Hydrate reservoirs are different from the host reservoirs of all other fossil energy sources because the characteristics of hydrate reservoirs are generally controlled by deep-sea fine-grained sedimentation. In such r...Hydrate reservoirs are different from the host reservoirs of all other fossil energy sources because the characteristics of hydrate reservoirs are generally controlled by deep-sea fine-grained sedimentation. In such reservoirs, the reliability of the classical logging evaluation models established for diagenetic reservoirs is questionable. This study used well W8 in the Qiongdongnan Basin to explore the clay content, porosity, saturation, and hydrate-enriched layer identification of a logging-based hydrate reservoir, and it was found that considering the effect of the clay content on the log response is necessary in the logging evaluation of hydrate reservoirs. In the evaluation of clay content, a method based on the optimization inversion method can obtain a more reliable clay content than other methods. Fine-grained sediment reservoirs have a high clay content, and the effect of clay on log responses must be considered when calculating porosity. In addition, combining density logging and neutron porosity logging data can obtain the best porosity calculation results, and the porosity calculation method based on sonic logging predicted that the porosity of the studied reservoir was low. It was very effective to identify hydrate layers based on resistivity, but the clay distribution and pore structure will also affect the relationship between resistivity, porosity and saturation, and it was suggested that the factors effecting the resistivity of different layers should be considered in the saturation evaluation and that a suitable model should be selected. This study also considered the lack of clarity of the relationships among the lithology, physical properties, hydrate-bearing occurrence properties, and log response properties of hydrate reservoirs and the lack of specialized petrophysical models. This research can directly help to improve hydrate logging evaluation.展开更多
It is always significant for assessing and evaluation of oil bearing layers, especially for well logging data processing and interpretation of non marine oil beds to get more accurate physical properties in thin and...It is always significant for assessing and evaluation of oil bearing layers, especially for well logging data processing and interpretation of non marine oil beds to get more accurate physical properties in thin and inter thin layers. This paper presents a definition of measures and the measure presents power law relation with the corresponded scale described by fractal theory. Thus, logging curves can be reconstructed according to this power law relation. This method uses the local structure nearby concurrent points to compensate the average effect of logging probes and measurement errors. As an example, deep and medium induced conductivity (IMPH and IDPH) curves in ODP Leg 127 Hole 797C are reconstructed or corrected. Corrected curves are with less adjacent effects through comparison of corrected curves with original one. And also, the power spectra of corrected well logging curve are abounding with more resolution components than the original one. Thus, fractal correction method makes the well logging more resoluble for thin beds.展开更多
The nonaxisymmetric acousto-electric field excited by an eccentric acoustic source in the borehole based on Pride seismoelectric theory is considered. It is shown that the acoustic field inside the borehole, converted...The nonaxisymmetric acousto-electric field excited by an eccentric acoustic source in the borehole based on Pride seismoelectric theory is considered. It is shown that the acoustic field inside the borehole, converted electric and magnetic fields and coupled fields outside the borehole are composed of an infinitude of multipole fields with different orders. The numerical results show that both the electromagnetic waves and the seismoelectric field in the borehole, and the three components of both electric field and magnetic field can be detected. Measurements on the borehole axis will be of advantage to determining shear velocity information. The components of the symmetric and nonsymmetric acoustic and electromagnetic fields can be strengthened or weakened by adding or subtracting the two full waveforms logged in some azimuths. It may be a new method of directly measuring the shear wave velocity by using the borehole seismoelectric effect.展开更多
Electromagnetic Computer Tomography (ECT) is a method to probe the interior of an inhomogeneous medium via surface measurement in a non-linear way. Due to the great differences in conductivity and permittivity betwe...Electromagnetic Computer Tomography (ECT) is a method to probe the interior of an inhomogeneous medium via surface measurement in a non-linear way. Due to the great differences in conductivity and permittivity between oil and water in the well, Electromagnetic Tomography Well Logging (ETWL), a new flow imaging measurement system, is proposed to describe the distribution and movement of oil/water two-phase flow in the well by scanning the detected region and applying a suitable data processing algorithm. The results of the numerical simulation and physical modeling show that the system could provide a clear image of the flow profile.展开更多
Based on the well logging knowledge graph of hydrocarbon-bearing formation(HBF),a Knowledge-Powered Neural Network Formation Evaluation model(KPNFE)has been proposed.It has the following functions:(1)extracting charac...Based on the well logging knowledge graph of hydrocarbon-bearing formation(HBF),a Knowledge-Powered Neural Network Formation Evaluation model(KPNFE)has been proposed.It has the following functions:(1)extracting characteristic parameters describing HBF in multiple dimensions and multiple scales;(2)showing the characteristic parameter-related entities,relationships,and attributes as vectors via graph embedding technique;(3)intelligently identifying HBF;(4)seamlessly integrating expertise into the intelligent computing to establish the assessment system and ranking algorithm for potential pay recommendation.Taking 547 wells encountered the low porosity and low permeability Chang 6 Member of Triassic in the Jiyuan Block of Ordos Basin,NW China as objects,80%of the wells were randomly selected as the training dataset and the remainder as the validation dataset.The KPNFE prediction results on the validation dataset had a coincidence rate of 94.43%with the expert interpretation results and a coincidence rate of 84.38%for all the oil testing layers,which is 13 percentage points higher in accuracy and over 100 times faster than the primary conventional interpretation.In addition,a number of potential pays likely to produce industrial oil were recommended.The KPNFE model effectively inherits,carries forward and improves the expert knowledge,nicely solving the robustness problem in HBF identification.The KPNFE,with good interpretability and high accuracy of computation results,is a powerful technical means for efficient and high-quality well logging re-evaluation of old wells in mature oilfields.展开更多
