Traditional deep learning methods pursue complex and single network architectures without considering the petrophysical relationship between different elastic parameters.The mathematical and statistical significance o...Traditional deep learning methods pursue complex and single network architectures without considering the petrophysical relationship between different elastic parameters.The mathematical and statistical significance of the inversion results may lead to model overfitting,especially when there are a limited number of well logs in a working area.Multitask learning provides an eff ective approach to addressing this issue.Simultaneously,learning multiple related tasks can improve a model’s generalization ability to a certain extent,thereby enhancing the performance of related tasks with an equal amount of labeled data.In this study,we propose an end-to-end multitask deep learning model that integrates a fully convolutional network and bidirectional gated recurrent unit for intelligent prestack inversion of“seismic data to elastic parameters.”The use of a Bayesian homoscedastic uncertainty-based loss function enables adaptive learning of the weight coeffi cients for diff erent elastic parameter inversion tasks,thereby reducing uncertainty during the inversion process.The proposed method combines the local feature perception of convolutional neural networks with the long-term memory of bidirectional gated recurrent networks.It maintains the rock physics constraint relationships among diff erent elastic parameters during the inversion process,demonstrating a high level of prediction accuracy.Numerical simulations and processing results of real seismic data validate the eff ectiveness and practicality of the proposed method.展开更多
Stochastic seismic inversion is the combination of geostatistics and seismic inversion technology which integrates information from seismic records, well logs, and geostatistics into a posterior probability density fu...Stochastic seismic inversion is the combination of geostatistics and seismic inversion technology which integrates information from seismic records, well logs, and geostatistics into a posterior probability density function （PDF） of subsurface models. The Markov chain Monte Carlo （MCMC） method is used to sample the posterior PDF and the subsurface model characteristics can be inferred by analyzing a set of the posterior PDF samples. In this paper, we first introduce the stochastic seismic inversion theory, discuss and analyze the four key parameters： seismic data signal-to-noise ratio （S/N）, variogram, the posterior PDF sample number, and well density, and propose the optimum selection of these parameters. The analysis results show that seismic data S/N adjusts the compromise between the influence of the seismic data and geostatistics on the inversion results, the variogram controls the smoothness of the inversion results, the posterior PDF sample number determines the reliability of the statistical characteristics derived from the samples, and well density influences the inversion uncertainty. Finally, the comparison between the stochastic seismic inversion and the deterministic model based seismic inversion indicates that the stochastic seismic inversion can provide more reliable information of the subsurface character.展开更多
Non-liner wave equation inversion,wavelet analysis and artificial neural networks were used to obtain stratum parameters and the distribution of thin coal seams.The lithology of the water-bearing/resisting layer in th...Non-liner wave equation inversion,wavelet analysis and artificial neural networks were used to obtain stratum parameters and the distribution of thin coal seams.The lithology of the water-bearing/resisting layer in the Quaternary system was also predicted.The implementation process included calculating the well log parameters,stratum contrasting the seismic data and the well logs,and extracting,studying and predicting seismic attributes.Seismic inversion parameters,including the layer velocity and wave impedance,were calculated and effectively used for prediction and analysis.Prior knowledge and seismic interpretation were used to remedy a dearth of seismic data during the inversion procedure.This enhanced the stability of the inversion method.Non-linear seismic inversion and artificial neural networks were used to interpret coal seismic lithology and to study the water-bearing/resisting layer in the Quaternary system.Interpretation of the 1～2 m thin coal seams,and also of the water-bearing/resisting layer in the Quaternary system,is provided.The upper mining limit can be lifted from 60 m to 45 m.The predictions show that this method can provide reliable data useful for thin coal seam exploitation and for lifting the upper mining limit,which is one of the principles of green mining.展开更多
To improve the accuracy of inversion results,geological facies distributions are considered as additional constraints in the inversion process.However,the geological facies itself also has its own uncertainty.In this ...To improve the accuracy of inversion results,geological facies distributions are considered as additional constraints in the inversion process.However,the geological facies itself also has its own uncertainty.In this paper,the initial sedimentary facies maps are obtained by integrated geological analysis from well data,seismic attributes,and deterministic inversion results.Then the fi rst iteration of facies-constrained seismic inversion is performed.According to that result and other data such as geological information,the facies distribution can be updated using cluster analysis.The next round of facies-constrained inversion can then be performed.This process will be repeated until the facies inconsistency or error before and after the inversion is minimized.It forms a new iterative facies-constrained seismic inversion technique.Compared with conventional facies-constrained seismic inversion,the proposed method not only can reduces the non-uniqueness of seismic inversion results but also can improves its resolution.As a consequence,the sedimentary facies will be more consistent with the geology.A practical application demonstrated that the superposition relationship of sand bodies could be better delineated based on this new seismic inversion technique.The result highly increases the understanding of reservoir connectivity and its accuracy,which can be used to guide further development.展开更多
During the process of coal prospecting and exploration, different measurement time, different logging instruments and series can lead to systematic errors in well logs. Accordingly, all logging curves need to be norma...During the process of coal prospecting and exploration, different measurement time, different logging instruments and series can lead to systematic errors in well logs. Accordingly, all logging curves need to be normalized in the mining area. By studying well-logging normalization methods, and focusing on the characteristics of the coalfield, the frequency histogram method was used in accordance with the condition of the Guqiao Coal Mine. In this way, the density and sonic velocity at marker bed in the non-key well were made to close to those in the key well, and were eventually equal. Well log normalization was completed when this method was applied to the entire logging curves. The results show that the scales of logging data were unified by normalizing coal logging curves, and the logging data were consistent with wave impedance inversion data. A satisfactory inversion effect was obtained.展开更多
