Using Artificial Neural Networks to Predict the Presence of Overpressured Zones in the Anadarko Basin,Oklahoma

Many sedimentary basins throughout the world exhibit areas with abnormal pore-fluid pressures (higher or lower than normal or hydrostatic pressure). Predicting pore pressure and other parameters (depth, extension, magnitude, etc.) in such areas are challenging tasks. The compressional acoustic (sonic) log (DT) is often used as a predictor because it responds to changes in porosity or compaction produced by abnormal pore-fluid pressures. Unfortunately, the sonic log is not commonly recorded in most oil and/or gas wells. We propose using an artificial neural network to synthesize sonic logs by identifying the mathematical dependency between DT and the commonly available logs, such as normalized gamma ray (GR) and deep resistivity logs (REID). The artificial neural network process can be divided into three steps: (1) Supervised training of the neural network; (2) confirmation and validation of the model by blind-testing the results in wells that contain both the predictor (GR, REID) and the target values (DT) used in the supervised training; and 3) applying the predictive model to all wells containing the required predictor data and verifying the accuracy of the synthetic DT data by comparing the back-predicted synthetic predictor curves (GRNN, REIDNN) to the recorded predictor curves used in training (GR, REID). Artificial neural networks offer significant advantages over traditional deterministic methods. They do not require a precise mathematical model equation that describes the dependency between the predictor values and the target values and, unlike linear regression techniques, neural network methods do not overpredict mean values and thereby preserve original data variability. One of their most important advantages is that their predictions can be validated and confirmed through back-prediction of the input data. This procedure was applied to predict the presence of overpressured zones in the Anadarko Basin, Oklahoma. The results are promising and encouraging.     

Artificial neural network forward modelling and inversion of electrokinetic logging data

An artificial neural network method is proposed as a computationally economic alternative to numerical simulation by the Biot theory for predicting borehole seismoelectric measurements given a set of formation properties. Borehole seismoelectric measurements are simulated using a finite element forward model, which solves the Biot equations together with an equation for the streaming potential. The results show that the neural network method successfully predicts the streaming potentials at each detector, even when the input pressures are contaminated with 10% Gaussian noise. A fast inversion methodology is subsequently developed in order to predict subsurface material properties such as porosity and permeability from streaming potential measurements. The predicted permeability and porosity results indicate that the method predictions are more accurate for the permeability predictions, with the inverted permeabilities being in excellent agreement with the actual permeabilities. This approach was finally verified by using data from a field experiment. The predicted permeability results seem to predict the basic trends in permeabilities from a packer test. As expected from synthetic results, the predicted porosity is less accurate. Investigations are also carried out to predict the zeta potential. The predicted zeta potentials are in agreement with values obtained through experimental self potential measurements.     

Load or concentration,logged or unlogged? Addressing ten years of uncertainty in neural network suspended sediment prediction

This paper addresses current inconsistencies in methodological approaches for neural network modelling of suspended sediment. An expansion in the number of case studies being published over the last decade has yet to result in agreed guidelines on whether suspended sediment load or concentration should be modelled, and whether log‐transformation of data is either necessary or potentially beneficial. This contrasts with the well‐recognized guidelines that direct traditional sediment rating curve studies. The paper reports a comprehensive set of single‐input single‐output neural network suspended sediment modelling experiments performed on two catchments in Puerto Rico. It examines the impact of internal complexity, input variable choice and data transformation on the form, consistency and physical rationality of model outputs, the existence of localized overfitting and the usefulness of global performance metrics. Sound guidance on whether to model sediment load or concentration, and whether to model log‐transformed data is provided. Copyright © 2011 John Wiley & Sons, Ltd.     

Calibration of high-resolution geophysical data with tracer test measurements to improve hydrological predictions

Relationships between porosity and hydraulic conductivity tend to be strongly scale- and site-dependent and are thus very difficult to establish. As a result, hydraulic conductivity distributions inferred from geophysically derived porosity models must be calibrated using some measurement of aquifer response. This type of calibration is potentially very valuable as it may allow for transport predictions within the considered hydrological unit at locations where only geophysical measurements are available, thus reducing the number of well tests required and thereby the costs of management and remediation. Here, we explore this concept through a series of numerical experiments. Considering the case of porosity characterization in saturated heterogeneous aquifers using crosshole ground-penetrating radar and borehole porosity log data, we use tracer test measurements to calibrate a relationship between porosity and hydraulic conductivity that allows the best prediction of the observed hydrological behavior. To examine the validity and effectiveness of the obtained relationship, we examine its performance at alternate locations not used in the calibration procedure. Our results indicate that this methodology allows us to obtain remarkably reliable hydrological predictions throughout the considered hydrological unit based on the geophysical data only. This was also found to be the case when significant uncertainty was considered in the underlying relationship between porosity and hydraulic conductivity.     

