JOURNAL OF THE GLOBAL POWER AND PROPULSION SOCIETY

journal.gpps.global

Creep-damage modelling for micro gas turbine

combustion chambers lifetime prediction
Review article Daniele Cirigliano1,*, Herol Lawerence D’Souza2, Felix Grimm1, Peter Kutne1,
Manfred Aigner1
Article history:
1
German Aerospace Center (DLR), Pfaffenwaldring 38-40, 70569 Stuttgart, Germany
Submission date: 6 October 2022 2
Ruhr University-Bochum, Universitätsstraße 150, 44801 Bochum, Germany
Acceptance date: 5 April 2023
Publication date: 9 June 2023
This is the updated version of a paper Abstract
originally presented at the Global Power
and Propulsion Technical Conference, Micro Gas Turbines (MGTs) are nowadays largely used for electrical and
GPPS Chania22, September 11–14, 2022. thermal energy production in small buildings and households. Their reliabil-
ity and compactness allow them to operate for thousands of hours with
minimal maintenance. However, the long exposure at high temperatures in
combustion chambers can promote creep, which can induce thermal
*Correspondence: fatigue and potential failure of these components. Creep-induced damage
DC: daniele.cirigliano@dlr.de
in MGTs has not yet been thoroughly investigated, due to the lack of
numerical tools able to model these strongly coupled phenomena. This
study presents the development of a Fortran-based subroutine integrated
Peer review:
into ANSYS APDL. The code allows for a life assessment based on the
Single blind
Lemaitre-Chaboche creep damage model. Secondary creep and stress
Copyright: relaxation are modeled for the high-temperature resistant alloy Inconel718.
© 2023 Cirigliano et al. This is an open A new set of temperature-dependent parameters for the Norton equation is
access article distributed under the Creative provided, and the method to obtain these parameters from creep rupture
Commons Attribution License (CC-BY 4.0), tests is outlined. The model is validated and shows good agreement with
which permits unrestricted use, distribution, experimental data. The subroutine correctly reproduces visco-plasticity,
and reproduction in any medium, provided stress relaxation and damage under typical MGTs operating temperatures.
the original work is properly cited and its This model constitutes the foundation of a life-assessment analysis for
authors credited. combustion chambers. The results highlight the impact of temperature and
creep on the component’s life and the importance of integrating life assess-
Keywords: ment analysis into the preliminary design of combustion chambers.
thermomechanical-fatigue; creep; damage;
MGT; FEM
Introduction
Citation:
Cirigliano D., D’Souza H. L., Grimm F.,
Micro Gas Turbines (MGTs) for co-, tri- or polygeneration are becom-
Kutne P., and Aigner M. (2023). Creep- ing progressively more widespread because of their reliability, flexibility,
damage modelling for micro gas turbine compactness and very low NOx emissions (Sun et al., 2012). Despite
combustion chambers lifetime prediction. the on-going energy transition and the short-term impact due to the lin-
Journal of the Global Power and Propulsion gering COVID-19 pandemic, the global production of MGTs (especially
Society. 7: 166–176. in Europe) shows a promising increasing trend for the next decade
https://doi.org/10.33737/jgpps/163088 (Palmer, 2021). Although the electrical efficiency is generally lower than
the one for similar size internal combustion engines (Ofualagba, 2012),
the high total efficiency of MGTs, which can reach 80% thanks to the
extraction of the high exhaust gases thermal content through a recuper-
ator (Beith, 2011) is one of the main reason for their increasing adop-
tion. Moreover, exhaust heat can be further extracted by means of a heat
exchanger for the production of hot water.
However, if on one side recuperation is beneficial for the overall
system efficiency, on the other side it reduces the ability to cool down
the hot mechanical components, such as the combustion chamber,
being the inlet air largely warmer than what it would be without

