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

被引：0

作者：

Bai Y. ^{[1
]}

Tan M. ^{[1
]}

Xiao C. ^{[2
]}

Han C. ^{[2
]}

Wu H. ^{[3
]}

Luo W. ^{[2
]}

Xu B. ^{[3
]}

机构：

[1] School of Geophysics and Information Technology, China University of Geosciences, Beijing

[2] Research Institute of Exploration and Development, Tarim Oilfield Company, PetroChina, Korla

[3] Research Institute of Petroleum Exploration and Development, PetroChina Co Ltd., Beijing

来源：

Acta Geophysica Sinica | 2021年 / 64卷 / 05期

关键词：

Clustering; Dynamic classification committee machine; Fluid identification; Intelligent algorithm; Tight gas;

D O I：

10.6038/cjg2021O0311

中图分类号：

学科分类号：

摘要：

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. © 2021, Science Press. All right reserved.

引用

页码：1745 / 1758

页数：13

共 25 条

[1] Ansari H R., Use seismic colored inversion and power law committee machine based on imperial competitive algorithm for improving porosity prediction in a heterogeneous reservoir, Journal of Applied Geophysics, 108, pp. 61-68, (2014)
[2] Barzegar R, Moghaddam A A, Baghban H., A supervised committee machine artificial intelligent for improving DRASTIC method to assess groundwater contamination risk: a case study from Tabriz plain aquifer, Iran, Stochastic Environmental Research and Risk Assessment, 30, 3, pp. 883-899, (2016)
[3] Chai M R, Cheng D, Zhang C M, Et al., Prediction of debris composition in glutenite by machine learning method: a case study in Baikouquan formation of Well X723 in the NW Margin of Junggar basin, Journal of Xi'an Shiyou University (Natural Science Edition), 32, 5, pp. 22-28, (2017)
[4] Chen J, Xie R C, Liu C C, Et al., Logging fluid identification and quantitative evaluation of Jurassic tight sandstone gas reservoirs in Zhongjiang Gas Field, Natural Gas Industry, 39, S1, pp. 136-141, (2019)
[5] Dashti A, Harami H R, Rezakazemi M., Accurate prediction of solubility of gases within H<sub>2</sub>-selective nanocomposite membranes using committee machine intelligent system, International Journal of Hydrogen Energy, 43, 13, pp. 6614-6624, (2018)
[6] Ding S C., Evaluation method study of tight sandstone oil and gas reservoir of block Ⅰ in Middle Zhungaer basin, (2014)
[7] Fan R, Zhou L, Wu J, Et al., Research on tight sandstone reservoir fluids identification in Xujiahe Formation, northeastern Sichuan basin, Petroleum Geology and Recovery Efficiency, 22, 3, pp. 67-71, (2015)
[8] Ghiasi-Freez J, Kadkhodaie-Ilkhchi A, Ziaii M., The application of committee machine with intelligent systems to the prediction of permeability from petrographic image analysis and well logs data: a case study from the South Pars Gas field, south Iran, Petroleum Science and Technology, 30, 20, pp. 2122-2136, (2012)
[9] Guo H X, Li Y J, Li Y A, Et al., BPSO-Adaboost-KNN ensemble learning algorithm for multi-class imbalanced data classification, Engineering Applications of Artificial Intelligence, 49, pp. 176-193, (2016)
[10] Hong Z, Zhang M G, Zhu X M., Prediction on lithology and fluid probabilities of tight clastic gas reservoir based on rock physics, Geophysical Prospecting for Petroleum, 54, 6, pp. 735-744, (2015)

← 1 2 3 →