PROCEEDINGS OF ECOS 2023 - THE 36TH INTERNATIONAL CONFERENCE ON

EFFICIENCY, COST, OPTIMIZATION, SIMULATION AND ENVIRONMENTAL IMPACT OF ENERGY SYSTEMS

25-30 JUNE, 2023, LAS PALMAS DE GRAN CANARIA, SPAIN

Free and Open-Source Teaching: Understanding Exergy

using Thermal Engineering Systems in Python (TESPy)

Mathias Hofmanna , Francesco Witteb , Malte Fritzc,d , Jonas Freißmannc,d ,

Ilja Tuschyc,d , George Tsatsaronise
a
Technische Universität Berlin, Berlin, Germany, [email protected], CA
b
German Aerospace Center (DLR), Institute of Networked Energy Systems, Oldenburg, Germany
c
Hochschule Flensburg, University of Applied Sciences, Flensburg, Germany
d
Center for Sustainable Energy Systems (ZNES), Europa-Universität & Hochschule Flensburg, Flensburg,
Germany
e
Technische Universität Berlin, Berlin, Germany

Abstract:
In energy or chemical engineering, a representation of highly complex processes is often only possible through
a simulation. Conducting experiments or building demonstrators is far too costly and time-consuming. Apply-
ing exergy-based methods is beneficial to reveal thermodynamic inefficiencies and to propose appropriate
optimization approaches for energy conversion processes. Therefore, teaching theory and software-based
application of such methods are essential in engineering study programs.
This paper presents a didactic concept for open educational resources on exergy analysis covering the in-
troduction to the exergy method, its distinction from the energy analysis, and its application to various case
studies. The course aims to combine thermodynamic understanding with numerical mathematics and object-
oriented programming. Students learn to build and run models of energy conversion systems and conduct
respective exergy analyses using the software Thermal Engineering Systems in Python (TESPy). The course
material is developed using Jupyter notebooks, which offer a flexible connection between theory and code.
Results are reproducible and can be tested and developed further by the open-source community.
The advantages of exergy-based methods can be illustrated by evaluating and visualizing real thermodynamic
losses using Grassmann or waterfall diagrams. Process understanding can be deepened further by parameter
analysis showing their impact on components and the overall process. Finally, the acquired knowledge is
transferred to more complex problems with multiple components or more than one product.

Keywords:
Thermodynamics, Exergy, Exergy-based methods, Simulation, Python, Online class

1. Introduction and Motivation

In the last years, particularly since the start of the COVID-19 pandemic, the need for using electronically-based
teaching materials has increased. Nevertheless, the didactic concepts for teaching events have also changed
fundamentally in a concise time [1, 2]. Specific methods, e.g., flipped/inverted classroom, were previously
only used by an interested group of lecturers and played a subordinate role. The pandemic has significantly
accelerated the development and led to the increased use of numerous previously underrepresented didactic
methods. As a result, the didactic quality of teaching courses was improved almost imperceptibly, especially in
the field of engineering.
But what principles should be applied to the design of a lecture or seminar at a university today? Von Thun [3]
gives some advice on this. He considers the development of one’s own design principles as fundamental, which
correspond to the requirements and possibilities of your subject, your contents, and your person. The lecture
or seminar should not only be given but also created. Developing those design principles is an exciting and
valuable part of the professional biography of a university teacher.
When designing a course, it is necessary to determine which competency goals are to be achieved. According
to Anderson and Krathwohl [4] there are six categories of cognitive process dimensions to choose from:
first remember, second understand, third apply, fourth analyze, fifth evaluate, sixth create. Many courses at
universities, especially introductory classes, usually are within the third category. However, the study’s goal
should be to reach the 6th level, i.e., to plan a research paper on a given topic, i.e., to work scientifically on
one’s own. Concerning engineering sciences, it is, therefore, essential to get into independent practice-relevant
action as early as possible in the curriculum, e.g., to be able to program independently.

195 https://doi.org/10.52202/069564-0019
 In engineering programs, especially in the field of energy and chemical engineering, thermodynamics is a
fundamental discipline. In many of these courses, the exergy concept [5–9] is taught to aspiring engineers as a
helpful and powerful tool. In this context, the conclusive teaching of the concept is one of the major challenges
for the lecturers. There needs to be more than a conventional thermodynamics lecture to make the students
apply the exergy concept and explore exergy. Following the previous remarks, we need at least one more
advanced course for teaching the exergy concept. Independent activities such as programming should be
included. The students must be enabled to apply general principles of thermodynamics independently, evaluate
them, and finally, create energy conversion processes independently according to the given requirements and
check them with the help of exergy-based methods.
In today’s highly interconnected and globalized world, it should be widely accepted that research and teaching
are not the exclusive preserve of established societies or segregated groups with limited social access. Indeed,
even in the beginnings of the universities in the Middle Ages, admission restrictions were unusual [10, 11].
Ultimately, the composition of the student body was shaped by social realities, such as the exclusion of women.
Of course, numerous social barriers have been dismantled in the meantime. That it is nevertheless not self-
evident is shown by the 4th United Nations Sustainable Development Goal1 : “Ensure inclusive and equitable
quality education and promote lifelong learning opportunities for all.” Addressing lifelong learning shows a
need to make knowledge and knowledge transfer accessible to all without barriers, even outside appropriate
institutions. For example, this can be made possible through a digital offering independent of end devices. Free
and open-source access should be a standard feature.
Furthermore, in research projects funded by third parties, work is often expected today to be transdisciplinary.
Consequently, social shareholders are also involved, and data and results should be available to everyone for
subsequent use. For such cooperative collaboration across spatial or temporal distances, several technical
solutions (cloud, git, web) exist today that enable participation and curation in various forms simultaneously.
Remember, understand, apply, analyze, evaluate, and create no longer necessarily have to take place in the
lecture hall or seminar room on the chalkboard.
An online course was developed based on the findings formulated up to this point. The open educational
resources on exergy analysis covering the introduction to the exergy method, its distinction from the energy
analysis, and its application to various case studies. The interactive digital learning experience offered to
students in the course described here is a building block in the development of professional competencies for
future engineers.
The present contribution is structured as follows: A didactic concept for the developed online course is presented
in the next section. Information about the online implementation based on Python and Jupyter Notebooks
is collected in Section 3. Some illustrative results from the course material are shown in the results section,
followed by the conclusions.
All developed materials and used libraries are available on a free and open-source basis. If researchers,
practitioners, teachers, or students use the content, the CC BY 4.0 guidelines2 should be respected. The
prepared content can be accessed online in a textbook format.3 Furthermore, everyone is invited to join the
developer team, share ideas and suggestions, or even bug reports via the GitHub repository.4