J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 166

Cirigliano et al. | Creep-damage modelling for lifetime prediction https://www.journalssystem.com/jgpps/163088,0,2.html

recuperation. In MGTs, the air entering the combustion chamber is often above 650°C (Enagi et al., 2017). The
lower cooling capability, in addition to the higher flame temperature due to warmer inlet gases, causes very high
temperatures in the metal components of the combustion chamber, with peaks of 1,200 K in some cases
(Cirigliano et al., 2020). Frequent starts and stops induce mechanical stresses at the boundary between materials
due to their different thermal expansion, which, if taking place too fast, may cause local permanent plastic defor-
mations. Moreover, long exposure at high temperatures in combustion chambers can promote creep, which can
induce thermal fatigue and potential failure of these components. The accumulation of damage in combustion
chambers can hence be divided into two categories: the plasticity taking place during transient states (maneuvers,
starts and stops) and the viscoplastic-based damage (creep damage) at steady state operation.
It is then necessary to take into account the material degradation due to the progressively increasing viscoplas-
tic deformation; this can be done by adding a scalar parameter to the mechanical model. Commercial Finite
Element Methods (FEM) codes, like ANSYS, allow the user to write their custom mechanical model. In this
work, the development of a Fortran-based subroutine integrated into ANSYS APDL is presented. On so doing,
creep, damage, and material degradation can be modeled. In this work, Lemaitre-Chaboche creep damage model
is adopted (Lemaitre and Chaboche, 1978). Secondary creep and isotropic hardening are modeled for the high-
temperature resistant alloy Inconel 718. The material degradation is considered by coupling the Young’s
modulus with the introduced damage.
The novelty of this paper consists in the determination of Norton coefficients for Inconel 718, which instead
of being assumed constant, are temperature dependent and based on experimental data. It is shown that the
creep model and subroutine based on these new coefficients correctly reproduces creep, stress relaxation and
damage under typical MGT operating temperatures. This model constitutes the foundation of a life-assessment
analysis for combustion chambers under creep damage.

Numerical methods

Creep modelling
Creep is described as a time-dependent deformation of a material subjected to high temperature and loads below
its yield strength for an extended period of time. Temperature, stress, time and material properties influence the
rate of creep deformation. The creep phenomena is mainly divided into three stages; i.e., primary I, secondary II,
and tertiary III. The three stages correspond to strain rates that are decreasing, constant, and increasing, respect-
ively. Figure 1 depicts a typical creep strain versus time curve. The stage I in the curve indicates the primary
creep where the creep strain rate decreases to a minimum value, due the fact that the material is experiencing an
increase in creep resistance, attributed to a reduction in the density of free dislocations. The stage II is known as
steady-state creep or secondary creep. This regime is also termed as a state of balance between the rate of disloca-
tion generation and rate of recovery, where the former contributes to hardening and the latter to softening of the
test material (Abe, 2014). In the tertiary creep stage, the creep rate increases rapidly until failure. This stage
describes a softening stage, where the cracks are formed and propagate along the grain boundaries (Betten,
2008). In addition, the microscopic voids accumulate within the material, resulting in a decrease in the effective
cross-sectional area (i.e., Necking), which raises the effective stress.

Figure 1. Typical creep strain vs. time curve showing the three stages of creep.

J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 167

Cirigliano et al. | Creep-damage modelling for lifetime prediction https://www.journalssystem.com/jgpps/163088,0,2.html

In order to capture the damage caused by creep, the modified Lemaitre-Chaboche model (Chaboche and
Rousselier, 1983; Lemaitre, 1985) is used in this study. The effective stress concept Lemaitre and Chaboche
(1978) defines the creep damage Dcr starting from the effective section S~ and the effective uniaxial stress σ:

S σ
σ~ ¼ σ ¼ (1)
S~ 1  Dcr

where Dcr represents the macroscopic effect of the mechanical degradation due to creep voids, and
S~ ¼ S  Svoids . Hence, strictly speaking, Dcr would be defined as Dcr ¼ Svoids =S. Given the difficulty to obtain
a reliable measure of Svoids during operation, it is more practical to describe the status of Dcr as the integral of its
time evolution:
ðt
Dcr (t) ¼ D_ cr (t)dt (2)
0

