Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Predicting Wall Thickness Changes in Cold Forging Processes: An Integrated FEM and Neural Network approach

Sasa Ilic, Abdulkerim Karaman, Johannes Pöppelbaum, Jan Niclas Reimann, Michael Marré, Andreas Schwung

TL;DR

The neural network architecture is extended by directly incorporating information about the nosing process by adding different types of edges and their corresponding encoders to model object interactions and enhances model accuracy and opens the possibility for employing precise surrogate models within closed-loop production processes.

Abstract

This study presents a novel approach for predicting wall thickness changes in tubes during the nosing process. Specifically, we first provide a thorough analysis of nosing processes and the influencing parameters. We further set-up a Finite Element Method (FEM) simulation to better analyse the effects of varying process parameters. As however traditional FEM simulations, while accurate, are time-consuming and computationally intensive, which renders them inapplicable for real-time application, we present a novel modeling framework based on specifically designed graph neural networks as surrogate models. To this end, we extend the neural network architecture by directly incorporating information about the nosing process by adding different types of edges and their corresponding encoders to model object interactions. This augmentation enhances model accuracy and opens the possibility for employing precise surrogate models within closed-loop production processes. The proposed approach is evaluated using a new evaluation metric termed area between thickness curves (ABTC). The results demonstrate promising performance and highlight the potential of neural networks as surrogate models in predicting wall thickness changes during nosing forging processes.

Predicting Wall Thickness Changes in Cold Forging Processes: An Integrated FEM and Neural Network approach

TL;DR

The neural network architecture is extended by directly incorporating information about the nosing process by adding different types of edges and their corresponding encoders to model object interactions and enhances model accuracy and opens the possibility for employing precise surrogate models within closed-loop production processes.

Abstract

This study presents a novel approach for predicting wall thickness changes in tubes during the nosing process. Specifically, we first provide a thorough analysis of nosing processes and the influencing parameters. We further set-up a Finite Element Method (FEM) simulation to better analyse the effects of varying process parameters. As however traditional FEM simulations, while accurate, are time-consuming and computationally intensive, which renders them inapplicable for real-time application, we present a novel modeling framework based on specifically designed graph neural networks as surrogate models. To this end, we extend the neural network architecture by directly incorporating information about the nosing process by adding different types of edges and their corresponding encoders to model object interactions. This augmentation enhances model accuracy and opens the possibility for employing precise surrogate models within closed-loop production processes. The proposed approach is evaluated using a new evaluation metric termed area between thickness curves (ABTC). The results demonstrate promising performance and highlight the potential of neural networks as surrogate models in predicting wall thickness changes during nosing forging processes.

Paper Structure

This paper contains 33 sections, 21 equations, 22 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Nosing Processes
  4. Finite Element Surrogate Models
  5. Analysis of Nosing Processes
  6. Description of the nosing process
  7. Process limits
  8. Dimensional accuracy
  9. Wall thickness change during nosing
  10. Finite element models
  11. Model description
  12. Mesh convergence study
  13. Parameters
  14. Graph Neural Network Architecture to approximate FEM Simulations
  15. Network Requirements and challenges
  16. ...and 18 more sections

Figures (22)

  • Figure 1: Approach for creating the AI Model.
  • Figure 2: Nosing of tubes without a Mandrel – principle and key parameters based on Handbuchumformen.
  • Figure 3: Process limits. Picture left: Nosed tube with non-deformed shape (maximum equivalent stress: 485 MPa). Picture right: Nosed tube with cross folds caused by excessive axial compressive stresses (maximum equivalent stress: 951 MPa), which lead to initial buckling, total failure due to buckling, and a deformation area exhibiting significantly increased equivalent stress.
  • Figure 4: Independent formation of a bending radius $r$Haa83. Tube coloured in red, die coloured in blue.
  • Figure 5: Sizes of the dimensional deviation after forging Alb90.
  • ...and 17 more figures