2. Didactic Concept
2.1. Target Group, Prerequisites and Integration into Curriculum
The course is intended for students in undergraduate engineering programs, especially power engineering,
process engineering, and chemical engineering. In addition, the course can be a possible advanced training,
since it is available in an open-source form. Nevertheless, the content is not primarily aimed at that, meaning
that examples could seem too simple for a practitioner.
Students should already completed coursework in basic mathematics, thermodynamics, and object-oriented
programming knowledge. Furthermore, they should be familiar with the purpose of using apparatus and
machines as well as processes of energy engineering. Finally, they are expected to be interested in going
deeper into the exergy concept.
Consequently, the course fits into the curriculum of the study program according to Figure 1. In deviation from
the illustration, it is conceivable that the course is offered in the third or fourth semester. In any case, the stated
prerequisites should be taken into account. The advantage of an early start of the course is the practice-oriented
relevance of the contents. Thus, a hands-on mentality is developed, and the students are taught the relevant
engineering skills at an early stage. They learn to solve problems in teams and practice modeling, programming,
1 https://sdgs.un.org/goals/goal4
2 https://creativecommons.org/licenses/by/4.0/
3 https://fwitte.github.io/TESPy teaching exergy/
4 https://github.com/fwitte/TESPy teaching exergy

https://doi.org/10.52202/069564-0019 196
and scientific computing. Also, they get familiar with from scratch process design and find alternative solutions.
Finally, the students further develop existing structures.

5th to 6th ... Understanding exergy ...

semester using TESPy

3rd to 4th ... Heat Turbo- Energy engineering (including ...

semester transfer machines applications of exergy analysis)

1st to 2nd ... Linear Thermodynamics (including Computer science Numerical ...
semester Algebra fundamentals of exergy analysis) and programming mathematics

Figure 1: Integration into the curriculum of a typical engineering undergraduate program

2.2. Learning Outcomes

After finishing the course, the students are familiar with the basics of steady-state process simulation and know
different methods for calculating the thermodynamic properties of substances. Also, they can build, initialize and
solve process flowsheets independently, using the software Thermal Engineering Systems in Python (TESPy)
as a process simulation tool for modeling, calculating, and analyzing complex energy engineering processes
and interpreting the results obtained. The participants identify and solve problems that occur during process
simulation and can independently use aspects of the exergy concept for their future research questions. They
are familiar with physical and chemical exergy and the splitting of physical exergy. They can evaluate processes
and components with the help of exergy analysis and exergy-based parameters and present the analyses
graphically. Finally, using a version control system, the students apply group work methods such as team
communication, milestone processing, and cooperative programming.
In their best efforts, students complete the module by adding new elements to the existing course, and thus
reach the highest level of competence according to Anderson and Krathwohl [4], see Section 1.
2.3. Content and Competencies
Following the course content of fundamentals of thermodynamics and energy conversion, the focus is on
simulating single state changes, simple cycles, and complex energy conversion systems. At the same time, the
participants will practice methods of exergetic process evaluation.
The mathematical representation of individual components of energy engineering processes, e.g., heat ex-
changers, pumps, turbines, waste heat boilers, steam generators, condensers, is subsequently combined into
the representation of overall processes, e.g., gas turbines, gas and steam turbine power plants, steam power
processes, and integration of renewable energy sources. The methods of setting up process simulations, solving
mass and energy balances, and calculating thermodynamic property data are covered. The simulation program
used is the free and open-source software TESPy, see Section 3.
The work can be done individually or in small groups. We recommend working in teams of 2 to 3 people to
intensify social skills and test suitable collaboration methods. Figure 2 shows the distribution of the competence
fields addressed during the course. Depending on the individual characteristics of the course, shares may shift
slightly.

Knowledge and understanding

20% Analysis and methodology
20%
Development and design
25% Application and practice
10%
25% Social competence

Figure 2: Pie chart showing the competency fields addressed in the progress of the course.

197 https://doi.org/10.52202/069564-0019
 2.4. Schedule
The Gannt chart in Figure 3 shows a sample schedule for designing the class for one semester (2 contact hours
per week). In its first version, the course includes four sections and three milestones. The kick-off meeting
can include a get-to-know-you and orientation session if the students work in teams. Afterward, there is the
possibility to configure the own hardware and run tests regarding the functionality of the libraries to be used.

Semester, 15 Weeks
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Kick-Off Meeting
Presentation of First Results
Final Presentation
1. Getting Started
Setup and Run Test
Repetition on Exergy Method
2. Components
Adiabatic Pipe Flow
Heat Exchanger
Compressed Air Production
3. Simple Cycles
Clausius Rankine
Heat Pump
Gas Turbine
4. Processes
Electrolysis Heat Recovery
ORC Combined Heat and Power

Figure 3: Gannt chart showing the possible progress of the course in one term.

