See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/357932834

Application of Deep Neural Network to Predict the High-Cycle Fatigue Life of

AISI 1045 Steel Coated by Industrial Coatings

Article in Journal of Marine Science and Engineering · January 2022

DOI: 10.3390/jmse10020128

CITATIONS READS

4 358

5 authors, including:

Erfan Maleki Okan Unal

Politecnico di Milano Karabuk University
74 PUBLICATIONS 1,235 CITATIONS 45 PUBLICATIONS 1,103 CITATIONS

SEE PROFILE SEE PROFILE

Kazem reza kashyzadeh

Peoples' Friendship University of Russia
117 PUBLICATIONS 642 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

ANN modelling of welding process View project

Stability of Structures under their Own Weight View project

All content following this page was uploaded by Erfan Maleki on 19 January 2022.

The user has requested enhancement of the downloaded file.

 Journal of
Marine Science
and Engineering

Article
Application of Deep Neural Network to Predict the High-Cycle
Fatigue Life of AISI 1045 Steel Coated by Industrial Coatings
Erfan Maleki 1 , Okan Unal 2,3 , Seyed Mahmoud Seyedi Sahebari 4 , Kazem Reza Kashyzadeh 5, *
and Igor Danilov 5

1 Mechanical Engineering Department, Politecnico di Milano, 20156 Milan, Italy; erfan.maleki@polimi.it

2 Mechanical Engineering Department, Karabuk University, Karabuk 78050, Turkey; unalokan78@gmail.com
3 Modern Surface Engineering Laboratory, Karabuk University, Karabuk 78050, Turkey
4 Department of Mechanical and Manufacturing Engineering, Ontario Tech University,
Oshawa, ON L1G 0C5, Canada; Seyedmahmoud.seyedisahebari@ontariotechu.net
5 Department of Transport, Academy of Engineering, Peoples’ Friendship University of Russia
(RUDN University), 6 Miklukho-Maklaya Street, 117198 Moscow, Russia; danilov-ik@rudn.ru
* Correspondence: reza-kashi-zade-ka@rudn.ru or kazem.kashyzadeh@gmail.com

Abstract: In this study, deep learning approach was utilized for fatigue behavior prediction, analysis,
and optimization of the coated AISI 1045 mild carbon steel with galvanization, hardened chromium,
and nickel materials with different thicknesses of 13 and 19 µm were used for coatings and afterward
fatigue behavior of related specimens were achieved via rotating bending fatigue test. Experimental
results revealed fatigue life improvement up to 60% after applying galvanization coat on untreated
material. Obtained experimental data were used for developing a Deep Neural Network (DNN)
modelling and accuracy of more than 99%.was achieved. Predicted results have a fine agreement



with experiments. In addition, parametric analysis was carried out for optimization which indicated


that coating thickness of 10–15 µm had the highest effects on fatigue life improvement.
Citation: Maleki, E.; Unal, O.; Seyedi
Sahebari, S.M.; Reza Kashyzadeh, K.;
Danilov, I. Application of Deep Keywords: fatigue life; coating; deep neural network; optimization; prediction
Neural Network to Predict the
High-Cycle Fatigue Life of AISI 1045
Steel Coated by Industrial Coatings. J.
Mar. Sci. Eng. 2022, 10, 128. https:// 1. Introduction
doi.org/10.3390/jmse10020128 Most of the used components in some major industries such as marine and offshore
Academic Editor: José A.F.O. industries which primarily comprises the offshore oil and gas and the offshore wind energy
Correia industries are affected by the repeated applied loadings that make fatigue failure. Moreover,
the corrosion phenomenon is an unavoidable factor for the failure of structures in the vicin-
Received: 13 December 2021
ity of water, especially seawater due to its salinity [1,2]. However, combination of fatigue
Accepted: 16 January 2022
and corrosion simultaneously can afford faster failure and generally, amounts of damages
Published: 19 January 2022
caused by corrosion fatigue are higher [3]. Therefore, considering effective approaches to
Publisher’s Note: MDPI stays neutral improve the corrosion resistance and fatigue have critical roles in this area [4,5]. To im-
with regard to jurisdictional claims in prove the corrosion resistance of conventionally manufactured materials (specially steels),
published maps and institutional affil- varieties of treatments can be performed such as shot peening [6–14], carburizing [15–17],
iations. and plating [18–22]. In addition, rather than conventional materials, recently different post-
processing methods were also applied on the additively manufactured metallic materials
as well to improve the mechanical properties [23–32].
Plating due its low cost, mostly used by providing a protective layer of corrosion-
Copyright: © 2022 by the authors.
resistant material in the outer surface of the components that has contact with the corrosive
Licensee MDPI, Basel, Switzerland.
environments. In electroplating as one of widely used methods of plating, an ionic metal
This article is an open access article
is prepared with electrons to make a non-ionic coating on a regraded substrate by using
distributed under the terms and
conditions of the Creative Commons
chemical solution [33]. Because of initiation of micro-cracks in the coat layer by apply load-
Attribution (CC BY) license (https://
ing and afterward it’s interpenetration to the substrate, fatigue life of the coated material,
creativecommons.org/licenses/by/
specifically steels, might be reduced although the corrosion resistance was improved [34]. It
4.0/). is approved by many researches that different coatings such as nickel, hardened chromium,