With the development of computational power, there has been an increased focus on data-fitting related seismic inversion techniques for high fidelity seismic velocity model and image, such as full-waveform inversion a...With the development of computational power, there has been an increased focus on data-fitting related seismic inversion techniques for high fidelity seismic velocity model and image, such as full-waveform inversion and least squares migration. However, though more advanced than conventional methods, these data fitting methods can be very expensive in terms of computational cost. Recently, various techniques to optimize these data-fitting seismic inversion problems have been implemented to cater for the industrial need for much improved efficiency. In this study, we propose a general stochastic conjugate gradient method for these data-fitting related inverse problems. We first prescribe the basic theory of our method and then give synthetic examples. Our numerical experiments illustrate the potential of this method for large-size seismic inversion application.展开更多
In order to improve the fine structure inversion ability of igneous rocks for the exploration of underlying strata, based on particle swarm optimization(PSO), we have developed a method for seismic wave impedance inve...In order to improve the fine structure inversion ability of igneous rocks for the exploration of underlying strata, based on particle swarm optimization(PSO), we have developed a method for seismic wave impedance inversion. Through numerical simulation, we tested the effects of different algorithm parameters and different model parameterization methods on PSO wave impedance inversion, and analyzed the characteristics of PSO method. Under the conclusions drawn from numerical simulation, we propose the scheme of combining a cross-moving strategy based on a divided block model and high-frequency filtering technology for PSO inversion. By analyzing the inversion results of a wedge model of a pitchout coal seam and a coal coking model with igneous rock intrusion, we discuss the vertical and horizontal resolution, stability and reliability of PSO inversion. Based on the actual seismic and logging data from an igneous area, by taking a seismic profile through wells as an example, we discuss the characteristics of three inversion methods, including model-based wave impedance inversion, multi-attribute seismic inversion based on probabilistic neural network(PNN) and wave impedance inversion based on PSO.And we draw the conclusion that the inversion based on PSO method has a better result for this igneous area.展开更多
In this paper seismic inversion was used as a key technique and the seismic wavelet most suitable to the actual underground situation was extracted with the higher-order statistics algorithm. The wavelets extracted in...In this paper seismic inversion was used as a key technique and the seismic wavelet most suitable to the actual underground situation was extracted with the higher-order statistics algorithm. The wavelets extracted in this way and the wavelets extracted with the seismic statistics techniques were used separately for inverting the seismic data of the southern part of Tahe oilfield, Tarim basin. The results showed that the resolution of the wavelet inversion with the higher-order statistics method was greatly improved, and the wavelet-inverted section could better distinguish the thin sandstone reservoirs of the upper and lower Carboniferous and their lateral distribution, providing a reliable basis of analysis for the study of thin sandstone reservoirs.展开更多
Geoscientific evidence shows that various parameters such as compaction,buoyancy effect,hydrocarbon maturation,gas effect and tectonic activities control the pore pressure of sub-surface geology.Spatially controlled g...Geoscientific evidence shows that various parameters such as compaction,buoyancy effect,hydrocarbon maturation,gas effect and tectonic activities control the pore pressure of sub-surface geology.Spatially controlled geoscientific data in the tectonically active areas is significantly useful for robust estimation of pre-drill pore pressure.The reservoir which is tectonically complex and pore pressure is changing frequently that circumference motivated us to conduct this study.The changes in pore pressure have been captured from the fine-scale to the broad scale in the Jaisalmer sub-basin.Pore pressure variation has been distinctly observed in pre-and post-Jurassic age based on the current study.Post-stack seismic inversion study was conducted to capturing the variation of pore pressure.Analysis of low-frequency spectrum and integrated interval velocity model provided a detailed feature of pore pressure in each compartment of the study area.Pore pressure estimated from well log data was correlated with seismic inversion based result.Based on the current study one well has been proposed where pore pressure was estimated and two distinguished trends are identified in the study zone.The approaches of the current study were analysed thoroughly and it will be highly useful in complex reservoir condition where pore pressure varies frequently.展开更多
The objective of this work is to implement a pseudo-forward equation which is called PFE to transform data （similarity attribute） to model parameters （porosity） in a gas reservoir in the F3 block of North Sea. Thi...The objective of this work is to implement a pseudo-forward equation which is called PFE to transform data （similarity attribute） to model parameters （porosity） in a gas reservoir in the F3 block of North Sea. This equation which is an experimental model has unknown constants in its structure; hence, a least square solution is applied to find the best constants. The results derived from solved equa- tions show that the errors on measured data are mapped into the errors of estimated constants; hence, Tikhonov regularization is used to improve the estimated parameters. The results are compared with a conventional method such as cross plotting between acoustic impedance and porosity values to validate the PFE model. When the testing dataset in sand units was used, the correlation coefficient between two variables （actual and predicted values） was obtained as 0.720 and 0.476 for PFE model and cross-plotting analysis, respectively. Therefore, the testing dataset validates rela- tively well the PFE optimized by Tikhonov regularization in sand units of a gas reservoir. The obtained results indi- cate that PFE could provide initial information about sandstone reservoirs. It could estimate reservoir porosity distribution approximately and it highlights bright spots and fault structures such as gas chimneys and salt edges.展开更多
