Article(id=1228279668579693512, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1228279664221815452, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2407065, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1726848000000, receivedDateStr=2024-09-21, revisedDate=1747065600000, revisedDateStr=2025-05-13, acceptedDate=null, acceptedDateStr=null, onlineDate=1770774293322, onlineDateStr=2026-02-11, pubDate=1754582400000, pubDateStr=2025-08-08, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1770774293322, onlineIssueDateStr=2026-02-11, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1770774293322, creator=13701087609, updateTime=1770774293322, updator=13701087609, issue=Issue{id=1228279664221815452, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='22', pageStart='9211', pageEnd='9648', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1770774292283, creator=13701087609, updateTime=1770777611996, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1228293588207992892, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1228279664221815452, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1228293588207992893, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1228279664221815452, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=9335, endPage=9341, ext={EN=ArticleExt(id=1228279669695378398, articleId=1228279668579693512, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Pore Pressure Prediction Model Based on CNN-Attn Neural Network, columnId=1228279669582132187, journalTitle=Science Technology and Engineering, columnName=Papers·Petroleum and Natural Gas Industry, runingTitle=null, highlight=null, articleAbstract=

In the exploration and exploitation of oil and gas, artificial intelligence models are extensively employed in the prediction of formation pore pressure. Among them, single models tend to encounter problems such as overfitting or unstable prediction outcomes, leaving room for improvement in aspects like prediction accuracy and generalization ability. To enhance the prediction accuracy of formation pore pressure, a CNN-Attn neural network-based formation pore pressure prediction model was established by virtue of deep learning technology. In this research, five types of logging and while-drilling data were optimally selected, and the linear correlation between the data and formation pore pressure was verified using the Pearson correlation coefficient method. Through the optimization of the structure of the one-dimensional CNN, the model can effectively capture the local characteristics of the data and, when combined with the self-attention mechanism, strengthen the model’s ability to capture global dependencies, thereby elevating the model’s expressiveness and comprehension. To validate the prediction accuracy of this model, two wells in the Bayan block were subjected to prediction. The average absolute errors of the prediction results were both less than 1 MPa, the root mean square errors were both less than 1 MPa, the average relative errors were both less than 1.3%, and the determination coefficients were both greater than 0.9, with higher accuracy compared to the BP, CNN, and LSTM models. This model has improved the prediction accuracy of formation pore pressure and provided data support for drilling safety.

, correspAuthors=Zhong-hui LI, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=Kai TANG, Zhong-hui LI, Tian-bao CAO, Peng-jie HU), CN=ArticleExt(id=1228279673537359948, articleId=1228279668579693512, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=石油、天然气工业基于CNN-Attn神经网络的孔隙压力预测, columnId=1228279669867344865, journalTitle=科学技术与工程, columnName=论文·石油、天然气工业, runingTitle=null, highlight=null, articleAbstract=

在石油和天然气的勘探开发中,人工智能模型被广泛应用于地层孔隙压力的预测。其中单一模型容易出现过拟合或预测结果不稳定的问题,在预测精度和泛化能力方面仍有提升空间。为了提高地层孔隙压力预测精度,基于深度学习技术,建立了CNN-Attn神经网络地层孔隙压力预测模型。优选了5种测井和随钻数据,使用Pearson相关系数法验证数据与地层孔隙压力的线性相关性。通过对一维CNN的结构进行优化,使模型能够有效捕捉数据的局部特征,并与自注意力机制结合,增强模型对全局依赖关系的捕捉能力,从而提高模型的表现力和理解能力。为了验证该模型的预测精度,对巴彦区块两口井进行预测,预测结果的平均绝对误差均小于1 MPa,均方根误差均小于1 MPa,平均相对误差均小于1.3%,决定系数均大于0.9,比BP、CNN和LSTM模型精度高。该模型提升了地层孔隙压力预测精度,并为钻井安全性提供了数据支持。

, correspAuthors=李忠慧, authorNote=null, correspAuthorsNote=
* 李忠慧(1977—),男,汉族,吉林公主岭人,博士,教授,研究方向:岩石力学与钻完井工程。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=fF2ZwhrUwU/xt/DFdz8SaA==, magXml=lYFW3YzeLhxXKTPFQsgnWQ==, pdfUrl=null, pdf=Lmes8Rpma0/NIaFZxf4eag==, pdfFileSize=6191102, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=L8GAuD/pyYtMy8Z0CLWXNg==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=M3BERsOhDW74RtLuAOxDgg==, mapNumber=null, authorCompany=null, fund=null, authors=

唐凯(2001—),男,汉族,湖北仙桃人,硕士研究生。研究方向:人工智能技术与钻井工程。E-mail:

, authorsList=唐凯, 李忠慧, 曹天宝, 胡棚杰)}, authors=[Author(id=1228369775643001084, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=1430600329@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1228369775739470083, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, authorId=1228369775643001084, language=EN, stringName=Kai TANG, firstName=Kai, middleName=null, lastName=TANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Key Laboratory of Drilling and Production Engineering for Oil and Gas, Wuhan 430100, China
2 School of Petroleum Engineering, Yangtze University, National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan 430100, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1228369775814967561, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, authorId=1228369775643001084, language=CN, stringName=唐凯, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 油气钻采工程湖北省重点实验室, 武汉 430100
2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100, bio={"content":"

唐凯(2001—),男,汉族,湖北仙桃人,硕士研究生。研究方向:人工智能技术与钻井工程。E-mail:

