Accurate estimation of mineralogy from geophysical well logs is crucial for characterizing geological formations,particularly in hydrocarbon exploration,CO_(2) sequestration,and geothermal energy development.Current t...Accurate estimation of mineralogy from geophysical well logs is crucial for characterizing geological formations,particularly in hydrocarbon exploration,CO_(2) sequestration,and geothermal energy development.Current techniques,such as multimineral petrophysical analysis,offer details into mineralogical distribution.However,it is inherently time-intensive and demands substantial geological expertise for accurate model evaluation.Furthermore,traditional machine learning techniques often struggle to predict mineralogy accurately and sometimes produce estimations that violate fundamental physical principles.To address this,we present a new approach using Physics-Integrated Neural Networks(PINNs),that combines data-driven learning with domain-specific physical constraints,embedding petrophysical relationships directly into the neural network architecture.This approach enforces that predictions adhere to physical laws.The methodology is applied to the Broom Creek Deep Saline aquifer,a CO_(2) sequestration site in the Williston Basin,to predict the volumes of key mineral constituents—quartz,dolomite,feldspar,anhydrite,illite—along with porosity.Compared to traditional artificial neural networks (ANN),the PINN approach demonstrates higher accuracy and better generalizability,significantly enhancing predictive performance on unseen well datasets.The average mean error across the three blind wells is 0.123 for ANN and 0.042 for PINN,highlighting the superior accuracy of the PINN approach.This method reduces uncertainties in reservoir characterization by improving the reliability of mineralogy and porosity predictions,providing a more robust tool for decision-making in various subsurface geoscience applications.展开更多
Pharmaceutical pollution is becoming an increasing threat to aquatic environments since inactive compounds do not break down,and the drug products are accumulated in living organisms.The ability of a drug to dissolve ...Pharmaceutical pollution is becoming an increasing threat to aquatic environments since inactive compounds do not break down,and the drug products are accumulated in living organisms.The ability of a drug to dissolve in water(i.e.,LogS)is an important parameter for assessing a drug’s environmental fate,biovailability,and toxicity.LogS is typically measured in a laboratory setting,which can be costly and time-consuming,and does not provide the opportunity to conduct large-scale analyses.This research develops and evaluates machine learning models that can produce LogS estimates and may improve the environmental risk assessments of toxic pharmaceutical pollutants.We used a dataset from the ChEMBL database that contained 8832 molecular compounds.Various data preprocessing and cleaning techniques were applied(i.e.,removing the missing values),we then recorded chemical properties by normalizing and,even,using some feature selection techniques.We evaluated logS with a total of several machine learning and deep learning models,including;linear regression,random forests(RF),support vector machines(SVM),gradient boosting(GBM),and artificial neural networks(ANNs).We assessed model performance using a series of metrics,including root mean square error(RMSE)and mean absolute error(MAE),as well as the coefficient of determination(R^(2)).The findings show that the Least Angle Regression(LAR)model performed the best with an R^(2) value close to 1.0000,confirming high predictive accuracy.The OMP model performed well with good accuracy(R^(2)=0.8727)while remaining computationally cheap,while other models(e.g.,neural networks,random forests)performed well but were too computationally expensive.Finally,to assess the robustness of the results,an error analysis indicated that residuals were evenly distributed around zero,confirming the results from the LAR model.The current research illustrates the potential of AI in anticipating drug solubility,providing support for green pharmaceutical design and environmental risk assessment.Future work should extend predictions to include degradation and toxicity to enhance predictive power and applicability.展开更多
Pore pressure is a decisive measure to assess the reservoir’s geomechanical properties,ensures safe and efficient drilling operations,and optimizes reservoir characterization and production.The conventional approache...Pore pressure is a decisive measure to assess the reservoir’s geomechanical properties,ensures safe and efficient drilling operations,and optimizes reservoir characterization and production.The conventional approaches sometimes fail to comprehend complex and persistent relationships between pore pressure and formation properties in the heterogeneous reservoirs.This study presents a novel machine learning optimized pore pressure prediction method with a limited dataset,particularly in complex formations.The method addresses the conventional approach's limitations by leveraging its capability to learn complex data relationships.It integrates the best Gradient Boosting Regressor(GBR)algorithm to model pore pressure at wells and later utilizes ContinuousWavelet Transformation(CWT)of the seismic dataset for spatial analysis,and finally employs Deep Neural Network for robust and precise pore pressure modeling for the whole volume.In the second stage,for the spatial variations of pore pressure in the thin Khadro Formation sand reservoir across the entire subsurface area,a three-dimensional pore pressure prediction is conducted using CWT.The relationship between the CWT and geomechanical properties is then established through supervised machine learning models on well locations to predict the uncertainties in pore pressure.Among all intelligent regression techniques developed using petrophysical and elastic properties for pore pressure prediction,the GBR has provided exceptional results that have been validated by evaluation metrics based on the R^(2) score i.e.,0.91 between the calibrated and predicted pore pressure.Via the deep neural network,the relationship between CWT resultant traces and predicted pore pressure is established to analyze the spatial variation.展开更多
Karst fractures serve as crucial seepage channels and storage spaces for carbonate natural gas reservoirs,and electrical image logs are vital data for visualizing and characterizing such fractures.However,the conventi...Karst fractures serve as crucial seepage channels and storage spaces for carbonate natural gas reservoirs,and electrical image logs are vital data for visualizing and characterizing such fractures.However,the conventional approach of identifying fractures using electrical image logs predominantly relies on manual processes that are not only time-consuming but also highly subjective.In addition,the heterogeneity and strong dissolution tendency of karst carbonate reservoirs lead to complexity and variety in fracture geometry,which makes it difficult to accurately identify fractures.In this paper,the electrical image logs network(EILnet)da deep-learning-based intelligent semantic segmentation model with a selective attention mechanism and selective feature fusion moduledwas created to enable the intelligent identification and segmentation of different types of fractures through electrical logging images.Data from electrical image logs representing structural and induced fractures were first selected using the sliding window technique before image inpainting and data augmentation were implemented for these images to improve the generalizability of the model.Various image-processing tools,including the bilateral filter,Laplace operator,and Gaussian low-pass filter,were also applied to the electrical logging images to generate a multi-attribute dataset to help the model learn the semantic features of the fractures.The results demonstrated that the EILnet model outperforms mainstream deep-learning semantic segmentation models,such as Fully Convolutional Networks(FCN-8s),U-Net,and SegNet,for both the single-channel dataset and the multi-attribute dataset.The EILnet provided significant advantages for the single-channel dataset,and its mean intersection over union(MIoU)and pixel accuracy(PA)were 81.32%and 89.37%,respectively.In the case of the multi-attribute dataset,the identification capability of all models improved to varying degrees,with the EILnet achieving the highest MIoU and PA of 83.43%and 91.11%,respectively.Further,applying the EILnet model to various blind wells demonstrated its ability to provide reliable fracture identification,thereby indicating its promising potential applications.展开更多
