Physics-Informed Neural Networks for Structural Mechanics and Construction: Modeling the Deflection of a Single-Span Beam Physics-Informed Neural Networks

This article presents the development and analysis of a Physics-Informed Neural Network (PINN) model for calculating the deflection of a simply supported beam under uniformly distributed load. The training dataset was synthesized based on analytical principles of structural mechanics and included th...

Full description

Saved in:
Bibliographic Details
Main Authors: F. N. Zakharov, Qian Jie, Xu Yi
Format: Article
Language:English
Published: Moscow State University of Civil Engineering (MGSU) 2025-03-01
Series:Железобетонные конструкции
Subjects:
Online Access:https://www.g-b-k.ru/jour/article/view/69
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849434703035105280
author F. N. Zakharov
Qian Jie
Xu Yi
author_facet F. N. Zakharov
Qian Jie
Xu Yi
author_sort F. N. Zakharov
collection DOAJ
description This article presents the development and analysis of a Physics-Informed Neural Network (PINN) model for calculating the deflection of a simply supported beam under uniformly distributed load. The training dataset was synthesized based on analytical principles of structural mechanics and included the following parameters: relative length of the measurement section l, number of measurement points N, and noise level R. The training dataset consisted of 1,296 rows describing random points within the beam span. In this study, 480 PINN models were trained to evaluate the impact of the weight of the physics-informed loss function, the number of measurements, and the noise level on prediction accuracy. The results demonstrated that PINN models achieve high accuracy (R2 ≥ 0.88) even with high noise levels (R > 20 %) and exhibit robustness to low and moderate noise levels. The study identified that adjusting the weight of the physics-informed loss function is a key parameter for achieving an optimal balance between the loss functions of physical laws and experimental data. Increasing the number of measurement points positively influences accuracy at low noise levels. However, an increase in the number of measurement points under high noise levels reduces the prediction accuracy of the model. The scientific novelty of the study lies in proposing an approach for structural analysis using PINN, which integrates physical laws into the training process. The findings confirm the potential of using PINN for engineering calculations, particularly under limited data conditions.
format Article
id doaj-art-b1abc37b1faa4f59b0dd6a4cf089ec06
institution Kabale University
issn 2949-1622
2949-1614
language English
publishDate 2025-03-01
publisher Moscow State University of Civil Engineering (MGSU)
record_format Article
series Железобетонные конструкции
spelling doaj-art-b1abc37b1faa4f59b0dd6a4cf089ec062025-08-20T03:26:34ZengMoscow State University of Civil Engineering (MGSU)Железобетонные конструкции2949-16222949-16142025-03-0191354810.22227/2949-1622.2025.1.35-4863Physics-Informed Neural Networks for Structural Mechanics and Construction: Modeling the Deflection of a Single-Span Beam Physics-Informed Neural NetworksF. N. Zakharov0Qian Jie1Xu Yi2Bluetown Leju Construction Management Co. Ltd. Research InstituteBluetown Leju Construction Management Co. Ltd. Research InstituteZhejiang Province Architectural Design and Research InstituteThis article presents the development and analysis of a Physics-Informed Neural Network (PINN) model for calculating the deflection of a simply supported beam under uniformly distributed load. The training dataset was synthesized based on analytical principles of structural mechanics and included the following parameters: relative length of the measurement section l, number of measurement points N, and noise level R. The training dataset consisted of 1,296 rows describing random points within the beam span. In this study, 480 PINN models were trained to evaluate the impact of the weight of the physics-informed loss function, the number of measurements, and the noise level on prediction accuracy. The results demonstrated that PINN models achieve high accuracy (R2 ≥ 0.88) even with high noise levels (R > 20 %) and exhibit robustness to low and moderate noise levels. The study identified that adjusting the weight of the physics-informed loss function is a key parameter for achieving an optimal balance between the loss functions of physical laws and experimental data. Increasing the number of measurement points positively influences accuracy at low noise levels. However, an increase in the number of measurement points under high noise levels reduces the prediction accuracy of the model. The scientific novelty of the study lies in proposing an approach for structural analysis using PINN, which integrates physical laws into the training process. The findings confirm the potential of using PINN for engineering calculations, particularly under limited data conditions.https://www.g-b-k.ru/jour/article/view/69physics-informed neural networksstructural mechanicslimited dataartificial intelligencebeam calculation
spellingShingle F. N. Zakharov
Qian Jie
Xu Yi
Physics-Informed Neural Networks for Structural Mechanics and Construction: Modeling the Deflection of a Single-Span Beam Physics-Informed Neural Networks
Железобетонные конструкции
physics-informed neural networks
structural mechanics
limited data
artificial intelligence
beam calculation
title Physics-Informed Neural Networks for Structural Mechanics and Construction: Modeling the Deflection of a Single-Span Beam Physics-Informed Neural Networks
title_full Physics-Informed Neural Networks for Structural Mechanics and Construction: Modeling the Deflection of a Single-Span Beam Physics-Informed Neural Networks
title_fullStr Physics-Informed Neural Networks for Structural Mechanics and Construction: Modeling the Deflection of a Single-Span Beam Physics-Informed Neural Networks
title_full_unstemmed Physics-Informed Neural Networks for Structural Mechanics and Construction: Modeling the Deflection of a Single-Span Beam Physics-Informed Neural Networks
title_short Physics-Informed Neural Networks for Structural Mechanics and Construction: Modeling the Deflection of a Single-Span Beam Physics-Informed Neural Networks
title_sort physics informed neural networks for structural mechanics and construction modeling the deflection of a single span beam physics informed neural networks
topic physics-informed neural networks
structural mechanics
limited data
artificial intelligence
beam calculation
url https://www.g-b-k.ru/jour/article/view/69
work_keys_str_mv AT fnzakharov physicsinformedneuralnetworksforstructuralmechanicsandconstructionmodelingthedeflectionofasinglespanbeamphysicsinformedneuralnetworks
AT qianjie physicsinformedneuralnetworksforstructuralmechanicsandconstructionmodelingthedeflectionofasinglespanbeamphysicsinformedneuralnetworks
AT xuyi physicsinformedneuralnetworksforstructuralmechanicsandconstructionmodelingthedeflectionofasinglespanbeamphysicsinformedneuralnetworks