Incorporation of Physics into Machine Learning for Production Prediction from Unconventional Reservoirs: A Brief Review of the Gray-Box Approach

2021 ◽  
pp. 1-12
Author(s):  
Hui-Hai Liu ◽  
Jilin Zhang ◽  
Feng Liang ◽  
Cenk Temizel ◽  
Mustafa A. Basri ◽  
...  
Keyword(s):  
Machine Learning ◽  
Oil And Gas ◽  
Oil And Gas Industry ◽  
Complex Problem ◽  
Unconventional Reservoirs ◽  
Future Research ◽  
Gas Industry ◽  
Well Production ◽  
Determination Of Parameters ◽  
Production Prediction

Summary Prediction of well production from unconventional reservoirs is often a complex problem with an incomplete understanding of physics and a considerable amount of data. The most effective way for dealing with it is to use the gray-box approach that combines the strengths of physics-based models and machine learning (ML) used for dealing with certain components of the prediction where physical understanding is poor or difficult. However, the development of methodologies for the incorporation of physics into ML is still in its infancy, not only in the oil and gas industry, but also in other scientific and engineering communities, including the physics community. To set the stage for further advancing the use of combining physics-based models with ML for predicting well production, in this paper we present a brief review of the current developments in this area in the industry, including ML representation of numerical simulation results, determination of parameters for decline curve analysis (DCA) models with ML, physics-informed ML (PIML) that provides an efficient and gridless method for solving differential equations and for discovering governing equations from observations, and physics-constrained ML (PCML) that directly embeds a physics-based model into a neural network. The advantages and potential limitations of the methods are discussed. The future research directions in this area include, but are not limited to, further developing and refining methodologies, including algorithm development, to directly embed physics-based models into ML; exploring the usefulness of PIML for reservoir simulations; and adapting the new developments of how the physics and ML are incorporated in other communities to the well-production prediction. Finally, the methodologies we discuss in the paper can be generally applied to conventional reservoirs as well, although the focus here is on unconventional reservoirs.

Value Co-Creation Practices in the Decommissioning of Offshore Platforms: A Case Study Approach

2021 ◽  
pp. 239496432110320
Author(s):  
Francesca Loia ◽  
Vincenzo Basile ◽  
Nancy Capobianco ◽  
Roberto Vona
Keyword(s):  
Oil And Gas ◽  
Oil And Gas Industry ◽  
Future Research ◽  
Offshore Platforms ◽  
Gas Industry ◽  
Study Approach ◽  
Collaborative Interactions ◽  
Exploratory Approach ◽  
Italian Context

Over the years, value co-creation practices have become increasingly more important by supporting collaborative interactions and the achievement of sustainable and mutual competitive advantage between the ecosystem’ actors. In this direction, the oil and gas industry is proposing a sustainable re-use of offshore platforms based on value co-creation and resources exchange between the actors involved. According to this consideration, this work aims at re-reading the decommissioning of offshore platforms in the light of value co-creation practices, trying to capture the factors that governments and companies can leverage to pursue a sustainable development of local communities. To reach this goal, this work follows an exploratory approach by using, in particular, the case study. Specifically, one of the most notably projects in the Italian context have been chosen, the Paguro platform, in order to provide empirical insights into the nature of these value co-creation processes. Five value co-creation practices have been identified which highlight the importance of synergistic efforts of institutions, companies and technology-based platforms for improving the ability to co-create and capture value in the process of decommissioning. This exploratory work establishes a foundation for future research, and offers theoretical and managerial guidance in this increasingly important area.

Applied Mechanics Problems in the Oil and Gas Industry

1986 ◽  
Vol 39 (11) ◽  
pp. 1687-1696 ◽  
Author(s):  
Jean-Claude Roegiers
Keyword(s):  
Oil And Gas ◽  
Petroleum Industry ◽  
Oil Industry ◽  
Oil And Gas Industry ◽  
Production Equipment ◽  
Paper Briefly ◽  
Research Activity ◽  
Gas Industry ◽  
Well Production ◽  
Current Research Activity

The petroleum industry offers a broad spectrum of problems that falls within the domain of expertise of mechanical engineers. These problems range from the design of well production equipment to the evaluation of formation responses to production and stimulation. This paper briefly describes various aspects and related difficulties with which the oil industry has to deal, from the time the well is spudded until the field is abandoned. It attempts to delineate the problems, to outline the approaches presently used, and to discuss areas where additional research is needed. Areas of current research activity also are described; whenever appropriate, typical or pertinent case histories are used to illustrate a point.

