Prediction Of Reservoir Parameters Research Articles

Simultaneous prediction of petrophysical properties and formation layered thickness from acoustic logging data using a modular cascading residual neural network (MCARNN) with physical constraints

Predicting petrophysical properties from logging data using deep neural networks (DNNs) improves the efficiency of subsurface exploration. Traditional methods, however, heavily rely on data pre-processing. Therefore, a novel Modular Cascading Residual Neural Network (MCARNN) is proposed to simultaneously retrieve layered thickness and petrophysical properties of porous media directly from raw acoustic logging data without pre-processing. The MCARNN mainly comprises two cascading modules: the first is tasked with retrieving layered thickness, while the second one further reconstructs petrophysical properties based on the thickness information. To more accurately simulate the uncertainties of the real world, we incorporate some known geophysical knowledge, including the distribution range of actual petrophysical parameters, physical constraints among porous medium parameters, configurations of logging instruments, and stratified geological models. During this process, all Biot's porous parameters are independently randomized to cover a wide range of geological rock types. Furthermore, to enhance MCARNN's generalization ability under different random noise distributions, we independently generate the training and testing datasets using distinct numerical algorithms, thus preventing the “inverse crime” phenomenon. Additionally, varying levels of Gaussian white noises are added to the training dataset, and the transmitter-receiver configuration of the logging instrument is varied to enhance the training dataset. Two numerical experiments are implemented to validate the reliability and superiority of the MCARNN. It is shown that the proposed MCARNN can not only accurately predict porous petrophysical parameters when the background layer configurations are known but also be able to simultaneously reconstruct layered thickness and petrophysical parameters with low model misfit. Notably, MCARNN successfully predicts the porous velocity dispersion curves within a broad frequency band based on limited frequency points—a task typically requiring auxiliary equations in traditional inversion methods, demonstrating the advanced predictive capability of the proposed MCARNN. Through this research, we not only address the deficiencies in existing real-world logging datasets in terms of sample volume and label reliability but also effectively integrate rock physics models with data-driven neural networks, providing a novel and effective pathway for reservoir parameter prediction in the field of geophysical logging.

Inter-well reservoir parameter prediction based on LSTM-Attention network and sedimentary microfacies

The essence of predicting inter-well reservoir parameters is to find the distribution pattern of these parameters in three-dimensional space, which is closely related to the distribution of sedimentary microfacies. Existing research on neural network-based prediction of reservoir parameters can be divided into two directions: vertical and horizontal. The former predicts the logging curves of individual wells, while the latter predicts average data points between wells. However, there is a lack of research on prediction methods for logging curves inter-wells within the entire three-dimensional space. This paper aims to incorporate geological conceptual information, such as sedimentary microfacies, into the spatial prediction of reservoir parameters, and to study the prediction method of well-logging curves, taking porosity as an example. The goal is to achieve the effect of obtaining a predicted well-log porosity curve for a designated location in the study area by inputting spatial coordinates and sedimentary microfacies information. The research method combines the Long Short-Term Memory (LSTM) network and Attention Mechanism, uses real logging data for experiments, conducts multi-method comparisons, discusses the impact of sedimentary microfacies and different neural network methods on the prediction effect of inter-well reservoir parameters, and carries out generalization experiments of the method in new areas. The experimental results show that the research method is effective and can achieve the purpose of describing the spatial distribution of reservoir parameters and guiding geological exploration and development work.

Reservoir parameters prediction based on spatially transferred long short-term memory network.

Reservoir reconstruction, where parameter prediction plays a key role, constitutes an extremely important part in oil and gas reservoir exploration. With the mature development of artificial intelligence, parameter prediction methods are gradually shifting from previous petrophysical models to deep learning models, which bring about obvious improvements in terms of accuracy and efficiency. However, it is difficult to achieve large amount of data acquisition required for deep learning due to the cost of detection, technical difficulties, and the limitations of complex geological parameters. To address the data shortage problem, a transfer learning prediction model based on long short-term memory neural networks has been proposed, and the model structure has been determined by parameter search and optimization methods in this paper. The proposed approach transfers knowledge from historical data to enhance new well prediction by sharing some parameters in the neural network structure. Moreover, the practicality and effectiveness of this method was tested by comparison based on two block datasets. The results showed that this method could significantly improve the prediction accuracy of the reservoir parameters in the event of data shortage.

