Accurate prediction of formation pore pressure is essential to predict fluid flow and manage hydrocarbon production in petroleum engineering.Recent deep learning technique has been receiving more interest due to the g...Accurate prediction of formation pore pressure is essential to predict fluid flow and manage hydrocarbon production in petroleum engineering.Recent deep learning technique has been receiving more interest due to the great potential to deal with pore pressure prediction.However,most of the traditional deep learning models are less efficient to address generalization problems.To fill this technical gap,in this work,we developed a new adaptive physics-informed deep learning model with high generalization capability to predict pore pressure values directly from seismic data.Specifically,the new model,named CGP-NN,consists of a novel parametric features extraction approach(1DCPP),a stacked multilayer gated recurrent model(multilayer GRU),and an adaptive physics-informed loss function.Through machine training,the developed model can automatically select the optimal physical model to constrain the results for each pore pressure prediction.The CGP-NN model has the best generalization when the physicsrelated metricλ=0.5.A hybrid approach combining Eaton and Bowers methods is also proposed to build machine-learnable labels for solving the problem of few labels.To validate the developed model and methodology,a case study on a complex reservoir in Tarim Basin was further performed to demonstrate the high accuracy on the pore pressure prediction of new wells along with the strong generalization ability.The adaptive physics-informed deep learning approach presented here has potential application in the prediction of pore pressures coupled with multiple genesis mechanisms using seismic data.展开更多
In this paper, a new concept called numerical structure of seismic data is introduced and the difference between numerical structure and numerical value of seismic data is explained. Our study shows that the numerical...In this paper, a new concept called numerical structure of seismic data is introduced and the difference between numerical structure and numerical value of seismic data is explained. Our study shows that the numerical seismic structure is closely related to oil and gas-bearing reservoir, so it is very useful for a geologist or a geophysicist to precisely interpret the oil-bearing layers from the seismic data. This technology can be applied to any exploration or production stage. The new method has been tested on a series of exploratory or development wells and proved to be reliable in China. Hydrocarbon-detection with this new method for 39 exploration wells on 25 structures indi- cates a success ratio of over 80 percent. The new method of hydrocarbon prediction can be applied for: (1) depositional environment of reservoirs with marine fades, delta, or non-marine fades (including fluvial facies, lacustrine fades); (2) sedimentary rocks of reservoirs that are non-marine clastic rocks and carbonate rock; and (3) burial depths range from 300 m to 7000 m, and the minimum thickness of these reservoirs is over 8 m (main frequency is about 50 Hz).展开更多
Seismic data structure characteristics means the waveform character arranged in the time sequence at discrete data points in each 2-D or 3-D seismic trace. Hydrocarbon prediction using seismic data structure character...Seismic data structure characteristics means the waveform character arranged in the time sequence at discrete data points in each 2-D or 3-D seismic trace. Hydrocarbon prediction using seismic data structure characteristics is a new reservoir prediction technique. When the main pay interval is in carbonate fracture and fissure-cavern type reservoirs with very strong inhomogeneity, there are some difficulties with hydrocarbon prediction. Because of the special geological conditions of the eighth zone in the Tahe oil field, we apply seismic data structure characteristics to hydrocarbon prediction for the Ordovician reservoir in this zone. We divide the area oil zone into favorable and unfavorable blocks. Eighteen well locations were proposed in the favorable oil block, drilled, and recovered higher output of oil and gas.展开更多