In the second section, students are encouraged to investigate thermodynamic fundamentals for simple ap-
plications and program them independently. All analyses are carried out at the component level and usually
involve exactly one thermodynamically describable change of state. With the help of simple exergetic analyses
based on the solution of the mass and energy balances and simple graphical evaluations, differences between
energetic and exergetic descriptions can be made easily understandable.
In the second section, the students will partially write down the correlations in an elementary way and later
moving forward to a object-oriented programming. The advantages of a Python library (e.g., TESPy) become
clear with the transition from the second to the third section.
In the third section of the course, the exergy analysis is to be applied to simple multi-component systems. At the
latest in this section, the students should be aware of the advantages of the exergy-based evaluation compared
to a purely energetic consideration. A further focus is then on the graphical evaluation of the results. How can
the findings be visualized in the best possible way? The students can try out various approaches independently
here.
Since the interaction of several components is now being investigated, the focus is on how the components’
properties impact the overall process. For instance, which components significantly influence the exergy
destruction of the overall process and why. It is necessary to determine which thermodynamic parameters are
the most important for improving the overall process.
The last section examines complex processes with more than one product. If a process provides different forms
of energy as products, for example electricity and district heating, an energy-based assessment of the quality of
the overall process is misleading. Exergy is the only way to ensure comparability of the different material and
energy flows. The students are to conduct these evaluations for processes such as heat recovery in electrolysis
or processes with combined heat and power.
2.5. Further Reading, Online Tutorials, Hardware, Software
During their work the students can use several textbooks or online tutorials for fundamental or further reading.
We recommend the following:

• Thermodynamic fundamentals [12, 13]

• Exergy concept [6–9]
• Python [14] and Jupyter Notebooks [15]
• TESPy [16–18]

https://doi.org/10.52202/069564-0019 198
For scientific computing in Python environments, it is a convenient option to use ready-made distributions
that integrate relevant packages and library management in a user-friendly way. A simple start succeeds with
miniforge5 , available for all current operating systems. To edit and run the code students can install a code editor
such as Pycharm6 (free of charge for university members) or Visual Studio Code7 or run a local server allowing
them to work in a browser. The online documentation of the course includes all relevant installation instructions.

3. Implementation
In recent years, there has been a concern [19] in the energy system analysis community, among others, with
the public provision of data and publicly funded research’s traceability. Various initiatives are engaging in this
area, developing free and open-source work paradigms and taking them into other domains. Aspects initially
discussed only in research are spilled into teaching. For example, the Journal of Open Source Education [20]
specifically addresses lecturers to publish online educational resources connected with free and open-source
software. In a proposal, Morrison [21] formulates the idea of combining publicly available data and voluntary
contributions from experts into a “student-centered” energy system model for education. The actual energy
system model will be programmed in Julia, and the workflow will be set up in Jupyter. It is explicitly not a
web-based scenario generator or gamification exercise. The students will perform exercises in a curated way
and develop the code of the energy system model piece by piece.
Inspired by these ideas, the present course for energy technology and exergy analysis was developed. In
particular, the idea for the structural content and the technical project architecture is based on Morrison’s
proposal. The entire content of the course is embedded in a Python-based structure. The prepared content
can be accessed online in a textbook format8 and the source files are available through the respective GitHub
repository.9 Figure 4 shows the implementation.

Jupyter
ZeroMQ Python
Server
Kernel
s

(local)
et
ck
So
eb
W
&
TP

Python Libraries
HT

User Browser TESPy

CoolProp
NumPy
.ipynb files matplotlib
GitHub Repository
(local clone)

Figure 4: Implementation of the resources into a Python-based architecture including a powerful and user-
friendly front-end

The files10 are provided as a Jupyter notebook. It is an editable format where not only Python code is stored and
can be executed, but visualization, markdown, and collaboration capabilities exist. Consequently, this format is
particularly suitable for course formats like the one presented here. Outlines can be created, and text can be
formatted, images and videos can be added, mathematical formulas and referencing can be used.
To create their own code, students can download the Jupyter files to their machine and locally run a Jupyter
server, which allows to access a Python kernel and run their code in the browser or in code editors.
The open-source software TESPy [16] is used for the component-based simulation of the considered energy
conversion processes. The software is developed as part of the open energy modeling framework (oemof) in
the scientific field and by industrial users.11 The user can build up arbitrary thermodynamic processes from
single predefined components like heat exchangers, compressors, valves, or turbines. In addition to heat
pumps, geothermal or solar thermal power plants, heat networks, or chillers can also be mapped. The software
5 https://github.com/conda-forge/miniforge
6 https://www.jetbrains.com/community/education/#students
7 https://code.visualstudio.com/
8 https://fwitte.github.io/TESPy teaching exergy
9 https://github.com/fwitte/TESPy teaching exergy
10 file extension .ipynb
11 https://oemof.org

199 https://doi.org/10.52202/069564-0019
 accesses the likewise open-source substance properties database CoolProp [22]. Detailed documentation with
more examples and tutorials is available for the software.12
Following process simulation and design, exergy analysis can be used to identify sources of thermodynamic
losses. The exergy analysis is performed automatically. Only the exergetic fuel, product, and loss of the overall
process have to be passed to the tool [17, 18]. Furthermore, the simulation models can be combined with
optimization methods [23].

4. Featured Course Material

The course is based on three relevant pillars whose approach enables change management and innovation
processes in their respective disciplines or application areas. The three pillars are

• object-oriented programming with Python,

• steady-state power plant simulation with open-source software (TESPy, CoolProp), and
• thermodynamic evaluation using exergy-based methods.