Equivalent Von-Mises stress and equivalent strain are defined in terms of their stress and strain components:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
(σ 1  σ 2 )2 þ (σ 2  σ 3 )2 þ (σ 3  σ 1 )2
σ eq ¼ (3)
2

pffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
ϵeq ¼ (ϵ1  ϵ2 )2 þ (ϵ2  ϵ3 )2 þ (ϵ3  ϵ1 )2 : (4)
3

The total equivalent strain ϵeq is given by the sum of elastic, plastic, thermal and creep equivalent strains:

ϵeq ¼ ϵeleq þ ϵeqpl þ ϵth

eq þ ϵeq
cr
(5)

Creep mechanisms can generally be grouped into two categories: diffusion creep and dislocation creep.
Diffusion creep is dominant at lower stresses, and is caused by the movement of vacancies; dislocation creep is
present at higher stresses and is caused by the movement of dislocations through the lattice. When a dislocation
encounters an obstacle (for example another defect), the former can either climb it (at mid stresses) or glide
through it (at high stresses). The creep mechanisms acting in a material can be identified for a range of stresses
and temperatures in a deformation map, according to the methods described by Frost and Ashby (1982). A
deformation map of Inconel 718 is not available to the authors, but similar Nickel-based superalloys, such as
IN738LC Carey et al. (1990) or MAR-M-200 Cieśla et al. (2016), show the onset of dislocation creep at half of
the melting point (about 600 C) and at stresses above 50 MPa. The operative conditions at which MGT com-
bustion chambers are operated (metal temperature and stresses due to thermal strains) lay above these values.
Moreover, it has been shown in other studies that the validity of creep power laws, for example the Norton law,
can be extended by accounting for temperature-dependent parameters Golan et al. (1996) and Alain (1998) or
by introducing recovery processes associated to the dislocation structure Das et al. (2022). Barbosa et al. (1988)
used physics-based four stress- and temperature-dependent parameters to describe creep curves. McLean et al.
(1992) modelled creep curves of single crystal superalloys using eight temperature- and stress-dependent para-
meters. On so doing, power law can be used to describe the creep behavior over an extensive range of stress and
temperature. For these reasons, this work aims to model dislocation creep only, and to adapt the Norton law
material parameters to different temperatures.
The derivative of creep equivalent strain in time ( primary and tertiary creep) can be described in the form of
its dependency on equivalent stress, temperature and either creep strain or time, by means of a power law:

ϵ_ creq ¼ C1 σ Ceq2 ϵcreq C3 e C4 =T , or ϵ_ creq ¼ C1 σ Ceq2 t C3 e C4 =T (6)

J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 168

Cirigliano et al. | Creep-damage modelling for lifetime prediction https://www.journalssystem.com/jgpps/163088,0,2.html

where T is the temperature and t is the time. For secondary creep, however, the strain rate is constant, hence,
C3 = 0:

ϵ_ creq ¼ C1 σ Ceq2 e C4 =T (7)

where C1 , C2 and C4 are material constants whose values are determined using creep tests. Equation 7 is the
most important and widely used formulation to predict the secondary creep and is known as Norton power law.
Table 1 shows the values of C1 , C2 and C4 for Inconel 718 according to (Liu et al., 2015) and based on creep
test data (Li et al., 2010). Note that for these parameters, stress is expressed in MPa, temperature in K, and
strain in units per hour.
The creep damage evolution is given by:
σ r
D_ cr ¼ (1  Dcr )k
eq
(8)
A

where Dcr is the scalar creep damage parameter. D_ cr is expressed in units per second and k is given by:

k ¼ a0 þ a1 (σ eq  z) þ a2 (σ eq  z)2 (9)

where a0 , a1 , a2 , z, r and A are material constants which are determined again by creep tests. For Inconel 718
these values are reported in Table 2 (Zhang, 1995).
The scalar creep damage variable Dcr is zero for undamaged material and increases in time with Equation 8 up
to an arbitrary value, usually 1, at which the material is assumed to fail. Basing on the damage theory of
(Kachanov, 1999) and taking into account the concept of effective stress (Rabotnov et al., 1970) and the prin-
ciple of strain equivalence (Lemaitre and Chaboche, 1978), changes in mechanical behavior can be measured
through the evolution of the elastic modulus Lemaitre (1985):