We propose to use a Few-Shot Learning(FSL)method for the pre-stack seismic inversion problem in obtaining a high resolution reservoir model from recorded seismic data.Recently,artificial neural network(ANN)demonstrate...We propose to use a Few-Shot Learning(FSL)method for the pre-stack seismic inversion problem in obtaining a high resolution reservoir model from recorded seismic data.Recently,artificial neural network(ANN)demonstrates great advantages for seismic inversion because of its powerful feature extraction and parameter learning ability.Hence,ANN method could provide a high resolution inversion result that are critical for reservoir characterization.However,the ANN approach requires plenty of labeled samples for training in order to obtain a satisfactory result.For the common problem of scarce samples in the ANN seismic inversion,we create a novel pre-stack seismic inversion method that takes advantage of the FSL.The results of conventional inversion are used as the auxiliary dataset for ANN based on FSL,while the well log is regarded the scarce training dataset.According to the characteristics of seismic inversion(large amount and high dimensional),we construct an arch network(A-Net)architecture to implement this method.An example shows that this method can improve the accuracy and resolution of inversion results.展开更多
Seismic inversion and basic theory are briefly presented and the main idea of this method is introduced. Both non-linear wave equation inversion technique and Complete Utilization of Samples Information (CUSI) neural ...Seismic inversion and basic theory are briefly presented and the main idea of this method is introduced. Both non-linear wave equation inversion technique and Complete Utilization of Samples Information (CUSI) neural network analysis are used in lithological interpretation in Jibei coal field. The prediction results indicate that this method can provide reliable data for thin coal exploitation and promising area evaluation.展开更多
As sandstone layers in thin interbedded section are difficult to identify,conventional model-driven seismic inversion and data-driven seismic prediction methods have low precision in predicting them.To solve this prob...As sandstone layers in thin interbedded section are difficult to identify,conventional model-driven seismic inversion and data-driven seismic prediction methods have low precision in predicting them.To solve this problem,a model-data-driven seismic AVO(amplitude variation with offset)inversion method based on a space-variant objective function has been worked out.In this method,zero delay cross-correlation function and F norm are used to establish objective function.Based on inverse distance weighting theory,change of the objective function is controlled according to the location of the target CDP(common depth point),to change the constraint weights of training samples,initial low-frequency models,and seismic data on the inversion.Hence,the proposed method can get high resolution and high-accuracy velocity and density from inversion of small sample data,and is suitable for identifying thin interbedded sand bodies.Tests with thin interbedded geological models show that the proposed method has high inversion accuracy and resolution for small sample data,and can identify sandstone and mudstone layers of about one-30th of the dominant wavelength thick.Tests on the field data of Lishui sag show that the inversion results of the proposed method have small relative error with well-log data,and can identify thin interbedded sandstone layers of about one-15th of the dominant wavelength thick with small sample data.展开更多
Sokor Formation reservoir intervals are intrinsically anisotropic,heterogeneous and with a characteristic of Low Contrast Low Resistivity(LCLR)log responses in parts of the Termit basin.Discriminating sands from shale...Sokor Formation reservoir intervals are intrinsically anisotropic,heterogeneous and with a characteristic of Low Contrast Low Resistivity(LCLR)log responses in parts of the Termit basin.Discriminating sands from shales/mudstones and hydrocarbon sands from brine sands as well as accurately evaluating the distribution of relevant reservoir properties using conventional seismic interpretation are complicated,and undermines reservoir characterization in such reservoirs.To enhance reservoir evaluation and reduce development planning and production risks,rock physics analysis was intergrated into the petrophysical workflow,which fed higher fidelity inputs into a post stack seismic inversion workflow.Rock Physics Diagnostics(RPD)analysis revealed that the reservoir interval of interest has grain size distribution of different lithologies,which is related to the environment of deposition and burial history,and could be best described by the constant cement sand model.The rock physics analyses revealed that facies were most effectively discriminated based on their Vp/Vs ratios and acoustic impedance.Particularly,hydrocarbon saturated sandstones,brine saturated shaly sandstones and shales/mudstones which exhibit similar acoustic impedance characteristics,were clearly discriminated by their Vp/Vs.The inverted seismic attributes as well as Seismic Based-Rock Physics Templates(RPT),clearly delineated the hydrocarbon fields,predicted new prospects beyond the existing well locations,which could be considered for field appraisal or development opportunities in the basin.These results demonstrate the value of the robust application of rock physics diagnostic modeling and seismic inversion in quantitative reservoir characterization and may be quite useful in undrilled locations in the basins and fields with similar geology.展开更多
The brittleness index(BI)is crucial for predicting engineering sweet spots and designing fracturing operations in shale oil reservoir exploration and development.Seismic amplitude variation with offset(AVO)inversion i...The brittleness index(BI)is crucial for predicting engineering sweet spots and designing fracturing operations in shale oil reservoir exploration and development.Seismic amplitude variation with offset(AVO)inversion is commonly used to obtain the BI.Traditionally,velocity,density,and other parameters are firstly inverted,and the BI is then calculated,which often leads to accumulated errors.Moreover,due to the limited of well-log data in field work areas,AVO inversion typically faces the challenge of limited information,resulting in not high accuracy of BI derived by existing AVO inversion methods.To address these issues,we first derive an AVO forward approximation equation that directly characterizes the BI in P-wave reflection coefficients.Based on this,an intelligent AVO inversion method,which combines the advantages of traditional and intelligent approaches,for directly obtaining the BI is proposed.A TransUnet model is constructed to establish the strong nonlinear mapping relationship between seismic data and the BI.By incorporating a combined objective function that is constrained by both low-frequency parameters and training samples,the challenge of limited samples is effectively addressed,and the direct inversion of the BI is stably achieved.Tests on model data and applications on field data demonstrate the feasibility,advancement,and practicality of the proposed method.展开更多
