Artificial neural network model for reservoir petrophysical properties: porosity, permeability and water saturation prediction

Okon, Anietie N.; Adewole, Steve; Uguma, Emmanuel M.

doi:10.1007/s40808-020-01012-4

Cited by 69 publications

(31 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence, the results proved that Bayesian Regularization gave better results as compared with the other two approaches. is finding is similar to [28,50] but different from [33]. GSD faces heterogeneous work environments that results in multiple challenges which have been highlighted in this research study.…”

Section: Discussionmentioning

confidence: 56%

“…In [ 49 ], authors applied these three ANN techniques for the prediction of Flash Floods and found Bayesian Regularization an efficient technique as compared with the other two techniques. Authors in [ 50 ] used ANN for reservoir petro-physical properties such as porosity, permeability, and water saturation. In [ 51 ], authors compared relative predictive abilities of Levenberg–Marquardt and Bayesian Regularization methods for data on prices of four cryptocurrencies, namely, Bitcoin, Bitcoin Cash, Litecoin, and Ripple and found that the Bayesian Regularization method gives less error in prediction as compared with Levenberg–Marquardt in case of large data but both the methods are found nearly equally efficient for small data.…”

Section: Artificial Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Risk Prediction by Using Artificial Neural Network in Global Software Development

Iftikhar

Alam

Ahmed

et al. 2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

The demand for global software development is growing. The nonavailability of software experts at one place or a country is the reason for the increase in the scope of global software development. Software developers who are located in different parts of the world with diversified skills necessary for a successful completion of a project play a critical role in the field of software development. Using the skills and expertise of software developers around the world, one could get any component developed or any IT-related issue resolved. The best software skills and tools are dispersed across the globe, but to integrate these skills and tools together and make them work for solving real world problems is a challenging task. The discipline of risk management gives the alternative strategies to manage risks that the software experts are facing in today’s world of competitiveness. This research is an effort to predict risks related to time, cost, and resources those are faced by distributed teams in global software development environment. To examine the relative effect of these factors, in this research, neural network approaches like Levenberg–Marquardt, Bayesian Regularization, and Scaled Conjugate Gradient have been implemented to predict the responses of risks related to project time, cost, and resources involved in global software development. Comparative analysis of these three algorithms is also performed to determine the highest accuracy algorithms. The findings of this study proved that Bayesian Regularization performed very well in terms of the MSE (validation) criterion as compared with the Levenberg–Marquardt and Scaled Conjugate Gradient approaches.

show abstract

Section: Discussionmentioning

confidence: 56%

Section: Artificial Neural Networkmentioning

confidence: 99%

Risk Prediction by Using Artificial Neural Network in Global Software Development

Iftikhar

Alam

Ahmed

et al. 2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…To decide on the final model, the RMSE, ADD, and R 2 values were comparatively analyzed for the topologies of QP-4-4-1, IBP-4-7-1, BBP-4-7-1, GA-6-14-1, and LM-4-5-1. Furthermore, to determine the R 2 values, the topologies projection and actual values of the D p were plotted for the testing data group [56] (Fig. 2).…”

Section: Model Selectionmentioning

confidence: 99%

Porosity Estimation of Mesoporous TiO₂‐ZnO Nanocrystalline by Artificial Neural Network Modeling

et al. 2021

View full text Add to dashboard Cite

Artificial neural network was introduced for modeling and verifying the relative significance of operational variables to attain the best possible average pore diameter (Dp) of the mesoporous TiO2‐ZnO nanocrystalline synthesized by microwave‐assisted hydrothermal method. To optimize the system, the autonomous variables calcination temperature and concentrations of D‐glucose, polyethylene glycol, and Zn‐Ti were employed as input factors, Dp was designated as output. Dp was achieved from the implementation of the experimental model of the variables by the Barrett‐Joyner‐Halenda method. The network was trained by IBP, BBP, LM, QP, and GA algorithms as a model. The BBP‐4‐7‐1 model provided the optimum numerical properties among four other algorithms. Furthermore, the calcination temperature acted as the most efficient variable whilst D‐glucose concentration showed the least effect. By validating the microwave‐assisted hydrothermal conditions, in situ polymerization of poly(sodium 4‐styrene sulfonate) was introduced to activate mesopore walls by –SO3H functional groups used for ester production by transesterification of used cooking oil, resulting in a methyl ester content of 97.4 % and excellent reusability for ten sequential reactions without any further regeneration.