J. Mar. Sci. Eng. 2022, 10, 128. https://doi.org/10.3390/jmse10020128 https://www.mdpi.com/journal/jmse

J. Mar. Sci. Eng. 2022, 10, 128 2 of 12

warm galvanization, and titanium can be improved the corrosion resistance [35–38] but few
works were carried out to study the effects of coatings on fatigue behavior of the mantellic
materials [39,40].
On the other hand, recently artificial intelligence-based approaches such as neural
networks (NNs) are extensively implemented to analyze and optimize multi-objective
and complex problems [41–44]; as well as their wide applications in fatigue life estima-
tion [45–50]. Based on the available data in the literature different types of NNs were
utilized for fatigue behavior prediction and analyses [51–57]. It should be mentioned that
there are also different methods for fatigue life assessment such as XFEM [58,59] and phase
field method [60,61]. By applying NN modelling or other approaches such as XFEM rather
than fatigue life prediction, other types of analyses such as parametric and sensitivity
analyses can be carried out by generating corresponded model function or further finite
element analyses. However, by using other common methods such S-N or Manson-coffin
curves only life predictions can be obtained.
In general, a neural network consists of three major layers: input, hidden, and output
layers [62]. As the primary generation of artificial neural networks, shallow neural networks
(SNNs) that have 1 or 2 hidden layers were mostly used in simulations of the ill-defined
problems [63]. However, it is approved by developing Deep Neural Network (DNN) that
has more hidden layers (more than 2), higher accuracy in the predicted results can be
obtained with same or smaller data set [64,65]. The NNs are the most frequently employed
machine learning approach for fatigue life prediction in the last decade [66]. A major reason
for the popularity of the application of NNs on fatigue behavior analysis is that NNs are
mostly utilized for universal function approximation. NNs have significant self-learning
properties, adaptivity, fault tolerance, nonlinearity, and flexibility of nonlinear mapping [67].
Therefore, as phenomenon of fatigue is highly sensitive to various parameters including
material, loading, temperature, humidity, vibration etc., NNs can be applied to all fatigue
analysis. On the other hand, as reported in several recent studies [68,69], applications
of DNN based modelling of fatigue behavior attracts lots of attention lately due to their
higher performance and accuracy compared to the SNNs [70,71]. In addition, it should
be mentioned that applications of NN-based models for fatigue behavior analyses are
investigated in the comprehensive review study carried out by Chen and Liu [72].
Following the authors previous studies [47,73] which surveyed the applications of
SNNs on fatigue life prediction of the coated carbon steel, in this study, influences of
three different coatings of nickel, hardened chromium, and warm galvanization on fatigue
behavior of the AISI 1045 mild carbon steel which has extensive applications in different
industries due its cost and properties, were investigated experimentally and a DNN model
was developed for prediction of fatigue life of the coated steel with different coatings
thicknesses and amplitude stresses.

2. Experimental Study
Fatigue test specimens (as shown in Figure 1a) based on the BS3518 standard [59]
were prepared using AISI 1045 carbon steel with chemical composition of 0.45 C, 0.8 Mn,
0.04 P, 0.05 S and Bal. Fe (% weight). The used material had yield and ultimate stresses
of 475 and 515 MPa respectively. AISI 1045 is widely employed in different engineering
components such as tools, railway wheels, and offshore structures due to its good properties
and low cost [60–62]. Three sets of specimens were considered for each desired coating,
including one control specimen without coating (as-received) and two others with different
coating thickness. Fatigue tests were performed via rotating cantilever beam equipment (as
presented in Figure 1b) with stress ratio of R=-1, frequency of 58 Hz, room temperature,
50% humidity, and pure bending load with stress amplitude range of 330–450 MPa in all
considered sets of specimens. To obtain S-N curves, in each regarded sets 13 samples were
utilized. Three coatings of nickel, hardened chromium, and warm galvanization, with
thicknesses of 13 and 19 µm were plated on the steel substrate via electroplating in same
 MPa in all considered sets of specimens. To obtain S-N curves, in each regarded sets 13
samples were utilized. Three coatings of nickel, hardened chromium, and warm galvani-
J. Mar. Sci. Eng. 2022, 10, 128 3 of 12
zation, with thicknesses of 13 and 19 µm were plated on the steel substrate via electroplat-
ing in same conditions of humidity, temperature, etc. Images of Scanning Electron Micros-
copy (SEM) of the coated specimens with considered thicknesses are presented in Figure
conditions of humidity, temperature, etc. Images of Scanning Electron Microscopy (SEM)
1c. of the coated specimens with considered thicknesses are presented in Figure 1c.

Figure 1. (a) Fatigue test specimen and (b) Schematic illustration of cantilever rotating bending
Figure 1. (a) Fatigue test specimen and (b) Schematic illustration of cantilever rotating bending fa-
fatigue equipment used in this study. (c) Observed coating thicknesses by SEM, from up to bottom 13
tigue equipment used in this study. (c) Observed coating thicknesses by SEM, from up to bottom 13
and 19 µm, respectively.
and 19 µm, respectively.
Obtained S-N curves of the coated AISI 1045 steel with different coatings as results of
Obtained S-N curves
fatigue testsofisthe coatedinAISI
presented 1045
Figure steel with
2 [41,63]. different
The results coatings
of fatigue as resultsreveal that
experiments
of fatigue tests isfatigue
presented in Figure
life improves 2 [41,63].
in the The results
coated specimens by of fatigue
warm experiments
galvanization reveal
in both thicknesses of
that fatigue life improves in the coated specimens by warm galvanization in both thick- coating
13 and 19 µm. Also, in the coated specimen by nickel with 13 µm thickness. Nickel
nesses of 13 andwith19
19 µm. thickness
µmAlso, in theapproximately
coated specimenhas nobyeffects on variation
nickel with 13 µm of fatigue life. However,
thickness.
fatigue life of coated specimens with hardened chromium in both considered thicknesses is
Nickel coating with19 µm thickness approximately has no effects on variation of fatigue
reduced. Considering stress level of 330 MPa, the results reveal that after applying coatings
life. However, fatigue life of coated specimens with hardened chromium in both consid-
on the untreated material, the fatigue behavior can be improved up to 40–60 and 5–50% in
ered thicknesses theis reduced. Considering
coated specimens stressgalvanization
with warm level of 330andMPa, the respectively.
nickel, results reveal that
after applying coatings on the untreated material, the fatigue behavior can be improved
up to 40–60 and 5–50% in the coated specimens with warm galvanization and nickel, re-
spectively.
J.J.Mar.
Mar.Sci.
Sci.Eng. 2022,10,
Eng.2022, 9, x128
FOR PEER REVIEW 44 of
of 12
12

Figure2.2.S-N
Figure S-Ncurves
curvesof
ofthe
thecoated
coatedAISI
AISI1045
1045steel
steelwith
withdifferent
differentcoatings
coatingsand
andthe
thecorrespondence
correspondence
thicknesses.
thicknesses.

