3.19.2.

0
Thermoeconomic Analysis
Antonio Valero and Csar Torres
Center of Research for Energy Resources and Consumption, Centro Politcnico
Superior, Universidad de Zaragoza, Spain.
Keywords: Thermoeconomics, Exergy, Cost Accounting, Optimization of Energy
Systems, Energy Auditing, Diagnosis of Energy Systems.
Contents
1. Introduction .................................................................................................................. 1
2. An historical overview ................................................................................................. 4
3. What is the Exergy Cost? ............................................................................................. 6
4. Cost Accounting ........................................................................................................... 9
4.1. The process of cost formation ............................................................................. 14
4.2. The principle of non equivalence of the irreversibilities..................................... 16
5. Thermoeconomic Diagnosis ....................................................................................... 16
6. Thermoeconomic Optimization.................................................................................. 19
7. Final reflections and conclusions ............................................................................... 24
Glosary ........................................................................................................................... 27
Nomenclature.................................................................................................................. 29
Bibliography ................................................................................................................... 30
Summary
This paper introduces the basis of thermoeconomics analysis. It provides an in-depth
summary of the state-of-art and the progress that has been made in this field.
The concept of exergy costing and the cost formation process are presented. A brief
introduction of the applications of thermoeconomic analysis, that will be developed in
the articles of this chapter, is made. It includes cost accounting, design, optimization
and diagnosis of energy systems.
1. Introduction
The increasing demand of natural resources by current energy conversion technologies
and the concern for the impact on the environment due to emission, waste disposal and
signs of global warming has brought the creation of new disciplines that help to
understand how to improve the design and operation of energy systems and prevent
residues from damaging the environment.

Thermoeconomics is, in its widest possible sense, the science of natural resources
saving that connects physics and economics by means of the Second Law of
Thermodynamics.
A thermal power plant or a chemical plant are examples of energy systems formed from
a set of subsystems or processes. These systems interact with their environment,
consuming some external resources, which are then transformed into products. The final
purpose of this transformation is to increase the economic utility.
The production process of a complex energy system can be analysed in terms of its
economic profitability and efficiency with respect to resource consumption. An
economic analysis can calculate the cost of fuel, investment, operation and maintenance
for the total plant or even individual components but provide no means on how to
allocate costs among them and its products. On the other hand, thermodynamic analysis
let us calculate the efficiency of the individual process of the plant and locates and
quantifies the irreversibilities but it cannot evaluate their significance in terms of the
overall production process.
Thermoeconomic analysis combines economic and thermodynamic analysis by
applying the concept of cost, originally an economic property, to exergy (see Exergy,
Energy System Analysis and Optimization). Most analysts agree that exergy is an
adequate thermodynamic property to which allocate cost because it accounts for energy
quality. The exergy of a thermodynamic flow is the minimum amount of technical work
needed for its production, from the reference environment. Once the reference
environment is defined, exergy is the thermodynamic function of state which makes
possible to formulate the equivalence between different energy and/or matter flow
streams of a plant. Two flows are thermodynamically equivalent, that is, it is
theoretically possible to get one from the other without additional consumption of
energy resources if, and only if, they have the same exergy. Exergetic efficiency
compare a real process to an ideal process, i.e. reversible, of the same type. An exergy
analysis locates and quantifies irreversibilities in a process.
The physical magnitude connecting thermodynamics and economics is entropy
generation or more specifically irreversibility. It represents the useful energy lost or
destroyed, in all physical processes, and it has been used for pinpointing the true
inefficiencies of industrial processes. Since all common processes in an actual plant are
not reversible, exergy is destroyed and some natural resources are consumed and lost
forever, which involves a cost in economic terms. The more irreversible a process is, the
more natural resources are consumed.
The exergy balance accounts for the degradation of the exergy. The incoming exergy
will always be greater than the leaving one:
Exergy Input - Exergy Output = Irreversibilities > 0
This expression only keeps in mind the irreversibilities of the process. The purpose of
this process is set by means of the definition of its efficiency. This is to say, that there is
an implicit classification of the flows crossing the boundary of the system: the flows
that are the production objective, the resources required to carry out the production and
those that are residual. This information is not implicit in the second law and is the most
important conceptual leap separating and at the same time uniting physics with
economics.

The following equation:

Resources(F) Products(P) = Residues(R) + Irreversibilities(I) > 0
is of utmost importance because it places purpose in the heart of thermodynamics.
The concept of efficiency defined as
Efficiency = Product / Resource
is older than thermodynamics and measures the quality of a process. The desireto
produce a certain product is external to the system, and must be defined beforehand.
Once this has been done, the design of the system and its functional structure will fit the
aim of using available resources (capital, raw material, man power,). Every definition
of efficiency demands a comparison of the product obtained with the resources needed
to obtain it. Its inverse value is:
Unit Consumption = Resources / Product
This expression is also a definition of the unit average cost when resources refer to the
overall plant instead of individual processes. This concept is the key of
thermoeconomics. A logical chain of concepts can be established, see Figure 1, which
allows connecting physics with economics.

Purpose

Second Law
Efficiency

Exergy
Cost

Economic
Cost

Figure 1. Logical chain of thermoeconomic concepts

Thermoeconomics deals with man-made energy systems. Its efficiency is a purposive

concept and so they are the thermoeconomics analyses.
Thus, thermoeconomics assesses the cost of consumed resources, money and system
irreversibilities in terms of the overall production process. They help to point out how
resources may be used more effectively in order to save them. Money costs express the
economic effect of inefficiencies and are used to improve the cost effectiveness of
production processes. Assessing the cost of the flow streams and processes in a plant
helps to understand the process of cost formation, from the input resources to the final
products.
These analyses can solve problems related to complex energy systems that could not be
solved by using conventional energy analyses. Among other applications
thermoeconomics are used for:

Rational prices assessment of plant products based on physical criteria.

Optimisation of specific process unit variables to minimize the final product cost,
i.e. global and local optimisation.

Detection of inefficiencies and calculation of their economic effects in operating

plants, i.e. plant operation thermoeconomic diagnosis.

Evaluation of various design alternatives or operation decisions and profitability

maximization.

Energy audits.

These applications of thermoeconomic analysis are briefly analysed in this article, and
will be developed in detail in the articles of thisTopic.
2. An historical overview
For introduction to this section, the contributions of Gaggioli and El Sayed (see
bibliography) have been followed. The history of Thermoeconomics is relevant to
understand how the current state of art has been reached (see bibliography). Also, its
history helps us to know why it is very closely related with second law analysis and the
development of the concept of exergy. It is generally agreed that the basic concept of
exergy was first formed, independently, by Gibbs in U.SA, and Maxwell in England in
the last quarter of the 19th century.
The first proposal in the literature to use second law analysis for costing purposes was a
paper of Keenan in 1932. While he does not do exergy costing therein, he refers to it
explicitly as the means for appropriately apportioning cost associated with the
cogeneration of electric power and steam for distribution. Engineers thought that
obviously the fuel cost should be allocated to the steam and the power in proportion
to their energy content. The result, however, was the cogenerated electricity costed in
this manner, was far less expensive than electricity produced in conventional power
plants. Keenan pointed out that the value of the steam and the electricity lied in the
availability not in their energy. Although the proposal was based on Second Law
analysis, it is strangethat in practice the use of second law concepts was circumvented.
The use of the second law and specially entropy and exergy in thermal and mechanical
engineering has been resisted for a long time, because of the complicated means
whereby they have been developed and explained. On the other hand, the first law
analysis and the energy magnitude have been used because they are conceptually
comfortable. However, while it appeared obvious that the attribution of the fuel cost
to co-generated steam and electricity should be in proportion to their energies, to the
cost accountant this resulted in gross inconsistencies. A method known as Lost
kilowatts was adopted as rational for overcoming the inconsistencies of energy costing.
The basis of the method is that the co generated electricity cost is obtained as if it had
been produced in a condensing turbine and the remaining costs are assigned to the
steam.
The idea of coupling exergy and cost streams was first introduced by M. Benedict in
1948 in a seminar at M.I.T. He determined the total cost attributable to the
irreversibilities of an air separation plant and used this cost for optimal design.
Unfortunately, the contents of the seminar were not published until the 80s. The early
work by Keenan and Benedict was extended by his students, specially E.P.
Gyftopoulos, who worked on the concept of availability.
The interest to formulate the interaction between cost and efficiency was first
highlighted by Tribus and Evans at UCLA, in the early 60s. They were studying
desalination processes, and making exergy analysis, which led them to the idea of
exergy costing and its applications to engineering economics, for which they coined the
word Thermoeconomics. The essence of the Evans-Tribus procedure was to trace the
flow of money, fuel cost and operation and amortized capital cost through a plant,
associating the utility of each stream with its exergy. Y. El-Sayed, professor of
Mechanical Engineering in Egypt, corresponded with Evans and Tribus in connection
with their work in desalination and he came to work with them in the U.S.A. They
published in 1970 a key paper, multiple times referenced, called Thermoeconomics
4

