temperature differ-
ences that will result in a smaller total surface area.
Here the engineer implicitly recognizes that the first
carrier has a higher exergy content and is willing to
pay more for using this carrier than the second one.
This example indicates that exergy, rather than mass
or energy, should serve as a basis for assigning costs
to energy carriers.
For the adiabatic mixing unit (component IV)
shown in Fig. 5, the first law of thermodynamics
detects no losses. Thus, if the cost calculation is based
on the energy content, the cost per unit energy of
stream 6 will be equal to the average cost per unit
energy of streams 3 and 5, if the capital cost of the
mixing unit is ignored. The second law of thermody-
namics, however, identifies exergy destruction during
the mixing process, and with the aid of a cost balance,
concludes that the cost per unit of exergy must be
higher after the mixing process (stream 6) than the
average cost per exergy unit before it (streams 3 and 5).
An increase in the exergy destruction in the mixing
unit (i.e. an increase in the pressure drop or an
increase in the temperature difference between

streams 3 and 5) leads to an increase in the cost per
exergy unit between inlet and outlet of the mixing
unit and consequently between fuel (methane) and
product (net power) of the entire gas-turbine system
in Fig. 5. Thus, costing based on exergy sheds light
on the cost formation process while costing using the
first law could lead to erroneous conclusions, x.40,i 30
Exergy costing is based on the notion that exergy
is the only rational basis for assigning costs to energy
carriers and to 'energy waste' (exergy destruction
and exergy losses). Thus, exergy costing uses costs
per exergy unit. Exergoeconomics is based on exergy
costing and is usually applied at the plant-component
level, x'2'8'15
The simplest way to apply exergy costing is to
charge throughout the plant for exergy destruction
and exergy losses at a uniform cost per exergy unit
equal to the average cost per exergy unit of the fuel
of the total plant. This approach, however, does not
consider that the importance of exergy destruction
and exergy loss, from both the thermodynamic and
economic viewpoints, depends on the relative posi-
tion of the subsystem where the exergy destruction
occurs within the total plant. For example, one MW
of exergy destruction rate in the low-pressure steam
turbine affects the cost of electricity more than an
exergy destruction rate of one MW in the boiler of a
steam power plant. Therefore, more sophisticated
approaches to exergy costing are required.
In thermoeconomics a cost (monetary) value is
assigned to each material and energy stream in the
energy-conversion system (process) being considered.
This value represents the total cost required to pro-
duce this stream. The total cost flow rate associated
with the total exergy flow rate of the nth material
stream is denoted by ~7~°T and given by
¢ 7 o , = ~,Pn + ~7~n+ CKN+ t~ar (50a)
/Z.~oT = d o * / ~ . TM
- P H L-'PH - C H ~TCH
= Cn r,n +On ~n +On
Here (7x, cx and L'x are the cost rate, average cost
per exergy unit and exergy rate, respectively, associ-
ated with the Xth form of exergy (X = total (TOT),
physical (PH), chemical (CH), kinetic (KN) and po-
tential (PT) exergy) of the nth stream. Thus, (7x is
the product of E x and cx for the nth energy or
material stream
Cx = E . e . X X
withX=W,Q, TOT, CH, PH, M o r T .
When the physical exergy is separated into thermal
and mechanical exergy, the following relationship
holds
= = E.c. + E.c..
The cost rates associated with material and energy
streams are used in cost balances and in calculating
various exergoeconomic variables which assist in the
evaluation and optimization of an energy system.
•1pEcs 1 9 : 3 - E
Energy systems 241
5.1. C o s t B a l a n c e
A cost balance expresses that the total cost of the
output streams in an energy-conversion system (pro-
cess) equals the total cost of the input streams plus
the appropriate charges due to capital investment
and operating and maintenance expenses (Z). For
the system shown in Fig. l, the following cost balance
can be formulated assuming that electric power (IV)
and heat (0) are supplied to the system
NO NI
~]or= ~w+u~+ ~ ~TOT+2 (53a)
j=l i=1
with d"T°r (n = i or j ) according to Eq. 50. Thus Eq.
53a can be written as follows:
NO NI
y~ (cyO~/r~o~) = ew/rw + ~OEO + y~ cTO~:To~) + 2.
j=l i=l
(53b)
The term Z is calculated by dividing the sum of
annual capital investment charges and operating and
maintenance expenses by the average number of time
units (hours or seconds) of plant operation per year.
All items in a cost balance are, in general, positive.
Thus, if power or useful heat are transferred from
the system shown in Fig. 1 to another system, the
terms ~w = cWEW and G ° = c°L"° in Eqs 53a and
53b should appear with a positive sign on the left
side of these equations.
In the exergetic evaluation, a net fuel and a net
product were defined for each component of a
system. The cost flow rates associated with the fuel
(Cv) and product (Cp) of a component are calculated
by replacing the exergy flow rates (variable L') by
cost flow rates (variable (7). Then the cost balance be-
comes 2,8,61
(Tp = (Tv + 2 - (TL. (54)
The cost flow rate associated with the fuel (or
- K N I:,~KN PT ~ ' P T
~ +C, Jz~ . product) of a plant component contains the corre-
(50b) sponding cost rates of the same streams used in the
same order and with the same sign as in the definition
of the exergy of fuel (or product). The following are
some examples of defining the cost flow rate of fuel
and product for selected plant components; compare
the following Eqs 55 through 58 with Eqs 22, 23, 44
and 45, respectively.
Compressor, p u m p or f a n (Fig. 3a)
Cp,A= c2~TOT ~"TOT2- - el"TOT /"~IL-'TOT= ~"~OT- - CTOT
(51) (55)
(?F,A = CwIV = ~,W. (56)
Boiler (Fig. 3 f )
CP,F = (~,~or _ ~71OT) + (~aTOr -- ~OX) (57)
(52)
d'~.~ = ~:I °T + d : P T - (~:~oT +/~loT). (58)
The cost flow rates ~rox and ~[OX can be negative if
additional costs (i.e. costs not already considered in
the cost balance for the boiler) are associated with
242
handling and final deposition of ash and flue-gas
cleaning (desulfurization, particulate removal or NOx
reduction).* Thus, the cost flow rate associated with
the fuel of a boiler, Eq. 58, increases when additional
costs are incurred during the final ash deposition
and flue-gas release in the environment. These costs
are finally charged to the product of the boiler
through the cost balance, Eq. 54.
It is apparent that in defining the terms Cv,k and
~P.k after the terms J~F.k and /~P.k have been defined,
we simply substitute for the terms L"x (X = TOT,
PH, CH, M, T, W, Q) in the expressions for/~F.k and
/~P.k by the corresponding terms ~,x in the expressions
for (?F.kand /2'p.k,respectively.
5.2. Average Cost of Fuel and Product
The average cost of fuel for the kth component
(CF,k)expresses the average cost at which each exergy
unit of fuel (as defined in the exergetic efficiency) is
supplied to the kth component.
~F,k
- --.
CF.k ~- EF,k
Similarly, the cost of product (Ca.k) is the average
cost at which each exergy unit of the product of the
kth component was generated.
Cp,k
Cp.k = EP.k
Using Eqs 59 and 60, the cost balance (Eq. 54) can
be written for the kth component as follows:
Cp,kF--,P.k = CF,kF--.V.k "Jr-Z k -- CL.k.
Equations 21 and 61 are very convenient general
formulations of the exergy and cost balances.
5.3. Costing of Material and Energy-Transfer
Streams
A cost balance such as Eq. 53b can be formulated
for every component of a system. The unknown
variables in this equation are the costs per exergy
unit, cT°r (cPn, ccH, cKN and cPr when physical,
chemical, kinetic and potential exergy are considered
separately, or cr, cM, ccn, cKr~ and cPT when the
physical exergy is split into thermal and mechanical
exergy), for a material stream or cw and c° for the
corresponding energy transfer. Since the number of
material and energy-transfer streams is larger than
the number of components, additional (auxiliary)
equations are needed to calculate the unknown vari-
ables. Such equations were considered implicitly16 or
explicitly28'4°'7a in past publications.
*The costs per exergy unit c3 and c4 are negative in these
cases and represent penalty costs associated with the treat-
ment of streams 3 and 4 before their release to the environ-
ment.
G. TSATSARONIS
To simplify the following discussion we will assume
that the changes in kinetic and potential energy (and,
thus, the terms A(~tN and A ~ Pr) are negligible. We will
also assume that the cost rates of all entering streams in
a component are known (from the components they
exit or from their purchase costs if the stream enters
the total plant) and that the unknown variables of the
exiting streams need to be calculated.
In the base case where only total exergy is consid-
ered, no auxiliary equation is required for a compo-
nent with only one exiting stream. The cost per
exergy unit for this stream can be calculated from
the cost balance of the component being considered.
For each additional exiting stream, one auxiliary
equation must be written. Thus, the total number of
auxiliary equations is equal to the number of exiting
streams minus one.
When a distinction between physical and chemical
exergy is made, one auxiliary equation per exiting
stream is required in addition to the auxiliary equa-
tions required for the same component in the base
case (considering only total exergy). Thus, the total
number of auxiliary equations per plant component
(59)
is two times the total number of exiting streams
minus one.
If three exergy forms (chemical, thermal and me-
chanical exergy) are considered in the cost balances,
two auxiliary equations per exiting stream must be
formulated in addition to the auxiliary equations
required for the same component in the base case.
(60)
For each component, the total number of auxiliary
equations required in this case is three times the
number of streams exiting the component minus one.
The following are examples of auxiliary equations
(61) to be used for the components shown in Fig. 3.
Compressor, pump or fan (Fig. 3a)
ccn = ccH (62)
(63)
Equation 63 implies that the addition of thermal and
mechanical exergy to the working fluid occurs at the
same average cost per added exergy unit.
Heat exchanger (Fig. 3b)
cx = cx with X = TOT or (PH and CH) or (T, M
and CH) (64)
ccn = ccn (65)
c~2 = c~ (no phase change). (66)
If there is phase change (e.g. vaporization) in the
cold stream of Fig. 3b, Eq. 63 applied to the heat
exchanger of Fig. 3b should replace Eq. 66.
Mixing device (Fig. 3c)
~ n = t~xcn+ ~22n (67)
~7~ = ~'~ + ~7~. (68)
 Energy systems 243

