List of Articles Neural Network

• Open Access Article
  • Abstract Page
  • Full-Text
  
  1 - Sonic wave velocity estimation using intelligent system and multi resolution graph base clustering: A case study from one of Iranian south field
  مرتضی نوری مینا کریمی خالدی
  20.1001.1.22518738.1392.3.6.3.5
  Abstract Compressional and shear velocity are two fundamental parameters, which have many applications in petrophysical, geophysical, and geomechanical operations. These two parameters can be obtained using Dipole Sonic Imaging tool (DSI), but unfortunately this tool More

  Abstract Compressional and shear velocity are two fundamental parameters, which have many applications in petrophysical, geophysical, and geomechanical operations. These two parameters can be obtained using Dipole Sonic Imaging tool (DSI), but unfortunately this tool is run just in few wells of a field. Therefore it is important to predict compressional and shear velocity indirectly from the other conventional well logs that have good correlation with these parameters in wells without these logs. Classical methods to predict the mentioned parameters are utilizing correlations and regression analysis. However, the best tool is intelligent systems including Artificial Neural Network, Fuzzy Logic, Adaptive Neuro Fuzzy Inference System, and Multi resolution graph base clustering for performing such tasks. In this paper 1321 data points from Kangan and Dalan formations which have compressional and shear velocity are used. These data are divided into two groups: 995 and 326 data points were used for construction of intelligent systems and model testing, respectively. The results showed that despite differences in concept, all of the intelligent techniques were successful for estimation of compressional and shear velocities. The Multi resolution graph base clustering. The method had the best performance among the others due to precise clustering the data points. Using this method, the compressional and shear velocity were correlated with correlation factor of 0.9505 and 0.9407, respectively. The developed model does not incorporate depth or lithological data as a part of the inputs to the network. This means that utilized methodology is applicable to any field. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  2 - Porosity modeling in Azadegan oil field: a comparative study of Bayesian theory of data fusion, multi layer neural network, and multiple linear regression techniques
  عطیه  مظاهری طرئی حسین معماریان بهزاد تخم چی بهزاد مشیری
  20.1001.1.22518738.1391.2.3.4.3
  Porosity parameter is an important reservoir property that can be obtained by studying the well core. However, all wells in a field do not have a core. Additionally, in some wells such as horizontal wells, measuring the well core is practically impossible. However, for More

  Porosity parameter is an important reservoir property that can be obtained by studying the well core. However, all wells in a field do not have a core. Additionally, in some wells such as horizontal wells, measuring the well core is practically impossible. However, for almost all wells, log data is available. Usually these logs are used to estimate porosity. The porosity value obtained from this method is influenced by factors such as temperature, pressure, fluid type, and amount of hydrocarbons in shale formations. Thus it is slightly different from the exact value of porosity. Thus, estimates are prone to error and uncertainty. One of the best and yet most practical ways to reduce the amount of uncertainty in measurement is using various sources and data fusion techniques. The main benefit of these techniques is that they increase confidence and reduce risk and error in decision making. In this paper, in order to determine porosity values, data from four wells located in Azadegan oil field are used. First, multilayer neural network and multiple linear regressions are used to estimate the values and then the results of these techniques are compared with a data fusion method (Bayesian theory). To check if it would be possible to generalize these three methods on other data, the porosity parameter of another independent well in this field is also estimated by using these techniques. Number of input variables to estimate porosity in both the neural network and the multiple linear regressions methods is 7, and in the data fusion technique, a maximum of 7 input variables is used. Finally, by comparing the results of the three methods, it is concluded that the data fusion technique (Bayesian theory) is a considerably more accurate technique than multilayer neural network, and multiple linear regression, when it comes to porosity value estimation; Such that the results are correlated with the ground truth greater than 90%. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  3 - Application of Artificial Intelligence during History matching in One of fractured oil Reservoirs
  ناصر اخلاقی ریاض خراط صدیقه مهدوی
  Nowadays different methods of soft computing to reduce time and calculation content are widely used in oil and gas industry. One of the main applications of these methods is prediction of the results of different processes in oil industry which their estimation with More