and the Design of Heat Systems, where the mathematical foundation for thermal
system optimization is given.
Also in the sixties M. Obert and R. Gaggioli were working in the optimal design of
power plant steam piping, They proposed costing the steam exergy at a value to that of
power produced, penelizing irreversibilities for electricity which therefore, will not be
produced. R. Gaggioli directed, in the University of Wisconsin, the Ph.D. Theses of G.
Reistad (1970) and W. Wepfer (1979) on Second Law Costing methods, that include
the definition of rules to provide a rational distribution of the cost.
Meanwhile in Europe, since the fifties, a largenumber of works in second law analysis
has been developed, specially in East Europe. In 1952, Rant introduced the name of
exergy, just as we know it currently, defined as external useful work in opposition to the
energy (internal work). Other outstanding authors are: Beyer, Baehr, Brodiansky,
Szargut, Knoche among others. Some of the works, that also included thermoeconomic
analysis, were compiled by T.J. Kotas in 1985 in the book The Exergy Method of
Thermal Plant Analysis, that is one of the basic references in exergy analysis and
thermoeconomics.
The comprehensive effort to apply thermoeconomics to the analysis, optimization and
design of thermal systems did not start until the eighties. It startedin 1985, when R.
Gaggioli, who led the Systems Analysis Technical Committee of the Advanced Energy
Systems Division (AESD) of the American Society of Mechanical Engineers (ASME),
gathered and strived to broaden the participation of non U.S. scientists and research
groups that are working in advanced second law analysis and thermoeconomics. Under
this idea a series of AESD annual international meetings was created, that focused on
modern aspects of thermal sciences with particular emphasis on engineering
thermodynamics including exergy analysis and thermoeconomics. The first one took
place in Rome in 1997, and was chaired by Enrico Sciubba and Michael Moran.
It is at that time thatthermoeconomics actually took off. The interest and works
regarding to thermoeconomic analysis highly increased: G. Tsatsaronis (1985),
introduces the key concept of Fuel and Product. C. Frangopoulos (1983) and M. Von
Spakovsky (1986), whose Ph.D theses directed by R. Evans applied and formalized the
autonomous method of Evans and El-Sayed. In 1986 A. Valero and co-workers
published another key paper A General Theory of Energy Saving where the Theory of
Exergy Cost was introduced.
The International Mechanical Engineering Congress & Exposition (IMECE) formerly
Winter Annual Meeting of ASME and its non-US counterpart conferences on
Efficiency, Cost, Optimization and Simulation of Energy Systems (ECOS) are the key
references to follow the current state-of-art since its beginnings. ECOS and closely
related conferences have been held in Italy, China, Greece, Spain, Poland, Turkey,
Germany, France, Japan, Mexico and Scandinavian countries. In these countries, among
others, there are active research groups in advanced energy systems including exergy
analysis and thermoeconomics.
Thermoeconomic methods are generally subdivided in two categories, those based on
cost accounting, e.g. Exergy Cost Theory, Average Cost Approach, Last-In-First-Out
Approach (see bibliography) and those based on optimization techniques e.g.
Thermoeconomic Functional Analysis, Engineering Functional Analysis. Cost
accounting methods help to determine the actual cost of products and provide a rational
basis for pricing, while optimisation methods are used to find optimum design or
operation conditions.
5

In the 90s an important work starts in order to achieve a greater standardization and
formalism. Many articles published (see bibliography) compare, analyze and unify the
different methodologies (see Structural Theory of Thermoeconomics).
One of the most interesting initiatives is the project CGAM (1993), led by: C.
Frangopoulos, G. Tsatsaronis, A.Valero and M. Von Spakovsky (see bibliography)
whose objective was to show how the methodologies of each group of research could be
applied by solving a predefined and simple problem of optimization of gas turbine
cycle. In the final analysis, the aim was the unification of the different methodologies.
In the same direction, in the year 2001 another project called TADEUS (in honor to
Proffesor Tadeus Kotas) was initiated. Its aim is to apply procedures from different
research groups in thermoeconomic analysis to the diagnosis of the energy system
malfunction and inefficiencies. The objective of this new effort is to establish the
common concepts and nomenclature, compare the results and highlight the main
characteristics of each approach.
3. What is the Exergy Cost?
To introduce thermoeconomics let us start from the simplest physical concept of cost
asking ourselves: What is the exergy content of a beautifully designed bohemian glass?
Or that of a stone sculpture? Or that of gold? The answer is zero in practical terms.
Many things we value, thermodynamics does not. The source of value may be or may
not be related to its exergy content, even for the case of fuels. The only thing that
physics can do is to assess the physical cost of objects, i.e. the amount of energy units
required to produce a given product, namely embodied energy.
The concept of embodied energy comes from the 70s, when it was a great concern with
the first global energy crisis. The problem with energy is the lack of techniques to
allocate values of embodied energy when two products are produced simultaneously. A
more precise concept then came: the exergy cost proposed by Valero or the
cumulative exergy consumption proposed by Szargut (see bibliography) which are in
fact the same concepts as embodied energy but using exergy.
Cost could be defined as the amount of resources needed to obtain a functional product.
From one hand, resources take a general meaning. From the other hand, cost is
associated to the purpose of production. It is associated neither to price nor to the
resources that could be saved if the production process were less efficient or more
conventional one.
Cost is an emergent property. It cannot be measured as a physical magnitude of a flow
stream as temperature or pressure, it depends of the system structure and appears as an
outcome of the system analysis. Therefore, it needs precise rules for calculating it from
physical data. Cost is a property that cannot be found in the product itself.
The exergy cost of a mass or energy stream is the amount of exergy required to produce
it. For example in the case of the cogeneration plant depicted in Figure 3, the exergy
cost of the net power is the exergy provided by coal or natural gas to generate the
electrical power delivered to the network by the cogeneration plant.
The unit exergy cost of a mass or energy stream represents the amount of exergy
required to obtain a unit of exergy of the product stream. If Ei represents the exergy of
the i-th product stream and Ei* its exergy cost, the unit average exergy cost is written as:

ki* =

Ei*
Ei

(1)

The monetary or exergoeconomic cost, takes into account the economic cost of the
consumed fuel cF (i.e. its market price $/MJ) as well as the cost of the installation and
the operation of the plant, Z ($/s), and defines the amount of money to generate a mass
and/or energy flow. The economic cost balance could be written as:
C P = cF E F + Z
(2)
These costs measure the economic efficiency of a process. Similarly, the unit monetary
cost (also called exergoeconomic cost) of mass and/or energy flow is the amount of
monetary units per unit of exergy required to obtain the referred flow:
c E +Z
cP = F F
(3)
EP
One can further distinguish between average cost, which is a ratio and express the
average amount of resources per unit of product, and marginal cost which indicates the
additional resources E0 required to obtain one additional unit of a product stream Ei
under specified conditions, defined as:
E
ki* = 0
(4)
Ei cond
Figure 2 shows the differences between average and marginal cost for a generic costing
function. Other types of cost should be also considered: the opportunity cost refers to
the highest rate of return one could obtain per extra unit available for some resource.
This concept can be used for allocating cost to by-products. The abatement cost is
defined as the amount of resources required to eliminate an undesired output, waste or
residue.

Total Cost

CP (c/kWh)

Average
Cost

Marginal
Cost

0
0

P (kW)

Figure 2. Marginal and Average Cost

Then, what is the exergy cost of electricity? A first answer would be 3 units of exergy,
because we suppose that it is produced in a conventional thermal power plant burning
fuel-oil or coal, with an efficiency of 33%. If we consider electricity at the consumers
point, we must add the losses in transport. This adds a 10% to the unit cost giving now a
cost of 3.3 units. However coal or fuel-oil must be mined, treated and transported to the
plant. Additionally, each time we produce electricity we produce wastes that need to be
disposed. Disposal of wastes requires more exergy: Flue Gas Desulfuration, Special
NOx burners and CO2 capture and storage. Different studies indicate that electricity of
thermal origin at the consumers point can range between 4.5 and 5.5, which means that
we need 4.5-5.5 units of exergy of coal in mine to produce one unit of electrical exergy.
On the other hand, electricity is a raw fuel used in most industrial processes. We can
question ourselves what is the physical cost measured in exergy units of all
manufactured products as well as the exergy needed for their use, maintenance, repair
and disposal. This comprehensive analysis is named Exergy Life Cycle Assessment (see
bibliography) and provides average exergy costs because it is focused in obtaining
round numbers which could be used as ecological indices of sustainability.
However, the problem comes up when two or more products, by-products and residues
are produced simultaneously. How to allocate costs? We need rules mathematically
supported rather than considering concepts of uselfulness. We need to look inside the
system in order to understand the process of cost formation, by identifying the internal
relationships of all the structure components.
Indeed, the main problem of allocating costs has been to find a function that adequately
characterises every one of the internal flows in a system and distributes cost
proportionally. This function needs to be universal, sensitive and additive. That is, it
needs to have an objective value for every possible material manifestations, it needs to
be able to vary when these manifestations do so and each internal flow property needs
to be represented additively. There is a wide international consensus that the best
function, at least for energy systems, is exergy, which can contain in its own analytical
structure the flow history.
If we face the problem of calculating approximate average costs, we probably can stop
our analysis by disaggregating our system at not very detailed level. Also, and
commonly, cumulative exergy consumption analysis does not go into process details but
focuses on the overall exergy consumption.
Not only physical components can be disaggregated, exergy itself can be disaggregated
into its mechanical, thermal, chemical or physical components that have their own
history of formation. Each sub-process has consumed its resources to produce a
particular increase of pressure, temperature or composition. The flow history and its
physical cost can always be reconstructed i.e. the amount of given resources to produce
the flow. By a systematic account of these consumed resources, we can associate a
physical cost to each identified flow, representing the sum of the resources needed to
produce it under the given circumstances.
The problem with allocating costs to flow bifurcations can be solved classically. If the
bifurcation does not affect the quality of the bifurcated flow, the costs are proportional
to the quantity of each exiting flow. If the bifurcation affects quality, we must perform a
detailed analysis of the change of exergy components and their proportion to allocate a
cost to the bifurcated flow. Each exergy component is a reflection of a qualitative flow
property which, in turn, has passed on to the flow throughout the history of the
productive process with a specific resource consumption cost.