TABLE 4. Cost flow rates (C') associated with the total exergy, physical exergy and chemical
exergy as well as corresponding costs per exergy unit (c) for the streams shown in Fig. 5.
Results are for case AI: the physical exergy is not split into mechanical and thermal exergy
and the old approach to exergy costing is used. The L~ values considered here are given in
Table 3
Stream ~,ror ~PH ~H cTOT ¢CH cPH
no. [$/hr] [$/hr] [$/hr] [$/GJ] IS/GJ] [$/GJ]

1 0.00 0.00 0.00 0.00 0.00 0.00

2 24.40 24.40 0.00 6.68 6.76 0.00
3 43.30 43.30 0.00 6.78 6.82 0.00
4 33.07 0.00 33.07 3.00 0.00 3.00
5 33.86 0.79 33.07 3.05 11.27 3.00
6 77.17 44.10 33.07 4.47 6.95 3.03
7 77.47 77.47 0.00 5.58 5.60 0.00
8 31.24 31.24 0.00 5.56 5.60 0.00
9 12.94 12.94 0.00 5.51 5.60 0.00
I;VI 23.40 5.95
~VIII 0.49 6.19
l;Vw 47.63 5.95
I:V.~, 24.23 5.95

TABLE5. Cost per exer[gy unit of product (cp) and fuel (CF), relative cost difference (r), cost rates of exergy
destruction (C'D.Fand Co.p)and exergoeconomic factor f for components of the gas-turbine system shown
in Fig. 5. The values of this table are consistent with those shown in Table 4 (case AI; Z values from
Table 3)

Component r = cp -- CF
Cp CF CF CD.F CD.p f
No. Name [S/G J] [S/G J] [%] [$/hr] [$/hr] [%]

I Air compressor 6.76 5.95 13.6 1.91 2.17 34.31

II Regenerator 6.92 5.60 23.6 3.00 3.71 16.64
III Fuel compressor 11.27 6.20 81.8 0.06 0.10 83.91
IV Mixing 6.95 6.71 3.7 1.57 1.62 0.00
V Combustion 4.45 3.04 46.2 10.25 14.99 2.84
VI Gas turbine 5.95 5.60 6.3 1.42 1.51 49.59

Turbine (Fig. 3d) average cost per supplied exergy unit. Similar assump-
c x = c x = c x with X = T O T or (CH and PH) tions are made when the physical exergy is separated
into thermal and mechanical exergy.
or (T, M and CH). (69)
t~-tl /:6M - t ~ tl-tl
Combustion Chamber/Gasifier (Fig. 3e) - - - - - (75)

cc" = cca (incomplete combustion) (70a)

c~ = c~7 (76)
c cn = 0 (complete combustion) (70b)
= ~ = 0. (77)
c cH = ~ " (gasification) (70c)
The costing of exergy losses such as the losses repre-
~ ' ~ = ~ ' ~ + ( 7 2M if L'~t < / ~ + E 2 M (71a) sented by streams 3 and 4 in Fig. 3f is discussed in
t~ - (t~ + t~) t~ - (iT + tl) the following Section 5.4.
The auxiliary Eqs 62 through 69, 70b and 71b
were used to generate the results in Tables 4 through
if /~ > ~ + ~7~. (71b) 7 that refer to the old approach to exergy costing.
Giaggioli and Reistad 4° and Wepfer 43 applied various
Boiler (Fig. Zf ) concepts, called the extraction, the equality and the
cC6H = cO5H and ccH = c crl (72) by-product methods, to establish auxiliary equations.
C~H = 6,CH ..~ C TM (incomplete combustion) Valero and co-workers 71'13z used a systematic pro-
(73a) cedure for formulating auxiliary equations. Detailed
discussions of auxiliary assumptions are included in
c3c " = e c " = 0 (complete combustion) (73b) Refs 1, 8, 32, 40, 43, 67, 71 and 107.
t~" - C~"
P" t~" - t~" In 1990 the author developed a new approach to
= (74)
~PH -- j~PH ~S -- ~7 exergy costing which does not require auxiliary equa-
tions. This approach, which follows the cost forma-
Equation 74 means that exergy is supplied to both tion process and is based only on exergy and cost
the main steam and the reheat steam at the same accounting principles, is presented in the following
244 G. TSATSARONIS

TABLE6. Comparison of the cost per total exergy unit cr°r

tor the streams shown in Fig. 5 and for the following four dx = - , ( ~ x X = T O T , PH, CH, M , T . (79)
cases when capital costs are neglected (Z = 0 for all compo- rh,
nents). All values are given in $/GJ
Similarly, the exergy difference in the regenerator
Stream CTOT (II) ~H _ e~n (=0.1518 M J kg -1) is supplied to the
no. A1 A2 B1 B2 air at an average cost per unit of added exergy of
cp.n given by the following relationships:
1 0.00 0.00 0.00 0.00
2 5.72 5.47 5.85 5.18 d'~" - d'~" d'~" - d:~H + 2,1
3 5.91 7.04 5.87 5.66 Cp,li -- =
4 3.00 3.00 3.00 3.00 E~P - g~" -~" - E~" (80)
5 3.02 3.02 3.02 3.02
6 4.13 4.55 4.12 4.04 In the same way we find that the entire physical
7 5.13 5.65 5.12 5.02 exergy of methane at state 5 (e~H - ee4H = 0.3294 M J
8 5.11 6.73 4.90 5.54
9 5.07 5.46 4.86 5.92 kg -~) is supplied at an average cost of
l,f/t 5.31 5.08 5.43 4.81
~/lll 5.52 5.28 5.65 5.00 ~ " - ~" d: w" + 2i.
/'/¢',,,t 5.31 5.08 5.43 4.81 CP,m E~"- E~ - ~"- Ep. (81)

Case AI: The physical exergy is not split and the old
approach to stream costing is used.
Case A2: The physical exergy is not split and the new
approach to stream costing is used.
Case BI: The physical exergy is split into thermal and
mechanical exergy and the old approach to stream costing
is used.
Case B2: The physical exergy is split into thermal and
mechanical exergy and the new approach to stream costing
is used.
To account for the fact that the methane compressor
(III) is driven by an electric motor, it is assumed in
the calculations that cw.m = Cw,,~t/0.96.
In the mixing process, exergy is removed from
stream 3 and supplied to stream 5. First, we need to
determine the average cost per unit of physical exergy
of stream 5 (ep.w) at which the exergy ~H _ ~ n is
supplied to stream 5

?J~5 - - - -
with the aid of the gas turbine system shown in Fig.
~/~/6 m5 /
5. The following discussion will focus on physical Cp,IV ~ ~--

exergy.
/hs(eg H -- egH)
Initially, we follow the exergy supply to the air up • H

to state 3. In the compressor (I) the specific air
exergy increases from ~H (which is zero) to e~n. The
average cost per unit of physical exergy at which the
difference ~ a - e~H (=0.2007 M J kg -l) is supplied
to the air is given by Cp.i
d,~H _ C?~. ~w, +
Cp,, / ~ n _ ~?~n = E~H _ E~a"
In the above relationships, the second equation ex-
presses the cost balance for the compressor while the
first equation defines the average cost at which physi-
cal exergy is supplied to the air stream.
The cost is shown in Fig. 6 with a dotted horizontal
line. The area below this line (A - 2 - e~H - 0) corre-
sponds to (~H /rh2. This mass specific cost is indi-
cated here by the symbol d
m3 "- - Zlv
~m3 tn6 /
• (82)
m s ( e ~ ~ - e~")
Before the mixing process we have two separate
histories of exergy supply to the air stream and the
21 methane stream. After the mixing process, a single
(78) stream (the mixture) exists. Therefore, we need to
calculate the history of exergy supply to the mixture.
This is done by combining the two lines showing the
history of the costs per unit of physical exergy for air
and methane in Fig. 6 into one line (solid line) which
refers to the mixture of air and methane. The combi-
nation is done using the mass rates as relative
weights.
The average cost at which an exergy unit is sup-
plied to the mixture in the specific-exergy interval

TABLE 7. Comparison of the relative cost difference r between fuel and product and of the cost rate of
exergy destruction ~a.v for the components of the gas-turbine system shown in Fig. 5 for the four cases
(A1, A2, B1, B2) presented in Table 6 and for zero capital costs

Component r = cp or-cF [%] ~'D,F [S/hr]

no. AI A2 BI B2 AI A2 BI B2

I 8.9 8.9 8.9 8.9 1.71 1.63 1.75 1.54

II 19.6 19.6 19.1 19.1 2.76 4.10 2.66 2.83
III 13.2 13.2 13.2 13.2 0.05 0.05 0.05 0.05
IV 3.7 3.7 6.2 6.2 1.36 1.61 1.30 1.35
V 44.9 44.9 44.9 44.9 10.25 10.25 10.25 10.25
VI 3.2 3.2 3.2 3.2 1.31 1.25 1.34 1.18
 Energy systems 245

e pH = 0 . 2 0 0 7

e pH 0.3525 Cp.iv(=216.96)
e pH 0.3294
e; H 0.5482
COST Mixture
PER 14.0
CM,56 (=13.71) 6
PHYSICAL
EXERGY 12.0 CH 4 Cp. iii(=I0.76) c.P.I.I.
(.~II=29).
UNIT 3
10,0 r ......
c PH
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

M1xture M.25
8.0
[SlOJ] Mixture CM02 ( = 6 " 3 4 ) / (=11.28)
6,0 .... ]2
4.0

2.0

I I I [
0.0
0.1 ~0.2 ' 0.~3+ ~_ 014
PH PH PH
ep8 e5 e6 e 5
SPECIFIC PHYSICAL EXEROY e PH [Md/kg]

FIG. 6. Average costs at which each unit of physical exergy is supplied to the air and methane streams
from inlet to state 6 for the gas-turbine system shown in Fig. 5. The solid line corresponds to the mixture
of air and methane.