  Nowadays different methods of soft computing to reduce time and calculation content are widely used in oil and gas industry. One of the main applications of these methods is prediction of the results of different processes in oil industry which their estimation with usual methods is too difficult and has either no single response or their response finding need more time and cost due to their nonlinear of the related problems. Because of much uncertainty on information which used in simulators, the results of these simulation models may have lot errors so production data (Pressure, Production Rate, Water Oil Ratio (WOR), Gas Oil Ratio (GOR) and etc.) during reservoir life is used to historical accommodation between simulator results and actual data. The main purpose of this study is investigation and feasibility study of a usual method of artificial intelligence in oil industry, which is based on the soft computing. In this study, Artificial Neural Network (ANN) is used to make a predicting model for bottom hole pressure and for one of the fractured oil reservoirs with the seven years history of production. Some unconditional parameters such as fracture porosity, horizontal and vertical fracture permeability, height of matrix and matrix-fracture dual porosity were applied as input data of the networks, and pressure was applied as an output in network making. Applied data in network making is achieved from the 50 runs with simulator. The conclusion of this study showed that predicting model of ANN with error less than 4% and reduces the time of process, has a good ability to history matching. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  4 - Application of Artificial Intelligence during History matching in One of fractured oil Reservoirs
  ناصر اخلاقی Reyaz kharata Sedigheh Mahdavi
  20.1001.1.22518738.1390.1.1.5.3
  Nowadays different methods of soft computing to reduce time and calculation content are widely used in oil and gas industry. One of the main applications of these methods is prediction of the results of different processes in oil industry which their estimation with u More

  Nowadays different methods of soft computing to reduce time and calculation content are widely used in oil and gas industry. One of the main applications of these methods is prediction of the results of different processes in oil industry which their estimation with usual methods is too difficult and has either no single response or their response finding need more time and cost due to their nonlinear of the related problems. Because of much uncertainty on information which used in simulators, the results of these simulation models may have lot errors so production data (Pressure, Production Rate, Water Oil Ratio (WOR), Gas Oil Ratio (GOR) and etc.) during reservoir life is used to historical accommodation between simulator results and actual data. The main purpose of this study is investigation and feasibility study of a usual method of artificial intelligence in oil industry, which is based on the soft computing. In this study, Artificial Neural Network (ANN) is used to make a predicting model for bottom hole pressure and for one of the fractured oil reservoirs with the seven years history of production. Some unconditional parameters such as fracture porosity, horizontal and vertical fracture permeability, height of matrix and matrix-fracture dual porosity were applied as input data of the networks, and pressure was applied as an output in network making. Applied data in network making is achieved from the 50 runs with simulator. The conclusion of this study showed that predicting model of ANN with error less than 4% and reduces the time of process, has a good ability to history matching. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  5 - Improve the detection of buried channel, using Artificial Neural Networks and seismic attributes
  Alireza Ghazanfari Abdolrahim Javaherian Mojtaba Seddigh Arabani
  20.1001.1.22518738.1396.7.14.2.8
  Channels are one of the most important stratigraphic and morphological events. If channels place in a suitable position such as enclosed in impermeable place can make suitable oil and gas reservoir; So identifying channels are crucial. Different tools such as filters, s More

  Channels are one of the most important stratigraphic and morphological events. If channels place in a suitable position such as enclosed in impermeable place can make suitable oil and gas reservoir; So identifying channels are crucial. Different tools such as filters, seismic attributes, artificial neural networks, and meta-attributes have played an important role in this regard. In this paper dip-steering cube, dip-steer median filter, dip-steer diffusion filter, and fault enhancement filter, have been used. Then, various seismic attributes such as similarity, texture, spectral decomposition, energy and polar dip have been defined and studied. Therefore, work on F3 real seismic data of Dutch part of the North sea for detecting channels has been started by detecting suitable attributes. For identifying the channel in data, it has been used from compilation and combination of seismic attributes using supervised ANN (multi-layer perceptron), and development of mata-attributes, then recombine meta-attributes created along the channel, and using different interpretation point, for eliminating the impact of facies and lithology changes along the channel. Among the advantages and the reasons for using this kind of neural network (supervised), which increases the effect of the neural network and improves the result, is the ability to train the network by specifying the channel and non-channel points used in this paper. Finally, using the above methods, the identification of the channel examined in the above seismic data has been improved, and the channel has been properly detected and extracted throughout its entire length. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  6 - Compilation of artificial neural networks and the thinned Fault likelihood auto-tracking algorithm, for identification, interpretation and extraction of faults
  Alireza Ghazanfari Hoseyn Mohammadrezaei Hamidreza Ansari
  20.1001.1.22518738.1397.8.15.5.0
  Fault identification and investigating their evolution is of special importance in the exploration and development of hydrocarbon resources. Success in exploration and development of hydrocarbon fields, need to recognition of petroleum systems and in this regard one of More

