A Physics–Data Hybrid Framework for Friction Interface Temperature-Field Prediction Based on Experimentally Corrected Finite Element Analysis

Accurate characterization of the friction interface temperature field is essential for assessing thermal conditions and ensuring the reliable operation of wet friction components. However, direct acquisition of the complete transient temperature field remains difficult under multiple operating conditions. To address this limitation, a physics–data hybrid framework based on experimentally corrected finite element data is proposed. Experimental temperature measurements are used to validate and spatially correct the finite element method (FEM)-based temperature field, yielding an experimentally constrained complete-field dataset. The proper orthogonal decomposition (POD) method extracts the dominant spatial modes, while the radial basis function (RBF) model maps pressure, rotational speed, and sliding duration to the modal coefficients, enabling rapid reconstruction under unseen operating conditions. The POD-RBF model achieves a coefficient of determination (R2) of 0.956, with a mean absolute error (MAE) of 3.301 °C and a root mean square error (RMSE) of 7.380 °C. Sensitivity analysis indicates that prediction accuracy is most sensitive to the RBF shape parameter, followed by the smoothing parameter and the number of retained POD modes; among the operating-condition parameters, sliding duration has the greatest influence. Further analysis shows that increasing pressure and rotational speed intensifies both the temperature rise and spatial temperature nonuniformity at the friction interface. The proposed framework provides an effective approach for rapid prediction of complete temperature fields under multiple operating conditions.

Authors

Institutions

Publication Details

Journal
Lubricants
Published
2026-11-18
DOI
https://doi.org/10.3390/lubricants14110439
Primary Topic
Model Reduction and Neural Networks
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A Physics–Data Hybrid Framework for Friction Interface Temperature-Field Prediction Based on Experimentally Corrected Finite Element Analysis

Jianpeng Wu, Nikolaos Rogkas, Ao Ding, Xinchao Lin et al.
5 links
Lubricants
Model Reduction and Neural Networks
article

A Physics–Data Hybrid Framework for Friction Interface Temperature-Field Prediction Based on Experimentally Corrected Finite Element Analysis

Jianpeng Wu, Nikolaos Rogkas, Ao Ding, Xinchao Lin, Xingyuan Jia
article en

Abstract

Accurate characterization of the friction interface temperature field is essential for assessing thermal conditions and ensuring the reliable operation of wet friction components. However, direct acquisition of the complete transient temperature field remains difficult under multiple operating conditions. To address this limitation, a physics–data hybrid framework based on experimentally corrected finite element data is proposed. Experimental temperature measurements are used to validate and spatially correct the finite element method (FEM)-based temperature field, yielding an experimentally constrained complete-field dataset. The proper orthogonal decomposition (POD) method extracts the dominant spatial modes, while the radial basis function (RBF) model maps pressure, rotational speed, and sliding duration to the modal coefficients, enabling rapid reconstruction under unseen operating conditions. The POD-RBF model achieves a coefficient of determination (R2) of 0.956, with a mean absolute error (MAE) of 3.301 °C and a root mean square error (RMSE) of 7.380 °C. Sensitivity analysis indicates that prediction accuracy is most sensitive to the RBF shape parameter, followed by the smoothing parameter and the number of retained POD modes; among the operating-condition parameters, sliding duration has the greatest influence. Further analysis shows that increasing pressure and rotational speed intensifies both the temperature rise and spatial temperature nonuniformity at the friction interface. The proposed framework provides an effective approach for rapid prediction of complete temperature fields under multiple operating conditions.

LubricantsVol. 14(11)
National Technical University of Athens (GR), Centre for Research and Technology Hellas (GR), Beijing Information Science & Technology University (CN)
National Natural Science Foundation of China
Model Reduction and Neural Networks
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon