Optimized Flare Performance Analysis Through Multi-Modal Machine Learning and Temporal Standard Deviation Enhancements

Flaring is a routine practice in the upstream gas industry to dispose of waste gases, but its efficiency can drop significantly under non-ideal conditions such as crosswinds, over-aeration, or over-steaming. These inefficiencies lead to incomplete combustion, producing harmful substances like carbon...

Full description

Saved in:
Bibliographic Details
Main Authors: Said Boumaraf, Pengfei Li, Muaz Al Radi, Fares Oussama Abdelhafez, Abderaouf Behouch, Khalid Yousef Al Awadhi, Hamad Karki, Sajid Javed, Jorge Dias, Naoufel Werghi
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10879335/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850034023584235520
author Said Boumaraf
Pengfei Li
Muaz Al Radi
Fares Oussama Abdelhafez
Abderaouf Behouch
Khalid Yousef Al Awadhi
Hamad Karki
Sajid Javed
Jorge Dias
Naoufel Werghi
author_facet Said Boumaraf
Pengfei Li
Muaz Al Radi
Fares Oussama Abdelhafez
Abderaouf Behouch
Khalid Yousef Al Awadhi
Hamad Karki
Sajid Javed
Jorge Dias
Naoufel Werghi
author_sort Said Boumaraf
collection DOAJ
description Flaring is a routine practice in the upstream gas industry to dispose of waste gases, but its efficiency can drop significantly under non-ideal conditions such as crosswinds, over-aeration, or over-steaming. These inefficiencies lead to incomplete combustion, producing harmful substances like carbon monoxide and unburned methane, which contribute significantly to global warming. Current solutions for monitoring flare efficiency are often complex or expensive, limiting their widespread adoption. This work introduces a novel framework for estimating flare combustion efficiency (CE) using a multi-modal machine learning architecture enhanced by a Temporal Standard Deviation (TSD) preprocessing technique. Our approach combines synchronized visual data with minimal field measurements (FM) for accurate efficiency estimation. We first extract sequential frames from an RGB video stream of flares and process them to extract TSD images, which essentially highlight the variability and dynamic changes in the combustion process. Next, we extract image feature representations from TSD images using the state-of-the-art Vision Transformer (ViT) and fuse them with experimentally selected FM data to create a comprehensive combustion dataset. A CatBoost regression model is then trained on this dataset to estimate the final CE. Our proposed framework is validated using real-world data from industrial flare upstream operations. The results demonstrate significant improvements in estimation accuracy and reliability compared to traditional methods, achieving an R-squared score of 94.77% with minimal FM data. This approach not only enhances the understanding of combustion dynamics but also offers a scalable, cost-effective solution for continuous flare monitoring and optimization.
format Article
id doaj-art-5d3adaea68fa4911b8a53e39bd5db80f
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-5d3adaea68fa4911b8a53e39bd5db80f2025-08-20T02:57:59ZengIEEEIEEE Access2169-35362025-01-0113343623437710.1109/ACCESS.2025.354055810879335Optimized Flare Performance Analysis Through Multi-Modal Machine Learning and Temporal Standard Deviation EnhancementsSaid Boumaraf0https://orcid.org/0000-0001-8154-7195Pengfei Li1https://orcid.org/0009-0009-9458-8557Muaz Al Radi2Fares Oussama Abdelhafez3https://orcid.org/0009-0002-3420-7053Abderaouf Behouch4Khalid Yousef Al Awadhi5Hamad Karki6Sajid Javed7https://orcid.org/0000-0002-0036-2875Jorge Dias8https://orcid.org/0000-0002-2725-8867Naoufel Werghi9https://orcid.org/0000-0002-5542-448XC2PS, Department of Computer Science, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesC2PS, Department of Computer Science, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesC2PS, Department of Computer Science, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesDepartment of Mechanical and Nuclear Engineering, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesDepartment of Physics, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesADNOC Technology, Abu Dhabi, United Arab EmiratesDepartment of Mechanical and Nuclear Engineering, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesDepartment of Computer Science, KUCARS and C2PS, Khalifa University, Abu Dhabi, United Arab EmiratesDepartment of Computer Science, KUCARS and C2PS, Khalifa University, Abu Dhabi, United Arab EmiratesDepartment of Computer Science, KUCARS and C2PS, Khalifa