  Fault identification and investigating their evolution is of special importance in the exploration and development of hydrocarbon resources. Success in exploration and development of hydrocarbon fields, need to recognition of petroleum systems and in this regard one of the most important topics is identifying faults and their extension condition as a main fluid migration path, specially in deeper zones. Faults and fractures have crucial role in making high permeable and porous segments and cut reservoir and cap rock in the fluid migration path. In addition, for maximizing the production of hydrocarbon from reservoirs and also for reducing the risk of drilling, it is necessary to gain information about geometry and nature of faults of reservoirs. In this paper, the purpose is investigating the performance of combination of neural networks and Fault Likelihood auto-tracking algorithm for identification and interpretation of faults in seismic data. At first using the Dip-steering feature of software, the early filter for accurate identification of dip of structures in the data, have been designed and applied. Then with designing and applying the appropriate filters, the seismic data have been improved. After that proper seismic attributes for fault identification have been calculated from seismic data. With picking fault and non-fault points from data, a supervised neural network using the selected attributes was formed and after training the network, the appropriate output achieved. Then the output of neural network has been used as a input for Thinned Fault Likelihood auto-tracking algorithm. The output of this part contains a volume of tracked faults. Finally using sub-tools of TFL and optimal setting of parameters, 3D fault planes has been interpreted and extracted. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  7 - Permeability estimation using petrophysical logs and artificial intelligence methods: A case study in the Asmari reservoir of Ahvaz oil field
  Abouzar Mohsenipour Bahman Soleimani iman Zahmatkesh Iman  Veisi
  20.1001.1.22518738.1399.10.20.4.3
  Permeability is one of the most important petrophysical parameters that play a key role in the discussion of production and development of hydrocarbon fields. In this study, first, the magnetic resonance log in Asmari reservoir was evaluated and permeability was calcula More

  Permeability is one of the most important petrophysical parameters that play a key role in the discussion of production and development of hydrocarbon fields. In this study, first, the magnetic resonance log in Asmari reservoir was evaluated and permeability was calculated using two conventional methods, free fluid model (Coates) and Schlumberger model or mean T2 (SDR). Then, by constructing a simple model of artificial neural network and also combining it with Imperialist competition optimization (ANN-ICA) and particle swarm (ANN-PSO) algorithms, the permeability was estimated. Finally, the results were compared by comparing the estimated COATES permeability and SDR permeability with the actual value, and the estimation accuracy was compared in terms of total squared error and correlation coefficient. The results of this study showed an increase in the accuracy of permeability estimation using a combination of optimization algorithms with artificial neural network. The results of this method can be used as a powerful method to obtain other petrophysical parameters. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  8 - Petrophysical Modeling of Lower Zone of Ratawi Formation, using Neural Network Method in Assimilating Seismic and Geological Well Log Data
  Javid Hanachi Alireza Bashari
  20.1001.1.22518738.1401.12.23.4.3
  Esfandiar field is located at the northern part of the Persian . This field is a single large anticline with Lulu field of Saudi Arabia, with , 20 KM length and 7 KM width. The field was discovered in 1966 by drilling of well E1, on the northern culmination of t More

  Esfandiar field is located at the northern part of the Persian . This field is a single large anticline with Lulu field of Saudi Arabia, with , 20 KM length and 7 KM width. The field was discovered in 1966 by drilling of well E1, on the northern culmination of the field. wells E3 and E2 were drilled at the top of structure in the southern part of the field. DSTs tests results of E1 proved that the top of Lower Ratawi formation contain 15 m oil column. E3 well test result regards as a dry hole DSTs test results of E2 were not conclusive due to inadequate testing plans . E4 Appraisal well contained, 14 m oil column at the Lower Ratawi. Log interpretations results indicated, E2 and E3 wells contains oil in Yamama formation in the southern part of the field which has not been tested properly. Lower Ratawi (Top oil-bearing zone ), Zone 'B' of Lower Ratawi (Oil bearing zone at bottom), Yamama were constructed based on the existing data. Petrophysical and geophysical data has been used for the Lower Ratawi reservoir, as a result the geological models (structural and porosity models), with applying, related software’s and neural network geophysical method are generated . At the conclusion, the recommended plan consists of horizontal drilling wells for oil production in Lower Ratawi in the north of the field has been proposed. Manuscript profile