Therefore three conditions are needed to allocate costs. First the definition of the
boundaries of the system. Second, a structure of the system in which all the components
or processes are described in terms of black boxes interacting to each other through
energy flows ( or more generalized: energy, economic or information flows). And third
the definition of the purpose of production for each and every component.
In thermoeconomics the words: history, degradation, exergy, quality, cost, resource,
consumption, purpose and causality are related between them. In the cost formation
process, it is essential to analytically search for the locations and physical mechanisms
that make up a specific productive flow. The resources are used up to provide physicochemical qualities to the intermediate products until a finished product is obtained. The
main problem to be solved using exergy is how to measure and homogenize the
accounting of these qualities.
The product cost obtained in this way is an average value, that it is does not distinguish
between the first and the last units produced. Neither does it provide reasons to explain
why the sequence of sub-processes is as it is. There are no a priori reasons to add
qualities to a productive flow in a determined order and the analysis of the average costs
does not provide it per se. It only reflects the facts as they are produced and takes them
into account. It is only an additional knowledge and an engineering good practice of the
analyst that can justify why, for example, first we compress, then we heat and then we
change composition and successively so on.
At a given condition of the plant, that is, an instantaneous photograph of it, the
resources and the products are counted analytically and numerical indices are given to
each identified flow, which we call average costs. The problem now is to find a use for
these values. Evidently they are useful but cannot be overestimated since they do not
reflect the cost evolution of the next productive unit nor do they give us hints about why
the process is like it is nor if it can be improved. Nonetheless, they are useful for
benchmarking, that is to compare situations, either between two productive processes
that are not too different or two states of the same process.
4. Cost Accounting
For introducing this section, the contributions of Tsatasaronis, Kotas and Gaggioli were
mainly followed (see bibliography).
Cost accounting consists of procedures for determining, or better for estimating, the
total cost of production per unit of output for each product from a thermal system (e.g.,
for the electricity, steam, hot water, chilled water,). All of the capital and operating
costs, which are incurred to operate a thermal system, must be allocated to the final
products. Thus, for each product there are the direct costs, those which are clearly
attributable to the product such as resources and materials devoted specifically to it, and
there are all the other indirect costs. A main challenge to the cost accountant is to assign
each indirect cost in an equitable manner.
The purpose of cost accounting could be stated in broad terms as:

Determining the actual cost of products.

Providing a rational basis for pricing products and/or evaluating their profitability.

Providing means for controlling costs.

Forming a basis for operating decisions and their evaluation.

For a cogeneration plant, see Figure 3, it is clearly important to determine what it costs
to produce each unit of electricity and steam. These costs are significant, in as much as
the price of each product will be affected, though generally not determined, by the cost
of producing each. And clearly the profitability of individual products requires
knowledge of the cost of each one. The determination or estimation of product costs are
relevant not only for plants in operation but also for plants at the design stage.
There is not only interesting to find the costs of final products, it is also valuable to have
them for internal products. Then the build up of costs for each final product could be
traced through the energy system. If each cost of the internal streams of the system is
assessed, it could be used, by comparing them with standard or reference cost values, to
control and to avoid excessive resources consumption. This is the purpose of
thermoeconomic diagnosis.
The determination of all costs of the streams is also useful to make trade-off analysis of
the economics of the subsystem components. In an existing plant, such analysis can be
used, for example, for maintenance and retrofit decisions, as well as for developing and
implementing operation and control strategies. Likewise, it can be used for the
discovery of improvements on system concept and optimising the design of a particular
component and or the system as a whole.
The cost balance for the total system or for each component of the system could be
formulated as:
(5)
ci Ei + Z = c j E j
iIN

jOUT

where Ei represents the exergy of the inputs flows of the system/component and E j the
exergy of the output flows. ci represent the unit costs of the flows, that are known for
the inputs, and c j must be determined for the outputs.
In the case of a single product (output) of the plant, the unit cost of the product can be
determined by Eq. (5). In the case of a multi-product plant, the balance equation is not
sufficient. Additional criteria are required to determine the relationship between the unit
costs of the different products. This is where exergy can be used as a basis for cost
allocation of the products.
We will illustrate this with the example of a simple cogeneration plant depicted in
Figure 3.
First we will show that exergy and not energy is appropriate to take as the basis for
costing co-generated products. Energy costing rules consider that a unit of heat is
equivalent to a unit of work, then the equations to determine the cost of heat and work
could be written as:
CFa = CWa + CQa
(6)
a a
CW CQ = W Q
As aforementioned, Keenan was the first author that proposed to use exergy or
available energy to allocate costs. The exergy costing equations are:
b
b
b
CF = CW + CQ
(7)
b b
CW CQ = W EQ

10

If we allocate costs in proportion to their energy content the unit cost of electricity is
1.21 units of fuel, that is 2.5 times less expensive than electricity produced in a
conventional Rankine cycle. Furthermore this value is not sensible to the variation on
the quality of the steam produced or to any malfunction that would occur in the turbine.
According to this criteria it is the same to produce 1 kW of thermal energy than 1 kW
of mechanical energy.
T2 = 460 C
P2 = 50 bar

Generator

= 85%
Z1 = 10 c/s

Turbine
2

Boiler
1

= 97 %

is = 90%

Z1 = 0.3 c/s

Z2 = 0.6 c/s

T3 = 252 C

P3 = 10 bar
Pump

Figure 3. Physical Diagram of a simple cogenaration plant

Figure 4 shows a curve comparing the cost of steam using energy and exergy costing
criteria to allocate costs in the cogeneration plant, for different steam pressures. The
irrationality of energy costing is evident from the behaviour of the energy costing curve
at low turbine exhaust pressures. At a pressure of 1 bar, for example, the unit cost on a
energy basis for steam is close to 3 c/kg, even though such steam has very limited
usefulness. Unlike exergy costing, energy costing does not make distinctions about the
usefulness of energy transfer. We need cost accounting criteria that will be sensible to
the quality and degradation that occurs in the system.
Exergy cost accounting provides a wide and clear vision of the use and degradation of
energy and in consequence of natural resources. Exergy provides a thermodynamic
value of any energy stream with respect to reference conditions, and can be rigorously
obtained from the laws of thermodynamics, which allows precise measurements.
Exergy analysis provides two important messages: one is that it allows to quantify and
locate thermodynamic losses. The second is, it allows us to concentrate on the relevant
part of the energy, namely useful energy. Exergy may regarded as a measure of the
capacity of a given form of energy to produce work. It is also exergy which is lost, or
consumed, in order to make a particular process, and it is therefore reasonable to assess
the price of energy products on the basis of its exergy content.

