[1]祝元宠,咸玉席,李清宇,等. 基于大数据的页岩气产能预测[J].油气井测试,2019,28(01):1-6.[doi:10.19680/j.cnki.1004-4388.2019.01.001]
 ZHU Yuanchong,XIAN Yuxi,LI Qingyu,et al.Shale gas productivity forecast based on big data[J].Well Testing,2019,28(01):1-6.[doi:10.19680/j.cnki.1004-4388.2019.01.001]
点击复制

 基于大数据的页岩气产能预测()
分享到:

《油气井测试》[ISSN:1006-6977/CN:61-1281/TN]

卷:
28
期数:
2019年01期
页码:
1-6
栏目:
出版日期:
2019-02-25

文章信息/Info

Title:
Shale gas productivity forecast based on big data
文章编号:
1004-4388(2019)01-0001-06
作者:
 祝元宠 咸玉席 李清宇 卢徳唐
 中国科学技术大学石油天然气研究中心 安徽合肥 230026
Author(s):
ZHU Yuanchong XIAN Yuxi LI Qingyu LU Detang
 Research Center of Petroleum, University of Science and Technology of China, Hefei, Anhui 230026
关键词:
页岩气 产能预测 大数据 非参数拟合 支持向量机 长宁气田
Keywords:
shale gas productivity forecast big data nonparametric matching support vector machine (SVM) Changning shale gas field
分类号:
TE353
DOI:
10.19680/j.cnki.1004-4388.2019.01.001
文献标志码:
A
摘要:
 

参数拟合的传统页岩气井产能预测方法存在一定的局限性,引入基于支持向量机的非参数大数据分析方法进行页岩气井产能预测研究。根据

生产数据记录以及井底压力随生产过程的变化规律,建立eSVR支持向量回归模型,对长宁页岩气某区块实际生产数据分别进行了单井和多井的训练及预测检验,其中单

井回归检验的相关度系数达到0958 556,体现了该方法优秀的回归能力;多井学习模型在前95 d区间内对单井数据的预测也达到了接近单井回归的效果,体现了该方

法在密集数据区间内较好的预测能力,为页岩气产能预测提供了新的思路。

Abstract:
The traditional productivity forecast method of shale gas well based on parameter matching has some limitations. This paper introduced a nonparametric big data analysis method based on support vector machine (SVM). According to the recorded production data and the change law of bottomhole pressure with the production process, the eSVR (support vector regression) model was established, and it was tested by singlewell and multiwell training and forecasting based on actual production data of a block of Changning shale gas field. The correlation coefficient of singlewell regression test reaches 0958 556, which reflects the excellent regression performance of this method. The forecasting of the multiwell learning model based on single well data in the first 95 d interval is close to the effect of the singlewell regression model, which reflects the better forecasting performance of the method in the dense data interval. The proposed method provides a new idea for forecasting shale gas productivity.

相似文献/References:

