Process Optimisation Under Equipment Degradation
Lead Research Organisation:
Imperial College London
Department Name: Computing
Abstract
Reference:
"Process optimisation in the presence of equipment degradation and failure" (voucher number 17000145) submitted to ICASE 2017 by Imperial College London.
Description:
The studentship is associated with the University's EPSRC Centre for Doctoral Training in High Performance Embedded and Distributed Systems (EP/L016796/1). While the application area of the project falls outside the key application areas of the CDT, the CDT's major themes of design and optimisation, and analysis and verification, are extremely relevant to the project.
The motivation for this project is as follows: A key target in Drilling operations in the Oil and Gas industry is to deliver the safely drilled well in the least possible time. A significant portion of lost time results from tool failures, particularly if the tool fails while in the well because of the time it takes to retrieve the tool from the well and then wait for a substitute tool or repair the failed one. As a result there is a strong interest in being able to control the tool failure so that it fails only when convenient, for example when a well section has been finished drilling. This objective, however, has to be balanced with obtaining the maximum tool output in order to drill faster. Using industrial understanding of tool failure mechanisms, health management aims to optimize operation so as to maximize the overall operational gains.
The aim of this project is to develop an optimization framework under significant uncertainty. The main case study will require a prognosis model, that is, a quantitative description of the influence of operating conditions on the degradation of the tool, and a diagnosis model, that is, a quantitative relation between the state of degradation of the tool and the measured symptoms. Schlumberger will supply both models. These models often have large uncertainty due to the complexity of the drilling tools and of their operating environment. As a result, the major science in the scope of this project is developing optimization methods dealing with multiple opposing objectives under significant uncertainty.
"Process optimisation in the presence of equipment degradation and failure" (voucher number 17000145) submitted to ICASE 2017 by Imperial College London.
Description:
The studentship is associated with the University's EPSRC Centre for Doctoral Training in High Performance Embedded and Distributed Systems (EP/L016796/1). While the application area of the project falls outside the key application areas of the CDT, the CDT's major themes of design and optimisation, and analysis and verification, are extremely relevant to the project.
The motivation for this project is as follows: A key target in Drilling operations in the Oil and Gas industry is to deliver the safely drilled well in the least possible time. A significant portion of lost time results from tool failures, particularly if the tool fails while in the well because of the time it takes to retrieve the tool from the well and then wait for a substitute tool or repair the failed one. As a result there is a strong interest in being able to control the tool failure so that it fails only when convenient, for example when a well section has been finished drilling. This objective, however, has to be balanced with obtaining the maximum tool output in order to drill faster. Using industrial understanding of tool failure mechanisms, health management aims to optimize operation so as to maximize the overall operational gains.
The aim of this project is to develop an optimization framework under significant uncertainty. The main case study will require a prognosis model, that is, a quantitative description of the influence of operating conditions on the degradation of the tool, and a diagnosis model, that is, a quantitative relation between the state of degradation of the tool and the measured symptoms. Schlumberger will supply both models. These models often have large uncertainty due to the complexity of the drilling tools and of their operating environment. As a result, the major science in the scope of this project is developing optimization methods dealing with multiple opposing objectives under significant uncertainty.
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/R511961/1 | 01/10/2017 | 31/03/2023 | |||
| 2013307 | Studentship | EP/R511961/1 | 01/10/2017 | 30/09/2021 | Johannes Wiebe |
| Description | This work has achieved three types of results: 1) Exploration of the effect of uncertainty in equipment degradation on economic decision-making processes. We have shown that equipment degradation uncertainty has a large effect on the economic evaluation of processes optimization and that decisions can be improved by taking this uncertainty into account. 2) Development of novel methods for optimization under uncertainty. We have developed novel methods for optimization under uncertainty that allow uncertain data-based models, i.e. machine learning models, to be incorporated into decision-making processes. 3) Development of software tools to make methods for optimization under uncertainty more accessible, especially for less experienced users. |
| Exploitation Route | The insights from this work are already being taken forward by our industry partners and are of interest to many other industrial decision-makers in industries where equipment degradation of relevance. The methods developed from this work are freely available and may be extended and improved by other academics for diverse applications with uncertain data-based models. The software tools developed are also open source and freely available and may be used by anyone interested in optimization under uncertainty. |
| Sectors | Agriculture, Food and Drink,Chemicals,Digital/Communication/Information Technologies (including Software),Energy,Environment,Government, Democracy and Justice,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology,Transport |
| Description | The findings of this award have been developed in close collaboration with our industry partners at Schlumberger Research Cambridge. The findings are of considerable economic interest to them and are being integrated into their decision-making process. |
| First Year Of Impact | 2019 |
| Sector | Energy |
| Impact Types | Economic |
| Title | ROmodel |
| Description | The software allows formulating and solving of robust optimisation problems within the Pyomo modelling language |
| Type Of Technology | Software |
| Year Produced | 2021 |
| Open Source License? | Yes |
| Impact | The software makes robust optimisation techniques accessible to a wider audience |
| URL | https://github.com/johwiebe/romodel |