Computational Intelligence and Neuroscience

Computational Intelligence and Neuroscience / 2021 / Article

Research Article | Open Access

Volume 2021 |Article ID 9960478 | https://doi.org/10.1155/2021/9960478

Hany Gamal, Salaheldin Elkatatny, Ahmed Alsaihati, Abdulazeez Abdulraheem, "Intelligent Prediction for Rock Porosity While Drilling Complex Lithology in Real Time", Computational Intelligence and Neuroscience, vol. 2021, Article ID 9960478, 12 pages, 2021. https://doi.org/10.1155/2021/9960478

Intelligent Prediction for Rock Porosity While Drilling Complex Lithology in Real Time

Academic Editor: Qiangqiang Yuan
Received20 Mar 2021
Revised27 Apr 2021
Accepted07 Jun 2021
Published15 Jun 2021

Abstract

Rock porosity is an important parameter for the formation evaluation, reservoir modeling, and petroleum reserve estimation. The conventional methods for determining the rock porosity are considered costly and time-consuming operations during the well drilling. This paper aims to predict the rock porosity in real time while drilling complex lithology using machine learning. In this paper, two intelligent models were developed utilizing the random forest (RF) and decision tree (DT) techniques. The drilling parameters include weight on bit, torque, standpipe pressure, drill string rotation speed, rate of penetration, and pump rate. Two datasets were employed for building the models (3767 data points) and for validating the developed models (1676 data points). Both collected datasets have complex lithology of carbonate, sandstone, and shale. Sensitivity and optimization on different parameters for each technique were conducted to ensure optimum prediction. The models’ performance was checked by four performance indices which are coefficient of determination (R2), average absolute percentage error (AAPE), variance account for (VAF), and a20 index. The results indicated the strong porosity prediction capability for the two models. DT model showed R2 of 0.94 and 0.87 between the predicted and actual porosity values with AAPE of 6.07 and 9% for training and testing, respectively. Generally, RF provided a higher level of strong prediction than DT as RF achieved R2 of 0.99 and 0.90 with AAPE of 1.5 and 7% for training and testing, respectively. The models’ validation proved a high prediction performance as DT achieved R2 of 0.88 and AAPE of 8.58%, while RF has R2 of 0.92 and AAPE of 6.5%.

1. Introduction

The porosity of the rock is commonly defined as the ratio between the void pore spaces in the rock to the total bulk volume for the rock, and this space will provide the storage capacity for the petroleum fluids if it is connected. The rock porosity is a vital petrophysical property as it has a great impact on the reservoir reserve estimation, and as a result, for the field development decision-making [1, 2]. The precise determination of the rock porosity will significantly affect the petroleum reserve estimation and economics [3], and hence, the accuracy of the porosity determination will play a huge role.

Determining the rock porosity can be achieved practically by direct and indirect measurement or prediction using empirical equations. Laboratory measurements for the cored rock samples are the direct way to measure the rock porosity [4]; however, determining the porosity from the porosity logs is considered an indirect way to acquire the porosity values [5]. Each technique has its pros and cons from technical and economic aspects as the lab direct measurements for the porosity is considered the most relative accurate way; however, this technique is costly and time-consuming and covers only the cored interval within the reservoir or the drilled sections [6, 7]. On the contrary, the porosity determination by the well logging tools has many types of error as the operational calibration for the logging tool, in addition to the borehole and mud impact on the measurement [8, 9]. A recent technique is introduced to the field applications of rock characterization and rock porosity measurement by employing the drilled cuttings; however, the technique required special cuttings size and advanced sample preparation [10].

Determining the rock porosity from the well logs data approach was studied in the literature where the rock porosity was obtained based on other petrophysical logs [11, 12]. Nuclear magnetic resonance measurement was introduced for determining the rock porosity [13]. However, such techniques required the logging data or lab measurements to determine the rock porosity values that required extra cost and time.

The applications of artificial intelligence (AI) techniques provided huge contributions for dealing with petroleum data in different disciplines. AI tools such as artificial neural networks (ANNs), fuzzy logic (FL), expert systems, support vector machines (SVMs), functional networks (FN), and case-based reasoning provided high performance and accurate prediction results [14]. The implementation of such tools contributed to solving many technical problems such as estimation and optimization of drilling parameters [1519], predicting and monitoring the drilling fluids properties [2024], reservoir fluid properties [2530], rock permeability estimation [31, 32], and rock strength and geomechanical properties [3337]. These applications provided different approaches in terms of the AI techniques and the input parameters.

