Data-driven machine learning approach to predict mineralogy of organic-rich shales: An example from Qusaiba Shale, Rub’ al Khali Basin, Saudi Arabia

Mustafa, Ayyaz; Tariq, Zeeshan; Mahmoud, Mohamed; Radwan, Ahmed E.; Abdulraheem, Abdulazeez; Abouelresh, Mohamed O.

doi:10.1016/j.marpetgeo.2021.105495

Cited by 34 publications

(20 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“… 17 − 23 Adaptive neuro-fuzzy inference systems (ANFIS), artificial neural network (ANN), fuzzy logic, and group method of data handling techniques have been effective in obtaining the mineralogy of organic-rich shales, the oil formation volume factor, the fractured well productivity, the natural gas density of pure and mixed hydrocarbons, the breakdown pressure of unconventional reservoirs, and the critical total drawdown for the sand production. 24 − 30 …”

Section: Introductionmentioning

confidence: 99%

“…Nowadays, AI-based models have become a hot topic in engineering applications and are efficiently applied in many petroleum engineering calculations. − Deep learning and gradient boosting methods were successfully conducted to determine complex carbonate rock’s permeability, capillary pressure, relative permeability, and the optimum operational conditions for CO 2 foam enhanced oil recovery. − Adaptive neuro-fuzzy inference systems (ANFIS), artificial neural network (ANN), fuzzy logic, and group method of data handling techniques have been effective in obtaining the mineralogy of organic-rich shales, the oil formation volume factor, the fractured well productivity, the natural gas density of pure and mixed hydrocarbons, the breakdown pressure of unconventional reservoirs, and the critical total drawdown for the sand production. − …”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Determination of the Gas–Oil Ratio below the Bubble Point Pressure Using the Adaptive Neuro-Fuzzy Inference System (ANFIS)

et al. 2022

View full text Add to dashboard Cite

Determining the solution gas–oil ratio ( R s ) below the bubble point is a vital requirement that aids in multiple production engineering and reservoir analysis issues. Currently, there are some models available for the determination of the solution gas–oil ratio under the bubble point. However, they still may prove unreliable due to the applied assumptions and their specification to operate only under a particular range of data. In this study, the neuro-fuzzy, i.e., the adaptive neuro-fuzzy inference system (ANFIS) approach, is utilized to develop an accurate and dependable model for determining the R s below the bubble point pressure. A total of 376 pressure–volume–temperature datasets from Sudanese oil fields were used to establish the proposed ANFIS model. The trend analysis was applied to affirm the proper relationships between the inputs and outputs. Furthermore, using different statistical error analyses, the developed model was benchmarked against widely used empirical methods to evaluate the proposed method’s performance in predicting the R s at pressures below the bubble point. The proposed ANFIS model performs with an average absolute percent relative error of 10.60% and a correlation coefficient of 99.04%, surpassing the previously studied correlations.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Determination of the Gas–Oil Ratio below the Bubble Point Pressure Using the Adaptive Neuro-Fuzzy Inference System (ANFIS)

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Furthermore, Mahmoud et al applied machine learning to acid cracking test data points to establish several conductivity correlations and proposed an artificial neural network model to predict the fracture conductivity of carbonate rocks and pointed out that all AI models are data driven. Tariq et al and Mustafa et al , used data-driven machine learning to predict fracture pressure and mineralogy. Anemangely et al and Bajolvand et al , performed drilling rate prediction with the multilayer perceptron neural network and convolutional neural network, and these methods illustrate that the application of artificial neural models for ROP prediction is a feasible and useful method.…”

Section: Introductionmentioning

confidence: 99%

An Advanced Long Short-Term Memory (LSTM) Neural Network Method for Predicting Rate of Penetration (ROP)