As a consequence, the contents of the course lead to the conclusion that these three pillars can be interlinked
with each other exceptionally well for prospective engineers to teach the exergy concept. We want to summarize
this by presenting selected tasks from the course and illustrating the results comprehensibly.
Within all listed thermodynamic models below we assume steady-state processes, all components are adiabatic,
there are no pressure losses in the heat exchangers, and changes in kinetic and potential energies can be
neglected. Ambient conditions are set to T0 = 25 ◦ C and p0 = 1.013 bar. For the calculation of chemical exergies
the model of Ahrendts [24] is used. The corresponding flow diagrams and parameters of the processes can be
found in Figures 5, 8, 11, and 12 as well as Tables 1, 2, and 3. Due to space limitations, not all source codes
are given below. All tasks and solutions can be obtained from the git repository.
4.1. Highlight 1: Adiabatic Pipe Flow
Consider a well insulated pipeline transporting a fluid, i.e., water steam or air. The state change from 1 to 2
because of friction can be described as an adiabatic throttling valve as shown in 5.

1 2

T0

Figure 5: Flowchart of adiabatic pipe flow

Table 1: Adiabatic pipe flow parameters

Parameter Symbol Unit Value

◦
Inlet temperature T1 C 195
Inlet pressure p1 bar 10
Constant (or varying) outlet pressure p2 bar 6 (or p1 ... p0 )
Mass flow rate m1 kg/s 1

First, setting up the parameters of the pipe flow and calculating the inlet enthalpy, if necessary install CoolProp.13
from CoolProp . CoolProp import PropsSI as PSI
f l u i d = "water"
p i n = 10 * 1e5
T i n = 195 + 273.15
h i n = PSI ( "H" , "P" , p i n , "T" , T i n , f l u i d )
p o u t = 6 * 1e5

Considering the above assumptions, and the energy balance, we determine the outlet temperature for isenthalpic
throttling using CoolProp.
h out = h in
PSI ( "T" , "P" , p out , "H" , h out , f l u i d ) − 273.15
12 https://tespy.rtfd.io
13 As an alternative a simple, full-featured, lightweight CoolProp wrapper for Python is available via https://pypi.org/project/pyfluids/.

https://doi.org/10.52202/069564-0019 200
Subsequently, the following exercises, among others, can be performed. Implementing a function that splits
the physical exergy in its thermal and mechanical parts [25]. Creating a plot showing how the thermal and
mechanical exergies are affected within a range from ambient pressure to inlet pressure. And re-performing the
tasks by using air instead of water steam.
Using a Python function the calculation of thermal and mechanical exergy can be performed for different
temperatures, pressures and fluids.
def c a l c s p l i t t e d p h y s i c a l e x e r g y ( p , h , p0 , T0 , f l u i d ) :
r""" Calculate specific physical exergy according to splitting rule."""
s = PSI ( "S" , "P" , p , "H" , h , f l u i d )
h T0 p = PSI ( "H" , "P" , p , "T" , T0 , f l u i d )
s T0 p = PSI ( "S" , "P" , p , "T" , T0 , f l u i d )
ex therm = ( h − h T0 p ) − T0 * ( s − s T0 p )
h0 = PSI ( "H" , "P" , p0 , "T" , T0 , f l u i d )
s0 = PSI ( "S" , "P" , p0 , "T" , T0 , f l u i d )
ex mech = ( h T0 p − h0 ) − T0 * ( s T0 p − s0 )
r e t u r n ex therm , ex mech

e x T i n , e x M i n = c a l c s p l i t t e d p h y s i c a l e x e r g y ( p i n , h i n , p0 , T0 , "water" )

Figure 6a and 6b show the difference in the values between inlet and outlet of the thermal and mechanical
exergies, representing the corresponding exgery destruction in the pipe flow for varying pressure ratios. Note
the different scales of the y-axes. The thermal exergy of water steam changes by up to 300 kW, while the
mechanical exergy difference is 0.9 kW at maximum. For air the thermal exergy difference reaches a maximum
of around 0.5 kW, the mechanical exergy changes up to 200 kW.

300 0.9 0.5 200

ĖDM = Ė1M − Ė2M [kW]

ĖDM = Ė1M − Ė2M [kW]

ĖDT = Ė1T − Ė2T [kW]

ĖDT = Ė1T − Ė2T [kW]

150
200 0.6
0.25 100
100 0.3
50

0 0 0 0
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9
Pressure loss whithin pipe, p1 − p2 [bar] Pressure loss whithin pipe, p1 − p2 [bar]
(a) Water steam (b) Air

Figure 6: Thermal and mechanical exergy destruction within the the pipe for varying pressure losses

The fundamental difference of the substance characteristic leads to the fact that the exergy destruction for water
steam in the pipe flow is mainly triggered by a decrease of the thermal exergy. In the case of air, the mechanical
exergy decreases significantly more.
A comparison of the exergy destruction rate for the pipe flow of water steam and air is shown in Figure 7. The
ratio of both exergy destruction rates is almost constant and is about 1.5.

300
ĖD = Ė1 − Ė2 [kW]

Water steam
200
Air
100

0
0 1 2 3 4 5 6 7 8 9
Pressure loss whithin pipe, p1 − p2 [bar]

Figure 7: Exergy destruction for water steam and air within the adiabatic pipe flow

Lessons Learned from this task:

• Physical exergy of a mass flow can be split into a thermal and a mechanical part.
• At the same pressure and temperature, the shares of thermal and mechanical exergies depend on the
working fluid.

201 https://doi.org/10.52202/069564-0019
 4.2. Highlight 2: Open-Cycle Gas Turbine
A simple gas turbine model should be developed within this task. Compressed ambient air is mixed up with
methane. The gases after the combustion are used in a gas turbine expander to supply mechanical, respectively
electrical energy.