Reconstruction of Missing Well-Logs Using Facies-Informed Discrete Wavelet Transform and Time Series Regression

Summary Geophysical logging is widely used in lithofacies identification, reservoir parameter prediction, and geological modeling. However, it is common to have well-log sections with low-quality and/or missing segments. Repeating the well-log measurements is not only expensive but might also be impossible depending on the condition of the borehole walls. In these situations, reliable and accurate well-log prediction is, therefore, necessary in different stages of the geomodeling workflow. In this study, we propose a time series regression model to predict missing well-log data, incorporating facies information as an additional geological input and using discrete wavelet transform (DWT) to denoise the input data set. The main contributions of this work are threefold: (i) We jointly use facies information with well logs as the input data set; (ii) we use DWT to denoise the input data and consequently improve the signal-to-noise ratio of the input data; and (iii) we regard the depth domain as the time domain and use a time series regression algorithm for log reconstruction modeling. We show a real application example in two distinct scenarios. In the first, we predict missing well-log intervals. In the second, we predict complete well logs. The experimental results show the ability of the proposed prediction model to recover missing well-log data with high accuracy levels.

High-Fidelity Permeability and Porosity Prediction Using Deep Learning With the Self-Attention Mechanism.

Accurate estimation of reservoir parameters (e.g., permeability and porosity) helps to understand the movement of underground fluids. However, reservoir parameters are usually expensive and time-consuming to obtain through petrophysical experiments of core samples, which makes a fast and reliable prediction method highly demanded. In this article, we propose a deep learning model that combines the 1-D convolutional layer and the bidirectional long short-term memory network to predict reservoir permeability and porosity. The mapping relationship between logging data and reservoir parameters is established by training a network with a combination of nonlinear and linear modules. Optimization algorithms, such as layer normalization, recurrent dropout, and early stopping, can help obtain a more accurate training model. Besides, the self-attention mechanism enables the network to better allocate weights to improve the prediction accuracy. The testing results of the well-trained network in blind wells of three different regions show that our proposed method is accurate and robust in the reservoir parameters prediction task.

Seismic full-waveform inversion based on superresolution for high-precision prediction of reservoir parameters

The frequency band limitation of seismic data limits the resolution of full-waveform inversion (FWI) results. In addition, the high computational cost seriously affects the practical application of FWI. To alleviate these concerns, an FWI method based on superresolution (SR-FWI) is developed to improve the prediction efficiency and accuracy of the reservoir parameters. A channel attention mechanism is introduced for the multifrequency characteristics of the model images. A constrained residual channel attention network (CRCAN) is built for superresolution (SR) by adding structural constraints to the loss function of a deep learning network. A total of 65,000 sets of geologic models and natural images constitute the network training data, 90% of which are used for training with the rest used for testing. The iterative calculation for FWI is time-consuming; hence, SR is applied to the iterative process to reduce the number of iterations and accelerate the model update. Low-resolution images along with the synthetic and field data are used for the evaluation of the CRCAN and SR-FWI algorithms, respectively. The test results find that CRCAN can effectively improve the image resolution, whereas SR-FWI is beneficial due to its high efficiency and precision, especially in predicting the stratum edge and small-scale anomalies. Therefore, SR-FWI is a powerful means of reservoir static and dynamic detection and can provide high-resolution information for projects, such as resource development and CO2 storage.

The prediction of shale gas reservoir parameters through a multilayer transfer learning network

SUMMARY Due to the strong heterogeneity of the reservoir, it is difficult to extend the classical reservoir parameter prediction model to the new work area. Traditional geological methods rely on experience, while machine learning algorithms lack consideration of reservoir heterogeneity. This makes it impossible for these methods to achieve good predictive performance in new work areas. In this paper, we discuss designing a multilayer deep transfer learning network (MDTL). The network enables accurate prediction of reservoir parameters in new areas based on core and logging data from mature areas. For the architectural design, we establish a network structure that realizes data interaction in different work areas. For network parameter optimization, we design a multilayer maximum mean discrepancy loss function. It guides network training to learn feature knowledge about data in different work areas. For logging data processing, we explore a support vector machine method to remove abnormal noise in logging data. We apply MDTL to predict shale gas porosity and total organic carbon content in southern Sichuan. Compared to state-of-the-art reservoir parameter prediction models, MDTL achieves the best performance in new well areas. The network integrates the idea of multilayer transfer learning, reduces the influence of reservoir heterogeneity and effectively ensures the prediction accuracy.

