本文详细阐述了渗流力学经典理论——达西定律的发展历程及其适用条件,并从Navier-Stokes(N-S)方程推导出了达西定律在多孔介质中的毛细管渗流和裂缝渗流中的数学表达式。文章指出了当前达西定律应用中存在的8大问题,并综合分析了渗流...本文详细阐述了渗流力学经典理论——达西定律的发展历程及其适用条件,并从Navier-Stokes(N-S)方程推导出了达西定律在多孔介质中的毛细管渗流和裂缝渗流中的数学表达式。文章指出了当前达西定律应用中存在的8大问题,并综合分析了渗流力学理论在油气田开发中的主要挑战。针对这些挑战,本文提出了一系列对策和思考。文章强调指出:构建多尺度、多物理场耦合模型并借助AI科学计算是揭示油气储层复杂真实流动机制,填补目前理论空白的必由之路。建议指出:进一步发展核磁共振、电镜扫描及智能数据与图像处理等高精度实验技术,以直观展现流体在储层中的流动行为和过程。最后,建议综合运用实验研究、新理论模型建立和AI科学研究方法(AI for Science),突破油气渗流力学理论中目前遇到的挑战。研究成果可为我国高校、科研机构和研究者开展石油科学理论研究和课题立项提供重要参考,同时可为我国油气资源可持续进行科学和技术战略规划提供强有力的技术支撑。展开更多
Since the oil production of single well in water flooding reservoir varies greatly and is hard to predict, an oil production prediction method of single well based on temporal convolutional network(TCN) is proposed an...Since the oil production of single well in water flooding reservoir varies greatly and is hard to predict, an oil production prediction method of single well based on temporal convolutional network(TCN) is proposed and verified. This method is started from data processing, the correspondence between water injectors and oil producers is determined according to the influence radius of the water injectors, the influence degree of a water injector on an oil producer in the month concerned is added as a model feature, and a Random Forest(RF) model is built to fill the dynamic data of water flooding. The single well history is divided into 4 stages according to its water cut, that is, low water cut, middle water cut, high water cut and extra-high water cut stages. In each stage, a TCN based prediction model is established, hyperparameters of the model are optimized by the Sparrow Search Algorithm(SSA). Finally, the models of the 4 stages are integrated into one whole-life model of the well for production prediction. The application of this method in Daqing Oilfield, NE China shows that:(1) Compared with conventional data processing methods, the data obtained by this processing method are more close to the actual production, and the data set obtained is more authentic and complete.(2) The TCN model has higher prediction accuracy than other 11 models such as Long Short Term Memory(LSTM).(3) Compared with the conventional full-life-cycle models, the model of integrated stages can significantly reduce the error of production prediction.展开更多
A deep learning method for predicting oil field production at ultra-high water cut stage from the existing oil field production data was presented,and the experimental verification and application effect analysis were...A deep learning method for predicting oil field production at ultra-high water cut stage from the existing oil field production data was presented,and the experimental verification and application effect analysis were carried out.Since the traditional Fully Connected Neural Network(FCNN)is incapable of preserving the correlation of time series data,the Long Short-Term Memory(LSTM)network,which is a kind of Recurrent Neural Network(RNN),was utilized to establish a model for oil field production prediction.By this model,oil field production can be predicted from the relationship between oil production index and its influencing factors and the trend and correlation of oil production over time.Production data of a medium and high permeability sandstone oilfield in China developed by water flooding was used to predict its production at ultra-high water cut stage,and the results were compared with the results from the traditional FCNN and water drive characteristic curves.The LSTM based on deep learning has higher precision,and gives more accurate production prediction for complex time series in oil field production.The LSTM model was used to predict the monthly oil production of another two oil fields.The prediction results are good,which verifies the versatility of the method.展开更多
Through reviewing the generation process and essential characteristics of waterflooding curves, the essence and characteristics of Zhang Jinqing waterflooding curve and Yu Qitai waterflooding curve recommended in Chin...Through reviewing the generation process and essential characteristics of waterflooding curves, the essence and characteristics of Zhang Jinqing waterflooding curve and Yu Qitai waterflooding curve recommended in Chinese Petroleum Industry Standard 'Calculation methods for Recoverable Oil Reserves(SY/T5367—1998)' were discussed, and some technical issues related to the curves were examined in-depth. We found that:(1) All the waterflooding curves are based on empirical formulas derived from oilfield production experience and statistics methods, and can characterize oil displacement features by water quite well.