Underground fractures play an important role in the storage and movement of hydrocarbon fluid. Fracture rock physics has been the useful bridge between fracture parameters and seismic response. In this paper, we aim t...Underground fractures play an important role in the storage and movement of hydrocarbon fluid. Fracture rock physics has been the useful bridge between fracture parameters and seismic response. In this paper, we aim to use seismic data to predict subsurface fractures based on rock physics. We begin with the construction of fracture rock physics model. Using the model, we may estimate P-wave velocity, S-wave velocity and fracture rock physics parameters. Then we derive a new approximate formula for the analysis of the relationship between fracture rock physics parameters and seismic response, and we also propose the method which uses seismic data to invert the elastic and rock physics parameters of fractured rock. We end with the method verification, which includes using well-logging data to confirm the reliability of fracture rock physics effective model and utilizing real seismic data to validate the applicability of the inversion method. Tests show that the fracture rock physics effective model may be used to estimate velocities and fracture rock physics parameters reliably, and the inversion method is resultful even when the seismic data is added with random noise. Real data test also indicates the inversion method can be applied into the estimation of the elastic and fracture weaknesses parameters in the target area.展开更多
Deep learning has achieved great success in a variety of research fields and industrial applications.However,when applied to seismic inversion,the shortage of labeled data severely influences the performance of deep l...Deep learning has achieved great success in a variety of research fields and industrial applications.However,when applied to seismic inversion,the shortage of labeled data severely influences the performance of deep learning-based methods.In order to tackle this problem,we propose a novel seismic impedance inversion method based on a cycle-consistent generative adversarial network(Cycle-GAN).The proposed Cycle-GAN model includes two generative subnets and two discriminative subnets.Three kinds of loss,including cycle-consistent loss,adversarial loss,and estimation loss,are adopted to guide the training process.Benefit from the proposed structure,the information contained in unlabeled data can be extracted,and adversarial learning further guarantees that the prediction results share similar distributions with the real data.Moreover,a neural network visualization method is adopted to show that the proposed CNN model can learn more distinguishable features than the conventional CNN model.The robustness experiments on synthetic data sets show that the proposed method can achieve better performances than other methods in most cases.And the blind-well experiments on real seismic profiles show that the predicted impedance curve of the proposed method maintains a better correlation with the true impedance curve.展开更多
Although the ambiguity of seismic inversion is widely recognized in both theory and practice, so far as a concrete inversion example is concerned, there is not any objective, controllable method or any standard for ho...Although the ambiguity of seismic inversion is widely recognized in both theory and practice, so far as a concrete inversion example is concerned, there is not any objective, controllable method or any standard for how to evaluate and determine its ambiguity and reliability, especially for the high frequency components beyond the effective seismic frequency band. Taking log-constrained impedance inversion as an example, a new appraisal method is proposed on the basis of analyzing a simple geological model. Firstly, the inverted impedance model is transformed to a reflection coefficient series. Secondly, the maximum effective frequency of the real seismic data is chosen as a cutoff point and the reflection coefficient series is decomposed into two components by low-pass and high-pass filters. Thirdly, the geometrical reflection characteristics of the high-frequency components and that of the real seismic data are compared and analyzed. Then, the reliability of the inverted impedance model is appraised according to the similarity of geometrical characteristics between the high-frequency components and the real seismic data. The new method avoids some subjectivity in appraising the inverted result, and helps to enhance the reliability of reservoir prediction by impedance inversion technology.展开更多
Deep learning is widely used for seismic impedance inversion,but few work provides in-depth research and analysis on designing the architectures of deep neural networks and choosing the network hyperparameters.This pa...Deep learning is widely used for seismic impedance inversion,but few work provides in-depth research and analysis on designing the architectures of deep neural networks and choosing the network hyperparameters.This paper is dedicated to comprehensively studying on the significant aspects of deep neural networks that affect the inversion results.We experimentally reveal how network hyperparameters and architectures affect the inversion performance,and develop a series of methods which are proven to be effective in reconstructing high-frequency information in the estimated impedance model.Experiments demonstrate that the proposed multi-scale architecture is helpful to reconstruct more high-frequency details than a conventional network.Besides,the reconstruction of high-frequency information can be further promoted by introducing a perceptual loss and a generative adversarial network from the computer vision perspective.More importantly,the experimental results provide valuable references for designing proper network architectures in the seismic inversion problem.展开更多
Under the condition of thin interbeds with great lateral changes in terrestrial basins,a seismic meme inversion method is established based on the analysis of seismic sedimentology technology.The relationship between ...Under the condition of thin interbeds with great lateral changes in terrestrial basins,a seismic meme inversion method is established based on the analysis of seismic sedimentology technology.The relationship between seismic waveform and high-frequency well logs is established through dynamic clustering of seismic waveform to improve the vertical and horizontal resolution of inversion results;meanwhile,by constructing the Bayesian inversion framework of different seismic facies,the real facies controlled inversion is realized.The forward model verification results show that the seismic meme inversion can realize precise prediction of 3 m thick thin interbeds,proving the rationality and high precision of the method.The application in the Daqing placanticline shows that the seismic meme inversion could identify 2 m thin interbeds,and the coincidence rates of inversion results and drilling data were more than 80%.The seismic meme inversion method can improve the accuracy of reservoir prediction and provides a useful mean for thin interbeds prediction in terrestrial basins.展开更多