We propose an integrated method of data-driven and mechanism models for well logging formation evaluation,explicitly focusing on predicting reservoir parameters,such as porosity and water saturation.Accurately interpr...We propose an integrated method of data-driven and mechanism models for well logging formation evaluation,explicitly focusing on predicting reservoir parameters,such as porosity and water saturation.Accurately interpreting these parameters is crucial for effectively exploring and developing oil and gas.However,with the increasing complexity of geological conditions in this industry,there is a growing demand for improved accuracy in reservoir parameter prediction,leading to higher costs associated with manual interpretation.The conventional logging interpretation methods rely on empirical relationships between logging data and reservoir parameters,which suffer from low interpretation efficiency,intense subjectivity,and suitability for ideal conditions.The application of artificial intelligence in the interpretation of logging data provides a new solution to the problems existing in traditional methods.It is expected to improve the accuracy and efficiency of the interpretation.If large and high-quality datasets exist,data-driven models can reveal relationships of arbitrary complexity.Nevertheless,constructing sufficiently large logging datasets with reliable labels remains challenging,making it difficult to apply data-driven models effectively in logging data interpretation.Furthermore,data-driven models often act as“black boxes”without explaining their predictions or ensuring compliance with primary physical constraints.This paper proposes a machine learning method with strong physical constraints by integrating mechanism and data-driven models.Prior knowledge of logging data interpretation is embedded into machine learning regarding network structure,loss function,and optimization algorithm.We employ the Physically Informed Auto-Encoder(PIAE)to predict porosity and water saturation,which can be trained without labeled reservoir parameters using self-supervised learning techniques.This approach effectively achieves automated interpretation and facilitates generalization across diverse datasets.展开更多
Machine learning algorithms are widely used to interpret well logging data.To enhance the algorithms'robustness,shuffling the well logging data is an unavoidable feature engineering before training models.However,...Machine learning algorithms are widely used to interpret well logging data.To enhance the algorithms'robustness,shuffling the well logging data is an unavoidable feature engineering before training models.However,latent information stored between different well logging types and depth is destroyed during the shuffle.To investigate the influence of latent information,this study implements graph convolution networks(GCNs),long-short temporal memory models,recurrent neural networks,temporal convolution networks,and two artificial neural networks to predict the microbial lithology in the fourth member of the Dengying Formation,Moxi gas field,central Sichuan Basin.Results indicate that the GCN model outperforms other models.The accuracy,F1-score,and area under curve of the GCN model are 0.90,0.90,and 0.95,respectively.Experimental results indicate that the time-series data facilitates lithology prediction and helps determine lithological fluctuations in the vertical direction.All types of logs from the spectral in the GCN model and also facilitates lithology identification.Only on condition combined with latent information,the GCN model reaches excellent microbialite classification resolution at the centimeter scale.Ultimately,the two actual cases show tricks for using GCN models to predict potential microbialite in other formations and areas,proving that the GCN model can be adopted in the industry.展开更多
We propose a novel workflow for fast forward modeling of well logs in axially symmetric 2D models of the nearwellbore environment.The approach integrates the finite element method with deep residual neural networks to...We propose a novel workflow for fast forward modeling of well logs in axially symmetric 2D models of the nearwellbore environment.The approach integrates the finite element method with deep residual neural networks to achieve exceptional computational efficiency and accuracy.The workflow is demonstrated through the modeling of wireline electromagnetic propagation resistivity logs,where the measured responses exhibit a highly nonlinear relationship with formation properties.The motivation for this research is the need for advanced modeling al-gorithms that are fast enough for use in modern quantitative interpretation tools,where thousands of simulations may be required in iterative inversion processes.The proposed algorithm achieves a remarkable enhancement in performance,being up to 3000 times faster than the finite element method alone when utilizing a GPU.While still ensuring high accuracy,this makes it well-suited for practical applications when reliable payzone assessment is needed in complex environmental scenarios.Furthermore,the algorithm’s efficiency positions it as a promising tool for stochastic Bayesian inversion,facilitating reliable uncertainty quantification in subsurface property estimation.展开更多
Distinguishing between web traffic generated by bots and humans is an important task in the evaluation of online marketing campaigns.One of the main challenges is related to only partial availability of the performanc...Distinguishing between web traffic generated by bots and humans is an important task in the evaluation of online marketing campaigns.One of the main challenges is related to only partial availability of the performance metrics:although some users can be unambiguously classified as bots,the correct label is uncertain in many cases.This calls for the use of classifiers capable of explaining their decisions.This paper demonstrates two such mechanisms based on features carefully engineered from web logs.The first is a man-made rule-based system.The second is a hierarchical model that first performs clustering and next classification using human-centred,interpretable methods.The stability of the proposed methods is analyzed and a minimal set of features that convey the classdiscriminating information is selected.The proposed data processing and analysis methodology are successfully applied to real-world data sets from online publishers.展开更多
Pore size analysis plays a pivotal role in unraveling reservoir behavior and its intricate relationship with confined fluids.Traditional methods for predicting pore size distribution(PSD),relying on drilling cores or ...Pore size analysis plays a pivotal role in unraveling reservoir behavior and its intricate relationship with confined fluids.Traditional methods for predicting pore size distribution(PSD),relying on drilling cores or thin sections,face limitations associated with depth specificity.In this study,we introduce an innovative framework that leverages nuclear magnetic resonance(NMR)log data,encompassing clay-bound water(CBW),bound volume irreducible(BVI),and free fluid volume(FFV),to determine three PSDs(micropores,mesopores,and macropores).Moreover,we establish a robust pore size classification(PSC)system utilizing ternary plots,derived from the PSDs.Within the three studied wells,NMR log data is exclusive to one well(well-A),while conventional well logs are accessible for all three wells(well-A,well-B,and well-C).This distinction enables PSD predictions for the remaining two wells(B and C).To prognosticate NMR outputs(CBW,BVI,FFV)for these wells,a two-step deep learning(DL)algorithm is implemented.Initially,three feature selection algorithms(f-classif,f-regression,and mutual-info-regression)identify the conventional well logs most correlated to NMR outputs in well-A.The three feature selection algorithms utilize statistical computations.These algorithms are utilized to systematically identify and optimize pertinent input features,thereby augmenting model interpretability and predictive efficacy within intricate data-driven endeavors.So,all three feature selection algorithms introduced the number of 4 logs as the most optimal number of inputs to the DL algorithm with different combinations of logs for each of the three desired outputs.Subsequently,the CUDA Deep Neural Network Long Short-Term Memory algorithm(CUDNNLSTM),belonging to the category of DL algorithms and harnessing the computational power of GPUs,is employed for the prediction of CBW,BVI,and FFV logs.This prediction leverages the optimal logs identified in the preceding step.Estimation of NMR outputs was done first in well-A(80%of data as training and 20%as testing).The correlation coefficient(CC)between the actual and estimated data for the three outputs CBW,BVI and FFV are 95%,94%,and 97%,respectively,as well as root mean square error(RMSE)was obtained 0.0081,0.098,and 0.0089,respectively.To assess the effectiveness of the proposed algorithm,we compared it with two traditional methods for log estimation:multiple regression and multi-resolution graph-based clustering methods.The results demonstrate the superior accuracy of our algorithm in comparison to these conventional approaches.This DL-driven approach facilitates PSD prediction grounded in fluid saturation for wells B and C.Ternary plots are then employed for PSCs.Seven distinct PSCs within well-A employing actual NMR logs(CBW,BVI,FFV),in conjunction with an equivalent count within wells B and C utilizing three predicted logs,are harmoniously categorized leading to the identification of seven distinct pore size classification facies(PSCF).this research introduces an advanced approach to pore size classification and prediction,fusing NMR logs with deep learning techniques and extending their application to nearby wells without NMR log.The resulting PSCFs offer valuable insights into generating precise and detailed reservoir 3D models.展开更多
