Pitting corrosion behaviour modelling of stainless steel with support vector machines

Jiménez‐Come, María Jesús; Turias, Ignacio J.; Ruiz-Aguilar, Juan Jesús

doi:10.1002/maco.201407788

Cited by 7 publications

(6 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Wen et al (2009) predicted the corrosion rate of 3C steel under different seawater environment with SVR models and found the prediction error is smaller than back-propagation neural networks (BPNN) models for the identical training and test dataset, and the generalization ability of SVR model is also superior to that of BPNN model. Applications of SVR in predicting the pitting corrosion behavior of stainless steel (Jimenez-Come et al 2015 also obtained high prediction accuracy and it is useful to determine the critical factors that influence the corrosion process via sensitivity analysis (Jimenez-Come et al 2019). Fang et al (2008) hybrid genetic algorithms and SVR to forecast atmospheric corrosion of zinc and steel which provides better prediction capability than ANN models.…”

Section: Statistical Learning Modelmentioning

confidence: 99%

Atmospheric corrosion prediction: a review

Cai

Zhao

et al. 2020

Corrosion Reviews

View full text Add to dashboard Cite

The atmospheric corrosion of metallic materials causes great economic loss every year worldwide. Thus, it is meaningful to predict the corrosion loss in different field environments. Generally, the corrosion prediction method includes three parts of work: the modelling of the corrosive environment, the calibration of the corrosion effects, and the establishment of the corrosion kinetics. This paper gives an overview of the existing methods as well as promising tools and technologies which can be used in corrosion prediction. The basic corrosion kinetic model is the power function model and it is accurate for short-term corrosion process. As for the long-term corrosion process, the general linear models are more appropriate as they consider the protective effect of the corrosion products. Most corrosion effect models correlate the environmental variables, which are characterized by the annual average value in most cases, with corrosion parameters by linear equations which is known as the dose-response function. Apart from these conventional methods, some mathematical and numerical methods are also appropriate for corrosion prediction. The corrosive environment can be described by statistical distributions, time-varying functions and even geographic information system (GIS), while the corrosion effect can be captured via response surface models and statistical learning methods.

show abstract

Section: Statistical Learning Modelmentioning

confidence: 99%

Atmospheric corrosion prediction: a review

Cai

Zhao

et al. 2020

Corrosion Reviews

View full text Add to dashboard Cite

show abstract

“…Using real time atmospheric environmental elements, some of the stable classifier models, such as Naive Bayes (NB), K-Nearest Neighbor (KNN) and support vector machine (SVM), which can be used to find out an accurate grade subdivision model for average corrosion rate. Among them, the most commonly used one in the field of corrosion research is SVM (Fang et al , 2008; Jiménez-Come et al , 2015; Chou et al , 2017). SVM determines a boundary as the hyperplane that maximizes the margin of separation in a feature space.…”

Section: Data Modeling Research and Analysismentioning

confidence: 99%

Modeling research of initial atmospheric corrosion of Q235 carbon steel based on electrical resistance probe

Pei

2021

ACMM

View full text Add to dashboard Cite

Purpose This paper aims to discover the mathematical model for Q235 carbon steel corrosion date acquired in the initial stage of atmospheric corrosion using electrical resistance probe. Design/methodology/approach In this paper, mathematical approaches are used to construct a classification model for atmospheric environmental elements and material corrosion rates. Findings Results of the experiment show that the corrosion data can be converted into corrosion depth for calculating corrosion rate to obtain corrosion kinetics model and conform corrosion acceleration phase. Combined with corresponding atmospheric environmental elements, a real time grade subdivision model for corrosion rate can be constructed. Originality/value These mathematical models constructed by real time corrosion data can be well used to research the characteristics about initial atmospheric corrosion of Q235 carbon steel.

show abstract

“…Therefore, corrosion evaluation and life prediction through machine learning have become a research hotspot in recent years. Jiménez et al established a pitting model for stainless steel in various environments through machine learning. Kamrunnahar and Urquidi‐Macdonald used a backpropagation neural network method to predict the corrosion behaviour of metal alloys and classify and prioritize parameters (pH, temperature, time, electrolyte composition, metal composition, etc.…”

Section: Introductionmentioning

confidence: 99%

Machine learning assistance for electrochemical curve simulation of corrosion and its application

Gong

Dong

et al. 2019

Materials & Corrosion

View full text Add to dashboard Cite

In this paper, we used machine learning algorithms such as k‐nearest neighbour, decision tree, gradient boosting decision tree, random forest, and support vector machine in a scikit‐learn module of Python to construct polarization curves and electrochemical impedance spectra. After application of this method to a high‐level nuclear waste disposal tank material of pure copper, the polarization curves, and electrochemical impedance spectra of pure copper under different chloride ion concentrations, sulfide concentrations, and temperature environments were used as training sets. The combination of the cross‐validation and experimental validation results revealed that the RF algorithm had the best effect on predicting the polarization curve and EIS. Through the input weight analysis, it was found that the sulfide concentration had the greatest influence on the polarization curve, followed by the chloride ion concentration, and the temperature influence was the smallest. For electrochemical impedance spectroscopy, sulfide and temperature had a large effect, while chloride ions had little effect. The result of the weight analysis was consistent with traditional electrochemical results.

show abstract

Pitting corrosion behaviour modelling of stainless steel with support vector machines

Cited by 7 publications

References 36 publications

Atmospheric corrosion prediction: a review

Atmospheric corrosion prediction: a review

Modeling research of initial atmospheric corrosion of Q235 carbon steel based on electrical resistance probe

Machine learning assistance for electrochemical curve simulation of corrosion and its application

Contact Info

Product

Resources

About