基金supported by National Key R&D Program of China(2018YFA0702501)National Natural Science Foundation of China (41974140)+1 种基金Science and Technology Management Department,China National Petroleum Corporation(2022DQ0604-01)China National Petroleum Corporation-China University of Petroleum (Beijing) Strategy。
文摘Traditional deep learning methods pursue complex and single network architectures without considering the petrophysical relationship between different elastic parameters.The mathematical and statistical significance of the inversion results may lead to model overfitting,especially when there are a limited number of well logs in a working area.Multitask learning provides an eff ective approach to addressing this issue.Simultaneously,learning multiple related tasks can improve a model’s generalization ability to a certain extent,thereby enhancing the performance of related tasks with an equal amount of labeled data.In this study,we propose an end-to-end multitask deep learning model that integrates a fully convolutional network and bidirectional gated recurrent unit for intelligent prestack inversion of“seismic data to elastic parameters.”The use of a Bayesian homoscedastic uncertainty-based loss function enables adaptive learning of the weight coeffi cients for diff erent elastic parameter inversion tasks,thereby reducing uncertainty during the inversion process.The proposed method combines the local feature perception of convolutional neural networks with the long-term memory of bidirectional gated recurrent networks.It maintains the rock physics constraint relationships among diff erent elastic parameters during the inversion process,demonstrating a high level of prediction accuracy.Numerical simulations and processing results of real seismic data validate the eff ectiveness and practicality of the proposed method.
基金supported by the National Major Science and Technology Project of China on Development of Big Oil-Gas Fields and Coalbed Methane (No. 2008ZX05010-002)
文摘Stochastic seismic inversion is the combination of geostatistics and seismic inversion technology which integrates information from seismic records, well logs, and geostatistics into a posterior probability density function （PDF） of subsurface models. The Markov chain Monte Carlo （MCMC） method is used to sample the posterior PDF and the subsurface model characteristics can be inferred by analyzing a set of the posterior PDF samples. In this paper, we first introduce the stochastic seismic inversion theory, discuss and analyze the four key parameters： seismic data signal-to-noise ratio （S/N）, variogram, the posterior PDF sample number, and well density, and propose the optimum selection of these parameters. The analysis results show that seismic data S/N adjusts the compromise between the influence of the seismic data and geostatistics on the inversion results, the variogram controls the smoothness of the inversion results, the posterior PDF sample number determines the reliability of the statistical characteristics derived from the samples, and well density influences the inversion uncertainty. Finally, the comparison between the stochastic seismic inversion and the deterministic model based seismic inversion indicates that the stochastic seismic inversion can provide more reliable information of the subsurface character.
基金Projects 40574057 and 40874054 supported by the National Natural Science Foundation of ChinaProjects 2007CB209400 by the National Basic Research Program of ChinaFoundation of China University of Mining and Technology (OF4471)
文摘Non-liner wave equation inversion,wavelet analysis and artificial neural networks were used to obtain stratum parameters and the distribution of thin coal seams.The lithology of the water-bearing/resisting layer in the Quaternary system was also predicted.The implementation process included calculating the well log parameters,stratum contrasting the seismic data and the well logs,and extracting,studying and predicting seismic attributes.Seismic inversion parameters,including the layer velocity and wave impedance,were calculated and effectively used for prediction and analysis.Prior knowledge and seismic interpretation were used to remedy a dearth of seismic data during the inversion procedure.This enhanced the stability of the inversion method.Non-linear seismic inversion and artificial neural networks were used to interpret coal seismic lithology and to study the water-bearing/resisting layer in the Quaternary system.Interpretation of the 1～2 m thin coal seams,and also of the water-bearing/resisting layer in the Quaternary system,is provided.The upper mining limit can be lifted from 60 m to 45 m.The predictions show that this method can provide reliable data useful for thin coal seam exploitation and for lifting the upper mining limit,which is one of the principles of green mining.
基金This research is supported by the Joint Funds of the National Natural Science Foundation of China(No.U20B2016)the National Natural Science Foundation of China(No.41874167)the National Natural Science Foundation of China(No.41904130).
文摘To improve the accuracy of inversion results,geological facies distributions are considered as additional constraints in the inversion process.However,the geological facies itself also has its own uncertainty.In this paper,the initial sedimentary facies maps are obtained by integrated geological analysis from well data,seismic attributes,and deterministic inversion results.Then the fi rst iteration of facies-constrained seismic inversion is performed.According to that result and other data such as geological information,the facies distribution can be updated using cluster analysis.The next round of facies-constrained inversion can then be performed.This process will be repeated until the facies inconsistency or error before and after the inversion is minimized.It forms a new iterative facies-constrained seismic inversion technique.Compared with conventional facies-constrained seismic inversion,the proposed method not only can reduces the non-uniqueness of seismic inversion results but also can improves its resolution.As a consequence,the sedimentary facies will be more consistent with the geology.A practical application demonstrated that the superposition relationship of sand bodies could be better delineated based on this new seismic inversion technique.The result highly increases the understanding of reservoir connectivity and its accuracy,which can be used to guide further development.
