白洋^1,,
谭茂金^1,,,
肖承文²,
韩闯²,
武宏亮³,
罗伟平²,
徐彬森³
1. 中国地质大学(北京)地球物理与信息技术学院, 北京 100083
2. 中国石油塔里木油田公司勘探开发研究院, 新疆库尔勒 841000
3. 中国石油勘探开发研究院, 北京 100083

基金项目: 国家自然科学基金（41774144），中国石油塔里木油田分公司委托课题（041018120073）及页岩油气富集机理与有效开发国家重点实验室开放课题（20-YYGZ-KF-GC-10）资助


详细信息

作者简介: 白洋, 男, 1996年生, 博士研究生.E-mail: baiyang@cugb.edu.cn

通讯作者: 谭茂金, 男, 1973年生, 教授, 博士生导师, 主要从事地球物理测井理论、复杂油气藏和非常规测井解释与岩石物理研究.E-mail: tanmj@cugb.edu.cn

中图分类号: P631

收稿日期:2020-08-14
修回日期:2021-01-08
上线日期:2021-05-10

Dynamic classification committee machine-based fluid typing method from wireline logs for tight sandstone gas reservoir

BAI Yang^1,,
TAN MaoJin^1,,,
XIAO ChengWen²,
HAN Chuang²,
WU HongLiang³,
LUO WeiPing²,
XU BinSen³
1. School of Geophysics and Information Technology, China University of Geosciences, Beijing 100083, China
2. Research Institute of Exploration and Development, Tarim Oilfield Company, PetroChina, Xinjiang Korla 841000, China
3. Research Institute of Petroleum Exploration and Development, PetroChina Co Ltd., Beijing 100083, China


More Information

Corresponding author: TAN MaoJin,E-mail:tanmj@cugb.edu.cn

MSC: P631

--> Received Date: 14 August 2020
Revised Date: 08 January 2021
Available Online: 10 May 2021


摘要
摘要:致密砂岩流体识别难度大，智能算法能够较好地建立其流体识别模型.相较于单一智能算法，分类委员会机器通过联合多个专家（智能算法）有助于提升智能模型整体性能.而针对分类委员会机器中单个专家性能难以提升的问题，添加门网络构建动态分类委员会机器是一种更有效的模块化学习方式.本研究首先采用门网络将输入数据划分为多个子数据集，然后联合决策树、概率神经网络、贝叶斯分类、BP神经网络、最近邻算法分别训练子数据集得到多个子模型，最后利用组合器最优化子模型组合得到最佳的流体识别模型.针对塔里木盆地库车坳陷大北、克深、博孜地区致密砂岩地层测井数据和测试数据，采用平均影响值法优选敏感测井系列作为输入，构建了动态的测井流体识别模型，其训练、验证准确率分别为96.29%和91.39%.利用此模型以BZ9井为例进行流体类型判别，预测结果与测试结果一致.该方法将无监督与有监督学习相结合，引入门网络提高了数据集利用效率，避免了数据集分布不均衡对模型构建的影响；采用投票机制集成多种专家，建立了子模型与专家的适应关系，流体识别模型预测精度和泛化能力大大提高.
关键词: 致密气/
流体识别/
智能算法/
聚类/
动态分类委员会机器

Abstract:Tight sandstone fluid identification is difficult, and the intelligent algorithms can help to establish the fluid identification model of tight sandstone. Compared with the individual intelligent algorithm, the classification committee machine could improve the overall performance of the intelligent model by combining multiple experts (intelligent algorithms). For the limitation that the performance of individual expert is difficult to improve in the classification committee machine, adding a gate network to build the dynamic classification committee machine is a more effective approach to modular learning. In this study, firstly, the input data is divided into multiple sub-datasets by using the gate network, and then the sub-datasets are trained by the decision tree, probabilistic neural network, Bayesian classifier, BP neural network, and nearest neighbor algorithm to obtain multiple submodels. Finally, the combiner is used to optimize the submodel combination to obtain the best fluid recognition model. Based on the logs and test data of the tight sandstone formation in the Dabei, Keshen and Bozi block of Kuqa Depression in Tarim Basin, the mean influence value method was used to obtain the sensitive logs data, and then a dynamic fluid typing model was constructed. The training and verification accuracy was 96.29% and 91.39%. Taking well BZ9 as an example, the fluid type was predicted, and the prediction result is agreement with the test result. This method combines unsupervised and supervised learning, introduces the gate network to improve the dataset utilization, and avoids the impact of uneven data distribution on model training. The voting strategy is used to integrate multiple experts and establish the correspondence between the submodels and the experts. The prediction accuracy and generalization of the fluid identification model are greatly improved.
Key words:Tight gas/
Fluid identification/
Intelligent algorithm/
Clustering/
Dynamic classification committee machine



PDF全文下载地址:

http://www.geophy.cn/data/article/export-pdf?id=2c2ece41-2b7c-497d-bb24-08bf931185e2