E ¼ E0 (1  Dcr ) (10)

where E is the effective elastic modulus and E0 is the Young’s modulus of undamaged material. By differentiating
Equation 10 in time, one obtains:
σ r
E_ ¼ E0 D_ cr ¼ E0 (1  Dcr )k
eq
(11)
A

By remembering the constitutive relation between σ, ϵel and the elasticity matrix [D], which contains E
(Chandrupatla and Belegundu, 2012), one can derive:
:
σ_ ¼ [D] ϵel þ [D]_ϵel (12)

Equation 12, together with Equations 7 and 8, constitute a system of nonlinear differential equations in σ eq ,
Dcr and ϵcreq , where T can be constant or vary with time too. Also by remembering the fact that ϵeleq and ϵcreq
are related by Equation 5, it is clear how this highly coupled problem needs to be solved by a dedicated code.

Table 1. Norton coefficients for Inconel 718 according to (Liu et al., 2015).

C1 C2 C4

2.147·10−7 10.171 50,825.89

J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 169

Cirigliano et al. | Creep-damage modelling for lifetime prediction https://www.journalssystem.com/jgpps/163088,0,2.html

Table 2. Material constants for damage evolution equation.

a0 a1 a2 r Z A

13.2478 0.7865·10−4 0.1924·10−3 13.19 733.25 1,209

Material parameters identification

In applying the Norton formulation (Equation 7) to the analysis of structures one should bear in mind that the
material parameters C1 , C2 and C4 are obtained by interpolating experimental creep tests, hence their validity is
limited to the narrow range of stresses and temperatures the tests were conducted. In the previous section, a set
of these parameters is proposed in Table 1. However, the use of only three constants to cover the whole spectrum
of stresses and temperatures a structure can be subjected to might introduce relevant inaccuracies in the model.
For this reason, in this paper a whole new set of parameters was derived from the creep tests of Inconel 718 at
different temperatures performed by (Brinkman et al., 1991). In Figure 2, the results of the creep tests are
shown. In this section, a method to derive Norton parameters from creep tests is highlighted.
Since the secondary creep stage is characterized by a constant creep strain rate, one can assume
d ϵcreq =dt ≃ ϵcreq,ss =tss , where ϵcreq,ss and tss are the equivalent creep strain and the time to the onset of tertiary creep,
respectively. Hence, the Norton relation (Equation 7) can be rewritten in the form:

!1=C2
ϵcreq,ss exp(C4 =T )
σ eq ¼ (13)
C1 tss

where T, tss and σ eq are expressed in Kelvin, hours and MPa respectively. The time to the onset of tertiary creep
can be related to the time to rupture for Inconel 718 by (Brinkman et al., 1991):

tss ¼ Atrβ (14)

with A = 0,442 and β = 1,04 up to tr ¼ 105 h, and A = 0,7 and β = 1 above. Relating ϵcreq,ss with the creep strain
to rupture, ϵcreq,r , remains an harder task, due to the large amount of scatter which is typical in creep measure-
ments. For temperatures between 650 C and 760 C, some creep tests indicate ϵcreq,r between 3% and 9%,
increasing with temperature (Kim et al., 2008; Sugahara et al., 2012; Caliari et al., 2016). A more handy correl-
ation is given again by (Brinkman et al., 1991):

