PSI - Issue 48

First page Table of contents 1 Next page Last page

Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023)

Available online at www.sciencedirect.com Available online at www.sciencedirect.com ScienceDirect Structural Integrity Procedia 00 (2023) 000 – 000 Available online at www.sciencedirect.com ScienceDirect Structural Integrity Procedia 00 (2023) 000 – 000

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

ScienceDirect

Procedia Structural Integrity 48 (2023) 1–3

Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023) Editorial of the 2nd IRAS conference Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023) Editorial of the 2nd IRAS conference

Branislav Đorđević a , Aleksandar Sedmak b, *, Simon Sedmak a a Innovation Center of the Faculty of Mechanical Engineering, Kraljice Marije 16, Belgrade 11120, Serbia b University of Belgrade, Faculty of Mechanical Engineering, Kraljice Marije 16, Belgrade 11120, Serbia Branislav Đorđević a , Aleksandar Sedmak b, *, Simon Sedmak a a Innovation Center of the Faculty of Mechanical Engineering, Kraljice Marije 16, Belgrade 11120, Serbia b University of Belgrade, Faculty of Mechanical Engineering, Kraljice Marije 16, Belgrade 11120, Serbia

© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers Abstract The Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023) took place between 2-4 April 2023 at the Faculty of Mechanical Engineering of the University of Belgrade. The guest editors of the IRAS 2023 deeply acknowledge all members of the International Scientific Committee, Thematic Sessions Organizers, Keynote Speakers and authors that contributed to the success of this event, which gathered more than 90 participants presenting 88 papers on the spot or online. Sponsors are also fully acknowledged for their important contributions. The ESIS/TC12 2023 winners of the “Robert Moskovic” Award, Award of Merit TC12, and ESIS/TC12 Young Scientist Award were announced during the conference. Additionally, guest editors (conference chairs) sincerely thank the tireless efforts of the Organizing Committee members, as well as students and other staff involved in the organization. Finally, the guest editors are pleased to inform that the third edition of the IRAS event will be organized by Prof. Grzegorz Lesiuk (University of Wroclaw, Poland) and will take place in Wroclaw in Poland in 2025. © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers Keywords: Risk analysis; Safety; Structural integrity; Fatigue; Fracture mechanics; Complex structures; Structural components 1. Introduction The First International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2019) was organised by ESIS TC12, and took place between 1-2 July 2019 at the Faculty of Engineering of the University of Porto, Portugal. This conference attracted a large number of participants and was an excellent forum for discussion of the recent advances in risk analysis and safety of large Structures and Components. Having in mind its big success, TC12 Secretariat (co-chairmen Jose Correia, Aleksandar Sedmak and Vladimir Moskvichev, as well as scientific secretaries Abilio de Jesus, Elena Fedorova and Snežana Kirin) decided to organize the 2 nd IRAS Symposium in Belgrade in 2021. Unfortunately, due to Covid 19 pandemia, we had to postpone it until 2-4 April 2023 Abstract The Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023) took place between 2-4 April 2023 at the Faculty of Mechanical Engineering of the University of Belgrade. The guest editors of the IRAS 2023 deeply acknowledge all members of the International Scientific Committee, Thematic Sessions Organizers, Keynote Speakers and authors that contributed to the success of this event, which gathered more than 90 participants presenting 88 papers on the spot or online. Sponsors are also fully acknowledged for their important contributions. The ESIS/TC12 2023 winners of the “Robert Moskovic” Award, Award of Merit TC12, and ESIS/TC12 Young Scientist Award were announced during the conference. Additionally, guest editors (conference chairs) sincerely thank the tireless efforts of the Organizing Committee members, as well as students and other staff involved in the organization. Finally, the guest editors are pleased to inform that the third edition of the IRAS event will be organized by Prof. Grzegorz Lesiuk (University of Wroclaw, Poland) and will take place in Wroclaw in Poland in 2025. © 2023 The Authors. Published by ELSEVIER B.V. Keywords: Risk analysis; Safety; Structural integrity; Fatigue; Fracture mechanics; Complex structures; Structural components 1. Introduction The First International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2019) was organised by ESIS TC12, and took place between 1-2 July 2019 at the Faculty of Engineering of the University of Porto, Portugal. This conference attracted a large number of participants and was an excellent forum for discussion of the recent advances in risk analysis and safety of large Structures and Components. Having in mind its big success, TC12 Secretariat (co-chairmen Jose Correia, Aleksandar Sedmak and Vladimir Moskvichev, as well as scientific secretaries Abilio de Jesus, Elena Fedorova and Snežana Kirin) decided to organize the 2 nd IRAS Symposium in Belgrade in 2021. Unfortunately, due to Covid 19 pandemia, we had to postpone it until 2-4 April 2023

* Corresponding author E-mail address: asedmak@mas.bg.ac.rs * Corresponding author E-mail address: asedmak@mas.bg.ac.rs