Gated recurrent unit neural network (GRU) based on quantile regression (QR) predicts reservoir parameters through well logging data

The prediction of reservoir parameters is the most important part of reservoir evaluation, and porosity is very important among many reservoir parameters. In order to accurately measure the porosity of the core, it is necessary to take cores for indoor experiments, which is tedious and difficult. To solve this problem, this paper introduces machine learning models to estimate porosity through logging parameters. In this paper, gated recurrent unit neural network based on quantile regression method is introduced to predict porosity. Porosity measurement is implemented by taking cores for indoor experiments. The data is divided into training set and test set. The logging parameters are used as the input parameters of the prediction model, and the porosity parameters measured in the laboratory are used as the output parameters. Experimental results show that the quantile regression method improves the accuracy of the gated recurrent unit neural network, and the RMSE (Root Mean Square Error) of the unoptimized GRU neural network is 0.1774, after optimization, the RMSE is 0.1061. By comparing with the most widely used BP neural network, the accuracy of the method proposed in this paper is much higher than that of BP neural network. This shows that the gated recurrent neural network method based on quantile regression is excellent in predicting reservoir parameters.

ОТДЕЛЬНЫЕ ВОПРОСЫ ДИНАМИЧЕСКОЙ ИНВЕРСИИ ВОЛНОВОГО ПОЛЯ

Currently, reservoir parameter prediction methods based on the amplitude wave field inversion play an important role in solving geological problems in the joint interpretation of seismic and well data. From a formal point of view, well data are used to obtain additional conditions for solving inverse problems of seismic. But the formal description of solution is conditional, there is rather an adjustment of results obtained independently from seismic and well data. The paper considers the following issues: stability of the simultaneous inversion parameter estimation; calculation features of background model and coordination of background and seismic solution components; compliance with restrictions on the maximum sourcereceiver distances by kinematic muting and the range of reflection angles during AVO analysis. The AVO effect is analyzed, it displays itself in a change of the reflection coefficient sign and, correspondently, phase inversion of reflected wave.

Porosity and permeability prediction using a transformer and periodic long short-term network

Effective reservoir parameter prediction is important for subsurface characterization and understanding fluid migration. However, conventional methods for obtaining porosity and permeability are based on either core measurements or mathematical/petrophysical modeling, which are expensive or inefficient. In this study, we develop a reliable and low-cost deep learning (DL) framework for reservoir permeability and porosity prediction from real logging data at different regions. We leverage an advanced learning architecture (i.e., the transformer model) and design a new regression network (RPTransformer) that is sensitive to the depth period change of the logging data. The RPTransformer is composed of 1D convolutional, long short-term memory (LSTM), and transformer layers. First, we use a 1D convolutional layer for the first layer of the network to extract significant features from the logging data. Then, the nonlinear mapping relationships between logging data and reservoir parameters are established using several LSTM layers with a period parameter. Afterward, we use the encoder in the vision transformer with the self-attention mechanism to further extract logging data features. The developed network is a data-driven supervised learning framework and indicates highly accurate and robust prediction results when applied to different geographic regions. To demonstrate the reliable prediction performance of our network, we compare it with several classic machine learning and state-of-the-art DL methods, e.g., random forest, multilayer LSTM, and long short-term time-series network (LSTNet). More importantly, we find the generalization and uncertainty of the network in real-world applications through comprehensive numerical experiments.

Coalbed Methane Reservoir Parameter Prediction and Sweet-Spot Comprehensive Evaluation Based on 3D Seismic Exploration: A Case Study in Western Guizhou Province, China

As a kind of clean energy, the exploration and development of coalbed methane (CBM) are of great importance and significance. In this paper, the CBM reservoir parameters of a working area in Western Guizhou Province, China, were predicted by using 3D seismic exploration technology, and the sweet-spot area was evaluated based on the prediction results. In terms of data interpretation, while the traditional technology is extended to the study of CBM, new technologies and methods suitable for CBM were also actively explored, especially in the quantitative prediction methods of parameters such as coal seam thickness, CBM content, coal body structure, in situ stress, etc., and a three-highs quantitative prediction technology of CBM sweet spots characterized by high precision, high resolution, and high coincidence was explored, which is based on logging evaluation and petrophysics. In addition, in the comprehensive evaluation of multiparameter sweet spots, the multiparameter weighted step-by-step evaluation method based on inversion was innovatively proposed, which made the prediction of CBM sweet spots more focused, effective, and practical, and the distribution law of sweet spots was more consistent with the geological law, providing enough basis for subsequent well location deployment. Based on this method, the C409 coal seam was selected as the sweet spot.

