基于不同机器学习模型的石油测井数据岩性分类对比研究

doi:10.11720/wtyht.2024.1492

Abstract
Figure/Table
References
Related Citation (12)

Download: PDF (2671 KB) HTML (0 KB)
Export: BibTeX | EndNote (RIS)

Abstract

Specific computational tools assist geologists in identifying and classifying the lithology of rocks in oil well exploration,reducing costs,and enhancing operational efficiency. Machine learning methods integrate a vast amount of information,enabling efficient pattern recognition and accurate decision-making. This article categorizes the lithology of five oil wells in the Norwegian Sea,randomly dividing the data into a training set (70%) and a test set (30%). Using multivariate well log parameter data for training and validation,the application effectiveness of models such as Multilayer Perceptron (MLP),Decision Tree,Random Forest,and XGBoost is compared. The research results indicate that the XGBoost model outperforms others in terms of data generalization,achieving an accuracy of 95%. The Random Forest model follows with an accuracy of 94%. Meanwhile,Multilayer Perceptron (MLP) and Decision Tree models exhibit good robustness,with accuracies of 92% and 90%,respectively.

Key words： lithology identification machine learning oil logging XGBoost gorithm random forest

Received: 22 November 2023 Published: 16 April 2024

ZTFLH:

P631

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Li JIANG
	Zhi-Mo ZHANG
	Qi-Wei WANG
	Zhi-Bing FENG
	Bo-Cheng ZHANG
	Teng-Fei REN

Cite this article:

Li JIANG,Zhi-Mo ZHANG,Qi-Wei WANG, et al. Comparative study on lithology classification of oil logging data based on different machine learning models[J]. Geophysical and Geochemical Exploration, 2024, 48(2): 489-497.

URL:

https://www.wutanyuhuatan.com/EN/10.11720/wtyht.2024.1492 OR https://www.wutanyuhuatan.com/EN/Y2024/V48/I2/489

Schematic diagram of the three-layer perception machine

岩性	统计量	CALI/ (ft·s^-1)	RD/ (Ω·m)	RHOB/ (g·cm^-3)	GR/ (API)	CN/%	PEF/ (mV)	DTC/ (m·s^-1)	SP/ (mV)	样本总数
砂岩	平均值	11.51	1.86	2.20	37.96	0.30	5.29	101.91	71.49	5837
	最小值	7.45	0.12	1.49	10.61	0.06	1.12	57.04	15.39
	最大值	23.70	87.51	2.84	94.13	0.76	34.19	158.18	132.99
泥质	平均值	9.62	1.35	2.32	58.98	0.28	4.71	94.34	92.82	3338
砂岩	最小值	7.45	0.31	1.56	25.40	0.08	1.16	62.60	24.44
	最大值	22.28	11.72	2.77	146.75	0.73	27.35	171.05	135.48
页岩	平均值	13.26	1.08	2.10	75.38	0.44	4.10	132.66	69.11	32491
	最小值	5.94	0.33	1.43	24.85	-0.05	1.32	7.41	28.90
	最大值	25.71	84.12	2.95	804.29	0.80	39.77	230.43	137.08
泥岩	平均值	10.20	3.16	2.46	34.51	0.19	4.10	80.03	99.51	1512
	最小值	7.32	0.70	1.62	8.05	0.07	1.32	60.18	22.89
	最大值	16.80	13.33	2.64	81.83	0.46	39.77	110.39	122.81
白云岩	平均值	9.53	5.06	2.45	42.59	0.15	4.18	71.61	102.21	98
	最小值	8.47	0.90	1.52	21.10	0.06	2.81	54.73	38.61
	最大值	15.07	13.13	2.90	82.36	0.40	7.67	117.03	121.94
石灰岩	平均值	11.52	3.65	2.45	22.24	0.18	5.56	76.26	85.46	6222
	最小值	7.48	0.39	1.52	5.59	0.001	1.97	40.76	17.69
	最大值	23.71	66.42	2.90	126.06	0.62	166.99	167.77	120.68
白垩岩	平均值	11.70	4.12	2.53	16.03	0.12	8.20	66.47	104.60	1924
	最小值	7.75	1.15	1.51	5.78	0.03	3.56	54.26	83.06
	最大值	14.22	14.04	2.63	33.26	0.33	86.82	81.83	116.41
盐岩	平均值	11.13	13.45	1.91	49.80	0.16	15.95	60.81	127.01	20
	最小值	8.67	2.61	1.60	15.79	0.008	11.99	51.28	121.07
	最大值	12.66	41.02	2.28	78.28	0.6	21.35	73.23	130.41
凝灰岩	平均值	14.01	0.81	2.19	45.77	0.41	4.12	118.26	62.12	719
	最小值	12.23	0.38	1.57	21.31	0.29	2.73	61.40	36.35
	最大值	23.31	1.41	2.36	78.12	0.65	9.67	149.06	97.41
煤炭	平均值	8.56	11.22	1.81	46.55	0.48	2.64	114.47	120.64	49
	最小值	8.18	2.32	1.42	26.14	0.26	1.29	77.18	113.24
	最大值	9.38	27.73	2.58	67.76	0.60	7.01	127.48	128.37