Conventional time-space domain and frequency-space domain prediction filtering methods assume that seismic data consists of two parts, signal and random noise. That is, the so-called additive noise model. However, whe...Conventional time-space domain and frequency-space domain prediction filtering methods assume that seismic data consists of two parts, signal and random noise. That is, the so-called additive noise model. However, when estimating random noise, it is assumed that random noise can be predicted from the seismic data by convolving with a prediction error filter. That is, the source-noise model. Model inconsistencies, before and after denoising, compromise the noise attenuation and signal-preservation performances of prediction filtering methods. Therefore, this study presents an inversion-based time-space domain random noise attenuation method to overcome the model inconsistencies. In this method, a prediction error filter （PEF）, is first estimated from seismic data; the filter characterizes the predictability of the seismic data and adaptively describes the seismic data＇s space structure. After calculating PEF, it can be applied as a regularized constraint in the inversion process for seismic signal from noisy data. Unlike conventional random noise attenuation methods, the proposed method solves a seismic data inversion problem using regularization constraint; this overcomes the model inconsistency of the prediction filtering method. The proposed method was tested on both synthetic and real seismic data, and results from the prediction filtering method and the proposed method are compared. The testing demonstrated that the proposed method suppresses noise effectively and provides better signal-preservation performance.展开更多
地震是极具破坏性与不确定性的自然现象,在人们毫无察觉的情况下地震发生在人口稠密区时,将严重危害人们生命财产安全。人们不断努力了解地震的物理特征和物理危害与环境之间的相互作用,以便在地震发生前发出适当的警报。可靠的地震预...地震是极具破坏性与不确定性的自然现象,在人们毫无察觉的情况下地震发生在人口稠密区时,将严重危害人们生命财产安全。人们不断努力了解地震的物理特征和物理危害与环境之间的相互作用,以便在地震发生前发出适当的警报。可靠的地震预测应包含对地震信号的分析,但是这些信号在地震发生前不明显;因此使用数据驱动机器学习的方法来分析这些信号与地震的联系并预测地震。通过建立观测台网连续监测与地震发生相关的各种物理量或化学量,据此获取的地震前兆信息是地震预测的研究基础。地震发生前,地球物理场发生显著变化,伴随电磁和地声等多种前兆信号,其中电磁和地声信号具有临震特性,是开展地震临震观测预测研究的重要数据来源;因此对地下的电磁扰动和地声信号进行实时监测,获取长期观测数据用于数据驱动机器学习方法预测地震。该文基于AETA数据的临震模型预报,针对多分量地震监测预测系统(Acoustic and Electromagnetic Testing All in one system,AETA)在川滇地区记录的电磁和地声数据,提取时域和频域特征,采用基于随机森林算法、轻量级梯度提升决策树和极度随机树的集成学习方法共同预测该区域的发震情况,选取发震概率最大的子区域中心位置作为震中预测结果,进一步训练LightGBM回归模型以预测此子区域的震级,按周对地震三要素进行预测。实验结果表明,该方法在川滇地区地震风险预测上,准确率可达0.64,震级预测的平均误差为0.38,最小误差为0.00,具有良好的预测效果。展开更多
Conventional machine learning(CML)methods have been successfully applied for gas reservoir prediction.Their prediction accuracy largely depends on the quality of the sample data;therefore,feature optimization of the i...Conventional machine learning(CML)methods have been successfully applied for gas reservoir prediction.Their prediction accuracy largely depends on the quality of the sample data;therefore,feature optimization of the input samples is particularly important.Commonly used feature optimization methods increase the interpretability of gas reservoirs;however,their steps are cumbersome,and the selected features cannot sufficiently guide CML models to mine the intrinsic features of sample data efficiently.In contrast to CML methods,deep learning(DL)methods can directly extract the important features of targets from raw data.Therefore,this study proposes a feature optimization and gas-bearing prediction method based on a hybrid fusion model that combines a convolutional neural network(CNN)and an adaptive particle swarm optimization-least squares support vector machine(APSO-LSSVM).This model adopts an end-to-end algorithm structure to directly extract features from sensitive multicomponent seismic attributes,considerably simplifying the feature optimization.A CNN was used for feature optimization to highlight sensitive gas reservoir information.APSO-LSSVM was used to fully learn the relationship between the features extracted by the CNN to obtain the prediction results.The constructed hybrid fusion model improves gas-bearing prediction accuracy through two processes of feature optimization and intelligent prediction,giving full play to the advantages of DL and CML methods.The prediction results obtained are better than those of a single CNN model or APSO-LSSVM model.In the feature optimization process of multicomponent seismic attribute data,CNN has demonstrated better gas reservoir feature extraction capabilities than commonly used attribute optimization methods.In the prediction process,the APSO-LSSVM model can learn the gas reservoir characteristics better than the LSSVM model and has a higher prediction accuracy.The constructed CNN-APSO-LSSVM model had lower errors and a better fit on the test dataset than the other individual models.This method proves the effectiveness of DL technology for the feature extraction of gas reservoirs and provides a feasible way to combine DL and CML technologies to predict gas reservoirs.展开更多