Complex Resistivity Anisotropy Response Characteristics of Wufeng-Longmaxi Formation Shale in Southern Sichuan

Electrical exploration has become an important means of shale gas reservoir exploration and evaluation, and is expected to play a key role in the later stages of reservoir fracturing and development. At present, the research on the electrical response characteristics of shale gas reservoirs and their relationship with reservoir parameters is extensive and in-depth, but there is little research on their complex resistivity anisotropy characteristics and influencing factors, which restricts petrophysical modeling and reservoir parameter prediction, and reduces the reliability of shale gas exploration and reservoir evaluation by electromagnetic methods. In this paper, shale samples from the Longmaxi Formation and the Wufeng Formation of shale gas wells in southern Sichuan were collected, the complex resistivity of 34 shales in bedding direction and vertical bedding direction were measured, and the induced polarization (IP) parameters of shales were extracted by inversion. The electrical anisotropy response characteristics under different temperature and pressure conditions were analyzed, and the influencing factors and laws of complex resistivity anisotropy of shales were revealed. Combined with the test results of shale porosity and permeability, the evaluation model of resistivity, polarizability and porosity and permeability parameters was established. The research results have formed a set of testing methods and analysis techniques for electrical anisotropy of shale reservoirs, which are mainly based on complex resistivity parameter testing. It is helpful to understand the electrical anisotropy characteristics of shale gas reservoirs in southern Sichuan; this will provide the theoretical and physical basis for shale gas reservoir evaluation and fracturing monitoring by electrical exploration.

Simultaneous prediction of multiple physical parameters using gated recurrent neural network: Porosity, water saturation, shale content

Reservoir parameter prediction is of significant value to oil and gas exploration and development. Artificial intelligence models are developing rapidly in reservoir parameter prediction. Unfortunately, current research has focused on multi-input single-output prediction models. Meaning, these models use a large amount of logging or seismic data to predict the petrophysical properties of a single reservoir. Another prominent problem is that most mechanistic learning studies have focused on using logging data (e.g., gamma ray and resistivity) to make predictions of reservoir parameters. Although these studies have yielded promising accuracy, a great shortcoming is the inability to obtain such data in logs by seismic inversion. The value of our research work is to achieve a complete description of the reservoir using the elastic parameters from the seismic inversion. We developed a deep learning method based on gated recurrent neural network (GRNN) suitable for simultaneous prediction of porosity, saturation and shale content in the reservoir. GRNN is based on Gated Recurrent Unit (GRU), which can automatically update and reset the hidden state. The input parameters to the model are compressive wave velocity, shear wave velocity and density. The model is trained to fit nonlinear relationships between input parameters and multiple physical parameters. We employed two wells: one for testing and the other for training. 20% of the data in the training wells were used as the validation set. In preprocessing, we performed z-score whitening on the input data. During the training phase, the model hyperparameters were optimized based on the mean absolute error (MAE) box plots of the validation set. Experiments on the test data show that the model has superior robustness and accuracy compared to the conventional recurrent neural network (RNN). In the GRNN prediction results of the test set, the MAE is 0.4889 and the mean squared error (MSE) is 0.5283. Due to the difference in input parameters, our prediction is weaker than the research method using logging data. However, our proposed method has higher practical value in exploration work.

FTCN: A Reservoir Parameter Prediction Method Based on a Fusional Temporal Convolutional Network

Predicting reservoir parameters accurately is of great significance in petroleum exploration and development. In this paper, we propose a reservoir parameter prediction method named a fusional temporal convolutional network (FTCN). Specifically, we first analyze the relationship between logging curves and reservoir parameters. Then, we build a temporal convolutional network and design a fusion module to improve the prediction results in curve inflection points, which integrates characteristics of the shallow convolution layer and the deep temporal convolution network. Finally, we conduct experiments on real logging datasets. The results indicate that compared with the baseline method, the mean square errors of FTCN are reduced by 0.23, 0.24 and 0.25 in predicting porosity, permeability, and water saturation, respectively, which shows that our method is more consistent with the actual reservoir geological conditions. Our innovation is that we propose a new reservoir parameter prediction method and introduce the fusion module in the model innovatively. Our main contribution is that this method can well predict reservoir parameters even when there are great changes in formation properties. Our research work can provide a reference for reservoir analysis, which is conducive to logging interpreters’ efforts to analyze rock strata and identify oil and gas resources.