"}, bioImg=null, bioContent=

唐凯(2001—),男,汉族,湖北仙桃人,硕士研究生。研究方向:人工智能技术与钻井工程。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1228369775374565610, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, xref=1, ext=[AuthorCompanyExt(id=1228369775382954219, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775374565610, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Key Laboratory of Drilling and Production Engineering for Oil and Gas, Wuhan 430100, China), AuthorCompanyExt(id=1228369775387148525, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775374565610, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 油气钻采工程湖北省重点实验室, 武汉 430100)]), AuthorCompany(id=1228369775483617521, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, xref=2, ext=[AuthorCompanyExt(id=1228369775496200435, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775483617521, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Petroleum Engineering, Yangtze University, National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan 430100, China), AuthorCompanyExt(id=1228369775504589044, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775483617521, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100)])]), Author(id=1228369775911436561, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=lizhonghui@yangtzeu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1228369776007905562, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, authorId=1228369775911436561, language=EN, stringName=Zhong-hui LI, firstName=Zhong-hui, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 Key Laboratory of Drilling and Production Engineering for Oil and Gas, Wuhan 430100, China
2 School of Petroleum Engineering, Yangtze University, National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan 430100, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1228369776112763169, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, authorId=1228369775911436561, language=CN, stringName=李忠慧, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 油气钻采工程湖北省重点实验室, 武汉 430100
2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1228369775374565610, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, xref=1, ext=[AuthorCompanyExt(id=1228369775382954219, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775374565610, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Key Laboratory of Drilling and Production Engineering for Oil and Gas, Wuhan 430100, China), AuthorCompanyExt(id=1228369775387148525, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775374565610, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 油气钻采工程湖北省重点实验室, 武汉 430100)]), AuthorCompany(id=1228369775483617521, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, xref=2, ext=[AuthorCompanyExt(id=1228369775496200435, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775483617521, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Petroleum Engineering, Yangtze University, National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan 430100, China), AuthorCompanyExt(id=1228369775504589044, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775483617521, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100)])]), Author(id=1228369776205037864, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1228369776305701169, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, authorId=1228369776205037864, language=EN, stringName=Tian-bao CAO, firstName=Tian-bao, middleName=null, lastName=CAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Key Laboratory of Drilling and Production Engineering for Oil and Gas, Wuhan 430100, China
2 School of Petroleum Engineering, Yangtze University, National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan 430100, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1228369776410558775, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, authorId=1228369776205037864, language=CN, stringName=曹天宝, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 油气钻采工程湖北省重点实验室, 武汉 430100
2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1228369775374565610, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, xref=1, ext=[AuthorCompanyExt(id=1228369775382954219, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775374565610, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Key Laboratory of Drilling and Production Engineering for Oil and Gas, Wuhan 430100, China), AuthorCompanyExt(id=1228369775387148525, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775374565610, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 油气钻采工程湖北省重点实验室, 武汉 430100)]), AuthorCompany(id=1228369775483617521, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, xref=2, ext=[AuthorCompanyExt(id=1228369775496200435, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775483617521, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Petroleum Engineering, Yangtze University, National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan 430100, China), AuthorCompanyExt(id=1228369775504589044, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775483617521, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100)])]), Author(id=1228369776523804991, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1228369776628662597, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, authorId=1228369776523804991, language=EN, stringName=Peng-jie HU, firstName=Peng-jie, middleName=null, lastName=HU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Key Laboratory of Drilling and Production Engineering for Oil and Gas, Wuhan 430100, China
2 School of Petroleum Engineering, Yangtze University, National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan 430100, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1228369776825794896, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, authorId=1228369776523804991, language=CN, stringName=胡棚杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 油气钻采工程湖北省重点实验室, 武汉 430100
2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1228369775374565610, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, xref=1, ext=[AuthorCompanyExt(id=1228369775382954219, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775374565610, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Key Laboratory of Drilling and Production Engineering for Oil and Gas, Wuhan 430100, China), AuthorCompanyExt(id=1228369775387148525, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775374565610, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 油气钻采工程湖北省重点实验室, 武汉 430100)]), AuthorCompany(id=1228369775483617521, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, xref=2, ext=[AuthorCompanyExt(id=1228369775496200435, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775483617521, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Petroleum Engineering, Yangtze University, National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan 430100, China), AuthorCompanyExt(id=1228369775504589044, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775483617521, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100)])])], keywords=[Keyword(id=1228369778293801305, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, orderNo=1, keyword=formation pore pressure), Keyword(id=1228369778444796257, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, orderNo=2, keyword=intelligent prediction), Keyword(id=1228369778528682342, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, orderNo=3, keyword=deep learning), Keyword(id=1228369778654511466, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, orderNo=4, keyword=convolutional neural network), Keyword(id=1228369778792923505, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, orderNo=5, keyword=self-attention mechanism), Keyword(id=1228369778910364021, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, orderNo=1, keyword=地层孔隙压力), Keyword(id=1228369779027804542, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, orderNo=2, keyword=智能预测), Keyword(id=1228369779124273538, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, orderNo=3, keyword=深度学习), Keyword(id=1228369779233325448, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, orderNo=4, keyword=卷积神经网络), Keyword(id=1228369779317211535, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, orderNo=5, keyword=自注意力机制)], refs=[Reference(id=1228369783570235931, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=194111, pageEnd=null, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=Zhao Y, Sun T, Yang J, journalName=SPE/IADC Drilling Conference and Exhibition, refType=null, unstructuredReference=Zhao Y, Sun T, Yang J, et al. Combining drilling big data and machine learning method toimprove the timeliness of drilling[C]// SPE/IADC Drilling Conference and Exhibition. Hague: SPE, 2019: 194111-MS., articleTitle=Combining drilling big data and machine learning method toimprove the timeliness of drilling, refAbstract=null), Reference(id=1228369783658316321, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=null, volume=29, issue=4, pageStart=337, pageEnd=347, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=余金秋, 孙夕平, 于永才, journalName=海相油气地质2024, refType=null, unstructuredReference=余金秋, 孙夕平, 于永才, 等. 