Physics-Based AI-Driven Surrogate Modeling for Structural Displacement Prediction in Mechanical Systems With Limited Sensor Data
This study introduces a machine learning (ML)-based surrogate model for finite element analysis, designed to predict structural strain distributions using a minimal number of strategically placed virtual sensors. The proposed approach eliminates the dependency on external force measurements, leverag...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11084809/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849245879313104896 |
|---|---|
| author | Ali Hashemi Javad Beheshti Mahdieh Mohammadi |
| author_facet | Ali Hashemi Javad Beheshti Mahdieh Mohammadi |
| author_sort | Ali Hashemi |
| collection | DOAJ |
| description | This study introduces a machine learning (ML)-based surrogate model for finite element analysis, designed to predict structural strain distributions using a minimal number of strategically placed virtual sensors. The proposed approach eliminates the dependency on external force measurements, leveraging local strain measurements to infer global strain responses across a two-dimensional truss structure. Various regression algorithms, including decision trees and deep neural networks (DNNs), were evaluated, with DNNs achieving superior accuracy (<inline-formula> <tex-math notation="LaTeX">$\text {R}^{2}$ </tex-math></inline-formula> > 0.996, MAE < 4%, RMSE < 5.5%). The methodology significantly reduces sensor requirements and computational overhead, offering a practical, scalable solution for structural health monitoring (SHM) in complex mechanical systems. The results underscore the potential of ML-based surrogate models to enhance the efficiency and accuracy of continuous monitoring and dynamic analysis in large-scale infrastructure applications, setting the stage for future advancements in sensor optimization and three-dimensional system extensions. |
| format | Article |
| id | doaj-art-beb8e083f7954406a3c7d54677fffe64 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-beb8e083f7954406a3c7d54677fffe642025-08-20T03:58:40ZengIEEEIEEE Access2169-35362025-01-011313058513060210.1109/ACCESS.2025.359066411084809Physics-Based AI-Driven Surrogate Modeling for Structural Displacement Prediction in Mechanical Systems With Limited Sensor DataAli Hashemi0https://orcid.org/0000-0003-1115-5393Javad Beheshti1https://orcid.org/0009-0006-0600-766XMahdieh Mohammadi2Department of Civil, Environmental and Geomatics Engineering, Florida Atlantic University, Boca Raton, FL, USAIT and Computer Engineering Department, Islamic Azad University, Tafresh, IranFaculty of Built Environment, University of Malaya, Kuala Lumpur, MalaysiaThis study introduces a machine learning (ML)-based surrogate model for finite element analysis, designed to predict structural strain distributions using a minimal number of strategically placed virtual sensors. The proposed approach eliminates the dependency on external force measurements, leveraging local strain measurements to infer global strain responses across a two-dimensional truss structure. Various regression algorithms, including decision trees and deep neural networks (DNNs), were evaluated, with DNNs achieving superior accuracy (<inline-formula> <tex-math notation="LaTeX">$\text {R}^{2}$ </tex-math></inline-formula> > 0.996, MAE < 4%, RMSE < 5.5%). The methodology significantly reduces sensor requirements and computational overhead, offering a practical, scalable solution for structural health monitoring (SHM) in complex mechanical systems. The results underscore the potential of ML-based surrogate models to enhance the efficiency and accuracy of continuous monitoring and dynamic analysis in large-scale infrastructure applications, setting the stage for future advancements in sensor optimization and three-dimensional system extensions.https://ieeexplore.ieee.org/document/11084809/Machine learningsurrogate modelsfinite element methodsstructural dynamicsvirtual sensor placementartificial neural networks |
| spellingShingle | Ali Hashemi Javad Beheshti Mahdieh Mohammadi Physics-Based AI-Driven Surrogate Modeling for Structural Displacement Prediction in Mechanical Systems With Limited Sensor Data IEEE Access Machine learning surrogate models finite element methods structural dynamics virtual sensor placement artificial neural networks |
| title | Physics-Based AI-Driven Surrogate Modeling for Structural Displacement Prediction in Mechanical Systems With Limited Sensor Data |
| title_full | Physics-Based AI-Driven Surrogate Modeling for Structural Displacement Prediction in Mechanical Systems With Limited Sensor Data |
| title_fullStr | Physics-Based AI-Driven Surrogate Modeling for Structural Displacement Prediction in Mechanical Systems With Limited Sensor Data |
| title_full_unstemmed | Physics-Based AI-Driven Surrogate Modeling for Structural Displacement Prediction in Mechanical Systems With Limited Sensor Data |
| title_short | Physics-Based AI-Driven Surrogate Modeling for Structural Displacement Prediction in Mechanical Systems With Limited Sensor Data |
| title_sort | physics based ai driven surrogate modeling for structural displacement prediction in mechanical systems with limited sensor data |
| topic | Machine learning surrogate models finite element methods structural dynamics virtual sensor placement artificial neural networks |
| url | https://ieeexplore.ieee.org/document/11084809/ |
| work_keys_str_mv | AT alihashemi physicsbasedaidrivensurrogatemodelingforstructuraldisplacementpredictioninmechanicalsystemswithlimitedsensordata AT javadbeheshti physicsbasedaidrivensurrogatemodelingforstructuraldisplacementpredictioninmechanicalsystemswithlimitedsensordata AT mahdiehmohammadi physicsbasedaidrivensurrogatemodelingforstructuraldisplacementpredictioninmechanicalsystemswithlimitedsensordata |