0
research-article

Estimation of Flow-Accelerated Corrosion Rate in Nuclear Piping System

[+] Author and Article Information
Indranil Hazra

PhD Student Department of Civil and Environmental Engineering, University of Waterloo, Waterloo, ON N2L 3G1, Canada
ihazra@uwaterloo.ca

Mahesh Pandey

Professor Department of Civil and Environmental Engineering, University of Waterloo, Waterloo, ON N2L 3G1, Canada
mdpandey@uwaterloo.ca

Mikko I. Jyrkama

Research Associate Department of Civil and Environmental Engineering, University of Waterloo, Waterloo, ON N2L 3G1, Canada
mjyrkama@uwaterloo.ca

1Corresponding author.

ASME doi:10.1115/1.4044407 History: Received March 01, 2019; Revised July 20, 2019

Abstract

Flow-accelerated corrosion (FAC) is a life-limiting factor for the piping network of the primary heat transport system (PHTS) in CANDU® reactors. The pipe wall thinning caused by FAC is monitored by carrying out periodic in-service inspections (ISI) to ensure the fitness-for-service of the piping system. Accurate prediction of the lifetime of various components in the PHTS piping network requires estimation of FAC thinning rate. The traditional Bayesian inference techniques commonly employed for parameter estimation are computationally costly. This paper presents an inexpensive and intuitive simulation-based Bayesian approach to FAC rate estimation, called approximate Bayesian computation using Markov chain Monte Carlo (ABC-MCMC). ABC-MCMC is a likelihood-free Bayesian computation scheme that generates samples directly from an approximate posterior distribution by simulating data sets from a forward model. The efficiency of ABC-MCMC is demonstrated by presenting a comparison with a likelihood-based Bayesian computation scheme, Metropolis-Hastings (MH) algorithm, using a practical data-based example. Furthermore, an innovative step has been proposed for reducing the Markov chain burn-in time in the proposed scheme. To indicate the need of a Bayesian approach in quantifying the uncertainties related to the FAC model parameters, results from the linear regression method, a common industrial approach, is also presented in this study. The numerical results show a notable reduction in computational time, suggesting that ABC-MCMC is an efficient alternative to the traditional Bayesian inference methods, specifically for handling noisy degradation data.

Copyright (c) 2019 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Tables

Errata

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In