Alumina ceramics corrosion behaviour estimated by artificial neural networks (CROSBI ID 40340)
Prilog u knjizi | izvorni znanstveni rad
Podaci o odgovornosti
Žmak, Irena ; Ćurković, Lidija
engleski
Alumina ceramics corrosion behaviour estimated by artificial neural networks
Artificial neural network models were used for estimation of corrosion behaviour of a cold isostatically pressed (CIP) high-purity alumina ceramics in aqueous HCl solution. Corrosion tests were performed with initial mass concentrations of HCl aqueous solution of 2, 10 and 20 wt. % at room temperature. Immersion times were 24, 48, 72, 120, 168 and 240 hours. Chemical stability was monitored by the amount of Al3+, Mg2+, Ca2+, Na+, Si4+ and Fe3+ ions eluted in different concentrations of HCl solution by means of atomic apsorption spectrometry (AAS), expressed as the amount of eluted ions in mg per square centimetre of test alumina area (μg Mn+/cm2). The initial HCl aqueous solution concentration and immersion time were inputs to the neural network, and the output was the amount of eluted ions (μg Mn+/cm2). Error back-propagation learning algorithm, with Levenberg –Marquardt method, was applied to the feed forward neural networks.
alumina ceramics, acid corrosion, corrosion kinetics, artificial neural networks, estimation
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
549-560.
objavljeno
Podaci o knjizi
DAAAM International Scientific Book 2009
Katalinić, Branko
Beč: DAAAM International Vienna
2009.
978-3-901509-69-8