SRC-F
3
System Boundary

AC Air Compressor
2 4 CC Combustion Chamber
EXP EXP Expander
CC G Generator
SRC-A Source, Air
AC Ẇel SRC-F Source, Fuel
SNK-E Sink, Exhaust

1 5
SRC-A SNK-E

Figure 8: Flowchart of open-cycle gas turbine

Table 2: Open-cylce gas turbine parameters

Parameter Symbol Unit Value

Net power output Ẇnet MW 150
Air compressor, isentropic efficiency ηs,AC – 0.85
Air compressor, pressure ratio rp,AC – 18
Combustion chamber, thermal efficiency ηCC – 0.98
Combustion chamber, pressure ratio rp,CC – 0.95
Expander, isentropic efficiency ηs,AC – 0.9
◦
Expander, inlet temperature T4 C 1400
Expander, outlet pressure p5 bar 1.013
Fuel, inlet pressure p3 bar 20
◦
Fuel, inlet temperature T3 C 25
Fuel, mass analysisa yi,fuel – (... , 1.0)
⎛ ⎞
0.7551
⎜0.2314⎟
⎜ ⎟
⎜0.0129⎟
Air, mass analysisa yi,air – ⎜ ⎟
⎜0.0006⎟
⎜ ⎟
⎝ 0.0 ⎠
0.0
a
 T
All fluid compositions are given as yi = yN2 , yO2 , yAr , yCO2 , yH2 O , yCH4

TESPy is used to set up the thermodynamic model of the gas turbine. The solution is carried out for the full load
case. The documentation of the source code can be found in the appendix.
A parameter study should be performed based on the results for the air compressor pressure ratio and
expander inlet temperature. The solution is a multiple run of the simulation above, while varying pressures and
temperatures.
Figure 9 depicts a parameter plot for the given gas turbine system varying expander inlet temperature from
900 ◦ C to 1400 ◦ C and the air compressor pressure ratio from 5 to 30. The reference point corresponds to
the data from Table 2. For the plot only the marked points were calculated, the lines in between are linear
regressions. If necessary, the accuracy of the graph can be increased by performing more calculations. An
optimal combination of expander inlet temperature and pressure ratio exists to maximize the specific work and
the efficiency of the gas turbine.

https://doi.org/10.52202/069564-0019 202
 0.45

Gas turbine efficiency, ηGT [–]

0.4 30
Reference
15
0.35
10

C
◦
0.3

0
0
0

0
1000

40
120

130
110
900

=1
T4
0.25
r p,AC = 5

100 200 300 400 500 600

Gas turbine specific work, wGT [kJ/kg]

Figure 9: Gas turbine parameter plot for varying expander inlet temperature air compressor pressure ratio

However, with this initially only energetic evaluation of the process, based on the specific power and the net
efficiency, the real thermodynamic losses cannot be quantified and the performance of the system components
cannot be compared with each other. The exergy analysis answers these open questions.
In case TESPy is used for the simulation of the thermodynamic performance, an exergy-based evaluation can
be carried out using the class ExergyAnalysis. The students have to set up the exergy flow rates crossing
the system boundary and define the exergetic fuel, the exergetic product and the exergetic loss of the overall
process.
The busses for the net work rate and the fuel input are already defined. In case of fuel the chemical exergy must
be considered. Additionally, an exergy loss bus must be defined for the exhaust gas.
The generic exergy analysis of TESPy can be set up within a few lines of code, defining the above mentioned
exergy flow rates and the ambient conditions.
from t e s p y . t o o l s import E x e r g y A n a l y s i s
ean = E x e r g y A n a l y s i s ( g a s t u r b i n e , E P = [ w o r k n e t ] , E F = [ e x f u e l ] , E L = [ e x l o s s ] )
ean . analyse ( pamb=1.013 , Tamb=25)

The graphical evaluation of the exergy analysis can be done, for example, with a waterfall diagram. Penkuhn
et al. [26] show this for a cogeneration plant. Another form of visualization is the representation of exergy
rates in a Sankey diagram, often called a Grassmann diagram.14 The exergy analysis results are linked to the
representation of the process flow diagram. The different material and energy streams of the flowsheet are
brought to a uniform basis with the help of the exergy. The width represents the exergy rate. In addition, the
exergy destruction is directly assigned to each component. Figure 10 shows this for the gas turbine process
under investigation. The highest exergy destruction rate occurs in the combustion chamber, followed by the
expander and air compressor. The exhaust gases represent an exergy loss for the overall process.

fuel combustion chamber

ED

EF

exhaust EL

EP

air turbine

air compressor work rate

Figure 10: Grassmann diagram of gas turbine process showing the results of the exergy-based assessment

14 Grassmann [27] transferred the idea of Sankey to the exergy concept. Exergy flow diagrams are therefore associated with his name.

203 https://doi.org/10.52202/069564-0019
 The light tint indicates physical exergy, and the dark tint indicates chemical exergy, for example, visible in the fuel
exergy flow rate or the exergy flow rate between the combustion chamber and the expander. It can be observed
that all flows are either nearly pure chemical or pure physical exergy.
Lessons Learned from this task:

• Learn that TESPy can do a lot of work for you.

• Relevant thermodynamic process parameters identified and numerically evaluated.
• Applied exergy analysis for a simple process for the first time, and identified the locations and magnitudes
of real thermodynamic inefficiencies.

4.3. Highlight 3: Process Design Study for a Combined Heat and Power Plant
This example contains a process design study. A proposal for a power plant to supply electricity and district
heating is to be developed. The heat source is a geothermal field.
The design proposal is to be modeled, simulated, and evaluated independently by the students. The design
based on the principle of cogeneration is to be compared with a process design with complete condensation
operation (without heat extraction). For this purpose, the energetic and exergetic efficiency of the overall process
will be defined.
Ẇnet + Q̇DH
η= (1)
Ḣ1 − Ḣ2

 
Ẇnet + ĖFL − ĖRL
ε= (2)
Ė1 − Ė2

All predefined parameters and boundary conditions can be found in Figure 11 and Table 3.