(2) A new waterflooding curve can be derived by combining waterflooding parameters and using different mathematical calculations as long as the parameter combinations and mathematical operation meet a linear relationship, so proposing new waterflooding curves by changing the combination mode has no practical significance anymore.(3) The upwarp of waterflooding curve in the extremely high water cut stage is because the mobility ratio curve has an inflection point with the rapid rise of water cut after reaching a certain value, and the later rapid rise of mobility ratio changes the original two-phase flow dynamics.(4) After entering into water cut stage, all the waterflooding curves with linear relationship can be used to make prediction, even curves with inflection points, as long as they have a straight section above the inflection point.(5) Actual data of waterflooding oilfields has proved that Type A, Zhang Jinqing and Yu Qitai waterflooding curves all can predict accurately oil recoverable reserves in extremely high water cut stage and can be promoted.展开更多
After analyzing many studies of fluid flow theory of multi-porous media in low and extra-low permeability reservoirs and the numerical simulation of non-Darcy flow, we found that a negative flow rate occurs in the exi...After analyzing many studies of fluid flow theory of multi-porous media in low and extra-low permeability reservoirs and the numerical simulation of non-Darcy flow, we found that a negative flow rate occurs in the existing non-Darcy flow equation, which is unreasonable. We believe that the existing equation can only be considered as a discriminant to judging Darcy flow or non-Darcy flow, and cannot be taken as a fluid flow governing equation of multi-porous media. Our analysis of the experimental results shows that the threshold pressure gradient(TPG) of low and extra-low permeability reservoirs is excessively high, and does not conform to fluid flow through multi-porous media in the actual reservoir situation. Therefore, we present a reasonable TPG ranging from 0.006 to 0.04 MPa/m at the well depth of 1500 m and oil drainage distance of 500 m. The results of our study also indicate that the non-Darcy flow phenomenon will disappear when the TPG reaches a certain value. In addition, the TPG or non-Darcy flow in low and extra-low permeability reservoirs does not need to be considered in the productivity prediction and reservoir numerical simulation. At present, the black oil model or dual-porous media is suitable for simulating low and extra-low permeability reservoirs.展开更多
文摘本文详细阐述了渗流力学经典理论——达西定律的发展历程及其适用条件,并从Navier-Stokes(N-S)方程推导出了达西定律在多孔介质中的毛细管渗流和裂缝渗流中的数学表达式。文章指出了当前达西定律应用中存在的8大问题,并综合分析了渗流力学理论在油气田开发中的主要挑战。针对这些挑战,本文提出了一系列对策和思考。文章强调指出:构建多尺度、多物理场耦合模型并借助AI科学计算是揭示油气储层复杂真实流动机制,填补目前理论空白的必由之路。建议指出:进一步发展核磁共振、电镜扫描及智能数据与图像处理等高精度实验技术,以直观展现流体在储层中的流动行为和过程。最后,建议综合运用实验研究、新理论模型建立和AI科学研究方法(AI for Science),突破油气渗流力学理论中目前遇到的挑战。研究成果可为我国高校、科研机构和研究者开展石油科学理论研究和课题立项提供重要参考,同时可为我国油气资源可持续进行科学和技术战略规划提供强有力的技术支撑。
基金Major Unified Construction Project of Petro China(2019-40210-000020-02)。
文摘Since the oil production of single well in water flooding reservoir varies greatly and is hard to predict, an oil production prediction method of single well based on temporal convolutional network(TCN) is proposed and verified. This method is started from data processing, the correspondence between water injectors and oil producers is determined according to the influence radius of the water injectors, the influence degree of a water injector on an oil producer in the month concerned is added as a model feature, and a Random Forest(RF) model is built to fill the dynamic data of water flooding. The single well history is divided into 4 stages according to its water cut, that is, low water cut, middle water cut, high water cut and extra-high water cut stages. In each stage, a TCN based prediction model is established, hyperparameters of the model are optimized by the Sparrow Search Algorithm(SSA). Finally, the models of the 4 stages are integrated into one whole-life model of the well for production prediction. The application of this method in Daqing Oilfield, NE China shows that:(1) Compared with conventional data processing methods, the data obtained by this processing method are more close to the actual production, and the data set obtained is more authentic and complete.(2) The TCN model has higher prediction accuracy than other 11 models such as Long Short Term Memory(LSTM).(3) Compared with the conventional full-life-cycle models, the model of integrated stages can significantly reduce the error of production prediction.