Fracture-frequency prediction from borehole wireline logs using artificial neural networks

Borehole-wall imaging is currently the most reliable means of mapping discontinuities within boreholes. As these imaging techniques are expensive and thus not always included in a logging run, a method of predicting fracture frequency directly from traditional logging tool responses would be very useful and cost effective. Artificial neural networks (ANNs) show great potential in this area. ANNs are computational systems that attempt to mimic natural biological neural networks. They have the ability to recognize patterns and develop their own generalizations about a given data set. Neural networks are trained on data sets for which the solution is known and tested on data not previously seen in order to validate the network result. We show that artificial neural networks, due to their pattern recognition capabilities, are able to assess the signal strength of fracture-related heterogeneity in a borehole log and thus fracture frequency within a borehole. A combination of wireline logs (neutron porosity, bulk density, P-sonic, S-sonic, deep resistivity and shallow resistivity) were used as input parameters to the ANN. Fracture frequency calculated from borehole televiewer data was used as the single output parameter. The ANN was trained using a back-propagation algorithm with a momentum learning function. In addition to fracture frequency within a single borehole, an ANN trained on a subset of boreholes in an area could be used for prediction over the entire set of boreholes, thus allowing the lateral correlation of fracture zones.     

Inversion of nuclear well-logging data using neural networks

This work looks at the application of neural networks in geophysical well‐logging problems and specifically their utilization for inversion of nuclear downhole data. Simulated neutron and γ‐ray fluxes at a given detector location within a neutron logging tool were inverted to obtain formation properties such as porosity, salinity and oil/water saturation. To achieve this, the forward particle‐radiation transport problem was first solved for different energy groups (47 neutron groups and 20 γ‐ray groups) using the multigroup code EVENT. A neural network for each of the neutron and γ‐ray energy groups was trained to re‐produce the detector fluxes using the forward modelling results from 504 scenarios. The networks were subsequently tested on unseen data sets and the unseen input parameters (formation properties) were then predicted using a global search procedure. The results obtained are very encouraging with formation properties being predicted to within 10% average relative error. The examples presented show that neural networks can be applied successfully to nuclear well‐logging problems. This enables the implementation of a fast inversion procedure, yielding quick and reliable values for unknown subsurface properties such as porosity, salinity and oil saturation.     

利用多元地震属性预测测井特性

通过寻找井旁地震数据与测井曲线的关系,将这一关系应用到远离井的区域(只有地质数据,但无测井)来预测测井的有关特性,其方法有单属性分析和多属性分析[1]。本文通过实例描述了多属性分析的特点及预测结果。从单属性回归到多属性预测、再到神经网络预测过渡时,预测能力持续提高。同时对地震属性的选择和有效性进行了讨论,将结果应用到整个二维地震剖面上,能更好地确定井以外区域的测井特性。     

Neural networks for quick earthquake damage estimation

This paper proposes the use of neural networks to predict damage due to earthquakes from the indices of recorded ground motion. Since the relationship between ground motion indices and resulting damage is difficult to express in mathematical form, neural networks are conveniently applied for this problem. Simulated earthquake ground motions are used to have a well-distributed data set and the ductility factor from non-linear analysis of two single-degree-of-freedom structural models is used to represent the damage. A sensitivity analysis procedure is described to identify qualitatively the input parameters that have a greater influence on the damage. The result of the trained neural network is then verified by using several recorded earthquake ground motions. It is found that some instability in the prediction can occur. Instability occurs when input values exceed the range of the training data. The neural network model using PGA and SI as input give the best performance in the recall tests using actual earthquake ground motion, demonstrating the usefulness of neural network models for the quick estimation of damage through earthquake intensity monitoring.     

