Metaheuristic algorithm integrated neural networks for well-test analyses of petroleum reservoirs

Pandey, Rakesh; Aggarwal, Shrey; Nath, Griesha; Kumar, Anil; Vaferi, Behzad

doi:10.1038/s41598-022-21075-w

Cited by 13 publications

(4 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The CNN has k filters of size R1 × 1 that are convolved with the input matrix to produce k feature maps. The rectified linear unit (ReLU) activated CNN gives the output of shape v × k. The LSTM layer with p units and a hyperbolic tangent (Tanh) activation function 39 provides feedback connection to carry forward the relevant information. Finally, the LSTM output delivers to the DNN layer with D neurons and a softmax activation function to provide the final classification results.…”

Section: Deep Learning-based Paradigmmentioning

confidence: 99%

Supervised deep learning-based paradigm to screen the enhanced oil recovery scenarios

Pandey

Gandomkar

Vaferi

et al. 2023

Sci Rep

Self Cite

View full text Add to dashboard Cite

High oil prices and concern about limited oil reserves lead to increase interest in enhanced oil recovery (EOR). Selecting the most efficient development plan is of high interest to optimize economic cost. Hence, the main objective of this study is to construct a novel deep-learning classifier to select the best EOR method based on the reservoir’s rock and fluid properties (depth, porosity, permeability, gravity, viscosity), and temperature. Our deep learning-based classifier consists of a one-dimensional (1D) convolutional neural network, long short-term memory (LSTM), and densely connected neural network layers. The genetic algorithm has been applied to tune the hyperparameters of this hybrid classifier. The proposed classifier is developed and tested using 735 EOR projects on sandstone, unconsolidated sandstone, carbonate, and conglomerate reservoirs in more than 17 countries. Both the numerical and graphical investigations approve that the structure-tuned deep learning classifier is a reliable tool to screen the EOR scenarios and select the best one. The designed model correctly classifies training, validation, and testing examples with an accuracy of 96.82%, 84.31%, and 82.61%, respectively. It means that only 30 out of 735 available EOR projects are incorrectly identified by the proposed deep learning classifier. The model also demonstrates a small categorical cross-entropy of 0.1548 for the classification of the involved enhanced oil recovery techniques. Such a powerful classifier is required to select the most suitable EOR candidate for a given oil reservoir with limited field information.

show abstract

Section: Deep Learning-based Paradigmmentioning

confidence: 99%

Supervised deep learning-based paradigm to screen the enhanced oil recovery scenarios

Pandey

Gandomkar

Vaferi

et al. 2023

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…The integration of machine learning and metaheuristics is a new area that has become increasingly important in recent years. For example, in Kumar Pandey et al ( 2022 ), a genetic algorithm was used to improve the prediction accuracy of a deep network. A tuning of hyperparameters based on metaheuristics was performed and its advantages were discussed.…”

Section: Metaheuristic Integration To Optimize Deep Network: a Case S...mentioning

confidence: 99%

Discovering anomalies in big data: a review focused on the application of metaheuristics and machine learning techniques

Cavallaro,

Cutello,

Pavone

et al. 2023

Front. Big Data

View full text Add to dashboard Cite

With the increase in available data from computer systems and their security threats, interest in anomaly detection has increased as well in recent years. The need to diagnose faults and cyberattacks has also focused scientific research on the automated classification of outliers in big data, as manual labeling is difficult in practice due to their huge volumes. The results obtained from data analysis can be used to generate alarms that anticipate anomalies and thus prevent system failures and attacks. Therefore, anomaly detection has the purpose of reducing maintenance costs as well as making decisions based on reports. During the last decade, the approaches proposed in the literature to classify unknown anomalies in log analysis, process analysis, and time series have been mainly based on machine learning and deep learning techniques. In this study, we provide an overview of current state-of-the-art methodologies, highlighting their advantages and disadvantages and the new challenges. In particular, we will see that there is no absolute best method, i.e., for any given dataset a different method may achieve the best result. Finally, we describe how the use of metaheuristics within machine learning algorithms makes it possible to have more robust and efficient tools.

show abstract

“…A atualização contínua de modelos geológicos de reservatório é necessária em campos em produção, onde a frequência das perfurações de poços e o comportamento dinâmico decorrente da produção exigem atualizações rápidas. A incorporação de novos dados é beneficiada pelo desenvolvimento de ferramentas baseadas em machine learning e IA, mas a distribuição de informações geológicas ainda demonstra ser muito dependente da homogeneidade e previsibilidade do meio que está sendo representado (e.g., Pandey et al, 2022).…”

Section: Atualização De Modelos Geológicosunclassified

Machine learning, inteligência artificial (IA), e a caracterização da subsuperfície: aplicações, possibilidades e riscos

Berton

2023

Lat. Am. J. Energy Res.

View full text Add to dashboard Cite

Soluções computacionais baseadas em machine learning e inteligência artificial (IA) têm sido adaptadas para o processamento e interpretação das informações de subsuperfície. Diante do que pode se tornar uma revolução tecnológica, é necessário pensar nos prováveis impactos das novas soluções para as atividades relacionadas à caracterização de reservatórios e sistemas petrolíferos. A IA já prova ter utilidade na indústria de óleo e gás ao lidar com grandes volumes de dados geológicos de padrão homogêneo, poupando o usuário humano de tarefas repetitivas. Essa característica faz com que os programas sejam úteis para o aumento de eficiência e da segurança do trabalho, mas da forma como funcionam atualmente, eles estão longe de serem capazes de lidar com a frequente complexidade geológica que pode representar riscos ou oportunidades em subsuperfície. Nem mesmo os melhores programas de IA são capazes de solucionar problemas inerentes aos dados indiretos, como a baixa resolução sísmica ou a baixa representatividade de poços. Tampouco conseguem encontrar soluções plausíveis para situações geológicas complexas. As novas soluções tecnológicas devem ser vistas como ferramentas para facilitar a vida dos profissionais de subsuperfície. Como qualquer ferramenta, sua existência tem propósitos específicos que não abrangem toda a complexidade dos sistemas geológicos. As interpretações geológicas derivadas de programas baseados em machine learning e IA devem ser avaliadas como aproximações geoestatísticas, não como a representação total da realidade construída a partir de múltiplos processos naturais. Essa característica faz com que os profissionais das geociências sigam sendo essenciais para aplicar as ferramentas de IA corretamente e filtrar as informações produzidas por elas.

show abstract

Metaheuristic algorithm integrated neural networks for well-test analyses of petroleum reservoirs

Cited by 13 publications

References 46 publications

Supervised deep learning-based paradigm to screen the enhanced oil recovery scenarios

Supervised deep learning-based paradigm to screen the enhanced oil recovery scenarios

Discovering anomalies in big data: a review focused on the application of metaheuristics and machine learning techniques

Machine learning, inteligência artificial (IA), e a caracterização da subsuperfície: aplicações, possibilidades e riscos

Contact Info

Product

Resources

About