Evaluation of Machine-Learning Tools for Predicting Sand Production

2021 ◽  
Author(s):  
Afungchwi Ronald Ngwashi ◽  
David O. Ogbe ◽  
Dickson O. Udebhulu
Keyword(s):  
Machine Learning ◽  
Niger Delta ◽  
Oil And Gas ◽  
Back Propagation ◽  
Oil And Gas Industry ◽  
Learning Tools ◽  
Sand Production ◽  
Data Set ◽  
Test Set ◽  
Gas Industry

Abstract Data analytics has only recently picked the interest of the oil and gas industry as it has made data visualization much simpler, faster, and cost-effective. This is driven by the promising innovative techniques in developing artificial intelligence and machine-learning tools to provide sustainable solutions to ever-increasing problems of the petroleum industry activities. Sand production is one of these real issues faced by the oil and gas industry. Understanding whether a well will produce sand or not is the foundation of every completion job in sandstone formations. The Niger Delta Province is a region characterized by friable and unconsolidated sandstones, therefore it's more prone to sanding. It is economically unattractive in this region to design sand equipment for a well that will not produce sand. This paper is aimed at developing a fast and more accurate machine-learning algorithm to predict sanding in sandstone formations. A two-layered Artificial Neural Network (ANN) with back-propagation algorithm was developed using PYTHON programming language. The algorithm uses 11 geological and reservoir parameters that are associated with the onset of sanding. These parameters include depth, overburden, pore pressure, maximum and minimum horizontal stresses, well azimuth, well inclination, Poisson's ratio, Young's Modulus, friction angle, and shale content. Data typical of the Niger Delta were collected to validate the algorithm. The data was further split into a training set (70%) and a test set (30%). Statistical analyses of the data yielded correlations between the parameters and were plotted for better visualization. The accuracy of the ANN algorithm is found to depend on the number of parameters, number of epochs, and the size of the data set. For a completion engineer, the answer to the question of whether or not a well will require sand production control is binary-either a well will produce sand or it does not. Support vector machines (SVM) are known to be better suited as the machine-learning tools for binary identification. This study also presents a comparative analysis between ANN and SVM models as tools for predicting sand production. Analysis of the Niger Delta data set indicated that SVM outperformed ANN model even when the training data set is sparse. Using the 30% test set, ANN gives an accuracy, precision, recall, and F1 - Score of about 80% while the SVM performance was 100% for the four metrics. It is then concluded that machine learning tools such as ANN with back-propagation and SVM are simple, accurate, and easy-to-use tools for effectively predicting sand production.

scholarly journals Characterization of fatal injuries in oil and gas industry-related helicopter accidents in the Gulf of Mexico, 2004–2014

2020 ◽  
Vol 7 (1) ◽  
Author(s):  
Kristin Yeoman ◽  
Mary B. O’Connor ◽  
Sara Sochor ◽  
Gerald Poplin
Keyword(s):  
Lower Extremity ◽  
Gulf Of Mexico ◽  
Oil And Gas ◽  
Oil And Gas Industry ◽  
Body Region ◽  
Future Research ◽  
Gas Industry ◽  
Body Regions ◽  
Safety Features