ϵcreq,ss ¼ 0:2 þ A Btr(βα) (15)

where A and β are the same as shown above, 0.2 is the elastic strain offset, and B and α are reported in Table 3.
In this way, the determination of temperature-dependent parameters C1 , C2 and C4 can finally take place as
follows:
1. Obtain from creep tests points (σ, tr ) at different temperatures;
2. compute tss from Equation 14 with the corresponding A, β;
3. compute ϵcreq,ss from Equation 15 with the corresponding B,α;
4. substitute tss and ϵcreq,ss in Equation 13;
5. for every temperature T, find the values {C1 , C2 , C4 } giving a σ eq as close as possible to the one of the tests
( point 1).
The key point of this approach is noting that creep tests are performed at constant stress, hence, one can use
eq
the same σ eq of the creep rupture tests to relate the tr to the state on onset of tertiary creep, tss and ϵeq,ss of
Norton formulation, which is generally not known. The new set of parameters for Inconel 718 is presented in
Table 4. By applying these parameters in Equation 13, interpolating curves are generated and plotted in
Figure 2, showing excellent fit. A comparison between the use of constant parameters and temperature-
dependent parameters can be well appreciated in a 3D space (Figure 3).
J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 170
Cirigliano et al. | Creep-damage modelling for lifetime prediction https://www.journalssystem.com/jgpps/163088,0,2.html

Figure 2. Creep-rupture lifetime data and numerical interpolation for Inconel 718.

Table 3. B and α parameters.

tr (h) T ( C) B α

<11,000 ≥593 2.142 1.151

<11,000 ≤593 34.182 1.443

≥11,000 Any 2.142 1.151

J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 171

Cirigliano et al. | Creep-damage modelling for lifetime prediction https://www.journalssystem.com/jgpps/163088,0,2.html

Table 4. Norton equation coefficients at

different temperatures for Inconel 718.

T (°C) C1 C2 C4

427 1.10·10−9 21.8 109,000

482 3.00·10−9 15.8 84,000

538 4.00·10−9 14.4 81,000

593 2.00·10−9 12.3 71,000

649 1.00·10−7 5.4 34,500

704 5.00·10−6 3.7 29,000

760 2.00·10−7 4.5 28,000

User programmable feature USERMAT in ANSYS

In this section, the implementation of a damage-based material model in the commercial software ANSYS
APDL is outlined. User Programmable Features (UPFs) are a highly effective and flexible tool to tailor the behav-
ior of the APDL program to suit individual requirements. This is performed by writing a custom subroutine in
the C, C++, or Fortran programming languages. One of such subroutines is USERMAT, which is particularly
used to define non-linear stress-strain relationships of elastic-plastic materials, custom damage evolution and
creep laws, like in this study.
USERMAT subroutine is called at every iteration and executed on each element of the computational grid.
The input parameters, such as loads and temperature, are defined by the user during the modelling step. Current
stresses, strains and strain increments are the inputs at the start of the timestep. At each iteration, a new elastic,
plastic, thermal and creep strain, an effective Young’s modulus and damage evolution are computed based on the
constitutive equations and material model described in the previous sections. USERMAT then updates the stres-
ses and the material Jacobian matrix and these values are sent back to the main Finite Element code as outputs
(Lin, 1999). The status of every element is checked at every time increment using a strength lifetime failure cri-
terion: when the maximum damage of the structure reaches a threshold value, the subroutine is stopped and the
present time is recorded as the lifetime of the component. In this analysis, Dcr ¼ 0:4 is considered as the the
maximum allowable damage.

Figure 3. Creep-tests can be plotted in a three-dimensional space, where each point is represents a test conducted
at a specific σ eq , tr and T. By computing the creep strain with the Norton law, a 3D surface can be obtained and
overlapped to experimental data. A comparison between constant parameters (a) and temperature-dependent para-
meters (b) shows a much better fit of the latter. (a) Constant coefficients (Table 1). (b) Variable coefficients (Table 4).

J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 172

Cirigliano et al. | Creep-damage modelling for lifetime prediction https://www.journalssystem.com/jgpps/163088,0,2.html

The geometry chosen is a rectangular bar of Inconel 718 alloy with a base of 10 × 10 mm and a length of
100 mm. The bar is discretized with 2 hexahedral elements for each short side and with 20 along the length (80
elements in total). Since the application of the present work is three-dimensional and plain strain, SOLID186
elements (20 nodes each, second-order shape functions) are used. The bar is fixed at one end and a displacement
is applied to the opposite end to obtain the uni-axial stress state. Material properties of Inconel 718 such as
density, thermal expansion coefficients and Young’s modulus (of undamaged material) are taken from data sheets
(VDM Metals, 2020).