[1]邹顺良. 页岩气微注压降测试方法[J].油气井测试,2018,27(01):37.[doi:10.19680/j.cnki.1004-4388.2018.01.006]
 ZOU Shunliang. Shale gas injection/fall off testing method[J].Well Testing,2018,27(01):37.[doi:10.19680/j.cnki.1004-4388.2018.01.006]
[2]谢德湘,靳凯,易志东,等.考虑多效应作用的页岩气变流量压力数据分析方法[J].油气井测试,2018,27(03):14.[doi:10.19680/j.cnki.1044-4388.2018.03.003]
 XIE Dexiang,JIN Kai,YI Zhidong,et al.Pressure data analysis method for shale gas in production with variable rate with consideration of multieffect[J].Well Testing,2018,27(01):14.[doi:10.19680/j.cnki.1044-4388.2018.03.003]
[3]任勇,郭彪,石孝志,等.页岩气套变水平井连续油管钻磨复合桥塞技术[J].油气井测试,2018,27(04):61.[doi:10.19680/j.cnki.1004-4388.2018.04.010]
 [J].Well Testing,2018,27(01):61.[doi:10.19680/j.cnki.1004-4388.2018.04.010]
[4]肖寒.威远区块页岩气水平井基于灰色关联分析的产能评价方法[J].油气井测试,2018,27(04):73.[doi:10.19680/j.cnki.1004-4388.2018.04.012]
 [J].Well Testing,2018,27(01):73.[doi:10.19680/j.cnki.1004-4388.2018.04.012]
[5]王伟佳.页岩气井无限级固井滑套压裂技术 [J].油气井测试,2018,27(05):37.[doi:10.19680/j.cnki.1004-4388.2018.05.007]
 WANG Weijia.Infinitestage cementing sleeve fracturing technology for shale gas wells[J].Well Testing,2018,27(01):37.[doi:10.19680/j.cnki.1004-4388.2018.05.007]
[6]章雨,李少华,李俊仪,等.环江油田长6储层基于多元回归分析的产能评价[J].油气井测试,2019,28(02):68.[doi:10.19680/j.cnki.1004-4388.2019.02.012]
 ZHANG Yu,LI Shaohua,LI Junyi,et al.Productivity evaluation of Chang 6 formation in Huanjiang Oilfield based on multivariate regression analysis[J].Well Testing,2019,28(01):68.[doi:10.19680/j.cnki.1004-4388.2019.02.012]
[7]杜成良,李琳琳,徐凤兰,等. 底水驱基岩气藏全生命周期产能预测方法[J].油气井测试,2019,28(04):64.[doi:10.19680/j.cnki.1004-4388.2019.04.011]
 DU Chengliang,LI Linlin,XU Fenglan,et al.Full lifecycle productivity forecast for bedrock gas reservoir with bottomaquifer drive[J].Well Testing,2019,28(01):64.[doi:10.19680/j.cnki.1004-4388.2019.04.011]
[8]刘长印.焦石坝地区页岩气水平井压裂快速选段方法[J].油气井测试,2019,28(06):9.[doi:10.19680/j.cnki.1004-4388.2019.06.002]
 LIU Changyin.Rapid fracturing layer selecting method for horizontal shale gas wells in Jiaoshiba area[J].Well Testing,2019,28(01):9.[doi:10.19680/j.cnki.1004-4388.2019.06.002]
[9]曾凌翔.页岩气水平井增产改造关键参数分析[J].油气井测试,2020,29(05):27.[doi:10.19680/j.cnki.1004-4388.2020.05.005]
 ZENG Lingxiang.Analysis of key stimulation parameters for shale gas horizontal wells[J].Well Testing,2020,29(01):27.[doi:10.19680/j.cnki.1004-4388.2020.05.005]
[10]胡小虎,王妍妍,刘华,等.考虑拟时间的页岩气压裂水平井生产数据分析方法[J].油气井测试,2021,30(01):66.[doi:10.19680/j.cnki.1004-4388.2021.01.012]
 HU Xiaohu,WANG Yanyan,LIU Hua,et al.Production data analysis method of shale gas horizontal well fracturing considering pseudo time[J].Well Testing,2021,30(01):66.[doi:10.19680/j.cnki.1004-4388.2021.01.012]

备注/Memo

备注/Memo:
 


2018-10-25 收稿,2018-12-20 修回,2019-01-09 接受,2019-02-18 网络版发表
国家科技重大专项“致密油气藏多尺度介质复杂结构井数值试井分析方法及应用研究”(2017ZX05009005-002)、中石油-中科院重大战略合作

项目“页岩气钻完井井壁稳定与开发工程技术研究”(2015A-4812)、中国科学院战略先导科技专项“页岩气勘探开发基础理论与关键技术”(XDB10030402)

祝元宠,男,1988年出生,中国科学技术大学近代力学系在读硕士研究生,主要研究方向为致密油气藏

试井及产能预测。电话:158567984326;Email:redsky@mail.ustc.edu.cn。通信地址:安徽省合肥市中国科学技术大学西区力四楼616,邮政编码:230026。

更新日期/Last Update: 2019-02-12