A hybrid neural network model for typhoon-rainfall forecasting

A hybrid neural network model is proposed in this paper to forecast the typhoon rainfall. Two different types of artificial neural networks, the self-organizing map (SOM) and the multilayer perceptron network (MLPN), are combined to develop the proposed model. In the proposed model, a data analysis technique is developed based on the SOM, which can perform cluster analysis and discrimination analysis in one step. The MLPN is used as the nonlinear regression technique to construct the relationship between the input and output data. First, the input data are analyzed using a SOM-based data analysis technique. Through the SOM-based data analysis technique, input data with different properties are first divided into distinct clusters, which can help the multivariate nonlinear regression of each cluster. Additionally, the topological relationships among data are discovered from which more insight into the typhoon-rainfall process can be revealed. Then, for each cluster, the individual relationship between the input and output data is constructed by a specific MLPN. For evaluating the forecasting performance of the proposed model, an application is conducted. The proposed model is applied to the Tanshui River Basin to forecast the typhoon rainfall. The results show that the proposed model can forecast more precisely than the model developed by the conventional neural network approach.     

Quantitative geophysical pore‐type characterization and its geological implication in carbonate reservoirs

This paper discusses and addresses two questions in carbonate reservoir characterization: how to characterize pore‐type distribution quantitatively from well observations and seismic data based on geologic understanding of the reservoir and what geological implications stand behind the pore‐type distribution in carbonate reservoirs. To answer these questions, three geophysical pore types (reference pores, stiff pores and cracks) are defined to represent the average elastic effective properties of complex pore structures. The variability of elastic properties in carbonates can be quantified using a rock physics scheme associated with different volume fractions of geophysical pore types. We also explore the likely geological processes in carbonates based on the proposed rock physics template. The pore‐type inversion result from well log data fits well with the pore geometry revealed by a FMI log and core information. Furthermore, the S‐wave prediction based on the pore‐type inversion result also shows better agreement than the Greensberg‐Castagna relationship, suggesting the potential of this rock physics scheme to characterize the porosity heterogeneity in carbonate reservoirs. We also apply an inversion technique to quantitatively map the geophysical pore‐type distribution from a 2D seismic data set in a carbonate reservoir offshore Brazil. The spatial distributions of the geophysical pore type contain clues about the geological history that overprinted these rocks. Therefore, we analyse how the likely geological processes redistribute pore space of the reservoir rock from the initial depositional porosity and in turn how they impact the reservoir quality.     

Estimation of anisotropic hydraulic conductivity using geophysical data in a coastal aquifer of Karnataka,India

Estimation of hydraulic parameters is essential to understand the interaction between groundwater flow and seawater intrusion. Though several studies have addressed hydraulic parameter estimation, based on pumping tests as well as geophysical methods, not many studies have addressed the problem with clayey formations being present. In this study, a methodology is proposed to estimate anisotropic hydraulic conductivity and porosity values for the coastal aquifer with unconsolidated formations. For this purpose, the one-dimensional resistivity of the aquifer and the groundwater conductivity data are used to estimate porosity at discrete points. The hydraulic conductivity values are estimated by its mutual dependence with porosity and petrophysical parameters. From these estimated values, the bilinear relationship between hydraulic conductivity and aquifer resistivity is established based on the clay content of the sampled formation. The methodology is applied on a coastal aquifer along with the coastal Karnataka, India, which has significant clayey formations embedded in unconsolidated rock. The estimation of hydraulic conductivity values from the established correlations has a correlation coefficient of 0.83 with pumping test data, indicating good reliability of the methodology. The established correlations also enable the estimation of horizontal hydraulic conductivity on two-dimensional resistivity sections, which was not addressed by earlier studies. The inventive approach of using the established bilinear correlations at one-dimensional to two-dimensional resistivity sections is verified by the comparison method. The horizontal hydraulic conductivity agrees with previous findings from inverse modelling. Additionally, this study provides critical insights into the estimation of vertical hydraulic conductivity and an equation is formulated which relates vertical hydraulic conductivity with horizontal. Based on the approach presented, the anisotropic hydraulic conductivity of any type aquifer with embedded clayey formations can be estimated. The anisotropic hydraulic conductivity has the potential to be used as an important input to the groundwater models.     