The porosity prediction by employing artificial intelligence techniques was studied in the literature as shown in Table 1. The table shows the input parameters for predicting the rock porosity, data points, rock formation type for the study, correlation coefficient (R) between the predicted and the actual porosity values, and the AI techniques that were employed to build the prediction models.


Input parametersData pointsFormationRMethodsRef.

Density, neutron porosity, and sonic compressional time1700Carbonate0.98ANN, SVM, ANFIS[38]
Deep resistivity, density, neutron porosity, and gamma-ray420Sandstone0.93Fuzzy, ANN[39]
GR, bulk density, resistivity, neutron porosity, and sonic travel time plus lithofacies and stratigraphic information1000Carbonate and unconventional0.98ANN[40]
ROP, Q, RPM, SPP, T, WOB2800Carbonate0.96ANN[41]
ROP, RPM, WOB, T, depth, SPP, Q, and mechanical specific energy89549Sandstone and shale0.6ANN[42]

The studies investigated the core porosity prediction using the well logging data as density, neutron porosity, sonic time, resistivity log, gamma-ray (GR), and stratigraphic information [3840]. In addition, the drilling data was employed for predicting the formation porosity using drilling parameters as the rate of penetration (ROP), pump rate (Q), drill string rotating speed (RPM), standpipe pressure (SPP), torque (T), weight on bit (WOB), and mechanical specific energy [41, 42].

As shown in the literature, drilling data was employed but for carbonate formation during drilling horizontal well [41], and another study for sandstone and shale formations but with incorporating the mechanical specific energy as an additional input to the drilling data; furthermore, the model accuracy was low with a correlation coefficient between the predicted and actual porosity values of 0.6 [42]. The novel contributions for this research are generating the formation neutron porosity log from only the available surface drilling parameters for complex lithology drilled rocks with high accuracy using AI models. The current study predicted the porosity using a collected drilling data during drilling complex lithology formations that have carbonate, sand, and shale formations. The study introduced two AI models for the porosity prediction using decision tree (DT) and random forest (RF) tools. The obtained models from this study will help to save the operational cost and time to log or measure the rock porosity in the lab.

2. Materials and Methods

This research proposed two prediction models for the rock porosity using the drilling data as inputs. The study employed ANN and RF as AI tools for building the prediction model. Using these models, the porosity profile was generated with high accuracy for the whole drilled section that contains complex formation. Figure 1 represents the processing flow to provide robust models for rock porosity prediction. The data gathering and preprocessing include collecting the data from the drilling sensors and log data for data cleaning to remove the illogic values and outliers. The next step is to build the AI model structure and optimize the model parameters and learning algorithms in order to have good prediction results. The model went through training and testing processes and the results were checked by the statistical performance indices and if the accuracy level of the results is not high enough; then the model should be retrained for enhancing the accuracy level. By the end, the best model parameters and learning algorithm should be saved and reported.

2.1. Data Description and Statistics

The data in this study was collected during a drilling phase that covered the intermediate section for vertical wells. The drilled formations contain more than one rock type as sandstone, shale, and limestone that can be considered complex lithologies. The data covered 3767 readings for all the drilling parameters with the neutron porosity log after the data cleaning and preprocessing was used for building the machine learning models. Another dataset of 1670 data points was collected from the same drilling phase that was employed for validating the developed models. The drilling parameters include the surface drilling parameters as the weight on bit (WOB) in klb, torque (T) in kft.lbf, standpipe pressure (SPP) in psi, drill string rotary speed (RPM) in min−1, drilling rate of penetration (ROP) in ft/h, and mudflow rate (Q) in gpm.

2.2. Data Quality, Preprocessing, and Statistics

The collected data from the drilling sensors suffered from operational measurement and tool errors. And hence, the data should be preprocessed for removing the missing measurements, noise, and outliers by using a developed MATLAB code to ensure the data quality for developing the AI models.

Statistical analysis for the cleaned data shows the minimum, maximum, mean, standard deviation, kurtosis, skewness, and the data range for each parameter as shown in Table 2.