This study examines the turbidite dynamics and hydrocarbon reservoir formation in Ghana’s Tano Basin, which is located in coastal West Africa. Through an exploration of geological processes spanning millions of years...This study examines the turbidite dynamics and hydrocarbon reservoir formation in Ghana’s Tano Basin, which is located in coastal West Africa. Through an exploration of geological processes spanning millions of years, we uncover key factors shaping hydrocarbon accumulation, including source rock richness, temperature, pressure, and geological structures. The research offers valuable insights applicable to exploration, management, and sustainable resource exploitation in coastal West Africa. It facilitates the identification of exploration targets with higher hydrocarbon potential, enables the anticipation of reservoir potential within the Tano Basin, and assists in tailoring exploration and management strategies to specific geological conditions of the Tano Basin. Analysis of fluvial channels sheds light on their impact on landscape formation and hydrocarbon exploration. The investigation into turbidite systems unveils intricate interactions involving tectonics, sea-level fluctuations, and sedimentation patterns, influencing the development of reservoirs. An understanding of sediment transport and depositional settings is essential for efficient reservoir management. Geomorphological features, such as channels, submarine canyons, and distinct channel types, are essential in this situation. A detailed examination of turbidite channel structures, encompassing canyons, channel complexes, convex channels, and U-shaped channels, provides valuable insights and aids in identifying exploration targets like basal lag, channel levees, and lobes. These findings underscore the enduring significance of turbidite systems as conduits for sediment transport, contributing to enhanced reservoir management and efficient hydrocarbon production. The study also highlights how important it is to examine the configuration of sedimentary layers, stacking patterns, and angular laminated facies to identify turbidites, understand reservoir distribution, and improve well design. The dynamic nature of turbidite systems, influenced by basin characteristics such as shape and slope, is highlighted. The research provides valuable insights essential for successful hydrocarbon exploration, reservoir management, and sustainable resource exploitation in coastal West Africa.展开更多
基金funded by the National Natural Science Foundation of China(General Program:No.52074314,No.U19B6003-05)National Key Research and Development Program of China(2019YFA0708303-05)。
文摘Accurate prediction of formation pore pressure is essential to predict fluid flow and manage hydrocarbon production in petroleum engineering.Recent deep learning technique has been receiving more interest due to the great potential to deal with pore pressure prediction.However,most of the traditional deep learning models are less efficient to address generalization problems.To fill this technical gap,in this work,we developed a new adaptive physics-informed deep learning model with high generalization capability to predict pore pressure values directly from seismic data.Specifically,the new model,named CGP-NN,consists of a novel parametric features extraction approach(1DCPP),a stacked multilayer gated recurrent model(multilayer GRU),and an adaptive physics-informed loss function.Through machine training,the developed model can automatically select the optimal physical model to constrain the results for each pore pressure prediction.The CGP-NN model has the best generalization when the physicsrelated metricλ=0.5.A hybrid approach combining Eaton and Bowers methods is also proposed to build machine-learnable labels for solving the problem of few labels.To validate the developed model and methodology,a case study on a complex reservoir in Tarim Basin was further performed to demonstrate the high accuracy on the pore pressure prediction of new wells along with the strong generalization ability.The adaptive physics-informed deep learning approach presented here has potential application in the prediction of pore pressures coupled with multiple genesis mechanisms using seismic data.
基金Mainly presented at the 6-th international meeting of acoustics in Aug. 2003, and The 1999 SPE Asia Pacific Oil and GasConference and Exhibition held in Jakarta, Indonesia, 20-22 April 1999, SPE 54274.
