Manually picking regularly and densely distributed first breaks(FBs)are critical for shallow velocitymodel building in seismic data processing.However,it is time consuming.We employ the fullyconvolutional Seg Net to a...Manually picking regularly and densely distributed first breaks(FBs)are critical for shallow velocitymodel building in seismic data processing.However,it is time consuming.We employ the fullyconvolutional Seg Net to address this issue and present a fast automatic seismic waveform classification method to pick densely-sampled FBs directly from common-shot gathers with sparsely distributed traces.Through feeding a large number of representative shot gathers with missing traces and the corresponding binary labels segmented by manually interpreted fully-sampled FBs,we can obtain a welltrained Seg Net model.When any unseen gather including the one with irregular trace spacing is inputted,the Seg Net can output the probability distribution of different categories for waveform classification.Then FBs can be picked by locating the boundaries between one class on post-FBs data and the other on pre-FBs background.Two land datasets with each over 2000 shots are adopted to illustrate that one well-trained 25-layer Seg Net can favorably classify waveform and further pick fully-sampled FBs verified by the manually-derived ones,even when the proportion of randomly missing traces reaches50%,21 traces are missing consecutively,or traces are missing regularly.展开更多
An important application of spectral decomposition(SD)is to identify subsurface geological anomalies such as channels and karst caves,which may be buried in full-band seismic data.However,the classical SD methods incl...An important application of spectral decomposition(SD)is to identify subsurface geological anomalies such as channels and karst caves,which may be buried in full-band seismic data.However,the classical SD methods including the wavelet transform(WT)are often limited by relatively low time-frequency resolution,which is responsible for false high horizonassociated space resolution probably indicating more geological structures,especially when close geological anomalies exist.To address this issue,we impose a constraint of minimizing an lp(0<p<1)norm of time-frequency spectral coefficients on the misfit derived by using the inverse WT and apply the generalized iterated shrinkage algorithm to invert for the optimal coefficients.Compared with the WT and inverse SD(ISD)using a typical l1-norm constraint,the modified ISD(MISD)using an lp-norm constraint can yield a more compact spectrum contributing to detect the distributions of close geological features.We design a 3 D synthetic dataset involving frequency-close thin geological anomalies and the other3 D non-stationary dataset involving time-close anomalies to demonstrate the effectiveness of MISD.The application of 4 D spectrum on a 3 D real dataset with an area of approximately 230 km2 illustrates its potential for detecting deep channels and the karst slope fracture zone.展开更多
Offshore exploration in Norway and Denmark-in the North Sea, the Norwegian Sea and the Barents Sea-has involved drilling about 850 wildcat wells, resulting in about 300 oil and gas finds, of which 84 are fields with p...Offshore exploration in Norway and Denmark-in the North Sea, the Norwegian Sea and the Barents Sea-has involved drilling about 850 wildcat wells, resulting in about 300 oil and gas finds, of which 84 are fields with production. The recoverable resources of all these finds total about 65 billion barrels of oil equivalent. Almost all these hydrocarbons come from a Jurassic source and the main reservoirs and traps are Jurassic sandstones in fault blocks and Paleocene sandstones or Cretaceous chalks in gentle domes. The article describes four major fields-Ekofisk, Gullfaks, Ormen Lange and SnФhvitto illustrate some of the many challenges in developing and producing the hydrocarbons. Elsewhere in Norden, there has been much less exploration. Drilling results have mostly been negative in mainland Sweden, onshore Denmark, onshore Svalbard and on- and offshore West Greenland. Minor oil finds have been made in Palaeozoic rocks in the Baltic Sea. The first wells have recently been drilled off the Faroe Islands, resulting in one discovery. No drilling has taken place on- or offshore East Greenland. As a result of the hydrocarbon activities in Norway and Denmark, petroleum geoscience there has flourished, with 2000 geoscientists currently employed in the industry, many technical innovations made, a wealth of publically available information and a great increase in the understanding of the geology.展开更多
The Devonian Woodford Shale in the Anadarko Basin is a highly organic,hydrocarbon source rock.Accurate values of vitrinite reflectance(R_o)present in the Woodford Shale penetrated by 52 control wells were measured dir...The Devonian Woodford Shale in the Anadarko Basin is a highly organic,hydrocarbon source rock.Accurate values of vitrinite reflectance(R_o)present in the Woodford Shale penetrated by 52 control wells were measured directly.These vitrinite reflectance values,when plotted against borehole resistivity for the middle member of the Woodford Shale in the wells,display a rarely reported finding that deep resistivity readings decrease as R_o increases when R_o is greater than 0.90%.This phenomenon may be attributed to that aromatic and resin compounds containing conjugated pi bonds generated within source rocks are more electrically conductive than aliphatic compounds.And aromatic and resin fractions were generated more than aliphatic fraction when source rock maturity further increases beyond oil peak.The finding of the relationship between deep resistivity and R_o may re-investigate the previously found linear relationship between source rock formation and aid to unconventional play exploration.展开更多
Demand and shortage of oil in China Rapid economic development requires huge energy Since the early 1980s, China has consistently been the most rapidly growing economy on earth, sustaining an average annual growth