Products
Electricity
Ẇnet ≥ 80 MW

H2 O Combined heat and

ṁ1 = 500 kg/ s power plant District Heating
p1 = 10.8 br
FL Q̇DH ≥ 400 MW
1 = 0.45 pFL = 10 br
RL TRL = 50 ◦ C
ΔTFL,RL = 50 K
1 2

T2 ≥ 55 ◦ C Coolant
p2 = 30 br
SNK-COOL
Geothermal Field

SRC-GF SNK-GF SRC-COOL

T = 25 ◦ C
H2 O or Air

Figure 11: Process design study for a combined heat and power plant to provide electricity and district heating.

The following subtasks are to be solved within the scope of the assigned problem: Selection of suitable working
fluids and selection and reasoning of the use of thermodynamic properties models for the simulation; creation of
a process design, modeling, and simulation in TESPy; definition and use of relevant variables for the evaluation
of the process designs and subsequent comparison and discussion.
There is no generally valid standard solution for solving this task. Two possible designs with and without heat
extraction are shown in Figure 12. In both variants, the heat flow of the liquid part of the geothermal fluid is
absorbed by an upstream organic Rankine cycle (ORC) and used to produce electricity. The steam phase is
expanded in a steam turbine. In the case of simultaneous provision of district heating, a heating condenser with
preheating follows before the geothermal fluid is returned. In the design for electricity generation only, a second
ORC is inserted after the steam turbine.
Both process designs are simulated and compared both energetically and exergetically. The results are shown
in Table 4. According to Eq. (1), the energetic evaluation treats two different energy flows (electricity and heat)

https://doi.org/10.52202/069564-0019 204
 Table 3: Process design study specifications and constraints

Component Parameter Unit Value

Auxiliary installations Electrical power requirements MW 3
Generators Mechanical-electrical efficiency – 0.985
Motors Electrical-mechanical efficiency – 0.975
Steam turbines Isentropic efficiency – 0.9
Mechanical efficiency – 0.99
Minimum steam quality – 0.9
Pumps Isentropic efficiency – 0.8
Mechanical efficiency – 0.99
Heat exchangers Minimal temperature difference liquid–liquid K 10
liquid–condensing K 5
liquid–steam K 20
liquid–gaseous K 25

12

ST

SRC-GF T1
11

SEP
35
17
D1
36 13
34
EV1 SNK-DH
63
CD-DH
33

18 14 62
SNK-CL
32
ECO-DH
ECO1

61
M2
SRC-CL SRC-DH
31
M4 19 M3
37

P2
21 20 16 15

SNK-GF
P4 P3

(a) Geothermal combined heat and power plant

12

ST

SRC-GF T1 T2
11

SEP
35 45
17
D1 D2
36 13 46
34 44
EV1 EV2

33 43

18 14
SNK-CL1 SNK-CL2
32 42
ECO1

ECO2

M2 M2
SRC-CL1 SRC-CL2
31 CD1 41 CD2
M4 19 M3
37 47

SNK-GF P2 P2
21 20 16 15

P4 P3

(b) Geothermal power plant

Figure 12: Flowcharts of possible process designs

205 https://doi.org/10.52202/069564-0019
 equally. The efficiency is significantly higher in the case of heat extraction. It gives the impression that the
process is much closer to the thermodynamic optimum, which is not the case, as the comparison of the exergetic
efficiencies shows. Here, both exergetic products can be directly converted into each other. The result is that
the process designs have approximately similar range of exergetic efficiencies, and therefore the misestimation
of the energetic evaluation can be revealed.

Table 4: Comparison of energetic and exergetic efficiency for developed process designs

Process design Energetic efficiency η [%] Exergetic efficiency ε [%]

Geothermal combined heat and power plant 80.4 70.3
Geothermal power plant 19.0 62.8

Lessons Learned from this task:

• The exergy analysis assists in the selection and improvement of a suitable process design.
• Different material and energy streams can be consistently evaluated based on exergy.
• Various process designs with different fuels or products remain comparable with the help of exergy.

5. Conclusion
The idea presented in this article is based on the notion that research and teaching should be accessible to all
with as few barriers as possible. At the same time, the paper addresses the exergy concept, one of the essential
analysis tools of thermodynamic engineering. Object-oriented programming with Python, particularly steady-
state power plant simulation with open-source software, is linked to exergy as an established thermodynamic
quantity in the concept presented.
A modular structure is developed to facilitate the integration of the resources into existing courses and for
self-learners to choose their appropriate entry point. The modular structure covers exergy analysis of single
component operation as the entry-level, where the fundamental understanding of the exergy concept is conveyed.
Based on that, students will learn to model simple thermodynamic cycles and to carry out first and second-
law analyses by evaluating the simulation results. It is further supported by applying parametric analysis to
understand the connection between individual component parameters to the thermodynamic performance of the
overall process.
Further processing possibilities arise from the potential integration of additional exercises, the extension to the
exergoeconomic analysis [28], the exergoenvironmental analysis [29], or the advanced exergy analysis [30].
At the same time, all interested students, researchers, and practitioners are invited to try the course, and to
contribute by making comments or reporting bugs and developing it further collaboratively.

CRediT author statement

Mathias Hofmann: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation,
Resources, Data Curation, Writing - Original Draft, Writing - Review & Editing, Visualization Francesco Witte:
Conceptualization, Methodology, Software, Validation, Formal analysis, Data Curation, Writing - Review & Editing
Malte Fritz: Software, Validation, Formal analysis, Data Curation Jonas Freißmann: Software, Validation,
Formal analysis, Data Curation Ilja Tuschy: Review & Editing, Supervision George Tsatsaronis: Review &
Editing, Supervision
Malte Fritz and Jonas Freißmann contributed equally to this work.