Results and discussion

The application of the described UPF using the temperature-dependent coefficients is presented in this section.
Temperature is set to 750 K. The comparison between the experimental tests, the Norton-Damage model with
constant parameters of (Liu et al., 2015) and the Norton-Damage model of this analysis is depicted in Figure 4.
In the experiments, a displacement is applied to one end in order to generate an initial stress of 700 MPa;
shortly after, the stress relaxes to a much lower value, around 635 MPa, showing the typical primary creep phase.
Afterwards, the stress stabilizes and slowly decreases, due to further stress relaxation. Finally, a rapid decrease in
the cross-sectional area of the probe brings the material to failure.
For the two simulations, since the Norton equation only models the secondary creep, an initial σ ¼ 635 MPa
is set. The decrease of equivalent stress is due to a decrease in effective Young’s modulus, in turn due to damage
accumulation, as expected (see Equation 10). Both the stress decrease and the damage increase are represented
with good accuracy when compared to the experimental data, except from the areas at the very beginning and at
the end of the simulations, where the stress decreases relatively fast. Those are the areas corresponding to primary
and tertiary creep. It must be remembered that this UPF is developed to model secondary creep, hence, the high
strain rates typical of primary and tertiary creep are out of the scope of this paper.
The temperature-dependent parameters allow the user to span over a wider range of (in particular to higher)
temperatures, compared to those parameters of (Liu et al., 2015), which are optimized for 750 K only. As an
example, creep strain and damage over time at three different temperatures and for an initial stress of 650 MPa
are shown in Figure 5. It can be seen that temperature has an enormous impact on creep and on creep strain, as
expected. Higher temperatures promote a much faster creep deformation. This is depicted well from the picture
on the right. On the left, the damage shows again a non-linear increasing trend with time. Since the
Chaboche-Lemaitre damage Dcr depends on equivalent stress only (at least directly, see Equations 8 and 9), the
damage evolution is, in the first phase, very similar for all the temperatures. Over time, the damage increases
more at higher temperatures, both due to higher creep and lower Young’s modulus with temperature.

Conclusions

In this paper, a method for the determination of temperature-dependent Norton coefficients for Inconel 718 was
shown. The procedure takes rupture creep test data as input. The so obtained coefficients are then implemented

Figure 4. Comparison between experimental data, the use of the constant parameters from (Liu et al., 2015) and the
new, temperature-dependent parameters, at 750 K and initial σ = 700 Mpa.

J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 173

Cirigliano et al. | Creep-damage modelling for lifetime prediction https://www.journalssystem.com/jgpps/163088,0,2.html

Figure 5. Damage and creep strain at three different temperatures and initial σ = 650 Mpa.

in a User Programmable Feature inside the ANSYS APDL environment, enabling the user to write a custom
mechanical model. In this study, secondary creep together with material damage and Young’s modulus reduction
were combined.
The results show excellent agreement with the experimental data, strictly concerning the secondary creep
portion of the tests. The UPF allows the simulation of creep damage in a wide range of temperatures, in particu-
lar up to 760°C, which are close to the temperatures typical of a gas turbine combustion chamber. This method
has the potential to be extended to even higher temperatures, if creep tests are provided. The range 760–1,000°C
is currently under development.
This study constitutes the foundation to explore the life of Inconel 718 structures exposed at high temperature
creep, where stress relaxation, damage and creep are taken into account for a realistic representation of this
coupled problem. The direct application of this USERMAT subroutine in ANSYS to MGT combustion cham-
bers is currently under development and will be published on basis of a separate study.