Reef-bank reservoirs are an important target for petroleum exploration in marine carbonates and also an essential supplemental area for oil and gas production in China. Due to the diversity of reservoirs and the extre...Reef-bank reservoirs are an important target for petroleum exploration in marine carbonates and also an essential supplemental area for oil and gas production in China. Due to the diversity of reservoirs and the extreme heterogeneity of reef-banks, it is very difficult to discriminate the sedimentary facies and lithologies in reef-bank reservoirs using conventional well logs. The borehole image log provides clear identification of sedimentary structures and textures and is an ideal tool for discriminating sedimentary facies and lithologies. After examining a large number of borehole images and cores, we propose nine typical patterns for borehole image interpretation and a method that uses these patterns to discriminate sedimentary facies and lithologies in reeI^bank reservoirs automatically. We also develop software with user-friendly interface. The results of applications in reef-bank reservoirs in the middle Tarim Basin and northeast Sichuan have proved that the proposed method and the corresponding software are quite effective.展开更多
In order to investigate the influences of caliper, formation thickness and invaded zone on the form of dual laterologs, forward modeling technique were applied to calculate the dual laterologs for different cases. The...In order to investigate the influences of caliper, formation thickness and invaded zone on the form of dual laterologs, forward modeling technique were applied to calculate the dual laterologs for different cases. The result shows that the resistivity logs become smoother and lower as the borehole diameter increases, the increase of the contrast between mud resistivity and formation resistivity induce the logs to be more pointed. When the formation thickness is less than lm, the two-peak on the logs for resistive invasion vanished, and for thickness between 1 m and 4 m, the form of logs does not vary significantly. If the formation thickness is greater than 4 m, a platform appears on the logs at the middle of the formation. The thinner the invaded zone is, the more obvious the invasion feature on the laterologs is. For thick invaded zone the form of logs tend to be that of an uninvaded resistive formation. The form and amplitude of logs depend on the resistivity contrast between invaded zone, uninvaded formation and adjacentlayers.展开更多
Assessment of reservoir and fracture parameters is necessary to optimize oil production,especially in heterogeneous reservoirs.Core and image logs are regarded as two of the best methods for this aim.However,due to co...Assessment of reservoir and fracture parameters is necessary to optimize oil production,especially in heterogeneous reservoirs.Core and image logs are regarded as two of the best methods for this aim.However,due to core limitations,using image log is considered as the best method.This study aims to use electrical image logs in the carbonate Asmari Formation reservoir in Zagros Basin,SW Iran,in order to evaluate natural fractures,porosity system,permeability profile and heterogeneity index and accordingly compare the results with core and well data.The results indicated that the electrical image logs are reliable for evaluating fracture and reservoir parameters,when there is no core available for a well.Based on the results from formation micro-imager(FMI)and electrical micro-imager(EMI),Asmari was recognized as a completely fractured reservoir in studied field and the reservoir parameters are mainly controlled by fractures.Furthermore,core and image logs indicated that the secondary porosity varies from 0%to 10%.The permeability indicator indicates that zones 3 and 5 have higher permeability index.Image log permeability index shows a very reasonable permeability profile after scaling against core and modular dynamics tester mobility,mud loss and production index which vary between 1 and 1000 md.In addition,no relationship was observed between core porosity and permeability,while the permeability relied heavily on fracture aperture.Therefore,fracture aperture was considered as the most important parameter for the determination of permeability.Sudden changes were also observed at zones 1-1 and 5 in the permeability trend,due to the high fracture aperture.It can be concluded that the electrical image logs(FMI and EMI)are usable for evaluating both reservoir and fracture parameters in wells with no core data in the Zagros Basin,SW Iran.展开更多
The Ordovician Majiagou Formation is one of the main gas-producing strata in the Ordos Basin,China.The identification of hydrocarbon-bearing intervals via conventional well logs is a challenging task.This study descri...The Ordovician Majiagou Formation is one of the main gas-producing strata in the Ordos Basin,China.The identification of hydrocarbon-bearing intervals via conventional well logs is a challenging task.This study describes the litholog of Ma 5(Member 5 of Majiagou Formation)dolostones,and then analyzes the responses of various conventional well logs to the presences of natural gas.The lithology of the gas bearing layers is dominantly of the dolomicrite to fine to medium crystalline dolomite.Natural gas can be produced from the low resistivity layers,and the dry layers are characterized by high resistivities.Neutron-density crossovers are not sensitive to the presences of natural gas.In addition,there are no significant increases in sonic transit times in natural gas bearing layers.NMR(nuclear magnetic resonance)logs,DSI(Dipole Sonic Imager)logs and borehole image logs(XRMI)are introduced to discriminate the fluid property in Majiagou dolostone reservoirs.The gas bearing intervals have broad NMR T2(transverse relaxation time)spectrum with tail distributions as well as large T2gm(T2 logarithmic mean values)values,and the T2 spectrum commonly display polymodal behaviors.In contrast,the dry layers and water layers have low T2gm values and very narrow T2 spectrum without tails.The gas bearing layers are characterized by low Vp/Vs ratios,low Poisson’s ratios and low P-wave impedances,therefore the fluid property can be discriminated using DSI logs,and the interpretation results show good matches with the gas test data.The apparent formation water resistivity(AFWR)spectrum can be derived from XRMI image logs by using the Archie’s formula in the flushed zone.The gas bearing layers have broad apparent formation water resistivity spectrum and tail distributions compared with the dry and water layers,and also the interpretation results from the image logs exhibit good agreement with the gas test data.The fluid property in Majiagou dolostone reservoirs can be discriminated through NMR logs,DSI logs and borehole image logs.This study helps establish a predictable model for fluid property in dolostones,and have implications in dolostone reservoirs with similar geological backgrounds worldwide.展开更多
We investigated the quantity and quality 0f fallen l0gs in different Tsuga l0ngibracteata f0rest c0mmunities in the Tianba0yan Nati0nal Nature Reserve. We used redundancy analysis t0 determine the spatial distributi0n...We investigated the quantity and quality 0f fallen l0gs in different Tsuga l0ngibracteata f0rest c0mmunities in the Tianba0yan Nati0nal Nature Reserve. We used redundancy analysis t0 determine the spatial distributi0n 0f fallen l0gs in the different f0rest c0mmunities and t0 analyze the relati0nships am0ng stand structure, t0p0graphic fact0rs and human disturbance. The v0lume, c0vered area, mean l0g length and number 0f fallen l0gs differed significantly am0ng f0rest types （P 〈 0.05）, but mean diameter at breast height sh0wed n0 significant difference （P 〉 0.05）. The l0g v0lume and c0vered area in different f0rest types sh0wed the f0ll0wing trend： T. l0ngibracteata pure f0rest 〈 T. l0ngibracteata ＋ Olig0staehyum scabrifl0rur 〈 T. l0ngibraeteata ＋ hardw00d 〈 Rh0d0dendr0n simiarum ＋ T. l0ngibraeteata 〈 T. l0ngibraeteata ＋ Phyll0stachys heter0cycla pubescens. The spatial distributi0n patterns 0f l0gs quantity and quality indicated that l0g v0lume and c0vered area were str0ngly affected by envir0nmental fact0rs in the f0ll0wing 0rder： human disturbance 〉 elevati0n 〉 sl0pe p0siti0n 〉 b0le height 〉 tree height 〉 sl0pe aspect 〉 density 〉 basal area 〉 sl0pe gradient. The relative c0ntributi0n 0f envir0nmental variables 0n the t0tal variance was t0p0graphy （76%） 〉 disturbance （42%） 〉 stand structure （35%）. T0p0graphy and disturbance c0mbined explained 8.2% 0f the variance. Fallen l0~s auantitv and aualitvwere negatively related t0 elevati0n and sl0pe p0siti0n, and p0sitively ass0ciated t0 human disturbance. The l0g v0lume decreased fr0m n0rthern t0 s0uthern sl0pes. Envir0nmental fact0rs had the highest impact 0n class I （slightly decayed）, and l0west impact 0n class V （highly decayed）.展开更多