Reservoir Parameter Prediction Based on the Neural Random Forest Model

Porosity and saturation are the basis for describing reservoir properties and formation characteristics. The traditional, empirical, and formulaic methods are unable to accurately capture the nonlinear mapping relationship between log data and reservoir physical parameters. To solve this problem, in this study, a novel hybrid model (NRF) combining neural network (NN) and random forest (RF) was proposed based on well logging data to predict the porosity and saturation of shale gas reservoirs. The database includes six horizontal wells, and the input logs include borehole diameter, neutron, density, gamma-ray, and acoustic and deep investigate double lateral resistivity log. The porosity and saturation were chosen as outputs. The NRF model with independent and joint training was designed to extract key features from well log data and physical parameters. It provides a promising method for forecasting the porosity and saturation with R2above 0.94 and 0.82 separately. Compared with baseline models (NN and RF), the NRF model with joint training obtains the unsurpassed performance to predict porosity with R2above 0.95, which is 1.1% higher than that of the NRF model with independent training, 3.9% higher than RF, and superiorly greater than NN. For the prediction of saturation, the NRF model with joint training is still superior to other algorithms, with R2above 0.84, which is 2.1% higher than that of the NRF model with independent training and 7.0% higher than RF. Furthermore, the NRF model has a similar data distribution with measured porosity and saturation, which demonstrates the NRF model can achieve greater stability. It was proven that the proposed NRF model can capture the complex relationship between the logging data and physical parameters more accurately, and can serve as an economical and reliable alternative tool to give a reliable prediction.

Log interpretation method of resistivity low-contrast oil pays in Chang 8 tight sandstone of Huanxian area, Ordos Basin by support vector machine

Resistivity low-contrast oil pays are a kind of unconventional oil resource with no obvious difference in physical and electrical properties from water layers, which makes it difficult to be identified based on the characteristics of the geophysical well logging response. In this study, the support vector machine (SVM) technology was used to interpret the resistivity low-contrast oil pays in Chang 8 tight sandstone reservoir of Huanxian area, Ordos Basin. First, the input data sequences of logging curves were selected by analyzing the relationship between reservoir fluid types and logging data. Then, the SVM classification model for fluid identification and SVR regression model for reservoir parameter prediction were constructed. Finally, these two models were applied to interpret the resistivity low-contrast oil pays in the study area. The application results show that the fluid recognition accuracy of the SVM classification model is higher than that of the logging cross plot method, back propagation neural network method and radial basis function neural network method. The calculation accuracy of permeability and water saturation predicted by the SVR regression model is higher than that based on the experimental fitting model, which indicates that it is feasible to carry out logging interpretation and evaluation of the resistivity low-contrast oil pays by the SVM method. The research results not only provide an important reference and basis for the review of old wells but also provide technical support for the exploration and development of new strata.

Petrophysical experiment-based evaluation method for the saturation of gas hydrate

The seas around China contain abundant non-diagenetic gas hydrate deposits. The traditional logging and evaluation techniques based on conventional oil and gas no longer satisfy the evaluation requirements of offshore gas hydrate layers, and it remains difficult to calculate the hydrate saturation accurately, which seriously restricts the accurate evaluation of the resources. Petrophysical experiments are an important means to study gas hydrate. By simulating the hydrate formation conditions in sediments under natural conditions, the study of various petrophysical properties of gas hydrate is essential to the exploration and development of hydrates and the prediction of reservoir parameters. In this study, a tetrahydrofuran (THF) solution, montmorillonite, and natural sea sand were selected to simulate the growth of gas hydrate based on its characteristics in the study area, and in situ CT scanning experiments were carried out at different hydrate formation stages to determine the microscopic morphologies of gas hydrates within the pore space. In addition, the acoustic wave and resistivity of hydrates with different saturations were measured. Forward modeling of a differential effective medium (DEM) model was conducted for the hydrate reservoirs with different saturations, and the gas hydrate morphology was analyzed. Thereafter, the hydrate saturation predictions based on the petrophysical experiments were analyzed. The results of these analyses indicate that the microscopic morphology of hydrates in the study area is primarily pore-filling form. Based on the saturation estimations using the Archie's law, Wood equation, time-average equation and the DEM method, and under the constraint of the acoustic and electrical measurement results of the petrophysical experiments, an electric acoustic joint model was built by integrating the resistivity and acoustic waves. The estimation results indicate that the joint electro–acoustic model is the best method for calculating the saturation in the study area. • Design of experiments to simulate gas hydrate generation with THF, montmorillonite, and natural sea sand. • Multiple hydrate saturation prediction methods based on acoustic, electrical, and petrophysical models. • Determination of microscopic morphologies of gas hydrate. • A joint acoustic and resistivity hydrate saturation prediction model based on hydrate petrophysical experiments.

