| Full text | |
| Author(s): |
Colombo, Danilo
[1]
;
Alves Lima, Gilson Brito
[2]
;
Pereira, Danillo Roberto
[3]
;
Papa, Joao P.
[4]
Total Authors: 4
|
| Affiliation: | [1] CENPES PETROBRAS, Rio De Janeiro - Brazil
[2] Fluminense Fed Univ, Dept Prod Engn, Niteroi, RJ - Brazil
[3] Univ Western Sao Paulo, Presidente Prudente - Brazil
[4] Sao Paulo State Univ, UNESP, Dept Comp, Sao Paulo - Brazil
Total Affiliations: 4
|
| Document type: | Journal article |
| Source: | RELIABILITY ENGINEERING & SYSTEM SAFETY; v. 198, JUN 2020. |
| Web of Science Citations: | 0 |
| Abstract | |
Downhole Safety Valve (DHSV) stands for a device widely used in offshore wells to ensure the integrity and avoid uncontrolled leaks of oil and gas to the environment, known as blowouts. The reliability estimation of such valves can be used to predict the blowout occurrence and to evaluate the workover demand, as well as to assist decision-making actions. In this paper, we introduce FEMaR, a Finite Element Machine for regression problems, which figures no training step, besides being parameterless. Another main contribution of this work is to evaluate several machine learning models to estimate the reliability of DHSVs for further comparison against traditional statistical methods. The experimental evaluation over a dataset collected from a Brazilian oil and gas company showed that machine learning techniques are capable of obtaining promising results, even in the presence of censored information, and they can outperform the statistical approaches considered in this work. Such findings also investigated using uncertainty analysis, evidenced that we can save economic resources and increase the safety at the offshore well operations. (AU) | |
| FAPESP's process: | 14/12236-1 - AnImaLS: Annotation of Images in Large Scale: what can machines and specialists learn from interaction? |
| Grantee: | Alexandre Xavier Falcão |
| Support Opportunities: | Research Projects - Thematic Grants |
| FAPESP's process: | 13/07375-0 - CeMEAI - Center for Mathematical Sciences Applied to Industry |
| Grantee: | Francisco Louzada Neto |
| Support Opportunities: | Research Grants - Research, Innovation and Dissemination Centers - RIDC |
| FAPESP's process: | 16/19403-6 - Energy-based learning models and their applications |
| Grantee: | João Paulo Papa |
| Support Opportunities: | Regular Research Grants |