A reliable and effective model for reservoir physical property prediction is a key to reservoir characterization and management.At present,using well logging data to estimate reservoir physical parameters is an import...A reliable and effective model for reservoir physical property prediction is a key to reservoir characterization and management.At present,using well logging data to estimate reservoir physical parameters is an important means for reservoir evaluation.Based on the characteristics of large quantity and complexity of estimating process,we have attempted to design a nonlinear back propagation neural network model optimized by genetic algorithm(BPNNGA)for reservoir porosity prediction.This model is with the advantages of self-learning and self-adaption of back propagation neural network(BPNN),structural parameters optimizing and global searching optimal solution of genetic algorithm(GA).The model is applied to the Chang 8 oil group tight sandstone of Yanchang Formation in southwestern Ordos Basin.According to the correlations between well logging data and measured core porosity data,5 well logging curves(gamma ray,deep induction,density,acoustic,and compensated neutron)are selected as the input neurons while the measured core porosity is selected as the output neurons.The number of hidden layer neurons is defined as 20 by the method of multiple calibrating optimizations.Modeling results demonstrate that the average relative error of the model output is 10.77%,indicating the excellent predicting effect of the model.The predicting results of the model are compared with the predicting results of conventional multivariate stepwise regression algorithm,and BPNN model.The average relative errors of the above models are 12.83%,12.9%,and 13.47%,respectively.Results show that the predicting results of the BPNNGA model are more accurate than that of the other two,and BPNNGA is a more applicable method to estimate the reservoir porosity parameters in the study area.展开更多
How to fit a properly nonlinear classification model from conventional well logs to lithofacies is a key problem for machine learning methods.Kernel methods(e.g.,KFD,SVM,MSVM)are effective attempts to solve this issue...How to fit a properly nonlinear classification model from conventional well logs to lithofacies is a key problem for machine learning methods.Kernel methods(e.g.,KFD,SVM,MSVM)are effective attempts to solve this issue due to abilities of handling nonlinear features by kernel functions.Deep mining of log features indicating lithofacies still needs to be improved for kernel methods.Hence,this work employs deep neural networks to enhance the kernel principal component analysis(KPCA)method and proposes a deep kernel method(DKM)for lithofacies identification using well logs.DKM includes a feature extractor and a classifier.The feature extractor consists of a series of KPCA models arranged according to residual network structure.A gradient-free optimization method is introduced to automatically optimize parameters and structure in DKM,which can avoid complex tuning of parameters in models.To test the validation of the proposed DKM for lithofacies identification,an open-sourced dataset with seven con-ventional logs(GR,CAL,AC,DEN,CNL,LLD,and LLS)and lithofacies labels from the Daniudi Gas Field in China is used.There are eight lithofacies,namely clastic rocks(pebbly,coarse,medium,and fine sand-stone,siltstone,mudstone),coal,and carbonate rocks.The comparisons between DKM and three commonly used kernel methods(KFD,SVM,MSVM)show that(1)DKM(85.7%)outperforms SVM(77%),KFD(79.5%),and MSVM(82.8%)in accuracy of lithofacies identification;(2)DKM is about twice faster than the multi-kernel method(MSVM)with good accuracy.The blind well test in Well D13 indicates that compared with the other three methods DKM improves about 24%in accuracy,35%in precision,41%in recall,and 40%in F1 score,respectively.In general,DKM is an effective method for complex lithofacies identification.This work also discussed the optimal structure and classifier for DKM.Experimental re-sults show that(m_(1),m_(2),O)is the optimal model structure and linear svM is the optimal classifier.(m_(1),m_(2),O)means there are m KPCAs,and then m2 residual units.A workflow to determine an optimal classifier in DKM for lithofacies identification is proposed,too.展开更多
This paper presents an overview of petrophysical research and exploration achievements of low resistivity pay （LRP） zone by well logs in China. It includes geological characteristics and characteristics of well log ...This paper presents an overview of petrophysical research and exploration achievements of low resistivity pay （LRP） zone by well logs in China. It includes geological characteristics and characteristics of well log response of the low resistivity pay zones discovered and evaluated in recent years, as well as the problems in recognizing and evaluating low resistivity pay zones by well logs. The research areas mainly include the Neogene formations in the Bohai Bay Basin, the Triassic formations in the northern Tarim Basin and the Cretaceous formations in the Junggar Basin, The petrophysical research concerning recognition and evaluation of the low resistivity pays, based on their genetic types, is introduced in this paper.展开更多
The tight sandstone in the Tarim Basin has the characteristics of large burial depth and development of nature fractures due to concentrated in-situ stress. Identifying the present-day in-situ stress orientation is im...The tight sandstone in the Tarim Basin has the characteristics of large burial depth and development of nature fractures due to concentrated in-situ stress. Identifying the present-day in-situ stress orientation is important in hydrocarbon exploration and development, but also a key scientific question in understanding naturally fractured reservoirs. This paper presents a case study where we integrate various methods using wireline and image-log data, to identify present-day in-situ stress direction of ultra-deep fractured tight sandstone reservoirs, in the Kuqa depression. We discuss the formation mechanism of the elliptical borehole, compares the advantages and applicable conditions of the double caliper method,resistivity image logs and array sonic logs method. The well borehole diameter is measured orthogonally,then the ellipse is fitted, and the in-situ stress orientation is identified by the azimuth of the short-axis borehole, but it fails in the borehole expansion section, the fracture development section and the borehole collapse section. The micro-resistivity image logs method reveals the borehole breakouts azimuth, and also the strike of induced fractures, which are used to determine the orientation of in-situ stress. However, under water-based mud conditions, it’s hard to distinguish natural fractures from induced fractures by image logs. Under oil-based mud conditions, the induced fractures are difficult to identify due to the compromised image quality. As for the sonic log, shear waves will split when passing through an anisotropic formation, shear waves will split during propagation, and the azimuth of fast shear waves is consistent with the orientation of in-situ stress. However, it is usually affected by the anisotropy caused by the excessively fast rotation of the well log tools, so that the azimuth of fast shear wave cannot effectively reflect the orientation of the in-situ stress. Based on comprehensive assessment and comparison, in this paper we propose a method integrating various logging data to identify the orientation of in-situ stress. Among various types of logging data, the breakouts azimuth identified by image logs is proved to be the most credible in identifying the orientation of in-situ stress, while using the direction of induced fractures under water-based mud conditions is also viable. However, the azimuth of the fast shear wave is consistent with the orientation of maximum in-situ stress only when the rotation speed of the logging tool is low. The caliper method can be used as a reference for verifying the other two methods. Using this integrated method to study the orientation of in-situ stress in the Keshen8 trap, the results show that faults are an important factor affecting the direction of in-situ stress, while multi-level faults will produce superimposed effects that cause the current direction of in-situ stress to change.展开更多
基金the North Dakota Industrial Commission (NDIC) for their financial supportprovided by the University of North Dakota Computational Research Center。
文摘Accurate estimation of mineralogy from geophysical well logs is crucial for characterizing geological formations,particularly in hydrocarbon exploration,CO_(2) sequestration,and geothermal energy development.Current techniques,such as multimineral petrophysical analysis,offer details into mineralogical distribution.However,it is inherently time-intensive and demands substantial geological expertise for accurate model evaluation.Furthermore,traditional machine learning techniques often struggle to predict mineralogy accurately and sometimes produce estimations that violate fundamental physical principles.To address this,we present a new approach using Physics-Integrated Neural Networks(PINNs),that combines data-driven learning with domain-specific physical constraints,embedding petrophysical relationships directly into the neural network architecture.This approach enforces that predictions adhere to physical laws.The methodology is applied to the Broom Creek Deep Saline aquifer,a CO_(2) sequestration site in the Williston Basin,to predict the volumes of key mineral constituents—quartz,dolomite,feldspar,anhydrite,illite—along with porosity.Compared to traditional artificial neural networks (ANN),the PINN approach demonstrates higher accuracy and better generalizability,significantly enhancing predictive performance on unseen well datasets.The average mean error across the three blind wells is 0.123 for ANN and 0.042 for PINN,highlighting the superior accuracy of the PINN approach.This method reduces uncertainties in reservoir characterization by improving the reliability of mineralogy and porosity predictions,providing a more robust tool for decision-making in various subsurface geoscience applications.
