Tight glutenite reservoirs are widely developed in Bohai Bay Basin,East China.They are mostly huge thick and rely on hydraulic fracturing treatment for commercial exploitation.To investigate the propagation behavior o...Tight glutenite reservoirs are widely developed in Bohai Bay Basin,East China.They are mostly huge thick and rely on hydraulic fracturing treatment for commercial exploitation.To investigate the propagation behavior of hydraulic fractures in these glutenite reservoirs,the geological feature of reservoirs in Bohai Bay Basin is studied firstly,including the reservoir vertical distribution feature and the heterogeneous lithology.Then,hydraulic fracturing treatments in block Yan 222 are carried out and the fracturing processes are monitored by the microseismic system.Results show the hydraulic fractures generated in the reservoirs are mostly in X shape.The cause of X-shaped hydraulic fractures in this study is mainly ascribed to(I)the reservoir heterogeneity and(II)the stress shadow effect of two close hydraulic fractures propagating in the same orientation,which is confirmed by the following numerical simulation and related research in detail.This study can provide a reference for the research on the fracturing behavior of the deep thick glutenite reservoirs.展开更多
海底电缆作为各类海上平台能源供给的生命线,一旦发生故障将产生巨大的经济及战略影响,准确预测海底电缆运行状态有助于提前把握其运行风险,从而实现预防性维护。在充分挖掘海底电缆运维数据中的动、静态特征的基础上,提出了一种基于注...海底电缆作为各类海上平台能源供给的生命线,一旦发生故障将产生巨大的经济及战略影响,准确预测海底电缆运行状态有助于提前把握其运行风险,从而实现预防性维护。在充分挖掘海底电缆运维数据中的动、静态特征的基础上,提出了一种基于注意力机制和卷积神经网络-门控循环神经网络(convolutional neural networks-gated recurrent unit, CNN-GRU)海底电缆运行状态预测方法。首先,考虑在线监测、巡检指标、静态试验三类关键影响因素,建立海底电缆运行状态评估指标体系;然后,基于改进层次分析法及多层次变权评估思想构建海底电缆运行状态评估模型;最后,建立基于注意力机制和CNN-GRU组合神经网络模型,将历史运行参数及状态量化结果作为输入特征参量,实现海底电缆运行状态的演化趋势预测。算例分析表明,所提方法可有效预测海底电缆的运行状态,平均百分数误差低至1.04%,与全连接神经网络、CNN、CNN-长短期记忆神经网络(long short term memory, LSTM)等算法相比均具备更优的预测精度。展开更多
基金Projects(51879041,51774112,U1810203)supported by the National Natural Science Foundation of ChinaProject(2020M672224)supported by the China Postdoctoral Science FoundationProject(B2020-41)supported by the Doctoral Fund of Henan Polytechnic University,China。
文摘Tight glutenite reservoirs are widely developed in Bohai Bay Basin,East China.They are mostly huge thick and rely on hydraulic fracturing treatment for commercial exploitation.To investigate the propagation behavior of hydraulic fractures in these glutenite reservoirs,the geological feature of reservoirs in Bohai Bay Basin is studied firstly,including the reservoir vertical distribution feature and the heterogeneous lithology.Then,hydraulic fracturing treatments in block Yan 222 are carried out and the fracturing processes are monitored by the microseismic system.Results show the hydraulic fractures generated in the reservoirs are mostly in X shape.The cause of X-shaped hydraulic fractures in this study is mainly ascribed to(I)the reservoir heterogeneity and(II)the stress shadow effect of two close hydraulic fractures propagating in the same orientation,which is confirmed by the following numerical simulation and related research in detail.This study can provide a reference for the research on the fracturing behavior of the deep thick glutenite reservoirs.
文摘海底电缆作为各类海上平台能源供给的生命线,一旦发生故障将产生巨大的经济及战略影响,准确预测海底电缆运行状态有助于提前把握其运行风险,从而实现预防性维护。在充分挖掘海底电缆运维数据中的动、静态特征的基础上,提出了一种基于注意力机制和卷积神经网络-门控循环神经网络(convolutional neural networks-gated recurrent unit, CNN-GRU)海底电缆运行状态预测方法。首先,考虑在线监测、巡检指标、静态试验三类关键影响因素,建立海底电缆运行状态评估指标体系;然后,基于改进层次分析法及多层次变权评估思想构建海底电缆运行状态评估模型;最后,建立基于注意力机制和CNN-GRU组合神经网络模型,将历史运行参数及状态量化结果作为输入特征参量,实现海底电缆运行状态的演化趋势预测。算例分析表明,所提方法可有效预测海底电缆的运行状态,平均百分数误差低至1.04%,与全连接神经网络、CNN、CNN-长短期记忆神经网络(long short term memory, LSTM)等算法相比均具备更优的预测精度。
