0
RESEARCH PAPERS

Incorporating Neural Network Material Models Within Finite Element Analysis for Rheological Behavior Prediction

[+] Author and Article Information
B. Scott Kessler, A. Sherif El-Gizawy, Douglas E. Smith

 Kestek, 4602 W 121st Street, Suite 111, Overland Park, KS 66209

J. Pressure Vessel Technol 129(1), 58-65 (Feb 25, 2006) (8 pages) doi:10.1115/1.2389004 History: Received September 12, 2005; Revised February 25, 2006

The accuracy of a finite element model for design and analysis of a metal forging operation is limited by the incorporated material model’s ability to predict deformation behavior over a wide range of operating conditions. Current rheological models prove deficient in several respects due to the difficulty in establishing complicated relations between many parameters. More recently, artificial neural networks (ANN) have been suggested as an effective means to overcome these difficulties. To this end, a robust ANN with the ability to determine flow stresses based on strain, strain rate, and temperature is developed and linked with finite element code. Comparisons of this novel method with conventional means are carried out to demonstrate the advantages of this approach.

Copyright © 2007 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

(a) Strain flattening and (b) strain softening behavior 2

Grahic Jump Location
Figure 2

Schematic of a simple artificial neural network architecture

Grahic Jump Location
Figure 3

Hardlim transfer function within MATLAB, where W is the weight(s), p is the input(s), and b is the bias

Grahic Jump Location
Figure 4

Linear transfer function, single-input purelin neuron

Grahic Jump Location
Figure 5

Tan-sigmoid transfer function, single-input tansig neuron

Grahic Jump Location
Figure 6

Example demonstrating overfitting

Grahic Jump Location
Figure 7

6061 aluminum flow stress as a function of strain for 300°C

Grahic Jump Location
Figure 8

6061 aluminum flow stress as a function of strain for 550°C

Grahic Jump Location
Figure 9

Curve fits for all strains at 450°C

Grahic Jump Location
Figure 10

50 neuron LM network output using data set A for 300°C

Grahic Jump Location
Figure 11

50 neuron LM network output using data set A for 550°C

Grahic Jump Location
Figure 12

50 neuron LM network output using data set A for 300°C at intermediate strain rate values

Grahic Jump Location
Figure 13

50 neuron LM network output using data set A for 550°C at intermediate strain rate values

Grahic Jump Location
Figure 14

8-3 BR neuron network output using data set B for 300°C

Grahic Jump Location
Figure 15

8-3 BR neuron network output using data set B for 550°C

Grahic Jump Location
Figure 16

8-3 BR neuron network output using data set B for 300°C at intermediate strain rate values

Grahic Jump Location
Figure 17

8-3 BR neuron network output using data set B for 550°C at intermediate strain rate values

Grahic Jump Location
Figure 18

8-3 BR neuron network output using data set B for 375°C at intermediate strain rates

Grahic Jump Location
Figure 19

8-3 BR neuron network output using data set B for 450°C at intermediate strain rates

Grahic Jump Location
Figure 20

15-3 BR neuron network output using data set C for 300°C

Grahic Jump Location
Figure 21

15-3 BR neuron network output using data set C for 550°C

Grahic Jump Location
Figure 22

Linear regression for the 15-3 BR neuron network

Grahic Jump Location
Figure 23

15-3 BR neuron network output using data set C for 300°C at intermediate strain rates

Grahic Jump Location
Figure 24

15-3 BR neuron network output using data set C for 550°C at intermediate strain rates

Grahic Jump Location
Figure 25

15-3 BR neuron network output using data set C for 375°C

Grahic Jump Location
Figure 26

15-3 BR neuron network output using data set C for 450°C

Grahic Jump Location
Figure 27

Compression at 450°C for power law model compared with experimental data

Grahic Jump Location
Figure 28

Compression at 450°C for power law model compared with experimental data

Grahic Jump Location
Figure 29

Compression at 450°C for the ANN model compared with experimental data

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In