深层地层压力预测技术研究进展[J]. 海相油气地质2024, 29(4): 337-347., articleTitle=深层地层压力预测技术研究进展, refAbstract=null), Reference(id=1228369783775756834, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2024, volume=29, issue=4, pageStart=337, pageEnd=347, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=Yu Jinqiu, Sun Xiping, Yu Yongcai, journalName=Marine Origin Petroleum Geology, refType=null, unstructuredReference=Yu Jinqiu, Sun Xiping, Yu Yongcai, et al. Research progress on deep formation pressure prediction technology[J]. Marine Origin Petroleum Geology, 2024, 29(4): 337-347., articleTitle=Research progress on deep formation pressure prediction technology, refAbstract=null), Reference(id=1228369783868031526, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=2, pageStart=494, pageEnd=501, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=毕臣臣, journalName=科学技术与工程, refType=null, unstructuredReference=毕臣臣. 基于深度学习的页岩储层总有机碳含量预测方法[J]. 科学技术与工程, 2023, 23(2): 494-501., articleTitle=基于深度学习的页岩储层总有机碳含量预测方法, refAbstract=null), Reference(id=1228369784052580910, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=2, pageStart=494, pageEnd=501, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=Bi Chenchen, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Bi Chenchen. Total organic carbon content prediction method of shale reservoir based on deep learning[J]. Science Technology and Engineering, 2023, 23(2): 494-501., articleTitle=Total organic carbon content prediction method of shale reservoir based on deep learning, refAbstract=null), Reference(id=1228369784140661297, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=7, pageStart=2609, pageEnd=2617, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=马陇飞, 萧汉敏, 陶敬伟, journalName=科学技术与工程, refType=null, unstructuredReference=马陇飞, 萧汉敏, 陶敬伟, 等. 基于深度学习岩性分类的研究与应用[J]. 科学技术与工程, 2022, 22(7): 2609-2617., articleTitle=基于深度学习岩性分类的研究与应用, refAbstract=null), Reference(id=1228369784224547383, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=7, pageStart=2609, pageEnd=2617, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Ma Longfei, Xiao Hanmin, Tao Jingwei, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Ma Longfei, Xiao Hanmin, Tao Jingwei, et al. Research and application of lithology classification based on deep learnning[J]. Science Technology and Engineering, 2022, 22(7): 2609-2617., articleTitle=Research and application of lithology classification based on deep learnning, refAbstract=null), Reference(id=1228369784291656251, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=32, pageStart=14174, pageEnd=14183, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=杨琨, 罗山贵, 花凌旭, journalName=科学技术与工程, refType=null, unstructuredReference=杨琨, 罗山贵, 花凌旭, 等. 致密砾岩油藏压裂甜点预测研究——以玛18井区为例[J]. 科学技术与工程, 2022, 22(32): 14174-14183., articleTitle=致密砾岩油藏压裂甜点预测研究——以玛18井区为例, refAbstract=null), Reference(id=1228369784371348032, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=32, pageStart=14174, pageEnd=14183, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=Yang Kun, Luo Shangui, Hua Lingxu, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Yang Kun, Luo Shangui, Hua Lingxu, et al. Investigation on the fracturing sweet spot predliction of conglomerate tight oil reservoir: a case study of the Mal8 Well Block[J]. Science Technology and Engineering, 2022, 22(32): 14174-14183., articleTitle=Investigation on the fracturing sweet spot predliction of conglomerate tight oil reservoir: a case study of the Mal8 Well Block, refAbstract=null), Reference(id=1228369784467817028, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2021, volume=48, issue=1, pageStart=1, pageEnd=11, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=匡立春, 刘合, 任义丽, journalName=石油勘探与开发, refType=null, unstructuredReference=匡立春, 刘合, 任义丽, 等. 人工智能在石油勘探开发领域的应用现状与发展趋势[J]. 石油勘探与开发, 2021, 48(1): 1-11., articleTitle=人工智能在石油勘探开发领域的应用现状与发展趋势, refAbstract=null), Reference(id=1228369784555897418, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2021, volume=48, issue=1, pageStart=1, pageEnd=11, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=Kuang Lichun, Liu He, Ren Yili, journalName=Petroleum Exploration and Development, refType=null, unstructuredReference=Kuang Lichun, Liu He, Ren Yili, et al. Application and development trend of artificial intelligence in petroleum exploration and development[J]. Petroleum Exploration and Development, 2021, 48(1): 1-11., articleTitle=Application and development trend of artificial intelligence in petroleum exploration and development, refAbstract=null), Reference(id=1228369784639783502, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2023, volume=51, issue=4, pageStart=35, pageEnd=47, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=李根生, 宋先知, 祝兆鹏, journalName=石油钻探技术, refType=null, unstructuredReference=李根生, 宋先知, 祝兆鹏, 等. 智能钻完井技术研究进展与前景展望[J]. 石油钻探技术, 2023, 51(4): 35-47., articleTitle=智能钻完井技术研究进展与前景展望, refAbstract=null), Reference(id=1228369784727863892, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2023, volume=51, issue=4, pageStart=35, pageEnd=47, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=Li Gensheng, Song Xianzhi, Zhu Zhaopeng, journalName=Petroleum Drilling Techniques, refType=null, unstructuredReference=Li Gensheng, Song Xianzhi, Zhu Zhaopeng, et al. Research progress and the prospect of intelligent drilling and completion techno-logies[J]. Petroleum Drilling Techniques, 2023, 51(4): 35-47., articleTitle=Research progress and the prospect of intelligent drilling and completion techno-logies, refAbstract=null), Reference(id=1228369784811749975, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2024, volume=52, issue=4, pageStart=1, pageEnd=10, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=张瑞, 祝兆鹏, 李大钰, journalName=石油机械, refType=null, unstructuredReference=张瑞, 祝兆鹏, 李大钰, 等. 基于改进时序网络的钻进参数可解释实时预测[J]. 石油机械, 2024, 52(4): 1-10., articleTitle=基于改进时序网络的钻进参数可解释实时预测, refAbstract=null), Reference(id=1228369784916607577, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2024, volume=52, issue=4, pageStart=1, pageEnd=10, url=null, language=null, rfNumber=[8], rfOrder=14, authorNames=Zhang Rui, Zhu Zhaopeng, Li Dayu, journalName=China Petroleum Machinery, refType=null, unstructuredReference=Zhang Rui, Zhu Zhaopeng, Li Dayu, et al. Interpretable real-time prediction of dilling parameters based on improved sequential network[J]. China Petroleum Machinery, 2024, 52(4): 1-10., articleTitle=Interpretable real-time prediction of dilling parameters based on improved sequential network, refAbstract=null), Reference(id=1228369785063408228, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=11, pageStart=16, pageEnd=22, url=null, language=null, rfNumber=[9], rfOrder=15, authorNames=陈凯枫, 杨学文, 宋先知, journalName=石油机械, refType=null, unstructuredReference=陈凯枫, 杨学文, 宋先知, 等. 基于工程录井数据的井漏智能诊断方法[J]. 石油机械, 2022, 50(11): 16-22., articleTitle=基于工程录井数据的井漏智能诊断方法, refAbstract=null), Reference(id=1228369785168265831, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=11, pageStart=16, pageEnd=22, url=null, language=null, rfNumber=[9], rfOrder=16, authorNames=Chen KaiFeng, Yang Xuewen, Song Xianzhi, journalName=China Petroleum Machinery, refType=null, unstructuredReference=Chen KaiFeng, Yang Xuewen, Song Xianzhi, et al. An intelligent diagnosis method for lost circulation based on engineering logging data[J]. China Petroleum Machinery, 2022, 50(11): 16-22., articleTitle=An intelligent diagnosis method for lost circulation based on engineering logging data, refAbstract=null), Reference(id=1228369785294094958, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2024, volume=52, issue=5, pageStart=1, pageEnd=8, url=null, language=null, rfNumber=[10], rfOrder=17, authorNames=李萍, 于琛, 王建龙, journalName=石油机械, refType=null, unstructuredReference=李萍, 于琛, 王建龙, 等. 基于实时钻进参数的孔隙压力智能预测技术[J]. 石油机械, 2024, 52(5): 1-8., articleTitle=基于实时钻进参数的孔隙压力智能预测技术, refAbstract=null), Reference(id=1228369785424118390, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2024, volume=52, issue=5, pageStart=1, pageEnd=8, url=null, language=null, rfNumber=[10], rfOrder=18, authorNames=Li Ping, Yu Chen, Wang Jianlong, journalName=China Petroleum Machinery, refType=null, unstructuredReference=Li Ping, Yu Chen, Wang Jianlong, et al. Intelligent prediction of pore pressure using real-time drilling parameters[J]. China Petroleum Machinery, 2024, 52(5): 1-8., articleTitle=Intelligent prediction of pore pressure using real-time drilling parameters, refAbstract=null), Reference(id=1228369785575113336, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=135, pageStart=852, pageEnd=868, url=null, language=null, rfNumber=[11], rfOrder=19, authorNames=Karmakar M, Maiti S, journalName=Measurement, refType=null, unstructuredReference=Karmakar M, Maiti S, Short term memory efficient pore pressure prediction via Bayesian neural networks at Bering Sea slope of IODP expedition 323[J]. Measurement, 2019(135): 852-868., articleTitle=Short term memory efficient pore pressure prediction via Bayesian neural networks at Bering Sea slope of IODP expedition 323, refAbstract=null), Reference(id=1228369785667388029, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2013, volume=17, issue=null, pageStart=675, pageEnd=686, url=null, language=null, rfNumber=[12], rfOrder=20, authorNames=Keshavarzi R, Jahanbakhshi R, journalName=European Journal of Environmental and Civil Engineering, refType=null, unstructuredReference=Keshavarzi R, Jahanbakhshi R. Real-time prediction of pore pressure gradient through an artificial intelligence approach: a case study from one of middle east oil fields[J]. European Journal of Environmental and Civil Engineering, 2013, 17: 675-686., articleTitle=Real-time prediction of pore pressure gradient through an artificial intelligence approach: a case study from one of middle east oil fields, refAbstract=null), Reference(id=1228369785734496898, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=44, issue=4, pageStart=506, pageEnd=514, url=null, language=null, rfNumber=[13], rfOrder=21, authorNames=罗发强, 刘景涛, 陈修平, journalName=石油钻采工艺, refType=null, unstructuredReference=罗发强, 刘景涛, 陈修平, 等. 