Shale brittleness prediction using machine learning - A Middle East basin case study

Mustafa, Ayyaz; Tariq, Zeeshan; Abdulraheem, Abdulazeez; Mahmoud, Mohamed E.; Kalam, Shams; Khan, Rizwan Ahmed

doi:10.1306/12162120181

Cited by 5 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, the use of gas hydrate field data for machine learning models is discussed to provide state-of-the-art knowledge and advance machine learning in gas hydrate applications. For hydrate predictions, just like other predictive models in drilling, fracturing, and shale studies, ANN models are mostly used. − An ANN and SVR machine learning model was developed by Qin using field data from a dry tree facility in the Gulf of Mexico. The models were programmed to predict or detect gas hydrate formation plugs and the formation conditions.…”

Section: Machine Learning In Hydrate Field Data Predictionmentioning

confidence: 99%

“…For hydrate predictions, just like other predictive models in drilling, fracturing, and shale studies, ANN models are mostly used. 84 − 88 An ANN and SVR machine learning model was developed by Qin 50 using field data from a dry tree facility in the Gulf of Mexico. The models were programmed to predict or detect gas hydrate formation plugs and the formation conditions.…”

Section: Machine Learning In Hydrate Field Data Predictionmentioning

confidence: 99%

See 1 more Smart Citation

Research Advances in Machine Learning Techniques in Gas Hydrate Applications

Osei,

Bavoh,

Lal

2024

ACS Omega

View full text Add to dashboard Cite

The complex modeling accuracy of gas hydrate models has been recently improved owing to the existence of data for machine learning tools. In this review, we discuss most of the machine learning tools used in various hydrate-related areas such as phase behavior predictions, hydrate kinetics, CO 2 capture, and gas hydrate natural distribution and saturation. The performance comparison between machine learning and conventional gas hydrate models is also discussed in detail. This review shows that machine learning methods have improved hydrate phase property predictions and could be adopted in current and new gas hydrate simulation software for better and more accurate results.

show abstract

Section: Machine Learning In Hydrate Field Data Predictionmentioning

confidence: 99%

Section: Machine Learning In Hydrate Field Data Predictionmentioning

confidence: 99%

Research Advances in Machine Learning Techniques in Gas Hydrate Applications

Osei,

Bavoh,

Lal

2024

ACS Omega

View full text Add to dashboard Cite

show abstract

“…In 1959, Arthur Samuel proposed the idea of machine learning for the first time . This method can learn appropriate and effective characteristics from large amounts of complex data. − Since its introduction, machine learning has solved many complex prediction problems in mining, such as drilling fluid lost-circulation, coalbed methane production, coal mining, nuclear magnetic resonance porosity, shale brittleness, etc.…”

Section: Introductionmentioning

confidence: 99%

Mining Technology Evaluation for Steep Coal Seams Based on a GA-BP Neural Network

Li,

Wang,

et al. 2024

ACS Omega

View full text Add to dashboard Cite

Many mines in Guizhou Province are in urgent need of renovation to ensure harmonious operation and prolong their lifespan. The key to successful renovation lies in the prudent selection of the appropriate mining technologies. Therefore, a comprehensive investigation was conducted on steep coal mines in Guizhou Province, and a comprehensive evaluation framework was established. Spearman correlation analysis was performed on various factors, selecting geological conditions and working face parameters with high correlation as the input variables and mining methods as the output variables. The optimal values of each hyperparameter were determined through orthogonal experiments, and the neural network structure was confirmed to be “17-9-3”. Five variants of backpropagation (BP) algorithms were meticulously tested, and a genetic algorithm optimizing the BP neural network (GA-BP) was further assessed to improve the model’s prediction accuracy. The accuracy of the model was evaluated via the coefficient of determination (R 2) and mean squared error (MSE). The research results indicated that the variable step–size algorithm with a momentum term (VSS + MT) was the optimal algorithm for the BP neural network. Additionally, the MSE values of the artificial neural network and GA-BP neural network in the testing phase were 0.06 and 0.04, with prediction success rates of 70 and 90%, respectively, and R 2 values of 0.79 and 0.85, respectively. Thus, the GA-BP neural network demonstrated superior performance. Finally, industrial application of the model was conducted on a working face in the Zhong-Yu coal mine. The evaluation index for the working face was “0.847, 0.09, 0.111”, suggesting that fully mechanized mining should be adopted. The evaluation results were consistent with the current production status of the mine, verifying the reliability of the model in practical applications.

show abstract