Branislav Đorđević et al. / Procedia Structural Integrity 48 (2023) 1– 3 Branislav Đorđević et al / Structural Int egrity Procedia 00 (2023) 000 – 000

in mixed mode, on the spot and online. 2 nd IRAS is organized by European Structural Integrity Society (ESIS) TC12 and (Serbian) Society for Structural Integrity and Life “Prof. Dr. Stojan Sedmak”. The IRAS 2023 event had two keynote lecture sessions, with 7 presentations in total:  Liviu Marsavaina ( University Politehnica Timisoara, Romania ) “The Size and Notch Effect on Additive Manufactured Polymers”  Abilio M.P. de Jesus ( Faculty of Engineering of the University of Porto, Portugal ), “Exploring the Local Fatigue Approaches to Improve the Structural Integrity of Metallic Structures and Mechanical Components”  Chao Gao ( Norwegian University of Science and Technology of Trondheim, Norway ) “Bioinspired strat egy to break trade- off between strength and toughness”  Dražan Kozak ( University of Slavonski Brod, Mechanical Engineering Faculty, Croatia ) “Structural Health Monitoring by Embedded System for Remote Strain Gauge Measurement”  Nenad Gubeljak ( University of Maribor, Faculty of Mechanical Engineering, Slovenia ) ”Fatigue lifetime of a howitzer cannon”  Sreten Mastilović ( Institute for Multidisciplinary Research, University of Belgrade, Serbia ) “Size -Effect in Fracture Mechanics Testing by Using the Weibull J c Cu mulative Distribution Function”  Aleksandar Milivojević ( Faculty of Mechanical Engineering, University of Belgrade, Serbia ) “The use of hydrogen as an additive” Within 7 technical and 2 promotional sessions, 88 talks were presented by more than 90 participants on various aspects of risk analysis and safety of complex structures and components. The second edition of the IRAS Symposium gathered participants from 24 countries demonstrating the worldwide acceptance of this event. This conference is firstly followed by Book of abstracts, while this Volume of Procedia Structural Engineering contains 52 manuscripts, selected out of 63 submitted. The Organizing Committee of the IRAS 2023 deeply acknowledges all authors that contributed to the success of this event. The members of the International Scientific Committee are also fully acknowledged for their support to the IRAS 2023 event. Sponsors are also fully acknowledged for their important contributions. Finally, the guest editors sincerely thank the tireless efforts of Organizing Committee members, as well as students and other staff from the Faculty of Mechanical Engineering, University of Belgrade. 2. Topics of the conference The topics of the conference included:  Methods for reliability and probabilistic safety assessment  Sensitivity analysis  Data collection and analysis  Hydrogen Embrittlement  Design and evaluation of technical systems and structures  Design innovation for safety and reliability  Codes, standards and safety criteria  Engineering Structures  Structural Integrity and durability  Management and life-cycle performance  Models for ageing and life extension  Fatigue, fracture and damage mechanics  Damage evaluation and fatigue design  Dynamic and fatigue reliabilities  Safety assessment (loads and environmental effects; material properties; prediction of response and performance)  Analytical and numerical simulation  Structural health monitoring  Rock and soil structural integrity

Branislav Đorđević et al. / Procedia Structural Integrity 48 (2023) 1– 3 Branislav Đorđević et al / Structural Integrity Procedia 00 (2023) 000 – 000

3. TC12/ESIS awards Awards (“Robert Moskovic” Award; Award of Merit and Young Scientist Award) are another important activity of TC12. In 2020 the “Robert Moskovic” Award was attributed to:  Prof. Neil James, Plymouth University, UK, for contributions in the Fatigue and Fracture of Structural Alloys and Materials field;  Prof. Rui Calçada, University of Porto, Portugal, for contributions in the Integrity of Railway Structures and Infrastructures field;  Prof. Vladimir Moskvichev, Russian Academy of Sciences, Russia, for contributions in the Risk Analysis and Safety of Technical Systems field. In 2021 the “Robert Moskovic” Award was attributed to:  Prof. Filippo Berto, Norwegian University of Science and Technology, Norway, for contributions in the Local Approaches for Fracture and Fatigue Assessment field;  Prof. Milan Veljkovic, Delft University of Technology, Netherlands, for contributions in the Integrity of Steel Structures field;  Prof. Maria Feng, Columbia University, USA, for contributions in the Safety and Sustainability of Civil Infrastructure Systems field. In 2022, the Robert Moskovic Award is attributed to:  Prof. Hryhoriy Nykyforchyn, Karpenko Physico-Mechanical Institute, National Academy of Sciences of Ukraine, Ukraine, for contributions in the Corrosion-Hydrogen Degradation and Materials Protection field;  Dr. John Michopoulos, Naval Research Laboratory, Washington DC, USA, for contributions in the Computational, Theoretical and Experimental Multiphysics field;  Prof. Grzegorz Lesiuk, Wroclaw University of Science and Technology, Poland, for contributions in the Fatigue Crack Growth Analysis field. In 2022, the Robert Moskovic Award is attributed to:  Prof. Doctor Aleksandar Grbovic, University of Belgrade, Serbia, for an outstanding contribution to Fatigue Crack Growth Simulation  Prof. Doctor Marc André Meyers, University of California San Diego, USA for an outstanding contribution to Dynamic Behavior of Materials and Characterization of New Materials  Doctor Motomichi Koyama, Institute for Materials Research, Tohoku, University, Japan for an outstanding contribution to Material Resistance to Hydrogen Cracking The Award of Merit TC12 was attributed to the following researchers for their contributions to IRAS in 2023:  Prof. Abílio De Jesus, University of Porto, Portugal  Doctor Branislav Djordjevic, Innovation Center of the Faculty of Mechanical Engineering, Belgrade, Serbia  Doctor Simon Sedmak, Innovation Center of the Faculty of Mechanical Engineering, Belgrade, Serbia The ESIS/TC12 Young Scientist Award was attributed to Dr. Dorin Radu, Faculty of Civil Engineering, University of Brasov, Romania, for his scientific work. Acknowledgements The guest editors of this Volume would like to express a special thanks to Professor Francesco Iacoviello, Editor in-Chief of Procedia Structural Integrity journal, and to Elsevier staff for their support during the preparation of this issue. On behalf of the Organizing committee, guest editors send their gratitude to European Structural Integrity Society (ESIS), as well as all the sponsors of the event: TankMont (Serbia); Mont-R (Serbia); TrokutTest Group (Serbia); Metal Rehabilitation and Testing Ltd (Serbia); Institute de Soudure/Welding Institute (Serbia); Weld-Ing (Serbia); Innovation Center of Facluty of Mechanical Engineering in Belgrade (Serbia); Faculty of Mechanical Engineering in Belgrade (Serbia); The Bay Zoltan Nonprofit Ltd. for Applied Research (Hungary). Special thanks is attributed to Ministry of Science, Technological Development and Innovation of the Republic of Serbia.