The total organic carbon content usually determines the hydrocarbon generation potential of a formation.A higher total organic carbon content often corresponds to a greater possibility of generating large amounts of o...The total organic carbon content usually determines the hydrocarbon generation potential of a formation.A higher total organic carbon content often corresponds to a greater possibility of generating large amounts of oil or gas.Hence,accurately calculating the total organic carbon content in a formation is very important.Present research is focused on precisely calculating the total organic carbon content based on machine learning.At present,many machine learning methods,including backpropagation neural networks,support vector regression,random forests,extreme learning machines,and deep learning,are employed to evaluate the total organic carbon content.However,the principles and perspectives of various machine learning algorithms are quite different.This paper reviews the application of various machine learning algorithms to deal with total organic carbon content evaluation problems.Of various machine learning algorithms used for TOC content predication,two algorithms,the backpropagation neural network and support vector regression are the most commonly used,and the backpropagation neural network is sometimes combined with many other algorithms to achieve better results.Additionally,combining multiple algorithms or using deep learning to increase the number of network layers can further improve the total organic carbon content prediction.The prediction by backpropagation neural network may be better than that by support vector regression;nevertheless,using any type of machine learning algorithm improves the total organic carbon content prediction in a given research block.According to some published literature,the determination coefficient(R^(2))can be increased by up to 0.46 after using machine learning.Deep learning algorithms may be the next breakthrough direction that can significantly improve the prediction of the total organic carbon content.Evaluating the total organic carbon content based on machine learning is of great significance.展开更多
In order to solve the problems of the fine division of sedimentary sequence cycles and their change in two-dimensional space as well as lateral extension contrast, we developed a method of wavelet depth-frequency anal...In order to solve the problems of the fine division of sedimentary sequence cycles and their change in two-dimensional space as well as lateral extension contrast, we developed a method of wavelet depth-frequency analysis. The single signal and composite signal of different Milankovitch cycles are obtained by numerical simulation. The simulated composite signal can be separated into single signals of a single frequency cycle. We also develop a well-seismic calibration insertion technology which helps to realize the calibration from the spectrum characteristics of a single well to the seismic profile. And then we determine the change and distribution characteristics of spectrum cycles in the two-dimensional space. It points out the direction in determining the variations of the regional sedimentary sequence cycles, underground strata structure and the contact relationship.展开更多
In the early exploration of many oilfields,low-resistivity-low-contrast(LRLC)pay zones are easily overlooked due to the resistivity similarity to the water zones.Existing identification methods are model-driven and ca...In the early exploration of many oilfields,low-resistivity-low-contrast(LRLC)pay zones are easily overlooked due to the resistivity similarity to the water zones.Existing identification methods are model-driven and cannot yield satisfactory results when the causes of LRLC pay zones are complicated.In this study,after analyzing a large number of core samples,main causes of LRLC pay zones in the study area are discerned,which include complex distribution of formation water salinity,high irreducible water saturation due to micropores,and high shale volume.Moreover,different oil testing layers may have different causes of LRLC pay zones.As a result,in addition to the well log data of oil testing layers,well log data of adjacent shale layers are also added to the original dataset as reference data.The densitybased spatial clustering algorithm with noise(DBSCAN)is used to cluster the original dataset into 49 clusters.A new dataset is ultimately projected into a feature space with 49 dimensions.The new dataset and oil testing results are respectively treated as input and output to train the multi-layer perceptron(MLP).A total of 3192 samples are used for stratified 8-fold cross-validation,and the accuracy of the MLP is found to be 85.53%.展开更多
Since gas hydrate exists in three different forms at the same time such as pore filling,particle support and separate stratification,the calculation method of hydrate saturation using traditional shaly sand formation ...Since gas hydrate exists in three different forms at the same time such as pore filling,particle support and separate stratification,the calculation method of hydrate saturation using traditional shaly sand formation interpretation models is equivalent to considering only the simple case that hydrate exists as pore filling,and does not consider other complex states.Based on the analysis of hydrate resistivity experimental data and the general form of the resistivity-oil(gas)saturation relationship,the best simplified formula of hydrate saturation calculation is derived,then the physical meaning of the three items are clarified:they respectively represent the resistivity index-saturation relationship when hydrate particles are completely distributed in the pores of formation rocks,supported in the form of particles,and exist in layers,corresponding quantitative evaluation method of hydrate saturation is built.The field application shows that the hydrate saturation calculated by this method is closer to that obtained by sampling analysis.At the same time,it also provides a logging analysis basis for the effective development after hydrate exploration.展开更多