11

Temperature (C)
183

4,5

252

298

333

362

386

408

427

444

4
3,5

Cost (kJ/kg)

3
2,5
2
1,5
Energy
1

Exergy

0,5
0

10

15

20

25

30

35

40

45

Pressure (bar)

Figure 4. Energy vs. Exergy Cost of Cogenerated Steam

The basic concepts underlying the explicit determination of auxiliary equations based
on exergy criteria were presented, for example (see bibliography) by Gaggioli and
Wepfer in 1980. These methods or rules are called the equality, the extraction and the
by-product methods. They required a judgment regarding the purpose of each unit of the
plant and distribute the costs proportionally to exergy contents of the flows.
We illustrate these methods using the turbine of the cogeneration plant. The balance
cost Eq. (5) for this component is:
c2 E2 + Z 2 = c3 E3 + c5 E5
Now there is a problem since, assuming the cost of the inlet steam c2 is known, the cost
balance equations have two unknowns, namely c3 and c5 . Thus an assumption must be
made regarding the allocation of the cost to the two outputs: shaft power E5 and
exhausted steam E3 .
In the equality method, the generation of the two products is considered to have the
same priority, so the cost of the high pressure steam and the capital cost of the turbine
are charged to the generation of both outputs (products) proportionally to their exergies,
leading to the relationship:
c E + Z2
c3 = c5 = 2 2
E3 + E5
In the extraction method, it is considered that the purpose of the turbine is to generate
shaft power and thus the whole cost of the turbine and the irreversibility occurring in it
are charged against it. This results in assuming the unit cost of exergy of steam entering
the turbine to be the same as that of that of the steam leaving it, thus:

12

c3 = c2
c ( E E2 ) + Z 2
c5 = 2 3
E5
In the by-product method the cost of one of the outputs is assumed to be known. It
could be assumed that the generation of process steam is essential even if no electric
power is generated. Therefore, the exergy of the process steam is costed as if it were
produced alone in a hypothetical low pressure boiler at the required pressure and
temperature. Therefore, the cost of electricity is determined by the cost balance. If the
cost of electricity calculated by this method proves to be competitive with the purchase
of electricity, then a decision in favor of co-generation should be made. An equivalent
procedure could be made in case of estimating the electricity cost.
The Exergy Cost Theory developed a general procedure (see The Process of Cost
Formation) for the determination of the auxiliary equations. One key step in the
application of this procedure is to make a rational judgment on how the efficiency
should be defined for each component of the plant. That is, for each unit to decide
which inlet streams should viewed as fuel streams and which as feedstocks. And for the
outlets which are products, and which are unspent fuel. To make these definitions is
equivalent to deciding what the productive purpose of each component is, and choosing
between the aforementioned extraction and equality methods. The other key step is to
decide which streams should be viewed as residue streams, either wastes or byproducts, for which the unit cost can be assigned rather than evaluated. This is
equivalent to the by-product method or making the assignment to zero. These
procedures are also patterned in part from the ideas of Fuel and Product developed by
Tsatsaronis and Winhold (see bibliography). Furthermore the method is formulated in a
mathematical manner which is significantly useful for computer implementation.
Table 1. Exergy and Exergoeconomic costs for the cogeneration plant flows
Nr

E(MW)

E* (MW)

k* (kW/kW)

C(c/s)

c(c/MJ)

100

100

45

0,45

33,96

100

2,94

55

1,62

23,36

68,79

2,94

37,84

1,62

9,97

31,21

3,13

17,76

1,78

9,67

31,21

3,23

18,06

1,87

The system of equations, which determine the cost of heat and work for the
cogeneration plant as well as for internal flows, are:
c1 = c fuel
c4 = cwater
c2 E2 c1 E1 c4 E4 = Z1
c3 E3 + c5 E5 c2 E2 = Z 2
c6 E6 c5 E5 = Z 3
c2 c3 = 0
Table 1, shows the exergy and exergoeconomic cost of all flows considered in the
simple cogeneration plant, in the case of the turbine the extraction rule is used. In this

13

case the cost of electricity is higher than the cost of steam, because in the first case we
include the irreversibilities of the turbine and the generator.
Figure 5 depicted a cost-irreversibility diagram obtained form the results on Table 1,
which shows how the cost is formed. The cost of steam produced on the boiler is
distributed in proportion to its exergy content. A 70% is allocated to generated steam,
the rest is allocated to cost of the electricity generated. The cost of the steam includes its
exergy content plus a 70% of the irreversibility generated in the boiler, meanwhile the
cost of electricity includes its exergy content plus a 30% of the irreversibilities on the
boiler plus the irreversibilities on turbine and generator. This is the reason why the unit
cost of electricity is 15,4% higher than the cost of steam. The 9,9% is due to
irreversibilities and 5,5% to investment cost of the turbo-generator.

CQ = 68,79 MW

100 MW
EQ = 23,36 MW

Exergy

Acumulated
Irreversibility

CW = 31,21 MW
W = 9,67 MW
2

Figure 5. Irreversibility-Cost Diagram of Cogeneration Plant

4.1. The process of cost formation

Cost accounting methodologies are based on cost allocation rules, which attribute to the
useful product of each system component the sacrifice of resources required to obtain it,
and distribute its costs proportionally to the exergy. They are mainly numerical
techniques that calculate values in an accurate way, but they lack a mathematical
structure. Therefore, it is not easy with this blind information to identify the process of
cost formation. In order to do that, general relationships are required that relate the
overall efficiency of the plant and the cost of the products with the efficiency and
irreversibilities of each individual component that form the system.
In order to analyze how the costs in a productive process are formed, we consider a
sequential process, as it is shown in Figure 6. Let us name the exergy of the resources
(fuel) of each component as F and its product as P. Correspondingly F* and P* will
name the exergy cost (MW) of the fuel and product streams.

14

Ir
FT

F1

P1

F2

P2

Fr

Pr

Fn

Pn

PT

Figure 6.- Sequential processes system diagram

The unit exergy costs of the product of a generic unit i can be expressed as follows:
k P* ,i = Pi* Pi = Fi * Pi = k F* ,i Fi Pi = k F* ,i ki
(8)
and the unit exergoeconomic cost ($/MJ) as:
cP ,i = CP ,i Pi = (CF ,i + Z i ) Pi = cF ,i ki + Z i Pi

(9)

The increment of the unit cost in a process could be expressed as:

k P* ,i k F* ,i = k F* ,i I i Pi

(10)

and

cP ,i cF ,i = ( cF ,i I i + Z i ) Pi

(11)

These equations reveal that the unit cost of the product is always greater than the unit
cost of the fuel. Any difference is calculated in terms of the irreversibility and capital
costs.
As in thermoeconomics all the magnitudes are in proportion to exergy, all the reasoning
can be made using exergy costs (kW) instead of the more difficult to obtain
exergoeconomic costs measured in monetary units ($/s). Then from now on, we develop
the concepts using the monetary stripped exergy cost.
In the case of the sequential system, where the product of a process is the fuel of the
next one, the exergy cost of the product of the process i, could be written as:
i

Pi * = Pi + I r

(12)

r =1

It shows that the exergy cost contains both the exergy of the product, and all the
irreversibilities generated in the processes required to obtain it. Note that the left hand
side of Eq. (12) has an economic meaning, meanwhile the right hand side of this
equation is purely physics. This connects thermodynamics with economics.
On the other hand, Eq. (13) shows, that the unit exergy cost of the product of a
component equals the product of the unit exergy consumption of the components taking
part in the production:
i

k P* ,i = kr

(13)

r =1

These expressions could be generalized for any plant, it does not matter how complex
is it. (see Symbolic Exergoeconomics Analysis of Energy Systems).

15

For example, if we consider a full example of the cogeneration plant, including

feedwater pump recirculation, its cost formulae are as follows:
(1 r )k1k2 k3
(1 r )k1
k P* ,W =
k P* ,Q =
1 rk2 k3 k4
1 rk2 k3 k4
where r is the ratio of mechanical exergy to the total exergy of the high pressure
steam. The denominator of these formulae represent the contribution of recirculation. If
this value is close to zero the system could be considered as a sequential process.
Once these formulae are obtained, it is possible to know easily the effect of the variation
of the efficiency of a component on the overall system. This question is analyzed in the
next section.

4.2. The principle of non equivalence of the irreversibilities

The exergy balance of the overall system is given by FT = PT + IT . If the quantity of
final product PT , is kept fixed, any variation in the design or in operation of the plant
will cause an increase or decrease in the consumption of the resources equal to the
variation of the total irreversibility, FT = IT .
We could suppose that this change is due only to the variation in the exergy
consumption of the unit i, given by ki , then, the variation of resources entering the
system could be written as:
FT = k1 ki 1ki ki +1 kn PT = k F* ,i ki Pi
(14)
Under the conditions stated, the unit exergy cost of fuel k F* ,i and the product Pi of the
unit i will remain constant, and the increase of the resources consumption will be
given by Fi = I i = Pi ki . A first conclusion from this analysis state that there is no
equivalence among the irreversibilities of the process of a system. The more advanced
the process is, the larger is the impact on resources consumption for the same increase
of the local irreversibilities. Therefore, the unit exergy cost is a measure of the impact of
local malfunctions in terms of overall resources consumption. A second outcome is that
when modifying the efficiency of a process, a variation of the production demand
occurs and the irreversibilities of all units that precede it will change. This effect is
called structural disfunction.
The application of these ideas to performance tests and energy audits in actual
installations will allow to identify and quantify the causes of the increase of the
consumption resources in a system. That is the basis of thermoeconomic diagnosis.

5. Thermoeconomic Diagnosis
The general objective of energy audits and diagnosis is to discover the degradations in
energy systems. There are two main techniques to energy system diagnosis of
deteriorations:

Thermo-mechanical monitoring conditions are usually adopted in power plants in

order to predict failures.