ej -- ei is denoted by cu,u and calculated by for the mixture. The costs associated with the
specific-exergy value intervals 0-e~n; een2- e pH',5 et a _
Cp,a,ijr~/a "~- Cp,f,ii/~/f e~n; and e7PH- e 6 Pn are CM,O2, CM,25, CM,56 and CM,67,
cu,u = (83)
respectively and are calculated using Eq. 83. Accord-
ing to Fig. 7, the cost flow rates associated with
Here, CP.a,u and Cp.t.u are the average costs at which states 8 and 9 can be calculated directly from the
the air (subscript a) and the fuel (subscript f) were following equations.
supplied in the specific-exergy interval ej - ei.
In the combustion unit, the physical exergy of the ~ P H = mS[CM.O2(ePH __ 0 )
mixture increases by egn - e~6H at an average cost per + cM,zs(eP. -- e2PH)] = rhsd PH (86a)
supplied exergy unit of
or
~PH -- ~PH ~ 6CH Jl- Z v
~'PH = [~Pa - - / h s [ c M , 6 7 ( e P 8 _ eP6n)
Cp,V "~" ~ P H __ E g H - - ~ P . __ ~ P H " (84)
PH - ePH)] (86b)
+ cu,56(e~8 - e~H) + cM,25(es
At the outlet of the combustion unit the gas mixture and
reaches its highest physical exergy in the entire gas-
turbine system. In the downstream units (gas turbine ~'~H = eM,02e9PH"m 9 = d P H ? n 9 (87a)

and regenerator), physical exergy is removed from or

the mixture.
As explained in Ref. 132, when physical exergy is
removed from a stream, we should assume that the
exergy units supplied last to the stream are removed
first (last-in-first-out principle of accounting). In addi-
tion, an exergy unit is removed from a stream at the
same average cost at which it was previously supplied
to the same stream. These principles are applied here
to the mixing unit (for the exergy removal from state
3 to state 6), the gas turbine (from 7 to 8) and the
regenerator (from 8 to 9). The cost of exergy removal
in the mixing unit (Fig. 6) is given by:
-- = cp,nms(e3
/~3k~ 3 1~/6 /
Figure 7 shows the cost history of exergy addition
~,Pgn = ~,gn _ rh9[cM,25(ePn _ e~n)
+ CM,02~e2
l ~PH - - ePgH)]. (87b)
The system of linear equations as expressed by
Eqs 78 through 87 is solved simultaneously to calcu-
late the unknown costs. The results are given in
Tables 6 through 9 and in Figs 6 and 7. Equations
86 and 87 make redundant any auxiliary equations
of the form of Eq. 69 for the turbine and Eq. 64 for
the regenerator. In addition, this approach allows
marginal costs to be considered in the analysis, if
they are available. 132 When thermal and mechanical
exergy are separated, the procedure discussed above
- - e6). (85) for the physical exergy has to be repeated for thermal
and mechanical exergy separately. More details on
this new approach are given in Refs 132 and 135.
246 G. TSATSARONIS

e PH = 0 . 2 0 0 7

e; H = 0.3294

COST e pH = 0.3482
PER 14.0 0M,56 (=13,71)
6 e PH = 0.7600
PHYSICAL
EXEROY 12.0 PH
e8 = 0.2;064
UNIT CM,25 ( =I 1.28
10.0 PH
PH e 9 = 0.1269
c
c
[S/OJ]
8.0 M,02
(=6.54)
6.0 1
iI Cp.y= CM,F~7 ( = 4 . 4 5 ) 7
I
4.0 i 1I

2.0 1I 1I
1 1
i , I
0.0
t to.2_ttt .4
OH OH PH PH PH
' 0'6 1' 018 '
PH
e9 e 2 e8 e5 e6 e7
SPECIFIC PHYSICAL EXEROY e pH [Md/kg]

FIG. 7. History of the average cost per unit of physical exergy of the combustion gas.

TABLE 8. Cost flow rates associated with the total, chemical, chemical exergy. This a p p r o a c h leads to costs that
thermal and mechanical exergy for the streams shown in
Fig. 5 and for case B h The physical exergy is split into are independent of the reference state used to calculate
thermal and mechanical exeqg)' and the new approach to the chemical exergy values.
exergy costing is used. The Z values considered here are Sometimes it is convenient to use, in addition to
given in Table 3. All cost flow rates are given in $/h the cost flow rate associated with the total exergy o f
Stream a material stream, a cost flow rate due to non-exer-
no. ~TOT ~H (~T ~'M getic effects (e.g., costs associated with a non-energy
intensive process or with non-exergy related material
1 0.00 0.00 0.00 0.00 properties). In this case the total cost flow rate
2 21.74 0.00 5.80 15.94
a c c o m p a n y i n g the nth material stream is denoted by
3 40.53 0.00 25.05 15.48
4 33.07 33.07 0.00 0.00 L', (without superscript) a n d given by
5 33.81 33.07 0.00 0.73 C. = ~o~ + ~E. (88)
6 74.34 33.07 25.61 16.65
7 74.64 0.00 58.89 15.75 T h e cost rate C'T°T is calculated according to Eq. 50,
8 33.83 0.00 33.35 0.48
9 15.64 0.00 15.45 0.19 whereas ~NE is the cost rate due to non-exergetic
I;I,'~ 20.74 effects. Examples o f this term include the cost flow
~,z 0.44 rate associated with (a) a water stream at the outlet
l;Vvl 42.21 o f a water t r e a t m e n t unit, (b) an oxygen stream at
~;V,~, 21.45
the outlet o f a n air separation unit, (c) limestone
supplied to a fluidized-bed boiler or gasifier, (d) iron
The a u t h o r a n d co-workers1 a a introduced the reac- supplied to an energy-intensive metallurgical process
tion exergy as the economically relevant p a r t o f a n d (e) an inorganic chemical supplied to a n energy-

TABLE 9. Cost per exersy unit of product (cv) and fuel (cv), relative cost difference (r), cost rates of exergy
destruction ~'a.v and Cc.v and exergoeconomic factor f for the components of the gas-turbine system
shown in Fig. 5. The values of this table are consistent with those shown in Table 8 (case B2; Z values
from Table 3)

Cp -- CF
r =
Component cp CF CF ~D,F CD,V f
no. [$/GJ] [S/G J] [~o] [$/hr] [$/hr] [%]