Prediction of Reservoir Parameters of Cambrian Sandstones Using Petrophysical Modelling—Geothermal Potential Study of Polish Mainland Part of the Baltic Basin

Important factors controlling the effective utilization of geothermal energy are favorable reservoir properties of rock formations, which determine both the availability and the transfer opportunities of reservoir fluids. Hence, crucial to the successful utilization of a given reservoir is the preliminary recognition of distribution of reservoir parameters as it enables the researchers to select the prospective areas for localization of future geothermal installations and to decide on their characters. The objectives of this paper are analyses and discussion of the properties of quartz sandstones buried down to a depth interval from about 3000 to under 5000 m below surface. These sandstones belong to Ediacaran–Lowery Cambrian Łeba, Kluki and Żarnowiec formations. The source data from the Słupsk IG-1 provided the basis for 1D reconstruction of burial depth and paleothermal conditions as well as enabled the authors to validate of the results of 2D models. Then, porosity distribution within the reservoir formation was determined using the modelings of both the mechanical and chemical compactions along the 70 km-long B’-B part of the A’-A cross-section Bornholm-Słupsk IG-1 well. The results confirmed the low porosities and permeabilities as well as high temperatures of the analyzed rock formations in the Słupsk IG-1 well area. Towards the coast of the Baltic Sea, the porosity increases to more than 5%, while the temperature decreases, but is still relatively high, at about 130 °C. This suggests the application of an enhanced geothermal system or hot dry rocks system as principal methods for using geothermal energy.

Prediction of reservoir parameters in gas hydrate sediments using artificial intelligence (AI): A case study in Krishna–Godavari basin (NGHP Exp-02)

The estimation of accurate reservoir parameters is essential for conventional and non-conventional hydrocarbon prospects. An artificial neural network has been developed to predict the reservoir parameters (porosity and saturation of gas hydrates) in a silty-sand, sandy-silt and pelagic-poor clay reservoir at two neighbour wells using the petrophysical information at another well in the Krishna–Godavari basin. The well log data were acquired during the Expedition-02 of Indian National Gas Hydrates Program (NGHP Exp-02). The estimation of gas hydrate saturation using Archie’s equation may be erroneous, as it is valid for the quantification of conventional hydrocarbons in the clean sand reservoir. Since the study area is clay dominated, it is subjective to adjust Archie’s exponents so that it matches with the saturation, measured from the core data. To overcome this problem of estimating the reservoir parameters in such a scenario, first of all we have derived porosity from the density log data and estimated saturation by employing modified Archie’s equation to the resistivity log data at one well. In order to train the network, the log data at one well are taken as inputs and corresponding porosity and saturation are taken as outputs. The reservoir parameters are then predicted at two neighbour wells using the wireline log data as input in those two wells. The predicted porosity and saturation of gas hydrates are alike to the traditionally estimated porosity and saturation at the neighbour wells. The predicted porosity in the studied region varies between 33 and 76%, whereas the saturation of gas hydrates ranges between 3.39 and 86.92%. This shows that the designed network can be used to estimate the reservoir parameters directly from the well log data in the same reservoirs.

Semi-Supervised Learning Based on Generative Adversarial Network and Its Applied to Lithology Recognition

Lithology recognition is an essential part of reservoir parameter prediction. Compared to conventional algorithms, deep learning that needs a large amount of training data as support can extract features automatically. In the process of real data acquisition, the labeled data account for only a small portion due to high drilling cost, and it is difficult to achieve the data size required for deep learning training, resulting in a significant variance of the recognition model. In this paper, for this shortage, a semi-supervised algorithm based on generative adversarial network (GAN) with Gini-regularization is proposed, called SGAN_G, which takes borehole-side data as labeled data and seismic data as unlabeled data. First, the SGAN_G is trained by Adam (a method for stochastic optimization) algorithm and utilizes a discriminator to lithology recognition. And, we add the entropy regularization to the initial loss function which enhances the convergence speed and accuracy of the model. Eventually, we propose a novel sampling approach which employs multiple sampling points of seismic data as inputs to use the stratum information implicitly. Through the experimental comparison with a variety of supervised approaches, we can see that the SGAN_G can achieve higher prediction accuracy by using unlabeled data effectively.