Thermodynamic monitoring methodologies are mainly suitable to analyze

anomalies causing reduction of system efficiency.

16

Thermoeconomic diagnosis has its foundations on second law analysis and its aim is the
detection of efficiency deviation, its economic impact, the identification and the
location of its main causes. Thermoeconomic diagnosis procedures are based on the
comparison between two working states: the actual operating state and a reference
operating state, corresponding to the system without any deterioration or operation
anomaly, and the fact shown in previous section, that irreversibilities have a different
fuel impact depending on the location of the irreversibility in the process chain.
The difference between thermoeconomic analysis and conventional second law analysis
for energy audit management stems from the following ideas:
Technical exergy saving: Not all exergy saving which are thermodynamically possible
can be achieved in practice. In fact, the real saving potential of each unit is limited by
technical and/or economic constraints. Thus, for a subsystem or installation, the
technical exergy saving can be expressed as I I I 0 , where I 0 is the design base
irreversibility and I is the irreversibility increase when plant operating under design
conditions supplies the same total product. Then it will be possible to express the
additional consumption of overall resources as follows:
FT = FT FT0 = I j
(15)
j

Malfunction and disfunction: The malfunction of a process is defined as the variation of

its irreversibilities due to a degradation of its efficiency. The disfunction of a process is
defined as the variation of its irreversibility due to the changes in its production
demand.
The exergy balance for a component i can be written as I i = Fi Pi = (ki 1) Pi . If the
operating conditions of the plant vary, the efficiency and the amount of product change.
Then, the technical exergy saving could be broken down into as:
I i = Pi 0 ki + (ki 1)Pi = MFi + DFij
(16)
where MFi represents the malfunction of the i-th process, and DFij the disfunction
generated by the j-th process on the i-th. (see Thermoeconomic Diagnosis of Energy
Systems)

Malfunction Cost is the effect of a local malfunction on the overall consumption of

resources feeding a plant, given by:
MFi* = k F* ,i ki Pi 0
(17)
and Fuel Impact is the effect of all local malfunctions on the overall consumption of
resources feeding a plant.
Equation (17) is valid for the case of a sequential process, while the general equation of
fuel impact is presented and discussed in Thermoeconomic Diagnosis of Energy
Systems. For the general case the technical exergy saving coincides with the sum of the
malfunction cost of all components:
FT = I j = MFj*
(18)
j

Figure 7 shows that a variation of 1% in the isentropic efficiency of the cogeneration

turbine, causes a fuel impact of 0.24 MW, and compares the values of the technical
exergy saving and malfunction costs for each component It explains that the
malfunction is a better parameter than the irreversibility variation, in order to locate the

17

causes of fuel impact. In case of technical saving analysis the causes of the fuel impact
are located in the boiler, meanwhile in the case of the malfunction cost analysis, they
are located in the turbine and in the additional steam produced to maintain the
electricity production.

Steam
Production

Generator
Irreversivility
Malfunction
Malfunction Cost
Turbine

Boiler

50

100

150

200

250

Fuel Impact (kW)

Figure 7. Fuel Impact and Technical Saving

The efficiency variation of a component may have different causes, either external to
the plant (variation of environmental conditions, plant production demand and fuel
quality) or internal, which are the presence of anomalies due to the component
degradation (also called intrinsic malfunctions), efficiency variation induced by
modification of the component operation conditions, and control system intervention.
The development of thermoeconomic diagnosis begun with the work of the research
group led by A. Valero (see bibliography). The mathematical formulation of the fuel
impact formula, was developed in several works: Lozano et al. (1994), Reini et al.
(1995) and Torres et al. (1999). The major effort was concentrated in developing
procedures to localize anomalies and quantify effects both in the component where the
malfunction is originated and those indirectly affected by malfunctions, and the
definitions and calculations procedures for concepts such as: intrinsic malfunction,
induced malfunction, disfunction and malfunction cost (see Thermoeconomic Diagnosis
of Energy Systems)
The TADEUS problem, an acronym of Thermoeconomic Approach to the Diagnosis in
Energy Utility Systems, is a project aimed at integrating various experiences
accumulated by several groups of researchers working in thermoeconomic diagnosis. As
a result a set of articles were published in 2004, showing different approaches, each one
having particular characteristics that are, nonetheless, complementary to each other.

18

6. Thermoeconomic Optimization
For intruducing this section the contributions of Frangopoulos, El-Sayed and Von
Spakovsky were mainly used (see bibliography).
Thermal system design required to answer questions such as: What processes or
equipment items should be selected and how should they be arranged? What is the
preferred size of a component? What is the best temperature, pressure, flow rate, and
chemical composition of each stream in the system? To answer these questions,
engineers need to formulate an appropriate optimization problem.
The first step in the definition of optimization problem is to define clearly the
boundaries of the system to be optimized. All the subsystems that significantly affect
the performance of the system under study should be included in the optimization
problem. The selection of criteria on the basis of which the system design will be
evaluated and optimized is the key element in formulating an optimization problem.
Optimization criteria may be economic (total capital investment, total annual levelized
costs, annual levelized net profit,) technological (thermodynamic efficiency,
production time, production rate, fuel consumption,) and environmental (rates of
emitted pollutants). An optimized design is characterized by a minimum or maximum
value, as appropriate for each selected criterion.
Another essential element in formulating the optimization problem is the selection of
the design variables that adequately characterize the possible design options. In
selecting these variables, it is necessary to include all the important variables that affect
the efficiency and the cost effectiveness of the system. Each component and the system
as a whole is defined by a set of quantities. Some of them are fixed by external
conditions (e.g. environmental pressure and temperature, fuel price) and are called
parameters. The remaining are variables, i.e. their value may change during the
optimization procedure. Those variables, the values of which do not depend on other
variables or parameters, are called independent or design variables. The rest can be
determined by the solution of a set of equality constraints and they are called dependent
variables.
The mathematical model for an optimization problem consists of:

An objective function to be minimized.

A set of equality constraints.

A set of inequality constraints.

Thermoeconomic optimization methods use as a primary criterion an optimization

performance measure: minimize the total levelized cost of the system products that
includes the cost of external fuel resources, capital investment and maintenance cost.
Also multicriteria optimization and environmental constraints can be considered.
The objective function expresses the optimization criterion as a function of the design
variables:
Min C0 ( x) = ce FT ( x) + Z i ( x)
(19)
i

where ce represents the unit cost/price of the external resources, FT the exergy of the
external resources, and Z i the capital and maintenance cost of each component.

19

In order to define the equipment cost functions Z i , we will assume the investment costs
to increase with increasing capacity and decreasing exergy consumption (or increasing
efficiency), and can be represented, by the following relation:
i

Z i = Z i ,0 + i i ,0 Pi i
k k
i i ,0

(20)

where ki is the unit consumption of the component, Pi the product (size) of the
component and Z i ,0 is the part of the cost independent of the product. Both production
and efficiency are functions of the design variables.
Another approach of costing functions that handles itself in optimization is:
Z i = ca ,i Ai

(21)

where ca is the unit manufacture cost, and A represents a characterizing dimension of

the component, and it is expressed in terms of thermodynamic variables, following a
typical Cobb-Douglas production function:

Ai = i x j j
(22)
j

where i is a characteristic parameter of the characteristics of the device. The number

of thermodynamic parameters depends on the process; these are essentially sizing
parameters (mass rate, heat rate or power) and efficiency parameters (adiabatic or
isentropic efficiencies, pressure losses, heat transfer temperature differences,). An
example of cost function (see Cost Functions of components for optimal system design)
for a heat exchanger will be:
A = QTm1Pt 0.15 Ps0.15
The exchanger surface A is expressed in terms of the heat load, the logarithmic mean
temperature difference and the pressures losses on the shell side and the tube side.