Acknowledgments
The authors would like to thank Björn Kiehne (Berliner Zentrum für Hochschullehre) for his remarks on Section 1.
and 2. Mathias Penkuhn (Technische Universität Berlin) has to be credited for the conceptual design of the task
given in Section 4.3.

Nomenclature
Abbreviations

A Air
AC Air compressor
CC Combustion chamber

https://doi.org/10.52202/069564-0019 206
 CD Condenser
CL Coolant
D Drum
DH District heating
E Exhaust
ECO Economizer
EXP Expander
EV Evaporator
F Fuel
FL Feed line
G Generator
GF Geothermal fluid
GT Gas turbine
M Motor
ORC Organic Rankine cycle
P Pump
SEP Separator
SNK Sink
SRC Source
ST Steam turbine
T Turbine
RL Return line

Letter symbols

Ė Exergy rate, W
Ḣ Enthalpy rate, W
ṁ Mass flow rate, kg/s
p Pressure, bar
Q̇ Heat rate, W
r Ratio, –
T Temperature, ◦ C
w Specific work, kJ/kg
Ẇ Work rate, W
x Quality, –
yi Mass fraction, –

Greek symbols

Δ Difference
ε Exergetic efficiency
η Energetic efficiency

Subscripts and superscripts

0 At ambient state
D Destruction
F Fuel
i Stream
p Pressure
P Product
s Isentropic

207 https://doi.org/10.52202/069564-0019
 References
[1] D. Abdrasheva, M. Escribens, E. Sabzalieva, D. V. do Nascimento, and C. Yerovi. Resuming or Reforming? Tracking
the global impact of the COVID-19 pandemic on higher education after two years of disruption. Retrieved: January
27th, 2023. 2022. URL: https://unesdoc.unesco.org/ark:/48223/pf0000381749.
[2] K. H. Mok. “Impact of COVID-19 on Higher Education: Critical Reflections”. In: High. Educ. Policy 35 (2022), pp. 563–
567. DOI: 10.1057/s41307-022-00285-x.
[3] F. S. von Thun. “Wie gestalte ich meine Vorlesung – und halte die Hörerschaft und mich selbst bei Laune?” In: Die
Qualität Akademischer Lehre. Ed. by M. Merkt and K. Mayrberger. in German. Innsburck, Austria: Studienverlag,
2007, pp. 115–131.
[4] L. W. Anderson and D. R. Krathwohl, eds. A Taxonomy for Learning, Teaching, and Assessing: A Revision of Bloom’s
Taxonomy of Educational Objectives. New York: Longman, 2001.
[5] Z. Rant. “Exergie, ein neues Wort für technische Arbeitsfähigkeit”. In: Forschg. Ing.-Wes. 22.1 (1956). in German,
pp. 36–37.
[6] T. J. Kotas. The Exergy Method of Thermal Plant Analysis. London, UK: Butterworths, 1985.
[7] W. Fratzscher, M. Brodjanskij, and K. Michalek. Exergie. in German. Leipzig, Germany: Verlag für Grundstoffindustrie,
1986.
[8] A. Bejan, G. Tsatsaronis, and M. Moran. Thermal design and optimization. New York, USA: Wiley, 1996.
[9] J. Szargut. Exergy Method. Southampton, UK: WIT Press, 2005.
[10] W. Rüegg, ed. Geschichte der Universität in Europa. Vol. 1, Mittelalter. in German. München, Germany: Beck, 1993.
[11] R. A. Müller. Geschichte der Universität. in German. München, Germany: Callwey, 1990.
[12] M. J. Moran, H. N. Shapiro, D. D. Boettner, and M. B. Bailey. Fundamentals of Engineering Thermodynamics. 9th ed.
New York, USA: John Wiley, 2018.
[13] H. D. Baehr and S. Kabelac. Thermodynamik. in German. Berlin, Germany: Springer, 2016. DOI: 10.1007/978-3-662-
49568-1.
[14] E. Matthes. Python Crash Course. San Francisco, USA: No Starch Press, 2023.
[15] A. Galea. Beginning Data Analysis with Python and Jupyter. Birmingham, UK: Packt Publishing, 2018.
[16] F. Witte and I. Tuschy. “TESPy: Thermal Engineering Systems in Python”. In: J. Open Source Softw. 5 (2020), p. 2178.
DOI : 10.21105/joss.02178.

[17] F. Witte, M. Hofmann, J. Meier, I. Tuschy, and G. Tsatsaronis. “Generic and Open-Source Exergy Analysis—Extending
the Simulation Framework TESPy”. In: Energies 15 (2022), p. 4087. DOI: 10.3390/en15114087.
[18] M. Hofmann, F. Witte, K. Shawky, I. Tuschy, and G. Tsatsaronis. “Thermal Engineering Systems in Python (TESPy):
The implementation and validation of the chemical exergy”. In: Proceedings of ECOS 2022. Copenhagen, Denmark,
2022, pp. 257–269.
[19] S. Pfenninger, J. DeCarolis, L. Hirth, S. Quoilin, and I. Staffell. “The importance of open data and software: Is energy
research lagging behind?” In: Energy Policy 101 (2017), pp. 211–215. DOI: 10.1016/j.enpol.2016.11.046.
[20] The Journal of Open Source Education. Retrieved: March 23rd, 2023. URL: https://jose.theoj.org/.
[21] R. Morrison. einfach : an energy system model for education. 2020. DOI: 10.5281/zenodo.7746137.
[22] I. H. Bell, J. Wronski, S. Quoilin, and V. Lemort. “Pure and Pseudo-pure Fluid Thermophysical Property Evaluation and
the Open-Source Thermophysical Property Library CoolProp”. In: Ind. Eng. Chem. Res. 53.6 (2014), pp. 2498–2508.
DOI : 10.1021/ie4033999.