文摘In this paper, a new concept called numerical structure of seismic data is introduced and the difference between numerical structure and numerical value of seismic data is explained. Our study shows that the numerical seismic structure is closely related to oil and gas-bearing reservoir, so it is very useful for a geologist or a geophysicist to precisely interpret the oil-bearing layers from the seismic data. This technology can be applied to any exploration or production stage. The new method has been tested on a series of exploratory or development wells and proved to be reliable in China. Hydrocarbon-detection with this new method for 39 exploration wells on 25 structures indi- cates a success ratio of over 80 percent. The new method of hydrocarbon prediction can be applied for: (1) depositional environment of reservoirs with marine fades, delta, or non-marine fades (including fluvial facies, lacustrine fades); (2) sedimentary rocks of reservoirs that are non-marine clastic rocks and carbonate rock; and (3) burial depths range from 300 m to 7000 m, and the minimum thickness of these reservoirs is over 8 m (main frequency is about 50 Hz).
基金This reservoir research is sponsored by the National 973 Subject Project (No. 2001CB209).
文摘Seismic data structure characteristics means the waveform character arranged in the time sequence at discrete data points in each 2-D or 3-D seismic trace. Hydrocarbon prediction using seismic data structure characteristics is a new reservoir prediction technique. When the main pay interval is in carbonate fracture and fissure-cavern type reservoirs with very strong inhomogeneity, there are some difficulties with hydrocarbon prediction. Because of the special geological conditions of the eighth zone in the Tahe oil field, we apply seismic data structure characteristics to hydrocarbon prediction for the Ordovician reservoir in this zone. We divide the area oil zone into favorable and unfavorable blocks. Eighteen well locations were proposed in the favorable oil block, drilled, and recovered higher output of oil and gas.
基金supported by the National Natural Science Foundation of China(No.41474109)the China National Petroleum Corporation under grant number 2016A-33
文摘Conventional time-space domain and frequency-space domain prediction filtering methods assume that seismic data consists of two parts, signal and random noise. That is, the so-called additive noise model. However, when estimating random noise, it is assumed that random noise can be predicted from the seismic data by convolving with a prediction error filter. That is, the source-noise model. Model inconsistencies, before and after denoising, compromise the noise attenuation and signal-preservation performances of prediction filtering methods. Therefore, this study presents an inversion-based time-space domain random noise attenuation method to overcome the model inconsistencies. In this method, a prediction error filter （PEF）, is first estimated from seismic data; the filter characterizes the predictability of the seismic data and adaptively describes the seismic data＇s space structure. After calculating PEF, it can be applied as a regularized constraint in the inversion process for seismic signal from noisy data. Unlike conventional random noise attenuation methods, the proposed method solves a seismic data inversion problem using regularization constraint; this overcomes the model inconsistency of the prediction filtering method. The proposed method was tested on both synthetic and real seismic data, and results from the prediction filtering method and the proposed method are compared. The testing demonstrated that the proposed method suppresses noise effectively and provides better signal-preservation performance.
文摘地震是极具破坏性与不确定性的自然现象,在人们毫无察觉的情况下地震发生在人口稠密区时,将严重危害人们生命财产安全。人们不断努力了解地震的物理特征和物理危害与环境之间的相互作用,以便在地震发生前发出适当的警报。可靠的地震预测应包含对地震信号的分析,但是这些信号在地震发生前不明显;因此使用数据驱动机器学习的方法来分析这些信号与地震的联系并预测地震。通过建立观测台网连续监测与地震发生相关的各种物理量或化学量,据此获取的地震前兆信息是地震预测的研究基础。地震发生前,地球物理场发生显著变化,伴随电磁和地声等多种前兆信号,其中电磁和地声信号具有临震特性,是开展地震临震观测预测研究的重要数据来源;因此对地下的电磁扰动和地声信号进行实时监测,获取长期观测数据用于数据驱动机器学习方法预测地震。该文基于AETA数据的临震模型预报,针对多分量地震监测预测系统(Acoustic and Electromagnetic Testing All in one system,AETA)在川滇地区记录的电磁和地声数据,提取时域和频域特征,采用基于随机森林算法、轻量级梯度提升决策树和极度随机树的集成学习方法共同预测该区域的发震情况,选取发震概率最大的子区域中心位置作为震中预测结果,进一步训练LightGBM回归模型以预测此子区域的震级,按周对地震三要素进行预测。实验结果表明,该方法在川滇地区地震风险预测上,准确率可达0.64,震级预测的平均误差为0.38,最小误差为0.00,具有良好的预测效果。
基金funded by the Natural Science Foundation of Shandong Province (ZR2021MD061ZR2023QD025)+3 种基金China Postdoctoral Science Foundation (2022M721972)National Natural Science Foundation of China (41174098)Young Talents Foundation of Inner Mongolia University (10000-23112101/055)Qingdao Postdoctoral Science Foundation (QDBSH20230102094)。
