Birkbeck, University of London Featured PhD Programmes
Imperial College London Featured PhD Programmes
University of Kent Featured PhD Programmes
Norwich Research Park Featured PhD Programmes
European Molecular Biology Laboratory (Heidelberg) Featured PhD Programmes

A novel technique for handling missing data for structural health monitoring of HIC / SOHIC

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
  • Full or part time
    Dr S Soua
    Ms S Walker
  • Application Deadline
    Applications accepted all year round

Project Description


Hydrogen Induced Cracking (HIC) / Stress Oriented Hydrogen Induced Cracking (SOHIC) is the process of hydrogen forming metals becoming brittle as a result of the diffusion mechanisms of hydrogen into the microstructure of steel structures. In the oil and gas industry with several piping configurations, HIC / SOHIC leads to a drastic reduction of the steel pipes life span. If not monitored closely, could lead to eventual fracture and leakage leading to huge loss of lives, equipment and properties.
Structural Health Monitoring (SHM) approaches for industrial processes are typically derived from routinely collected data. One of the biggest problems hindering the performance of condition monitoring is dealing with missing data. The presence of missing data may compromise the information contained within a data set, resulting in bias that impacts on the quality of learned patterns or/and the performance of fault detection and prognostics. Therefore, it is necessary to implement effective imputation techniques that provide estimations for missing values by reasoning from observed data.


The project will focus on improved understanding of the hydrogen embrittlement process including but not limited to
 Hydrogen source and exposure time
 Diffusion process in metallic structures - dissociation, accentuation pressure, absorption and adsorption
 Material grading - chemical and mechanical property analysis
 Hydrogen embrittlement growth process / mechanism
 Damage visibility analysis

The project will also focuses on the development of a novel machine-learning imputation technique that is applicable to HIC monitoring system. Through exploiting both the prior model information and observed data, the developed model should be able to provide accurate estimations for missing values when more than one variable has missing data. This general objective can be broken down to three more specific objectives that would together achieve the overall goal of the project as follows:

• Conduct a review of existing methods for addressing the problem of missing data.
• Develop a novel machine-learning imputation algorithm that can be used to impute missing values when there is more than one variable with missing data.
• Provide a benchmark case to demonstrate the ability of different missing data analysis algorithms to impute missing values using data collected from operational industrial gas compressors.
The fundamental understanding gained along with the algorithm to generate missing data will be utilised to develop conditional monitoring techniques for development of enhanced condition monitoring and inspection procedure / technique.


The primary project deliverables are:

• Fundamental understanding of HIC / SOHIC in steel and factors (pressure and temperature) contributing to enhancing HIC / SOHIC
• Risk assessment of HIC / SOHIC and mitigation techniques
• A unified machine-learning imputation approach that can provide accurate estimations for missing values in real industrial systems.
• A benchmark case for the comparison of the performance of different imputation methods using real data acquired in industrial facilities.


• Contribute to industrial guidelines, regulations and standards applicability
• Development of treatment for exposure surfaces
• Establish mitigation processes for effective localisation monitoring of HIC / SOHIC
• The use of the developed imputation technique can overcome the limitations of standard missing data analysis methods by allowing multiple imputation for missing data being present in more than one variables
• The project generates a benchmark case study for the development of new imputation techniques and its implementation in real industrial facilities
• By implementing the developed imputation technique, the information extracted from the data pre-processing module can be more reliable, thereby allowing improved machine availability and reliability and reduced the overall maintenance cost.

About Industrial Sponsor

The Lloyd’s Register Foundation funds the advancement of engineer-related education and research and supports work that enhances safety of life at sea, on land and in the air, because life matters. Lloyd’s Register Foundation is partly funded by the profits of their trading arm Lloyd’s Register Group Limited, a global engineering, technical and business services organisation.


NSIRC is a state-of-the-art postgraduate engineering facility established and managed by structural integrity specialist TWI, working closely with, top UK and International Universities and a number of leading industrial partners. NSIRC aims to deliver cutting edge research and highly qualified personnel to its key industrial partners.

Candidate Requirements

Candidates should have a relevant degree at 2.1 minimum, or an equivalent overseas degree in engineering or materials science. Candidates with suitable work experience and strong capacity in numerical modelling and experimental skills are particularly welcome to apply. Overseas applicants should also submit IELTS results (minimum 6.5) if applicable.

Funding Notes

Funding Notes

This project is funded by Lloyds Register Foundation, TWI and academic partners. The studentship will provide successful Home/EU students with a stipend of £16k/year and will cover the cost of tuition fees. Overseas applicants are welcome to apply, with total funding capped at £24k/year.

FindAPhD. Copyright 2005-2019
All rights reserved.