WOB (klbf)T (kft.lbf)SPP (psi)RPM (1/min)ROP (ft/h)Q (gpm)

Minimum1.54.32140.277.926.1627.00.055
Maximum26.711.03076.0162.5119.6854.00.429
Mean11.87.42600.9128.565.8724.90.207
Standard deviation7.31.8201.415.818.273.40.067
Kurtosis−1.3−1.1−0.7−0.6−0.9−1.30.828
Skewness0.30.2−0.1−0.50.30.60.552
Range25.16.7935.884.693.4227.00.375

From the data statistics, the drilling parameters and porosity indicated the wide range for the data that will enhance the prediction capabilities of the developed AI models. The statistics show that WOB ranged from 1.5 to 26.7 (klbf), T from 4.3 to 11.0 (kft.lbf), SPP from 2140 to 3076 psi, pipe speed from 77.9 to 162.5 (1/min), ROP from 26.1 to 119.6 (ft/h), flow rate ranged from 627 to 854 (gpm), and the target parameter from 0.055 to 0.429 that covered very tight rock class to high porous rock scale. Figure 2 represents the porosity histogram for the recorded value. The histogram shows that the frequency of the porosity profile changed with the recorded porosity; the porosity values below 0.2 recorded 44% of the total recorded frequency, 49% of the total frequencies were recorded for the porosity values from 0.2 to 0.3, and only 8% from the total frequency was observed for the higher porosity values greater than 0.3. Hence, the porosity database covered a wide range for the rock porosity data that enhances the capability of the prediction models.

The relationships between the drilling parameters (model inputs) and the rock porosity (model output) show a direct linear relationship between the porosity and drilling parameters as Q, RPM, WOB, ROP, and T with a correlation coefficient (R) of 0.299, 0.233, 0.151, 0.144, and 0.086, respectively. However, the porosity shows a very weak indirect relationship with SPP by R of −0.003. As represented in Figure 3. However, it worthly mentioned that the relationship between the porosity and drilling parameters might reveal a nonlinear relationship.

2.3. Building and Evaluating the Artificial Intelligent Models

Two techniques were implemented in this study for building rock porosity models which are decision tree and random forest as they have the same tree-based technique. These techniques are well known and applied in many petroleum studies [43].

Decision tree (DT) is considered one of the AI algorithms and it is a simple approach for application [44]. The technique employed straightforward rules for decision-making based on inferred decision instructions [45]. The technique has a hierarchical construction that contains root node, decision nodes, leaf nodes, and branches as shown in Figure 4. Optimizing these parameters will lead to enhancing the model performance for better prediction [46].

Random forest is another supervised machine learning algorithm for classification and regression purposes. RF was introduced in 1995 and had several modifications over time [4750]; it is designed to overcome the overfitting that usually happens in classical discussion trees [51]. Similar to the other machine learning algorithms, several applications of random forest in the oil industry have been reported for classification [52] and regression [5355]. Figure 5 represents the layout for the random forest as it includes a number N of decision trees in the model structure.

The developed models were evaluated by determining four statistical parameters which are correlation coefficient of determination (R2), average absolute percentage error (AAPE), variance account for (VAF), and a20 index. These parameters are calculated as follows:where N is the number of data points in the dataset, is the actual output, is the predicted output, is the mean value, and m20 is the number of data points that have ().

3. Results and Discussion

The cleaned data after the preprocessing phase is plotted as shown in Figure 6 to illustrate the problem complexity for modeling the porosity prediction from the drilling data. The boundaries of the drilling parameters in addition to the rock porosity have a great impact on the models’ performance. The plot represents the boundaries for the training set parameters that are the output (porosity) versus input variables (drilling data).

3.1. DT Model

The data was randomly distributed to training and testing sets by 70 : 30% as 2637 data points for training and 1130 points for the testing set from all the model dataset of 3767 recordings. Max_depth, max_features, min_sample _split, and min_sample_leaf are the main parameters for the DT structure, where max_depth is the most effective parameter as it controls the distance between the root and the leaf node and hence has a great impact on the tree growth. Many sensitivity runs were executed to determine the optimum DT model parameters and the best parameters were recorded and are listed in Table 3.


