The Hong Kong Polytechnic University Featured PhD Programmes
University of Dundee Featured PhD Programmes
University College London Featured PhD Programmes
University of Oxford Featured PhD Programmes
University of Leeds Featured PhD Programmes

Oil Drilling Process Condition Detection and Prediction based on Real-time Dynamic Electrical Resistivity Tomography

  • Full or part time
  • Application Deadline
    Applications accepted all year round
  • Self-Funded PhD Students Only
    Self-Funded PhD Students Only

Project Description

Oil drilling is an essential process for Petroleum engineering and it is a complex time-variant process with multi-operational points. Most of the existing methods are based on the off-line trained model, where the real-time dynamics of the drilling process cannot be featured. However, the drilling process is risky while the fault condition would damage the drills and oil wells. Therefore, it is significant in practice to develop an on-line data learning based modelling approach for real-time condition detection and prediction. Using the on-line model, the optimal decision can be made to protect the device and reduce the drilling cost.

Technically speaking, this project focuses on Electrical Resistivity Tomography (ERT) technology, where the real-time images will return to the ground using the embedded-system-based drilling system. Based on the real-time images, the machine learning methods will be developed to achieve the fault diagnosis and condition prediction. The challenges are summarised as follows:
1) The high pressure and high temperature environment will affect the embedded system, especially the sensors. It results in the uncertainties and noises in the image which reduce the accuracy of the detection and prediction.
2) The faults underground e.g. crack, collapse, etc. would happen dependently which means that identified the specified fault is difficult. The features have to be distinguished based on the measured data subjected to the uncertainties and randomness.
3) The real-time modelling is challenged with limited data. To guarantee the performance and accuracy of the prediction, the simulation would be an option to train the fundamental model with basic functions however the experiments of the oil drilling process are very expensive.
To solve the fore-mentioned problems, the AI technologies will be used for ERT image where the new framework will be designed dealing with the uncertainties and randomness. In particular, the limited data would lead to the non-Gaussian distribution for the random variable of the dynamic model. Meanwhile, the features of the image will be collected and identified which enhance the performance of the fault diagnosis and condition prediction. The model validation will be implemented by off-line simulation with the collected experimental data. The performance analysis will be obtained with the limited data and the new developed AI framework considering the uncertainties.

The following 3-year schedule has been made for completing the project as a PhD research programme. In year one, the literature reviews for the existing dynamic image processing methods and AI methods will be done while the widely-used CNN structure will be adopted firstly to achieve the basic functions. The novel AI framework considering the uncertainties and randomness will be developed in the second year in order to achieve the feature extraction for fault diagnosis. In the third year, the predictive model will be obtained which will be validated by the off-line simulation and the final results will be tested in real experiment.

This project is based on the collaboration with Sinopec Petroleum Engineering Technology Academe (SPETA). It is the research institute belongs to the China Petrochemical Corporation which is the world’s largest oil refining, gas and petrochemical conglomerate. The experimental data will be supplied by SPETA and the simulation results will be validated by SPETA. In the end, the practical testing will be implemented in China leading by SPETA.

Funding Notes

This is a self-funded PhD project; applicants will be expected to pay their own fees or have a suitable source of third-party funding. A bench fee may also apply in addition to tuition fees.


R.E. Langer On determination of earth conductivity from observed surface potentials, Bull Am Math Soc, 42, pp 747–754, 1936.
Mifeng Ren, Qichun Zhang & Jianhua Zhang (2019) An introductory survey of probability density function control, Systems Science & Control Engineering, 7:1, 158-170
Isermann, Rolf. "Model-based fault-detection and diagnosis–status and applications." Annual Reviews in control 29.1 (2005): 71-85.
Drilling Automation, Journal of Petroleum Technology. December 14, 2017.

Email Now

Insert previous message below for editing? 
You haven’t included a message. Providing a specific message means universities will take your enquiry more seriously and helps them provide the information you need.
Why not add a message here
* required field
Send a copy to me for my own records.

Your enquiry has been emailed successfully

FindAPhD. Copyright 2005-2019
All rights reserved.