skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Prediction of Aluminum Pitting in Natural Waters Via Artificial Neural Network Analysis

Abstract

One mission of the Department of Energy''s Savannah River Site (SRS) is to store spent nuclear fuel (SNF) and other waste products while permanent storage facilities for such materials are prepared. This extended storage increases the probability of pitting corrosion for aluminum-based SNF stored in natural (fresh) waters. The Back Propagation of Error method was used to train and test an Artificial Neural Network (ANN) model using archival pitting data. For aluminum AA1100 (UNS A91100), a database from the British Non-Ferrous Metals Research Association (BNFMRA) was used because it contained the relevant chemical species for pitting. A trained ANN, consisting of two hidden layers of six and four elements each, provided a better estimate of pit depth as a function of water chemistry after 150000 training cycles than a regression (linear, quadratic and cross-product) model.

Authors:
Publication Date:
Research Org.:
Savannah River Site (US)
Sponsoring Org.:
US Department of Energy (US)
OSTI Identifier:
9832
Report Number(s):
WSRC-MS-99-00579
TRN: US0103205
DOE Contract Number:  
AC09-96SR18500
Resource Type:
Journal Article
Journal Name:
Corrosion Journal
Additional Journal Information:
Other Information: PBD: 13 Aug 1999
Country of Publication:
United States
Language:
English
Subject:
36 MATERIALS SCIENCE; 11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS; ALUMINIUM; FORECASTING; NEURAL NETWORKS; PITTING CORROSION; SAVANNAH RIVER PLANT; STORAGE FACILITIES; WATER CHEMISTRY; SPENT FUEL STORAGE; RADIOACTIVE WASTE STORAGE

Citation Formats

Mickalonis, J.I. Prediction of Aluminum Pitting in Natural Waters Via Artificial Neural Network Analysis. United States: N. p., 1999. Web.
Mickalonis, J.I. Prediction of Aluminum Pitting in Natural Waters Via Artificial Neural Network Analysis. United States.
Mickalonis, J.I. Fri . "Prediction of Aluminum Pitting in Natural Waters Via Artificial Neural Network Analysis". United States. https://www.osti.gov/servlets/purl/9832.
@article{osti_9832,
title = {Prediction of Aluminum Pitting in Natural Waters Via Artificial Neural Network Analysis},
author = {Mickalonis, J.I.},
abstractNote = {One mission of the Department of Energy''s Savannah River Site (SRS) is to store spent nuclear fuel (SNF) and other waste products while permanent storage facilities for such materials are prepared. This extended storage increases the probability of pitting corrosion for aluminum-based SNF stored in natural (fresh) waters. The Back Propagation of Error method was used to train and test an Artificial Neural Network (ANN) model using archival pitting data. For aluminum AA1100 (UNS A91100), a database from the British Non-Ferrous Metals Research Association (BNFMRA) was used because it contained the relevant chemical species for pitting. A trained ANN, consisting of two hidden layers of six and four elements each, provided a better estimate of pit depth as a function of water chemistry after 150000 training cycles than a regression (linear, quadratic and cross-product) model.},
doi = {},
journal = {Corrosion Journal},
number = ,
volume = ,
place = {United States},
year = {1999},
month = {8}
}