ParameterOptimum value

Max_depth10
Max_featuresSqrt
Min_samples_split2
Min_samples_leaf1

Figure 7 represents the cross plot for the actual versus the predicted values of the rock porosity for training and testing datasets as AAPE was 6.07% for training and 9% for testing, R2 was 0.94 and 0.87 for training and testing, respectively. VAF was higher than 86.7% and a20 index greater than 0.89 for training and testing phases.

The obtained results showed a high degree of match between the actual and the predicted values for the porosity profile for the drilled section of different lithology formation types as presented in Figure 8.

3.2. RF Model

The same approach was followed to obtain the optimum RF model parameters. The hyperparameters in RF are used either to enhance the predictive power or to make the model faster to run. The n_ estimators parameter is a hyperparameter which is the number of trees that RF builds before computing the average of predictions. A high number of n_estimators enhances the performance of the model and makes the prediction more stable, but it slows down the process of computation. Max_features is another hyperparameter which is the number of features to be considered to split a node in each decision tree. If max_features is “Sqrt”, then the number of features to be considered is the square root of the number of input variables in a dataset. If max_features is “Log2”, then the number of features to be considered is the base-2 logarithm of the number of input variables in a dataset. Max_depth is another important hyperparameter that represents the depth of each tree in a forest. In practice, deep trees can capture more information about the dataset, but also can cause model over-fitting. Therefore, max_depth was tuned from 1 to32 to find the optimum value. Min_samples_split, which represents the minimum number of samples required to split an internal node, and min_samples_leaf, which is the minimum number of samples required to be at a leaf node, are also important hyperparameters for the RF model. Table 4 shows the optimized RF model parameters.


ParameterOptimum value

N_estimators100
Max_depth15
Max_featuresSqrt
Min_samples_split2
Min_samples_leaf1

The training and testing results showed that, for the best RF model parameters, R2 of 0.99 and 0.90 with AAPE of 1.5 and 7% was observed for the training and testing datasets, respectively, as shown in Figure 9. VAF recorded 99.44% and 95.76%, while the a20 index was 1 and 0.93 for training and testing phases, respectively.

Figure 10 represents the high performance of the RF porosity prediction model for the drilled section. Generally, the two AI-developed models for predicting the rock porosity provided a high level of accuracy during training and testing the model.

3.3. Model Validation

A different data-set from the same field that has the same penetrated rocks with complex lithology was utilized for validating the developed DT and RF models. This process enhances the practical application for employing the developed model. A cleaned dataset (1670 data points) was employed for validating the models, and the obtained results showed a strong prediction performance for the porosity log from the surface drilling parameters. The validation results showed R2 of 0.88 and 0.92, and AAPE of 8.58 and 6.5% for DT and RF models, respectively, while VAF and a20 index were higher for RF than DT as shown in Figure 11.

Figure 12 compares the two developed AI models for the validation process. Both DT and RF models are highly performed for predicting the rock porosity for complex lithology formation as shown for the porosity profile of the actual and predicted porosity.

As presented in the results section, the developed RF model showed a great accuracy level in terms of the coefficient of determinations and the average absolute percentage errors between the actual and predicted porosity values. The RF model outperformed the DT with R2 higher than 0.9 and AAPE less than 7% during the model building and validation phases. In addition, the developed model accuracy is considered a high level compared with the developed model in the literature for predicting the rock porosity; however, the current work targets the complex lithology that means breaking one of the limitations for a specific model for each formation type.

4. Conclusions

This study presented a novel approach for predicting the rock porosity from the drilling data during drilling complex lithology formations. The study presented two developed AI models named decision tree (DT) and random forest (RF). The recorded data covered complex lithology of sandstone, shale, and carbonate rock types. Two datasets were employed for building the models and for validating the developed models: a complete dataset of 3763 data points for building the models with a 70 : 30 ratio for training and testing, respectively, while a different dataset (1670 points) for validating the models. The study findings can be summarized as follows:(i)DT model parameters were optimized and the results showed R2 of 0.94 and 0.87 with AAPE of 6.07 and 9% for training and testing, respectively(ii)The optimized RF model achieved R2 of 0.99 and 0.90 with an AAPE of 1.5 and 7% for training and testing, respectively(iii)Validating the two developed models proved the strong prediction performance for the two models with R2 of 0.88 and 0.92 with AAPE of 8.58 and 6.5% for DT and RF, respectively(iv)The RF model outperformed the DT for all the key performance indices while training, testing, and validating the models