基金Supported by the National Basic Research Program of China （2009CB219603, 2010CB226800） the National Natural Science Foundation of China （40874071, 40672104）
文摘During the process of coal prospecting and exploration, different measurement time, different logging instruments and series can lead to systematic errors in well logs. Accordingly, all logging curves need to be normalized in the mining area. By studying well-logging normalization methods, and focusing on the characteristics of the coalfield, the frequency histogram method was used in accordance with the condition of the Guqiao Coal Mine. In this way, the density and sonic velocity at marker bed in the non-key well were made to close to those in the key well, and were eventually equal. Well log normalization was completed when this method was applied to the entire logging curves. The results show that the scales of logging data were unified by normalizing coal logging curves, and the logging data were consistent with wave impedance inversion data. A satisfactory inversion effect was obtained.
基金partially supported by the National Natural Science Foundation of China (No.41230318)
文摘With the development of computational power, there has been an increased focus on data-fitting related seismic inversion techniques for high fidelity seismic velocity model and image, such as full-waveform inversion and least squares migration. However, though more advanced than conventional methods, these data fitting methods can be very expensive in terms of computational cost. Recently, various techniques to optimize these data-fitting seismic inversion problems have been implemented to cater for the industrial need for much improved efficiency. In this study, we propose a general stochastic conjugate gradient method for these data-fitting related inverse problems. We first prescribe the basic theory of our method and then give synthetic examples. Our numerical experiments illustrate the potential of this method for large-size seismic inversion application.
基金provided by the National Science and Technology Major Project(No.2011ZX05004-004)China National Petroleum Corporation Key Projects(No.2014E2105)the Priority Academic Program Development of Jiangsu Higher Education Institutions(PAPD)
文摘In order to improve the fine structure inversion ability of igneous rocks for the exploration of underlying strata, based on particle swarm optimization(PSO), we have developed a method for seismic wave impedance inversion. Through numerical simulation, we tested the effects of different algorithm parameters and different model parameterization methods on PSO wave impedance inversion, and analyzed the characteristics of PSO method. Under the conclusions drawn from numerical simulation, we propose the scheme of combining a cross-moving strategy based on a divided block model and high-frequency filtering technology for PSO inversion. By analyzing the inversion results of a wedge model of a pitchout coal seam and a coal coking model with igneous rock intrusion, we discuss the vertical and horizontal resolution, stability and reliability of PSO inversion. Based on the actual seismic and logging data from an igneous area, by taking a seismic profile through wells as an example, we discuss the characteristics of three inversion methods, including model-based wave impedance inversion, multi-attribute seismic inversion based on probabilistic neural network(PNN) and wave impedance inversion based on PSO.And we draw the conclusion that the inversion based on PSO method has a better result for this igneous area.
文摘In this paper seismic inversion was used as a key technique and the seismic wavelet most suitable to the actual underground situation was extracted with the higher-order statistics algorithm. The wavelets extracted in this way and the wavelets extracted with the seismic statistics techniques were used separately for inverting the seismic data of the southern part of Tahe oilfield, Tarim basin. The results showed that the resolution of the wavelet inversion with the higher-order statistics method was greatly improved, and the wavelet-inverted section could better distinguish the thin sandstone reservoirs of the upper and lower Carboniferous and their lateral distribution, providing a reliable basis of analysis for the study of thin sandstone reservoirs.
文摘Geoscientific evidence shows that various parameters such as compaction,buoyancy effect,hydrocarbon maturation,gas effect and tectonic activities control the pore pressure of sub-surface geology.Spatially controlled geoscientific data in the tectonically active areas is significantly useful for robust estimation of pre-drill pore pressure.The reservoir which is tectonically complex and pore pressure is changing frequently that circumference motivated us to conduct this study.The changes in pore pressure have been captured from the fine-scale to the broad scale in the Jaisalmer sub-basin.Pore pressure variation has been distinctly observed in pre-and post-Jurassic age based on the current study.Post-stack seismic inversion study was conducted to capturing the variation of pore pressure.Analysis of low-frequency spectrum and integrated interval velocity model provided a detailed feature of pore pressure in each compartment of the study area.Pore pressure estimated from well log data was correlated with seismic inversion based result.Based on the current study one well has been proposed where pore pressure was estimated and two distinguished trends are identified in the study zone.The approaches of the current study were analysed thoroughly and it will be highly useful in complex reservoir condition where pore pressure varies frequently.
文摘The objective of this work is to implement a pseudo-forward equation which is called PFE to transform data （similarity attribute） to model parameters （porosity） in a gas reservoir in the F3 block of North Sea. This equation which is an experimental model has unknown constants in its structure; hence, a least square solution is applied to find the best constants. The results derived from solved equa- tions show that the errors on measured data are mapped into the errors of estimated constants; hence, Tikhonov regularization is used to improve the estimated parameters. The results are compared with a conventional method such as cross plotting between acoustic impedance and porosity values to validate the PFE model. When the testing dataset in sand units was used, the correlation coefficient between two variables （actual and predicted values） was obtained as 0.720 and 0.476 for PFE model and cross-plotting analysis, respectively. Therefore, the testing dataset validates rela- tively well the PFE optimized by Tikhonov regularization in sand units of a gas reservoir. The obtained results indi- cate that PFE could provide initial information about sandstone reservoirs. It could estimate reservoir porosity distribution approximately and it highlights bright spots and fault structures such as gas chimneys and salt edges.