Abstract Background Transportation events are the most common cause of offshore fatalities in the oil and gas industry, of which helicopter accidents comprise the majority. Little is known about injury distributions in civilian helicopter crashes, and knowledge of injury distributions could focus research and recommendations for enhanced injury prevention and post-crash survival. This study describes the distribution of injuries among fatalities in Gulf of Mexico oil and gas industry-related helicopter accidents, provides a detailed injury classification to identify potential areas of enhanced safety design, and describes relevant safety features for mitigation of common injuries. Methods Decedents of accidents during 2004–2014 were identified, and autopsy reports were requested from responsible jurisdictions. Documented injuries were coded using the Abbreviated Injury Scale (AIS), and frequency and proportion of injuries by AIS body region and severity were calculated. Injuries were categorized into detailed body regions to target areas for prevention. Results A total of 35 autopsies were coded, with 568 injuries documented. Of these, 23.4% were lower extremity, 22.0% were thorax, 13.6% were upper extremity, and 13.4% were face injuries. Minor injuries were most prevalent in the face, neck, upper and lower extremities, and abdomen. Serious or worse injuries were most prevalent in the thorax (53.6%), spine (50.0%), head (41.7%), and external/other regions (75.0%). The most frequent injuries by detailed body regions were thoracic organ (23.0%), thoracic skeletal (13.3%), abdominal organ (9.6%), and leg injuries (7.4%). Drowning occurred in 13 (37.1%) of victims, and drowning victims had a higher proportion of moderate brain injuries (7.8%) and lower number of documented injuries (3.8) compared with non-drowning victims (2.9 and 9.4%, respectively). Conclusions Knowledge of injury distributions focuses and prioritizes the need for additional safety features not routinely used in helicopters. The most frequent injuries occurred in the thorax and lower extremity regions. Future research requires improved and expanded data, including collection of detailed data to allow characterization of both injury mechanism and distribution. Improved safety systems including airbags and helmets should be implemented and evaluated for their impact on injuries and fatalities.

scholarly journals A Common Risk Classification Concept for Safety Related Gas Leaks and Fugitive Emissions?

Energies ◽  
2019 ◽  
Vol 12 (21) ◽  
pp. 4063 ◽  
Author(s):  
Log ◽  
Pedersen
Keyword(s):  
Oil And Gas ◽  
Cost Benefit ◽  
Oil And Gas Industry ◽  
Leak Rate ◽  
Future Research ◽  
Fugitive Emissions ◽  
Gas Industry ◽  
Cost Benefit Ratio ◽  
Gas Leaks ◽  
Categorization Model

Gas leaks in the oil and gas industry represent a safety risk as they, if ignited, may result in severe fires and/or explosions. Unignited, they have environmental impacts. This is particularly the case for methane leaks due to a significant Global Warming Potential (GWP). Since gas leak rates may span several orders of magnitude, that is, from leaks associated with potential major accidents to fugitive emissions on the order of 10−6 kg/s, it has been difficult to organize the leaks in an all-inclusive leak categorization model. The motivation for the present study was to develop a simple logarithmic table based on an existing consequence matrix for safety related incidents extended to include non-safety related fugitive emissions. An evaluation sheet was also developed as a guide for immediate risk evaluations when new leaks are identified. The leak rate table and evaluation guide were tested in the field at five land-based oil and gas facilities during Optical Gas Inspection (OGI) campaigns. It is demonstrated how the suggested concept can be used for presenting and analysing detected leaks to assist in Leak Detection and Repair (LDAR) programs. The novel categorization table was proven valuable in prioritizing repair of “super-emitter” components rather than the numerous minor fugitive emissions detected by OGI cameras, which contribute little to the accumulated emissions. The study was limited to five land based oil and gas facilities in Norway. However, as the results regarding leak rate distribution and “super-emitter” contributions mirror studies from other regions, the methodology should be generally applicable. To emphasize environmental impact, it is suggested to include leaking gas GWP in future research on the categorization model, that is, not base prioritization solely on leak rates. Research on OGI campaign frequency is recommended since frequent coarse campaigns may give an improved cost benefit ratio.

Russian Scanners for Eddy Current Testing of Drilling Pipe Elements in the Oil and Gas Industry

NDT World ◽  
2020 ◽  
pp. 5-8
Author(s):  
Aleksandr Kazachenko
Keyword(s):  
Control Charts ◽  
Oil And Gas ◽  
Oil And Gas Industry ◽  
Complex Problem ◽  
Point Of View ◽  
Destructive Testing ◽  
Gas Industry ◽  
Shewhart Control ◽  
Inspection Process ◽  
Composite Pipes

Composite materials appear to be an ideal solution to a complex problem with conflicting conditions: how to simultaneously obtain sufficient strength, reliability and durability of the structure, while providing the minimum possible mass of it. However, non-destructive testing of products from them raises more and more questions. In the mass production of composite pipes for pipelines, the only possible option from the point of view of ensuring the necessary reliability, information capacity of the results of the performed inspection of products and productivity is the automation of the inspection process, which includes special methods for identifying defects. Statistical methods, including capability ratio and Shewhart control charts, should be used to estimate the error in determining the size of defects.