The porosity estimation in real time will save cost and time for the porosity determination in reality by employing either the lab measurements or well logging operations. The limitations beyond this research can be concluded in the formations type studied in this work, data range for the parameters, and the wellbore geometry as the models were built for intermediate section for vertical wells. So, the study provides recommendations for future work for predicting the rock porosity for complex lithology for other wellbore profiles and drilling sections, in addition to wide range for the data.

Abbreviations

ANN:Artificial neural network
DT:Decision tree
RF:Random forest
SVM:Support vector machines
FL:Fuzzy logic
FN:Functional networks
R2:Coefficient of determination
VAF:Variance account for
AAPE:Average absolute percentage error
AI:Artificial Intelligence
GR:Gamma-ray
WOB:Weight on bit
RPM:Rotating speed in revolutions per minute
ROP:Rate of penetration
Q:Flow rate
SPP:Standpipe pressure
T:Torque.

Data Availability

Most of the data are included in the manuscript. For the input data for the model, a sample of the data will be provided upon request.

Conflicts of Interest

The authors declare no conflicts of interest.

Authors’ Contributions

The manuscript was written through the contributions of all authors. All authors have approved the final version of the manuscript.

Acknowledgments

The authors acknowledge King Fahd University of Petroleum and Minerals for permitting to publish this work.