文摘Pharmaceutical pollution is becoming an increasing threat to aquatic environments since inactive compounds do not break down,and the drug products are accumulated in living organisms.The ability of a drug to dissolve in water(i.e.,LogS)is an important parameter for assessing a drug’s environmental fate,biovailability,and toxicity.LogS is typically measured in a laboratory setting,which can be costly and time-consuming,and does not provide the opportunity to conduct large-scale analyses.This research develops and evaluates machine learning models that can produce LogS estimates and may improve the environmental risk assessments of toxic pharmaceutical pollutants.We used a dataset from the ChEMBL database that contained 8832 molecular compounds.Various data preprocessing and cleaning techniques were applied(i.e.,removing the missing values),we then recorded chemical properties by normalizing and,even,using some feature selection techniques.We evaluated logS with a total of several machine learning and deep learning models,including;linear regression,random forests(RF),support vector machines(SVM),gradient boosting(GBM),and artificial neural networks(ANNs).We assessed model performance using a series of metrics,including root mean square error(RMSE)and mean absolute error(MAE),as well as the coefficient of determination(R^(2)).The findings show that the Least Angle Regression(LAR)model performed the best with an R^(2) value close to 1.0000,confirming high predictive accuracy.The OMP model performed well with good accuracy(R^(2)=0.8727)while remaining computationally cheap,while other models(e.g.,neural networks,random forests)performed well but were too computationally expensive.Finally,to assess the robustness of the results,an error analysis indicated that residuals were evenly distributed around zero,confirming the results from the LAR model.The current research illustrates the potential of AI in anticipating drug solubility,providing support for green pharmaceutical design and environmental risk assessment.Future work should extend predictions to include degradation and toxicity to enhance predictive power and applicability.
基金funded by the Basic Science Centre Project of the National Natural Science Foundation of China(Grant No.72088101)supported by the Higher Education Commission,Pakistan(Grant No.20-14925/NRPU/R&D/HEC/2021-2021)+1 种基金the Researchers Supporting Project Number(Grant No.RSP2025R351)King Saud University,Riyadh,Saudi Arabia,for funding this research article.
文摘Pore pressure is a decisive measure to assess the reservoir’s geomechanical properties,ensures safe and efficient drilling operations,and optimizes reservoir characterization and production.The conventional approaches sometimes fail to comprehend complex and persistent relationships between pore pressure and formation properties in the heterogeneous reservoirs.This study presents a novel machine learning optimized pore pressure prediction method with a limited dataset,particularly in complex formations.The method addresses the conventional approach's limitations by leveraging its capability to learn complex data relationships.It integrates the best Gradient Boosting Regressor(GBR)algorithm to model pore pressure at wells and later utilizes ContinuousWavelet Transformation(CWT)of the seismic dataset for spatial analysis,and finally employs Deep Neural Network for robust and precise pore pressure modeling for the whole volume.In the second stage,for the spatial variations of pore pressure in the thin Khadro Formation sand reservoir across the entire subsurface area,a three-dimensional pore pressure prediction is conducted using CWT.The relationship between the CWT and geomechanical properties is then established through supervised machine learning models on well locations to predict the uncertainties in pore pressure.Among all intelligent regression techniques developed using petrophysical and elastic properties for pore pressure prediction,the GBR has provided exceptional results that have been validated by evaluation metrics based on the R^(2) score i.e.,0.91 between the calibrated and predicted pore pressure.Via the deep neural network,the relationship between CWT resultant traces and predicted pore pressure is established to analyze the spatial variation.
基金the National Natural Science Foundation of China(42472194,42302153,and 42002144)the Fundamental Research Funds for the Central Univer-sities(22CX06002A).
文摘Karst fractures serve as crucial seepage channels and storage spaces for carbonate natural gas reservoirs,and electrical image logs are vital data for visualizing and characterizing such fractures.However,the conventional approach of identifying fractures using electrical image logs predominantly relies on manual processes that are not only time-consuming but also highly subjective.In addition,the heterogeneity and strong dissolution tendency of karst carbonate reservoirs lead to complexity and variety in fracture geometry,which makes it difficult to accurately identify fractures.In this paper,the electrical image logs network(EILnet)da deep-learning-based intelligent semantic segmentation model with a selective attention mechanism and selective feature fusion moduledwas created to enable the intelligent identification and segmentation of different types of fractures through electrical logging images.Data from electrical image logs representing structural and induced fractures were first selected using the sliding window technique before image inpainting and data augmentation were implemented for these images to improve the generalizability of the model.Various image-processing tools,including the bilateral filter,Laplace operator,and Gaussian low-pass filter,were also applied to the electrical logging images to generate a multi-attribute dataset to help the model learn the semantic features of the fractures.The results demonstrated that the EILnet model outperforms mainstream deep-learning semantic segmentation models,such as Fully Convolutional Networks(FCN-8s),U-Net,and SegNet,for both the single-channel dataset and the multi-attribute dataset.The EILnet provided significant advantages for the single-channel dataset,and its mean intersection over union(MIoU)and pixel accuracy(PA)were 81.32%and 89.37%,respectively.In the case of the multi-attribute dataset,the identification capability of all models improved to varying degrees,with the EILnet achieving the highest MIoU and PA of 83.43%and 91.11%,respectively.Further,applying the EILnet model to various blind wells demonstrated its ability to provide reliable fracture identification,thereby indicating its promising potential applications.