This work concerns the collecting field of the Abrobakro site, the objective of which is to determine the thickness of the layers crossed during drilling from electrical logging in order to propose the equipment plan ...This work concerns the collecting field of the Abrobakro site, the objective of which is to determine the thickness of the layers crossed during drilling from electrical logging in order to propose the equipment plan for the various boreholes. The electrical logging data sheets, particularly those on resistivity and expeditious granulometry using a 1.25 mm and 2 mm mesh sieve, were used. The layer thicknesses are determined with the inflection points on the graphs. The electrical logging shows that the sands in the study area have resistivity values between 400 and 5000 Ω.m. The decrease in resistivity observed at 50 m for all boreholes shows that the static level of the groundwater is at this depth. The results of the accelerated granulometry show that the first 20 meters contain more fine particles and coarse to very coarse sands from 20 m. The granulometry of the screen laying areas shows that the 1.5 mm slot openings are best suited for all drilling in the Abrobakro collecting field. The diameter d10 of the aquiferous sands of the collecting field is close to 1.25 mm.展开更多
Electrical properties are important physical parameters of natural gas hydrate,and,specifically,resistivity has been widely used in the quantitative estimation of hydrate saturation.There are three main methods to stu...Electrical properties are important physical parameters of natural gas hydrate,and,specifically,resistivity has been widely used in the quantitative estimation of hydrate saturation.There are three main methods to study the electrical properties of gas hydrate-bearing sediments:experimental laboratory measurements,numerical simulation,and resistivity logging.Experimental measurements can be divided into three categories:normal electrical measurement,complex resistivity measurement,and electrical resistivity tomography.Experimental measurements show that the resistivity of hydrate-bearing sediment is affected by many factors,and its distribution as well as the hydrate saturation is not uniform;there is a distinct non-Archie phenomenon.The numerical method can simulate the resistivity of sediments by changing the hydrate occurrence state,saturation,distribution,etc.However,it needs to be combined with X-ray CT,nuclear magnetic resonance,and other imaging techniques to characterize the porous characteristics of the hydrate-bearing sediments.Resistivity well logging can easily identify hydrate layers based on their significantly higher resistivity than the background,but the field data of the hydrate layer also has a serious non-Archie phenomenon.Therefore,more experimental measurements and numerical simulation studies are needed to correct the parameters of Archie’s formula.展开更多
Survey and novel research data are used in the present study to classify/identify the lithological type of Verey age reservoirs’rocks.It is shown how the use of X-ray tomography can clarify the degree of heterogeneit...Survey and novel research data are used in the present study to classify/identify the lithological type of Verey age reservoirs’rocks.It is shown how the use of X-ray tomography can clarify the degree of heterogeneity,porosity and permeability of these rocks.These data are then used to elaborate a model of hydraulic fracturing.The resulting software can take into account the properties of proppant and breakdown fluid,thermal reservoir conditions,oil properties,well design data and even the filtration and elastic-mechanical properties of the rocks.Calculations of hydraulic fracturing crack formation are carried out and the results are compared with the data on hydraulic fracturing crack at standard conditions.Significant differences in crack formation in standard and lithotype models are determined.It is shown that the average width of the crack development for the lithotype model is 2.3 times higher than that for the standard model.Moreover,the coverage of crack development in height for the lithotype model is almost 2 times less than that for the standard model.The estimated fracture half-length for the lithotype model is 13.3%less than that of for the standard model.A higher dimensionless fracture conductivity is also obtained for the lithotype model.It is concluded that the proposed approach can increase the reliability of hydraulic fracturing crack models.展开更多
A new multi-exponential inversion method for NMR relaxation signals is presented and tested, which is based on a solid iteration rebuild technique (SIRT). The T2 spectra inversed by the new method are compared with MA...A new multi-exponential inversion method for NMR relaxation signals is presented and tested, which is based on a solid iteration rebuild technique (SIRT). The T2 spectra inversed by the new method are compared with MAP-II results. The T1 and T2 inversion results with different pre-assigned relaxation times and different SNR show that 16 to 64 logarithm equal spaced time constants is better obviously than MAP-II. And in particular, it can ensure the relaxation time distribution when the SNR of the measured signal is very low. The new algorithm has been applied in rock core NMR analysis and NMR logging data process and interpretation.展开更多
The acoustic tools widely used in borehole well logging and being developed in borehole acoustic reflection imaging do not have the function of azimuthal measurement due to a symmetric source, so they can not be used ...The acoustic tools widely used in borehole well logging and being developed in borehole acoustic reflection imaging do not have the function of azimuthal measurement due to a symmetric source, so they can not be used to evaluate the azimuthal character of borehole formation. In this paper, a 3D finite difference method was used to simulate the acoustic fields in a fluid-filled borehole generated by a traditional monopole source and a phased arc array. Acoustic waveforms were presented for both cases. The analysis of the simulated waveforms showed that different from the monopole source, the acoustic energy generated by the phased arc array transmitter mainly radiated to the borehole in a narrow azimuthal range, which was the key technique to implement azimuthal acoustic well logging. Similar to the monopole source, the waveforms generated by the phased arc array in the fluid-filled borehole also contain compressional (P) waves and shear (S) waves refracted along the borehole, which is the theoretical foundation of phased arc array acoustic well logging.展开更多