References

  1. S. O. Olatunji, A. Selamat, and A. Abdulraheem, “Modeling the permeability of carbonate reservoir using type-2 fuzzy logic systems,” Computers in Industry, vol. 62, no. 2, pp. 147–163, 2011. View at: Publisher Site | Google Scholar
  2. O. F. Adizua and L. Oruade, “Reservoir characterization of an offshore Niger Delta “X” field using well log data,” International Journal of Engineering and Applied Sciences, vol. 2, 2018. View at: Google Scholar
  3. P. Mahzari, A. AlMesmari, and M. Sohrabi, “Co-history matching: a way forward for estimating representative saturation functions,” Transport in Porous Media, vol. 125, no. 3, pp. 483–501, 2018. View at: Publisher Site | Google Scholar
  4. M. A. Andersen, B. Duncan, and R. McLin, “Core truth in formation evaluation,” Oilfield Review, vol. 25, 2013. View at: Google Scholar
  5. J. G. Patchett and E. B. Coalson, “The determination of porosity in sandstone and shaly sandstone part two effects of complex mineralogy and hydrocarbons,” in Proceedings of the Annual Logging Symposium, Corpus Christi, TX, USA, July 1982. View at: Google Scholar
  6. J. R. Hook, “An introduction to porosity,” Petrophysics, vol. 44, no. 3, 2003. View at: Google Scholar
  7. A. Gyllensten, P. Tilke, M. Al-Raisi, and D. Allen, “Porosity heterogeneity analysis using geostatistics,” in Proceedings of the Abu Dhabi International Conference and Exhibition, Abu Dhabi, UAE, October 2004. View at: Google Scholar
  8. R. T. Bonnecaze, M. M. Sharma, J. E. Butler, and G. Arboleda, “High resolution downhole measurements of porosity and fluid saturation while core drilling,” in Proceedings of the SPE Annual Technical Conference and Exhibition, San Antonio, TX, USA, September 2002. View at: Google Scholar
  9. J. A. Kane and J. W. Jennings, “A method to normalize log data by calibration to large-scale data trends,” in Proceedings of the SPE Annual Technical Conference and Exhibition, Dallas, TX, USA, October 2005. View at: Google Scholar
  10. H. Kesserwan, E. Alshanqaiti, G. Jin, and I. Nuic, “Porosity measurements on drill cuttings-comprehensive inputs to formation evaluation near real-time while drilling,” in Proceedings of the Abu Dhabi International Petroleum Exhibition & Conference, Abu Dhabi, UAE, November 2017. View at: Google Scholar
  11. M. R. J. Wyllie, A. R. Gregory, and G. H. F. Gardner, “AN experimental investigation OF factors affecting elastic wave velocities IN porous media,” Geophysics, vol. 23, no. 3, pp. 459–493, 1958. View at: Publisher Site | Google Scholar
  12. R. Gaymard and A. Poupon, “Response of neutron and formation density logs in hydrocarbon bearing formations,” The Log Analyst, vol. 9, no. 5, 1968. View at: Google Scholar
  13. A. Timur, “Producible porosity and permeability of sandstone investigated through nuclear magnetic resonance principles,” The Log Analyst, vol. 10, no. 1, 1969. View at: Google Scholar
  14. S. Mohaghegh, “Virtual-intelligence applications in petroleum engineering: part 1-artificial neural networks,” Journal of Petroleum Technology, vol. 52, no. 9, pp. 64–73, 2000. View at: Publisher Site | Google Scholar
  15. H. Gamal, S. Elkatatny, and A. Abdulraheem, “Rock drillability intelligent prediction for a complex lithology using artificial neural network,” in Proceedings of the Abu Dhabi International Petroleum Exhibition & Conference, Abu Dhabi, UAE, November 2020. View at: Google Scholar
  16. A. Al-Abduljabbar, H. Gamal, and S. Elkatatny, “Application of artificial neural network to predict the rate of penetration for S-shape well profile,” Arabian Journal of Geosciences, vol. 13, no. 16, 2020. View at: Publisher Site | Google Scholar
  17. A. A. Mahmoud, S. Elkatatny, A. Al-AbdulJabbar, T. Moussa, H. Gamal, and D. Al Shehri, “Artificial neural networks model for prediction of the rate of penetration while horizontally drilling carbonate formations,” in Proceeding of the 54th U.S. Rock Mechanics/Geomechanics Symposium, Golden, CO, USA, June 2020. View at: Google Scholar
  18. A. Ahmed, A. Ali, S. Elkatatny, and A. Abdulraheem, “New artificial neural networks model for predicting rate of penetration in deep shale formation,” Sustainability, vol. 11, no. 22, 6527 pages, 2019. View at: Publisher Site | Google Scholar
  19. A. Hassan, A. Al-Majed, M. Mahmoud, S. Elkatatny, and A. Abdulraheem, “Improved predictions in oil operations using artificial intelligent techniques,” in Proceeding of the SPE Middle East Oil and Gas Show and Conference, Manama, Bahrain, March 2019. View at: Google Scholar
