PSI - Issue 59

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 59 (2024) 444–451

© 2024 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 DMDP 2023 Organizers Abstract Some production losses at the enterprise associated with the provision of a complex of technical maintenance and repair operations for the units, systems, and assemblies of automobiles have been discussed in the article under consideration. To enhance the accuracy of defect recognition and prediction, the method of artificial neural networks has been used. It involves reducing the training time of the method and the need for data normalization and scaling using the sigmoid function as the activation function in the algorithm block diagram of the artificial neural network method within its operational range. Based on the theoretical work of Hecht-Nielsen, it has beenfound that the use of more than two hidden layers of neurons in the artificial neural network method is inexpedient. On this basis, the upper limit of the neurons number, the error evaluation function to be minimized, and the error backward propagation method have been defined. Their distribution based on the complex of technical maintenance and repair operations uses the previous algorithm with a newly formed database. © 2024 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 DMDP 2023 Organizers Keywords: Reliability, Surface strength, Wear, Tribo-coupling, System analysis, Self-organization Abstract Some production losses at the enterprise associated with the provision of a complex of technical maintenance and repair operations for the units, systems, and assemblies of automobiles have been discussed in the article under consideration. To enhance the accuracy of defect recognition and prediction, the method of artificial neural networks has been used. It involves reducing the training time of the method and the need for data normalization and scaling using the sigmoid function as the activation function in the algorithm block diagram of the artificial neural network method within its operational range. Based on the theoretical work of Hecht-Nielsen, it has beenfound that the use of more than two hidden layers of neurons in the artificial neural network method is inexpedient. On this basis, the upper limit of the neurons number, the error evaluation function to be minimized, and the error backward propagation method have been defined. Their distribution based on the complex of technical maintenance and repair operations uses the previous algorithm with a newly formed database. © 2024 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 DMDP 2023 Organizers Keywords: Reliability, Surface strength, Wear, Tribo-coupling, System analysis, Self-organization VII International Conference “In -service Damage of Materials: Diagnostics and Prediction ” (DMDP 2023) Prediction of recognized defect combinations in the parts of automobile units, systems, and assemblies using artificial neural network method Victor Aulin а , *, Mykola Mytnyk b , Andrii Hrynkiv a , Artem Holovatyi a , Sergii Lysenko a , Uliana Plekan b VII International Conference “In -service Damage of Materials: Diagnostics and Prediction ” (DMDP 2023) Prediction of recognized defect combinations in the parts of automobile units, systems, and assemblies using artificial neural network method Victor Aulin а , *, Mykola Mytnyk b , Andrii Hrynkiv a , Artem Holovatyi a , Sergii Lysenko a , Uliana Plekan b а Central Ukrainian National Technical University, 25006 Kropyvnytskiy, Ukraine b Ternopil Ivan Puluj National Technical University, 46001 Ternopil, Ukraine а Central Ukrainian National Technical University, 25006 Kropyvnytskiy, Ukraine b Ternopil Ivan Puluj National Technical University, 46001 Ternopil, Ukraine

* Corresponding author. Tel.: +380988997104. E-mail address: Aulinvv@gmail.com * Corresponding author. Tel.: +380988997104. E-mail address: Aulinvv@gmail.com

2452-3216 © 2024 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 DMDP 2023 Organizers 2452-3216 © 2024 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 DMDP 2023 Organizers

2452-3216 © 2024 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 DMDP 2023 Organizers 10.1016/j.prostr.2024.04.063