基金Wupport from the National Natural Science Foundation of China(Grant No.42130812,42174151,and 41804126).
文摘We propose to use a Few-Shot Learning(FSL)method for the pre-stack seismic inversion problem in obtaining a high resolution reservoir model from recorded seismic data.Recently,artificial neural network(ANN)demonstrates great advantages for seismic inversion because of its powerful feature extraction and parameter learning ability.Hence,ANN method could provide a high resolution inversion result that are critical for reservoir characterization.However,the ANN approach requires plenty of labeled samples for training in order to obtain a satisfactory result.For the common problem of scarce samples in the ANN seismic inversion,we create a novel pre-stack seismic inversion method that takes advantage of the FSL.The results of conventional inversion are used as the auxiliary dataset for ANN based on FSL,while the well log is regarded the scarce training dataset.According to the characteristics of seismic inversion(large amount and high dimensional),we construct an arch network(A-Net)architecture to implement this method.An example shows that this method can improve the accuracy and resolution of inversion results.
文摘Seismic inversion and basic theory are briefly presented and the main idea of this method is introduced. Both non-linear wave equation inversion technique and Complete Utilization of Samples Information (CUSI) neural network analysis are used in lithological interpretation in Jibei coal field. The prediction results indicate that this method can provide reliable data for thin coal exploitation and promising area evaluation.
文摘As sandstone layers in thin interbedded section are difficult to identify,conventional model-driven seismic inversion and data-driven seismic prediction methods have low precision in predicting them.To solve this problem,a model-data-driven seismic AVO(amplitude variation with offset)inversion method based on a space-variant objective function has been worked out.In this method,zero delay cross-correlation function and F norm are used to establish objective function.Based on inverse distance weighting theory,change of the objective function is controlled according to the location of the target CDP(common depth point),to change the constraint weights of training samples,initial low-frequency models,and seismic data on the inversion.Hence,the proposed method can get high resolution and high-accuracy velocity and density from inversion of small sample data,and is suitable for identifying thin interbedded sand bodies.Tests with thin interbedded geological models show that the proposed method has high inversion accuracy and resolution for small sample data,and can identify sandstone and mudstone layers of about one-30th of the dominant wavelength thick.Tests on the field data of Lishui sag show that the inversion results of the proposed method have small relative error with well-log data,and can identify thin interbedded sandstone layers of about one-15th of the dominant wavelength thick with small sample data.
文摘Sokor Formation reservoir intervals are intrinsically anisotropic,heterogeneous and with a characteristic of Low Contrast Low Resistivity(LCLR)log responses in parts of the Termit basin.Discriminating sands from shales/mudstones and hydrocarbon sands from brine sands as well as accurately evaluating the distribution of relevant reservoir properties using conventional seismic interpretation are complicated,and undermines reservoir characterization in such reservoirs.To enhance reservoir evaluation and reduce development planning and production risks,rock physics analysis was intergrated into the petrophysical workflow,which fed higher fidelity inputs into a post stack seismic inversion workflow.Rock Physics Diagnostics(RPD)analysis revealed that the reservoir interval of interest has grain size distribution of different lithologies,which is related to the environment of deposition and burial history,and could be best described by the constant cement sand model.The rock physics analyses revealed that facies were most effectively discriminated based on their Vp/Vs ratios and acoustic impedance.Particularly,hydrocarbon saturated sandstones,brine saturated shaly sandstones and shales/mudstones which exhibit similar acoustic impedance characteristics,were clearly discriminated by their Vp/Vs.The inverted seismic attributes as well as Seismic Based-Rock Physics Templates(RPT),clearly delineated the hydrocarbon fields,predicted new prospects beyond the existing well locations,which could be considered for field appraisal or development opportunities in the basin.These results demonstrate the value of the robust application of rock physics diagnostic modeling and seismic inversion in quantitative reservoir characterization and may be quite useful in undrilled locations in the basins and fields with similar geology.
基金supposed by the National Nature Science Foundation of China(Grant No.42304131)the Natural Science Foundation of Heilongjiang Province(Grant No.LH2023D012)+1 种基金the Heilongjiang Postdoctoral Fund(Grant No.LBH-Z22092)the Basic Research Fund for Universities in Xinjiang Uygur Autonomous Region(Grant No.XJEDU2023P166)。
文摘The brittleness index(BI)is crucial for predicting engineering sweet spots and designing fracturing operations in shale oil reservoir exploration and development.Seismic amplitude variation with offset(AVO)inversion is commonly used to obtain the BI.Traditionally,velocity,density,and other parameters are firstly inverted,and the BI is then calculated,which often leads to accumulated errors.Moreover,due to the limited of well-log data in field work areas,AVO inversion typically faces the challenge of limited information,resulting in not high accuracy of BI derived by existing AVO inversion methods.To address these issues,we first derive an AVO forward approximation equation that directly characterizes the BI in P-wave reflection coefficients.Based on this,an intelligent AVO inversion method,which combines the advantages of traditional and intelligent approaches,for directly obtaining the BI is proposed.A TransUnet model is constructed to establish the strong nonlinear mapping relationship between seismic data and the BI.By incorporating a combined objective function that is constrained by both low-frequency parameters and training samples,the challenge of limited samples is effectively addressed,and the direct inversion of the BI is stably achieved.Tests on model data and applications on field data demonstrate the feasibility,advancement,and practicality of the proposed method.
