Advanced Condition Assessment and Pipe Failure Prediction ProjectAdvanced Condition Assessment and Pipe Failure Prediction Project
  • Home
  • About
    • Partners
    • Governance
    • Structure
    • People
      • Committee of Management
      • Activity Leaders
      • Activity 1 team (Monash University)
      • Activity 2 Team (UTS)
      • Activity 3 Team (University of Newcastle)
      • Activity 4 Teams (refer to Activities 1-3)
      • Consultant to the project
  • Activities
    • Overview
    • Activity 1
      • Activity 1 team (Monash University)
    • Activity 2
      • Activity 2 Team (UTS)
    • Activity 3
      • Activity 3 Team (University of Newcastle)
    • Activity 4
      • Activity 4 Teams (refer to Activities 1-3)
    • Activity 5
  • Vision & Future
  • Publications
    • Conference Papers
    • Fact Sheets
    • Journal Papers
    • Bulletins
    • Videos
  • Intranet
  • News
  • Events
  • Contact

MULTIPLE DEFECT INTERPRETATION BASED ON GAUSSIAN PROCESSES FOR MFL TECHNOLOGY

by Buddhi Wijerathna, Teresa Vidal-Calleja, Sarath Kodagoda, Qiang Zhang and Jaime Valls Miro

Centre for Autonomous Systems, Faculty of Engineering and IT, University of Technology, Sydney, NSW 2007, Australia.

Published in:

Proc. SPIE 8694, Nondestructive Characterization for Composite Materials, Aerospace Engineering, Civil Infrastructure, and Homeland Security 2013The 9th IEEE Conference on Industrial Electronics and Applications (ICIEA 2014)

Date of Conference:

10 March 2013

Conference Location :

San Diego, California, USA

Key words:

Nondestructive testing, Magnetic ux leakage, Finite element analysis,  Data driven inference, Gaussian processes.

Abstract

Magnetic Flux Leakage (MFL) technology has been used in non-destructive testing for more than three decades. There have been several publications in detecting and sizing defects on metal pipes using machine learning techniques. Most of these literature focus on isolated defects, which is far from the real scenario. This study is towards the generalization of interpretation of the leakage ux in the presence of multiple defects based on simulation models, together with data-driven inference methodologies, such as Gaussian Process (GP) models. A MFL device has been simulated using both COMSOL Multiphysics and ANSYS software followed by prototyping the same device for experimental validations. Multiple defects with di erent geometrical con gurations were introduced on a cast iron pipe sample and both radial and axial components of the leakage eld have been measured. It was observed that both axial and radial components di er with di erent defect con gurations. We propose to use GP to solve the inverse model problem by capturing such behaviors, i.e. to recover the pro le of a cluster of defects from the measurements of a MFL device. The data was used to learn the non-parametric GP model with squared exponential covariance function and automatic relevance determination to solve this regression problem. Extensive quantitative and qualitative evaluations are presented using simulated and experimental data that validate the success of the proposed non-parametric methodology for interpreting the pro ling of clusters of defects with MFL technology.

Click here to download the Paper

Information about Pipes

In August 2011 international water research organisations, Australian water utilities and three Australian universities came together through a collaborative research agreement, and committed overall funding of $16 million (including $4 million cash) over five years to undertake this research through the Advanced Condition Assessment and Pipe Failure Prediction Project.

more info

Recent posts

  • The final meeting of the Committee of Management

    December 6, 2016

  • Final Technical Assessment Committee meeting

    November 24, 2016

  • Critical Pipes Project wins B/HERT award

    November 16, 2016

Quick Links

  • About
  • Activities
  • Vision and Future
  • Resources
  • News
  • Events
  • Contact

Critical Pipes 2021 © - by Inspire Design

  • Home
  • About
  • Activities
  • Resources
  • Events
  • News
  • Contact