杨旺旺,张冲,杨梦琼,张亚男,汪明锐,孙康
YANG Wangwang,ZHANG Chong,YANG Mengqiong,ZHANG Ya'nan,WANG Mingrui,SUN Kang

• 1:油气资源与勘探技术教育部重点实验室(长江大学)
• 2:长江大学地球物理与石油资源学院

摘要(Abstract)：

伊拉克H油田碳酸盐岩储层孔隙结构复杂，孔隙类型多样，给渗透率测井评价工作带来了极大困难。针对这一问题，建立了基于测井序列信息的长短期记忆（LSTM）循环神经网络渗透率预测模型。从测井响应差异以及测井序列信息出发，优选敏感测井曲线，搭建LSTM循环神经网络，训练网络并优化网络参数，建立了基于LSTM循环神经网络的伊拉克H油田碳酸盐岩储层渗透率预测模型。应用该模型对伊拉克H油田进行渗透率测井评价，并将预测结果与灰色系统预测模型GM (0,N)进行对比。结果表明：相对于灰色系统预测模型的结果，基于LSTM循环神经网络的渗透率预测模型的均方根误差降低了29.47%，皮尔逊（Pearson）相关系数提高了6.59%，取得了较好的应用效果。该模型能够充分挖掘测井曲线与渗透率之间关系的信息，提升了储层渗透率评价精度。
Complex pore structure and various pore types of carbonate reservoir in Oilfield H of Iraq bring much challenge to permeability logging evaluation. To solve this problem,a permeability predicting model based on long short-term memory(LSTM)recurrent neural network is established. Starting from logging response difference and logging series information,sensitive logging curves are selected,LSTM recurrent neural network is built,network is trained,network parameters are optimized and permeability predicting model of carbonate reservoir in Oilfield H of Iraq based on LSTM recurrent neural network is established. The model is applied to permeability logging evaluation of Oilfield H in Iraq,and predicted results are compared with those of grey system predicting model GM(0,N). The results show that compared with grey system predicting model,the root mean square error of permeability predicting model based on LSTM recurrent neural network is reduced by 29.47% and Pearson correlation coefficient is increased by 6.59%,with better application effect. The model can fully mine the information between logging curve and reservior permeability to increase accuracy of permeability evaluation.

关键词(KeyWords)： 长短期记忆循环神经网络;伊拉克H油田碳酸盐岩储层;渗透率;测井评价
long short-term memory recurrent neural network;carbonate reservior in Oilfield H of Iraq;permeability;logging evaluation

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目“致密气储层岩石导电机理研究及饱和度评价”（41404084）;; 国家科技重大专项子课题“复杂碳酸盐岩储层测井评价关键技术研究与应用”（2017ZX05032-003-005）

作者(Author): 杨旺旺,张冲,杨梦琼,张亚男,汪明锐,孙康
YANG Wangwang,ZHANG Chong,YANG Mengqiong,ZHANG Ya'nan,WANG Mingrui,SUN Kang