I 6.02 5.27 14.2 1.70 1.94 37.08

II 6.85 5.57 23.0 2.99 3.68 16.70
III 10.47 5.49 90.6 0.05 0.10 85.48
IV 6.80 6.41 6.2 1.50 1.59 0.00
V 4.45 3.04 46.2 10.25 14.99 2.84
VI 5.27 4.94 6.7 1.26 1.34 52.71
 Energy systems
intensive chemical process. In addition, the term ~rq~
in Eq. 88 represents a convenient way in thermoeco-
nomics for charging some costs from one component
directly to another component which should be
charged with these costs. In this case the value of
(7~E remains unchanged as the nth material stream
crosses intermediate components. More details about
(7NE are given in Ref. 67
5.4. Costing of Exergy-Loss Streams
The exergy loss together with the exergy destruc-
tion represent what the layman calls 'energy waste'.
The exergy loss is the transfer of exergy out of the
total plant associated with a mass or energy stream
rejected to the environment. The term 'destruction' is
used to identify the unrecovered 'loss' of exergy
within a component, as distinct from the loss of
exergy through a plant output stream. The costing of
exergy loss depends on the purpose of the thermoeco-
nomic analysis. From the cost balance (Eq. 54) it is
apparent that the lower the monetary rate associated
with the exergy loss (~?L), the higher the cost rate of
the product (i.e., the cost rates of the desired output
streams which are part of the product). The purpose
of the thermoeconomic analysis could be:
(I) To calculate the costs of the total plant products
(for plants generating more than one product). The
easiest approach in this case is to assume that t~L = 0
for all exergy losses in the plant. Thus, the monetary
flow rates associated with exergy losses are auto-
matically charged to the final plant products. If
additional money needs to be spent until a material
stream representing an exergy loss is finally deposited
in the environment (e.g., for ash handling equipment
or for a flue gas desulfurization system that is not
included within the boundaries of the plant being
analyzed), the term L'L must be negative to charge
the additional expenses associated with final deposi-
tion to the products of the component where the loss
stream is originated (e.g., the boiler).
(2) To understand the cost formation process and
the cost flow in the plant or to optimize specific
variables in a single plant component. In this case, the
costing of all plant components is conducted as
discussed in Section 5.3 as if all exergy loss streams
were to be further used in the total plant. In so
doing, we calculate the monetary loss ((?L = cV-en~en~t
cnL-cHx
CF /-:,L ] accompanying each exergy loss. For in-
stance, the monetary loss accompanying heat transfer
from a plant component (operating at a temperature
T > To) to the surroundings is L'L Q = dHL~LQor, more
precisely, C'~ = c~-/~. This information can be used
to find the cost optimal insulation. In the end, all
monetary losses associated with exergy losses need to
be charged to the final products of the plant and
recovered through the sale of the products.
(3) To optimize the total plant. For overall optimiza-
tion puposes we assume that the exergy loss in a
247
plant component can be avoided by spending more
capital and we set in the cost balance
C'L.k = 0. (89)
In general, very few components have exergy losses
that need to be distinguished from the exergy destruc-
tion in practical applications. In the thermoeconomic
evaluation the concept of exergy loss is applicable to
the totalplant rather than to a single plant component
that happens to have an exiting stream that is not
further used in the plant. Often this component
should not be penalized for the exergy loss (e.g., flue
gas leaving the plant at the lowest allowable tempera-
ture). Then the costing approach discussed under
condition 2 should be applied.
5.5. Cost of Exergy Destruction
The cost associated with the exergy destruction in
a process is a 'hidden' cost, but a very important
one, that can be revealed only through a thermoeco-
nomic (exergoeconomic) analysis. In the cost bal-
ance, there is no cost term directly associated with
the exergy destruction; this, however, affects directly
the cost of product (Cp.k): assuming that the right
side of Eq. 61 remains constant, the higher the exergy
destruction, the lower the L'e.kvalue and, consequently,
the higher the Ca,k value. The cost flow rate associated
with the exergy destruction in the kth component ((?D.k)
can be represented either by the cost of the additional
fuel that needs to be supplied to this component to
cover the exergy destruction and generate the same
exergy flow rate of the product (Ep.k)
CD.k = CD F.k = CF,kED,k when Ep,k = constant,
(90a)
or by the monetary flow rate lost at the product side
(assuming that cp is not the cost but the price of the
product) because of the exergy destruction 4a
CD.k = CD.P.k = CP,kED,k,when ~'Vk = constant.
(90b)
The true cost of exergy destruction in the kth
component lies between the two values given by Eqs
90a and 90b; this cost could theoretically be calcu-
lated if we would split the total process in this
component into a large number of subprocesses and
would calculate the cost of exergy destruction for
each subprocess; the larger the number of subpro-
cesses, the smaller the difference between cv and cp.
This approach is impractical, however. In practical
applications we prefer to use Eq. 90a for estimating
the cost of exergy destruction. This represents a
conservative approach with respect to the required
optimal investment costs.
5.6. Relative Cost Difference
By combining Eqs 21, 59, 60 and 61, we calculate
248
the following relationships for the cost difference
(Ack) which expresses the increase in the average cost
per exergy unit between fuel and product:
2~ + c~.~(E~,~ +/rL.~) - ~L,~
ACk ~ Cp,k -- CF,k
gP.k
or
Ack =- Cp,k -- CF,k : component to improve the cost effectiveness of the
EF.k
Equations 91 a and b illustrate the real cost sources
in the kth component which are the cost rates associ-
ated with capital costs (Zk), exergy destruction
(CF,kED.k o r Cp.kF-~D.R), and exergy loss (CF.k/~L. k - -
(?e,k or Cp.kF.L,R- /£L.k)- In the following we will
assume that (TL.kis equal to zero.
The cost optimization of the kth component in-
volves finding the optimum trade-offs among these
cost sources. It is apparent that any part of exergy
destruction or of exergy loss that either does not
contribute to the reduction of capital investment
costs and operating and maintenance costs or that
can be eliminated by relatively small increases in
these costs is cost ineffective and should be elimi-
nated. A positive role of exergy destruction and
exergy loss is associated with their usefulness in
reducing the capital costs in the total plant or the
fuel costs in other components.
The relative cost difference, rk, between average
cost of product and average cost of fuel for the kth
component is obtained by dividing Eq. 91a by CF,k
and neglecting the CL,kterm
Cp.k -- CV.k Zk 1 -- Ck
r k - - - - - - + -
CF,k CF,kF--~p,k
with the exergetic efficiency ek according to Eq. 20.
The relative cost difference is a useful variable for
evaluating and optimizing the kth plant component.
6. T H E R M O E C O N O M I C EVALUATION
A complete thermoeconomic evaluation of an
energy-intensive process (system) is based on the
following variables calculated for the kth compo-
nent:
• exergetic efficiency, el
• exergy destruction and exergy loss, ED.k and
ELk
• exergy ratios ED,R/EF.tot and EL,k/EF,tot
• capital costs (Z) associated with capital invest-
ment and operating and maintenance expenses
• cost of exergy destruction, (70.k
• relative cost difference, rk, and
• exergoeconomic factor,~.
G. TSATSARONIS
J~ = . (93)
z~ + c~,~(ED.~ +/~L.~)
The exergoeconomic factor, j~, expresses (a) the
ratio of the capital cost to the total cost (due to
investment, O&M, exergy-destruction, and exergy-
(91a) loss costs) in the kth component and (b) the percent-
age contribution of the investment and O&M cost
rate to the relative cost difference, rk.
The following rules can be applied to the kth
total system:
(91b) (1) The higher the relative cost difference, the
more attention should be paid to this component,
particularly when also the cost rates Zk and (?D.k are
high.
(2) The higher the cost rates Zk and/or Co.k, the
more attention should be paid to this component.
(3) When the rk value is high, the exergoeconomic
factor (J~) identifies the major cost source (capital
cost or cost of exergy destruction) causing the high
rk value.
(4) If both the rk a n d ~ values are high, it might be
cost effective for the system to reduce the capital
investment for the kth component at the expense of
the component efficiency.
(5) If the rk value is high and the J~ value is low,
try to improve the component efficiency by increas-
ing the capital investment.
(6) Eliminate or reduce any steps (subprocesses)
that increase the exergy destruction or exergy loss
without contributing to the reduction of capital in-
vestment or fuel costs.
(7) If a component has a relatively low exergetic
efficiency, or a relatively large value of exergy-destruc-
tion or exergy-loss ratio (y), an increase in the exer-
getic efficiency might be cost effective.
(92) In the performance evaluation of an existing p l a n t
~k
the capital investment costs are sunk costs. If the
O&M expenses are also neglected, only fuel costs
are considered in the thermoeconomic evaluation
which is based now on the variables /S'~ and c~ for
the nth stream as well as Ac/, rk and ~6.k for the kth
component. The superscript - used in these var-
iables denotes that all investment and O&M costs
have been neglected in calculating the values of the
thermoeconomic variables. These variables assist in
understanding the effects of exergy destruction and
exergy loss (and, consequently, of a malfunction) in a
plant component on the performance of the other
components and the total system.
The ratio
c~=__ whenZ'k = 0 , k = l ..... ND (94)
E,
represents the cost per exergy unit for the nth flow
stream when the term ~, is zero for all plant compo-
nents, s The value of c~ is equal to unity for the raw
fuels supplied to the total plant and greater than one
 Energy systems
for all other streams in the plant (excluding some
exergy-loss streams). All increases in the values of c~
are caused by exergy destruction and exergy loss. In
the thermoeconomic evaluation of the design of an
energy system, c~ represents the lowest limit for the
cost per exergy unit of the nth stream.
Valero and co-workers 71'~34 call the variable ~
the 'exergetic cost'. This variable expresses the exergy
flow rate that needs to be supplied to the total plant
to generate the exergy flow rate E, of the nth stream
in the plant being evaluated.
Tables 4 through 9 contain results obtained from
the thermoeconomic analysis of the system shown in
Fig. 5 in different cases referring to (a) the treatment
of physical exergy (cases A and B), (b) the use of the
old (case 1) and the new (case 2) approach for
costing of exergy streams and (c) the treatment of
capital costs (zero in Tables 6 and 7). Among these
cases a preference should be given to case B2. The
results presented in the Tables will enable the reader
to understand better the concepts discussed here and
their application.
A thermoeconomic evaluation conducted for the
gas-turbine system (Fig. 5) and based on the results
presented in Tables 4 through 9 would conclude the
following. The fuel compressor (III) has the highest r
andfvalues among all components (Tables 5 and 9).
Therefore, it might be cost effective to use a less
expensive fuel compressor and accept a higher cost
of exergy destruction in this component. The combus-
tion chamber has the second highest r value and by
far the highest cost rate associated with exergy de-
struction (see values of either CD,F or CD,p in Tables
5 and 9). The logical conclusion would be to try to
decrease the exergy destruction in the combustion
process by spending more capital to increase the air-
preheating temperature T3 and by adding a reheat
stage. The regenerator has still significantly high r,
CD.V and CD.e values and a relatively low f value.
Thus, an increase in the exergetic efficiency of this
component (associated with a higher capital expendi-
ture to increase the total surface area) might be cost
effective. This conclusion agrees with and enhances
the previous conclusion from the thermoeconomic
variables for the combustion process.
Two final notes of caution are required, one refer-
ring to the gas-turbine system and another general
one. In any design of the gas-turbine system (including
the cost optimal design) the combustion process will
have the highest exergy destruction and cost of exergy
destruction rates among all components, as well as a
high r value and a very lowfvalue. The terms ~D,F,k
and ~D.P,k already include the effect of the exergy
destruction in the kth component on the total cost
(capital cost and cost of exergy destruction) of some
other components. It is, therefore, not allowed to add
changes in the values of ~'D,F,k and ~TD.p.kin order to
estimate the combined effect of simultaneous changes
in more than one plant component.
Thermoeconomic evaluations of energy systems
249
are presented among others in Refs 1, 8, 10, 23, 30,
31, 32, 34, 37, 39, 42, 49, 57, 60, 62, 63, 67, 70, 71,
73, 76-81, 84, 85, 87-91, 99, 100, 111, 121, 124, 130,
132 and 134-139.
7. T H E R M O E C O N O M I C O P T I M I Z A T I O N O F A N ENERGY
SYSTEM
The thermoeconomic evaluation discussed previ-
ously requires that all thermodynamic and cost data
are known for the plant configuration being evalu-
ated. For the optimization, however, we additionally
need to have a thermodynamic and a cost model The
thermodynamic performance simulation model will
allow a prediction of the effect of some important
thermodynamic variables on the plant design. The
cost model should permit detailed calculation of cost
values for each given set of the thermodynamic
values. In general, a thermodynamic model can be
developed more easily and is more reliable than a
cost model. The problems associated with a cost
model are that the cost information is not always
available or reliable; when it is available, it is often
not in the required form.
Several mathematical methods 14°-143 can be used
for optimizing the design and operation of energy
systems. When the system is very complex, however,
it becomes practically impossible to just use the
available mathematical tools because of the very
large number of equations, restrictions and variables
involved. In addition, the perfect cost information
which was assumed here is never available, at least
not in a form that would allow direct use of this in-
formation in a global optimization problem. Thermo-
economics offers some additional capabilities in
optimization that are not available in the conven-
tional optimization techniques. These capabilities are
discussed in the following.
It is apparent that for complex systems with a
large number of non-linear and, usually, non-explicit
variable relationships the term optimization implies
improvement rather than calculation of a global
mathematical optimum.
7.1. Cost-Optimal Exergetic Efficiency
The following approachS' 1o,24,61,6v.135 refers to the
optimization of a single plant component but pro-
vides useful information also for optimizing complex
plants. This approach is based on the following two
assumptions which constitute the cost model:
(1) The total net capital investment Ik, which is
associated with the kth plant component and is calcu-
lated at the beginning of the plant operation period,
is approximated by the following relationship:
[k = Bk( ~k y'j~.~. (95)
\1 --ek/
250 G. TSATSARONIS