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

ScienceDirect

Procedia Structural Integrity 48 (2023) 183–189

© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers Abstract There was modeled the stress-strain diagram of 6061-T651 aluminum alloy by machine learning methods. In this study, methods of k -nearest neighbors and random forest were applied to obtain the best model for predicting the stress-strain diagram of 6061 T651 aluminum alloy at six temperatures (20, 100, 150, 200, 250, 300ºС). The obtained results agree well with the experimental data. It was determined that the errors of 9.6% and 5.9% for linear and non-linear regions, respectively, were obtained by the method of k -nearest neighbors in the test sample. The errors of the random forest method were 15% and 13.7%. © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers Keywords: stress-strain diagram, machine learning, method of k -nearest neighbors, random forest, 6061-T651 aliminum alloy 1. Introduction Stress-strain diagrams of materials are built to determine their mechanical characteristics. In particular, the curve will have a different shape depending on the type of material, its state, and the conditions under which the testing was carried out. That is, these are the dependences of strength parameters on deformation ones. In general, stress-strain diagrams are constructed by different methods. In particular, in the paper by Molkov (2013), the conditional and the actual deformation curves of 65G spring carbon steel is plotted using the standard approach, and the digital image correlation technique (DIC) is constructed and shows a good correlation of the obtained results. In the paper by Pylypenko et al. (2009) the advantages of the complete stress-strain softening diagrams for estimation of limiting material damage under complex Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023) Application of machine learning for modeling of 6061-T651 aluminum alloy stress−strain diagram Oleh Yasniy a , Oleh Pastukh a , Iryna Didych a, *, Vasyl Yatsyshyn a , Ihor Chykhira a a Ternopil Ivan Puluj National Technical University, 56, Ruska Str., Ternopil, 46001, Ukraine Abstract There was modeled the stress-strain diagram of 6061-T651 aluminum alloy by machine learning methods. In this study, methods of k -nearest neighbors and random forest were applied to obtain the best model for predicting the stress-strain diagram of 6061 T651 aluminum alloy at six temperatures (20, 100, 150, 200, 250, 300ºС). The obtained results agree well with the experimental data. It was determined that the errors of 9.6% and 5.9% for linear and non-linear regions, respectively, were obtained by the method of k -nearest neighbors in the test sample. The errors of the random forest method were 15% and 13.7%. © 2023 The Authors. Published by ELSEVIER B.V. Keywords: stress-strain diagram, machine learning, method of k -nearest neighbors, random forest, 6061-T651 aliminum alloy 1. Introduction Stress-strain diagrams of materials are built to determine their mechanical characteristics. In particular, the curve will have a different shape depending on the type of material, its state, and the conditions under which the testing was carried out. That is, these are the dependences of strength parameters on deformation ones. In general, stress-strain diagrams are constructed by different methods. In particular, in the paper by Molkov (2013), the conditional and the actual deformation curves of 65G spring carbon steel is plotted using the standard approach, and the digital image correlation technique (DIC) is constructed and shows a good correlation of the obtained results. In the paper by Pylypenko et al. (2009) the advantages of the complete stress-strain softening diagrams for estimation of limiting material damage under complex Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023) Application of machine learning for modeling of 6061-T651 aluminum alloy stress−strain diagram Oleh Yasniy a , Oleh Pastukh a , Iryna Didych a, *, Vasyl Yatsyshyn a , Ihor Chykhira a a Ternopil Ivan Puluj National Technical University, 56, Ruska Str., Ternopil, 46001, Ukraine

* Corresponding author. Tel.: /; fax: /. E-mail address : iryna.didych1101@gmail.com * Corresponding author. Tel.: /; fax: /. E-mail address : iryna.didych1101@gmail.com

Oleh Yasniy et al. / Procedia Structural Integrity 48 (2023) 183–189 Yasniy et al / Structural Integrity Procedia 00 (2019) 000 – 000