The Hutubi (呼图壁) River reservoir of well block DX12 is a lithologic hydrocarbon reservoir that is under tectonic settings. The main oil-bearing sand body in this area is thin and has a poor transverse connectivit...The Hutubi (呼图壁) River reservoir of well block DX12 is a lithologic hydrocarbon reservoir that is under tectonic settings. The main oil-bearing sand body in this area is thin and has a poor transverse connectivity. Because of the complexity of the oil-water relationship, the oil reservoir presents a low resistivity feature, which brings great difficulties to hydrocarbon reservoir identification. This article develops an effective method of well log interpretation that can meet the requirement of low resistivity reservoir well logging evaluation. The authors combine the oil reservoir geology feature, the oil well logging curve characteristics and chemical analytical data to analyze the reasons for low resistivity, then establish the appropriate reservoir parameter explanation model, which uses different saturation computational methods according to different generations. When the clay content is more than 5%, we select W-S dual water model; when the shale content is more than 13%, we use the Schlumberger formula; when the shale content is less then 13%, we use Archie's formula. The well logging evaluation method of low resistivity reservoir has been improved by the irreducible water saturation formula which is established by the permeability, the porosity, the coefficient of pore structure and the shale content, hydrocarbon reservoir recognition charts, and the non-resistivity logging methods (repeat formation test (RFT); modular dynamic test (MDT), etc.). The coincidence rate for this arrangement of the well logging integrated interpretation is 82.6% in the well block DX12. It is a powerful direction for low resistivity well log interpretation.展开更多
The lithologies of the Chinese Continental Scientific Drilling main hole (CCSD-MH) are mainly comprised of orthogneiss,paragneiss,eclogite,amphibolite,and ultramafic rocks.The statistical results of logs of CCSD-MH ...The lithologies of the Chinese Continental Scientific Drilling main hole (CCSD-MH) are mainly comprised of orthogneiss,paragneiss,eclogite,amphibolite,and ultramafic rocks.The statistical results of logs of CCSD-MH indicate that ultramafic rocks are characterized by very high CNL (neutron log) and very low GR (gamma ray log) and RD (resistivity log);eclogites are characterized by high DEN (density),VP (P-wave velocity) and PE (photoelectric absorption capture cross section);orthogneiss and paragneiss are characterized by high GR,U (uranium content),Th (thorium content),K (potassium content) and RD,and low DEN,PE,and CNL;logging values of amphibolite are between the logging values of eclogites and paragneiss.In addition,the logs could reflect the degree of retrograde metamorphism of eclogites.The upper section (100-2 000 m) shows higher DEN,PE,VP,and lower GR,U,Th,K,RD than the lower section (2 000-5 000 m).Most logs of the upper section are more fluctuant than those of the lower section.This indicates that the upper section has more heterogeneities than the lower section.The cross plots of logs indicate that DEN,GR,K,and CNL are more powerful in identifying ultrahigh pressure metamorphic (UHPM) rocks at the CCSD-MH.GR value of the rocks from CCSD-MH shows obviously an increasing trend from ultramafic rock (the most mafic rocks at CCSD-MH) to orthogneiss (the most acid rocks at CCSD-MH).On the contrary,DEN value decreases from the ultramafic rock to the orthogneiss.CNL log is a good indicator of the content of structure water in crystalline rocks.展开更多
To promote adaptation of logging evaluation technologies to the development trend of unconventional oil and gas exploration and development era in China,the current situation and challenges of logging evaluation techn...To promote adaptation of logging evaluation technologies to the development trend of unconventional oil and gas exploration and development era in China,the current situation and challenges of logging evaluation technologies in China are analyzed systematically.Based on the concept of that demand drives technology development,and referring to the international leading technologies,development strategy of logging evaluation technology in China has been put forward.(1)Deepen petrophysics study:mobile 2 D NMR laboratory analysis technology for full diameter core should be developed,characteristic charts and evaluation standards of different fluid properties,different pore structures and different core exposure times should be established based on longitudinal and traverse relaxation spectra;in-depth digital rock experiment and mathematical and physical simulation research should be carried out to create innovative logging evaluation methods;acoustic and electrical anisotropy experimental analysis technology should be developed,and corresponding logging evaluation methods be innovated.(2)Strengthen target processing of logging data:precise inversion processing technology and sensitive information extraction technology of 2 D NMR logging should be developed to finely describe the micro-pore distribution in tight reservoir and accurately distinguish movable oil,bound oil,and bound water etc.The processing method of 3 D ultra-distance detection acoustic logging should be researched.(3)Develop special logging interpretation and evaluation methods:first,mathematical model for quantitatively describing the saturation distribution law of unconventional oil and gas near source and in source should be created;second,evaluation methods and standards of shale oil and deep shale gas"sweet-spots"with mobile oil content and gas content as key parameter separately should be researched vigorously;third,calculation methods of pore pressure under two high-pressure genetic mechanisms,under-compaction and hydrocarbon charging,should be improved;fourth,evaluation method of formation fracability considering the reservoir geologic and engineering quality,and optimization method of horizontal well fracturing stage and cluster based on comprehensive evaluation of stress barrier and lithologic barrier should be worked out.展开更多
基金funded by the Laboratory for Marine Geology,Qingdao National Laboratory for Marine Science and Technology(No.MGQNLM-KF202004)Hainan Provincial Natural Science Foundation of China(Nos.422RC746 and 421QN281)+2 种基金the National Natural Science Foundation of China(No.42106213)the China Postdoctoral Science Foundation(Nos.2021M690161 and 2021T140691)the Postdoctorate Funded Project in Hainan Province.