  20. M. Elzenary, S. Elkatatny, K. Z. Abdelgawad, A. Abdulraheem, M. Mahmoud, and D. Al-Shehri, “New technology to evaluate equivalent circulating density while drilling using artificial intelligence,” in Proceeding of the SPE Kingdom of Saudi Arabia Annual Technical Symposium and Exhibition, Dammam, Saudi Arabia, April 2018. View at: Google Scholar
  21. K. Al-azani, S. Elkatatny, A. Abdulraheem, M. Mahmoud, and D. Al-Shehri, “Real time prediction of the rheological properties of oil-based drilling fluids using artificial neural networks,” in Proceeding of the SPE Kingdom of Saudi Arabia Annual Technical Symposium and Exhibition, Dammam, Saudi Arabia, April 2018. View at: Google Scholar
  22. S. Elkatatny, “Real-time prediction of the rheological properties of water-based drill-in fluid using artificial neural networks,” Sustainability, vol. 11, no. 18, 5008 pages, 2019. View at: Publisher Site | Google Scholar
  23. A. Alsabaa, H. A. Gamal, S. M. Elkatatny, and A. Abdulraheem, “Real-time prediction of rheological properties of all-oil mud using artificial intelligence,” in Proceeding of the 54th U.S. Rock Mechanics/Geomechanics Symposium, Golden, CO, USA, June 2020. View at: Google Scholar
  24. A. Alsabaa, H. Gamal, S. Elkatatny, and A. Abdulraheem, “Real-time prediction of rheological properties of invert emulsion mud using adaptive neuro-fuzzy inference system,” Sensors, vol. 20, no. 6, 1669 pages, 2020. View at: Google Scholar
  25. M. A. Ahmadi, M. Pournik, S. R. Shadizadeh, M. Ali, M. Pournik, and S. Reza, “Toward connectionist model for predicting bubble point pressure of crude oils: application of artificial intelligence,” Petroleum, vol. 1, no. 4, pp. 307–317, 2015. View at: Publisher Site | Google Scholar
  26. F. S. Alakbari, S. Elkatatny, and S. O. Baarimah, “Prediction of bubble point pressure using artificial intelligence AI techniques,” in Proceeding of the SPE Middle East Artificial Lift Conference and Exhibition, Manama, Bahrain, November 2016. View at: Google Scholar
  27. D. A. Wood and A. Choubineh, “Transparent open-box learning network and artificial neural network predictions of bubble-point pressure compared,” Petroleum, vol. 6, 2020. View at: Publisher Site | Google Scholar
  28. M. R. Mahdiani and M. Norouzi, “A new heuristic model for estimating the oil formation volume factor,” Petroleum, vol. 4, no. 3, pp. 300–308, 2018. View at: Publisher Site | Google Scholar
  29. M. A. Oloso, M. G. Hassan, M. B. Bader-El-Den, and J. M. Buick, “Hybrid functional networks for oil reservoir PVT characterisation,” Expert Systems with Applications, vol. 87, pp. 363–369, 2017. View at: Publisher Site | Google Scholar
  30. S. Elkatatny, T. Moussa, A. Abdulraheem, and M. Mahmoud, “A self-adaptive artificial intelligence technique to predict oil pressure volume temperature properties,” Energies, vol. 11, no. 12, 3490 pages, 2018. View at: Publisher Site | Google Scholar
  31. H. Al Khalifah, P. W. J. Glover, and P. Lorinczi, “Permeability prediction and diagenesis in tight carbonates using machine learning techniques,” Marine and Petroleum Geology, vol. 112, 104096 pages, 2020. View at: Publisher Site | Google Scholar
  32. B. Shokooh Saljooghi and A. Hezarkhani, “A new approach to improve permeability prediction of petroleum reservoirs using neural network adaptive wavelet (wavenet),” Journal of Petroleum Science and Engineering, vol. 133, pp. 851–861, 2015. View at: Publisher Site | Google Scholar
  33. R. M. Alloush, S. Elkatatny, M. A. Mahmoud, T. M. Moussa, A. Z. Ali, and A. Abdulraheem, “Estimation of geomechanical failure parameters from well logs using artificial intelligence techniques,” in Proceedings of the SPE Kuwait Oil & Gas Show and Conference, Kuwait City, Kuwait, October 2017. View at: Google Scholar
  34. Z. Tariq, S. Elkatatny, M. Mahmoud, A. Z. Ali, and A. Abdulraheem, “A new technique to develop rock strength correlation using artificial intelligence tools,” in Proceedings of the SPE Reservoir Characterisation and Simulation Conference and Exhibition, Abu Dhabi, UAE, May 2017. View at: Google Scholar
  35. S. Elkatatny, Z. Tariq, M. Mahmoud, A. Abdulazeez, and I. M. Mohamed, “Application of artificial intelligent techniques to determine sonic time from well logs,” in Proceedings of the 50th U.S. Rock Mechanics/Geomechanics Symposium, Houston, TX, USA, June 2016. View at: Google Scholar
  36. Z. Tariq, S. Elkatatny, M. Mahmoud, A. Abdulraheem, and K. Fahd, “A new artificial intelligence based empirical correlation to predict sonic travel time,” in Proceedings of the International Petroleum Technology Conference, Bangkok, Thailand, November 2016. View at: Google Scholar
  37. A. Gowida, S. Elkatatny, and H. Gamal, “Unconfined compressive strength (UCS) prediction in real-time while drilling using artificial intelligence tools,” Neural Computing and Applications, vol. 7, 2021. View at: Google Scholar