文摘Conventional machine learning(CML)methods have been successfully applied for gas reservoir prediction.Their prediction accuracy largely depends on the quality of the sample data;therefore,feature optimization of the input samples is particularly important.Commonly used feature optimization methods increase the interpretability of gas reservoirs;however,their steps are cumbersome,and the selected features cannot sufficiently guide CML models to mine the intrinsic features of sample data efficiently.In contrast to CML methods,deep learning(DL)methods can directly extract the important features of targets from raw data.Therefore,this study proposes a feature optimization and gas-bearing prediction method based on a hybrid fusion model that combines a convolutional neural network(CNN)and an adaptive particle swarm optimization-least squares support vector machine(APSO-LSSVM).This model adopts an end-to-end algorithm structure to directly extract features from sensitive multicomponent seismic attributes,considerably simplifying the feature optimization.A CNN was used for feature optimization to highlight sensitive gas reservoir information.APSO-LSSVM was used to fully learn the relationship between the features extracted by the CNN to obtain the prediction results.The constructed hybrid fusion model improves gas-bearing prediction accuracy through two processes of feature optimization and intelligent prediction,giving full play to the advantages of DL and CML methods.The prediction results obtained are better than those of a single CNN model or APSO-LSSVM model.In the feature optimization process of multicomponent seismic attribute data,CNN has demonstrated better gas reservoir feature extraction capabilities than commonly used attribute optimization methods.In the prediction process,the APSO-LSSVM model can learn the gas reservoir characteristics better than the LSSVM model and has a higher prediction accuracy.The constructed CNN-APSO-LSSVM model had lower errors and a better fit on the test dataset than the other individual models.This method proves the effectiveness of DL technology for the feature extraction of gas reservoirs and provides a feasible way to combine DL and CML technologies to predict gas reservoirs.
文摘This study examines the turbidite dynamics and hydrocarbon reservoir formation in Ghana’s Tano Basin, which is located in coastal West Africa. Through an exploration of geological processes spanning millions of years, we uncover key factors shaping hydrocarbon accumulation, including source rock richness, temperature, pressure, and geological structures. The research offers valuable insights applicable to exploration, management, and sustainable resource exploitation in coastal West Africa. It facilitates the identification of exploration targets with higher hydrocarbon potential, enables the anticipation of reservoir potential within the Tano Basin, and assists in tailoring exploration and management strategies to specific geological conditions of the Tano Basin. Analysis of fluvial channels sheds light on their impact on landscape formation and hydrocarbon exploration. The investigation into turbidite systems unveils intricate interactions involving tectonics, sea-level fluctuations, and sedimentation patterns, influencing the development of reservoirs. An understanding of sediment transport and depositional settings is essential for efficient reservoir management. Geomorphological features, such as channels, submarine canyons, and distinct channel types, are essential in this situation. A detailed examination of turbidite channel structures, encompassing canyons, channel complexes, convex channels, and U-shaped channels, provides valuable insights and aids in identifying exploration targets like basal lag, channel levees, and lobes. These findings underscore the enduring significance of turbidite systems as conduits for sediment transport, contributing to enhanced reservoir management and efficient hydrocarbon production. The study also highlights how important it is to examine the configuration of sedimentary layers, stacking patterns, and angular laminated facies to identify turbidites, understand reservoir distribution, and improve well design. The dynamic nature of turbidite systems, influenced by basin characteristics such as shape and slope, is highlighted. The research provides valuable insights essential for successful hydrocarbon exploration, reservoir management, and sustainable resource exploitation in coastal West Africa.
