[1]于伟强,刘均荣,左翊寅,等.永久式井下压力计数据生产阶段识别方法[J].油气井测试,2018,27(02):14-21.[doi:10.19680/j.cnki.1004-4388.2018.02.003]
　YU Weiqiang,LIU Junrong,ZUO Yiyin,et al.Identification of production stages based on data from permanent downhole pressure gauge[J].Well Testing,2018,27(02):14-21.[doi:10.19680/j.cnki.1004-4388.2018.02.003]
点击复制

永久式井下压力计数据生产阶段识别方法()
分享到：

《油气井测试》[ISSN:1006-6977/CN:61-1281/TN]

卷:
27
期数:
2018年02期
页码:
14-21
栏目:
出版日期:
2018-04-25

文章信息/Info

Title:
Identification of production stages based on data from permanent downhole pressure gauge
文章编号:
1004-4388(2018)02-0014-08
作者:
于伟强1;  刘均荣2;  左翊寅1;  王东海3;  高鸿泉3
1. 中法渤海地质服务有限公司 天津 300452 
2. 中国石油大学（华东）石油工程学院 山东青岛 266580
3. 中国石油集团渤海钻探工程公司油气井测试分公司 河北廊坊 065007
Author(s):
YU Weiqiang1;  LIU Junrong2;  ZUO Yiyin1;  WANG Donghai3;  GAO Hongquan3
1.China France Bohai Geoservices Company, Tianjin 300452 China
2.China University of Petroleum(East China), Qingdao, Shandong 266580 China
3. Well Testing Branch of Bohai Drilling Engineering Co. Ltd., Langfang, Hebei 065000 China
关键词:
井下压力计; 识别; 油藏; 动态监测; 噪声; 小波; 卷积; 阈值; 反褶积
Keywords:
downhole pressure gauge;  identification;  reservoir;  dynamic monitoring;  noise;  wavelet;  convolution;  threshold value;  deconvolution
分类号:
TE353
DOI:
10.19680/j.cnki.1004-4388.2018.02.003
文献标志码:
A
摘要:
由于井底流动情况复杂，永久式井下压力计数据常含有不同程度噪声，使划分生产阶段的工作量巨大且困难，直接影响地层参数解释结果的准确性。分别在数据中加入方差为0005和0010的高斯白噪声，对比分析小波模极值方法和基于滤波器的分步卷积方法识别永久式压力数据生产阶段突变点过程中对噪声的敏感程度，探索生产阶段噪声敏感性较小的识别方法。结果表明，小波模极值方法错误识别概率较高，对噪声比较敏感，在划分生产阶段前需要进行数据降噪处理和优选小波类型及判断阈值；而改进的分步滤波卷积方法能够准确、有效地识别不同生产阶段，并且对噪声的敏感性小，不需要窗函数的优选过程，为自动处理庞大的永久式井下压力计数据提供了一种新的手段。
Abstract:
Due to downhole complexities, data from permanent downhole pressure gauge often contain certain noises. This makes the division of production stages extremely challenging and onerous, thus compromising the accuracy of formation parameter interpretation results. Gaussian white noises with variance of 0005 and 0010 were added to the data to figure out the sensitivities of the wavelet module extremum method and the filterbased stepwise convolution method to noises when they were used in identification of the breakthrough point of production stage based on the data from permanent pressure gauge. As revealed by the results, the wavelet module extremum method has a high probability of error and is quite sensitive to noises; moreover, it requires data denoising, selection of optimal wavelet types and determination of threshold value prior to division of production stages. In contrast, the modified filterbased stepwise convolution method can identify different production stages effectively and accurately, with low sensitivity to noises, and does not need the selection of optimal window function. Thus, the modified filterbased stepwise convolution method provides an innovative alternative for automatic processing of massive data from permanent downhole pressure gauge.

相似文献/References:

[1]王克林,杨向同,刘军严,等.库车山前高压气井钢丝投捞式试井技术[J].油气井测试,2018,27(04):15.[doi:10.19680.j.cnki.1004-4388.2018.04.003]
　[J].Well Testing,2018,27(02):15.[doi:10.19680.j.cnki.1004-4388.2018.04.003]
[2]姜楠. 新型低频电磁波井下数据无线传输技术[J].油气井测试,2022,31(06):54.[doi:10.19680/j.cnki.1004-4388.2022.06.011]
　[J].Well Testing,2022,31(02):54.[doi:10.19680/j.cnki.1004-4388.2022.06.011]

更新日期/Last Update: 2018-06-16