文摘Hydrate reservoirs are different from the host reservoirs of all other fossil energy sources because the characteristics of hydrate reservoirs are generally controlled by deep-sea fine-grained sedimentation. In such reservoirs, the reliability of the classical logging evaluation models established for diagenetic reservoirs is questionable. This study used well W8 in the Qiongdongnan Basin to explore the clay content, porosity, saturation, and hydrate-enriched layer identification of a logging-based hydrate reservoir, and it was found that considering the effect of the clay content on the log response is necessary in the logging evaluation of hydrate reservoirs. In the evaluation of clay content, a method based on the optimization inversion method can obtain a more reliable clay content than other methods. Fine-grained sediment reservoirs have a high clay content, and the effect of clay on log responses must be considered when calculating porosity. In addition, combining density logging and neutron porosity logging data can obtain the best porosity calculation results, and the porosity calculation method based on sonic logging predicted that the porosity of the studied reservoir was low. It was very effective to identify hydrate layers based on resistivity, but the clay distribution and pore structure will also affect the relationship between resistivity, porosity and saturation, and it was suggested that the factors effecting the resistivity of different layers should be considered in the saturation evaluation and that a suitable model should be selected. This study also considered the lack of clarity of the relationships among the lithology, physical properties, hydrate-bearing occurrence properties, and log response properties of hydrate reservoirs and the lack of specialized petrophysical models. This research can directly help to improve hydrate logging evaluation.
文摘It is always significant for assessing and evaluation of oil bearing layers, especially for well logging data processing and interpretation of non marine oil beds to get more accurate physical properties in thin and inter thin layers. This paper presents a definition of measures and the measure presents power law relation with the corresponded scale described by fractal theory. Thus, logging curves can be reconstructed according to this power law relation. This method uses the local structure nearby concurrent points to compensate the average effect of logging probes and measurement errors. As an example, deep and medium induced conductivity (IMPH and IDPH) curves in ODP Leg 127 Hole 797C are reconstructed or corrected. Corrected curves are with less adjacent effects through comparison of corrected curves with original one. And also, the power spectra of corrected well logging curve are abounding with more resolution components than the original one. Thus, fractal correction method makes the well logging more resoluble for thin beds.
基金Project supported by National Natural Science Foundation of China (Grant Nos 10534040 and 10272038) and Doctorate Foundation of the State Education Ministry of China (Grant Nos 20040183045 and 20030183052).
文摘The nonaxisymmetric acousto-electric field excited by an eccentric acoustic source in the borehole based on Pride seismoelectric theory is considered. It is shown that the acoustic field inside the borehole, converted electric and magnetic fields and coupled fields outside the borehole are composed of an infinitude of multipole fields with different orders. The numerical results show that both the electromagnetic waves and the seismoelectric field in the borehole, and the three components of both electric field and magnetic field can be detected. Measurements on the borehole axis will be of advantage to determining shear velocity information. The components of the symmetric and nonsymmetric acoustic and electromagnetic fields can be strengthened or weakened by adding or subtracting the two full waveforms logged in some azimuths. It may be a new method of directly measuring the shear wave velocity by using the borehole seismoelectric effect.
基金This work was supported by the National Natural Science Foundation of China(60472019).
文摘Electromagnetic Computer Tomography (ECT) is a method to probe the interior of an inhomogeneous medium via surface measurement in a non-linear way. Due to the great differences in conductivity and permittivity between oil and water in the well, Electromagnetic Tomography Well Logging (ETWL), a new flow imaging measurement system, is proposed to describe the distribution and movement of oil/water two-phase flow in the well by scanning the detected region and applying a suitable data processing algorithm. The results of the numerical simulation and physical modeling show that the system could provide a clear image of the flow profile.