Cost ($)

Total
Cost

Fuel Cost

Investement
Cost
0
1

Unit Consumption k

Figure 8.- Costing Functions

20

The balance between thermodynamic measures and capital expenditures is an economic

feature, which applies to the thermal system as a whole and to each component
individually. The costs of resources usually vary to the opposite direction of the cost of
equipment with respect to the design variables. An improvement on the structure or the
efficiency of the equipment implies a reduction of the resources consumption but an
increase of the capital investment. Figure 8 shows a graph of a generic function for the
cost of production.
The equality constraints are provided by appropriate thermodynamic and cost models as
well as appropriate boundary conditions. These models must include the flow rate and
energy balances for each component, relations associated with the engineering design,
such as the local efficiencies of the components. The model adopted by
thermoeconomic optimization (see Structural Theory of Thermoeconomics) relates the
input (fuels) of each components with its outputs and design variables:
Ei = gi ( E j , x)
(23)
The model can also contain inequality constraints that specify the allowable operating
ranges, the maximum and minimum performance requirements, and bounds on the
availability of resources. When optimum is reached with only equality constraints, we
obtain the shadow costs, one for each independent variable. If an inequality constraint is
active in the optimum, the cost becomes an opportunity cost for the constrained variable
(see bibliography).
There are cost optimization procedures which make no use of the exergy concept. So
cost-effectiveness of every change carried out on a plant component must be assessed in
terms of the overall system parameters, e.g. its effect on the consumption of fuel
resources. This makes optimization very complex and computer time consuming. With
thermoeconomic optimization these difficulties may be overcome. For example, with
proper themoeconomic analysis and under certain conditions, the decomposition is
applicable, which facilitates the solution of the problem, because it allows the
optimization problem of the whole system to be decomposed into a set of optimization
problems of subsystems or components, which are of smaller dimension (i.e. they have
fewer independent variables) and can be solved more easily.There was basically, at the
beginning, two different thermoeconomic approaches: the structural method that use the
local unit cost of the irreversibilities (see Design Optimization of Power and
Cogeneration Systems), and the autonomous method introduced by R. Evans and Y. ElSayed in 1970 that is the starting point of other state-of-the-art techniques. (see
Optimization Methods for Energy Systems and Functional Analysis).
We will illustrate this last method using a sequential process (fig. 6). Also, the
optimization will be restricted to one independently variable - xi - per component. With
the output of the system PT constant, the objective function is the total cost, per unit of
time and is given by:
n

C0 = c1 E1 + Z i (Ei +1 , xi )

(24)

i =1

where c1 ($/MJ) is the unit exergoeconomic cost of the fuel input, E1 the exergy input
rate, and Z i the cost of capital investment per unit of time of each component of the
system.

21

Assume that the optimum configuration of each component, and hence of the whole
system, can be obtained by choosing the values of - xi -, that satisfy the condition of
minimum for the objective function (24), given by:
C0
= 0 i = 1, , n
(25)
xi
With Ei as independent variables, n+1 constraining equations relating these variables
are required:
Ei = gi ( Ei +1 , xi ) i = 1, , n
(26)

En +1 = s = const

As a result a Lagrangian function L becomes the objective function instead of (24):

n

i =1

i =1

L(E, x, ) = c1 E1 + Z i (Ei +1 , xi ) + i ( gi ( Ei +1 , xi ) Ei ) + n +1 ( s En +1 )
If we introduce the function:
i ( Ei +1 , xi , i ) = Z i ( Ei +1 , xi ) + i gi ( Ei +1 , xi )

(27)

(28)

the Langrangian function (27) could be rewritten as:

n

L(E, x, ) = ( c1 1 ) E1 + ( i ( Ei +1 , xi , i ) i +1 Ei +1 ) + n +1 s

(29)

i =1

Minimum C0 occurs when the partial derivatives of L with respect of x and E variables
vanish. From the condition L Ei = 0 , we get:

1 = c1

Z i
gi

i +1 = E + i E
i +1
i +1

(30)

These equations determine the values of the Lagrange multipliers i or shadow costs,
that take the role of local marginal costs of exergy inputs of each component.
From the condition L xi = 0 , the minimum is obtained as:
i Z i
g
=
+ i i = 0
xi xi
xi

(31)

From the previous equations, it is shown that the system may be decomposed into
autonomous subsystems, whose local functions are defined in Eq. (28) and they are
optimized for the local variables xi .
A special case appears when capital cost and the exergy input increase linearly with the
exergy output of the components: Z i = f Z ,i ( xi ) Ei +1 and gi = ki ( xi ) Ei +1 , then Eq. (30), is
rewritten as:

1 = c1

i +1 = f Z ,i ( xi ) + i ki ( xi )

22

(32)

where ki represents the unit exergetic consumption, and therefore i represents the unit
average exergoeconomic cost. The general methodology is fully explained in the articles
Functional Analysis and Structural Theory of Thermoeconomics.
There are several approaches to Thermoeconomic optimization, that were presented in a
set of articles in 1993, as a result of the project CGAM.
The Exergoeconomic Optimization Approach, proposed by Tsatsaronis uses an iterative
design improvement procedure that does not aim at calculating the global optimum of a
predetermined objective function, as the conventional optimization methods do, but
tries to find a good solution for the overall system design. The basic idea, lies in a
commonly accepted concept from the cost view point: at constant capacity for a well
designed component, group of components, or total system, a higher investment cost
should correspond to a more efficient component and vice versa.
The Functional Analysis proposed for C. Frangopoulos and the Engineering Functional
Analysis proposed by M. von Spakovsky used the method of the Lagrange multipliers
and decomposition procedures.
A. Valero and coworkers, present a similar approach, but propose to use the unit
average exergy costs instead of the Lagrange multipliers.
Y. El- Sayed proposed also, in order to avoid problems with the isolation of the decision
variables (see Design Optimization of Power and Cogeneration Systems) to divide the
decision variables {x} into local variables {xL} and global variables {xG}, in general the
number of global variables is much smaller than the local variables, iterate to find the
local optimum of each component respect its local variables and the global optimum
respect to the global variables.
The decomposition strategy is based on the Principle of Thermoeconomic Isolation (TI)
introduced by R. Evans in 1980: A component of a thermal system is thermoeconomically isolated form the rest of the system if its production Pi and the unit cost
of the resources i are known quantities and independent from the rest of the
component variables.
It is an ideal condition which cannot be fully achieved for most of real systems, but the
more the TI conditions are fulfilled the fewer iterations are required to achieve the
optimal solution for the whole system. Therefore the thermoeconomic model of the
system is subdivided or decomposed into subgroups. Each subgroup is optimized in
turn, according to a sequential process, iterating around the system until the systems
internal economy converges, within prescribed tolerances, to a single set of values.
Decomposition may only approach the global optimum since the degree of
thermoeconomic isolation of the independent variables, the choice of the subgroups and
their functions, and the nature of the dependent variables greatly affects how close the
approach will be. Nonetheless, the advantages of this strategy facilitates the optimal
design of individual units in highly interdependent complex systems, and let the
designers to concentrate their efforts on designing the variables of single components,
while resting assured that these efforts improve the overall system.
These ideas could be used to design a strategy for local and global optimization
problems, that is described in Figure 9.

23

Define productive structure

Define Thermodynamic Model
Define Cost Functions
Select Decision Variables
Divide decision variables into local and global variables
{XL,1 , ...,XL,N,XG}
Chose initial values for all design variables
Solve termodynamic model

Compute the unit cost of products

For each component of the system

Find the optimun of the local function i
respect the local set of variables {XL,i}
Find the optimun of global cost function C0
respect the set of global variables {XG}

Until convergence criteria is satisfied

Figure 9.- Thermoeconomic Optimization Strategy

7. Final reflections and conclusions

Everyday life is immersed in economics. Our society tends to substitute human and
natural values by economic ones, i.e. prices. But capital is not the only resource. To say
the contrary is an arrogant way of reducing nature and society to nothing. We put prices
almost to everything even though we dont know its value. So, what is the alternative?
We can approach to know the value of things and living entities, if we try to reproduce
or replace them. When doing this, we realize how difficult it is. How many resources
of all kind of types are needed to get them? How much knowledge is necessary to
understand the mechanisms by which resources can be converted into entities? These
questions give a sense of ethics of conservation because we should not destroy what we
dont know how to construct it.
This is why it is so important to know costs measured as a general sacrifice of
resources, independently of the numeraire we use to quantify them. However, as it is
well known, the value of cost has its drawbacks. It depends on the limits of the system,
it also depends on the value exergy in our case and on the purpose (efficiency) of
each and every subsystem that interrelates the productive structure of the system.
Relativism is a main feature of resources accounting.
Besides of that, we have also another problem: are the costs of things stable with
production? In other words, when we say that the cost of electricity is three, does this
value change when we demand more electricity? Everybody knows that in small
production intervals the average cost of the product remains almost constant. So, what