Nomenclature

Latin symbols
[D] elasticity matrix
Dcr creep damage
E effective Young’s modulus
E0 undamaged Young’s modulus
MGT Micro Gas Turbine
tr rupture time
tss time at III creep onset
UPF User Programmable Feature

Greek symbols
ϵ strain tensor
ϵeq equivalent total strain
ϵcreq equivalent creep strain
ϵcreq,ss eq. creep strain at III creep onset
ϵeleq equivalent elastic strain
ϵeqpl equivalent plastic strain
ϵth
eq equivalent thermal strain
σ stress tensor
σ eq equivalent stress

J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 174

Cirigliano et al. | Creep-damage modelling for lifetime prediction https://www.journalssystem.com/jgpps/163088,0,2.html

Competing interests

Daniele Cirigliano declares that he has no conflict of interest. Herol Lawerence D’Souza declares that he has no conflict of
interest. Felix Grimm declares that he has no conflict of interest. Peter Kutne declares that he has no conflict of interest.
Manfred Aigner declares that he has no conflict of interest.

References

Abe F. (2014). Development of creep-resistant steels and alloys for use in power plants. In: Structural Alloys for Power Plants, Woodhead
Publishing Series in Energy, edited by Shirzadi A. and Jackson S. Cambridge, UK: Woodhead Publishing. 250–293. https://doi.org/
10.1533/9780857097552.2.250.
Alain I. (1998). The correlation between the power-law coefficients in creep: The temperature dependence. Journal of Materials Science.
33: 3201–3206. https://doi.org/10.1023/A:1004368511595.
Barbosa A. L., Taylor N. G., Ashby M. F., Dyson B. F., and Mclean M. (1988). A model based computer analysis of creep data
(cripsen): Applications to nickel-base superalloys. Superalloys. 88: 683–692.
Beith R. (2011). Small and micro combined heat and power (chp) systems. Woodhead Publishing Series in Energy. 42–69.
Betten J. (2008). Creep Mechanics: Springer Science & Business Media, Berlin, Heidelberg, Germany.
Brinkman C., Booker M., and Ding J. (1991). Creep and creep-rupture behavior of alloy718.
Caliari F. R., Candioto K. C. G., Couto A. A., Nunes C. Â., and Reis D. A. P. (2016). Effect of double aging heat treatment on the
short-term creep behavior of the inconel 718. Journal of Materials Engineering and Performance. 25 (6): 2307–2317. https://doi.org/10.
1007/s11665-016-2051-2.
Carey J., Sargent P., and Jones D. (1990). Deformation mechanism map for in738lc superalloy. Journal of Materials Science Letters. 9:
572–575. https://doi.org/doi:10.1007/BF00725881.
Chaboche J. L. and Rousselier G. (1983). On the plastic and viscoplastic constitutive equations—part i: Rules developed with internal
variable concept. Journal of Pressure Vessel Technology. 105 (2): 153–158. https://doi.org/10.1115/1.3264257.
Chandrupatla T. R. and Belegundu A. D. (2012). Introduction to Finite Elements in Engineering. 4th edn. Pearson: Prentice Hall, Upper
Saddle River, New Jersey 07458.
Cieśla M., Mańka M., Binczyk F., and Gradoń P. (2016). Creep behaviour of modified mar-247 superalloy. Archives of Metallurgy and
Materials. 61: 701–704.
Cirigliano D., et al. (2020). Thermo-structural analysis of a micro gas turbine jet- and recirculation- stabilized combustion chamber.
Proceedings of ASME Turbo Expo 2020 10B(GT2020-16285).
Das Y., Fernandez-Caballero A., Elmukashfi E., Jazaeri H., Forsey A., et al. (2022). Stress driven creep deformation and cavitation
damage in pure copper. Materials Science and Engineering: A. 833: 142543. https://www.sciencedirect.com/science/article/pii/
S0921509321018037. https://doi.org/10.1016/j.msea.2021.142543.
Enagi I. I., Al-attab K., and Zainal Z. A. (2017). Combustion chamber design and performance for micro gas turbine application.