3.3.Deep
DeepNeural
NeuralNetwork
Network
In
In this study,uncoated
this study, uncoatedAISI AISI1045
1045steel steelspecimen
specimenwas was considered
considered as as control
control specimen;
specimen;
thus,
thus, the coated specimens with different thicknesses and materials revealedthe
the coated specimens with different thicknesses and materials revealed theeffects
effectsofof
applied post-treatment under fatigue loading. Therefore, two
applied post-treatment under fatigue loading. Therefore, two parameters were considered parameters were considered
for inputs of the NNs modelling; one is related to the material and the other one is related
for inputs of the NNs modelling; one is related to the material and the other one is related
to the fatigue test. The difference of the treated materials rather than the coatings material
to the fatigue test. The difference of the treated materials rather than the coatings material
is the thickness of each coating which considered as one of the inputs. Also, a major
is the thickness of each coating which considered as one of the inputs. Also, a major pa-
parameter of stress amplitude which is related to the conditions of fatigue loading was also
rameter of stress amplitude which is related to the conditions of fatigue loading was also
regarded as inputs. Strong correlation of these two parameters as the main variables of this
regarded as inputs. Strong correlation of these two parameters as the main variables of
study with fatigue behavior can be clearly seen in the obtained S-N curves of the different
this study with fatigue behavior can be clearly seen in the obtained S-N curves of the dif-
sets of specimens. As it mentioned, fatigue behavior was investigated at 13 different stress
ferent sets of specimens. As it mentioned, fatigue behavior was investigated at 13 different
levels and with considering of coating three thicknesses of 0 (as-received specimens), 13,
stress levels and with considering of coating three thicknesses of 0 (as-received speci-
and 19 µm for coatings, there exist 39 samples for each of the regarded coatings materials
mens), 13, and 19 µm for coatings, there exist 39 samples for each of the regarded coatings
that 30 of them were used for training and the remained 9 samples, which not employed
materials that 30 of them were used for training and the remained 9 samples, which not
during training step, were used for network testing. Overall, 77 and 23% of obtained
employed during
experimental data training
was used step,
for were
networks used training
for network and testing. Overall, 77respectively.
testing processes, and 23% of
Employed methodology in this study is presented in Figure 3a (similar methodologyrespec-
obtained experimental data was used for networks training and testing processes, to the
tively.
one usedEmployed
by Maleki methodology
et al. [63,65]).inDifferent
this study is presented
SNNs and DNNs in Figure 3a (similar methodol-
by trial-and-error approach
ogy to
were the one used
developed; by Maleki
Figure 3b depictset al.the
[63,65]). Different
architecture of SNNs
the DNN and with
DNNs by hidden
four trial-and-error
layers
approach were developed; Figure 3b depicts the architecture
that w, b and f represent the weight matrixes, bias vectors, and transfer function of the DNN with fourinhidden
each
layers that
related layer,w,respectively.
b and f represent theknown
It is well weightthat matrixes,
number bias vectors, as
of neurons and transfer function
computational nodes in
each related layer, respectively. It is well known that number
of neural networks, have critical role on the efficiency of applied models [74]. Accuracy of neurons as computational
nodes
of of neuralresults
the predicted networks, have critical
of networks which role on the efficiency
determined of applied
via correlation models(R2)
coefficient [74].was
Ac-
curacy ofasthe
gathered thepredicted results ofofnetworks
factor of efficiency each developed which determined
network. Value via of
correlation coefficient
(R2) was calculated
(R2)
as was gathered
follows [75]: as the factor of efficiency of each developed network. Value of (R2) was
calculated as follows [75]: n
∑ ( f EXP,i − FEXP )( f ANN,i − FANN )
R 2 = s i =1 n
 f EXP,i  FEXP  f ANN ,i  FANN  (1)
n   2 2

∑R ( f EXP,i − FEXP ) ( f ANN,i − FANN )
2
n
i 1
(1)
i =1
  f  FEXP   f ANN ,i  FANN  
2 2
EXP,i
i 1
where, n is the number of fed samples, fEXP and fANN represent the experimental and
predicted
where, n values, respectively.
is the number of fedThe
samples, EXPfEXP
valuesfof fANNfANN
andand were determined
represent as follows
the experimental and [76]:
pre-
dicted values, respectively. The values of fEXP and fANN were determined as follows [76]:
1
FEXP =
n ∑ in=1 f EXP,i (2)
= ∑ , (2)
 J. Mar. Sci. Eng. 2022, 9, x FOR PEER REVIEW 5 of 12
J. Mar. Sci. Eng. 2022, 10, 128 5 of 12

1 ∑ n f ANN,i,
FANN == ∑ i =1
(3)
(3)
n

Figure 3.
Figure 3. (a)
(a) Employed
Employed methodology
methodology and
and (b)
(b) Schematic
Schematicof
ofaa DNN
DNN with
withfour
fourhidden
hiddenlayers.
layers.