24

does it mean small production intervals? This depends on the purpose of our study. In a
macro level, average costs can be used without much error. On the contrary if we are
interested in optimizing a power plant, we realize that the cost of electricity is very
sensitive to the efficiency of the plant components.
The purpose of the study advises the accounting methodology i.e. the accuracy of
results as well as the relaxation of the rules to assess costs. Thermoeconomics is
another viewpoint of matter analysis in Physics. When we try to understand the heat and
mass transfer mechanisms in a heat exchanger, we use the techniques provided by the
heat and mass transfer sciences. These sciences use and interpret pressure and
temperature as overall values of properties that spatially and temporally change. The
understanding of a system interconnection and efficiency is better attained using
thermodynamics rather than the sciences of heat and mass transfer. Thermodynamics
helps to understand how energy is used and degraded in the different parts of the
system. But thermodynamics requires an input-output abstraction of the system. Some
information can be obtained from its analyses, some other cannot. There is no
contradiction between the different types of analysis. They focus on different aspects of
reality. And the micro level science heat and mass transfer- is capable to explain the
magnitudes used at the next level science, for instance the interpretation of pressure,
temperature, energy and so on by the transfer sciences. There is not a disruption in
knowledge but continuity.
When we focus on exergy accounting, we are at the next (aggregated) level of
knowledge, now connected to thermodynamics. Thermoeconomics uses
thermodynamics but differs in the type of problems it tries to solve as well as the use of
new concepts like cost borrowed from economics.
Exergy could be considered as the bridge between thermodynamics and cost accounting
methodologies. This is, because exergy connects with intensive properties like pressure,
temperature, energy and so on and on the other side it can be rigorously defined and its
cumulative consumption calculated. When we investigate all the properties of exergy
costs we are giving a conceptual bridge between classical Physics and Economics. On
the other hand, and as explained before, we are giving a basis for a general resources
accounting methodology that can use any other numeraire or even relax the rules of
accounting themselves. That could be the case of emergy whose main interest is to
know how many units of solar energy are needed to produce any product.
Second law costing is deeply related to causality and not so related, as it might be
expected, to internal price allocation. In our point of view, when we are to assess the
impact that a malfunction has on the consumption of resources, you only need to
provide information about the structure of the plant and the efficiency of the
components. Exergy cost accounting is not so interested in arguments about how many
rewards you get by selling the products or by saving money because of joint production
instead of a separate one. Cost accounting propositions like "heat as a by-product" or
"work as a by-product" and so on, are needed when we are analyzing the price
formation process or better said, the price propagation process. But, when we are
interested in the cost formation process throughout the structure and we understand cost
as a sacrifice of resources, there is no doubt what are we finding.
There is a claim of the lack of words and for the dangerous use of the word "cost". In
fact, cost has different meanings for different people and practitioners. Cost is most of
the times related to money and not so much to physical resources, but in its broader
meaning cost is measured in resources, in time, in lives, etc. Besides of that, "cost" is in

25

many cases mistaken as "price" synonym. We mostly agree on its conceptual difference
but we forget it when we apply rules for cost accounting that include external
considerations about the finality of the production, or the market's utility of the products
we obtain.
So, what is the purpose of cost accounting? The answer must be deeply related to the
physical behavior of the system. Cost should be related to physical measurements like
mass flow rates, pressures, temperatures and compositions, then, related to actual
irreversibilities occurring in the system and then to causes. This is the only way that we
can provide physical roots to the accounting techniques. From this point we may use or
not the values of exergy costs as a basis for price setting, both external and internal
prices. Exergy cost accounting provides a set of indicators about how much impact on
resources has a perturbation on a given structure. Modern management techniques that
use internal prices to buy and sell intermediate products within the company can base its
price-settings on exergy costs but should not be taken directly instead of prices. Internal
economy techniques are one thing and exergy cost accounting another.
These are the ideas lying behind the exergy cost analysis. In the subsequent articles of
this chapter, we will give a physical basis to the rules for calculating costs, we will give
coherence to the connection of exergy cost with thermodynamics and we will
understand under what conditions cost changes.
The connection between the subsystem level and the overall system takes now a crucial
importance. Therefore we now concentrate on the relationship between efficiency
change and impact on resources.
There exists a relationship between additional local irreversibility and additional
consumption of resources (see Thermoeconomic Diagnosis of Energy Systems). The Eq.
(17) is quite important because it justifies the practical reasons for internal cost
accounting and thermoeconomic optimization based on decomposition strategies. In
other words it answers the question of how many additional expenses you must pay
because a malfunction appeared in a component. No more no less. The key word is
propagation. And the idea of costing is to quantitatively assess the propagation effects
on each and every subsystem in the plant.
From the point of view of thermoeconomic optimization, the exergy cost becomes the
shadow exergy cost, or Lagrangean exergy multiplier, when the thermodynamic
behavior of the plant is optimized. This connection is explained in the Structural Theory
of Thermoeconomics.
Knowledge of the shadow costs helps designers react to possible changes to constraints.
These constraints in the case of thermoeconomic analysis describe the exergy inputoutput structural relationships. The more advanced in the system is a process, the more
impact on fuel consumption its local irreversibilities have (principle of non equivalence
of the irreversibilities).
Therefore a crucial step in the optimization process, either in the design stage or in the
operation stage, is the sensitivity analysis, and shadow costs are used to weight changes
in the constraints. Relaxing the constraints leads to improvements in the optimum, or on
the contrary, degradations of the optimun are assesed as the value of the shadow costs
obtained.
A fascinating result is that the cost formation process, i.e. the formula that represents the
relationship of any cost with the efficiencies of the subprocess and other related
structural parameters have the same mathematical basis as that of the shadow costs.
26

Therefore the shadow costs obtained from the optimization problem provide the same
results as those of the exergy cost obtained from the cost accounting methodologies
(under given specifications as explained in the Structural Theory of Thermoeconomics).
This allows for the unification of thermoeconomic theories and for understanding why
under what conditions they do not match. This is possibly the best theoretical
contribution of thermoeconomic optimization.
Also, when we relate the exergy costs with the impact on resources feeding the plant,
i.e. the Fuel Impact formula, we are bridging the thermoeconomic optimization and
diagnosis. In other words if the loss of resources is thermodynamically and universally
measured in terms of irreversibility, the relationship between the local and global
irreversibility can be explained in terms of exergy cost and structural parameters (like
efficiencies, bifurcations and production parameters).
Summarizing, these are the reasons, why the concept of exergy cost is of paramount
importance on thermoeconomics:

It relates cost, an economic concept, with irreversibility, a physical concept. But

it is still measured in physical units and provides a rational basis for economic
allocation rules.

The formulae describing the cost formation process for a given system structure
coincide with these obtained for the optimization problem. Thus conceptually
connecting the optimization and the cost accounting methodologies. Also giving
a way to understand why are the reasons for discrepancies. This is because it is
easier to explain differences in formulae than in numbers.

If the relationship between local irreversibility and the overall impact on

resources is algebraically assessed through the exergy cost, we can understand
and also have a solid and meaningful basis for thermoeconomic diagnosis.

Even now, when we dispose of powerful computers and numerical methods for
optimization and simulation of real world process, that could overcome
thermoeconomics. Thermoeconomics becomes a crystal clear and unique way to
connect the physical universal measure of loss, i.e. irreversibility with the loss of
resources at the overall system level and then to economics. This is perhaps the stronger
reason for a continuing systematic research in this new science.

Glosary
Abatement Cost: Amount of resources required to eliminate an undesired output or
residue.
Average (unit) cost: Average amount of resources consumed to obtain (a unit of) the
product.
Capital cost: Monetary value of a process unit.
Characteristic equation: Equation describing exergy input(s) as a function of the
exergy output(s) of a process unit.
Cogeneration: It is the thermodynamically sequential production of two or more useful
forms of energy from a single primary energy source.
Constraints: Functional equalities or inequalities that have to be satisfied by the
optimum solution in an optimization problem.
Costing Functions: Functions which express the cost of a process unit in terms of
thermodynamic and physical variables.

27

Decomposition: Separation of an optimization problem (system) into subproblems

(subsystems) in order to facilitate the solution.
Disfunction: Increment of the irreversibility of a process when production demand
varies.
Embodied Energy: The amount of energy units required to produce a given product.
Energy system: A system that transforms various energy forms to other energy forms,
of which at least one is useful.
Exergoeconomic cost: Amount of resources apportioned in proportion to exergy-,
expressed in terms of monetary units, to obtain a product. In a broad sense it is
synonymous of thermoeconomic cost.
Exergy (unit) cost: Amount of resources, expressed in terms of exergy, to obtain (a
unit of) the product.
Exergy: The maximum theoretical useful work that can be obtained if a thermodynamic
system is brought into thermodynamic equilibrium with the environment, while the
system interacts with the environment only.
Functional analysis: The formal, documented determination of the functions of the
system as a whole and of each unit individually.
Impact on fuel or fuel impact: Additional fuel consumption because of an inefficiency
in a piece of equipment in a system.
Independent variables: Those variables appearing in the objective function that do not
depend on other variables or parameters; their optimum values, that is the values that
minimize (or maximize) the objective function, are determined by the optimization
procedure.
Induced Malfunction: Increment of the irreversibility of a process due to malfunction
of other process of the system.
Intrinsic Malfunction: Increment of the irreversibility due to the degradation of the
process itself.
Irreversibility: Is a synonymous of exergy destruction, which represents the energy
quality degradation occurring in whatever physical process.
Irreversibility: That part of exergy that is destroyed in a process.
Malfunction Cost: Increment of external resources due to the malfunction of a
component
Malfunction: It represents the increase of irreversibility in a device suffering an
inefficiency or efficiency degradation
Marginal cost: Additional resources consumed for producing one more unit of the
product under specified conditions.
Objective function: A mathematical function, the minimization (or maximization) of
which is the objective of optimization.
Opportunity Cost: The rate of degradation of the optimum per unit use of a nonoptimal
variable in the design
Optimization: The process of finding the conditions, i.e. the values of variables, that
give the minimum (or maximum) of the objective function.
Productive structure: Scheme of the function of the different process units and their
interactions in a system. It represents how the resources consumed by the system are
distributed among the different pieces of equipment and converted into the final plant
product.