Hydraulic fracturing(HF)is a commonly used technique to stimulate low permeability formations such as shale plays and tight formations.However,this method of well stimulation has also been used in high permeable uncon...Hydraulic fracturing(HF)is a commonly used technique to stimulate low permeability formations such as shale plays and tight formations.However,this method of well stimulation has also been used in high permeable unconsolidated sandstone formations to bypass near-wellbore formation damage and prevent sand production at some distance apart from the wellbore wall.The treatment is called frac-pack completion,where a short length but wide width fracture is formed by injecting aggressive concentrations of proppant into the fracture plane.This operation is known as tip screen-out(TSO).Detailed design of fluid and proppant,including an optimal pump schedule,is required to achieve satisfactory TSO.In this study,we first assess the lattice-based numerical method's capabilities for simulating hydraulic fracturing propagation in elastoplastic formation.The results will be compared with the same case simulation results using a pseudo 3D(P3D)model and analytical model.Second,we explore the Nolte(1986)design for frac-pack and TSO treatment using lattice-based software and the P3D model.The results showed that both models could simulate the hydraulic fracturing propagation in soft formation and TSO operation,while some differences were observed in generated geometry,the tip screenout time and net pressure profiles.The results are presented.It was noted that fracture propagation regime(viscosity/toughness),nonlocality and nonlinearity had an influence on the different geometries.The advantages of each model will be discussed.展开更多
Borehole acoustic logging-while-drilling (LWD) for formation evaluationhas become an indispensable part of hydrocarbon reservoir assessment [F. Citt ´a, C. Rus-sell, R. Deady and D. Hinz, The Leading Edge, 23 (20...Borehole acoustic logging-while-drilling (LWD) for formation evaluationhas become an indispensable part of hydrocarbon reservoir assessment [F. Citt ´a, C. Rus-sell, R. Deady and D. Hinz, The Leading Edge, 23 (2004), pp. 566-573]. However,the detection of acoustic formation arrivals over tool mode contamination has beena challenging problem in acoustic LWD technology. In this paper we propose a newmethod for separating tool waves from formation acoustic waves in acoustic LWD.This method is to measure the seismoelectric signal excited by the LWD acoustic waves.The LWD tool waves which propagate along the rigid tool rim can not excite any elec-tric signal. This is due to the effectively grounding of the drill string during the LWDprocess makes it impossible to accumulate any excess charge at the conductive tool —borehole fluid interface. Therefore, there should be no contribution by the tool modesto the recorded seismoelectric signals. To theoretically understand the seismoelectricconversion in the LWD geometry, we calculate the synthetic waveforms for the multi-pole LWD seismoelectric signals based on Pride’s theory [S. R. Pride, Phys. Rev. B, 50(1994), pp. 15678-15696]. The synthetic waveforms for the electric field induced by theLWD-acoustic-wave along the borehole wall demonstrate the absence of the tool mode.We also designed the laboratory experiments to collect simulated LWD monopole anddipole acoustic and seismoelectric signals in a borehole in sandstone. By analyzing thespectrum of acoustic and electric signals, we can detect and filter out the differencebetween the two signals, which are the mainly tool modes and noise.展开更多
We present theoretical and experimental studies on the effects of formationproperties on seismoelectric conversions in fluid-filled boreholes. First, we derive thetheoretical formulations for seismoelectric responses ...We present theoretical and experimental studies on the effects of formationproperties on seismoelectric conversions in fluid-filled boreholes. First, we derive thetheoretical formulations for seismoelectric responses for an acoustic source in a borehole.Then, we compute the electric fields in boreholes penetrating formations withdifferent permeability and porosity, and then we analyze the sensitivity of the convertedelectric fields to formation permeability and porosity. We also describe the laboratoryresults of the seismoelectric and seismomagnetic fields induced by an acousticsource in borehole models to confirm our theoretical and numerical developmentsqualitatively. We use a piezoelectric transducer to generate acoustic waves and a pointelectrode to receive the localized seismoelectric fields in layered boreholes and theelectric component of electromagnetic waves in a fractured borehole model. Numericalresults show that the magnitude ratio of the converted electric wave to the acousticpressure increases with the porosity and permeability increases in both fast and slowformations. Furthermore, the converted electric signal is sensitive to the formationpermeability for the same source frequency and formation porosity. Our experimentsvalidate our theoretical results qualitatively. An acoustic wave at a fracture intersectinga borehole induces a radiating electromagnetic wave.展开更多
基金financially supported by the National Key R&D Program of China(2018YFA0702504)the Fundamental Research Funds for the Central Universities(2462019QNXZ03)+1 种基金the National Natural Science Foundation of China(42174152 and 41974140)the Strategic Cooperation Technology Projects of CNPC and CUPB(ZLZX 2020-03)。
文摘Manually picking regularly and densely distributed first breaks(FBs)are critical for shallow velocitymodel building in seismic data processing.However,it is time consuming.We employ the fullyconvolutional Seg Net to address this issue and present a fast automatic seismic waveform classification method to pick densely-sampled FBs directly from common-shot gathers with sparsely distributed traces.Through feeding a large number of representative shot gathers with missing traces and the corresponding binary labels segmented by manually interpreted fully-sampled FBs,we can obtain a welltrained Seg Net model.When any unseen gather including the one with irregular trace spacing is inputted,the Seg Net can output the probability distribution of different categories for waveform classification.Then FBs can be picked by locating the boundaries between one class on post-FBs data and the other on pre-FBs background.Two land datasets with each over 2000 shots are adopted to illustrate that one well-trained 25-layer Seg Net can favorably classify waveform and further pick fully-sampled FBs verified by the manually-derived ones,even when the proportion of randomly missing traces reaches50%,21 traces are missing consecutively,or traces are missing regularly.