基金Supported by the National Science and Technology Major Project(2016ZX05007-004)。
文摘Based on the well logging knowledge graph of hydrocarbon-bearing formation(HBF),a Knowledge-Powered Neural Network Formation Evaluation model(KPNFE)has been proposed.It has the following functions:(1)extracting characteristic parameters describing HBF in multiple dimensions and multiple scales;(2)showing the characteristic parameter-related entities,relationships,and attributes as vectors via graph embedding technique;(3)intelligently identifying HBF;(4)seamlessly integrating expertise into the intelligent computing to establish the assessment system and ranking algorithm for potential pay recommendation.Taking 547 wells encountered the low porosity and low permeability Chang 6 Member of Triassic in the Jiyuan Block of Ordos Basin,NW China as objects,80%of the wells were randomly selected as the training dataset and the remainder as the validation dataset.The KPNFE prediction results on the validation dataset had a coincidence rate of 94.43%with the expert interpretation results and a coincidence rate of 84.38%for all the oil testing layers,which is 13 percentage points higher in accuracy and over 100 times faster than the primary conventional interpretation.In addition,a number of potential pays likely to produce industrial oil were recommended.The KPNFE model effectively inherits,carries forward and improves the expert knowledge,nicely solving the robustness problem in HBF identification.The KPNFE,with good interpretability and high accuracy of computation results,is a powerful technical means for efficient and high-quality well logging re-evaluation of old wells in mature oilfields.
基金This project was funded by the Open Fund of the Key Laboratory of Exploration Technologies for Oil and Gas Resources,the Ministry of Education(No.K2021-03)National Natural Science Foundation of China(No.42106213)+2 种基金the Hainan Provincial Natural Science Foundation of China(No.421QN281)the China Postdoctoral Science Foundation(Nos.2021M690161 and 2021T140691)the Postdoctorate Funded Project in Hainan Province.
文摘The total organic carbon content usually determines the hydrocarbon generation potential of a formation.A higher total organic carbon content often corresponds to a greater possibility of generating large amounts of oil or gas.Hence,accurately calculating the total organic carbon content in a formation is very important.Present research is focused on precisely calculating the total organic carbon content based on machine learning.At present,many machine learning methods,including backpropagation neural networks,support vector regression,random forests,extreme learning machines,and deep learning,are employed to evaluate the total organic carbon content.However,the principles and perspectives of various machine learning algorithms are quite different.This paper reviews the application of various machine learning algorithms to deal with total organic carbon content evaluation problems.Of various machine learning algorithms used for TOC content predication,two algorithms,the backpropagation neural network and support vector regression are the most commonly used,and the backpropagation neural network is sometimes combined with many other algorithms to achieve better results.Additionally,combining multiple algorithms or using deep learning to increase the number of network layers can further improve the total organic carbon content prediction.The prediction by backpropagation neural network may be better than that by support vector regression;nevertheless,using any type of machine learning algorithm improves the total organic carbon content prediction in a given research block.According to some published literature,the determination coefficient(R^(2))can be increased by up to 0.46 after using machine learning.Deep learning algorithms may be the next breakthrough direction that can significantly improve the prediction of the total organic carbon content.Evaluating the total organic carbon content based on machine learning is of great significance.
基金supported by the Fundamental Research Funds for the Central Universities(Grant No.2011YYL128)the CNPC Innovation Foundation(GrantNo.2012D-5006-0103)the Ministry of Land and Resources special funds for scientific research on public cause(Grant No.201311107)
文摘In order to solve the problems of the fine division of sedimentary sequence cycles and their change in two-dimensional space as well as lateral extension contrast, we developed a method of wavelet depth-frequency analysis. The single signal and composite signal of different Milankovitch cycles are obtained by numerical simulation. The simulated composite signal can be separated into single signals of a single frequency cycle. We also develop a well-seismic calibration insertion technology which helps to realize the calibration from the spectrum characteristics of a single well to the seismic profile. And then we determine the change and distribution characteristics of spectrum cycles in the two-dimensional space. It points out the direction in determining the variations of the regional sedimentary sequence cycles, underground strata structure and the contact relationship.
基金funded by the Strategic Cooperation Technology Projects of CNPC and CUPB(ZLZX2020-03)
文摘In the early exploration of many oilfields,low-resistivity-low-contrast(LRLC)pay zones are easily overlooked due to the resistivity similarity to the water zones.Existing identification methods are model-driven and cannot yield satisfactory results when the causes of LRLC pay zones are complicated.In this study,after analyzing a large number of core samples,main causes of LRLC pay zones in the study area are discerned,which include complex distribution of formation water salinity,high irreducible water saturation due to micropores,and high shale volume.Moreover,different oil testing layers may have different causes of LRLC pay zones.As a result,in addition to the well log data of oil testing layers,well log data of adjacent shale layers are also added to the original dataset as reference data.The densitybased spatial clustering algorithm with noise(DBSCAN)is used to cluster the original dataset into 49 clusters.A new dataset is ultimately projected into a feature space with 49 dimensions.The new dataset and oil testing results are respectively treated as input and output to train the multi-layer perceptron(MLP).A total of 3192 samples are used for stratified 8-fold cross-validation,and the accuracy of the MLP is found to be 85.53%.
文摘Since gas hydrate exists in three different forms at the same time such as pore filling,particle support and separate stratification,the calculation method of hydrate saturation using traditional shaly sand formation interpretation models is equivalent to considering only the simple case that hydrate exists as pore filling,and does not consider other complex states.Based on the analysis of hydrate resistivity experimental data and the general form of the resistivity-oil(gas)saturation relationship,the best simplified formula of hydrate saturation calculation is derived,then the physical meaning of the three items are clarified:they respectively represent the resistivity index-saturation relationship when hydrate particles are completely distributed in the pores of formation rocks,supported in the form of particles,and exist in layers,corresponding quantitative evaluation method of hydrate saturation is built.The field application shows that the hydrate saturation calculated by this method is closer to that obtained by sampling analysis.At the same time,it also provides a logging analysis basis for the effective development after hydrate exploration.