The constants Bk, nk, and mk depend on the compo- with

nent being considered. Equation 95 is assumed to be
n'+ 1 /(ilk "+- yk)Bkr/k
valid within a certain range of the design options or
Fk : ~f TCF,kf-'l,k m~' , (102)
design conditions for the kth component. The term
/~'.'k expresses the effect of component size on the
value of Ik. It is interesting to note that the reciprocal rOPT : qr/k- I Fk -- nk + l (!- e°PX'~
of the term leg~(1 -- ek)], which expresses the effect
r/k r/k \ e ° P ~ , / ' (103)
of efficiency on the capital investment of the kth . fl - eOPT~
component, is equal to the last term of Eq. 92. When A-~o.
PT ----Ep.kFk = E p . k [ x T ~ , (104)
some cost information on different design conditions
of the kth component is available, the least square
method can be used to calculate the values of the zOpT__C EJk__CF pk(1--' T (105)
constants r/k, mk and Bk in Eq. 95.135 Application of r/ /7 \ ,~OPT /1'
the least square method is simplified if the value of (1- ~OPT~
the capacity exponent mk is assumed. However, Eq. ~"OPT = CF.kL'p,kFk= CrkLTp~k - - , (106)
O,F,k \ eOcT //
95 could also be used in the absence of complete cost
information by assuming the values of the exponents
nk and ink. and
(2) The annual operating and maintenance costs 1
(excluding fuel costs) attributable to the kth plant foeT
k _ (107)
1 +nk"
component may be approximated by
Z~k M = Yk Ik q- (~OkrJ~p,k -~- Rk. (96) The variable Fk is the (r/k -{--l)th root of the so-called
thermoeconomic similarity number, initially formu-
Here, 7k is a coefficient that takes into account the lated in Ref. 24 for a total plant and for nk = mk =
part of the fixed operating and maintenance costs 1.
that depends on Ik; 09k is a constant expressing the The thermoeconomic similarity number considers
variable ievelized operating and maintenance costs the most important factors that affect the optimum
of the kth component, r is the annual time of plant values of the variables ek, rk, E'D,k, Zk, and C'D,F,k.
operation at the nominal capacity, and Rk includes The cost-optimal exergetic efficiency, for instance,
the remaining costs that have not been considered in increases with increasing fuel cost and annual
the first two terms of Eq. 96. number of hours of plant operation and with decreas-
These assumptions lead to the following expression ing capital recovery factor, fixed operating and main-
for the total annual costs (capital investment costs tenance cost factor, and exponent nk. It should be
and O&M costs, excluding fuel costs) attributable to noted here that the exponent r/k expresses the optimal
the kth component ratio between the cost of exergy destruction and the
capital cost in a component:
Z k = Z CI ~- Z OM : (ilk "~- 7k) lk -]- (DkrF--~p,k "~ Rk
'OPT
(97) D,F,k
nk -- (108)
zOPT
where flk is the capital recovery factor. The sum of
the t e r m s ,~kkI and ~OM for the kth component in a
Different plant components are characterized by
cost balance (e.g., Eqs 54 or 61) is obtained by
different values of the thermoeconomic exponents nk
dividing Eq. 97 by r.
and mk. Electric motor driven pumps, for instance,
Now the total costs associated with the kth compo-
will usually have nk values lower than 0.6, while
nent can be minimized by assuming
steam turbines are expected to shown nk values higher
//:P.k = constant (98) than 0.9. The nk values for heat exchangers and
chemical reactors will largely depend on the purpose
and
and the working substances used in these plant com-
CF.k = constant. (99) ponents.67
The minimization of the cost per exergy unit of prod- Equations 101 through 108 apply to the optimiza-
uct tion of a single component as illustrated in Fig. 8
which shows the variable cp as a function of the
CF,kJ~F,k "4- Z k
Cp.k -- -- m i n (100) exergetic efficiency e and the exergy destruction (ED).
Ep.k This optimization approach can also be used for a
group of components by appropriately adjusting the
under the above assumptions (Eqs 95 through 99)
definition of fuel and product and the calculation of
leads to the following expressions for the cost optimal
the capital investment in Eq. 95. However, the more
values of the variables ek, rk, ~D.k, Zk, CD,F,k andfk; components that are considered in a group, the more
1 difficult it is to develop an acceptably accurate cost
/~OPT = - - (101)
lq-Fk equation such as Eq. 95.
 Energy systems 251

PRODUCTt

/
COST / Total Cost
(Co)

Cp,A /
"I Fuel Cost
[ Capital Cost Dependent i
~ x e n g y D e ~

T ........... ¢o,,t°, cost

/ I I I oo.t~o. C ~ a l c0st ,...0.°.e.t

cF - l [ I " of Exergy DesLruction

i%
D,A
EXERGYDESTRUCTION (E"n )
EXERGETIC EFFICIENCY (C)

FIG. 8. Schematic of the contributions of fuel cost and capital cost to the total product cost as a function
of exergy destruction and exergetic efficiency.