基金financially supported by the National Key R&D Program of China(2018YFA0702504)the Fundamental Research Funds for the Central Universities(2462019QNXZ03)+2 种基金the Scientific Research and Technology Development Project of China National Petroleum Corporation(2017D-3504)the Major Scientific Research Program of Petrochina Science and Technology Management Department"Comprehensive Seismic Prediction Technology and Software Development of Natural Gas"(2019B-0607)the National Science and Technology Major Project(2017ZX05005-004)。
文摘An important application of spectral decomposition(SD)is to identify subsurface geological anomalies such as channels and karst caves,which may be buried in full-band seismic data.However,the classical SD methods including the wavelet transform(WT)are often limited by relatively low time-frequency resolution,which is responsible for false high horizonassociated space resolution probably indicating more geological structures,especially when close geological anomalies exist.To address this issue,we impose a constraint of minimizing an lp(0<p<1)norm of time-frequency spectral coefficients on the misfit derived by using the inverse WT and apply the generalized iterated shrinkage algorithm to invert for the optimal coefficients.Compared with the WT and inverse SD(ISD)using a typical l1-norm constraint,the modified ISD(MISD)using an lp-norm constraint can yield a more compact spectrum contributing to detect the distributions of close geological features.We design a 3 D synthetic dataset involving frequency-close thin geological anomalies and the other3 D non-stationary dataset involving time-close anomalies to demonstrate the effectiveness of MISD.The application of 4 D spectrum on a 3 D real dataset with an area of approximately 230 km2 illustrates its potential for detecting deep channels and the karst slope fracture zone.
文摘Offshore exploration in Norway and Denmark-in the North Sea, the Norwegian Sea and the Barents Sea-has involved drilling about 850 wildcat wells, resulting in about 300 oil and gas finds, of which 84 are fields with production. The recoverable resources of all these finds total about 65 billion barrels of oil equivalent. Almost all these hydrocarbons come from a Jurassic source and the main reservoirs and traps are Jurassic sandstones in fault blocks and Paleocene sandstones or Cretaceous chalks in gentle domes. The article describes four major fields-Ekofisk, Gullfaks, Ormen Lange and SnФhvitto illustrate some of the many challenges in developing and producing the hydrocarbons. Elsewhere in Norden, there has been much less exploration. Drilling results have mostly been negative in mainland Sweden, onshore Denmark, onshore Svalbard and on- and offshore West Greenland. Minor oil finds have been made in Palaeozoic rocks in the Baltic Sea. The first wells have recently been drilled off the Faroe Islands, resulting in one discovery. No drilling has taken place on- or offshore East Greenland. As a result of the hydrocarbon activities in Norway and Denmark, petroleum geoscience there has flourished, with 2000 geoscientists currently employed in the industry, many technical innovations made, a wealth of publically available information and a great increase in the understanding of the geology.