文摘This work concerns the collecting field of the Abrobakro site, the objective of which is to determine the thickness of the layers crossed during drilling from electrical logging in order to propose the equipment plan for the various boreholes. The electrical logging data sheets, particularly those on resistivity and expeditious granulometry using a 1.25 mm and 2 mm mesh sieve, were used. The layer thicknesses are determined with the inflection points on the graphs. The electrical logging shows that the sands in the study area have resistivity values between 400 and 5000 Ω.m. The decrease in resistivity observed at 50 m for all boreholes shows that the static level of the groundwater is at this depth. The results of the accelerated granulometry show that the first 20 meters contain more fine particles and coarse to very coarse sands from 20 m. The granulometry of the screen laying areas shows that the 1.5 mm slot openings are best suited for all drilling in the Abrobakro collecting field. The diameter d10 of the aquiferous sands of the collecting field is close to 1.25 mm.
基金the financial support provided by the National Natural Science Foundation of China(Grant Nos.42174133 and 41676032)China Geological Survey(Grant No.DD20190234)。
文摘Electrical properties are important physical parameters of natural gas hydrate,and,specifically,resistivity has been widely used in the quantitative estimation of hydrate saturation.There are three main methods to study the electrical properties of gas hydrate-bearing sediments:experimental laboratory measurements,numerical simulation,and resistivity logging.Experimental measurements can be divided into three categories:normal electrical measurement,complex resistivity measurement,and electrical resistivity tomography.Experimental measurements show that the resistivity of hydrate-bearing sediment is affected by many factors,and its distribution as well as the hydrate saturation is not uniform;there is a distinct non-Archie phenomenon.The numerical method can simulate the resistivity of sediments by changing the hydrate occurrence state,saturation,distribution,etc.However,it needs to be combined with X-ray CT,nuclear magnetic resonance,and other imaging techniques to characterize the porous characteristics of the hydrate-bearing sediments.Resistivity well logging can easily identify hydrate layers based on their significantly higher resistivity than the background,but the field data of the hydrate layer also has a serious non-Archie phenomenon.Therefore,more experimental measurements and numerical simulation studies are needed to correct the parameters of Archie’s formula.
基金The research was carried out within the framework of the grant for state support of leading scientific schools of the Russian Federation(Grant No.NSh-1010.2022.1.5).
文摘Survey and novel research data are used in the present study to classify/identify the lithological type of Verey age reservoirs’rocks.It is shown how the use of X-ray tomography can clarify the degree of heterogeneity,porosity and permeability of these rocks.These data are then used to elaborate a model of hydraulic fracturing.The resulting software can take into account the properties of proppant and breakdown fluid,thermal reservoir conditions,oil properties,well design data and even the filtration and elastic-mechanical properties of the rocks.Calculations of hydraulic fracturing crack formation are carried out and the results are compared with the data on hydraulic fracturing crack at standard conditions.Significant differences in crack formation in standard and lithotype models are determined.It is shown that the average width of the crack development for the lithotype model is 2.3 times higher than that for the standard model.Moreover,the coverage of crack development in height for the lithotype model is almost 2 times less than that for the standard model.The estimated fracture half-length for the lithotype model is 13.3%less than that of for the standard model.A higher dimensionless fracture conductivity is also obtained for the lithotype model.It is concluded that the proposed approach can increase the reliability of hydraulic fracturing crack models.
文摘A new multi-exponential inversion method for NMR relaxation signals is presented and tested, which is based on a solid iteration rebuild technique (SIRT). The T2 spectra inversed by the new method are compared with MAP-II results. The T1 and T2 inversion results with different pre-assigned relaxation times and different SNR show that 16 to 64 logarithm equal spaced time constants is better obviously than MAP-II. And in particular, it can ensure the relaxation time distribution when the SNR of the measured signal is very low. The new algorithm has been applied in rock core NMR analysis and NMR logging data process and interpretation.
基金supported by the National Natural ScienceFoundation of China(Grant Nos.10534040,40574049 and 40874097)the Research Fund for the Doctoral Programof Higher Education(Grant No.20070425028)the Foundation of State Key Laboratory of Petroleum Resourceand Prospecting,China University of Petroleum(Grant No.PRPDX2008-08).
文摘The acoustic tools widely used in borehole well logging and being developed in borehole acoustic reflection imaging do not have the function of azimuthal measurement due to a symmetric source, so they can not be used to evaluate the azimuthal character of borehole formation. In this paper, a 3D finite difference method was used to simulate the acoustic fields in a fluid-filled borehole generated by a traditional monopole source and a phased arc array. Acoustic waveforms were presented for both cases. The analysis of the simulated waveforms showed that different from the monopole source, the acoustic energy generated by the phased arc array transmitter mainly radiated to the borehole in a narrow azimuthal range, which was the key technique to implement azimuthal acoustic well logging. Similar to the monopole source, the waveforms generated by the phased arc array in the fluid-filled borehole also contain compressional (P) waves and shear (S) waves refracted along the borehole, which is the theoretical foundation of phased arc array acoustic well logging.