show abstract

“…By estimating the fluid types and their volume as well as the rock and mineral types, petrophysical studies will help to provide comprehensive information about the reservoir. Therefore, in various types of research studies, including reservoir characterization [5][6][7][8][9][10][11][12][13], hydraulic fracturing simulation/operation [14][15][16][17][18][19][20][21][22][23], and building, evaluating, and controlling numerical models [24,25], petrophysical parameters play a vital role. Porosity, permeability, and water saturation, which are considered as three main reservoir petrophysical properties, are provided by core or well-logging data.…”

Section: Introductionmentioning

confidence: 99%

Joint Application of Diagenetic, Petrophysical and Geomechanical Data for Selecting Hydraulic Fracturing Candidate Zone

Bakhshi¹,

Shahrabadi²,

Golsanami

et al. 2021

Int. J. Petrol. Technol.

View full text Add to dashboard Cite

The more comprehensive information on the reservoir properties will help to better plan drilling and design production. Herein, diagenetic processes and geomechanical properties are notable parameters that determine reservoir quality. Recognizing the geomechanical properties of the reservoir as well as building a mechanical earth model play a strong role in the hydrocarbon reservoir life cycle and are key factors in analyzing wellbore instability, drilling operation optimization, and hydraulic fracturing designing operation. Therefore, the present study focuses on selecting the candidate zone for hydraulic fracturing through a novel approach that simultaneously considers the diagenetic, petrophysical, and geomechanical properties. The diagenetic processes were analyzed to determine the porosity types in the reservoir. After that, based on the laboratory test results for estimating reservoir petrophysical parameters, the zones with suitable reservoir properties were selected. Moreover, based on the reservoir geomechanical parameters and the constructed mechanical earth model, the best zones were selected for hydraulic fracturing operation in one of the Iranian fractured carbonate reservoirs. Finally, a new empirical equation for estimating pore pressure in nine zones of the studied well was developed. This equation provides a more precise estimation of stress profiles and thus leads to more accurate decision-making for candidate zone selection. Based on the results, vuggy porosity was the best porosity type, and zones C2, E2 and G2, having suitable values of porosity, permeability, and water saturation, showed good reservoir properties. Therefore, zone E2 and G2 were chosen as the candidate for hydraulic fracturing simulation based on their E (Young’s modulus) and ν (Poisson’s ratio) values. Based on the mechanical earth model and changes in the acoustic data versus depth, a new equation is introduced for calculating the pore pressure in the studied reservoir. According to the new equation, the dominant stress regime in the whole well, especially in the candidate zones, is SigHmax>SigV>Sighmin, while according to the pore pressure equation presented in the literature, the dominant stress regime in the studied well turns out to be SigHmax>Sighmin>SigV.

show abstract

Artificial neural network model for reservoir petrophysical properties: porosity, permeability and water saturation prediction

Cited by 69 publications

References 48 publications

Risk Prediction by Using Artificial Neural Network in Global Software Development

Risk Prediction by Using Artificial Neural Network in Global Software Development

Porosity Estimation of Mesoporous TiO₂‐ZnO Nanocrystalline by Artificial Neural Network Modeling

Joint Application of Diagenetic, Petrophysical and Geomechanical Data for Selecting Hydraulic Fracturing Candidate Zone

Contact Info

Product

Resources

About

Artificial neural network model for reservoir petrophysical properties: porosity, permeability and water saturation prediction

Cited by 69 publications

References 48 publications

Risk Prediction by Using Artificial Neural Network in Global Software Development

Risk Prediction by Using Artificial Neural Network in Global Software Development

Porosity Estimation of Mesoporous TiO2‐ZnO Nanocrystalline by Artificial Neural Network Modeling

Joint Application of Diagenetic, Petrophysical and Geomechanical Data for Selecting Hydraulic Fracturing Candidate Zone

Contact Info

Product

Resources

About

Porosity Estimation of Mesoporous TiO₂‐ZnO Nanocrystalline by Artificial Neural Network Modeling