4.
4. Results
Results and
and Discussion
Discussion
Accuracy
Accuracy of of the
the results
results of
of several
several developed
developed SNNs
SNNs with
with 11 and
and 22 hidden
hidden layers
layers with
with
different number of neurons in each layer of them for nickel coated specimens
different number of neurons in each layer of them for nickel coated specimens is compared is compared
in
inFigure
Figure4a.4a.ItItcan
canbebeseen
seenthat
thatininthe
theSNNs,
SNNs,ininmost ofof
most the cases
the byby
cases increasing
increasingthethe
number
numberof
neurons, the efficiency of the network is become higher. Figure 4b illustrates
of neurons, the efficiency of the network is become higher. Figure 4b illustrates a compar- a comparison
between the accuracy
ison between of theof
the accuracy developed SNNs SNNs
the developed and DNNs with highest
and DNNs performance
with highest for all
performance
of the coating materials. From this figure, by implementation of a DNN with
for all of the coating materials. From this figure, by implementation of a DNN with struc- structure of
2-36-18-18-9-1, learning rate of 0.185, and Logaritmic Sigmod tarnsfer function, accuracy
of the predicted results of fatigue life for all of the coated specimens for each testing and
ci. Eng. 2022,J. 9,
Mar. Sci. PEER
x FOR Eng. 2022, 9, x FOR PEER REVIEW
REVIEW 6 of 12 6 of 12

J. Mar. Sci. Eng. 2022, 10, 128 6 of 12

ture oflearning
ture of 2-36-18-18-9-1, 2-36-18-18-9-1, learning
rate of 0.185, andrate of 0.185, Sigmod
Logaritmic and Logaritmic
tarnsferSigmod
function, tarnsfer
ac- function, ac-
curacy of
curacy of the predicted the predicted
results of fatigueresults
life forofallfatigue life for all
of the coated of the coated
specimens specimens
for each testing for each testing
training
andcan
and training steps steps
training
be can be
steps
reached toreached
can to approximately
be reached
approximately to0.99, 0.99,
approximately
which which is quiteisacceptable.
0.99,acceptable.
is quite which quite
Com- Comparison
acceptable. Com-
of the
parisonpredicted
of the and experimentally
predicted and values
experimentally of fatigue
values oflife for
fatigue
parison of the predicted and experimentally values of fatigue life for training and testing training
life for and testing
training andsteps is
testing
revealed
steps is revealedsteps in Figure
is revealed
in Figure 5. 5.
in Figure 5.

Figure4.4.of
Figure 4. (a) Performance
Figure (a)the
(a) Performance ofthe
developed of
Performance SNNsthedeveloped
developed SNNswith
with considering
SNNs withnumber
the considering thenumber
of neurons
considering the number ofneurons
in each
of neuronsin
ineach
each
layerand
layer and (b) Comparison
layer andof (b)
the
(b) Comparison of
accuracy ofof
Comparison theaccuracy
accuracy
developed
the SNNsofdeveloped
of developed
and SNNsand
DNNs.SNNs andDNNs.
DNNs.

Figure 5. Cont.
ci. Eng. 2022, 9, x FOR PEER REVIEW 7 of 12
J. Mar. Sci. Eng. 2022, 10, 128 7 of 12

Figure 5. Comparison of the predicted and experimental values of fatigue life for training and
Figure 5. Comparison of the predicted and experimental values of fatigue life for training and test-
testing steps considering the coated materials, including (a,b) Nickel, (c,d) Hardened chromium, and
ing steps considering the coated materials, including (a,b) Nickel, (c,d) Hardened chromium, and
(e,f) Warm galvanization.
(e,f) Warm galvanization.

To clearly observe the differences between the experimental and predicted results
To clearly observe the differences between the experimental and predicted results
through developing DNNs, combined results of training and testing processes for all
J. Mar. Sci. Eng. 2022, 9,through developing
x FOR PEER REVIEW DNNs, combined results of training and testing processes for all coat-
coating materials are shown in Figure 6. The results reveal highly acceptable and8 very
of 12 low
ing materials are shown in Figure 6. The results reveal highly acceptable and very low
data scattering in the obtained results.
data scattering in the obtained results.
Relevant model function of a DNN with 6 layers can be determined by the following
equation:
O (o(1)) = a6 = f6(w6f5(w5f4(w4f3 (w3f2(w2f1(w1i + b1) + b2) + b3) + b4) + b5) + b6) (4)
where a1 to a5 represent the outputs of the first to fifth layer, respectively; a6 is the sixth
layer output which is equal to the function O (o(1)). Function of o(1) expresses fatigue life.
Although in the used data set which fed to the network, there were only 3 and 13
values for coatings thicknesses and stress amplitude, respectively, by using generated
model function, parametric analyses can be carried out for the whole interval of the input
parameters as shown in Figure 7. According to the DNN results, AISI 1045 steel coated by
nickel and warm galvanization with coating thicknesses of 10 to 15 µm have the highest
effects on fatigue life improvement and the thickness values lower and higher than the
mentioned interval have not any considerable effects on the variation of the fatigue life.
However, in the hardened chromium coated material, fatigue life decreases by increasing
of coating thickness.

Figure
Figure 6. Combined
6. Combined results
results of training
of training and testing
and testing processes
processes forcoating
for all all coating materials
materials comparing
comparing the the
experimental and predicted values.
experimental and predicted values.
Mar. Sci. Eng. 2022, 9, x FOR PEER REVIEW 8 of 12

J. Mar. Sci. Eng. 2022, 10, 128 8 of 12

Relevant model function of a DNN with 6 layers can be determined by the following
equation:

O (o(1)) = a6 = f6 (w6 f5 (w5 f4 (w4 f3 (w3 f2 (w2 f1 (w1 i + b1 ) + b2 ) + b3 ) + b4 ) + b5 ) + b6 ) (4)

where a1 to a5 represent the outputs of the first to fifth layer, respectively; a6 is the sixth
layer output which is equal to the function O (o(1)). Function of o(1) expresses fatigue life.
Although in the used data set which fed to the network, there were only 3 and
13 values for coatings thicknesses and stress amplitude, respectively, by using generated
model function, parametric analyses can be carried out for the whole interval of the input
parameters as shown in Figure 7. According to the DNN results, AISI 1045 steel coated by
nickel and warm galvanization with coating thicknesses of 10 to 15 µm have the highest
effects on fatigue life improvement and the thickness values lower and higher than the
mentioned interval have not any considerable effects on the variation of the fatigue life.
However,
Figure 6. Combined in the
results hardened
of training andchromium coated
testing processes formaterial, fatigue
all coating lifecomparing
materials decreasesthe
by increasing
experimentalofand
coating thickness.
predicted values.