[23] F. Biscani and D. Izzo. “A parallel global multiobjective framework for optimization: pagmo”. In: J. Open Source Softw.
5.53 (2020), p. 2338. DOI: 10.21105/joss.02338.
[24] J. Ahrendts. “Reference states”. In: Energy 5 (1980), pp. 666–677. DOI: 10.1016/0360-5442(80)90087-0.
[25] T. Morosuk and G. Tsatsaronis. “Splitting physical exergy: Theory and application”. In: Energy 167 (2019), pp. 698–
707. DOI: 10.1016/j.energy.2018.10.090.
[26] M. Penkuhn, M. Hofmann, S. Meinke, and C. Lösche. “Prozessbewertung und Exergieanalyse für ein Heizkraftwerk”.
In: Kraftwerkstechnisches Kolloquium. Dresden, Germany, 2021, pp. 1–12. DOI: 10.14279/depositonce-15966.
[27] P. Grassmann. “Die Exergie und das Flussbild der technisch nutzbaren Leistung”. In: Allg. Wärmetechn. 9.4/5 (1959).
in German, pp. 79–86.
[28] G. Tsatsaronis. “Thermoeconomic analysis and optimization of energy systems”. In: Prog. Energy Combust. Sci. 19
(1993), pp. 227–257. DOI: 10.1016/0360-1285(93)90016-8.
[29] L. Meyer, G. Tsatsaronis, J. Buchgeister, and L. Schebek. “Exergoenvironmental analysis for evaluation of the
environmental impact of energy conversion systems”. In: Energy 34 (2009), pp. 75–89. DOI: 10.1016/j.energy.2008.
07.018.
[30] M. Penkuhn and G. Tsatsaronis. “A decomposition method for the evaluation of component interactions in energy
conversion systems for application to advanced exergy-based analyses”. In: Energy 133 (2017), pp. 388–403. DOI:
10.1016/j.energy.2017.03.144.

https://doi.org/10.52202/069564-0019 208
A Appendix: Open-Cycle Gas Turbine – TESPy Source Code
from t e s p y . networks import Network
from t e s p y . components import ( Sink , Source , Compressor , DiabaticCombustionChamber , T u r b i n e )
from t e s p y . c o n n e c t i o n s import Connection , Bus

# general c o n f i g u r a t i o n s
# substances , f l u i d s , network , u n i t s
f l u i d l i s t = [ ’N2’ , ’O2’ , ’Ar’ , ’CO2’ , ’H2O’ , ’CH4’ ]
g a s t u r b i n e = Network ( f l u i d s = f l u i d l i s t , p u n i t =’bar’ , T u n i t =’C’ , h u n i t =’kJ / kg’ )

# composition of given f l u i d s
a i r = {’N2’ : 0.7551 , ’O2’ : 0.2314 , ’Ar’ : 0.0129 , ’CO2’ : 0.0006 , ’H2O’ : 0 , ’CH4’ : 0}
f u e l = {’N2’ : 0 , ’O2’ : 0 , ’Ar’ : 0 , ’CO2’ : 0 , ’H2O’ : 0 , ’CH4’ : 1}

# d e f i n e sources and s i n k s
s r c a i r = Source ( ’air’ )
s r c f u e l = Source ( ’fuel ’ )
snk exhaust = Sink ( ’exhaust ’ )

# define components
cmp AC = Compressor ( ’air compressor ’ )
cmp CC = DiabaticCombustionChamber ( ’combustion chamber ’ )
cmp EX = T u r b i n e ( ’expander ’ )

# define connections
c1 = Connection ( s r c a i r , ’out1 ’ , cmp AC , ’in1’ , l a b e l =’1’ )
c2 = Connection ( cmp AC , ’out1 ’ , cmp CC , ’in1’ , l a b e l =’2’ )
c3 = Connection ( s r c f u e l , ’out1 ’ , cmp CC , ’in2’ , l a b e l =’3’ )
c4 = Connection ( cmp CC , ’out1 ’ , cmp EX , ’in1’ , l a b e l =’4’ )
c5 = Connection ( cmp EX , ’out1 ’ , snk exhaust , ’in1’ , l a b e l =’5’ )

# add c o n n e c t i o n s t o network
g a s t u r b i n e . add conns ( c1 , c2 , c3 , c4 , c5 )

# parameter o f components
cmp AC . s e t a t t r ( e t a s =0.85 , p r =18)
cmp CC . s e t a t t r ( eta =0.98 , pr =0.95)
cmp EX . s e t a t t r ( eta s =0.9)

# parameter o f c o n n e c t i o n s
c1 . s e t a t t r ( p =1.013 , T=25 , f l u i d = a i r )
c3 . s e t a t t r ( p=20 , T=25 , f l u i d = f u e l )
c4 . s e t a t t r ( T=1400)
c5 . s e t a t t r ( p =1.013)

# busses
w o r k n e t = Bus ( ’work netto ’ )
f u e l i n = Bus ( ’fuel input ’ )

w o r k n e t . add comps (
{’comp ’ : cmp AC , ’base ’ : ’bus’ , ’char ’ : 1} ,
{’comp ’ : cmp EX , ’char ’ : 1})

f u e l i n . add comps (
{’comp ’ : cmp CC , ’base ’ : ’bus’ } )

g a s t u r b i n e . add busses ( work net , f u e l i n )

# parameter o f busses
w o r k n e t . s e t a t t r (P=−150e6 )

# s o l v e network
g a s t u r b i n e . s o l v e ( ’design ’ )

# p r i n t network r e s u l t s
gas turbine . p r i n t r e s u l t s ( )

209 https://doi.org/10.52202/069564-0019