基金funded by the Foundation of State Key Laboratory of Petroleum Resources and Prospecting,China University of Petroleum,Beijing(No.PRP/open-1605)partly supported by the Open Fund of Key Laboratory of Exploration Technologies for Oil and Gas Resources(Yangtze University),Ministry of Education(No.K2017-18)Tight Oil Enrichment and Key Exploration and Development Technology Project of National Science and Technology Major Project(Nos.2016ZX05046-002 and 2016ZX05047-005)
文摘The Devonian Woodford Shale in the Anadarko Basin is a highly organic,hydrocarbon source rock.Accurate values of vitrinite reflectance(R_o)present in the Woodford Shale penetrated by 52 control wells were measured directly.These vitrinite reflectance values,when plotted against borehole resistivity for the middle member of the Woodford Shale in the wells,display a rarely reported finding that deep resistivity readings decrease as R_o increases when R_o is greater than 0.90%.This phenomenon may be attributed to that aromatic and resin compounds containing conjugated pi bonds generated within source rocks are more electrically conductive than aliphatic compounds.And aromatic and resin fractions were generated more than aliphatic fraction when source rock maturity further increases beyond oil peak.The finding of the relationship between deep resistivity and R_o may re-investigate the previously found linear relationship between source rock formation and aid to unconventional play exploration.
文摘Demand and shortage of oil in China Rapid economic development requires huge energy Since the early 1980s, China has consistently been the most rapidly growing economy on earth, sustaining an average annual growth
文摘Hydraulic fracturing(HF)is a commonly used technique to stimulate low permeability formations such as shale plays and tight formations.However,this method of well stimulation has also been used in high permeable unconsolidated sandstone formations to bypass near-wellbore formation damage and prevent sand production at some distance apart from the wellbore wall.The treatment is called frac-pack completion,where a short length but wide width fracture is formed by injecting aggressive concentrations of proppant into the fracture plane.This operation is known as tip screen-out(TSO).Detailed design of fluid and proppant,including an optimal pump schedule,is required to achieve satisfactory TSO.In this study,we first assess the lattice-based numerical method's capabilities for simulating hydraulic fracturing propagation in elastoplastic formation.The results will be compared with the same case simulation results using a pseudo 3D(P3D)model and analytical model.Second,we explore the Nolte(1986)design for frac-pack and TSO treatment using lattice-based software and the P3D model.The results showed that both models could simulate the hydraulic fracturing propagation in soft formation and TSO operation,while some differences were observed in generated geometry,the tip screenout time and net pressure profiles.The results are presented.It was noted that fracture propagation regime(viscosity/toughness),nonlocality and nonlinearity had an influence on the different geometries.The advantages of each model will be discussed.
文摘Borehole acoustic logging-while-drilling (LWD) for formation evaluationhas become an indispensable part of hydrocarbon reservoir assessment [F. Citt ´a, C. Rus-sell, R. Deady and D. Hinz, The Leading Edge, 23 (2004), pp. 566-573]. However,the detection of acoustic formation arrivals over tool mode contamination has beena challenging problem in acoustic LWD technology. In this paper we propose a newmethod for separating tool waves from formation acoustic waves in acoustic LWD.This method is to measure the seismoelectric signal excited by the LWD acoustic waves.The LWD tool waves which propagate along the rigid tool rim can not excite any elec-tric signal. This is due to the effectively grounding of the drill string during the LWDprocess makes it impossible to accumulate any excess charge at the conductive tool —borehole fluid interface. Therefore, there should be no contribution by the tool modesto the recorded seismoelectric signals. To theoretically understand the seismoelectricconversion in the LWD geometry, we calculate the synthetic waveforms for the multi-pole LWD seismoelectric signals based on Pride’s theory [S. R. Pride, Phys. Rev. B, 50(1994), pp. 15678-15696]. The synthetic waveforms for the electric field induced by theLWD-acoustic-wave along the borehole wall demonstrate the absence of the tool mode.We also designed the laboratory experiments to collect simulated LWD monopole anddipole acoustic and seismoelectric signals in a borehole in sandstone. By analyzing thespectrum of acoustic and electric signals, we can detect and filter out the differencebetween the two signals, which are the mainly tool modes and noise.
文摘We present theoretical and experimental studies on the effects of formationproperties on seismoelectric conversions in fluid-filled boreholes. First, we derive thetheoretical formulations for seismoelectric responses for an acoustic source in a borehole.Then, we compute the electric fields in boreholes penetrating formations withdifferent permeability and porosity, and then we analyze the sensitivity of the convertedelectric fields to formation permeability and porosity. We also describe the laboratoryresults of the seismoelectric and seismomagnetic fields induced by an acousticsource in borehole models to confirm our theoretical and numerical developmentsqualitatively. We use a piezoelectric transducer to generate acoustic waves and a pointelectrode to receive the localized seismoelectric fields in layered boreholes and theelectric component of electromagnetic waves in a fractured borehole model. Numericalresults show that the magnitude ratio of the converted electric wave to the acousticpressure increases with the porosity and permeability increases in both fast and slowformations. Furthermore, the converted electric signal is sensitive to the formationpermeability for the same source frequency and formation porosity. Our experimentsvalidate our theoretical results qualitatively. An acoustic wave at a fracture intersectinga borehole induces a radiating electromagnetic wave.