基于BP和LSTM神经网络的顺北油田5号断裂带地层孔隙压力智能预测方法[J]. 石油钻采工艺, 2022, 44(4): 506-514., articleTitle=基于BP和LSTM神经网络的顺北油田5号断裂带地层孔隙压力智能预测方法, refAbstract=null), Reference(id=1228369785818382982, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=44, issue=4, pageStart=506, pageEnd=514, url=null, language=null, rfNumber=[13], rfOrder=22, authorNames=Luo Faqiang, Liu Jingtao, Chen Xiuping, journalName=Oil Drilling & Production Technology, refType=null, unstructuredReference=Luo Faqiang, Liu Jingtao, Chen Xiuping, et al. Intelligent method for predicting formation pore pressure in No.5 fault zone in Shunbei oilfield based on BP and LSTM neural network[J]. Oil Drilling & Production Technology, 2022, 44(4): 506-514., articleTitle=Intelligent method for predicting formation pore pressure in No.5 fault zone in Shunbei oilfield based on BP and LSTM neural network, refAbstract=null), Reference(id=1228369787206697612, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=1, pageStart=12, pageEnd=23, url=null, language=null, rfNumber=[14], rfOrder=23, authorNames=宋先知, 姚学喆, 李根生, journalName=石油科学通报, refType=null, unstructuredReference=宋先知, 姚学喆, 李根生, 等. 基于LSTM-BP神经网络的地层孔隙压力计算方法[J]. 石油科学通报, 2022(1): 12-23., articleTitle=基于LSTM-BP神经网络的地层孔隙压力计算方法, refAbstract=null), Reference(id=1228369787290583696, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=1, pageStart=12, pageEnd=23, url=null, language=null, rfNumber=[14], rfOrder=24, authorNames=Song Xianzhi, Yao Xuezhe, Li Gensheng, journalName=Petroleum Science Bulletin, refType=null, unstructuredReference=Song Xianzhi, Yao Xuezhe, Li Gensheng, et al. A novel method to calculate formation pressure based on the LSTM-BP neural network[J]. Petroleum Science Bulletin, 2022(1): 12-23., articleTitle=A novel method to calculate formation pressure based on the LSTM-BP neural network, refAbstract=null), Reference(id=1228369787357692563, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2019, volume=12, issue=592, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=25, authorNames=Ahmed A, Elkatatny S, Ali A, journalName=Arabian Journal of Geosciences, refType=null, unstructuredReference=Ahmed A, Elkatatny S, Ali A, et al. Comparative analysis of artificial intelligence techniques for formation pressure prediction while drilling[J]. Arabian Journal of Geosciences, 2019, 12(592). DOI: 10.1007/s12517-019-4800-7., articleTitle=Comparative analysis of artificial intelligence techniques for formation pressure prediction while drilling, refAbstract=null), Reference(id=1228369787412218519, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2022, volume=211, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=26, authorNames=Matinkia M, Amraeiniya A, Behboud M M, journalName=Journal of Petroleum Science and Engineering, refType=null, unstructuredReference=Matinkia M, Amraeiniya A, Behboud M M, et al. A novel approach to pore pressure modeling based on conventional well logs using convolutional neural network[J]. Journal of Petroleum Science and Engineering, 211, 2022.DOI: 10.1016/j.petrol.2022.110156., articleTitle=A novel approach to pore pressure modeling based on conventional well logs using convolutional neural network, refAbstract=null), Reference(id=1228369787487715993, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=27, authorNames=Kiranyaz S, Avci O, Abdeljaber O, journalName=Mechanical Systems and Signal Processing, refType=null, unstructuredReference=Kiranyaz S, Avci O, Abdeljaber O, et al. 1D convolutional neural networks and applications: a survey[J]. Mechanical Systems and Signal Processing, 2021, 151.DOI: 10.1016/j.ymssp.2020.107398., articleTitle=1D convolutional neural networks and applications: a survey, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1228369775374565610, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, xref=1, ext=[AuthorCompanyExt(id=1228369775382954219, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775374565610, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Key Laboratory of Drilling and Production Engineering for Oil and Gas, Wuhan 430100, China), AuthorCompanyExt(id=1228369775387148525, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775374565610, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 油气钻采工程湖北省重点实验室, 武汉 430100)]), AuthorCompany(id=1228369775483617521, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, xref=2, ext=[AuthorCompanyExt(id=1228369775496200435, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775483617521, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Petroleum Engineering, Yangtze University, National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan 430100, China), AuthorCompanyExt(id=1228369775504589044, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, companyId=1228369775483617521, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100)])], figs=[ArticleFig(id=1228369779573064095, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Fig.1, caption=Data processing diagram, figureFileSmall=ZjInO8+6AptMipmF7tnq1w==, figureFileBig=2Qw12Y+Rg54EwZw8YhVDJA==, tableContent=null), ArticleFig(id=1228369779661144484, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=图1, caption=数据处理图, figureFileSmall=ZjInO8+6AptMipmF7tnq1w==, figureFileBig=2Qw12Y+Rg54EwZw8YhVDJA==, tableContent=null), ArticleFig(id=1228369779757613483, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Fig.2, caption=Pearson correlation coefficient, figureFileSmall=s34AoID60FU9RvOPv5siAQ==, figureFileBig=tNlyzAv7qqXi187C76LNDA==, tableContent=null), ArticleFig(id=1228369779858276782, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=图2, caption=Pearson相关系数, figureFileSmall=s34AoID60FU9RvOPv5siAQ==, figureFileBig=tNlyzAv7qqXi187C76LNDA==, tableContent=null), ArticleFig(id=1228369779971522994, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Fig.3, caption=CNN-Attn network structure, figureFileSmall=xeTc2idQkshfeJrrlvL3yw==, figureFileBig=eZ0ZtTwqV9dtUvha36EESg==, tableContent=null), ArticleFig(id=1228369780109935032, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=图3, caption=CNN-Attn网络结构, figureFileSmall=xeTc2idQkshfeJrrlvL3yw==, figureFileBig=eZ0ZtTwqV9dtUvha36EESg==, tableContent=null), ArticleFig(id=1228369780311261631, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Fig.4, caption=Forecast diagram, figureFileSmall=25kqBODRoGP0CGCqOqUOug==, figureFileBig=Ypg8G/g6Cc9wp25vnPeRHg==, tableContent=null), ArticleFig(id=1228369780416119234, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=图4, caption=预测示意图, figureFileSmall=25kqBODRoGP0CGCqOqUOug==, figureFileBig=Ypg8G/g6Cc9wp25vnPeRHg==, tableContent=null), ArticleFig(id=1228369780495811014, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Fig.5, caption=Comparison of predicted and actual values of the four models for Well A, figureFileSmall=A6IbPUutmjf2jlSIzOU9ZA==, figureFileBig=j0izwdW7tb+O0kPEXQGYBg==, tableContent=null), ArticleFig(id=1228369780651000268, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=图5, caption=4种模型在A井的预测值与实际值对比, figureFileSmall=A6IbPUutmjf2jlSIzOU9ZA==, figureFileBig=j0izwdW7tb+O0kPEXQGYBg==, tableContent=null), ArticleFig(id=1228369780810383829, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Fig.6, caption=Comparison of predicted and actual valuesof the four models for Well B, figureFileSmall=4KcwOra6Lk4zYpGazgXU1g==, figureFileBig=EEszKCGnyNwWDqEm79i3ZA==, tableContent=null), ArticleFig(id=1228369780936212955, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=图6, caption=4种模型在B井的预测值与实际值对比, figureFileSmall=4KcwOra6Lk4zYpGazgXU1g==, figureFileBig=EEszKCGnyNwWDqEm79i3ZA==, tableContent=null), ArticleFig(id=1228369781041070558, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Fig.7, caption=Generalized prediction results, figureFileSmall=VFfmNAV4jjxjAYr5ehcXhw==, figureFileBig=Vpk7w9zTQtMn9rJpOKwFOg==, tableContent=null), ArticleFig(id=1228369781154316774, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=图7, caption=泛化预测结果, figureFileSmall=VFfmNAV4jjxjAYr5ehcXhw==, figureFileBig=Vpk7w9zTQtMn9rJpOKwFOg==, tableContent=null), ArticleFig(id=1228369781263368685, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Table 1, caption=

Data of Bayan block

, figureFileSmall=null, figureFileBig=null, tableContent=
井深/m 钻压/kN 声波时差/
(μs·ft-1)
自然伽马/
API
岩石密度/
(g·cm-3)
地层压力/
(g·cm-3)
1 154 52 139.9 52.85 2.005 1.042
1 156 63 143.0 50.25 1.994 1.045
1 158 95 142.7 64.54 2.069 1.045
1 160 56 144.2 86.37 2.208 1.047
1 162 87 151.6 87.67 2.090 1.036
), ArticleFig(id=1228369781368226289, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=表1, caption=

巴彦区块数据

, figureFileSmall=null, figureFileBig=null, tableContent=
井深/m 钻压/kN 声波时差/
(μs·ft-1)
自然伽马/
API
岩石密度/
(g·cm-3)
地层压力/
(g·cm-3)
1 154 52 139.9 52.85 2.005 1.042
1 156 63 143.0 50.25 1.994 1.045
1 158 95 142.7 64.54 2.069 1.045
1 160 56 144.2 86.37 2.208 1.047
1 162 87 151.6 87.67 2.090 1.036
), ArticleFig(id=1228369782769123831, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Table 2, caption=

Superparameter optimization results for wells A and B

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 A井单元数 模型 B井单元数
BP 120 BP 300
CNN 70 CNN 70
LSTM 50 LSTM 80
CNN-Attn CNN层70、
全连接层70
CNN-Attn CNN层60、
全连接层80
), ArticleFig(id=1228369782882370045, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=表2, caption=

A井和B井超参数优选结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 A井单元数 模型 B井单元数
BP 120 BP 300
CNN 70 CNN 70
LSTM 50 LSTM 80
CNN-Attn CNN层70、
全连接层70
CNN-Attn CNN层60、
全连接层80
), ArticleFig(id=1228369782970450433, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Table 3, caption=

Results of four model evaluation indexes of well A

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 MAE/MPa RMSE/MPa MRE/% R2
BP 2.32 2.64 3.29 0.55
CNN 1.38 1.67 1.97 0.82
LSTM 1.34 1.71 1.95 0.81
CNN-Attn 0.78 0.97 1.13 0.94
), ArticleFig(id=1228369783104668169, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=表3, caption=

A井的4种模型评价指标结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 MAE/MPa RMSE/MPa MRE/% R2
BP 2.32 2.64 3.29 0.55
CNN 1.38 1.67 1.97 0.82
LSTM 1.34 1.71 1.95 0.81
CNN-Attn 0.78 0.97 1.13 0.94
), ArticleFig(id=1228369783234691599, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=EN, label=Table 4, caption=

Results of four model evaluation indexes of well B

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 MAE/MPa RMSE/MPa MRE/% R2
BP 3.46 3.88 7.81 -1.49
CNN 1.10 1.29 2.48 0.72
LSTM 2.06 2.72 4.71 -0.23
CNN-Attn 0.53 0.65 1.20 0.93
), ArticleFig(id=1228369783352132117, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1228279668579693512, language=CN, label=表4, caption=

B井的4种模型评价指标结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 MAE/MPa RMSE/MPa MRE/% R2
BP 3.46 3.88 7.81 -1.49
CNN 1.10 1.29 2.48 0.72
LSTM 2.06 2.72 4.71 -0.23
CNN-Attn 0.53 0.65 1.20 0.93
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2407065, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2407065, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2407065, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2407065, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
石油、天然气工业基于CNN-Attn神经网络的孔隙压力预测
收藏切换
PDF下载
唐凯 1, 2 , 李忠慧 1, 2, * , 曹天宝 1, 2 , 胡棚杰 1, 2
科学技术与工程 | 论文·石油、天然气工业 2025,25(22): 9335-9341
收起
收藏切换
科学技术与工程 | 论文·石油、天然气工业 2025, 25(22): 9335-9341
石油、天然气工业基于CNN-Attn神经网络的孔隙压力预测
全屏
唐凯1, 2 , 李忠慧1, 2, * , 曹天宝1, 2, 胡棚杰1, 2
作者信息
  • 1 油气钻采工程湖北省重点实验室, 武汉 430100
  • 2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100
  • 唐凯(2001—),男,汉族,湖北仙桃人,硕士研究生。研究方向:人工智能技术与钻井工程。E-mail:

通讯作者:

* 李忠慧(1977—),男,汉族,吉林公主岭人,博士,教授,研究方向:岩石力学与钻完井工程。E-mail:
Pore Pressure Prediction Model Based on CNN-Attn Neural Network
Kai TANG1, 2 , Zhong-hui LI1, 2, * , Tian-bao CAO1, 2, Peng-jie HU1, 2
Affiliations
  • 1 Key Laboratory of Drilling and Production Engineering for Oil and Gas, Wuhan 430100, China
  • 2 School of Petroleum Engineering, Yangtze University, National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan 430100, China
出版时间: 2025-08-08 doi: 10.12404/j.issn.1671-1815.2407065
文章导航
收藏切换

在石油和天然气的勘探开发中,人工智能模型被广泛应用于地层孔隙压力的预测。其中单一模型容易出现过拟合或预测结果不稳定的问题,在预测精度和泛化能力方面仍有提升空间。为了提高地层孔隙压力预测精度,基于深度学习技术,建立了CNN-Attn神经网络地层孔隙压力预测模型。优选了5种测井和随钻数据,使用Pearson相关系数法验证数据与地层孔隙压力的线性相关性。通过对一维CNN的结构进行优化,使模型能够有效捕捉数据的局部特征,并与自注意力机制结合,增强模型对全局依赖关系的捕捉能力,从而提高模型的表现力和理解能力。为了验证该模型的预测精度,对巴彦区块两口井进行预测,预测结果的平均绝对误差均小于1 MPa,均方根误差均小于1 MPa,平均相对误差均小于1.3%,决定系数均大于0.9,比BP、CNN和LSTM模型精度高。该模型提升了地层孔隙压力预测精度,并为钻井安全性提供了数据支持。

地层孔隙压力  /  智能预测  /  深度学习  /  卷积神经网络  /  自注意力机制

In the exploration and exploitation of oil and gas, artificial intelligence models are extensively employed in the prediction of formation pore pressure. Among them, single models tend to encounter problems such as overfitting or unstable prediction outcomes, leaving room for improvement in aspects like prediction accuracy and generalization ability. To enhance the prediction accuracy of formation pore pressure, a CNN-Attn neural network-based formation pore pressure prediction model was established by virtue of deep learning technology. In this research, five types of logging and while-drilling data were optimally selected, and the linear correlation between the data and formation pore pressure was verified using the Pearson correlation coefficient method. Through the optimization of the structure of the one-dimensional CNN, the model can effectively capture the local characteristics of the data and, when combined with the self-attention mechanism, strengthen the model’s ability to capture global dependencies, thereby elevating the model’s expressiveness and comprehension. To validate the prediction accuracy of this model, two wells in the Bayan block were subjected to prediction. The average absolute errors of the prediction results were both less than 1 MPa, the root mean square errors were both less than 1 MPa, the average relative errors were both less than 1.3%, and the determination coefficients were both greater than 0.9, with higher accuracy compared to the BP, CNN, and LSTM models. This model has improved the prediction accuracy of formation pore pressure and provided data support for drilling safety.

formation pore pressure  /  intelligent prediction  /  deep learning  /  convolutional neural network  /  self-attention mechanism
唐凯, 李忠慧, 曹天宝, 胡棚杰. 石油、天然气工业基于CNN-Attn神经网络的孔隙压力预测. 科学技术与工程, 2025 , 25 (22) : 9335 -9341 . DOI: 10.12404/j.issn.1671-1815.2407065
Kai TANG, Zhong-hui LI, Tian-bao CAO, Peng-jie HU. Pore Pressure Prediction Model Based on CNN-Attn Neural Network[J]. Science Technology and Engineering, 2025 , 25 (22) : 9335 -9341 . DOI: 10.12404/j.issn.1671-1815.2407065
在石油和天然气勘探开发中,准确预测地层孔隙压力对于钻井作业的成功至关重要。地层孔隙压力能够有效指导泥浆密度和套管程序的设计,提高钻井时效性[1]。随着向更深的地层钻进,地下环境变得更加复杂,钻井难度随之增大,未能准确预测地层孔隙压力可能导致井壁坍塌、井漏或井喷等井壁稳定问题,加大泥浆设计难度,威胁钻井作业的安全性和增加钻井成本。
当前的孔隙压力预测方法主要包括有效应力法、Eaton法和等效深度法[2]。但由于对经验参数的依赖,逐渐难以满足复杂地层的变化,亟需新的方法予以应对。人工智能技术可以构建出高度智能化的模型,这些模型能够从大数据中自主学习并深入理解相关参数与预测值之间复杂的关系,具有高效性和精准性。例如,毕臣臣[3]针对仅考虑单因素线性关系导致预测结果误差大的问题,提出了基于深度学习的TOC(total organic carbon)含量预测方法,提高了页岩储层TOC含量的预测精度及分辨率。马陇飞等[4]针对钻井取心成本大的问题,建立了测井数据与岩性数据的神经网络模型,准确率达到了71%。杨琨等[5]针对目前缺乏有效预测玛湖砾岩油藏压裂甜点方法的问题和提高水平井压裂改造效果的迫切需求,建立了基于机器学习的致密砾岩油藏压裂甜点预测模型,表现出较好的性能。因此,深度学习等前沿技术已广泛应用于测井、物探和钻完井等领域[6],并在钻井和地层参数预测中展现出巨大潜力[7-9]。近年来,人工智能技术的崛起为地层孔隙压力预测带来了新的可能。研究人员已尝试采用多种人工智能模型来预测孔隙压力,并取得了显著成果。例如,李萍等[10]采用随机森林、XGBoost(eXtreme gradient boosting)和LightGBM(light gradient boosting machine)3种机器学习模型进行预测,结果显示XGBoost和LightGBM的预测趋势更加稳定,而随机森林模型则存在过拟合,使得实际孔隙压力预测过程中误差较大。Karmakar等[11]采用贝叶斯神经网络,考虑了前序地层孔隙压力与当前地层压力之间的联系。Keshavarzi等[12]利用误差反向传播神经网络(back propagation,BP)预测伊朗Asmari油田的孔隙压力梯度,结果显示其预测精度优于Eaton法,但BP神经网络在长序列数据中容易出现梯度消失或梯度爆炸的问题,这会导致地层孔隙压力预测结果偏差大。罗发强等[13]研究了长短期记忆神经网络模型(long short-term memory,LSTM)在地层孔隙压力预测中的表现,发现LSTM模型相较于BP模型具有更好的记忆能力和预测效果。宋先知等[14]将LSTM模型与自适应学习能力的BP模型结合形成的LSTM-BP模型,优于单独的BP、LSTM和SVM模型,提高了地层孔隙压力预测的准确度,表明融合模型是提高预测准确度的一种途径。另一方面,Ahmed等[15] 采用支持向量机(support vector machine,SVM)及其他人工智能工具进行地层孔隙压力预测。这些方法无需依赖正常压力趋势,即可直接预测地层孔隙压力,并且SVM具有预测时间短、效率高的特点。Matinkia等[16]对比了9种人工智能模型,结果显示卷积神经网络(convolutional neural network,CNN)的表现优于SVM等其他模型,突显出CNN在数据研究范围内的高度泛化能力。
尽管BP神经网络、LSTM神经网络和SVM等多种人工智能方法已被广泛应用于地层压力预测,在智能钻井的发展方面有重要作用,但仍存在梯度爆炸的问题、模型的理解能力和泛化能力相对有限,难以精准关注预测中最关键的区域。此外,这些模型对数据的前后波动感知能力不足,难以全面捕捉地层压力的动态变化特征,从而影响预测的精度和可靠性。
针对以上问题,为了在多维度特征数据中构建出稳定的地层孔隙压力预测模型,现通过分析将CNN模型参数进行特定优化,并将CNN捕捉局部信息的能力与自注意力机制相结合,增强模型对数据的理解能力和泛化能力,建立CNN-Attn神经网络的地层孔隙压力预测模型。动态地关注对预测最为关键的区域和特征,增强对全局信息的理解。以期为井壁稳定性分析提供关键数据,帮助预测钻井过程中可能遇到的异常压力,防止井漏、井喷等事故的发生,确保勘探作业的安全性和稳定性。
卷积神经网络CNN是一种具有代表性的深度学习模型,其基本结构包括卷积层、池化层和全连接层。在一维卷积神经网络中[17],卷积层通过卷积核来识别局部模式和特征,卷积核沿输入序列滑动,执行逐元素相乘并求和的操作,生成卷积特征,这有助于网络捕捉输入序列中的时序模式。池化层的池化操作有助于降低卷积层输出的维度,减小计算负担,同时保留关键的特征。
自注意力机制是一种用于神经网络模型中的技术,主要用于捕捉输入数据中不同部分之间的依赖关系。它通过为每个输入元素分配一个权重,以决定该元素在输出中的重要性,从而增强模型的精度和鲁棒性。这些权重是通过计算查询(query)、键(key)和值(value)之间的相似性得出的。具体来说,输入数据被投影到查询、键和值的空间中,然后计算查询与键的点积来获得注意力得分。注意力得分经过softmax归一化后,与值相乘并求和,得到最终的注意力输出。自注意力机制在处理长序列数据时非常有效,因为它能够在一个步骤中考虑到序列中的所有元素之间的关系,避免了传统序列模型中的信息传递瓶颈。其公式为
Attention(Q,K,V)=Softmax$\left(\frac{Q{K}^{\mathrm{T}}}{\sqrt{{d}_{\mathrm{k}}}}\right)$V
式(1)中:QKV为输入矩阵,分别代表查询矩阵、键矩阵和值矩阵;dk为向量维度。
Attention公式的作用是通过对QK的相似度进行加权,来得到对应于输入的V的加权和。
在人工智能应用中,数据处理是一个不可或缺的环节。为了提高地层孔隙压力模型的计算精度和效率,必须对原始的测井数据进行处理,因为这些数据可能由于各种因素出现异常值。通过适当的数据处理,可以有效地去除噪声和异常值,从而为模型提供更准确、更可靠的输入数据。结合钻井领域的专业经验,本文研究中选用伊顿指数法、有效应力法和Dc指数法中用于计算地层孔隙压力的井深、岩石密度、声波时差、自然伽马和钻压5种参数作为输入数据,这些数据与地层孔隙压力相关,并且容易获取、相对完整。首先,对原数据采用滑动平均滤波技术处理;接着,进行相关性分析,计算得出各参数与地层孔隙压力之间的Pearson相关系数;最后,通过归一化处理,将数据范围调整到一个统一标准,避免由于特征范围差异带来的偏差,从而进一步提高地层孔隙压力模型的质量和计算效率。
采用滑动平均滤波技术进行数据处理,通过设定窗口阈值对数据点进行比较,若超过阈值则用窗口内的平均值替换数据点。这种方法有效平滑数据,减小波动,便于观察和分析,同时增强数据的趋势和模式,其算法简单、计算效率高,适用于多种数据类型。以巴彦区块A井数据为例,A井部分数据如表1所示,声波时差处理结果如图1所示,滑动平均滤波技术有效去除了数据中的噪声。
采用Pearson相关系数来评估井深、岩石密度、声波时差、自然伽马和钻压5种参数与地层孔隙压力之间的线性关联性。Pearson相关系数是一种最常用的线性相关系数,它可以反映两个随机变量之间的线性相关程度。Pearson相关系数计算公式为
ρx,y=$\frac{\mathrm{c}\mathrm{o}\mathrm{v}(x,y)}{{\sigma }_{x}{\sigma }_{y}}$
式(2)中:ρx,yx,y的Pearson相关系数;cov(x,y)为x,y的协方差;σxx的标准差;σyy的协方差。
计算结果如图2所示,由统计学知识可知,相关系数为0.4~0.6为中等程度相关,0.6~0.8为强相关,0.8~1.0为极强相关。井深、声波时差、自然伽马和岩石密度4种参数都与地层孔隙压力存在中等及以上的线性相关关系。虽然钻压的相关系数小于0.4,但不可忽略钻压与地层孔隙压力之间的物理联系。
为了减少计算复杂度,提高模型的训练速度,考虑钻井的各个参数是与深度值有很大关系的序列化数据,且不符合正态分布,所以使用最小-最大规范化处理数据。
x'=$\frac{x-\mathrm{m}\mathrm{i}\mathrm{n}\left(x\right)}{\mathrm{m}\mathrm{a}\mathrm{x}\left(x\right)-\mathrm{m}\mathrm{i}\mathrm{n}\left(x\right)}$
式(3)中:x为原始数据;min(x)为数据中的最小值;max(x)为数据中的最大值;x'为标准化后的结果。
针对使用多维度的测井数据预测地层孔隙压力的问题,提出了CNN-Attn组合模型。
一维卷积能够自动从输入数据中提取特征,无需手动设计特征,这使得它在处理复杂的、非线性的地层序列数据时尤为有效。通过卷积操作,1D-CNN能够检测到局部模式和趋势,如短期的变化和周期性特征,这对于预测地层孔隙压力非常有效。卷积核在整个输入序列上共享参数,这减少了模型参数的数量,从而降低了过拟合的风险,并提高了训练效率。因此,CNN可以有效地从这些以深度为序的测井数据中捕获局部信息,如数据变化的趋势或波动。自注意力机制则通过计算特征间的相似性,对特征进行重新加权,使模型能够聚焦于全局信息。对于泛化能力而言,它可以动态调整模型对不同特征的关注度,从而根据输入的变化自动识别最相关的特征,而非依赖所有输入特征,这意味着模型能够更有效地适应多样化的数据。尤其是在预测地层孔隙压力的任务中,数据往往存在不确定性和噪声,自注意力机制可以通过衡量不同特征之间的相关性,过滤掉不相关或噪声特征,让模型更加专注于对预测有用的信息,使模型在不确定或噪声数据的干扰下保持稳健。结合1D-CNN和自注意力机制,模型不仅能捕捉局部特征,还能整合全局依赖关系,显著提升了泛化能力和预测准确性。此外,这种结合使模型在应对复杂、多变的地层孔隙压力参数数据时展现出更强的鲁棒性。
CNN-BP组合模型的结构如图3所示,由输入层、1D-CNN层、自注意机制层和输出层构成。
地层孔隙压力与地层结构有很大关系,即当前井深的地层孔隙压力与前序的测井数据有一定的内在联系。故采用多对一的方式预测,在输入层将数据处理成[步长,特征维度],预测示意图如图4所示。
在1D-CNN层中,卷积核大小的选择对于模型性能至关重要。选择合适的卷积核大小能够有效地捕捉输入数据中的特征和模式。考虑到地质分层和前序地层测井数据与地层孔隙压力之间的短期依赖关系,较小的卷积核更为合适。较小的卷积核能够捕捉局部的细节特征,同时减少计算资源和时间的消耗。步幅决定了卷积核在输入数据上移动的步长,其选择对模型的性能和计算效率有重要影响。由于模型使用的数据维度为5,属于小维度数据,步幅为1可以确保卷积核在输入数据的每个位置都进行计算,从而不遗漏任何细节信息。
在自注意力机制层中,自注意力机制可以帮助模型更好地捕捉输入数据中的长期依赖关系,通过全连接的方式将信息综合,输出预测结果。
在训练过程中需要调整的参数主要有:卷积核大小、单元数、学习率、批量大小、步长、优化器和损失函数。
卷积核大小则影响卷积操作捕捉的特征尺寸,适当的大小有助于更好地理解数据的局部结构,通过控制变量法,当卷积核大小设置为2、3、4时,均方根误差会逐步增大,因此选用2。
适当的单元数可以提高模型容量和表达能力,有助于捕捉复杂的数据模式,本文使用随机搜索的方法确定单元数的值。
学习率设置过大可能导致训练过程不稳定,而设置过小可能使训练过程过于缓慢;批量大小是每次更新权重时使用的样本数,它直接影响训练的稳定性和速度,根据调参经验分别设置为10-4、64。
步长是指使用多少条数据预测一条数据,它影响模型对序列数据的理解程度。为更好地学习局部特征和模式,使用控制变量法得出步长为12、15、18时,均方根误差先减小后增大,因此设置为15。
优化器是用于更新权重的算法,选择适当的优化器有助于提高模型的收敛性和泛化能力,因此选择能自适应调整学习率的Adam优化器。
损失函数用于衡量模型的预测结果与实际结果的差距,选择适当的损失函数能够引导模型更好地学习任务目标,预测地层孔隙压力属于回归任务,故损失函数选用均方根误差。
为了验证本文的CNN-Attn组合模型预测地层孔隙压力的性能,将该模型与BP、CNN和LSTM模型进行对比,使用模型评价指标进行验证。
模型评价指标可以用来衡量模型是否有效,可以用来调整模型的参数。本文所研究的地层孔隙压力预测属于回归问题,评价指标用来衡量预测值与真实值之间的误差,故采用的模型评价指标为平均绝对误差(mean absolute error,MAE)、均方根误差(root mean squared error,RMSE)、平均相对误差(mean relative error,MRE)和决定系数(determination coefficient,R2),其中MAE、RMSE和MRE越小,R2越大,模型预测能力越强。
平均绝对误差(MAE)是实测值与预测值之间绝对误差的平均值,为预测值误差的体现,其计算公式为
$\mathrm{MAE}=\frac{1}{N}\sum_{i=1}^{N}\mid\hat{y}_{i}-y_{i}\mid$
式(4)中:${\dot{y}}_{i}$为参数数据的实测值;yi为模型预测值;N为数据数量。
均方根误差(RMSE)为实测值与真实值之间偏差的体现,计算公式为
RMSE=$\sqrt{\frac{1}{N}\stackrel{N}{\sum _{i=1}}({\dot{y}}_{i}-{y}_{i}{)}^{2}}$
平均相对误差指真实值与预测值的相对误差的平均值,其计算公式为
MRE=$\frac{1}{N}\stackrel{N}{\sum _{i=1}}\frac{|{\dot{y}}_{i}-{y}_{i}|}{{y}_{i}}$×100%
决定系数(R2)为衡量预测模型能力的指标。其计算公式为
R2=1-$\frac{\stackrel{N}{\sum _{i=1}}({\dot{y}}_{i}-{y}_{i}{)}^{2}}{\stackrel{N}{\sum _{i=1}}({y}_{i}-{\stackrel{-}{y}}_{i}{)}^{2}}$
使用巴彦区块A井和B井的数据进行实验,A井数据涵盖井深1 154~6 344 m,B井数据涵盖井深1 007~4 289 m。实验采用留出法,前80%的数据用于训练模型,其余20%用于测试。经过对数据进行预处理和模型超参数优选,使用CNN-Attn模型进行预测,并与BP、CNN和LSTM模型进行比较。
A、 B两口井按经验设定和随机搜索得出最优模型参数,如表2所示。
使用上述优选的模型参数进行地层孔隙压力预测,A、B两口井评价指标结果如表3表4所示,预测结果如图5图6所示。
结果显示CNN-Attn模型的MAE均小于1 MPa,RMSE均小于1 MPa,MRE均小于1.3%,决定系数均大于0.9,其评价指标数据均优于BP神经网络模型、CNN神经网络模型和LSTM神经网络模型,表明CNN-Attn模型能更好地处理测井的序列性数据,且预测的准确度更高。由图5图6可知,BP神经网络的预测结果在整体上有所偏差,可能由于无法保留和利用序列中的信息,所以预测效果相比于LSTM神经网络较差。CNN神经网络模型预测结果相对较好。CNN神经网络模型与自注意力机制结合后,可以根据特征的重要性动态调整权重,并有效捕捉短期和长期趋势,提高预测准确性。
单井模型以巴彦区块A井自身的数据训练,并预测自身的后序深度的地层孔隙压力。在实际钻井中,更希望能通过打一口井或若干口井对同一区块的地层孔隙压力进行预测,故将巴彦区块A井训练好的模型用来预测B井的地层孔隙压力,以测试模型的泛化能力。预测的结果如图7所示,其MAE为0.523 MPa,RMSE为0.628 MPa,MRE为1.208%,决定系数为0.935。由此可以得出CNN-Attn模型具有良好的泛化能力,在实际钻井中,可以为现场提供更多的数据参考。
(1)选定井深、岩石密度、声波时差、自然伽马和钻压5种数据作为输入参数并进行相关性分析,利用CNN发卷积操作提取特征并检测局部模式和趋势,结合自注意力机制擅长捕捉全局依赖关系,从而创建出CNN-Attn地层孔隙压力预测模型。
(2)使用巴彦区块两口井的数据验证模型的性能,对比BP、CNN和LSTM模型的结果显示,CNN-Attn模型能够有效地预测地层孔隙压力。使用当前井数据建立模型时,两口井的MAE均小于1 MPa,RMSE均小于1 MPa,MRE均小于1.3%,决定系数均大于0.9,说明模型具有较高的精度和稳定性;泛化能力较强,检验结果平均绝对误差小于1 MPa、决定系数大于0.9。综上所述,本文研究采用人工智能技术提供了一种新的方法来预测地层孔隙压力,为钻井安全提供了有力的数据参考。
(3)在后续的研究中,将考虑设计更合适的模型结构,多通道挖掘序列特征,以及更全面地考虑浅层和深层的地层信息,融合环境因素(如地质结构、岩性等)作为额外输入,进一步提高预测精度。
参考文献 引证文献
排序方式:
[1]
Zhao Y, Sun T, Yang J, et al. Combining drilling big data and machine learning method toimprove the timeliness of drilling[C]// SPE/IADC Drilling Conference and Exhibition. Hague: SPE, 2019: 194111-MS.
[2]
余金秋, 孙夕平, 于永才, 等. 深层地层压力预测技术研究进展[J]. 海相油气地质2024, 29(4): 337-347.
Yu Jinqiu, Sun Xiping, Yu Yongcai, et al. Research progress on deep formation pressure prediction technology[J]. Marine Origin Petroleum Geology, 2024, 29(4): 337-347.
[3]
毕臣臣. 基于深度学习的页岩储层总有机碳含量预测方法[J]. 科学技术与工程, 2023, 23(2): 494-501.
Bi Chenchen. Total organic carbon content prediction method of shale reservoir based on deep learning[J]. Science Technology and Engineering, 2023, 23(2): 494-501.