Figure 7. 2DFigure
contours7. 2D contours of
of influences ofinfluences of coating
coating thickness andthickness and stressonamplitude
stress amplitude onoffatigue
fatigue life AISI life of AISI
1045 steel for1045 steelmaterials
coating for coating
of materials
(a) Nickel,of(b)
(a)Warm
Nickel, (b) Warm galvanization,
galvanization, and (c)chromium.
and (c) Hardened Hardened chromium.

5. Conclusions
5. Conclusions
In the present study, influences of different conventional industrial coatings, including
In the present study, influences of different conventional industrial coatings, includ-
nickel, hardened chromium, and warm galvanization which are usually used for the
ing nickel, hardened chromium, and warm galvanization which are usually used for the
improvements of the corrosion resistance, were investigated on fatigue life of the coated
improvements of the corrosion resistance, were investigated on fatigue life of the coated
AISI 1045 carbon steel. Experimental results revealed that coatings of nickel and warm
AISI 1045 carbon steel. Experimental results revealed that coatings of nickel and warm
galvanization with a thickness of 13 µm improve the fatigue life. Moreover, using hardened
galvanization with a thickness of 13 µm improve the fatigue life. Moreover, using hard-
chromium coating leads to reduce the fatigue life of the AISI 1045 steel. Afterward, the
ened chromium coating leads to reduce the fatigue life of the AISI 1045 steel. Afterward,
process was modeled and analyzed via SNNs and DNNs that based on the obtained results
the process was modeled and analyzed via SNNs and DNNs that based on the obtained
of R2, DNN has higher performance rather than the SNN. DNN results illustrate that
results of R2, DNN has higher performance rather than the SNN. DNN results illustrate
highest fatigue life of coated steel with nickel and warm galvanization can be obtained
that highest fatigue life of coated steel with nickel and warm galvanization can be ob-
in the thicknesses of 10-15 µm and also by employing hardened chromium as coating, by
increasing of the thickness the fatigue life decreases. Eventually, it can be concluded that
J. Mar. Sci. Eng. 2022, 10, 128 9 of 12

DNN that has more efficiency than SNN can be used for prediction and analysis of the
fatigue life of coated materials finely.

Author Contributions: Conceptualization, E.M. and K.R.K.; methodology, E.M., O.U. and S.M.S.S.;
software, E.M. and O.U.; validation, E.M., O.U. and K.R.K.; formal analysis, E.M., O.U. and S.M.S.S.;
investigation, E.M., O.U., K.R.K., S.M.S.S. and I.D.; resources, E.M., O.U., K.R.K., S.M.S.S. and I.D.;
data curation, E.M. and K.R.K.; writing—original draft preparation, E.M., O.U. and S.M.S.S.; writing—
review and editing, E.M., K.R.K. and I.D.; visualization, E.M.; supervision, E.M., O.U. and K.R.K.;
project administration, K.R.K.; funding acquisition, K.R.K. and I.D. All authors have read and agreed
to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: The data that support the findings of this study are available from the
corresponding author upon reasonable request.
Acknowledgments: This paper has been supported by the RUDN University Strategic Academic
Leadership Program.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Amiri, N.; Shaterabadi, M.; Reza Kashyzadeh, K.; Chizari, M. A Comprehensive Review on Design, Monitoring, and Failure in
Fixed Offshore Platforms. J. Mar. Sci. Eng. 2021, 9, 1349. [CrossRef]
2. Abdollahnia, H.; Hadi, M.; Elizei, A.; Kashyzadeh, K.R. Multiaxial Fatigue Life Assessment of Integral Concrete Bridge with a
Real-Scale and Complicated Geometry Due to the Simultaneous Effects of Temperature Variations and Sea Waves Clash. J. Mar.
Sci. Eng. 2021, 9, 1433. [CrossRef]
3. Makhlouf, A.S.H. Handbook of Smart Coatings for Materials Protection; Woodhead Publishing: Cambridge, UK, 2014; ISBN 9780857096883.
4. Maleki, E.; Bagherifard, S.; Bandini, M.; Guagliano, M. Surface post-treatments for metal additive manufacturing: Progress,
challenges, and opportunities. Addit. Manuf. 2021, 37, 101619. [CrossRef]
5. Maleki, E.; Maleki, N.; Fattahi, A.; Unal, O.; Guagliano, M.; Bagherifard, S. Mechanical characterization and interfacial enzymatic
activity of AISI 316L stainless steel after surface nanocrystallization. Surf. Coat. Technol. 2021, 405, 126729. [CrossRef]
6. Unal, O.; Maleki, E.; Varol, R. Effect of severe shot peening and ultra-low temperature plasma nitriding on Ti-6Al-4V alloy.
Vacuum 2018, 150, 69–78. [CrossRef]
7. Maleki, E.; Unal, O.; Reza Kashyzadeh, K. Efficiency Analysis of Shot Peening Parameters on Variations of Hardness, Grain Size
and Residual Stress via Taguchi Approach. Met. Mater. Int. 2019, 25, 1436–1447. [CrossRef]
8. Maleki, E.; Unal, O.; Kashyzadeh, K.R.K.R. Effects of conventional, severe, over, and re-shot peening processes on the fatigue
behavior of mild carbon steel. Surf. Coat. Technol. 2018, 344, 62–74. [CrossRef]
9. Amanov, A.; Karimbaev, R.; Maleki, E.; Unal, O.; Pyun, Y.S.Y.-S.; Amanov, T. Effect of combined shot peening and ultrasonic
nanocrystal surface modification processes on the fatigue performance of AISI 304. Surf. Coat. Technol. 2019, 358, 695–705.
[CrossRef]
10. Hashemi, B.; Rezaee Yazdi, M.; Azar, V. The wear and corrosion resistance of shot peened-nitrided 316L austenitic stainless steel.
Mater. Des. 2011, 32, 3287–3292. [CrossRef]
11. Maleki, E.; Bagherifard, S.; Unal, O.; Bandini, M.; Farrahi, G.H.; Guagliano, M. Introducing gradient severe shot peening as a
novel mechanical surface treatment. Sci. Rep. 2021, 11, 22035. [CrossRef] [PubMed]
12. Maleki, E.; Unal, O.; Guagliano, M.; Bagherifard, S. The effects of shot peening, laser shock peening and ultrasonic nanocrystal
surface modification on the fatigue strength of Inconel 718. Mater. Sci. Eng. A 2021, 810, 141029. [CrossRef]
13. Maleki, E.; Unal, O.; Reza Kashyzadeh, K. Influences of Shot Peening Parameters on Mechanical Properties and Fatigue Behavior
of 316 L Steel: Experimental, Taguchi Method and Response Surface Methodology. Met. Mater. Int. 2021, 27, 4418–4440. [CrossRef]
14. Unal, O.; Maleki, E.; Karademir, I.; Husem, F.; Efe, Y.; Das, T. Effects of conventional shot peening, severe shot peening, re-shot
peening and precised grinding operations on fatigue performance of AISI 1050 railway axle steel. Int. J. Fatigue 2022, 155, 106613.
[CrossRef]
15. Nikam, V.V.; Reddy, R.G.; Collins, S.R.; Williams, P.C.; Schiroky, G.H.; Henrich, G.W. Corrosion resistant low temperature
carburized SS 316 as bipolar plate material for PEMFC application. Electrochim. Acta 2008, 53, 2743–2750. [CrossRef]
16. Martin, F.J.; Natishan, P.M.; Lemieux, E.J.; Newbauer, T.M.; Rayne, R.J.; Bayles, R.A.; Kahn, H.; Michal, G.M.; Ernst, F.; Heuer,
A.H. Enhanced corrosion resistance of stainless steel carburized at low temperature. Proc. Metall. Mater. Trans. A Phys. Metall.
Mater. Sci. 2009, 40, 1805–1810. [CrossRef]
J. Mar. Sci. Eng. 2022, 10, 128 10 of 12

