Session: 15-01: Robotic and Automated NDE

Paper Number: 149431

149431 - An Integrated Graphical User Interface for Evaluating Inspection and Monitoring Strategies 

Abstract: 

Corrosion is a widespread degradation phenomenon affecting numerous engineering structures, spanning extensive surface areas and large timescales. The conventional surveillance approach involves either infrequent scanning inspection of the entire structure surface or continuous monitoring with permanently installed sensors at specific locations, each method having its own drawbacks. Recent technological advancements present an alternative, hybrid approach: the deployment of resident robots capable of manipulating a fleet of sensors for 4D (spatial + temporal) encoded measurements. However, the efficacy of the hybrid approach has not been thoroughly explored. In addition, discussions on the efficacy of different NDE approaches remain inconclusive, partially due to the diverse characteristics of the degradation phenomena.   

In this presentation, a versatile framework is presented. The framework encompasses crucial steps in defining and evaluating NDE processes including modelling of the 4D surface evolution of corroding surfaces, simulation of ultrasonic thickness measurements, error analysis and performance assessment. A metric named the “unreliability function” is employed to evaluate the efficacy of different mixed inspection and monitoring scenarios. This makes it possible to highlight the potential benefits of the proposed hybrid NDE and monitoring approach. In addition, the framework has been packaged as a stand-alone graphical user interface (GUI), allowing other researchers and industrial users to use the tool themselves to run through the entire evaluation process on degradation phenomena that are of interest to them.

Presenting Author: Peter Huthwaite Imperia College London

Presenting Author Biography: Peter Huthwaite received his MEng degree in Mechanical Engineering from Imperial College in 2008. He begun a PhD with the Non-Destructive Testing group later that year, completing in early 2012. Subsequently he took up a Research Associate position with the lab. In early 2015 he won a 5-year EPSRC Early Career Fellowship "Quantitative non-destructive imaging with limited data" of nearly £1 million in value, and was appointed as a Lecturer in 2016, promoted to Senior Lecturer in 2018 and Reader in 2021. His interests include imaging in both ultrasonics and radiography, with a particular focus on limited data imaging. He also has worked on numerical methods for ultrasonic simulation, including finite difference and finite element methods in the time and frequency domains. He wrote the software package Pogo (http://www.pogo.software) which is a GPU accelerated FE solution.

Authors:

Yifeng Zhang Imperial College London
Peter Huthwaite Imperia College London
Frederic Cegla Imperial College London