Electrical Submersible Pump Prognostics and Health Monitoring Using Machine Learning and Natural Language Processing

2021 ◽  
Author(s):  
Rajeev Ranjan Sinha ◽  
Supriya Gupta ◽  
Praprut Songchitruksa ◽  
Saniya Karnik ◽  
Amey Ambade
Keyword(s):  
Machine Learning ◽  
Language Processing ◽  
Oil And Gas ◽  
Oil And Gas Industry ◽  
Operating Conditions ◽  
Learning Approaches ◽  
Submersible Pump ◽  
Gas Industry ◽  
Electrical Submersible Pump ◽  
Low Efficiency

Abstract Electrical Submersible Pump (ESP) systems efficiently pump high volumes of production fluids from the wellbore to the surface. They are extensively used in the oil and gas industry due to their adaptability, low maintenance, safety and relatively low environmental impact. They require specific operating conditions with respect to the power, fluid level and fluid content. Oilfield operation workflows often require extensive surveillance and monitoring by subject-matter experts (SMEs). Detecting issues like formation of unwanted gas and emulsions in ESPs requires constant analysis of downhole data by SMEs. The lack of adequate and accurate monitoring of the downhole pumps can lead to low efficiency, high lifting costs, and frequent repair and replacements. There are 3 workflows described in the paper which demonstrate that the maintenance costs of the ESPs can be significantly reduced, and production optimized with the augmentation of machine learning approaches typically unused in ESP surveillance and failure analysis.

Machine Learning as Accelerating Tool in Remote Operation Realisation through Monitoring Oil and Gas Equipments and Identifying its Failure Mode

2021 ◽  
Author(s):  
Ahmad Naufal Naufal ◽  
Samy Abdelhamid Samy ◽  
Nenisurya Hashim Nenisurya ◽  
Zaharuddin Muhammad Zaharuddin ◽  
Eddy Damsuri Eddy ◽  
...  
Keyword(s):  
Machine Learning ◽  
Failure Mode ◽  
Oil And Gas ◽  
Oil And Gas Industry ◽  
Support Vector ◽  
Environmental Damage ◽  
Equipment Failure ◽  
Statistical Regression ◽  
Gas Industry ◽  
Plant Equipment

Abstract Equipment failure, unplanned downtime operation, and environmental damage cost represent critical challenges in overall oil and gas business from well reservoir identification and drilling strategy to production and processing. Identifying and managing the risks around assets that could fail and cause redundant and expensive downtime are the core of plant reliability in oil and gas industry. In the current digital era; there is an essential need of innovative data-driven solutions to address these challenges, especially, monitoring and diagnosis of plant equipment operations, recognize equipment failure; avoid unplanned downtime; repair costs and potential environmental damage; maintaining reliable production, and identifying equipment failures. Machine learning-artificial intelligence application is being studied to develop predictive maintenance (PdM) models as innovative analytics solution based on real-data streaming to get to an elevated level of situational intelligence to guide actions and provide early warnings of impending asset failure that previously remained undetected. This paper proposes novel machine learning predictive models based on extreme learning/support vector machines (ELM-SVM) to predict the time to failure (TTF) and when a plant equipment(s) will fail; so maintenance can be planned well ahead of time to minimize disruption. Proper visualization with deep-insights (training and validation) processes of the available mountains of historian and real-time data are carried out. Comparative studies of ELM-SVM techniques versus the most common physical-statistical regression techniques using available rotating equipment-compressors and time-failure mode data. Results are presented and it is promising to show that the new machine learning (ELM-SVM) techniques outperforms physical-statistics techniques with reliable and high accurate predictions; which have a high impact on the future ROI of oil and gas industry.

Sign in / Sign up

Export Citation Format

Share Document