184

modes loading showed. In the paper by Yasnii et al. (2004), a procedure for predicting the jump-like deformation in the alloy based on the histograms of the distribution of dispersed particles in the initial material is put forward. In general, it is important to evaluate the behavior of the material under different types of stress states with a limited number of experiments. Therefore, in this case, it is advisable to use machine learning methods. Machine learning is an area of artificial intelligence by Smola et al. (2010). It is used for studying input data and constructing a model by continuously evaluating, optimizing, and settings parameters. In particular, due to the ability to interpret non-linear relationships between input and output data, machine learning methods can solve problems of fracture mechanics with high accuracy by Pidaparti et al. (1995), Mohanty et al. (2009). In particular, in the paper by Yasnii et al. (2018) using methods of machine learning (neural networks, boosted trees, random forests, support-vector machines, and k -nearest neighbors), diagrams of fatigue fracture of D16T aluminum alloy under regular loading with a stress ratio R = 0, 0.2, 0.4, and 0.6 were constructed, and it was found that the method of neural networks gives the slightest prediction error equal to 3.2 and 2.5% in tested samples. In papers by Seed et al. (1998), Pujol et al. (2011), the growth of short fatigue cracks and fatigue lifetime under step-stress conditions by method of neural networks are predicted, respectively. It is known that machine learning is often used to predict the dependencies of shape memory alloys (SMAs). In particular, in the paper by Hmede et al. (2022), functions of the SMAs are effectively modeled using machine and deep learning methods, whereas, in the paper by Trehern et al. (2022), an AI-enabled materials discovery framework was successfully used to identify both SMA chemistries and the associated thermo-mechanical processing steps that result in narrow transformation hysteresis and transformation range under applied stress. In addition, stress strain diagrams of aluminum alloy AMg6 by Yasniy et al. (2020), Didych et al. (2022), and aluminum alloy AL-6061 by Didych et al. (2022) were predicted by machine learning methods. Therefore, it is advisable to use them for numerical modeling of stress-strain diagrams of 6061-T651 aluminum alloy at different temperatures. The aim of this study is to predict the stress-strain diagram of 6061-T651 aluminum alloy at six different temperatures (20, 100, 150, 200, 250, 300 ºС) using machi ne learning methods, in particular, the method of k -nearest neighbors and random forests, and to compare the obtained results. 2. Material and methods It is known by Haykin (2006) that there are several main approaches that are widely used in the field of data prediction, that is, supervised learning, unsupervised and mixed. In the first case, it interprets the teacher's participation as the knowledge presented in the form of input-output couples. In particular, the network parameters are corrected by the difference between the desirable and output signals of the network. In contrast, the network with unsupervised learning cannot know the correct answers to each sample of the training sample. In the mixed case, part of the weights is determined by supervised learning, while the other is obtained when the network is self-learning. It is known that at the stage of machine learning model development, preparation of data, algorithm construction, training on the learning data, and verification using the test data are essential. The k – nearest neighbors method algorithm is based on comparing known elements with new ones. Its basic idea is that the new object to be predicted belongs to the class that is most common among k – nearest neighbors of the training sample (Fig.1). The distance between k -nearest neighbors is usually Euclidean. This machine learning method is the algorithm for supervised learning, so it requires a marked data set. In particular, regression problems are concerned with the result prediction of the dependent variable given a set of independent variables. The prediction results are supposed to be the average of the results of its k – nearest neighbors.

Fig.1. Examples of data with plus and minus signs and the query point marked by a blue triangle by Smola et al. (2010)

Oleh Yasniy et al. / Procedia Structural Integrity 48 (2023) 183–189 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

185

One of the powerful methods of machine learning is random forests. This ensemble algorithm uses the concepts of bagging and random subspaces, and the basic algorithms are decision trees by Alpaydin (2010). In the regression problem, their answers are averaged; in the classification problem, the decision is made by majority vote (Fig.2).

Fig. 2. Algorithm of random forests

A tree is built, as a rule, until the sample is exhausted (until representatives of only one class remain in the leaves). Still, in modern implementations, some parameters limit the tree's height, the number of objects in the leaves, and the number of things in the subsample at which the splitting is. 3. Results and discussion The stress-strain diagram of the 6061-T651 aluminum alloy is predicted by machine learning methods according to the experimental data obtained in the paper by Aakash et al. (2019). In the learning process, the data set was divided into two unequal parts - training and test samples. In addition, the stress-strain diagram was divided into two regions, that is, linear and non-linear, to improve the prediction quality. As a result, two networks were built using different machine learning methods. The sample contained 2018 elements in the linear region and 643 elements in the non linear region, of which 80% were randomly selected for the training sample, and 20% were left to assess the quality of the prediction. It is important that stress and temperature were chosen as the input parameter, while deformation is the output parameter. Several numerical experiments were conducted using input-output pairs to obtain the best algorithm architecture. It is found that the obtained results are in good agreement with the experimental data. The prediction error was calculated using the Mean Absolute Percent Error (MAPE) formula: = 100% ⋅ 1 ∑ | − | | | =1 (1) By using the methods of machine learning, we plotted the dependences of the experimental data on the predicted values of the linear region (0.01-0.08) of the strain for different temperatures (Fig. 3).

Oleh Yasniy et al. / Procedia Structural Integrity 48 (2023) 183–189 Yasniy et al / Structural Integrity Procedia 00 (2019) 000 – 000

186 4

0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.00 0.02 0.04 0.06 0.08 0.10 Strain (pred), mm/mm Strain (exp), mm/mm

Fig. 3. The predicted and experimental data of linear region (0.01-0.08) of the strain for different temperatures obtained by the method of k-nearest neighbors (left) and random forests (rigth) The predicted and experimental dependences of the linear region of the strain on the stress for different temperatures obtained by the methods of machine learning (Fig. 4). By using the methods of machine learning, we plotted the dependences of the experimental data on the predicted values of the non-linear region (0-0.01) of the strain for different temperatures (Fig. 5). The predicted and experimental dependences of the non-linear region (0-0.01) of the strain for different temperatures obtained by the methods of machine learning (Fig. 6). In addition, the dependences of the root mean square error on the number of trees for the linear (a) and non-linear (b) region obtained by the random forests method are shown in Fig. 7. The dependences of importance of input parameters for the linear (a) and non-linear (b) region obtained by the random forests method are shown in Fig. 8.

