Menu
Search
Filters
Close
Menu

Mechanistic Model as a Bias toMachine Learning Algorithm for Confident Prediction of Corrosion

Picture for Mechanistic Model as a Bias toMachine Learning Algorithm for Confident Prediction of Corrosion
Available for download

A constant challenge persists among corrosion engineers to estimate and predict field corrosion rates despite the huge advancements in corrosion science. This situation has compelled the corrosion engineers to opt for the machine learning (ML) algorithms for corrosion prediction. However, the “blackbox” ML algorithms are not appreciated in high stakes decisions because they use arbitrary fitting models rather than scientific principles.

Product Number: 51323-19108-SG
Author: Ishan Patel, Gheorghe Bota
Publication Date: 2023
$0.00
$20.00
$20.00

Bayesian network is employed to estimate a risk-based life cycle cost of corrosion for assets. It has been highly recognized that inclusion of mechanistic models to a Bayesian network can increase the confidence in estimation of corrosion rates. However, coefficients of mechanistic models are often unknown, especially when complex rate processes are involved, which discourages the usage of the model. A methodology is proposed here, to introduce a mechanistic model as a bias to a regressive machine learning (ML) algorithm. No attempts have been made to obtain phenomenological coefficients of the mechanistic model. Instead, a methodology is proposed to obtain a highly tuned parameter vector for a ML algorithm from a learning set of corrosion rate data.

Bayesian network is employed to estimate a risk-based life cycle cost of corrosion for assets. It has been highly recognized that inclusion of mechanistic models to a Bayesian network can increase the confidence in estimation of corrosion rates. However, coefficients of mechanistic models are often unknown, especially when complex rate processes are involved, which discourages the usage of the model. A methodology is proposed here, to introduce a mechanistic model as a bias to a regressive machine learning (ML) algorithm. No attempts have been made to obtain phenomenological coefficients of the mechanistic model. Instead, a methodology is proposed to obtain a highly tuned parameter vector for a ML algorithm from a learning set of corrosion rate data.

Product tags
Also Purchased
Picture for Development and Testing of a Mechanistic Based Atmospheric Corrosion Prediction System
Available for download

Development and Testing of a Mechanistic Based Atmospheric Corrosion Prediction System

Product Number: 51323-18962-SG
Author: A.R. Trueman
Publication Date: 2023
$20.00

The work here is the culmination of many years of prior effort in the development of an atmospheric corrosion model and accompanying sensors. Atmospheric corrosion is a complicated process where many factors interact to determine if it occurs and its severity. These factors can be separated into
three general categories: environmental, surface salts, and materials.
Picture for Corrosion Scale And Moisture Assessments – An Improvement To On-Stream Inspections For CUI Management
Available for download

Corrosion Scale And Moisture Assessments – An Improvement To On-Stream Inspections For CUI Management

Product Number: 51323-19167-SG
Author: Ahmad Raza Khan Rana, Graham Brigham, Omar Chaar, Syed Umair Niaz Bukhari
Publication Date: 2023
$20.00

CUI (Corrosion under insulation) is among the key damage mechanisms in the process and hydrocarbon industries which results in the localized corrosion (mainly) thereby reducing the service life of the assets (pipes, equipment, etc.) CUI is reportedly known as the reason behind 40-60% of failures in the facility piping whereas small bore piping (i.e., NPS< 4”) are even more sensitive to CUI failures, where up to 81% of reported failures in small-sized piping are known to be from CUI. Monetary spending to inspect and fix CUI-related failures cost 10% of overall maintenance budget in a typical medium-sized oil refinery. CUI risk is influenced by numerous operational and environmental factors which impedes its management in a typical AIM (Asset integrity management) program.
Picture for Practical Application Of AI To Transform Pipeline Integrity Management
Available for download

Practical Application Of AI To Transform Pipeline Integrity Management

Product Number: 51323-19039-SG
Author: Pankaj Tandon
Publication Date: 2023
$20.00

Management and operating teams, across all industries, but especially inside asset management organizations, must increasingly rely on data-driven assessment and analytics for decision making. Clearly, inefficiency and ineffectiveness associated with manual or poorly optimized business processes and fragmented business management Infrastructures is a barrier to formulating the safest, most efficient and cost effective asset management and operational decisions. According to Deloitte’s digital maturity analysis, the midstream industry and specially the pipeline sector is in its lowest stages of digital adoption (first 20-30% progress).
Categories
Industry
Recently viewed
Popular tags
View all

Association for Materials Protection and Performance

© AMPP All Rights Reserved.
Membership Terms of Content Use
Contact Customer Support