松辽盆地深层火山岩含气储层产能预测

松辽盆地深层火山岩是当前大庆地球物理、地质、地球化学研究的主要领域之一,已取得丰硕的成果,火成岩含气储层产能作为一个表示动态特征的参数,是储层评价的重要指标之一,本文讨论了火成岩含气储层的产能与测井响应之间的关系,探讨了根据测井资料应用人工神经网络技术预测火成岩含气储层产能的方法,利用已知气井测试结果和测井资料作为网络的训练样本,根据网络学习训练结果,输入测井资料等静态参数,可预测储集层的产能,根据这种关系采用神经网络技术实现了测井对产能的预测评价,从而为大庆深部火成岩含气储层的开发提供了一定的依据。     

Porosity and permeability prediction from wireline logs using artificial neural networks: a North Sea case study

Estimations of porosity and permeability from well logs are important yet difficult tasks encountered in geophysical formation evaluation and reservoir engineering. Motivated by recent results of artificial neural network (ANN) modelling offshore eastern Canada, we have developed neural nets for converting well logs in the North Sea to porosity and permeability. We use two separate back-propagation ANNs (BP-ANNs) to model porosity and permeability. The porosity ANN is a simple three-layer network using sonic, density and resistivity logs for input. The permeability ANN is slightly more complex with four inputs (density, gamma ray, neutron porosity and sonic) and more neurons in the hidden layer to account for the increased complexity in the relationships. The networks, initially developed for basin-scale problems, perform sufficiently accurately to meet normal requirements in reservoir engineering when applied to Jurassic reservoirs in the Viking Graben area. The mean difference between the predicted porosity and helium porosity from core plugs is less than 0.01 fractional units. For the permeability network a mean difference of approximately 400 mD is mainly due to minor core-log depth mismatch in the heterogeneous parts of the reservoir and lack of adequate overburden corrections to the core permeability. A major advantage is that no a priori knowledge of the rock material and pore fluids is required. Real-time conversion based on measurements while drilling (MWD) is thus an obvious application.     

Committee neural networks for porosity and permeability prediction from well logs

Neural computing has moved beyond simple demonstration to more significant applications. Encouraged by recent developments in artificial neural network (ANN) modelling techniques, we have developed committee machine (CM) networks for converting well logs to porosity and permeability, and have applied the networks to real well data from the North Sea. Simple three‐layer back‐propagation ANNs constitute the blocks of a modular system where the porosity ANN uses sonic, density and resistivity logs for input. The permeability ANN is slightly more complex, with four inputs (density, gamma ray, neutron porosity and sonic). The optimum size of the hidden layer, the number of training data required, and alternative training techniques have been investigated using synthetic logs. For both networks an optimal number of neurons in the hidden layer is in the range 8–10. With a lower number of hidden units the network fails to represent the problem, and for higher complexity overfitting becomes a problem when data are noisy. A sufficient number of training samples for the porosity ANN is around 150, while the permeability ANN requires twice as many in order to keep network errors well below the errors in core data. For the porosity ANN the overtraining strategy is the suitable technique for bias reduction and an unconstrained optimal linear combination (OLC) is the best method of combining the CM output. For permeability, on the other hand, the combination of overtraining and OLC does not work. Error reduction by validation, simple averaging combined with range‐splitting provides the required accuracy. The accuracy of the resulting CM is restricted only by the accuracy of the real data. The ANN approach is shown to be superior to multiple linear regression techniques even with minor non‐linearity in the background model.     

Detection and identification of seismic signals recorded at Krakatau volcano (Indonesia) using artificial neural networks

The Anak Krakatau volcano (Indonesia) has been monitored by a multi-parametric system since 2005. A variety of signal types can be observed in the records of the seismic stations installed on the island volcano. These include volcano-induced signals such as LP, VT, and tremor-type events as well as signals not originating from the volcano such as regional tectonic earthquakes and transient noise signals. The work presented here aims at the realization of a system that automatically detects and identifies the signals in order to estimate and monitor current activity states of the volcano. An artificial neural network approach was chosen for the identification task. A set of parameters was defined, describing waveform and spectrogram properties of events detected by an amplitude-ratio-based (STA/LTA) algorithm. The parameters are fed into a neural network which is, after a training phase, able to generalize input data and identify corresponding event types. The success of the identification depends on the network architecture and training strategy. Several tests have been performed in order to determine appropriate network layout and training for the given problem. The performance of the final system is found to be well suited to get an overview of the seismic activity recorded at the volcano. The reliability of the network classifier, as well as general drawbacks of the methods used, are discussed.     