基金supported by National Key Research and Development Program (2019YFA0708301)National Natural Science Foundation of China (51974337)+2 种基金the Strategic Cooperation Projects of CNPC and CUPB (ZLZX2020-03)Science and Technology Innovation Fund of CNPC (2021DQ02-0403)Open Fund of Petroleum Exploration and Development Research Institute of CNPC (2022-KFKT-09)
文摘We propose an integrated method of data-driven and mechanism models for well logging formation evaluation,explicitly focusing on predicting reservoir parameters,such as porosity and water saturation.Accurately interpreting these parameters is crucial for effectively exploring and developing oil and gas.However,with the increasing complexity of geological conditions in this industry,there is a growing demand for improved accuracy in reservoir parameter prediction,leading to higher costs associated with manual interpretation.The conventional logging interpretation methods rely on empirical relationships between logging data and reservoir parameters,which suffer from low interpretation efficiency,intense subjectivity,and suitability for ideal conditions.The application of artificial intelligence in the interpretation of logging data provides a new solution to the problems existing in traditional methods.It is expected to improve the accuracy and efficiency of the interpretation.If large and high-quality datasets exist,data-driven models can reveal relationships of arbitrary complexity.Nevertheless,constructing sufficiently large logging datasets with reliable labels remains challenging,making it difficult to apply data-driven models effectively in logging data interpretation.Furthermore,data-driven models often act as“black boxes”without explaining their predictions or ensuring compliance with primary physical constraints.This paper proposes a machine learning method with strong physical constraints by integrating mechanism and data-driven models.Prior knowledge of logging data interpretation is embedded into machine learning regarding network structure,loss function,and optimization algorithm.We employ the Physically Informed Auto-Encoder(PIAE)to predict porosity and water saturation,which can be trained without labeled reservoir parameters using self-supervised learning techniques.This approach effectively achieves automated interpretation and facilitates generalization across diverse datasets.
基金supported by National Natural Science Foundation of China(Nos.41872150,U2344209 and U19B6003)the PetroChina Southwest Oil and Gasfield Company(No.2020-54365)。
文摘Machine learning algorithms are widely used to interpret well logging data.To enhance the algorithms'robustness,shuffling the well logging data is an unavoidable feature engineering before training models.However,latent information stored between different well logging types and depth is destroyed during the shuffle.To investigate the influence of latent information,this study implements graph convolution networks(GCNs),long-short temporal memory models,recurrent neural networks,temporal convolution networks,and two artificial neural networks to predict the microbial lithology in the fourth member of the Dengying Formation,Moxi gas field,central Sichuan Basin.Results indicate that the GCN model outperforms other models.The accuracy,F1-score,and area under curve of the GCN model are 0.90,0.90,and 0.95,respectively.Experimental results indicate that the time-series data facilitates lithology prediction and helps determine lithological fluctuations in the vertical direction.All types of logs from the spectral in the GCN model and also facilitates lithology identification.Only on condition combined with latent information,the GCN model reaches excellent microbialite classification resolution at the centimeter scale.Ultimately,the two actual cases show tricks for using GCN models to predict potential microbialite in other formations and areas,proving that the GCN model can be adopted in the industry.
基金financially supported by the Russian federal research project No.FWZZ-2022-0026“Innovative aspects of electro-dynamics in problems of exploration and oilfield geophysics”.
文摘We propose a novel workflow for fast forward modeling of well logs in axially symmetric 2D models of the nearwellbore environment.The approach integrates the finite element method with deep residual neural networks to achieve exceptional computational efficiency and accuracy.The workflow is demonstrated through the modeling of wireline electromagnetic propagation resistivity logs,where the measured responses exhibit a highly nonlinear relationship with formation properties.The motivation for this research is the need for advanced modeling al-gorithms that are fast enough for use in modern quantitative interpretation tools,where thousands of simulations may be required in iterative inversion processes.The proposed algorithm achieves a remarkable enhancement in performance,being up to 3000 times faster than the finite element method alone when utilizing a GPU.While still ensuring high accuracy,this makes it well-suited for practical applications when reliable payzone assessment is needed in complex environmental scenarios.Furthermore,the algorithm’s efficiency positions it as a promising tool for stochastic Bayesian inversion,facilitating reliable uncertainty quantification in subsurface property estimation.
基金supported by the ABT SHIELD(Anti-Bot and Trolls Shield)project at the Systems Research Institute,Polish Academy of Sciences,in cooperation with EDGE NPDRPMA.01.02.00-14-B448/18-00 funded by the Regional Development Fund for the development of Mazovia.
文摘Distinguishing between web traffic generated by bots and humans is an important task in the evaluation of online marketing campaigns.One of the main challenges is related to only partial availability of the performance metrics:although some users can be unambiguously classified as bots,the correct label is uncertain in many cases.This calls for the use of classifiers capable of explaining their decisions.This paper demonstrates two such mechanisms based on features carefully engineered from web logs.The first is a man-made rule-based system.The second is a hierarchical model that first performs clustering and next classification using human-centred,interpretable methods.The stability of the proposed methods is analyzed and a minimal set of features that convey the classdiscriminating information is selected.The proposed data processing and analysis methodology are successfully applied to real-world data sets from online publishers.
文摘Pore size analysis plays a pivotal role in unraveling reservoir behavior and its intricate relationship with confined fluids.Traditional methods for predicting pore size distribution(PSD),relying on drilling cores or thin sections,face limitations associated with depth specificity.In this study,we introduce an innovative framework that leverages nuclear magnetic resonance(NMR)log data,encompassing clay-bound water(CBW),bound volume irreducible(BVI),and free fluid volume(FFV),to determine three PSDs(micropores,mesopores,and macropores).Moreover,we establish a robust pore size classification(PSC)system utilizing ternary plots,derived from the PSDs.Within the three studied wells,NMR log data is exclusive to one well(well-A),while conventional well logs are accessible for all three wells(well-A,well-B,and well-C).This distinction enables PSD predictions for the remaining two wells(B and C).To prognosticate NMR outputs(CBW,BVI,FFV)for these wells,a two-step deep learning(DL)algorithm is implemented.Initially,three feature selection algorithms(f-classif,f-regression,and mutual-info-regression)identify the conventional well logs most correlated to NMR outputs in well-A.The three feature selection algorithms utilize statistical computations.These algorithms are utilized to systematically identify and optimize pertinent input features,thereby augmenting model interpretability and predictive efficacy within intricate data-driven endeavors.So,all three feature selection algorithms introduced the number of 4 logs as the most optimal number of inputs to the DL algorithm with different combinations of logs for each of the three desired outputs.Subsequently,the CUDA Deep Neural Network Long Short-Term Memory algorithm(CUDNNLSTM),belonging to the category of DL algorithms and harnessing the computational power of GPUs,is employed for the prediction of CBW,BVI,and FFV logs.This prediction leverages the optimal logs identified in the preceding step.Estimation of NMR outputs was done first in well-A(80%of data as training and 20%as testing).The correlation coefficient(CC)between the actual and estimated data for the three outputs CBW,BVI and FFV are 95%,94%,and 97%,respectively,as well as root mean square error(RMSE)was obtained 0.0081,0.098,and 0.0089,respectively.To assess the effectiveness of the proposed algorithm,we compared it with two traditional methods for log estimation:multiple regression and multi-resolution graph-based clustering methods.The results demonstrate the superior accuracy of our algorithm in comparison to these conventional approaches.This DL-driven approach facilitates PSD prediction grounded in fluid saturation for wells B and C.Ternary plots are then employed for PSCs.Seven distinct PSCs within well-A employing actual NMR logs(CBW,BVI,FFV),in conjunction with an equivalent count within wells B and C utilizing three predicted logs,are harmoniously categorized leading to the identification of seven distinct pore size classification facies(PSCF).this research introduces an advanced approach to pore size classification and prediction,fusing NMR logs with deep learning techniques and extending their application to nearby wells without NMR log.The resulting PSCFs offer valuable insights into generating precise and detailed reservoir 3D models.