300

Exp Pred _20 Pred _100 Pred _150 Pred _200 Pred _250 Pred _300

Exp Pred_ 20 Pred_ 100 Pred_ 150 Pred_ 200 Pred_ 250 Pred_ 300

250

200

150

100 Stress, MPa

0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08

Strain, mm/mm

Fig. 4. The predicted and experimental dependences of the linear region of the strain on the stress for different temperatures obtained by the method of k-nearest neighbors (left) and random forests (right)

Oleh Yasniy et al. / Procedia Structural Integrity 48 (2023) 183–189 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

187 5

0.010

0.008

0.006

0.004

Strain, mm/mm

0.002

Strain (pred), mm/mm

0.000

0.005

0.010

0.000

0.005

0.010

Strain (exp), mm/mm

Strain, mm/mm

Fig. 5. The predicted and experimental data of the non-linear region (0-0.01) of the strain for different temperatures obtained by the method of k nearest neighbors (left) and random forests (right)

300

Exp Pred_20 Pred_100 Pred_150 Pred_200 Pred_250 Pred_300

250

Exp Pred_20 Pred_100 Pred_150 Pred_200 Pred_250 Pred_300

200

150

100

Stress, MPa

0.000 0.002 0.004 0.006 0.008 0.010

Strain, mm/mm

Fig. 6. The predicted and experimental dependences of the non-linear region (0-0.01) of the strain for different temperatures obtained by the method of k-nearest neighbors (left) and random forests (right)

Oleh Yasniy et al. / Procedia Structural Integrity 48 (2023) 183–189 Yasniy et al / Structural Integrity Procedia 00 (2019) 000 – 000

188

0,0000016

0,00012

Train data Test data

0,00011

0,0000014

Train data Test data

0,00010

0,0000012

0,00009

0,0000010

0,00008

0,0000008

0,00007

0,0000006 Average Squared Error

Average Squared Error

0,00006

0,0000004

0,00005

0,0000002

0,00004

10 20 30 40 50 60 70 80 90 100 Number of Trees

90 100

Number of Trees

Fig. 7. Dependences of the root mean square error on the number of trees for the linear (left) and non-linear (right region obtained by random forests

Fig. 8. The dependences of importance of input parameters for the linear (left) and non-linear (right) region obtained by random forests

The number of k -nearest neighbors in the linear and non-linear sections is 10 and 1, respectively. The number of trees in two cases is equal to 100. 4. Conclusions The stress-strain diagrams of 6061- T651 aluminum alloy at temperatures T = 20, 100, 150, 200, 250, and 300ºС were built by machine learning methods, particularly by k -nearest neighbors and random forests. It was found that the prediction results agreed with the experimental ones. It was determined that the errors of 9.6% and 5.9% for linear and non-linear regions were obtained by the k -nearest neighbors method in the test sample. The errors of the random forest method were 15% and 13.7%.

Oleh Yasniy et al. / Procedia Structural Integrity 48 (2023) 183–189 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

189

References

Molˈkov, Y.V., 2013. Application of the Method of Digital Image Correlation to the Construction of Stress – Strain Diagrams. Materials Science 48, 832 – 837. Pylypenko, A., Paratsa, V., 2009. The usage of complete stress - strain softening diagrams for an estimation of limiting damage of material under complex modes loading. Visnyk Ternopilskoho derzhavnoho tekhnichnoho universytetu 14, 45 - 51. [in Ukrainian] Yasnii, P. V., Fedak, S. I., Glad ’ o, V. B., Galushchak, M. P., 2004. Jumplike Deformation in AMg6 Aluminum Alloy in Tension. Strength of Materials 36, 113 – 118. Smola, A., Vishwanathan, S.V.N., 2010. Introduction to Machine Learning , Cambridge University Press, pp. 234. Pidaparti, R. M. V., Palakal, M. J., 1995. Neural network approach to fatigue-crack-growth predictions under aircraft spectrum loadings. Journal of Aircraft 32, 825-831. Mohanty, J. R., Verma, B. B., Parhi, D. R. K., Ray D. R., 2009. Application of artificial neural network for predicting fatigue crack propagation life of aluminum alloys. Archives of Computational Materials Science and Surface Engineering 1, 133 – 138. Yasnii О .P., Pastukh O. А ., Pyndus Yu. І ., Lutsyk N.S., Didych I.S., 2018. Prediction of the diagrams of fatigue fracture of D16T aluminum alloy by the methods of machine learning. Materials Science 54, 333-338. Seed, G. M., Murphy, G. S., 1998. The applicability of neural networks in modeling the growth of short fatigue cracks. Fatigue Fracture Engng Mater. Struct. 21, 183-190. Pujol, J. C. F., Pinto, J. M. A., 2011. A neural network approach to fatigue life prediction. International Journal of Fatigue 33, 313-322. Hmede, R., Chapelle, F., Lapusta, Yu., 2022. Review of Neural Network Modeling of Shape Memory Alloys. Sensors 22, 5610. Trehern, W., Ortiz-Ayala, R., Atli, K.C., Arroyave, R., Karaman, I., 2022. Data-driven shape memory alloy discovery using Artificial Intelligence Materials Selection (AIMS) framework. Acta Materialia 228, 117751. Yasniy O., Didych I., Fedak S., Lapusta Yu., 2020. Modeling of AMg6 aluminum alloy jump-like deformation properties by machine learning methods. Procedia Structural Integrity, 28,1392-1398. Didych, I., Yasniy, O., Fedak, S., Lapusta, Yu., 2022. Prediction of jump-like creep using preliminary plastic strain. Procedia Structural Integrity 36, 166-170. Didych, I., Yasniy, O., Pasternak, I., Sobashek, L., 2022. Modelling of AL-6061 aluminum alloy deformation diagrams by machine learning methods. Procedia Structural Integrity 42, 1344-1349. Haykin, S., 1999. Neural Networks: A Comprehensive Foundation. Second Ed., Prentice Hall, Canada, pp. 823. Alpayndin, E., 2010. Introduction to Machine Learning. The Knowledge Engineering Review 25, 353. Aakash, B. S., Connors, J. P., Shields, M. D. ,2019. Stress-strain data for aluminum 6061-T651 from 9 lots at 6 temperatures under uniaxial and plane strain tension. Data in Brief, 25.