17. Sharghi-Moshtaghin, R.; Kahn, H.; Ge, Y.; Gu, X.; Martin, F.J.; Natishan, P.M.; Rayne, R.J.; Michal, G.M.; Ernst, F.; Heuer,
A.H. Low-temperature carburization of the ni-base superalloy IN718: Improvements in surface hardness and crevice corrosion
resistance. Metall. Mater. Trans. A Phys. Metall. Mater. Sci. 2010, 41, 2022–2032. [CrossRef]
18. Balaraju, J.N.; Sankara Narayanan, T.S.N.; Seshadri, S.K. Evaluation of the corrosion resistance of electroless Ni-P and Ni-P
composite coatings by electrochemical impedance spectroscopy. J. Solid State Electrochem. 2001, 5, 334–338. [CrossRef]
19. Ono, S.; Tsuge, H.; Nishi, Y.; Hirano, S.I. Improvement of corrosion resistance of metals by an environmentally friendly silica
coating method. J. Sol-Gel Sci. Technol. 2004, 29, 147–153. [CrossRef]
20. Li, G.; Zhang, L.; Cai, F.; Yang, Y.; Wang, Q.; Zhang, S. Characterization and corrosion behaviors of TiN/TiAlN multilayer coatings
by ion source enhanced hybrid arc ion plating. Surf. Coat. Technol. 2019, 366, 355–365. [CrossRef]
21. Tang, P.T.; Watanabet, T.; Andersen, J.E.T.; Bech-Nielsen, G. Improved corrosion resistance of pulse plated nickel through
crystallisation control. J. Appl. Electrochem. 1995, 25, 347–352. [CrossRef]
22. Kashyzadeh, K.R.; Arghavan, A. Study of the effect of different industrial coating with microscale thickness on the CK45 steel by
experimental and finite element methods. Strength Mater. 2013, 45, 748–757. [CrossRef]
23. Bagherifard, S.; Beretta, N.; Monti, S.; Riccio, M.; Bandini, M.; Guagliano, M. On the fatigue strength enhancement of additive
manufactured AlSi10Mg parts by mechanical and thermal post-processing. Mater. Des. 2018, 145, 28–41. [CrossRef]
24. Maleki, E.; Bagherifard, S.; Razavi, S.M.J.; Riccio, M.; Bandini, M.; du Plessis, A.; Berto, F.; Guagliano, M. Fatigue behaviour of
notched laser powder bed fusion AlSi10Mg after thermal and mechanical surface post-processing. Mater. Sci. Eng. A 2022, 829,
142145. [CrossRef]
25. Maleki, E.; Bagherifard, S.; Sabouri, F.; Bandini, M.; Guagliano, M. Hybrid thermal, mechanical and chemical surface post-
treatments for improved fatigue behavior of laser powder bed fusion AlSi10Mg samples with notched geometry. Surf. Coat.
Technol. 2021, 430, 127962. [CrossRef]
26. Razavi, S.M.J.; Van Hooreweder, B.; Berto, F. Effect of build thickness and geometry on quasi-static and fatigue behavior of
Ti-6Al-4V produced by Electron Beam Melting. Addit. Manuf. 2020, 36, 101426. [CrossRef]
27. Solberg, K.; Berto, F. Notch-defect interaction in additively manufactured Inconel 718. Int. J. Fatigue 2019, 122, 35–45. [CrossRef]
28. Kahlin, M.; Ansell, H.; Moverare, J.J. Fatigue behaviour of notched additive manufactured Ti6Al4V with as-built surfaces. Int. J.
Fatigue 2017, 101, 51–60. [CrossRef]
29. Nicoletto, G. INFLUENCE OF ROUGH AS-BUILT SURFACES ON SMOOTH AND NOTCHED FATIGUE BEHAVIOR OF L-PBF
AlSi10Mg. Addit. Manuf. 2020, 34, 101251. [CrossRef]
30. Uzan, N.E.; Ramati, S.; Shneck, R.; Frage, N.; Yeheskel, O. On the effect of shot-peening on fatigue resistance of AlSi10Mg
specimens fabricated by additive manufacturing using selective laser melting (AM-SLM). Addit. Manuf. 2018, 21, 458–464.
[CrossRef]
31. Maleki, E.; Unal, O.; Bandini, M.; Guagliano, M.; Bagherifard, S. Individual and synergistic effects of thermal and mechanical
surface post-treatments on wear and corrosion behavior of laser powder bed fusion AlSi10Mg. J. Mater. Process. Technol. 2021, 302,
117479. [CrossRef]
32. Maleki, E.; Bagherifard, S.; Sabouri, F.; Guagliano, M. Effects of hybrid post-treatments on fatigue behaviour of notched LPBF
AlSi10Mg: Experimental and deep learning approaches. Procedia Struct. Integr. 2021, 34, 141–153. [CrossRef]
33. Genna, S.; Ucciardello, N. A thermographic technique for in-plane thermal diffusivity measurement of electroplated coatings.
Opt. Laser Technol. 2019, 113, 338–344. [CrossRef]
34. Arghavan, A.; Kashyzadeh, K.R.; Asfarjani, A.A. Investigating effect of industrial coatings on fatigue damage. Proc. Appl. Mech.
Mater. 2011, 87, 230–237. [CrossRef]
35. Weldon, D.G. Failure Analysis of Paints and Coatings: Revised Edition; Wiley: Hoboken, NJ, USA, 2009; ISBN 9780470697535.
36. Karger-Kocsis, J. Paints, coatings and solvents. Compos. Sci. Technol. 1994, 51, 613–614. [CrossRef]
37. Zadeh, M.K.; Yeganeh, M.; Shoushtari, M.T.; Esmaeilkhanian, A. Corrosion performance of polypyrrole-coated metals: A review
of perspectives and recent advances. Synth. Met. 2021, 274, 116723. [CrossRef]
38. Cha, S.C.; Erdemir, A. Coating Technology for Vehicle Applications; Springer: New York, NY, USA, 2015; ISBN 9783319147710.
39. Kovářík, O.; Haušild, P.; Čapek, J.; Medřický, J.; Siegl, J.; Mušálek, R.; Pala, Z.; Curry, N.; Björklund, S. Resonance bending fatigue
testing with simultaneous damping measurement and its application on layered coatings. Proc. Int. J. Fatigue 2016, 82, 300–309.
[CrossRef]
40. Yang, X.; Li, S.; Qi, H. Effect of MCrAlY coating on the low-cycle fatigue behavior of a directionally solidified nickel-base
superalloy at different temperatures. Int. J. Fatigue 2015, 75, 126–134. [CrossRef]
41. Maleki, E. Artificial neural networks application for modeling of friction stir welding effects on mechanical properties of 7075-T6
aluminum alloy. Proc. IOP Conf. Ser. Mater. Sci. Eng. 2015, 103, 012034. [CrossRef]
42. Maleki, N.; Kashanian, S.; Maleki, E.; Nazari, M. A novel enzyme based biosensor for catechol detection in water samples using
artificial neural network. Biochem. Eng. J. 2017, 128, 1–11. [CrossRef]
43. Maleki, E.; Unal, O. Optimization of Shot Peening Effective Parameters on Surface Hardness Improvement. Met. Mater. Int. 2020,
27, 3173–3185. [CrossRef]
44. Jahanshahi, M.; Maleki, E.; Ghiami, A. On the efficiency of artificial neural networks for plastic analysis of planar frames in
comparison with genetic algorithms and ant colony systems. Neural Comput. Appl. 2016, 28, 3209–3227. [CrossRef]
J. Mar. Sci. Eng. 2022, 10, 128 11 of 12