Fuel Processing Technology. 166 (2): 258–268. https://doi.org/10.1016/j.fuproc.2017.05.037.
Frost H. and Ashby M. (1982) Deformation-Mechanism Maps. Pergamon Press: Oxford, UK.
Golan O., Arbel A., Eliezer D., and Moreno D. (1996). The applicability of norton’s creep power law and its modified version to a single-
crystal superalloy type cmsx-2. Materials Science and Engineering: A. 216 (1): 125–130. https://www.sciencedirect.com/science/article/
pii/0921509396104007. https://doi.org/10.1016/0921-5093(96)10400-7.
Kachanov L. M. (1999). Rupture time under creep conditions. International Journal of Fracture. 97 (1–4): 11–18. https://doi.org/10.
1023/A:1018671022008.
Kim D.-H., Kim J.-H., Sa J.-W., Lee Y.-S., Park C.-K., and Moon S.-I. (2008). Stress rupture characteristics of inconel 718 alloy for
ramjet combustor. Materials Science and Engineering: A. 483-484: 262–265. 14th International Conference on the Strength of
Materials. https://doi.org/10.1016/j .msea.2006.12.159.
Lemaitre J. (1985). A continuous damage mechanics model for ductile fracture. Journal of Engineering Materials and Technology. 107 (1):
83–89. https://doi.org/10.1115/1.3225775.
Lemaitre J. and Chaboche J.-L. (1978). Aspect phénoménologique de la rupture par endommagement. Journal de Mecanique Appliquee.
2 (3): 317–365.
Li L., Song Y., and Huang X. (2010). Analysis on creep and creep life of rotors connecting bolts. Technology & New Process. 6: 75–79.
Lin G. (1999). Ansys user material subroutine usermat. Canonsburg, PA: ANSYS.
Liu D., Li H., and Liu Y. (2015). Numerical simulation of creep damage and life prediction of superalloy turbine blade. Mathematical
Problems in Engineering. 2015: 732502, 10 pages. https://doi.org/10.1155/2015/732502.
McLean M., Ghosh R., Curtis R., Basu-Conlin U., and Winstone M. (1992). Anisotropy of high temperature deformation of single
crystal superalloys – constitutive law, modelling and validation, The Minerals. Metals and Materials Society Superalloys. 1992: 609–618.
Ofualagba G. (2012). The modeling and simulation of a microturbine generation system. International Journal of Scientific & Engineering
Research. 2 (2): 1–7. https://doi.org/10.15373/22778179/FEB2013/1
Palmer C. (2021). Eworldwide gas turbine market report 2022. Turbomachinery Magazine, Handbook 2022. https://www.
turbomachinerymag.com/view/worldwide-gas-turbine-market-report-2022
Rabotnov Y. N., Leckie F. A., and Prager W. (1970). Creep problems in structural members. Journal of Applied Mechanics. 37 (1):
249–249. https://doi.org/10.1115/1.3408479.
Sugahara T., Martinolli K., Reis D. A., de Moura Neto C., Couto A. A., et al. (2012). Creep behavior of the inconel 718 superalloy. In:
Defect and Diffusion Forum. Vol. 326–328. edited by Andreas Öchsner, Graeme E. Murch, Ali Shokuhfar and João M.P.Q. Delgado.
Switzerland: Trans Tech Publications. 509–514. https://doi.org/10.4028/www.scientific.net/DDF.326-328.509.

J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 175

Cirigliano et al. | Creep-damage modelling for lifetime prediction https://www.journalssystem.com/jgpps/163088,0,2.html

Sun L., et al. (2012). Numerical simulation of a complete gas turbine engine with wet compression. Journal of Engineering for Gas
Turbines and Power. 135 (1): 012002. https://doi.org/10.1115/1.4007366.
VDM Metals (2020). VDM Alloy 718, Material Data Sheet No. 4127, Rev. 02. https://www.vdm-metals.com/fileadmin/user_upload/
Downloads/Data_Sheets/Data_Sheet_VDM_Alloy_718.pdf
Zhang L. (1995). Research of fatigue failure on high-temperature parts based on damage mechanics. PhD thesis, Beijing University of
Aeronautics and Astronautics.

J. Glob. Power Propuls. Soc. | 2023 | 7: 166–176 | https://doi.org/10.33737/jgpps/163088 176