ScienceDirect Structural Integrity Procedia 00 (2023) 000 – 000 Structural Integrity Procedia 00 (2023) 000 – 000 Available online at www.sciencedirect.com Available online at www.sciencedirect.com ScienceDirect Available online at www.sciencedirect.com ScienceDirect

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

Procedia Structural Integrity 48 (2023) 73–80

© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers Abstract Ensuring a higher level of safety during the transportation and handling of LNG (liquefied natural gas) fuel is critical. At a cryogenic temperature that can reach -163°C, LNG is stored in specialized Type-C independent tanks, and any accidental gas leak may cause damage to the ship's structure. Prolonged exposure to LNG flow may cause steel to become more brittle, and an accidental load on the ship's structure may result in structural collapse due to brittle fracture, leading to significant losses. As a mitigation measure against the risk of accidental LNG release on LNG-fueled ships, the use of higher strength and cryogenic temperature-resistant steel is necessary. This paper presents a discussion on the use of steel for shipbuilding, with a focus on the potential risks associated with accidental LNG release during transportation and the performance of steels subjected to low temperatures. Based on existing research, the discussion is organized into three topics, namely experiments conducted at cryogenic temperatures, including tensile tests and Charpy V-notch impact tests, and material modeling using finite element analysis. The paper provides a procedure and method regarding to these topics. © 2023 The Authors. Published by ELSEVIER B.V. This is an open-access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer review under the responsibility of the IRAS 2023 organizers Keywords: Brittle Fracture; DTBTT; Charpy V-Notch Test; Tensile Test; Cryogenic Temperature; Finite Element Analysis Abstract Ensuring a higher level of safety during the transportation and handling of LNG (liquefied natural gas) fuel is critical. At a cryogenic temperature that can reach -163°C, LNG is stored in specialized Type-C independent tanks, and any accidental gas leak may cause damage to the ship's structure. Prolonged exposure to LNG flow may cause steel to become more brittle, and an accidental load on the ship's structure may result in structural collapse due to brittle fracture, leading to significant losses. As a mitigation measure against the risk of accidental LNG release on LNG-fueled ships, the use of higher strength and cryogenic temperature-resistant steel is necessary. This paper presents a discussion on the use of steel for shipbuilding, with a focus on the potential risks associated with accidental LNG release during transportation and the performance of steels subjected to low temperatures. Based on existing research, the discussion is organized into three topics, namely experiments conducted at cryogenic temperatures, including tensile tests and Charpy V-notch impact tests, and material modeling using finite element analysis. The paper provides a procedure and method regarding to these topics. © 2023 The Authors. Published by ELSEVIER B.V. This is an open-access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer review under the responsibility of the IRAS 2023 organizers Keywords: Brittle Fracture; DTBTT; Charpy V-Notch Test; Tensile Test; Cryogenic Temperature; Finite Element Analysis Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023) A review on the hull structural steels for ships carrying liquefied gas: Materials performance subjected to low temperatures Haris Nubli a , Suryanto Suryanto b , Aprianur Fajri b , Jung Min Sohn a,c , Aditya Rio Prabowo b, * a Department of Marine Convergence Design Engineering, Pukyong National University, Busan, Republic of Korea b Department of Naval Architecture and Marine Systems Engineering, Pukyong National University, Busan, Republic of Korea c Department of Mechanical Engineering, Universitas Sebelas Maret, Surakarta, Indonesia Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023) A review on the hull structural steels for ships carrying liquefied gas: Materials performance subjected to low temperatures Haris Nubli a , Suryanto Suryanto b , Aprianur Fajri b , Jung Min Sohn a,c , Aditya Rio Prabowo b, * a Department of Marine Convergence Design Engineering, Pukyong National University, Busan, Republic of Korea b Department of Naval Architecture and Marine Systems Engineering, Pukyong National University, Busan, Republic of Korea c Department of Mechanical Engineering, Universitas Sebelas Maret, Surakarta, Indonesia

* Corresponding author. Tel.: /; fax: /. E-mail address: aditya@ft.uns.ac.id 1. Introduction * Corresponding author. Tel.: /; fax: /. E-mail address: aditya@ft.uns.ac.id 1. Introduction

2452-3216 © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers 2452-3216 © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers The world fleet's current status indicates that the use of liquefied natural gas (LNG) as an alternative fuel remains dominant, with 923 out of 1,349 ships in operation and 534 out of 1,046 ships on order (DNV, 2021). However, the increasing number of LNG-fueled ships requires an adequate supporting structure and infrastructure, which shipping through harsh sea route will make the ship prone to structural damage (Cao et al., 2016; Prabowo et al., The world fleet's current status indicates that the use of liquefied natural gas (LNG) as an alternative fuel remains dominant, with 923 out of 1,349 ships in operation and 534 out of 1,046 ships on order (DNV, 2021). However, the increasing number of LNG-fueled ships requires an adequate supporting structure and infrastructure, which shipping through harsh sea route will make the ship prone to structural damage (Cao et al., 2016; Prabowo et al.,

Haris Nubli et al. / Procedia Structural Integrity 48 (2023) 73–80 Nubli et al. / Structural Integrity Procedia 00 (2023) 000 – 000

74 2

2016;2018;2020;2022; Smaradhana et al., 2021; Ansori et al., 2022; Do et al., 2022; Carvalho et al., 2023; Faqih et al., 2023; Pratama et al., 2023). In terms of infrastructure, refuelling site is one of the important facilities, and among the modes available for replenishing LNG fuel, such as truck-to-ship (TTS), pipeline-to-ship (PTS), and ship-to-ship (STS), the last mode demonstrates better performance due to its large fuel quantity provision, operational flexibility, and lack of interference with cargo handling operations (EMSA, 2018; Fevre, 2018; Nubli et al., 2022a; Tam, 2020). Despite the advantages of STS replenishment, handling LNG fuel poses higher risks than conventional marine diesel fuel due to its volatility, low flashpoint, and cryogenic temperature (Nubli and Sohn, 2022). An accidental release of LNG could lead to damage to the ship's structure and subsequent structural collapse due to the brittle fracture of steel material (Park et al., 2021). Therefore, selecting high-strength and cryogenic temperature-resistant steel is crucial to mitigate the potential hazards associated with cryogenic temperatures subjected to an LNG carrying ship or LNG bunkering ship, which is related to the ductile to the brittle transition temperature (DBTT) of the steel material. The use of liquefied natural gas (LNG) as a fuel in the maritime industry requires special consideration due to the hazards associated with the cryogenic temperatures involved. At temperatures as low as -163°C, LNG must be stored and transported in specialized tanks, which pose a risk of brittle fracture due to DBTT of the steel used in the tank construction (Muttaqie et al., 2020; Nubli et al., 2022b). This hazard is particularly acute for ships carrying LNG or LNG bunkering ships, which are at risk of accidental gas leaks that can damage the ship's structure and result in a significant loss (Paik, 2020). To mitigate this risk, it is essential to develop high-strength or high-manganese steels, cryogenic temperature-resistant steel that can withstand prolonged exposure to LNG flow and maintain its ductile properties at low temperatures (Muttaqie et al., 2020). The IGC code provides detailed guidance on materials for ships carrying liquefied gases, including specifications for plate thickness, design temperature, and usage (International Maritime Organization, 2014). Of particular concern is the LNG storage tank, which must be constructed with materials capable of withstanding cryogenic temperatures as low as -165°C. Options for this application include 9% nickel steel, high manganese austenitic steel, and aluminum alloys, with the choice of material depending on the specific requirements of the Type-C independent tank (International Maritime Organization, 2014; Muttaqie et al., 2020). The IGC code specifies that these materials must be able to withstand a minimum absorbed energy of 41 J in the Charpy V-Notch test (International Maritime Organization, 2014). For example, high manganese austenitic steel exhibits an ultimate tensile stress of up to 1,500 MPa and a fracture strain of up to 0.45 mm/mm at -163°C. Careful selection of materials can be a key factor in mitigating the cryogenic hazards associated with these structures. In this paper, the utilization of steel in shipbuilding is examined, primarily addressing the potential hazards related to unintentional LNG discharge during transit and the behavior of steel when exposed to low temperatures. Drawing from available research, the analysis is structured around three main areas: experimental investigations performed at cryogenic temperatures encompassing tensile testing and Charpy V-notch impact testing, as well as material modeling employing finite element analysis. The paper outlines an approach and methodology concerning these subjects. 2. Hazard of the Liquefied Natural Gas The transport of liquefied natural gas (LNG) presents unique safety challenges compared to other cargo types due to its classification as a high-flammability material. LNG can be easily vaporized at normal pressure and room temperature and is capable of spontaneous ignition (Nubli et al., 2022b). The Electronic Major Accident Report System (eMARS) classifies accidents by material type, facility location, and activity type, with a focus on LNG bunkering incidents (European Commission, 2021). Bhardwaj et al. (2018) analyzed several gas release cases in offshore structures and classified them according to severity level, ranging from low to high. Hydrocarbon gas was found to have the highest number of releases, accounting for 50.5% of the 321 total cases in the FPSO's accident database. Moreover, fluid-release events frequently occur during normal operation, resulting in moderate severity (Bhardwaj et al., 2018). Potential hazards resulting from LNG releases include asphyxiation, cryogenic burns, structural damage, and fire, with catastrophic vapor cloud explosions (VCEs) possible when an ignition source and gas accumulation coincide (Park et al., 2018). An assessment of risk is essential to ensure the safe transportation and handling of liquefied natural gas (LNG) due to its high flammability and the potential hazards associated with cryogenic temperatures. One of the significant risks in the LNG transportation industry is generic accidents, which include collision, grounding, contact, fire/explosion, and accidents during loading/unloading. According to the SAFEDOR risk model, these accidents are the primary

Haris Nubli et al. / Procedia Structural Integrity 48 (2023) 73–80 Nubli et al. / Structural Integrity Procedia 00 (2023) 000 – 000

