MétaCan
Menu
Back to cohort
Record W4416174720 · doi:10.1016/j.jmapro.2025.10.104

Real-time and data-efficient springback prediction in tube bending using force measurement and physics-informed machine learning

2025· article· en· W4416174720 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

fundA Canadian funder is recorded on the work.
no affNo Canadian affiliation: this work is invisible to an affiliation-only frame.
No Canadian affiliation. An affiliation-only frame, the usual design, would never have seen this work. It is one of the works that make the case for inverting the frame.

Bibliographic record

VenueJournal of Manufacturing Processes · 2025
Typearticle
Languageen
FieldPhysics and Astronomy
TopicModel Reduction and Neural Networks
Canadian institutionsnot available
FundersNorges ForskningsrådAlexander Family FoundationNorges Teknisk-Naturvitenskapelige Universitet
KeywordsBendingDisplacement (psychology)Artificial neural networkConvolutional neural networkStability (learning theory)Deformation (meteorology)Tube (container)

Abstract

fetched live from OpenAlex

Tube bending is widely used in manufacturing structural components with complex curvatures and angles. However, springback is a major issue affecting the dimensional accuracy of components and production efficiency. Accurately predicting and controlling springback in tube bending is a critical challenge, as traditional analytical models, while useful, struggle with deformation complexities and manufacturing uncertainties, limiting their prediction accuracy. Meanwhile, machine learning models show great potential for addressing these limitations but often require large datasets, posing a challenge in data-limited industrial applications. In this research, we develop a physics-informed neural network (PINN) that integrates an analytical force-springback relationship into a convolutional neural network (CNN) framework. Firstly, a validated force-monitoring system is used to capture real-time force data during tube bending, while springback is measured using a soft sensor that combines displacement measurements with analytical calculation. Furthermore, an attention-based mechanism is applied to enhance the model's explainability, identifying the most influential forces and time steps in predicting springback. The PINN model is evaluated from three aspects, i.e., the overall predictive performance and explainability, the ability to learn effectively with limited training data, and the feasibility for real-time springback prediction utilizing truncated force sequences. The results demonstrate that PINN successfully predicts springback and adapts to variations in material properties, achieving comparable accuracy to CNN. Additionally, when trained with fewer samples, PINN benefits from the embedded physical constraints, thus exhibiting faster convergence, improved stability and more consistent predictions. These findings indicates that PINN provides a data-efficient and reliable approach for springback prediction, with promising applications in real-time monitoring, explainable AI, and closed-loop process control for tube bending. • Real-time springback prediction using physics-informed neural networks. • In-process monitoring system integrates real-time force data and soft sensor measurement. • Embedding analytical constraints enhance model stability and accuracy. • Channel-wise attention identifies critical forces, improving interpretability. • Early-stage prediction supports inline correction in tube bending.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.253
Threshold uncertainty score0.420

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.039
GPT teacher head0.287
Teacher spread0.248 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it