DOI: 10.19597/j.issn.1000-3754.202105003

参考文献(References)：

• [1]赵培强，李长文，沙峰，等.滨里海盆地东缘中区块碳酸盐岩储层渗透率预测研究[J].石油科学通报，2020,5(1):39-48.ZHAO Peiqiang, LI Changwen, SHA Feng, et al. Study of permeability prediction of carbonate reservoirs in the middle block of the eastern margin of the Caspian Basin[J]. Petroleum Science Bulletin,2020,5(1):39-48.
• [2]张凤生，隋秀英，段朝伟，等.高孔隙度低渗透率碳酸盐岩储层岩心核磁共振实验研究[J].测井技术，2018,42(5):497-502,529.ZHANG Fengsheng, SUI Xiuying, DUAN Chaowei, et al.NMR experimental study on carbonate reservoirs with high porosity and low permeability[J]. Well Logging Technology,2018,42(5):497-502,529.
• [3]孙建孟，闫国亮.渗透率模型研究进展[J].测井技术，2012,36(4):329-335.SUN Jianmeng,YAN Guoliang. Review on absolute permeability model[J]. Well Logging Technology, 2012, 36(4):329-335.
• [4]袁伟，何胜林，焦祥燕，等.北部湾盆地低渗-特低渗含砾砂岩储层渗透率评价方法[J].大庆石油地质与开发，2020,39(5):127-134.YUAN Wei,HE Shenglin,JIAO Xiangyan,et al. Evaluating method for the permeabilities of low permeability and ultra-low permeability pebbly sandstone reservoirs in Beibuwan Basin[J]. Petroleum Geology&Oilfield Development in Daqing,2020,39(5):127-134.
• [5]张冲，汤潇，倪国辉，等.基于Thomeer函数的中东区块碳酸盐岩储层渗透率评价[J].长江大学学报（自然科学版），2021,18(1):43-51ZHANG Chong,TANG Xiao,NI Guohui,et al.Evaluation of carbonate reservoir permeability in Middle East Block based on Thomeer function[J]. Journal of Yangtze University(Natural Science Edition),2021,18(1):43-51.
• [6]张冲，张占松，张超谟.基于等效岩石组分理论的渗透率解释模型[J].测井技术，2014,38(6):690-694.ZHANG Chong,ZHANG Zhansong,ZHANG Chaomo. A permeability interpretation model based on equivalent rock elements theory[J].Well Logging Technology,2014,38(6):690-694.
• [7]GUNTER G W,SPAIN D R,VIRO E J,et al. Winland pore throat prediction method-aproper retrospect:New examples from carbonates and complex systems[C].Abu Dhabi:SPWLA55th Annual Logging Symposium,2014.
• [8]SWANSON B F. A simple correlation between permeabilities and mercury capillary Pressures[J].Journal of Petroleum Technology,1981,33(12):2498-2504.
• [9]GUO B Y, AL G, DUAN S K. Correlation between sandstone permeability and capillary pressure curves[J].Journal of Petroleum Science and Engineering,2004,43(2):239-246.
• [10]朱林奇，张冲，胡佳，等.基于单元体模型的核磁共振测井渗透率评价方法[J].石油钻探技术，2016,44(4):120-126.ZHU Linqi,ZHANG Chong,HU Jia,et al.An NMR logging permeability evaluation method based on the representative elementary volume model[J]. Petroleum Drilling Techniques,2016,44(4):120-126.
• [11]严伟，张冲，胡挺，等.利用斯通利波估算火成岩地层渗透率[J].岩性油气藏，2013,25(1):107-110.YAN Wei, ZHANG Chong, HU Ting, et al. Calculation of permeability in igneous rock formation based on Stoneley wave[J]. Lithologic Reserviors,2013,25(1):107-110.
• [12]谷宇峰，张道勇，鲍志东.测井资料PSO-XGBoost渗透率预测[J].石油地球物理勘探，2021,56(1):26-37.GU Yufeng, ZHANG Daoyong, BAO Zhidong. Permeability prediction using PSO-XGBoost based on logging data[J].Oil Geophysical Prospecting,2021,56(1):26-37.
• [13]安鹏，曹丹平，赵宝银，等.基于LSTM循环神经网络的储层物性参数预测方法研究[J].地球物理学进展，2019,34(5):1849-1858.AN Peng, CAO Danping, ZHAO Baoyin, et al. Reservoir physical parameters prediction based on LSTM recurrent neural network[J]. Progress in Geophysics, 2019, 34(5):1849-1858.
• [14]王俊，曹俊兴，尤加春，等.基于门控循环单元神经网络的储层孔渗饱参数预测[J].石油物探，2020,59(4):616-627.WANG Jun,CAO Junxing,YOU Jiachun,et al.Prediction of reservoir porosity, permeability, and saturation based on a gated recurrent unit neural network[J].Geophysical Prospecting for Petroleum,2020,59(4):616-627.
• [15]HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
• [16]史静，李琥，李冰洁，等.基于长短期记忆神经网络的配电网负荷预测方法研究[J].供用电，2019, 36(7):71-77,90.SHI Jing,LI Hu,LI Bingjie,et al.Distribution network load forecasting based on LSTM neural network[J].Distribution&Utilization,2019,36(7):71-77,90.
• [17]邓聚龙.灰色控制系统[J].华中工学院学报，1982,10(3):9-18.DENG Julong. The grey control system[J].Journal of Central China Institute of Technology,1982,10(3):9-18.