28

Thermoeconomic Diagnosis: Part of the Thermoeconomic analysis which studies the

causes of the malfunctions of the thermal processes and its impact in the consumption
of natural resources.
Thermoeconomic isolation principle: This determines whether a component behavior
is independent or not with respect to the rest of the plant units. It can be defined as
follows. A unit of a thermal system is thermoeconomically isolated from the rest of the
system if the output product of the unit and the prices associated with the product and
the inputs coming from other units of the system are constant and known quantities.
Thermoeconomic model: Mathematical representation of the productive structure of a
system.
Thermoeconomic Optimization: Part of the Thermoeconomic analysis which studies
the optimum cost for design of thermal system.
Nomenclature
Scalars
Z
Capital cost rate of a component ($/s)
k*
C
c
E
F
I
m
n
P
Q
W
x

Unit average exergy cost of

Exergoeconomic cost ($/s)
Unit exergoeconomic cost ($/MJ)
Exergy of a flow (kW)
Fuel exergy of a component (kW)
Irreversibility and exergy destruction of a component (kW)
Number of flows
Number of components
Product exergy of a component (kW)
Heat
Work
Operational variable

Greek letters

Increment

Marginal exergy consumption

Exergy efficiency

Lagrange multiplier; marginal cost

Local Cost Function

Subscripts
0
Related to environment
e
Inlet
F, P
Related to fuel or product
i, j
Indexes for component, flow number
s
Outlet
T
Total system
Q, W Related to Head and Work
Superscripts
*
Exergy cost
29

Reference values

Bibliography
Bejan A., Tsatsaronis G., and Moran M.(1996). Thermal Design and Optimization,
Wiley, New York. [This book provides an outlook of thermoeconomics and its scope,
which is complementary to the approach presented in this paper. It also presents a good
review of thermoeconomic analysis methods.]
Cornelissen R.L. (1997) Thermodynamics and sustainable development. The use of
exergy analysis and the reduction of irreversibility. PhD Thesis. University of Twente,
the Nedtherlands. [Exergy Life Cycle analysis is exposed.]
El- Sayed Y. (2003). The Thermoeconomics of Energy Conversions. Elsevier Ltd.
Oxford U.K. [The outcome of more than 20 years of thermoeconomic development, this
book is an essential reference resource for practicing engineers and researchers.]
El-Sayed Y. M., and Evans R. B. (1970). Thermoeconomics and the design of heat
systems. Journal of Engineering for Power 92(1), 27 35. [One of the first works in
optimization of thermal systems by thermodynamic and economic considerations
combined.]
Erlach B., Serra L., and Valero A. (1999). Structural theory as standard for
thermoeconomics. Energy Conversion and Management 40, 16271649. [In this paper
it is proved how the mathematical formalism of the structural theory can be applied to
cost accounting methods, such as AVCO and LIFO, which are different from exergy
cost theory.]
Evans R. B. (1980). Thermoeconomic isolation and essergy analysis, Energy 5(89),
805822. [This paper introduces the concept of thermoeconomic isolation]
Frangopoulos C. A. (1987). Thermoeconomic Functional Analysis and Optimization.
Energy, 12(7), pp. 563571. Pergamon Press Ltd. [This paper presents the fundamentals
and a brief descriptions of TFA.]
Frangopoulos C. (1994) Application to the exergoeconomic functional approach to the
CGAM problem. Energy, 19(3), pp. 323342. [The TFA approach is applied to the
CGAM problem.]
Gaggioli R. A., and Wepfer W. J. (1980). Exergy economics. Energy: The International
Journal 5, 823 837. [Cost accounting and benefit cost analysis of energy systems and
conservation techniques based on exergy. Exergy cost accounting assessment rules are
introduced.]
Gaggioli R. A., and El-Sayed Y. M. (1987). A Critical Review of Second Law Costing
Methods. Proceedings of the IV International Symposium on Second Law Analysis of
Thermal Systems (ASME Book I00236), 59-73. ASME, New York. [Even though it has
been published several years ago, this paper presents an excellent review with a detailed
description of the scope, advantages a drawbacks of Second Law based methods.]
Kotas T. J. (1995). The Exergy Method of Thermal Plant Analysis, Malabar, Fla.:
Krieger Publishing Company. [Reprinted with corrections and additions to the first
edition of 1985. A presentation of the theory combined with numerical examples helps
in understanding exergy and the way it can be applied. It includes a chapter of
thermoeconomic applications of exergy, that after twenty years of its publication it is
still a reference text.]

30

Lazzaretto A., and Tsatsaronis G. (1997). On the quest for objective equations in exergy
costing. In Proceedings of the ASME Advanced Energy Systems Division, volume 37,
pages 197209. [This paper presents the fundamentals of AVCO, LIFO and SPECO
exergy cost accounting methods.]
Lazzaretto A., and Tsatsaronis, G. (2001). Comparison Between SPECO-Based and
Functional Exergoeconomic Approaches, Proceedings of the 2001 ASME International
Mechanical Engineering Congress and Exposition, November 11-16, 2001, New York,
N.Y., IMECE2001/AES 23650, pp. 1-16 [In this article a comparison between cost
accounting methods such as SPECO, AVCO or TCE and Functional Thermoeconomic
Approaches such as TFA is made, for better understanding and integration of
thermoeconomic methodologies.]
Lozano M. A., and Valero A. (1993). Theory of the Exergetic Cost. Energy, 18(9), pp.
939, 960. Pergamon Press Ltd. [A mature description of ECT with several examples of
its applications.]
Neufville R. (1990). Applied System Analysis. Engineering Planning and Technology
Management. McGraw-Hill. New York. [In this book the concepts of shadow, marginal,
average and opportunity costs are fully explained.]
Reini M., Lazzaretto A., and Macor A. (1995). Average strutural and marginal costs as a
result of a unified formulation of the thermoeconomic problem. In Proceedings of
Second Law Analysis of Energy Systems: Towards the 21st Century, Rome. [In this
paper the impact fuel formula is introduced.]
Serra L., Valero A., Torres C. and Uche J. (2003). Thermoeconomic Analysis:
Fundamentals. In Integrated Power and Desalination Plants. Edited by Husain A.
EOLSS Publisher. Chapter 8, pp. 427-459. [An introduction to thermoeconomic
analysis.]
Szargut J., Morris D. R., and Steward F. R. (1988). Exergy Analysis of Thermal,
Chemical and Metallurgical Processes, 332 pp. New York: Hemisphere
Publishing/Berlin: Springer-Verlag. [The concept of exergy and its application to
thermal systems, chemical and metallurgical processes are presented. The concept of
cumulative exergy is introduced.]
Tsatsaronis G. and Winhold M. (1985). Exergoeconomic analysis and evaluation of
energy conversion plants. Part I: A new general methodology; Part II: Analysis of a
coal-fired steam power plant. Energy 10(1), 69-94. [A method of combined exergetic
and economic analysis is proposed, based on the fuel-product approach.]
Tsatsaronis G. and Pisa J., (1994). Exergoeconomics evaluation and optimization of
energy systems application to the CGAM problem. Energy, 19(3) pp. 287-321. [The
exergoecomic optimization approach is applied to the CGAM problem.]
Valero A., Lozano M. A., and Muoz M. (1986). A General Theory of Exergy Saving I,
II and III. ASME Book n 40341C, WAM-1986, AES 2-3, pp. 1-21. [In this paper the
Exergy Cost Theory and Application are introduced.]
Valero A., Tsatsaronis G, Frangopoulos C. A., and Von Spakovsky M. R. (1994). The
CGAM Problem. Energy, 19(3), pp. 281381. Elsevier Science Ltd. [This issue of
Energy is a monograph in which the CGAM problem, is solved applying different
thermoeconomic methodologies. Each paper of the referred issue presents the solution
of each thermoeconomic method.]

31

Valero A., Lozano M. A., Serra L., and Torres C. (1994). Application of the Exergetic
Cost Theory to the CGAM problem. Energy, 19(3), pp. 365381. Elsevier Science Ltd.
[This paper shows the application of the Exergy Cost Theory applying the Symbolic
Thermoeconomics formulation to the optimization of energy systems.]
Valero A., Correas L., Lazareto A., Rangel V., Reini M., Taccani R., Toffolo A., Verda
V., and Zaleta. A. (2004). Thermoeconomic philosophy applied to operating analysis
and diagnosis of energy utilty systems. Int. J. Thermodynamics, 7(2) pp.33106. [A set
of articles where TADEUS problem and different approaches for thermoeconomic
diagnosis are presented.]
Von Spakovsky M. R. (1994). Application of engineering functional analysis to the
analysis and optimization of the CGAM problem. Energy, 19(3), pp. 343364. Elsevier
Science Ltd. [This paper shows the application of the engineering functional analysis to
the optimization of energy systems.]

32