基金Supported by China National Science and Technology Major Project(2016ZX05016-006)
文摘A deep learning method for predicting oil field production at ultra-high water cut stage from the existing oil field production data was presented,and the experimental verification and application effect analysis were carried out.Since the traditional Fully Connected Neural Network(FCNN)is incapable of preserving the correlation of time series data,the Long Short-Term Memory(LSTM)network,which is a kind of Recurrent Neural Network(RNN),was utilized to establish a model for oil field production prediction.By this model,oil field production can be predicted from the relationship between oil production index and its influencing factors and the trend and correlation of oil production over time.Production data of a medium and high permeability sandstone oilfield in China developed by water flooding was used to predict its production at ultra-high water cut stage,and the results were compared with the results from the traditional FCNN and water drive characteristic curves.The LSTM based on deep learning has higher precision,and gives more accurate production prediction for complex time series in oil field production.The LSTM model was used to predict the monthly oil production of another two oil fields.The prediction results are good,which verifies the versatility of the method.
基金Supported by China National Science and Technology Major Project(2016ZX05016-006)
文摘Through reviewing the generation process and essential characteristics of waterflooding curves, the essence and characteristics of Zhang Jinqing waterflooding curve and Yu Qitai waterflooding curve recommended in Chinese Petroleum Industry Standard 'Calculation methods for Recoverable Oil Reserves(SY/T5367—1998)' were discussed, and some technical issues related to the curves were examined in-depth. We found that:(1) All the waterflooding curves are based on empirical formulas derived from oilfield production experience and statistics methods, and can characterize oil displacement features by water quite well.(2) A new waterflooding curve can be derived by combining waterflooding parameters and using different mathematical calculations as long as the parameter combinations and mathematical operation meet a linear relationship, so proposing new waterflooding curves by changing the combination mode has no practical significance anymore.(3) The upwarp of waterflooding curve in the extremely high water cut stage is because the mobility ratio curve has an inflection point with the rapid rise of water cut after reaching a certain value, and the later rapid rise of mobility ratio changes the original two-phase flow dynamics.(4) After entering into water cut stage, all the waterflooding curves with linear relationship can be used to make prediction, even curves with inflection points, as long as they have a straight section above the inflection point.(5) Actual data of waterflooding oilfields has proved that Type A, Zhang Jinqing and Yu Qitai waterflooding curves all can predict accurately oil recoverable reserves in extremely high water cut stage and can be promoted.
基金sponsored by National Key Project of Science and Technology of the Ministry of Science and Technology(MOST)(Grant No.2011ZX05043-002)
文摘After analyzing many studies of fluid flow theory of multi-porous media in low and extra-low permeability reservoirs and the numerical simulation of non-Darcy flow, we found that a negative flow rate occurs in the existing non-Darcy flow equation, which is unreasonable. We believe that the existing equation can only be considered as a discriminant to judging Darcy flow or non-Darcy flow, and cannot be taken as a fluid flow governing equation of multi-porous media. Our analysis of the experimental results shows that the threshold pressure gradient(TPG) of low and extra-low permeability reservoirs is excessively high, and does not conform to fluid flow through multi-porous media in the actual reservoir situation. Therefore, we present a reasonable TPG ranging from 0.006 to 0.04 MPa/m at the well depth of 1500 m and oil drainage distance of 500 m. The results of our study also indicate that the non-Darcy flow phenomenon will disappear when the TPG reaches a certain value. In addition, the TPG or non-Darcy flow in low and extra-low permeability reservoirs does not need to be considered in the productivity prediction and reservoir numerical simulation. At present, the black oil model or dual-porous media is suitable for simulating low and extra-low permeability reservoirs.