Logging data for different lithologies

Cross-validation diagram

Decision tree modeling analysis of lithology confusion matrix

Random forest modeling analysis of lithological confusion matrix

Decision tree and random forest model classification report

Confusion matrix of lithology analyzed by MLP modeling

Confusion matrix of lithology analyzed by the XGBoost model

MLP and XGBoost model classification report

F1 values for various lithologies

[1]	徐德龙, 李涛, 黄宝华, 等. 利用交会图法识别国外 M 油田岩性与流体类型的研究[J]. 地球物理学进展, 2012, 27(3):1123-1132.
[1]	Xu D L, Li T, Huang B H, et al. Research on the identification of the lithology and fluid type of foreign oilfield by using the crossplot method[J]. Progress in Geophysics, 2012, 27(3):1123-1132.
[2]	刘子云, 王向公. 利用概率统计方法判断岩性[J]. 江汉石油学院学报, 1989, 11(2):35-40.
[2]	Liu Z Y, Wang X G. Determination of lithology through probability statistics[J]. Journal of Jianghan Petroleum Institute, 1989, 11(2):35-40.
[3]	寻知锋, 余继峰. 聚类和判别分析在测井岩性识别中的应用[J]. 山东科技大学学报:自然科学版, 2008, 27(5):10-13.
[3]	Xun Z F, Yu J F. The application of cluster and discriminant analyses in logging lithology recognition[J]. Journal of Shandong University of Science and Technology:Natural Science, 2008, 27(5):10-13.
[4]	关涛. 基于交会图和贝叶斯聚类分析法的岩性识别方法[J]. 科学技术与工程, 2013, 13(4):976-979.
[4]	Guan T. Method of lithologic identification based on crossplot and Bayesian cluster analysis algorithm[J]. Science Technology and Engineering, 2013, 13(4):976-979.
[5]	潘少伟, 王朝阳, 张允, 等. 基于长短期记忆神经网络补全测井曲线和混合优化XGBoost的岩性识别[J]. 中国石油大学学报:自然科学版, 2022, 46(3):62-71.
[5]	Pan S W, Wang Z Y, Zhang Y, et al. Lithology identification based on LSTM neural networks completing log and hybrid optimized XGBoost[J]. Journal of China University of Petroleum:Edition of Natural Science, 2022, 46(3):62-71.
[6]	李洪奇, 谭锋奇, 许长福, 等. 基于决策树方法的砾岩油藏岩性识别——以克拉玛依油田六中区克下组油藏为例[J]. 石油天然气学报, 2010, 32(3):73-79,408.
[6]	Li H Q, Tan F Q, Xu C F, et al. Lithological identification of conglomerate reservoirs base on decision tree method[J]. Journal of Oil and Gas Technology, 2010, 32(3):73-79,408.
[7]	赖强, 魏伯阳, 吴煜宇, 等. 基于随机森林的K-近邻算法划分火成岩岩性[J]. 特种油气藏, 2021, 28(6):62-69.
[7]	Lai Q, Wei B Y, Wu Y Y, et al. Classification of igneous rock lithology with K-nearest neighbor algorithm based on random forest (RF-KNN)[J]. Special Oil & Gas Reservoirs, 2021, 28(6):62-69.
[8]	Quinlan J R. Simplifying decision trees[J]. International Journal of Man-Machine Studies, 1987, 27(3):221-234.
[9]	杨霞霞, 苏锋, 黄戌霞. 基于改进随机森林算法的不平衡数据分类方法研究[J]. 网络安全技术与应用, 2020(10):70-71.
[9]	Yang X X, Su F, Huang X X. Research on imbalanced data classification method based on improved random forest algorithm[J]. Network Security Technology & Application, 2020(10):70-71.