University, Abu Dhabi, United Arab EmiratesFlaring is a routine practice in the upstream gas industry to dispose of waste gases, but its efficiency can drop significantly under non-ideal conditions such as crosswinds, over-aeration, or over-steaming. These inefficiencies lead to incomplete combustion, producing harmful substances like carbon monoxide and unburned methane, which contribute significantly to global warming. Current solutions for monitoring flare efficiency are often complex or expensive, limiting their widespread adoption. This work introduces a novel framework for estimating flare combustion efficiency (CE) using a multi-modal machine learning architecture enhanced by a Temporal Standard Deviation (TSD) preprocessing technique. Our approach combines synchronized visual data with minimal field measurements (FM) for accurate efficiency estimation. We first extract sequential frames from an RGB video stream of flares and process them to extract TSD images, which essentially highlight the variability and dynamic changes in the combustion process. Next, we extract image feature representations from TSD images using the state-of-the-art Vision Transformer (ViT) and fuse them with experimentally selected FM data to create a comprehensive combustion dataset. A CatBoost regression model is then trained on this dataset to estimate the final CE. Our proposed framework is validated using real-world data from industrial flare upstream operations. The results demonstrate significant improvements in estimation accuracy and reliability compared to traditional methods, achieving an R-squared score of 94.77% with minimal FM data. This approach not only enhances the understanding of combustion dynamics but also offers a scalable, cost-effective solution for continuous flare monitoring and optimization.https://ieeexplore.ieee.org/document/10879335/Flare stackscombustion efficiencytemporal standard deviationdeep learningCatBoost regression
spellingShingle Said Boumaraf
Pengfei Li
Muaz Al Radi
Fares Oussama Abdelhafez
Abderaouf Behouch
Khalid Yousef Al Awadhi
Hamad Karki
Sajid Javed
Jorge Dias
Naoufel Werghi
Optimized Flare Performance Analysis Through Multi-Modal Machine Learning and Temporal Standard Deviation Enhancements
IEEE Access
Flare stacks
combustion efficiency
temporal standard deviation
deep learning
CatBoost regression
title Optimized Flare Performance Analysis Through Multi-Modal Machine Learning and Temporal Standard Deviation Enhancements
title_full Optimized Flare Performance Analysis Through Multi-Modal Machine Learning and Temporal Standard Deviation Enhancements
title_fullStr Optimized Flare Performance Analysis Through Multi-Modal Machine Learning and Temporal Standard Deviation Enhancements
title_full_unstemmed Optimized Flare Performance Analysis Through Multi-Modal Machine Learning and Temporal Standard Deviation Enhancements
title_short Optimized Flare Performance Analysis Through Multi-Modal Machine Learning and Temporal Standard Deviation Enhancements
title_sort optimized flare performance analysis through multi modal machine learning and temporal standard deviation enhancements
topic Flare stacks
combustion efficiency
temporal standard deviation
deep learning
CatBoost regression
url https://ieeexplore.ieee.org/document/10879335/
work_keys_str_mv AT saidboumaraf optimizedflareperformanceanalysisthroughmultimodalmachinelearningandtemporalstandarddeviationenhancements
AT pengfeili optimizedflareperformanceanalysisthroughmultimodalmachinelearningandtemporalstandarddeviationenhancements
AT muazalradi optimizedflareperformanceanalysisthroughmultimodalmachinelearningandtemporalstandarddeviationenhancements
AT faresoussamaabdelhafez optimizedflareperformanceanalysisthroughmultimodalmachinelearningandtemporalstandarddeviationenhancements
AT abderaoufbehouch optimizedflareperformanceanalysisthroughmultimodalmachinelearningandtemporalstandarddeviationenhancements
AT khalidyousefalawadhi optimizedflareperformanceanalysisthroughmultimodalmachinelearningandtemporalstandarddeviationenhancements
AT hamadkarki optimizedflareperformanceanalysisthroughmultimodalmachinelearningandtemporalstandarddeviationenhancements
AT sajidjaved optimizedflareperformanceanalysisthroughmultimodalmachinelearningandtemporalstandarddeviationenhancements
AT jorgedias optimizedflareperformanceanalysisthroughmultimodalmachinelearningandtemporalstandarddeviationenhancements
AT naoufelwerghi optimizedflareperformanceanalysisthroughmultimodalmachinelearningandtemporalstandarddeviationenhancements