The conditions that optimize a single component
usually do not optimize the total plant. Single compo-
nent optimization can be used in a preliminary optimi-
zation phase for some important components that
dominate the cost picture, i.e. the components with
the highest Z and E'o values.
The assumptions expressed by Eqs 98 and 99 are,
mathematically speaking, not fulfilled for compo-
nents of complex plants. Therefore, the single compo-
nent optimization expressed by the set of Eqs 101
through 108 represents only an approximation when
applied to a system of components. We should not
forget, however, that an approximation that can be
easily implemented is always welcome, particularly
when all other conventional approaches fail to truly
optimize a single component of a complex system
(often just because they are based on energy costing
and, therefore, fail to identify the real fuel costs).
7.2. Design Optimization
The goal of design optimization of an energy system
is to find the structure and the values of the system
parameters that minimize the cost of the final products
considering restrictions imposed by the desired reliabil-
ity, availability, maintainability, operability and envi-
ronmental impact of the system. Finding the cost
optimal structure may be more important than opti-
mizing the parameters of a given structure. An exergy
analysis and a thermoeconomic evaluation are very
useful tools in solving both optimization problems.
References 134 through 137, published in the same
issue of Energy--The International Journal, present
four different thermoeconomic approaches applied
to the same problem, the design optimization of a
cogeneration plant. Reference 135 uses in the total-
plant optimization the cost optimal exergetic effi-
ciency (Eq. 101) and the cost optimal relative cost
difference (Eq. 103) for each plant component. A
comparison of the actual with the cost optimal values
for e and r identifies (a) in what direction (higher or
lower efficiency) the design of a component should
be changed to reduce the total costs associated with
it, and (b) the components that are far away from
their optimal r value. If some of the latter happen to
be major components (characterized by large (TD
and/or Z values), then these components should be
modified first in the direction indicated by the com-
parison between e and e°pr. Thus, the following
dimensionless variables (relative differences) are de-
fined:
r k - - r OPT
Ark -- - - (109)
r kOPT
OPT
~k - - k
Ark -- - - (110)
CkOPT
These variables, together with (:D and Z, assist in
improving the design structure, in estimating the
relative importance of decision variables in a given
design structure and in prioritizing the required
changes to improve the cost effectiveness of a given
design. The application of an iterative optimization
process using the variables Ark and Aek on a cogenera-
tion system is discussed in Ref. 135. The cogeneration
system, which is a relatively simple system, is used
just as an example to illustrate the use of the variables
Ark and A e k f o r optimization purposes.
7.2. !. Optimizationof major plant components
As mentioned before, an energy-conversion plant
may contain major components that dominate the
252 G. TSATSARONIS
overall cost picture because the cost rates Z, and CD
associated with these components are much larger
than the corresponding cost rates for the other plant
components. Major components usually are associ-
ated with chemical reactions and/or heat transfer.
Examples of such components include the boiler of a
steam power plant, the gasifier and the gas-turbine
system of an integrated gasification-combined-cycle
(IGCC) power plant, and the steam reformer of a
chemical plant.
In the preliminary plant optimization phase, the
design of major plant components should be opti-
mized in isolation f r o m - - a n d sometimes ahead of the
detailed design of--the remaining plant components.
Thermoeconomics facilitates--and often enables--
this isolated component optimization where we calcu-
late the values of the major decision variables directly
affecting the performance of a major component
that minimize the relative cost difference for the com-
ponent being considered. In this optimization we
should use the relative cost difference, r (instead of
the cost of product, cp that is used in Fig. 8) as the
variable to be minimized because the variable r
allows changes in the cost of fuel (cv) for a major
component to be appropriately considered in the
iterative optimization. Any conventional optimiza-
tion method could be applied to minimize the r value
of a major plant component as a function of some
decision variables.
The values of the decision variables obtained in
this preliminary component optimization will be very
close to their final optimal values. This approach
considerably reduces the time and efforts required
for optimization of a complex plant. References 138
and 144 discuss the application of this approach to
the gasifier of an IGCC power plant. The cost opti-
mal gasification temperature for the total system was
obtained by minimizing the r value for the gasifier.
The optimization of a major plant component is
facilitated when the component is at the front end of
the plant. For other major downstream components
we cannot automatically expect that the optimization
of a component in isolation from the remaining
components will lead to a plant optimization. Some
additional iterations will usually be necessary to opti-
mize the total plant. In this case, the decision variable
value obtained through the first optimization of a
major component in isolation represents a good start-
ing value for the iterative optimization of the total
plant.
7.2.2. Other optimization approaches
Most of the approaches discussed so far are based
on algebraic relationships and average costs. A math-
ematically rigorous cost optimization requires mar-
ginal costs and calculus procedures. The methodol-
ogy introduced by Y. M. E1-Sayed and R. B. Evans 22
and the subsequent developments are not discussed
here. An excellent review of the various aspects associ-
ated with this methodology is presented in Ref. 107.
Particularly interesting are also the optimization
approaches by R. B. Evans, C. A. Frangopoulos and
M. R. von Spakovsky, ~7"136"137'145 148 W. J. Wepfer
and co-workers, 5°'s2 A. Valero and co-work-
ers, 7 0 ' 7 1 ' 1 3 4 A. L. London and R. Shah, 51 A.
Bejan, x°a'149 Y. M. E1-Sayed, Is° S. Aceves-Saborio,
J. Ranasinghe and G. M. Reistad 151'152 and L.
Lin. 1~5 Suitability of each approach for a particular
application depends on (a) the amount of available
information on performance and cost of each system
component, (b) the complexity of the total system,
(c) the objectives of optimization and (d) the re-
quired accuracy.
The major problem in the optimization of complex
systems is the lack of cost data and appropriate
simulation techniques. There is a fundamental differ-
ence between the way manufacturing engineers calcu-
late costs and the way these data are needed for
optimization purposes, i.e. as functions of thermody-
namic variables. As long as the cost and performance
data required for a rigorous global optimization are
not available, a thermoeconomic evaluation based
on algebraic techniques (as discussed in Sections 5
through 7) combined with iterative optimization tech-
niques is probably the best approach to the improve-
ment of large energy systems.
If all required cost and performance data are avail-
able, conventional optimization procedures that do
not use costs of exergy might be preferable for the
optimization of relatively small systems. However,
compared with conventional procedures, the advan-
tages of a thermoeconomic approach increase with
increasing complexity of the energy system, particu-
larly when chemical reactions are involved, regardless
of the available information on cost and perform-
ance.
7.3. Guidelines for Designing Energy Systems
Several guidelines for developing and improving
the design of an energy system can be derived from
exergetic and thermoeconomic considerations. The
following is a list of general and process-specific
guidelines, some of which are based on Refs 138,
144, 153, 154 and 155.
7.3.1. General guidelines
• When assessing the possibility of improving the
performance and cost effectiveness of a plant
answer the following questions:
- - What is the purpose of each process?
- - Is each process necessary?
- - Does the exergy destruction in each process
offset capital investment or fuel costs?
- - What are the available options for each
process?
- - W h a t is the estimated performance and cost
effectiveness of each available option?
 Energy systems
-- What is the impact of modifications being con-
sidered on the remaining plant components?
• Use simplified exergy destruction and cost effec-
tiveness calculations as a guide to process modifica-
tions, particularly modifications in system struc-
ture.
• Some exergy destruction and exergy losses can be
avoided, while others cannot. Concentrate efforts
on those that can.
• Avoid processes using excessively large thermody-
namic driving forces (differences in temperature,
pressure and chemical composition).
• The larger the mass flow, the larger the need to
use its exergy effectively.
• Focus efforts initially on the cost dominant compo-
nents and then consider the improvement of minor
components.
• Prioritize the changes required to improve the
cost effectiveness of the entire plant using the
variables (?i~,Z, Ark and Aek.
• Optimize the cost dominant components by mini-
mizing the relative cost difference r.
• Distinguish between design parameters that affect
both efficiency and product costs and parameters
that affect only the product costs. Optimize the
first set of parameters first.
7.3.2. Process specific guidelines
• Avoid unnecessary or cost-ineffective heat transfer.
- - The higher the temperature in thermal processes
(T > To), the more valuable the heat transfer
and, consequently, the more a direct heat trans-
fer to the ambient, to cooling water or to a
refrigeration stream must be avoided.
- - T h e lower the temperature in a refrigeration
process, the more valuable the heat transfer
and, consequently, the more a direct heat trans-
fer to the ambient or to a hot stream must be
avoided.
--Avoid the use of intermediate heat transfer
fluids when exchanging heat between two
streams.
- - T h e cost optimal temperature difference in a
single heat exchanger increases with increasing
average temperature.
--In the design of heat exchanger networks
(HEN)
(a) try to match streams where the final tempera-
ture of one stream is close to the initial tempera-
ture of the other;
(b) if there is a big difference in the heat capac-
ity rates of two streams exchanging heat, con-
sider splitting the stream with the larger heat
capacity rate;
(c) for heat exchangers operating in the same
temperature range, try to reduce the heat transfer
rate in the heat exchangers with the highest ,~,/L'p
ratios by increasing the heat transfer rate in the
heat exchangers with the lowest Z./L'p ratios; and
253
(d) after developing a design configuration, use