[4]
马陇飞, 萧汉敏, 陶敬伟, 等. 基于深度学习岩性分类的研究与应用[J]. 科学技术与工程, 2022, 22(7): 2609-2617.
Ma Longfei, Xiao Hanmin, Tao Jingwei, et al. Research and application of lithology classification based on deep learnning[J]. Science Technology and Engineering, 2022, 22(7): 2609-2617.
[5]
杨琨, 罗山贵, 花凌旭, 等. 致密砾岩油藏压裂甜点预测研究——以玛18井区为例[J]. 科学技术与工程, 2022, 22(32): 14174-14183.
Yang Kun, Luo Shangui, Hua Lingxu, et al. Investigation on the fracturing sweet spot predliction of conglomerate tight oil reservoir: a case study of the Mal8 Well Block[J]. Science Technology and Engineering, 2022, 22(32): 14174-14183.
[6]
匡立春, 刘合, 任义丽, 等. 人工智能在石油勘探开发领域的应用现状与发展趋势[J]. 石油勘探与开发, 2021, 48(1): 1-11.
Kuang Lichun, Liu He, Ren Yili, et al. Application and development trend of artificial intelligence in petroleum exploration and development[J]. Petroleum Exploration and Development, 2021, 48(1): 1-11.
[7]
李根生, 宋先知, 祝兆鹏, 等. 智能钻完井技术研究进展与前景展望[J]. 石油钻探技术, 2023, 51(4): 35-47.
Li Gensheng, Song Xianzhi, Zhu Zhaopeng, et al. Research progress and the prospect of intelligent drilling and completion techno-logies[J]. Petroleum Drilling Techniques, 2023, 51(4): 35-47.
[8]
张瑞, 祝兆鹏, 李大钰, 等. 基于改进时序网络的钻进参数可解释实时预测[J]. 石油机械, 2024, 52(4): 1-10.
Zhang Rui, Zhu Zhaopeng, Li Dayu, et al. Interpretable real-time prediction of dilling parameters based on improved sequential network[J]. China Petroleum Machinery, 2024, 52(4): 1-10.
[9]
陈凯枫, 杨学文, 宋先知, 等. 基于工程录井数据的井漏智能诊断方法[J]. 石油机械, 2022, 50(11): 16-22.
Chen KaiFeng, Yang Xuewen, Song Xianzhi, et al. An intelligent diagnosis method for lost circulation based on engineering logging data[J]. China Petroleum Machinery, 2022, 50(11): 16-22.
[10]
李萍, 于琛, 王建龙, 等. 基于实时钻进参数的孔隙压力智能预测技术[J]. 石油机械, 2024, 52(5): 1-8.
Li Ping, Yu Chen, Wang Jianlong, et al. Intelligent prediction of pore pressure using real-time drilling parameters[J]. China Petroleum Machinery, 2024, 52(5): 1-8.
[11]
Karmakar M, Maiti S, Short term memory efficient pore pressure prediction via Bayesian neural networks at Bering Sea slope of IODP expedition 323[J]. Measurement, 2019(135): 852-868.
[12]
Keshavarzi R, Jahanbakhshi R. Real-time prediction of pore pressure gradient through an artificial intelligence approach: a case study from one of middle east oil fields[J]. European Journal of Environmental and Civil Engineering, 2013, 17: 675-686.
[13]
罗发强, 刘景涛, 陈修平, 等. 基于BP和LSTM神经网络的顺北油田5号断裂带地层孔隙压力智能预测方法[J]. 石油钻采工艺, 2022, 44(4): 506-514.
Luo Faqiang, Liu Jingtao, Chen Xiuping, et al. Intelligent method for predicting formation pore pressure in No.5 fault zone in Shunbei oilfield based on BP and LSTM neural network[J]. Oil Drilling & Production Technology, 2022, 44(4): 506-514.
[14]
宋先知, 姚学喆, 李根生, 等. 基于LSTM-BP神经网络的地层孔隙压力计算方法[J]. 石油科学通报, 2022(1): 12-23.
Song Xianzhi, Yao Xuezhe, Li Gensheng, et al. A novel method to calculate formation pressure based on the LSTM-BP neural network[J]. Petroleum Science Bulletin, 2022(1): 12-23.
[15]
Ahmed A, Elkatatny S, Ali A, et al. Comparative analysis of artificial intelligence techniques for formation pressure prediction while drilling[J]. Arabian Journal of Geosciences, 2019, 12(592). DOI: 10.1007/s12517-019-4800-7.
[16]
Matinkia M, Amraeiniya A, Behboud M M, et al. A novel approach to pore pressure modeling based on conventional well logs using convolutional neural network[J]. Journal of Petroleum Science and Engineering, 211, 2022.DOI: 10.1016/j.petrol.2022.110156.
[17]
Kiranyaz S, Avci O, Abdeljaber O, et al. 1D convolutional neural networks and applications: a survey[J]. Mechanical Systems and Signal Processing, 2021, 151.DOI: 10.1016/j.ymssp.2020.107398.
2025年第25卷第22期
PDF下载
145
65
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2407065
  • 接收时间:2024-09-21
  • 首发时间:2026-02-11
  • 出版时间:2025-08-08
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-09-21
  • 修回日期:2025-05-13
基金
作者信息
    1 油气钻采工程湖北省重点实验室, 武汉 430100
    2 长江大学石油工程学院油气钻完井技术国家工程研究中心, 武汉 430100

通讯作者:

* 李忠慧(1977—),男,汉族,吉林公主岭人,博士,教授,研究方向:岩石力学与钻完井工程。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2407065
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