Rock Properties and Seismic Attenuation: Neural Network Analysis

—Using laboratory data, the influence of rock parameters on seismic attenuation has been analyzed using artificial neural networks and regression models. The predictive capabilities of the neural networks and multiple linear regresssion were compared. The neural network outperforms the multiple linear regression in predicting attenuation values, given a set of input of rock parameters. The neural network can make complex decision mappings and this capability is exploited to examine the influence of various rock parameters on the overall seismic attenuation. The results indicate that the most influential rock parameter on the overall attenuation is the clay content, closely followed by porosity. Though grain size contribution is of lower importance than clay content and porosity, its value of 16 percent is sufficiently significant to be considered in the modeling and interpretation of attenuation data.     

地震与测井数据综合预测裂缝发育带

章针对前新生代海相残留盆地碳酸盐岩裂缝识别这一难题提出一种利用地震与测井数据综合预测裂缝发育带的方法.用测井资料标识出裂缝发育层段，同时利用地震资料具有空间上数据点多、分布均匀的特点通过地震多属性的人工神经网络方法将测井与地震数据结合起来综合预测裂缝发育带，充分利用测井资料和地震资料的各自优势，达到在剖面上和区域上预测裂缝发育带的目的.本方法经过实际资料的处理与预测证实比常规方法预测精度高.     

A semi-empirical velocity-porosity-clay model for petrophysical interpretation of P- and S-velocities

We design a velocity–porosity model for sand-shale environments with the emphasis on its application to petrophysical interpretation of compressional and shear velocities. In order to achieve this objective, we extend the velocity–porosity model proposed by Krief et al., to account for the effect of clay content in sandstones, using the published laboratory experiments on rocks and well log data in a wide range of porosities and clay contents. The model of Krief et al. works well for clean compacted rocks. It assumes that compressional and shear velocities in a porous fluid-saturated rock obey Gassmann formulae with the Biot compliance coefficient. In order to use this model for clay-rich rocks, we assume that the bulk and shear moduli of the grain material, and the dependence of the compliance on porosity, are functions of the clay content. Statistical analysis of published laboratory data shows that the moduli of the matrix grain material are best defined by low Hashin–Shtrikman bounds. The parameters of the model include the bulk and shear moduli of the sand and clay mineral components as well as coefficients which define the dependence of the bulk and shear compliance on porosity and clay content. The constants of the model are determined by a multivariate non-linear regression fit for P- and S-velocities as functions of porosity and clay content using the data acquired in the area of interest. In order to demonstrate the potential application of the proposed model to petrophysical interpretation, we design an inversion procedure, which allows us to estimate porosity, saturation and/or clay content from compressional and shear velocities. Testing of the model on laboratory data and a set of well logs from Carnarvon Basin, Australia, shows good agreement between predictions and measurements. This simple velocity-porosity-clay semi-empirical model could be used for more reliable petrophysical interpretation of compressional and shear velocities obtained from well logs or surface seismic data.     

基于模糊粗糙集的机器学习储层参数预测

因为地震数据的三维空间分布优势,地震属性已经被广泛应用于含油气性预测、储层厚度预测、孔隙度预测等。但也存在地震属性之间信息冗余、属性与储层物性参数关系模糊的问题。针对这两个问题,将模糊粗糙理论和机器学习引入到储层参数预测中来。通过模糊粗糙集理论对地震属性进行约简,去除冗余信息,得到最优化的地震属性组合;将约简后的属性作为机器学习的输入,实现从地震属性到储层物性参数的非线性映射。该方法既保留了地震属性中有效信息,又避免了因输入变量过多而导致的网络模型训练困难。实际数据应用表明,属性约简的机器学习预测结果分辨率更高,并与数据吻合更好。      

Feed forward neural network and interpolation function models to predict the soil and subsurface sediments distribution in Bam,Iran

An application of the artificial neural network (ANN) approach for predicting mean grain size using electric resistivity data from Bam city is presented. A feed forward back propagation network was developed employing 45 sets of input data. The input variables in the ANN model are the electrical resistivity, water table as a Boolean value and depth; the output is the mean grain size. To demonstrate the authenticity of this approach, the network predictions are compared with those from interpolation methods and the same data. This comparison shows that the ANN approach performs better results. The predicted and observed mean grain size values were compared and show high correlation coefficients. The ANN approach maps show a high degree of correlation with well data based grain size maps and can therefore be used conservatively to better understand the influence of input parameters on sedimentological predictions.