the 'pinch method'~ 56 to evaluate the HEN.
• Minimize the mixing of streams with differences
in temperature, pressure or chemical composition.
• Minimize the use of throttling; check whether
expanders are cost effective.
• Maximize the use of cogeneration of power and
process steam in a plant.
• In cogeneration systems consider steam generation
at the highest available pressure level.
• In chemical reactions consider preheating the reac-
tants.
• Minimize the use of combustion. When using com-
bustion processes, minimize the excess air.
• In selecting compressors, turbines and motors
keep in mind that the cost per unit of exergy
destruction for these components is, usually,
higher than in the remaining plant components;
therefore, the most efficient options for these com-
ponents could be the most cost-effective alterna-
tives.
8. CLOSURE
Exergy analysis and thermoeconomics are very
useful analytical tools in energy engineering. Their
practical importance has been demonstrated in
several studies. Exergy analysis is now often pre-
sented in introductory courses in thermodynamics
and may be considered as a fully developed disci-
pline. However, in order to convince more energy-
engineering practitioners to adopt exergy analysis as
a working tool, additional work is required to stand-
ardize and simplify the nomenclature and exergy
calculation methods.
The field of thermoeconomics has matured to the
point where it is a valuable tool for improving the
design, operation and maintenance of energy sys-
tems; however, it is not yet a fully developed disci-
pline. Thus, studies involving further development of
basic aspects of thermoeconomics are still being con-
ducted in parallel with thermoeconomic applications
to practical problems. Today, thermoeconomics has
reached a stage of development nearly free of some
arbitrary assumptions used in the past. Both exergy
analysis and thermoeconomics may be considered to
be general and objective methodologies for analyzing
and optimizing energy systems, However, the need
for simplification, standardization and further devel-
opment of thermoeconomic approaches is still large
and urgent.
The effectiveness of improving the efficiency and
reducing the costs in the design or operation of an
energy system increases when engineers understand
the real causes of inefficiencies and the true cost
sources. An exergy analysis identifies the location,
cause and magnitude of the real thermodynamic
losses (exergy destruction and exergy loss). A thermo-
economic evaluation identifies the location and
254 G. TSATSARONIS
cause of the cost sources, calculates their magnitude
and compares their effects on the costs of the prod-
ucts. All this information, complemented by the engi-
neer's intuition and judgment, assists in the improve-
ment of the efficiency and reduction of the product
costs in energy systems by identifying the required
changes in structure and parameter values much
faster than traditional approaches. Decisions about
the design, operation and repair or replacement of
equipment are facilitated.
In addition, thermoeconomics provides an objec-
tive cost allocation to more than one product of the
same process. For instance, a thermoeconomic analy-
sis of a cogeneration plant (which produces electricity
and process steam) will provide the cost of steam
and the cost of electricity separately. The cost ratio
of steam to electricity calculated by the analysis does
not necessarily have to be reflected in their selling
prices, but the company that operates the plant
should know the real cost of generating each form of
energy.
A thermoeconomic analysis also shows how much
raw fuel is required to produce each stream in the
system. Finally, thermoeconomics helps managers
decide how to allocate research and development
funds to improve plant components that contribute
most significantly to the product costs.
It is true that many conclusions obtained by a
thermoeconomic analysis could also be obtained
through a large number of conventional energy and
economic analyses. The advantage of thermoeconom-
ics is that it replaces an expensive and subjective
search for cost reduction with an objective, well-
informed, systematic and, therefore, shorter search in
which all of the cost sources are properly identified
and evaluated. The savings in both engineering and
computer time are significant. In particular, applica-
tion of thermoeconomic analysis to new energy
system concepts and complex plants (especially those
with important chemical reactions) results in signifi-
cant savings in design costs and costs of plant prod-
ucts.
Finally, compared with engineers using the tradi-
tional energy and economic analyses, those applying
thermoeconomics develop, in general, a better under-
standing of the performance of energy systems and
the interactions between performance and economics,
as well as more confidence in their ability to improve
energy systems. It is to be expected that in the future
more energy-engineering practitioners will apply ther-
moeconomic techniques to energy systems. The
reader should always keep in mind that controversial
issues (such as what reference model should be used
for calculating chemical exergy or what exergy-cost-
ing approach should be applied) are in almost all
cases practically unimportant for the conclusions
to be drawn from a thermoeconomic analysis.
These issues should not deter an engineer in any
way from applying thermoeconomics to an energy
system.
Acknowledgements--The author wishes to acknowledge the
useful comments made by Professor Richard A. Gaggioli
and two other reviewers on the first draft of this paper. The
assistance of Mr David W. Price in reviewing the final
paper and preparing the final figures is appreciated.
REFERENCES
1. GAGGIOLI,R. A., Efficiency and costing, A. C.S. Sympo-
sium Series 235, 3 (1983).
2. TSATSARONIS,G., Energy economics and management
in industry, Proc. European Conference I, 151 (1984).
3. HAYWOOD,R. W., J. Mech. Eng. Sci. 16, 258 (1974).
4. BOSNJAKOVlC,F., Arch. Wiirmewirt. 19, 1 (1938).
5. GASPAROVIC, N., Brennstoff-Wiirme-Kraft 13, 502
(1961).
6. WEPFER,W. J., Applications of the second law to the
analysis of energy systems, Ph.D. Dissertation, Univer-
sity of Wisconsin, Madison (1979).
7. LIu, Y. A. and WEPFER,W. J., Efficiency and costing,
R. A. Gaggioli, (Ed.), A.C.S. Symposium Series 235,
415 (1983).
8. TSATSARONIS,G. and WXNHOLD,M., Thermoeconomic
Analysis of Power Plants, EPR1 Final Report AP-3651,
Palo Alto, CA (1984).
9. WALL,G., Exergy-A Useful Concept, Physical Resource
Theory Group, Grteborg (1986).
10. SZARGUT, J., MORRIS, D. R. and STEWARD, F. R.,
Exergy Analysis of Thermal, Chemicaland Metallurgical
Processes, Hemisphere, New York (1988).
II. RANT,Z., Forsch.-Ing.-Wes. 22, 36 (1956).
12. MORAN, M. J. and SCIUBBA, E. (Eds), Second law
analysis of thermal systems, ASME (1987).
13. K~NAN, J. H., Trans. ASME54, 195 (1932).
14. BENEDICT,M. and GYFTOPOULOS,E. P., Thermodynam-
ics: second law analysis, R. A. Gaggioli, (Ed.), A.CS.
Symposium Series 122, 195 (1980).
15. EVANS,R. B. and TRmUS, M., A contribution to the
theory of thermoeconomics, UCLA, Dept. of Eng.
Report No. 6243, Los Angeles (1962).
16. EVANS,R. B. GRELLIN,G. L. and TRmus, M., Principles
of Desalination, K. S. Spiegler, (Ed.), Academic Press,
New York, 21 (1966).
17. OBERT, E. F. and GAGGIOLI,R. A., Thermodynamics,
McGraw-Hill, New York (1963).
18. GAGGIOLI,R. A., Thermodynamics and the non-equilib-
rium system, Ph.D. Dissertation, University of Wiscon-
sin, Madison (1961).
19. BERGMANN,E. SCHMIDT,K. R., Energie und Exergie,
VDI, Diisseldorf, 63 (1965).
20. FRATZSCHER,W. and KLrDITZ, D., Chem. Technol. 29,
654 (1968).
21. SZARGUT,J., Brennstoff-Wiirme-Kraft 19, 309 (1967).
22. EL-SAYED,Y. M. and EVANS, R. B., ASME J. Eng.
Power 92, 27 (1970).
23. REISTAD, G. M., Availability: concepts and applica-
tions, Ph.D. Dissertation, University of Wisconsin,
Madison (1970).
24. SZARGUT,J. Brennstoff-W?irme-Kraft23, 516 (1971).
25. BEWR,J., Energieanwendung21,179 (1972).
26. BEWR,J., Energieanwendung23, 274 (1974).
27. SZARCUT,J. Energieanwendung23, 306 (1974).
28. GAGGIOLI,R. A. Proc. Int. Conf. on Energy Use Manage-
ment, II, Tucson, AZ, October 24-28 (1977).
29. FErmlrqo,T. H. and GAC_~IOLI,R. A., Trans. ASME, J.
Eng. Power99, 482 (1977).
30. KNOC}~, K. F. and FtmK, J. E., Int. J. Hydrogen
Energy 2, 377 and 387 (1977).
31. GAt~t~IOLI,R. A. and FEHRING,T. H., Combustion 49,
35 (1978).
32. GA6~mLI, R. A., RODRmU~Z,L. and WEPFER,W. J., A
thermodynamic--economic analysis of the synthane
 Energy systems
process, Final Report, U.S. D.O.E., Contract No. EY-77-
S-02-4589-1 (I 978).
33. BEYER, J., Energieanwendung 27, 204 (1978); 28, 66
(1979); 28, 86 (1979); 28, 137 (1979).
34. EISERMANN,W., Analysis of processes for production
of synthetic gaseous fuels, Final Report, NA TO Fellow-
ship Contract No. 430/402/566/8 (1979).
35. LOPER,J. L., ASMEpaper 79-WA/TS-4 (1979).
36. WEPFER, W. J., GAGGIOLI, R. A. and OBERT, E. F.,
Trans. ASME, J. Eng. Industry 101,427 (1979).
37. EVANS,R. B., Energy 5, 805 (1980).
38. GAGGIOL1, R. A. and WEPFER, W. J., Energy 5, 823
(1980).
39. GAGGIOLI,R. m. and WEPFER,W. J., Chem. Eng. Prog.
Tech. Manual 6, 140 (1980).
40. REISTAD,G. M. and GAGGIOLI,R. A., Thermodynam-
ics: second law analysis, A. CS. Symposium Series 122,
143 (1980).
41. SZARGUT,J., Energy 5, 709 (1980).
42. TRmUS, M. and EL-SAYED, Y. M., Zhermo-
economic analysis of an industrial process, Center for
Advanced Engineering Study, M.I.T., Cambridge, MA
(1980).
43. WEPFER,W. J., Thermodynamics: second law analysis,
A. C. S. Symposium Series 122, 161 (1980).
44. TRIBUS, M. and EL-SAYEO, Y. M., Proc. 2nd Worm
Congress of Chemical Engineering, Montreal, Canada,
II, 278 (1981).
45. BAEHR, Proc. 2nd World Congress of Chemical Engi-
neering, Montreal, Canada, II, 238 (1981).
46. LONDON, A. L., Int. J. Heat Mass Transfer 25, 743
(1982)..
47. BARRERE, M., Rev. Gen. Therm. Fr. 225, 243
(1983).
48. EL-SAYED,Y. M. and TRIaUS, M., Efficiency and cost-
ing, A.CS. Symposium Series 235, 215 (1983).
49. EVANS, R. B., KADABA, P. V. and HENDRrX, W. A.,
Efficiency and costing, A.CS. Symposium Series 235,
239 (1983).
50. GARCEAU, R. M. and WEI'~R, W. J., Efficiency and
costing, A.CS. Symposium Series 235, 263 (1983).
51. LONDON,A. L. and SHAH, R., Heat Transfer Eng. 4, 59
(1983).
52. MOORE, B. B. and WEP~R, W. J., Efficiency and cost-
ing, A.CS. Symposium Series 235, 289 (1983).
53. SPIEGLER,K, S., Desalination 44, 3 (1983).
54. HESSELMANN,K., Energy Economics and Management
in Industry, 123 (1984).
55. MAGAEVA,S. and KARAIVANOV,S., Chem. Ing.-Tech.
56, 58 (1984).
56. TNAI31T, S. S. and STARK, J., Energy Economics and
Management in Industry, 165 (1984).
57 HESSELMANN,K.,W/irmeaustauschernetzwerke-eineexer-
gofkonomische betrachtung, Ph.D. Dissertation, Tech-
nical University of Aachen, Germany (1985).
58. KNocr~, K. F. and HESSELMANN,K., Chem.-Ing.-Techn.
57, 602 (1985).
59. LING, W., ZHU, K. X., TOO, J. R. and GAGGIOLI,R. A.,
Analysis of energy systems~lesigns and operation,
ASME, AES-Vol. 1, 167 (1985).
60. TSATSARONIS,G., Thermorkonomische analyse von ener-
gieumwandlungsprozessen, Dr. Habilitatus Thesis,
