New probabilistic fracture mechanics approach with neural networkbased crack modeling: Its application to multiple cracks problem
Abstract
Studies on efficient utilization and life extension of operating nuclear power plants (NPPs) have become increasingly important since ages of the firstgeneration NPPs are approaching their design lives. In order to predict a remaining life of each plant, it is necessary to select those critical components that strongly influence the plant life, and to evaluate their remaining lives by considering aging effects of materials and other factors. This paper proposes a new method to incorporate sophisticated crack models, such as interaction and coalescence of multiple surface cracks, into probabilistic fracture mechanism (PFM) computer programs using neural networks. First, hundreds of finite element (FE) calculations of a plate containing multiple surface cracks are performed by parametrically changing crack parameters such as sizes and locations. A fully automated 3D FE analysis system is effectively utilized here. Second, the backpropagation neural network is trained using the FE solutions, i.e. crack parameters vs. their corresponding stress intensity factors (SIFs). After a sufficient number of training iterations, the network attains an ability to promptly output SIFs for arbitrary combinations of crack parameters. The well trained network is then incorporated into the parallel PFM program which runs on one of massively parallel computers composed of 512more »
 Authors:

 Univ. of Tokyo (Japan)
 Science Univ. of Tokyo (Japan)
 Publication Date:
 OSTI Identifier:
 122630
 Report Number(s):
 CONF950740
ISBN 0791813355; TRN: 95:024247
 Resource Type:
 Conference
 Resource Relation:
 Conference: Joint American Society of Mechanical Engineers (ASME)/Japan Society of Mechanical Engineers (JSME) pressure vessels and piping conference, Honolulu, HI (United States), 2327 Jul 1995; Other Information: PBD: 1995; Related Information: Is Part Of Fatigue and fracture mechanics in pressure vessels and piping. PVPVolume 304; Mehta, H.S.; Wilkowski, G.; Takezono, S.; Bloom, J.; Yoon, K.; Aoki, S.; Rahman, S.; Nakamura, T.; Brust, F.; Yoshimura, S. [eds.]; PB: 594 p.
 Country of Publication:
 United States
 Language:
 English
 Subject:
 22 NUCLEAR REACTOR TECHNOLOGY; 99 MATHEMATICS, COMPUTERS, INFORMATION SCIENCE, MANAGEMENT, LAW, MISCELLANEOUS; FRACTURE MECHANICS; NEURAL NETWORKS; NUCLEAR POWER PLANTS; PRESSURE VESSELS; SYSTEMS ANALYSIS; PARAMETRIC ANALYSIS; THREEDIMENSIONAL CALCULATIONS; CRACK PROPAGATION; PARALLEL PROCESSING; SERVICE LIFE; COMPUTER CODES
Citation Formats
Yoshimura, Shinobu, Lee, J S, Yagawa, Genki, Sugioka, Kiyoshi, and Kawai, Tadahiko. New probabilistic fracture mechanics approach with neural networkbased crack modeling: Its application to multiple cracks problem. United States: N. p., 1995.
Web.
Yoshimura, Shinobu, Lee, J S, Yagawa, Genki, Sugioka, Kiyoshi, & Kawai, Tadahiko. New probabilistic fracture mechanics approach with neural networkbased crack modeling: Its application to multiple cracks problem. United States.
Yoshimura, Shinobu, Lee, J S, Yagawa, Genki, Sugioka, Kiyoshi, and Kawai, Tadahiko. Wed .
"New probabilistic fracture mechanics approach with neural networkbased crack modeling: Its application to multiple cracks problem". United States.
@article{osti_122630,
title = {New probabilistic fracture mechanics approach with neural networkbased crack modeling: Its application to multiple cracks problem},
author = {Yoshimura, Shinobu and Lee, J S and Yagawa, Genki and Sugioka, Kiyoshi and Kawai, Tadahiko},
abstractNote = {Studies on efficient utilization and life extension of operating nuclear power plants (NPPs) have become increasingly important since ages of the firstgeneration NPPs are approaching their design lives. In order to predict a remaining life of each plant, it is necessary to select those critical components that strongly influence the plant life, and to evaluate their remaining lives by considering aging effects of materials and other factors. This paper proposes a new method to incorporate sophisticated crack models, such as interaction and coalescence of multiple surface cracks, into probabilistic fracture mechanism (PFM) computer programs using neural networks. First, hundreds of finite element (FE) calculations of a plate containing multiple surface cracks are performed by parametrically changing crack parameters such as sizes and locations. A fully automated 3D FE analysis system is effectively utilized here. Second, the backpropagation neural network is trained using the FE solutions, i.e. crack parameters vs. their corresponding stress intensity factors (SIFs). After a sufficient number of training iterations, the network attains an ability to promptly output SIFs for arbitrary combinations of crack parameters. The well trained network is then incorporated into the parallel PFM program which runs on one of massively parallel computers composed of 512 processing units. To demonstrate its fundamental performances, the present computer program is applied to evaluate failure probabilities of aged reactor pressure vessels considering interaction and coalescence of two dissimilar semielliptical surface cracks.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {1995},
month = {11}
}