Corrosion current density prediction of 3C steel under different seawater environment via artificial neural network

The harsh seawater environment provides a corrosive medium to all structures built from metallic materials that operate within this environment. Due to its mechanical strength, ease of manufacture, formability, and low cost, 3C steel is the most widely used structural material in various sectors. Ho...

Full description

Saved in:
Bibliographic Details
Main Authors: Zulkifli, Mohammad Fakhratul Ridwan Bin, Wahab, Siti Norhaizam Binti Abd, Abdullah, Samsuri Bin, Jusoh, Suriani Binti Mat, Bin Wan Nik, Wan Mohd Norsani
Format: Conference Proceeding
Language:English
Subjects:
Algorithms
Artificial neural networks
Corrosion
Corrosion currents
Corrosion mechanisms
Corrosion rate
Corrosion resistant steels
Corrosion tests
Current density
Dissolved oxygen
Marine environment
Multilayer perceptrons
Multilayers
Neural networks
Offshore structures
Oxidation
Parameters
Regularization
Sea water corrosion
Seawater
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The harsh seawater environment provides a corrosive medium to all structures built from metallic materials that operate within this environment. Due to its mechanical strength, ease of manufacture, formability, and low cost, 3C steel is the most widely used structural material in various sectors. However, the corrosion problem cannot be avoided due to its interaction with seawater. Furthermore, due to the extensive range of parameters regulating the rate, predicting the corrosion rate of 3C steel structures in the marine environment is challenging. Hence, corrosion rate modelling improves better understanding in terms of corrosion behaviour. In this study, the corrosion rate prediction of 3C steel in seawater was conducted via an artificial neural network (ANN). This study aims to establish a corrosion current density prediction of 3C steel influenced by different seawater environment parameters. The data came from previous studies that looked at dissolved oxygen, salinity, pH value, temperature, and oxidation-reduction potential to see how these affected the corrosion rate of 3C steel in seawater. Three training algorithms were proposed in Multi-Layer Perceptron (MLP), which are Levenberg Marquardt (LM), Bayesian Regularization (BR), and Scaled Conjugate Gradient (SCG). The results show that the LM algorithm is the best training algorithm in this study according to its performance indicator represented by high regression, R, and low mean square error (MSE) value.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0188540