  38. S. Elkatatny, Z. Tariq, M. Mahmoud, and A. Abdulraheem, “New insights into porosity determination using artificial intelligence techniques for carbonate reservoirs,” Petroleum, vol. 4, no. 4, pp. 408–418, 2018. View at: Publisher Site | Google Scholar
  39. A. Ali, T. Aïfa, and K. Baddari, “Prediction of natural fracture porosity from well log data by means of fuzzy ranking and an artificial neural network in Hassi Messaoud oil field, Algeria,” Journal of Petroleum Science and Engineering, vol. 115, pp. 78–89, 2014. View at: Publisher Site | Google Scholar
  40. D. A. Wood, “Predicting porosity, permeability and water saturation applying an optimized nearest-neighbour, machine-learning and data-mining network of well-log data,” Journal of Petroleum Science and Engineering, vol. 184, 106587 pages, 2019. View at: Google Scholar
  41. A. Al-Abduijabbar, K. Al-Azani, and S. Elkatatny, “Estimation of reservoir porosity from drilling parameters using artificial neural networks,” Petrophysics, vol. 61, no. 3, pp. 318–329, 2020. View at: Google Scholar
  42. R. Kanfar, O. Shaikh, M. Yousefzadeh, and T. Mukerji, “Real-time well log prediction from drilling data using deep learning,” Petroleum Engineering, 2020. View at: Google Scholar
  43. H. H. Alkinani, A. T. T. Al-Hameedi, S. Dunn-Norman, M. T. Alsaba, and A. S. Amer, “Review of the applications of decision tree analysis in petroleum engineering with a rigorous analysis,” in Proceedings of the SPE Middle East Oil and Gas Show and Conference, Manama, Bahrain, March 2019. View at: Google Scholar
  44. X. Li and C. W. Chan, “Application of an enhanced decision tree learning approach for prediction of petroleum production,” Engineering Applications of Artificial Intelligence, vol. 23, no. 1, pp. 102–109, 2010. View at: Publisher Site | Google Scholar
  45. G. T. Reddy, M. P. K. Reddy, K. Lakshmanna et al., “Analysis of dimensionality reduction techniques on big data,” IEEE Access, vol. 8, pp. 54776–54788, 2020. View at: Publisher Site | Google Scholar
  46. D. Sun, “Investigating the applications of machine learning techniques to predict the rock brittleness index,” Applied Science, vol. 10, no. 5, 1691 pages, 2020. View at: Google Scholar
  47. L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001. View at: Publisher Site | Google Scholar
  48. T. K. Ho, “Random decision forests,” in Proceedings of the International Conference on Document Analysis and Recognition, ICDAR, vol. 1, pp. 278–282, Montreal, Canada, August 1995. View at: Google Scholar
  49. E. M. Kleinberg, “An overtraining-resistant stochastic modeling method for pattern recognition,” Annals of Statistics, vol. 24, no. 6, pp. 2319–2349, 1996. View at: Publisher Site | Google Scholar
  50. E. M. Kleinberg, “On the algorithmic implementation of stochastic discrimination,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 5, pp. 473–490, 2000. View at: Publisher Site | Google Scholar
  51. T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, Springer, Berlin, Germany, 2009.
  52. Y. Kim, R. Hardisty, E. Torres, and K. J. Marfurt, “Seismic facies classification using random forest algorithm,” in Proceedigs of the 2018 SEG International Exposition and Annual Meeting, Anaheim, CA, USA, August 2018. View at: Google Scholar
  53. C. Hegde, S. Wallace, and K. Gray, “Using trees, bagging, and random forests to predict rate of penetration during drilling,” in Proceedigs of the Society Of Petroleum Engineers-SPE Middle East Intelligent Oil And Gas Conference And Exhibition, Abu Dhabi, UAE, September 2015. View at: Google Scholar
  54. E. Nasir and C. Rickabaugh, “Optimizing drilling parameters using a random forests ROP model in the permian basin,” in Proceedigs of the SPE Liquids-Rich Basins Conference-North America, Midland, TX, USA, September 2018. View at: Google Scholar
  55. Y. Sun, G. Ma, L. Wang, F. Zhang, X. Li, and L. Mei, “Organic-matter content prediction based on the random forest algorithm: application to a Lower Silurian shale-gas reservoir,” in Proceedigs of the SEG International Exposition and Annual Meeting, San Antonio, TX, USA, September 2019. View at: Google Scholar

Copyright © 2021 Hany Gamal et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Related articles

No related content is available yet for this article.
 PDF Download Citation Citation
 Download other formatsMore
 Order printed copiesOrder
Views690
Downloads753
Citations

Related articles

No related content is available yet for this article.

Article of the Year Award: Outstanding research contributions of 2021, as selected by our Chief Editors. Read the winning articles.