75 3

contributors to risk, with high probabilities and severe consequences based on accident databases from 1964 to 2005 (Vanem et al., 2008). These generic accidents can occur on all ship types, but the risk of accident escalation is higher for LNG carriers due to cargo volatility. For instance, a collision on an LNG carrier that damages the cargo hold or LNG line may cause LNG leakage. Hence, it is crucial to prioritize risk mitigation measures in the design and operation of LNG carriers and implement effective safety management systems to prevent or minimize the consequences of potential accidents. In order to evaluate the hazards of cryogenic release on offshore or onshore units, Advanced Cryogenic Risk Analysis (ACRA) is suggested as a further risk analysis approach (Lloyd’s Register, 2015) . This method incorporates factors such as structures, equipment, barriers, and wind conditions into Computational Fluid Dynamics (CFD) simulation, as well as heat transfer calculation to estimate the DBTT of the exposed structure (Lloyd’s Register, 2015; Pujol et al., 2016). Figure 1 exhibits the example of CFD- based heat transfer analysis to the ship’s plate under cryogenic flow (Nubli et al., 2022a). The embrittlement of structural steel due to cryogenic exposure can weaken the steel's mechanical properties and potentially cause a structural collapse under an accidental load (Paik et al., 2020). Therefore, the temperature profile obtained from CFD simulation can be utilized as a load for Advanced Cryogenic Spill Protection Optimization (ACSPO). ACSPO considers a thermal-structural analysis simulated by the Nonlinear Finite Element Analysis (NLFEA) method and can be used to estimate the load capacity of a structure in the presence of cryogenic exposure (Lloyd’s Register, 2015) . In conclusion, the hazard of LNG is various which can result in various consequences. Particularly, the cryogenic temperature can weaken the structure. Thus, comp-rehensive structural and consequence analyses must be adopted in the preliminary design of the ships carrying LNG.

(a) (b) Fig. 1. Dispersion and heat transfer case of an accidental LNG release: (a) gas temperature, (b) steel temperature contours (Nubli et al., 2022a). 3. Cryogenic Tensile Test The use of carbon steel in the hull structures of ships that carry liquefied natural gas (LNG) can be problematic due to its susceptibility to the ductile-to-brittle transition temperature (DBTT) phenomenon. This type of steel can be prone to damage when exposed to cryogenic temperatures, such as those experienced by LNG, which can be as low as - 163°C. Hence, the use of higher-strength steel grades such as AH, DH, EH, and FH is recommended for these ships, governed by the IGC code. Recent studies have explored the effects of cryogenic temperatures on the mechanical properties of various steel grades. For example, full-scale collapse testing conducted by Paik et al. (2020) revealed the lower yield strength of AH32 in compression compared to tension under cryogenic temperatures. At sub-zero temperatures, steels exhibit a pronounced phenomenon known as hardening, which manifests as an increase in both yield and ultimate tensile strengths. Extensive research has been conducted to explore this behavior. For instance, Cho et al. (2014) conducted a tensile test on DH36 steel and observed an increase in yield and ultimate tensile strengths by 21.6% and 12.1%, respectively, when the temperature decreased from room temperature to -50oC. Similarly, Park et al. (2015) performed a study involving a tensile test on DH36 steel at temperatures ranging from room temperature to -60 oC. Their findings indicated significant enhancements in yield and ultimate tensile strengths, reaching up to 15.73% and 14.27%, respectively. Notably, both studies assumed quasi-static loading conditions,

Page 2 Page 3 Page 4 Page 5 Page 6 Page 7 Page 8 Page 9 Page 10 Page 11 Page 12 Page 13 Page 14 Page 15 Page 16 Page 17 Page 18 Page 19 Page 20 Page 21 Page 22 Page 23 Page 24 Page 25 Page 26 Page 27 Page 28 Page 29 Page 30 Page 31 Page 32 Page 33 Page 34 Page 35 Page 36 Page 37 Page 38 Page 39 Page 40 Page 41 Page 42 Page 43 Page 44 Page 45 Page 46 Page 47 Page 48 Page 49 Page 50 Page 51 Page 52 Page 53 Page 54 Page 55 Page 56 Page 57 Page 58 Page 59 Page 60 Page 61 Page 62 Page 63 Page 64 Page 65 Page 66 Page 67 Page 68 Page 69 Page 70 Page 71 Page 72 Page 73 Page 74 Page 75 Page 76 Page 77 Page 78 Page 79 Page 80 Page 81 Page 82 Page 83 Page 84 Page 85 Page 86 Page 87 Page 88 Page 89 Page 90 Page 91 Page 92 Page 93 Page 94 Page 95 Page 96 Page 97 Page 98 Page 99 Page 100 Page 101 Page 102 Page 103 Page 104 Page 105 Page 106 Page 107 Page 108 Page 109 Page 110 Page 111 Page 112 Page 113 Page 114 Page 115 Page 116 Page 117 Page 118 Page 119 Page 120 Page 121 Page 122 Page 123 Page 124 Page 125 Page 126 Page 127 Page 128 Page 129 Page 130 Page 131 Page 132 Page 133 Page 134 Page 135 Page 136 Page 137 Page 138 Page 139 Page 140 Page 141 Page 142 Page 143 Page 144 Page 145 Page 146 Page 147 Page 148 Page 149 Page 150 Page 151 Page 152 Page 153 Page 154 Page 155 Page 156 Page 157 Page 158 Page 159 Page 160 Page 161 Page 162 Page 163 Page 164 Page 165 Page 166 Page 167 Page 168 Page 169 Page 170 Page 171 Page 172 Page 173 Page 174 Page 175 Page 176 Page 177 Page 178 Page 179 Page 180 Page 181 Page 182 Page 183 Page 184 Page 185 Page 186 Page 187 Page 188 Page 189 Page 190 Page 191 Page 192 Page 193 Page 194 Page 195 Page 196 Page 197 Page 198 Page 199 Page 200

Made with FlippingBook Annual report maker