45. Maleki, E.; Unal, O.; Reza Kashyzadeh, K. Fatigue behavior prediction and analysis of shot peened mild carbon steels. Int. J.
Fatigue 2018, 116, 48–67. [CrossRef]
46. Kong, Y.S.; Abdullah, S.; Schramm, D.; Omar, M.Z.; Haris, S.M. Optimization of spring fatigue life prediction model for vehicle
ride using hybrid multi-layer perceptron artificial neural networks. Mech. Syst. Signal Process. 2019, 122, 597–621. [CrossRef]
47. Maleki, E.; Kashyzadeh, K.R.R. Effects of the hardened nickel coating on the fatigue behavior of CK45 steel: Experimental, finite
element method, and artificial neural network modeling. Iran. J. Mater. Sci. Eng. 2017, 14, 81–99. [CrossRef]
48. Maleki, E.; Unal, O. Fatigue limit prediction and analysis of nano-structured AISI 304 steel by severe shot peening via ANN. Eng.
Comput. 2020, 37, 2663–2678. [CrossRef]
49. Sbarufatti, C.; Manes, A.; Giglio, M. Performance optimization of a diagnostic system based upon a simulated strain field for
fatigue damage characterization. Mech. Syst. Signal Process. 2013, 40, 667–690. [CrossRef]
50. Fathalla, E.; Tanaka, Y.; Maekawa, K. Remaining fatigue life assessment of in-service road bridge decks based upon artificial
neural networks. Eng. Struct. 2018, 171, 602–616. [CrossRef]
51. Jimenez-Martinez, M.; Alfaro-Ponce, M. Fatigue damage effect approach by artificial neural network. Int. J. Fatigue 2019, 124,
42–47. [CrossRef]
52. Venkatesh, V.; Rack, H.J. Neural network approach to elevated temperature creep-fatigue life prediction. Int. J. Fatigue 1999, 21,
225–234. [CrossRef]
53. Vassilopoulos, A.P.; Georgopoulos, E.F.; Dionysopoulos, V. Artificial neural networks in spectrum fatigue life prediction of
composite materials. Int. J. Fatigue 2007, 29, 20–29. [CrossRef]
54. Figueira Pujol, J.C.; Andrade Pinto, J.M. A neural network approach to fatigue life prediction. Int. J. Fatigue 2011, 33, 313–322.
[CrossRef]
55. Barbosa, J.F.; Correia, J.A.F.O.; Júnior, R.C.S.F.; de Esus, A.M.P. Fatigue life prediction of metallic materials considering mean
stress effects by means of an artificial neural network. Int. J. Fatigue 2020, 135, 105527. [CrossRef]
56. Abdalla, J.A.; Hawileh, R. Modeling and simulation of low-cycle fatigue life of steel reinforcing bars using artificial neural
network. Proc. J. Frankl. Inst. 2011, 348, 1393–1403. [CrossRef]
57. Amiri, N.; Farrahi, G.H.; Kashyzadeh, K.R.; Chizari, M. Applications of ultrasonic testing and machine learning methods to
predict the static & fatigue behavior of spot-welded joints. J. Manuf. Process. 2020, 52, 26–34. [CrossRef]
58. Cervera, M.; Barbat, G.B.; Chiumenti, M.; Wu, J.Y. A Comparative Review of XFEM, Mixed FEM and Phase-Field Models for
Quasi-brittle Cracking. Arch. Comput. Methods Eng. 2021, 1–75. [CrossRef]
59. Pandey, V.B.; Singh, I.V.; Mishra, B.K.; Ahmad, S.; Venugopal Rao, A.; Kumar, V. A new framework based on continuum damage
mechanics and XFEM for high cycle fatigue crack growth simulations. Eng. Fract. Mech. 2019, 206, 172–200. [CrossRef]
60. Bui, T.Q.; Hu, X. A review of phase-field models, fundamentals and their applications to composite laminates. Eng. Fract. Mech.
2021, 248, 107705. [CrossRef]
61. Lo, Y.S.; Borden, M.J.; Ravi-Chandar, K.; Landis, C.M. A phase-field model for fatigue crack growth. J. Mech. Phys. Solids 2019,