基金supported by the National Basic Research Program of China(Grant Nos.2013CB228604,2014CB239201)the National Oil and Gas Major Projects of China(Grant No.2011ZX05014-001-010HZ)+2 种基金CNPC Innovation Foundation(Grant No.2011D-5006-0301)the Fundamental Research Funds for the Central Universities in China(Grant No.14CX06015A)SINOPEC Key Laboratory of Geophysics
文摘Underground fractures play an important role in the storage and movement of hydrocarbon fluid. Fracture rock physics has been the useful bridge between fracture parameters and seismic response. In this paper, we aim to use seismic data to predict subsurface fractures based on rock physics. We begin with the construction of fracture rock physics model. Using the model, we may estimate P-wave velocity, S-wave velocity and fracture rock physics parameters. Then we derive a new approximate formula for the analysis of the relationship between fracture rock physics parameters and seismic response, and we also propose the method which uses seismic data to invert the elastic and rock physics parameters of fractured rock. We end with the method verification, which includes using well-logging data to confirm the reliability of fracture rock physics effective model and utilizing real seismic data to validate the applicability of the inversion method. Tests show that the fracture rock physics effective model may be used to estimate velocities and fracture rock physics parameters reliably, and the inversion method is resultful even when the seismic data is added with random noise. Real data test also indicates the inversion method can be applied into the estimation of the elastic and fracture weaknesses parameters in the target area.
基金financially supported by the NSFC(Grant No.41974126 and 41674116)the National Key Research and Development Program of China(Grant No.2018YFA0702501)the 13th 5-Year Basic Research Program of China National Petroleum Corporation(CNPC)(2018A-3306)。
文摘Deep learning has achieved great success in a variety of research fields and industrial applications.However,when applied to seismic inversion,the shortage of labeled data severely influences the performance of deep learning-based methods.In order to tackle this problem,we propose a novel seismic impedance inversion method based on a cycle-consistent generative adversarial network(Cycle-GAN).The proposed Cycle-GAN model includes two generative subnets and two discriminative subnets.Three kinds of loss,including cycle-consistent loss,adversarial loss,and estimation loss,are adopted to guide the training process.Benefit from the proposed structure,the information contained in unlabeled data can be extracted,and adversarial learning further guarantees that the prediction results share similar distributions with the real data.Moreover,a neural network visualization method is adopted to show that the proposed CNN model can learn more distinguishable features than the conventional CNN model.The robustness experiments on synthetic data sets show that the proposed method can achieve better performances than other methods in most cases.And the blind-well experiments on real seismic profiles show that the predicted impedance curve of the proposed method maintains a better correlation with the true impedance curve.
基金supported by the Major Basic Research Development Program of China’s 973 Project(grant No.2007CB209608)the Science and Technology Innovation Foundation of CNPC(grant No.2010D-5006-0301)
文摘Although the ambiguity of seismic inversion is widely recognized in both theory and practice, so far as a concrete inversion example is concerned, there is not any objective, controllable method or any standard for how to evaluate and determine its ambiguity and reliability, especially for the high frequency components beyond the effective seismic frequency band. Taking log-constrained impedance inversion as an example, a new appraisal method is proposed on the basis of analyzing a simple geological model. Firstly, the inverted impedance model is transformed to a reflection coefficient series. Secondly, the maximum effective frequency of the real seismic data is chosen as a cutoff point and the reflection coefficient series is decomposed into two components by low-pass and high-pass filters. Thirdly, the geometrical reflection characteristics of the high-frequency components and that of the real seismic data are compared and analyzed. Then, the reliability of the inverted impedance model is appraised according to the similarity of geometrical characteristics between the high-frequency components and the real seismic data. The new method avoids some subjectivity in appraising the inverted result, and helps to enhance the reliability of reservoir prediction by impedance inversion technology.
基金supported by the National Natural Science Foundation of China under Grant No.42050104
文摘Deep learning is widely used for seismic impedance inversion,but few work provides in-depth research and analysis on designing the architectures of deep neural networks and choosing the network hyperparameters.This paper is dedicated to comprehensively studying on the significant aspects of deep neural networks that affect the inversion results.We experimentally reveal how network hyperparameters and architectures affect the inversion performance,and develop a series of methods which are proven to be effective in reconstructing high-frequency information in the estimated impedance model.Experiments demonstrate that the proposed multi-scale architecture is helpful to reconstruct more high-frequency details than a conventional network.Besides,the reconstruction of high-frequency information can be further promoted by introducing a perceptual loss and a generative adversarial network from the computer vision perspective.More importantly,the experimental results provide valuable references for designing proper network architectures in the seismic inversion problem.
文摘Under the condition of thin interbeds with great lateral changes in terrestrial basins,a seismic meme inversion method is established based on the analysis of seismic sedimentology technology.The relationship between seismic waveform and high-frequency well logs is established through dynamic clustering of seismic waveform to improve the vertical and horizontal resolution of inversion results;meanwhile,by constructing the Bayesian inversion framework of different seismic facies,the real facies controlled inversion is realized.The forward model verification results show that the seismic meme inversion can realize precise prediction of 3 m thick thin interbeds,proving the rationality and high precision of the method.The application in the Daqing placanticline shows that the seismic meme inversion could identify 2 m thin interbeds,and the coincidence rates of inversion results and drilling data were more than 80%.The seismic meme inversion method can improve the accuracy of reservoir prediction and provides a useful mean for thin interbeds prediction in terrestrial basins.