基金sponsored by the National S&T Major Special Project(No.2008ZX05020-01)
文摘Reef-bank reservoirs are an important target for petroleum exploration in marine carbonates and also an essential supplemental area for oil and gas production in China. Due to the diversity of reservoirs and the extreme heterogeneity of reef-banks, it is very difficult to discriminate the sedimentary facies and lithologies in reef-bank reservoirs using conventional well logs. The borehole image log provides clear identification of sedimentary structures and textures and is an ideal tool for discriminating sedimentary facies and lithologies. After examining a large number of borehole images and cores, we propose nine typical patterns for borehole image interpretation and a method that uses these patterns to discriminate sedimentary facies and lithologies in reeI^bank reservoirs automatically. We also develop software with user-friendly interface. The results of applications in reef-bank reservoirs in the middle Tarim Basin and northeast Sichuan have proved that the proposed method and the corresponding software are quite effective.
文摘In order to investigate the influences of caliper, formation thickness and invaded zone on the form of dual laterologs, forward modeling technique were applied to calculate the dual laterologs for different cases. The result shows that the resistivity logs become smoother and lower as the borehole diameter increases, the increase of the contrast between mud resistivity and formation resistivity induce the logs to be more pointed. When the formation thickness is less than lm, the two-peak on the logs for resistive invasion vanished, and for thickness between 1 m and 4 m, the form of logs does not vary significantly. If the formation thickness is greater than 4 m, a platform appears on the logs at the middle of the formation. The thinner the invaded zone is, the more obvious the invasion feature on the laterologs is. For thick invaded zone the form of logs tend to be that of an uninvaded resistive formation. The form and amplitude of logs depend on the resistivity contrast between invaded zone, uninvaded formation and adjacentlayers.
基金financial and data support from NISOC Oil Company.
文摘Assessment of reservoir and fracture parameters is necessary to optimize oil production,especially in heterogeneous reservoirs.Core and image logs are regarded as two of the best methods for this aim.However,due to core limitations,using image log is considered as the best method.This study aims to use electrical image logs in the carbonate Asmari Formation reservoir in Zagros Basin,SW Iran,in order to evaluate natural fractures,porosity system,permeability profile and heterogeneity index and accordingly compare the results with core and well data.The results indicated that the electrical image logs are reliable for evaluating fracture and reservoir parameters,when there is no core available for a well.Based on the results from formation micro-imager(FMI)and electrical micro-imager(EMI),Asmari was recognized as a completely fractured reservoir in studied field and the reservoir parameters are mainly controlled by fractures.Furthermore,core and image logs indicated that the secondary porosity varies from 0%to 10%.The permeability indicator indicates that zones 3 and 5 have higher permeability index.Image log permeability index shows a very reasonable permeability profile after scaling against core and modular dynamics tester mobility,mud loss and production index which vary between 1 and 1000 md.In addition,no relationship was observed between core porosity and permeability,while the permeability relied heavily on fracture aperture.Therefore,fracture aperture was considered as the most important parameter for the determination of permeability.Sudden changes were also observed at zones 1-1 and 5 in the permeability trend,due to the high fracture aperture.It can be concluded that the electrical image logs(FMI and EMI)are usable for evaluating both reservoir and fracture parameters in wells with no core data in the Zagros Basin,SW Iran.
基金This work is financially supported by the Science Foundation of China University of Petroleum, Beijing (Grant No. 2462017YJRC023)the Fundamental Research Funds for the Central Universities and the Opening Fund of Key Laboratory of Deep Oil & Gas (Grant No. 20CX02116A)
文摘The Ordovician Majiagou Formation is one of the main gas-producing strata in the Ordos Basin,China.The identification of hydrocarbon-bearing intervals via conventional well logs is a challenging task.This study describes the litholog of Ma 5(Member 5 of Majiagou Formation)dolostones,and then analyzes the responses of various conventional well logs to the presences of natural gas.The lithology of the gas bearing layers is dominantly of the dolomicrite to fine to medium crystalline dolomite.Natural gas can be produced from the low resistivity layers,and the dry layers are characterized by high resistivities.Neutron-density crossovers are not sensitive to the presences of natural gas.In addition,there are no significant increases in sonic transit times in natural gas bearing layers.NMR(nuclear magnetic resonance)logs,DSI(Dipole Sonic Imager)logs and borehole image logs(XRMI)are introduced to discriminate the fluid property in Majiagou dolostone reservoirs.The gas bearing intervals have broad NMR T2(transverse relaxation time)spectrum with tail distributions as well as large T2gm(T2 logarithmic mean values)values,and the T2 spectrum commonly display polymodal behaviors.In contrast,the dry layers and water layers have low T2gm values and very narrow T2 spectrum without tails.The gas bearing layers are characterized by low Vp/Vs ratios,low Poisson’s ratios and low P-wave impedances,therefore the fluid property can be discriminated using DSI logs,and the interpretation results show good matches with the gas test data.The apparent formation water resistivity(AFWR)spectrum can be derived from XRMI image logs by using the Archie’s formula in the flushed zone.The gas bearing layers have broad apparent formation water resistivity spectrum and tail distributions compared with the dry and water layers,and also the interpretation results from the image logs exhibit good agreement with the gas test data.The fluid property in Majiagou dolostone reservoirs can be discriminated through NMR logs,DSI logs and borehole image logs.This study helps establish a predictable model for fluid property in dolostones,and have implications in dolostone reservoirs with similar geological backgrounds worldwide.
基金supported by the National Natural Science Foundation of China (Grant No.31370624)the Specialized Research Fund for the Doctoral Program of Higher Education (Grant No. 20103515110005)the Natural Science Foundation of Fujian, China (Grant No. 2011J01071)
文摘We investigated the quantity and quality 0f fallen l0gs in different Tsuga l0ngibracteata f0rest c0mmunities in the Tianba0yan Nati0nal Nature Reserve. We used redundancy analysis t0 determine the spatial distributi0n 0f fallen l0gs in the different f0rest c0mmunities and t0 analyze the relati0nships am0ng stand structure, t0p0graphic fact0rs and human disturbance. The v0lume, c0vered area, mean l0g length and number 0f fallen l0gs differed significantly am0ng f0rest types （P 〈 0.05）, but mean diameter at breast height sh0wed n0 significant difference （P 〉 0.05）. The l0g v0lume and c0vered area in different f0rest types sh0wed the f0ll0wing trend： T. l0ngibracteata pure f0rest 〈 T. l0ngibracteata ＋ Olig0staehyum scabrifl0rur 〈 T. l0ngibraeteata ＋ hardw00d 〈 Rh0d0dendr0n simiarum ＋ T. l0ngibraeteata 〈 T. l0ngibraeteata ＋ Phyll0stachys heter0cycla pubescens. The spatial distributi0n patterns 0f l0gs quantity and quality indicated that l0g v0lume and c0vered area were str0ngly affected by envir0nmental fact0rs in the f0ll0wing 0rder： human disturbance 〉 elevati0n 〉 sl0pe p0siti0n 〉 b0le height 〉 tree height 〉 sl0pe aspect 〉 density 〉 basal area 〉 sl0pe gradient. The relative c0ntributi0n 0f envir0nmental variables 0n the t0tal variance was t0p0graphy （76%） 〉 disturbance （42%） 〉 stand structure （35%）. T0p0graphy and disturbance c0mbined explained 8.2% 0f the variance. Fallen l0~s auantitv and aualitvwere negatively related t0 elevati0n and sl0pe p0siti0n, and p0sitively ass0ciated t0 human disturbance. The l0g v0lume decreased fr0m n0rthern t0 s0uthern sl0pes. Envir0nmental fact0rs had the highest impact 0n class I （slightly decayed）, and l0west impact 0n class V （highly decayed）.