[10]	杜昕, 范廷恩, 董建华, 等. 基于多层感知机网络的薄储层预测[J]. 石油地球物理勘探, 2020, 55(6):1178-1187,1159.
[10]	Du X, Fan T E, Dong J H, et al. Characterization of thin sand reservoirs based on a multi-layer perceptron deep neural network[J]. Oil Geophysical Prospecting, 2020, 55(6):1178-1187,1159.
[11]	Chen T Q, Guestrin C. XGBoost:A scalable tree boosting system[C]// Proceedings of the 22^nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016:785-794.
[12]	张家臣, 邓金根, 谭强, 等. 基于XGBoost的测井曲线重构方法[J]. 石油地球物理勘探, 2022, 57(3):697-705,496.
[12]	Zhang J C, Deng J G, Tan Q, et al. Reconstruction of well logs based on XGBoost[J]. Oil Geophysical Prospecting, 2022, 57(3):697-705,496.
[13]	段忠义, 肖昆, 杨亚新, 等. 基于集成学习的松辽盆地砂岩型铀矿地层岩性自动识别研究[J]. 原子能科学技术, 2023, 57(12):2443-2454.
[13]	Duan Z Y, Xiao K, Yang Y X, et al. Automatic lithology identification of sandstone-type uranium deposit in Songliao basin based on ensemble learning[J]. Atomic Energy Science and Technology, 2023, 57(12):2443-2454.
[14]	Bormann P, Aursand P, Dilib F, et al. 2020 FORCE machine learning contest. https://github.com/bolgebrygg/Force-2020-Machine-Learning-competition.
[15]	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.
[15]	Ma L F, Xiao H M, Tao J W, et al. Research and application of lithology classification based on deep learning[J]. Science Technology and Engineering, 2022, 22(7):2609-2617.
[16]	Haykin S. Neural networks and learning machines,3/E[M]. Upper Sanddle River: Pearson Education, 2009.
[17]	Maria N J R, Pankaja R. Performance analysis of text classification algorithms using con-fusion matrix[J]. International Journal of Engineering and Technical Research (IJETR), 2016, 6(4):75-78.
[18]	安鹏, 曹丹平. 基于深度学习的测井岩性识别方法研究与应用[J]. 地球物理学进展, 2018, 33(3):1029-1034.
[18]	An P, Cao D P. Research and application of logging lithology identification based on deep learning[J]. Progress in Geophysics, 2018, 33(3):1029-1034.
[19]	武中原, 张欣, 张春雷, 等. 基于LSTM循环神经网络的岩性识别方法[J]. 岩性油气藏, 2021, 33(3):120-128.
[19]	Wu Z Y, Zhang X, Zhang C L, et al. Lithology identification based on LSTM recurrent neural network[J]. Lithologic Reservoirs, 2021, 33(3):120-128.
[20]	马陇飞, 萧汉敏, 陶敬伟, 等. 基于梯度提升决策树算法的岩性智能分类方法[J]. 油气地质与采收率, 2022, 29(1):21-29.
[20]	Ma L F, Xiao H M, Tao J W, et al. Intelligent lithology classification method based on GBDT algorithm[J]. Petroleum Geology and Recovery Efficiency, 2022, 29(1):21-29.
[21]	段忠义, 肖昆, 杨亚新, 等. 松辽盆地砂岩型铀矿钻孔岩性的测井识别[J]. 地球物理学进展, 2023, 38(6):2490-2501.
[21]	Duan Z Y, Xiao K, Yang Y X, et al. Logging identification of borehole lithology of sandstone-type uranium deposit in Songliao Basin[J]. Progress in Geophysics, 2023, 38(6):2490-2501.