Technical University of Aachen, Germany (1985).
61. TSATSARONIS,G. and WINHOLD, M., Energy 10, 69
(1985).
62. ZUBAIR,S. M. KADABA,P. V. and EVANS,R. B., A S M E
Paper 85- WA/HT-20 (1985).
63. GAGGIOLI,R. A., EL-SAYED,Y. M. EL-NASHAR,A. M.
and KAMALUDDIN,B., Computer-aided engineering of
energy systems, A SME, AES-Vol. 2-3, 77 (1986).
64. HESSELMANN,K., Computer-aided engineering of energy
systems, ASME, AES-Vol. 2-3, 23 (1986).
65. KNOCHE,K. F. and HESSELMANN,K., Computer-aided
255
engineering of energy systems, ASME, AES-Vol. 2-3,
35 (1986).
66. TAPIA,C. F. and MORAN,M. J., Computer-aided engi-
neering of energy systems, ASME, AES-VoL 2-1, 93
(1986).
67. TSATSARONIS,G., WINHOLD,M. and STOJANOFF,C. G.,
Thermoeconomic Analysis of a Gasification-
Combined-Cycle Power Plant, EPRI Final Report, AP-
4734, Palo Alto, CA (1986).
68. TSATSARONIS,G. and WINHOLD, M., World Congress
III of Chemical Engineering Proc., Tokyo, Japan, !, 402
(1986).
69. TSATSARONIS,G. and WINHOLD, M., Computer-aided
engineering of energy systems, ASME, AES-Vol. 2-3,
45 (1986).
70. VALERO, A., LOZANO, M. A. and MUNOZ, M.,
Computer-aided engineering of energy systems, A SME,
AES-Vol. 2-3, 1; 9; 17 (1986).
71. VALERO,A., LOZANO,M. A., ALCONCHEL,J. A., MUNOZ,
M. and TORRES, C., Computer-aided engineering of
energy systems, ASME, AES-Vol. 2-1, 43 (1986).
72. MORAN, M. J. and GAGGIOLI, R. A. (Eds), Analysis
and design of advanced energy systems: fundamentals,
ASME, AES- Vol. 3-1 (1987).
73. MORAN, M. J., STECCO, S. S. and REISTAD, G. M.
(Eds), Analysis and design of advanced energy systems:
applications, ASME, AES-Vol. 3-2 (1987).
74. WEPFER, W. J., TSATSARONIS,G. and BAJURA, R. A.
(Eds), Thermodynamic analysis of chemically reactive
systems, ASME, AES-Vol. 4 (1988).
75. WEPFER,W. J. LIOR, N. and BE/AN, A. (Eds), Second-
law analysis in heat/mass transfer and energy conver-
sion, ASME, AES- Vol. 6/HTD- Vol. 97 (1988).
76. WEPFER,W. J. and MORAN, M. J. (Eds), Approaches to
the design and optimization of thermal systems, ASME,
AES-Vol. 7(1988).
77. RUlXIAN,C. and MORAN, M. J. (Eds), Thermodynamic
Analysis and Improvement of Energy Systems, Inter-
national Academic Publishers, Beijing (1989).
78. BAJURA,R. m., MARSTON,C. n. and Tsatsaronis, G.
(Eds), Analysis and design of energy systems: computer
aided engineering, ASME, AES-Vol. 10-1 (1989).
79. BAJURA,R. A., VONSPAKOVSKY,M. R. and Geskin, E.
S. (Eds), Analysis and design of energy systems: thermo-
dynamic analysis of industrial processes, ASME, AES-
1Io1.10-3 (1989).
80. BOEI-IM,R. F. and EL-SAYED,Y. M. (Eds), Simulation
of thermal energy systems, ASME, HTD-Vol.
124/AES-Vol. 6 (1989).
81. STECCO,S. S. and MORAN, M. J. (Eds), A Future for
Energy, Pergamon, Oxford (1990).
82. TSATSARONIS,G., GAGGIOLI, R. A., EL-SAYED,Y. M.
and DROST, M. K. (Eds), Fundamentals of thermody-
namics and exergy analysis, ASME, AES-Vol. 19
(1990).
83. TSATSARONIS,G., MORAN, M. J. and BE/AN, A. (Eds),
Education in thermodynamics and energy systems,
ASME, AES-Vol. 20 (1990).
84. TSATSARONIS,G., BAJURA,R. m., KENNEY,W. F. and
REISTAD, G. M. (Eds), Computer-aided energy-systems
analysis, ASME, AES-Vol. 21 (1990).
85. KOUREMENOS,D. A., TSATSARONIS,G. and RAKOPOU-
LOS, C. D. (Eds), Analysis of thermal and energy sys-
tems, Proc. of the Intl. Conf., June 3~, 1991, Athens,
Greece, ASME, New York (1991).
86. REISTAD,G. M., GAGGIOLI,R. A., BE/AN,A. and TSATSA-
RONIS, G. (Eds), Thermodynamics and energy systems-
fundamentals, education and computer aided analysis,
ASME, AES-Vol. 24 (1991).
87. REISTAD, G. M., MORAN, M. J., WEPFER, W. J. and
L1OR,N. (Eds), Second-law analysis-industrial and envi-
ronmental applications, ASME, A ES- Vol. 25/ HTD- Vol.
191 (1991).
256 G. TSATSARONIS
88. VALERO, A. and TSATSARONIS,G. (Eds), ECOS '92,
Proc. of the Intl. Symp. on Efficiency, Costs, Optimiza-
tion and Simulation of Energy Systems, June 15-18,
1992, Zaragoza, Spain, ASME, New York (I 992).
89. STECCO,S. S. and MORAN, M. J. (Eds), Energyfor the
Transition Age, Nova Science, Commack (1992).
90. BOEnM, R. F. et al. (Eds), Thermodynamics and the
design, analysis and improvement of energy systems,
ASME, AES-VoL 27/HTD-Vol. 228 (1992).
91. SZARGUT,J. et aL (Eds), ENSEC '93, Proc. of the Intl.
Conf. on Energy Systems and Ecology, July 5-9, 1993,
Cracow, Poland (1993).
92. SHINSKEY,F. G., Energy Conservation Through Con-
trol Academic Press, New York (1978).
93. AHERN,J. E., The Exergy Method of Energy Systems
Analysis, Wiley, New York (1980).
94. CAMBEL,A. B. (Ed.), Energy 5 (1980).
95. GAGGIOLI,R. A. (Ed.), Thermodynamics: second-law
analysis, A CS Symposium Series 122 (1980).
96. BAEHR, H. D., Thermodynamik, Springer, Berlin
(1981).
97. BEJAN,A., Entropy Generation Through Heat and Fluid
Flow, Wiley, New York (1982).
98. GAGGIOLI,R. A. (Ed.), Efficiency and costing, ACS
Symposium Series 235 (1983).
99. KENNEY, W. F., Energy Conservation in the Process
Industries, Academic Press, Orlando (1984).
100. KOTAS, T. J., The Exergy Method of Thermal Plant
Analysis, Butterworths, London (1985).
101. KOTAS,T. J., Chem. Eng. Res. Des. 64, 212 (1986).
102. MORAN, M. J., BAJORA, R. A. and TSATSARONIS,G.
(Eds), Analysis and design of advanced energy sys-
tems: computer-aided analysis and design, ASME,
AES-Vol. 3-3 (1987)
103. BEJAN, A., Advanced Engineering Thermodynamics,
Wiley, New York (1988).
104. BOSNJAKOVIC,F. and KNOCHE,K. F., Technische Ther-
modynamik, Steinkopff, Darmstadt (1988).
105. MORAN, M. J. and SHAPIRO, H. N., Fundamentals of
Engineering Thermodyamics, Wiley, New York
(1988).
106. MORAN, M. J., Availability Analysis: A Guide to Effi-
cient Energy Use, ASME Press, New York (1989).
107. EL-SAVED,Y. M. and GAGGIOLI,R. A., J. Energy Res.
Technol. 111, 1 and 8 (1989).
108. TSATSARONIS,G., Second law analysis of thermal sys-
tems, ASME, 81 (1987).
109. BEVER, J., Energieanwendung, 358 (1970); 70 (1972);
179 (1972); 274 (1974); 204 (1978).
110. KNOCHE, K. F., Technische Thermodynamik, Vieweg,
Braunsehweig (1978).
111. TSATSARONIS,G., PiSA, J. J. and LIN, L., A Futurefor
Energy, pp. 771,783, Pergamon, Oxford (1990).
1i 2. BAEHR, H. n., Brenstoff-Wtirme-Kraft 31, 292
(1979).
l l 3. AHRENDTS,J., Die Exergie chemisch Reaktionsffdhiger
Systeme, VDI-Forschungsheft 579, VDI, Diisseldorf
(1977).
114. AHRENDTS,J., Energy 5, 667 (1980).
1! 5. BAEHR,H. D. and SCHMIDT,E. F., Brennstoff-Wiirme-
Kraft 15, 375 (1963).
116. BOSNJAKOVIC,F., Forsch. Ing.-Wes. 29, 151 (1963).
117. GAt3GIOLI, R. A. and l~rlr, P. J., Chemtech 7, 496
(1977).
118. RODRIGUEZ, L. S. J., Thermodynamics: second law
analysis, ACS Symposium Series 122, 39 (1980).
119. Snmn, J. H. and FAN, L. T., Energy Sources 6, 1
(1982).
120. TSATSARONIS,G. and TAWFIK, T. A., Application of
the exergy concept to the analysis of a gasifier, Interim
Report prepared for the U.S. Department of Energy,
Project DE-FC21-89MC26019(I 989).
121. FRANGOPOULOS,C. A., Analysis and design of energy
systems: thermodynamic analysis of industrial proc-
esses, ASME, AES-Vol. 10-3, 49 (1989).
122. KOTAS,T. J., Int. J. Heat Fluid Flow2, 147 (1980).
123. TAWEIK, T. A., Performance comparisons of inte-
grated gasification-combined cycle power plants,
Ph.D. Dissertation, Tennessee Technological Univer-
sity (1992).
124. PRICE, D. W., Exergoeconomic evaluation of Dow-
based IGCC power plants, MSc Thesis, Tennessee
Technological University (1992).
125. GRASSMANN,P., Kiiltechnik 4, 52 (1952).
126. EISERMAN, W., HASBERG, W. and TSATSARONIS,G.,
Brennstoff- Wtirme-Kraft 36, 45 (1984).
127. BARIN,I. and KNACKE,O., Thermochemical Properties
of Inorganic Substances, Springer, Berlin (1973).
128. Technical Assessment Guide (TAGrM), Electric Power
Research Institute, TR-100281, VoL 3, Palo Alto
(1991).
129. TSATSARONIS,G. and KRANE, R. J., Thermoeconomic
evaluation and optimization of energy-intensive pro-
cesses, AIChE Short Course Notes, New York (1993).
130. TSATSARONIS,G. and VALERO, A., Mech. Eng., 84
(1989).
131. VALERO,A. TORRES,C. and LOZANO, M. A., Simula-
tion of thermal energy systems, ASME, HTD-Vol.
124/AES-Vol. 6, 63 (1989).
132. TSATSARONIS,G., LIN, L. and PISA, J., J. Energy Res.
Technol. 115, 9 (1993).
133. TSATSARONIS,G., PISA, J. J. and GALLEGO, L. M.,
Thermodynamic Analysis and Improvement of Energy
Systems, 195 (1989).
134. SERRA,L., LOZANO,M., VALERO,A. and TORRES, C.,
Energy, submitted for publication (1993).
135. TSATSARONIS,G. and PISA, J., Energy, submitted for
publication (1993).
136. FRANGOPOULOS,C. A., Energy, submitted for publica-
tion (1993).
137. YONSPAKOVSKY,M. R., Energy, submitted for publica-
tion (1993).
138. TSATSARONIS,G., LIN, L., PISA, J. and TAWF1K, T.,
Thermoeconomic design optimization of a KRW-
based IGCC power plant, Final report submitted to
Southern Company Services and the U.S. Depart-
ment of Energy, DE-FC21-89MC26019 (November,
1991).
139. TSATSARONIS,G., KRAUSE, A., TAWFIK, T. and LIN,
L., Thermoeconomic evaluation of the design of a
pressurized fluidized-bed hydroretorting plant, Final
Report submitted to the Institute of Gas Technology
and the U.S. Department of Energy, DE-AC21-
87MC11089 (May, 1992).
140. EDGAR, T. F. and HIMMELBLAU,D. M., Optimization
of Chemical Processes, McGraw-Hill, New York
(1988).
141. REKLAITIS,G. V., RAVINDRAN,A. and RAGSDELL,K.
M., Engineering Optimization--Methods and Applica-
tions, Wiley-Interscience, New York (1983).
142. PIKE, R. W., Optimization for Engineering Systems,
van Nostrand Reinhold, New York (1986).
143. PAPALAMBROS,P. Y. and Wilde, D. J., Principles of
Optimal Design, Cambridge University Press, New
York (1988).
144. TSATSARONIS,G., PISA, J., LIN, L. and TAWFIK, T.,
Thermodynamics and the design analysis and improve-
ment of energy systems, ASME, AES-Vol. 27/HTD-
Vol. 228, 37; 55 (1992).
145. FRANGOPOULOS,C. A., Thermoeconomic functional
analysis, Ph.D. Dissertation, Georgia Institute of Tech-
nology (1983).
146. EVANS,R. B. and VON SPAKOVSKY,M. R., Second law
aspects of thermal design, ASME. HTD-VoL 33, 27
(1984).
147. EVANS,R. B. and FRANGOVOULOS,C. A., Second law
 Energy systems
aspects of thermal design, ASME, HTD-Vol. 33, 87
(1984).
148. yon SPAKOVSKV,M. R., A practical generalized analy-
sis approach to the optimal thermoeconomic design
and improvement of real-world thermal systems,
Ph.D. Dissertation, Georgia Institute of Technology
(1986).
149. BEJ^N, A., Second law analysis of thermal systems,
ASME, 1 (1987).
150. EL-SAYED, Y. M., J. Energy Res. Technol. ! 11, 111
(1989).
151. ACEWS-SABORIO, S., RANASINGHE, J. and REISTAD,
G. M., J. Heat Transfer I l l , 29 (1989).
152. RANASINGHE, J., ACEVES-SAaORIO, S. and REISTAD,
JPECS 19:3-F
257
G. M., J. Eng. Gas Turbines Power 111, 719 (1989).
153. SAMA,D. A., The use of the second law of thermo-
dynamics in the design of heat exchangers, heat ex-
changer networks, and processes, ENSEC '93, Proceed-
ings of the International Conference on Energy Systems
and Ecology, Cracow, Poland (July, 1993).
154. MORAN, M. J., personal communication (November,
1992).
155. LIN, L., Design optimization of advanced power
plants, Ph.D. Thesis, Tennessee Technological Univer-
sity (May, 1993).
156. LINNHOFF,B. et al., A User Guide on Process Integra-
tion for the Efficient Use of Energy, The Institution of
Chemical Engineers, London (1983).