基金supported by the National Natural Science Foundation of China(No.41002045)。
文摘A reliable and effective model for reservoir physical property prediction is a key to reservoir characterization and management.At present,using well logging data to estimate reservoir physical parameters is an important means for reservoir evaluation.Based on the characteristics of large quantity and complexity of estimating process,we have attempted to design a nonlinear back propagation neural network model optimized by genetic algorithm(BPNNGA)for reservoir porosity prediction.This model is with the advantages of self-learning and self-adaption of back propagation neural network(BPNN),structural parameters optimizing and global searching optimal solution of genetic algorithm(GA).The model is applied to the Chang 8 oil group tight sandstone of Yanchang Formation in southwestern Ordos Basin.According to the correlations between well logging data and measured core porosity data,5 well logging curves(gamma ray,deep induction,density,acoustic,and compensated neutron)are selected as the input neurons while the measured core porosity is selected as the output neurons.The number of hidden layer neurons is defined as 20 by the method of multiple calibrating optimizations.Modeling results demonstrate that the average relative error of the model output is 10.77%,indicating the excellent predicting effect of the model.The predicting results of the model are compared with the predicting results of conventional multivariate stepwise regression algorithm,and BPNN model.The average relative errors of the above models are 12.83%,12.9%,and 13.47%,respectively.Results show that the predicting results of the BPNNGA model are more accurate than that of the other two,and BPNNGA is a more applicable method to estimate the reservoir porosity parameters in the study area.
基金supported by the National Natural Science Foundation of China(Grant No.42002134)China Postdoctoral Science Foundation(Grant No.2021T140735)Science Foundation of China University of Petroleum,Beijing(Grant Nos.2462020XKJS02 and 2462020YXZZ004).
文摘How to fit a properly nonlinear classification model from conventional well logs to lithofacies is a key problem for machine learning methods.Kernel methods(e.g.,KFD,SVM,MSVM)are effective attempts to solve this issue due to abilities of handling nonlinear features by kernel functions.Deep mining of log features indicating lithofacies still needs to be improved for kernel methods.Hence,this work employs deep neural networks to enhance the kernel principal component analysis(KPCA)method and proposes a deep kernel method(DKM)for lithofacies identification using well logs.DKM includes a feature extractor and a classifier.The feature extractor consists of a series of KPCA models arranged according to residual network structure.A gradient-free optimization method is introduced to automatically optimize parameters and structure in DKM,which can avoid complex tuning of parameters in models.To test the validation of the proposed DKM for lithofacies identification,an open-sourced dataset with seven con-ventional logs(GR,CAL,AC,DEN,CNL,LLD,and LLS)and lithofacies labels from the Daniudi Gas Field in China is used.There are eight lithofacies,namely clastic rocks(pebbly,coarse,medium,and fine sand-stone,siltstone,mudstone),coal,and carbonate rocks.The comparisons between DKM and three commonly used kernel methods(KFD,SVM,MSVM)show that(1)DKM(85.7%)outperforms SVM(77%),KFD(79.5%),and MSVM(82.8%)in accuracy of lithofacies identification;(2)DKM is about twice faster than the multi-kernel method(MSVM)with good accuracy.The blind well test in Well D13 indicates that compared with the other three methods DKM improves about 24%in accuracy,35%in precision,41%in recall,and 40%in F1 score,respectively.In general,DKM is an effective method for complex lithofacies identification.This work also discussed the optimal structure and classifier for DKM.Experimental re-sults show that(m_(1),m_(2),O)is the optimal model structure and linear svM is the optimal classifier.(m_(1),m_(2),O)means there are m KPCAs,and then m2 residual units.A workflow to determine an optimal classifier in DKM for lithofacies identification is proposed,too.
基金Supported by CNPC Innovation Foundation,Research Projects of PetroChina,Xinjiang and Tarim Oil Companies
文摘This paper presents an overview of petrophysical research and exploration achievements of low resistivity pay （LRP） zone by well logs in China. It includes geological characteristics and characteristics of well log response of the low resistivity pay zones discovered and evaluated in recent years, as well as the problems in recognizing and evaluating low resistivity pay zones by well logs. The research areas mainly include the Neogene formations in the Bohai Bay Basin, the Triassic formations in the northern Tarim Basin and the Cretaceous formations in the Junggar Basin, The petrophysical research concerning recognition and evaluation of the low resistivity pays, based on their genetic types, is introduced in this paper.
基金financial support for this work comes from the Science Foundation of CUPB(No.2462017YJRC023)supported by the Exploration and Development Research Institute of Petro China Tarim Oilfield Branch Company
文摘The tight sandstone in the Tarim Basin has the characteristics of large burial depth and development of nature fractures due to concentrated in-situ stress. Identifying the present-day in-situ stress orientation is important in hydrocarbon exploration and development, but also a key scientific question in understanding naturally fractured reservoirs. This paper presents a case study where we integrate various methods using wireline and image-log data, to identify present-day in-situ stress direction of ultra-deep fractured tight sandstone reservoirs, in the Kuqa depression. We discuss the formation mechanism of the elliptical borehole, compares the advantages and applicable conditions of the double caliper method,resistivity image logs and array sonic logs method. The well borehole diameter is measured orthogonally,then the ellipse is fitted, and the in-situ stress orientation is identified by the azimuth of the short-axis borehole, but it fails in the borehole expansion section, the fracture development section and the borehole collapse section. The micro-resistivity image logs method reveals the borehole breakouts azimuth, and also the strike of induced fractures, which are used to determine the orientation of in-situ stress. However, under water-based mud conditions, it’s hard to distinguish natural fractures from induced fractures by image logs. Under oil-based mud conditions, the induced fractures are difficult to identify due to the compromised image quality. As for the sonic log, shear waves will split when passing through an anisotropic formation, shear waves will split during propagation, and the azimuth of fast shear waves is consistent with the orientation of in-situ stress. However, it is usually affected by the anisotropy caused by the excessively fast rotation of the well log tools, so that the azimuth of fast shear wave cannot effectively reflect the orientation of the in-situ stress. Based on comprehensive assessment and comparison, in this paper we propose a method integrating various logging data to identify the orientation of in-situ stress. Among various types of logging data, the breakouts azimuth identified by image logs is proved to be the most credible in identifying the orientation of in-situ stress, while using the direction of induced fractures under water-based mud conditions is also viable. However, the azimuth of the fast shear wave is consistent with the orientation of maximum in-situ stress only when the rotation speed of the logging tool is low. The caliper method can be used as a reference for verifying the other two methods. Using this integrated method to study the orientation of in-situ stress in the Keshen8 trap, the results show that faults are an important factor affecting the direction of in-situ stress, while multi-level faults will produce superimposed effects that cause the current direction of in-situ stress to change.
