1 State Key Laboratory of Continental Dynamics/Department of Geology, Northwest University, Xi’an 710069, China
2 Shaanxi Yanchang Petroleum (Group) Co., Ltd.,Gas Field Company, Yan’an 716000, China
3 Yan’an Natural Gas Exploration and Development Engineering Technology Research Center, Yan’an 716000, China
4 Tracy Energy Technologies Co., Ltd., Hangzhou 310000, China
5 School of Earth Resources, China University of Geosciences, Wuhan 430000, China
Show less
文章历史+
Received
Published
Revised
2024-12-14
2026-01-18
2025-10-13
Online Date
2026-09-20
PDF (7989K)
摘要
岩性识别是油气勘探开发中的重要任务,对储层评价和开发方案优化具有关键作用。然而,传统的测井解释方法在面对复杂储层条件(如低渗透、非均质性强的储层)时,往往难以精准提取关键特征并进行高效分类。针对上述问题,提出了一种结合卷积神经网络(convolutional neural networks,CNN)、长短时记忆网络(long short-term memory,LSTM)和注意力机制(attention)的深度学习模型,通过整合局部特征提取、时序依赖建模以及特征加权处理,实现对复杂储层岩性的精确分类和储层参数的高效预测。以气田2023—2024年已钻井的测井数据为基础,优选了伽马射线(gamma ray,GR)、声波时差(acoustic time difference,DT)和电阻率(resistivity,RT)作为输入特征,经过数据清洗、缺失值补全、小波去噪和标准化处理后形成高质量的数据集,以岩性分类结果作为模型输出,通过模型训练和验证,该方法在岩性分类任务中取得了94.89%的总体准确率,表明该模型不仅能够有效应对复杂储层的非线性特征,还能通过动态储层分析,为储层岩性识别和气藏开发提供科学依据。
Abstract
Rock identification plays a critical role in oil and gas exploration and development, particularly in reservoir evaluation and the optimization of development plans. However, traditional well logging interpretation methods often face challenges in accurately extracting key features and effectively classifying complex reservoir conditions, such as low-permeability and highly heterogeneous reservoirs. To address these challenges, a deep learning model was proposed that integrates CNN(convolutional neural networks), LSTM(long short-term memory) networks, and attention mechanisms. This model combines local feature extraction, temporal dependency modeling, and feature weighting to achieve accurate classification of complex reservoir lithology and efficient prediction of reservoir parameters. The dataset utilized in this study was based on well logging data from gas fields drilled between 2023 and 2024, with GR(gamma ray), DT(acoustic time difference), and RT(resistivity) selected as input features. After performing data cleaning, missing value imputation, wavelet denoising, and standardization, a high-quality dataset was generated. Lithology classification results were used as the output of the model. Through model training and validation, an overall accuracy of 94.89% is achieved for lithology classification tasks, demonstrating that the model is capable of handling the nonlinear characteristics of complex reservoirs. Furthermore, it provides a scientific foundation for reservoir lithology identification and gas reservoir development through dynamic reservoir analysis.
GAO Xiao-ping, ZHU Hong-bo, ZHAO Kai, HU Lei-lei, LIU Wei-kai, FANG Jun-ling, YANG Jing-hua.
Rock Identification Method Based on Convolutional Neural Network-long Short Term Memory Artificial Neural Network-self Attention Mechanism[J].
Science Technology and Engineering, 2026, 26(2): 544-552 DOI:10.12404/j.issn.1671-1815.2409281
YanYanfang, ShaoYanlin, WangQing, et al. Lithology identification of clastic rocks based on UAV images based on data fusion[J]. Science Technology and Engineering, 2024, 24(12): 4869-4875.
LiBozhi, YangMinghe, XuKai, et al. Mechanical drilling rate prediction method based on convolutional neural network with attention mechanism[J]. Science Technology and Engineering, 2024, 24(21): 8910-8916.
ZhaoYanlong, FangZhengkui, QiuZiyao, et al. Prediction of remaining service life of sucker rod under corrosion conditions based on long short-term memory network[J]. Science Technology and Engineering, 2022, 21(36): 15429-15433.
LeiMingfeng, ZhangYunbo, WangWeidong, et al. Research on intelligent identification method and application of rock lithology Mask R-CNN[J]. Journal of Railway Science and Engineering, 2022, 19(11): 3372-3382.
ZhangXiaofeng, PangChunyang, HuRui, et al. Research and application of complex lithology identification method based on CNN-GRU[J]. Well Logging Technology, 2023, 47(6): 662-670.
ZhangFengbo, MaXueling, DongZhenzhen, et al. Application of machine learning model based on CNN and LSTM in well logging lithology identification[J]. Journal of Xi’an Shiyou University(Natural Science Edition), 2024, 39(5): 96-103, 133.
WangTingting, WangZhenhao, LiFang, et al. Optuna-BiGRU well logging lithology identification based on enhanced multi-head attention mechanism[J]. Journal of Earth Sciences and Environment, 2024, 46(1): 127-142.
HeJing, LiWeiping, ZhangDaiyao, et al. Correlation analysis between ground resistance and soil volume water content based on Spearman correlation coefficient[J]. Journal of Mountain Meteo-rology, 2024, 48(3): 86-90.
HuangShouzhun. Application of radioactivity measurement in exploration of carbon aceoussiliceous mudstone-type uranium deposits in Ningguo-Jixi Area[J]. Western Resources, 2023(2): 140-142.
HanHonglai, LiuZhidi, ZhangZhiqiang, et al. Prediction of oil and gas reservoir productivity based on logging curve envelope area[J]. Science Technology and Engineering, 2024, 24(15): 6254-6262.
ChenGanghua, LiangShasha, WangJun, et al. Application of convolutional neural network in lithology identification[J]. Well Logging Technology, 2019, 43(2): 129-134.
WuHongwu, GaiShaoyan, DaFeipeng. Image super-resolution algorithm based on receptive field optimization and progressive feature fusion[J]. Journal of Xi’an Jiaotong University, 2025(1): 1-11.
LiRong. Significant region prediction method using convolutional neural network[J]. Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition), 2019, 31(1): 37-43.
ChenGanghua, ZhangYuxia, WangJun, et al. Application of bidirectional long short-term memory neural network in lithology identification of beach-bar sand reservoirs[J]. Well Logging Technology, 2023, 47(3): 319-325.
MaMinggang, PanYueliang, PengZebao, et al. Research on lithology classification based on adaptive meta-learning fusion attention[J]. Hydropower, 2022, 48(11): 50-54, 101.
LuCaiyun, ZhuWen, ZhangHaitian, et al. Research and application of optimization strategy for calculation of power prediction accuracy of renewable energy power generation[J]. Electrical Applications, 2024, 43(6): 54-59.
ZhouShengkai, FuLizhen, SongWen’ai. Short text semantic similarity calculation model based on deep learning[J]. Journal of Guangxi Normal University(Natural Science Edition), 2022, 40(3): 49-56.
ZhaoJing, KongTaoru. Research on computer big data analysis based on decision tree classification algorithm[J]. Automation and Instrumentation, 2024(10): 154-158.
ChengXiang, CaiJun. Research on rolling bearing fault diagnosis based on LightGBM[J]. Journal of Lanzhou University of Arts and Sciences(Natural Science Edition), 2024, 38(1): 59-65.
SongWei, SunXiongwei, WangHongna, et al. Deep coalbed methane geology-engineering integrated intelligent decision-making technology based on big data analysis algorithm[J]. Acta Petrolei Sinica, 2023, 44(11): 1949-1958.