Lou

Cheng

et al. 2022

ACS Omega

View full text Add to dashboard Cite

Rate of penetration (ROP) is an essential factor in drilling optimization and reducing the drilling cycle. Most of the traditional ROP prediction methods are based on building physical model and single intelligent algorithms, and the efficiency and accuracy of these prediction methods are very low. With the development of artificial intelligence, high-performance algorithms make reliable prediction possible from the data perspective. To improve ROP prediction efficiency and accuracy, this paper presents a method based on particle swarm algorithm for optimization of long short-term memory (LSTM) neural networks. In this paper, we consider the Tuha Shengbei block oilfield as an example. First, the Pearson correlation coefficient is used to measure the correlation between the characteristics and eight parameters are screened out, namely, the depth of the well, gamma, formation density, pore pressure, well diameter, drilling time, displacement, and drilling fluid density. Second, the PSO algorithm is employed to optimize the super-parameters in the construction of the LSTM model to the predict ROP. Third, we assessed model performance using the determination coefficient (R 2 ), root mean square error (RMSE), and mean absolute percentage error (MAPE). The evaluation results show that the optimized LSTM model achieves an R 2 of 0.978 and RMSE and MAPE are 0.287 and 12.862, respectively, hence overperforming the existing methods. The average accuracy of the optimized LSTM model is also improved by 44.2%, indicating that the prediction accuracy of the optimized model is higher. This proposed method can help to drill engineers and decision makers to better plan the drilling operation scheme and reduce the drilling cycle.

show abstract

“…Geomechanical studies in many areas worldwide are dependent on empirical equations for geomechanical parameter estimation, which may work in some places but not always (Sarkar et al 2012;Suorineni 2014a, b;Najibi et al 2015;Iramina 2018). Machine learning techniques have been widely used in the oil and gas industry as a powerful tool for prediction of several vital parameters in the energy industry (e.g., Vo Thanh et al 2020;Ashraf et al 2020Ashraf et al , 2021Rajabi et al 2021;Mustafa et al 2022;Safaei-Farouji et al 2022;Radwan et al 2022). Geoscientists and petroleum engineers have applied machine learning application on well logs and other parameters to infer the most critical geomechanics parameters (e.g., Miah 2020;Mohamadian et al 2021;Kor et al 2021;Radwan et al 2022).…”

Section: Introductionmentioning

confidence: 99%

Implications of machine learning on geomechanical characterization and sand management: a case study from Hilal field, Gulf of Suez, Egypt

Abdelghany

Hammed

Radwan

et al. 2022

J Petrol Explor Prod Technol

Self Cite

View full text Add to dashboard Cite

Sand production is one of the major challenges in the oil and gas industry, so a comprehensive geomechanical analysis is necessary to mitigate sand production in mature fields. As the pore pressure drastically decline in depleted reservoirs, the sand production risk becomes more critical and needs to be studied. However, the absence of key logs in many wells is a big challenge in the petroleum industry, and most geologists and engineers use empirical equations to predict missed log intervals. We conducted a comprehensive geomechanical modeling study on a full set of logs from two wells from the Hilal field, Gulf of Suez, Egypt, to infer the geomechanical elements and predict sand production. We have used the multi-arm calipers to calculate the actual depth of damage ratio to validate the geomechanical parameters in the prognosis model and confirm the stress orientations. We used machine learning approach to infer key sonic log in X-10 well to replace the empirical equations. The multi-arm calipers analysis showed an observed anisotropy in the hole diameter size with more enlargement in the ENE direction and fits with the minimum horizontal stress direction in the direction of N 60oE. The later also deduced the maximum horizontal stress direction in N150 ° based on the induced fractures from borehole image data in a nearby field. We developed and compared two sand management models: one using empirical equation and the other using machine learning. The model driven by the Gardner equation suggests sand production from day one, which is not matched with the production data, while the model driven by machine learning suggests no sand production risk, which is matched with the actual production data. Our results demonstrate the advantage of using machine learning technique in geomechanical studies on the classical empirical equations in the area of study that can be applied in other basins. The findings of this study can help with a better understanding of the implications of machine learning on geomechanical characterization and sand management.

show abstract

Data-driven machine learning approach to predict mineralogy of organic-rich shales: An example from Qusaiba Shale, Rub’ al Khali Basin, Saudi Arabia

Cited by 34 publications

References 45 publications

Determination of the Gas–Oil Ratio below the Bubble Point Pressure Using the Adaptive Neuro-Fuzzy Inference System (ANFIS)

Determination of the Gas–Oil Ratio below the Bubble Point Pressure Using the Adaptive Neuro-Fuzzy Inference System (ANFIS)

An Advanced Long Short-Term Memory (LSTM) Neural Network Method for Predicting Rate of Penetration (ROP)

Implications of machine learning on geomechanical characterization and sand management: a case study from Hilal field, Gulf of Suez, Egypt

Contact Info

Product

Resources

About