Numerical simulation data underlying the scientific paper Surrogate model for FEA analysis and damage calculation used for exhaust system validation

Zaletel, Jan (author), Oman, Simon (author), Nagode, Marko (author), Klemenc, Jernej (author)

The research data consist of the training and validation datasets and the source code used for building the surrogate model. They include both input and output data as well as surrogate model parameters. The data were generated for the purpose of training and validating a surrogate model (neural network) for predicting the fatigue life of an exhaust system. The goal of the methodology using a surrogate model is to replace time consuming finite element modeling and fatigue life calculations with a computationally simpler approach for determining the location and magnitude of the maximum damage.

• Language: English
• Keywords: material fatigue, damage, surrogate model, computational modeling, neural network, design, optimization
• Typology: 2.20 - Complete scientific database of research data
• Geographic coverage: Slovenija
• Time coverage: Od/From 2025-12-15 do/till 2025-03-26
• Organization: FS - Faculty of Mechanical Engineering
• Year: 2026
• PID: 20.500.12556/RUL-181332
• Data col. methods: Synthesis; Simulation
• Publication date in RUL: 01.04.2026

Licences

• License: CC BY 4.0, Creative Commons Attribution 4.0 International
• Link: http://creativecommons.org/licenses/by/4.0/
• Description: This is the standard Creative Commons license that gives others maximum freedom to do what they want with the work as long as they credit the author.

Secondary language

• Language: Slovenian
• Title: Podatki numeričnih simulacij, obravnavani v članku Surrogate model for FEA analysis and damage calculation used for exhaust system validation

Abstract

Raziskovalni podatki so zbirka učnih in validacijskih baz podatkov ter programske kode, uporabljenih pri grajenju nadomestnega (angl. surrogate) modela. Zajemajo tako vhodne kot izhodne podatke in parametre. Podatki so bili generirani z namenom učenja in validiranja nadomestnega modela (nevronska mreža) za napoved dobe trajanja izpušnega sistema. Cilj metodologije z uporabo nadomestnega modela je zamenjati časovno potratno modeliranje s končnimi elementi in dolgotrajen izračun dobe trajanja z računsko enostavnejšim pristopom za določitev lokacije in velikosti največje poškodbe.

• Keywords: utrujanje materiala, poškodba, nadomestni model, računalniško modeliranje, nevronska mreža, konstruiranje, optimizacija

Projects

• Funder: ARIS - Slovenian Research and Innovation Agency
• Project number: P2-0182
• Name: Razvojna vrednotenja