132, 103684. [CrossRef]
62. Maleki, E.; Unal, O. Shot Peening Process Effects on Metallurgical and Mechanical Properties of 316 L Steel via: Experimental and
Neural Network Modeling. Met. Mater. Int. 2019, 27, 262–276. [CrossRef]
63. Maleki, E.; Bagherifard, S.; Guagliano, M. Application of artificial intelligence to optimize the process parameters effects on
tensile properties of Ti-6Al-4V fabricated by laser powder-bed fusion. Int. J. Mech. Mater. Des. 2021, 1–24. [CrossRef]
64. Feng, S.; Zhou, H.; Dong, H. Using deep neural network with small dataset to predict material defects. Mater. Des. 2019, 162,
300–310. [CrossRef]
65. Maleki, E.; Unal, O.; Guagliano, M.; Bagherifard, S. Analysing the Fatigue Behaviour and Residual Stress Relaxation of Gradient
Nano-structured 316L Steel Subjected to the Shot Peening via Deep Learning Approach. Met. Mater. Int. 2021, 28, 112–131.
[CrossRef]
66. Kalayci, C.B.; Karagoz, S.; Karakas, Ö. Soft computing methods for fatigue life estimation: A review of the current state and
future trends. Fatigue Fract. Eng. Mater. Struct. 2020, 43, 2763–2785. [CrossRef]
67. Abiodun, O.I.; Jantan, A.; Omolara, A.E.; Dada, K.V.; Mohamed, N.A.E.; Arshad, H. State-of-the-art in artificial neural network
applications: A survey. Heliyon 2018, 4, e00938. [CrossRef] [PubMed]
68. Tao, C.; Zhang, C.; Ji, H.; Qiu, J. Fatigue damage characterization for composite laminates using deep learning and laser ultrasonic.
Compos. Part B Eng. 2021, 216, 108816. [CrossRef]
69. Lee, H.; Lim, H.J.; Skinner, T.; Chattopadhyay, A.; Hall, A. Automated fatigue damage detection and classification technique for
composite structures using Lamb waves and deep autoencoder. Mech. Syst. Signal Process. 2021, 163, 108148. [CrossRef]
70. Rovinelli, A.; Sangid, M.D.; Proudhon, H.; Ludwig, W. Using machine learning and a data-driven approach to identify the small
fatigue crack driving force in polycrystalline materials. Npj Comput. Mater. 2018, 4, 35. [CrossRef]
71. Zhan, Z.; Li, H. Machine learning based fatigue life prediction with effects of additive manufacturing process parameters for
printed SS 316L. Int. J. Fatigue 2021, 142, 10594. [CrossRef]
72. Chen, J.; Liu, Y. Fatigue modeling using neural networks: A comprehensive review. Fatigue Fract. Eng. Mater. Struct. 2021.
[CrossRef]
J. Mar. Sci. Eng. 2022, 10, 128 12 of 12

73. Reza Kashyzadeh, K.; Maleki, E. Experimental Investigation and Artificial Neural Network Modeling of Warm Galvanization and
Hardened Chromium Coatings Thickness Effects on Fatigue Life of AISI 1045 Carbon Steel. J. Fail. Anal. Prev. 2017, 17, 1276–1287.
[CrossRef]
74. Maleki, E.; Maleki, N. Artificial Neural Network Modeling of Pt/C Cathode Degradation in PEM Fuel Cells. J. Electron. Mater.
2016, 45, 3822–3834. [CrossRef]
75. Maleki, E.; Mirzaali, M.J.; Guagliano, M.; Bagherifard, S. Analyzing the mechano-bactericidal effect of nano-patterned surfaces on
different bacteria species. Surf. Coat. Technol. 2020, 408, 126782. [CrossRef]
76. Maleki, E.; Unal, O.; Reza Kashyzadeh, K. Surface layer nanocrystallization of carbon steels subjected to severe shot peening:
Analysis and optimization. Mater. Charact. 2019, 157, 109877. [CrossRef]

View publication stats