The implementation of reduced-order models in the development process of a rubber-metal component for an electric drive unit
DOI:
https://doi.org/10.14311/AP.2026.66.0253Keywords:
noise and vibrations, rubber-metal elements, innovative methodology, neural network, reduced order models, applicationAbstract
This thesis focuses on streamlining the development process of rubber-metal elements (RMEs) for mountelectric drive units (EDUs) in electric vehicles using reduced-order models (ROMs) and deep learning techniques. Traditional methods for developing these RME are both time-consuming and costly. Therefore, the aim of this work was to design and implement a more efficient approach utilising modern deep learning techniques. An innovative methodology was developed, which combines ROM with deep learning, enabling the creation of accurate and fast predictive models for RME behaviour without the need for extensive finite element (FE) analyses. The ROMs were developed and subsequently implemented into an application for predicting the static and dynamic characteristics of RMEs. The results demonstrated that this methodology significantly reduced the development time and improved simulation accuracy compared to conventional approaches.
Downloads
References
[1] K. Pietrzak, O. Pietrzak. Environmental effects of electromobility in a sustainable urban public transport. Sustainability 12(3):1052, 2020. https://doi.org/10.3390/su12031052
[2] P. Pavlínek. Transition of the automotive industry towards electric vehicle production in the east European integrated periphery. Empirica 50(1):35–73, 2023. https://doi.org/10.1007/s10663-022-09554-9
[3] A. Albatayneh, M. N. Assaf, D. Alterman, M. Jaradat. Comparison of the overall energy efficiency for internal combustion engine vehicles and electric vehicles. Environmental and Climate Technologies 24(1):669–680, 2020. https://doi.org/10.2478/rtuect-2020-0041
[4] A. Karki, S. Phuyal, D. Tuladhar, et al. Status of pure electric vehicle power train technology and future prospects. Applied System Innovation 3(3):35, 2020. https://doi.org/10.3390/asi3030035
[5] B. Heißing, M. Ersoy. Chassis Handbook: Fundamentals, driving dynamics, components, mechatronics, perspectives. Springer, 2011. ISBN 978-3-8348-0994-0. https://doi.org/10.1007/978-3-8348-9789-3
[6] W. S. Lee, S.-K. Youn, B.-K. Kim. Finite element analysis and design optimization of rubber components for vibration isolation. Archives of Mechanics 55(5–6):449–479, 2003. [2025-10-23]. https://am.ippt.pan.pl/index.php/am/article/view/v55p449
[7] C. Cernuda, I. Llavori, A.-C. Zăvoianu, et al. Critical analysis of the suitability of surrogate models for finite element method application in catalog-based suspension bushing design. In 2020 25th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA), vol. 1, pp. 829–836. 2020. https://doi.org/10.1109/ETFA46521.2020.9212166
[8] S. Fresca, G. Gobat, P. Fedeli, et al. Deep learning-based reduced order models for the real-time simulation of the nonlinear dynamics of microstructures. International Journal for Numerical Methods in Engineering 123(20):4749–4777, 2022. https://doi.org/10.1002/nme.7054
[9] J. Dobrovolný. Návrh inovatívnej metodiky pre vývoj gumokovového prvku uloženia elektrickej hnacej sústavy [In Slovak; Design of an innovative methodology for the development of the rubber-metal mount of the electric powertrain]. Master’s thesis, Slovak University of Technology in Bratislava, Faculty of Mechanical Engineering, 2024. [2025-24-10]. https://is.stuba.sk/zp/portal_zp.pl?podrobnosti=177316
[10] J. Danko, J. Dobrovolný, Ľ. Magdolen, et al. Parametric generation of datasets for the creation of reduced order models of rubber-metal elements. Strojnícky časopis – Journal of Mechanical Engineering 75(1):1–8, 2025. https://doi.org/10.2478/scjme-2025-0001
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Ján Danko, Juraj Dobrovolný, Tomáš Milesich, Igor Kevický
This work is licensed under a Creative Commons Attribution 4.0 International License.