基金supported by the PetroChina Xinjiang Oilfield Exploration & Production Research Institute
文摘The Hutubi (呼图壁) River reservoir of well block DX12 is a lithologic hydrocarbon reservoir that is under tectonic settings. The main oil-bearing sand body in this area is thin and has a poor transverse connectivity. Because of the complexity of the oil-water relationship, the oil reservoir presents a low resistivity feature, which brings great difficulties to hydrocarbon reservoir identification. This article develops an effective method of well log interpretation that can meet the requirement of low resistivity reservoir well logging evaluation. The authors combine the oil reservoir geology feature, the oil well logging curve characteristics and chemical analytical data to analyze the reasons for low resistivity, then establish the appropriate reservoir parameter explanation model, which uses different saturation computational methods according to different generations. When the clay content is more than 5%, we select W-S dual water model; when the shale content is more than 13%, we use the Schlumberger formula; when the shale content is less then 13%, we use Archie's formula. The well logging evaluation method of low resistivity reservoir has been improved by the irreducible water saturation formula which is established by the permeability, the porosity, the coefficient of pore structure and the shale content, hydrocarbon reservoir recognition charts, and the non-resistivity logging methods (repeat formation test (RFT); modular dynamic test (MDT), etc.). The coincidence rate for this arrangement of the well logging integrated interpretation is 82.6% in the well block DX12. It is a powerful direction for low resistivity well log interpretation.
基金supported by the Special Fund for Basic Scientific Research of Central Colleges (No. CUG090106)the National Basic Research Program of China (No. 2003CB716500)
文摘The lithologies of the Chinese Continental Scientific Drilling main hole (CCSD-MH) are mainly comprised of orthogneiss,paragneiss,eclogite,amphibolite,and ultramafic rocks.The statistical results of logs of CCSD-MH indicate that ultramafic rocks are characterized by very high CNL (neutron log) and very low GR (gamma ray log) and RD (resistivity log);eclogites are characterized by high DEN (density),VP (P-wave velocity) and PE (photoelectric absorption capture cross section);orthogneiss and paragneiss are characterized by high GR,U (uranium content),Th (thorium content),K (potassium content) and RD,and low DEN,PE,and CNL;logging values of amphibolite are between the logging values of eclogites and paragneiss.In addition,the logs could reflect the degree of retrograde metamorphism of eclogites.The upper section (100-2 000 m) shows higher DEN,PE,VP,and lower GR,U,Th,K,RD than the lower section (2 000-5 000 m).Most logs of the upper section are more fluctuant than those of the lower section.This indicates that the upper section has more heterogeneities than the lower section.The cross plots of logs indicate that DEN,GR,K,and CNL are more powerful in identifying ultrahigh pressure metamorphic (UHPM) rocks at the CCSD-MH.GR value of the rocks from CCSD-MH shows obviously an increasing trend from ultramafic rock (the most mafic rocks at CCSD-MH) to orthogneiss (the most acid rocks at CCSD-MH).On the contrary,DEN value decreases from the ultramafic rock to the orthogneiss.CNL log is a good indicator of the content of structure water in crystalline rocks.
文摘To promote adaptation of logging evaluation technologies to the development trend of unconventional oil and gas exploration and development era in China,the current situation and challenges of logging evaluation technologies in China are analyzed systematically.Based on the concept of that demand drives technology development,and referring to the international leading technologies,development strategy of logging evaluation technology in China has been put forward.(1)Deepen petrophysics study:mobile 2 D NMR laboratory analysis technology for full diameter core should be developed,characteristic charts and evaluation standards of different fluid properties,different pore structures and different core exposure times should be established based on longitudinal and traverse relaxation spectra;in-depth digital rock experiment and mathematical and physical simulation research should be carried out to create innovative logging evaluation methods;acoustic and electrical anisotropy experimental analysis technology should be developed,and corresponding logging evaluation methods be innovated.(2)Strengthen target processing of logging data:precise inversion processing technology and sensitive information extraction technology of 2 D NMR logging should be developed to finely describe the micro-pore distribution in tight reservoir and accurately distinguish movable oil,bound oil,and bound water etc.The processing method of 3 D ultra-distance detection acoustic logging should be researched.(3)Develop special logging interpretation and evaluation methods:first,mathematical model for quantitatively describing the saturation distribution law of unconventional oil and gas near source and in source should be created;second,evaluation methods and standards of shale oil and deep shale gas"sweet-spots"with mobile oil content and gas content as key parameter separately should be researched vigorously;third,calculation methods of pore pressure under two high-pressure genetic mechanisms,under-compaction and hydrocarbon charging,should be improved;fourth,evaluation method of formation fracability considering the reservoir geologic and engineering quality,and optimization method of horizontal well fracturing stage and cluster based on comprehensive evaluation of stress barrier and lithologic barrier should be worked out.