Dual-Layer CNN-LSTM-Based Rate of Penetration Modeling for Geological Drilling Processes
-
摘要: 钻速是表征钻进效率的关键指标，其变化受多种钻进参数的共同影响，具有显著的强耦合性和非线性特征. 当前时刻的钻速状态，不仅由该时刻的钻进参数和地层条件决定，还与前若干时刻的钻速变化、钻进参数变化、钻进状态等密切相关，呈现出明显的时间依赖与序列记忆特性. 针对实际钻进数据中工况复杂、噪声较大以及传统模型难以有效反映长期时序依赖的问题，以襄阳某井场实际钻进过程为研究对象，提出一种基于双层CNN-LSTM的钻速建模方法. 首先，针对原始钻进数据多工况混杂的问题，构建正常钻进工况识别与井段自动划分方法，并结合异常数据清洗、分段滤波及尺度变换，形成一套数据预处理方案. 其次，采用时滞互信息相关性分析方法，对钻速与多种钻进参数之间的非线性及时序相关关系进行定量分析，据此确定模型输入变量及时间窗口长度. 在此基础上，通过多尺度卷积特征提取与时序建模相结合，构建一种双层CNN并联与LSTM串联的钻速模型. 实验结果表明，所提出的模型在实际钻进数据上的精度和稳定性均优于传统机器学习模型及单一结构深度学习模型，能够有效反映钻进过程中钻速的时序变化规律，为复杂钻进条件下的钻速建模提供一种可行的数据驱动方法.Abstract: Rate of penetration (ROP) is a key indicator for evaluating drilling efficiency. Its variation is jointly influenced by multiple drilling parameters and exhibits strong coupling and pronounced nonlinear characteristics. The ROP at the current time step is determined not only by the instantaneous drilling parameters and formation conditions, but also closely related to the historical evolution of ROP, drilling parameters, and drilling states over previous time steps, showing evident temporal dependence and sequence memory effects. To address the challenges posed by complex operating conditions, high noise levels in real drilling data, and the difficulty of traditional models in effectively capturing long-term temporal dependencies, this study investigates real drilling operations from a drilling site in Xiangyang and proposes a dual-layer CNN-LSTM-based ROP modeling method. First, to cope with the coexistence of multiple drilling conditions in raw data, a normal drilling condition identification and automatic well-section segmentation approach is developed. Combined with abnormal data cleaning, segment-wise filtering, and scale transformation, a systematic data preprocessing scheme is established. Second, a time-lagged mutual information analysis is employed to quantitatively analyze the nonlinear and temporal correlations between ROP and multiple drilling parameters, based on which the model input variables and the length of the time window are determined. On this basis, a ROP modeling framework is constructed by integrating multi-scale convolutional feature extraction withtemporal sequence modeling through parallel dual-layer CNNs cascaded with an LSTM network. Experimental results demonstrate that the proposed model achieves superior accuracy and stability on real drilling data compared with traditional machine learning models and deep learning models with single network structures. The results indicate that the proposed approach can effectively capture the temporal evolution characteristics of ROP during drilling operations, providing a feasible data-driven solution for ROP modeling under complex drilling conditions.
-
图 3 双层CNN-LSTM钻速模型网络结构
Fig. 3. Network architecture of the dual-layer CNN-LSTM ROP model
图 6 钻速时滞互信息相关性分析热力图
Fig. 6. Heatmap of time-lagged mutual information between ROP and drilling parameters
表 1 网络超参数配置
Table 1. Hyperparameter configuration of the network
超参数类型 数值 卷积通道数 64 LSTM隐层维数 128 训练轮次 300 学习率初值 0.001 批大小 32 
下载: 导出CSV
表 2 5种建模方法评价指标对比
Table 2. Comparison of evaluation metrics for five ROP modeling methods
建模方法 MSE MAE R2 BP 0.137 22 0.253 17 0.562 54 SVR 3.513 0 1.492 4 -10.2 LSTM 0.032 27 0.113 69 0.897 12 单层CNN-LSTM 0.077 719 0.197 990 0.752 230 本文方法 0.011 548 0.074 203 0.963 190
下载: 导出CSV
-
Ahmed, O. S., Adeniran, A. A., Samsuri, A., 2019. Computational Intelligence Based Prediction of Drilling Rate of Penetration: a Comparative Study. Journal of Petroleum Science and Engineering, 172: 1-12. https://doi.org/10.1016/j.petrol.2018.09.027 Bai, K., Sheng, M., Zhang, H., et al., 2025. Real-Time Drilling Torque Prediction Ahead of the Bit with Just-in-Time Learning. Petroleum Science, 22: 430-441. https://doi.org/10.1016/j.petsci.2024.12.014 Barbosa, L. F. F. M., Nascimento, A., Mathias, M. H., et al., 2019. Machine Learning Methods Applied to Drilling Rate of Penetration Prediction and Optimization: a Review. Journal of Petroleum Science and Engineering, 183: 106332. https://doi.org/10.1016/j.petrol.2019.106332 Du, S., Ma, T. Y., Huang, C., 2026. Federated Dictionary Learning-Based Intelligent Monitoring for Geological Drilling Processes. Earth Science, https://doi.org/10.3799/dqkx.2026.002 (in Chinese with English abstract). Gan, C., Wang, X., Wang, L., et al., 2023. Multi-Source Information Fusion-Based Dynamic Model for Online Prediction of Rate of Penetration (ROP) in Drilling Process. Geoenergy Science and Engineering, 230: 212187. doi: 10.1016/j.geoen.2023.212187 Gu, Y. B., 2023. Real-Time Prediction and Control of Weight on Bit in Automatic Drilling System Based on Downhole Weight on Bit (Dissertation). Xi'an Shiyou University, Xi'an (in Chinese with English abstract). Guan, Z. C., Hu, H. G., Wang, B., et al., 2019. Experimental Study on Rock-Breaking Efficiency of PDC Bit Based on Mechanical Specific Energy and Sliding Frictional Coefficient. Journal of China University of Petroleum (Edition of Natural Science), 43(5): 92-100 (in Chinese with English abstract). Hegde, C., Daigle, H., Millwater, H., et al. ., 2017. Analysis of Rate of Penetration (ROP) Prediction in Drilling Using Physics-Based and Data-Driven Models. Journal of Petroleum Science and Engineering, 159: 295-306. https://doi.org/10.1016/j.petrol.2017.09.020 Ji, C., Ma, F. Y., Wang, J. H., et al. ., 2021. Real-Time Industrial Process Fault Diagnosis Based on Time Delayed Mutual Information Analysis. Processes, 9(6): 1027. https://doi.org/10.3390/pr9061027 Kor, K., Altun, G., 2020. Is Support Vector Regression Method Suitable for Predicting Rate of Penetration? Journal of Petroleum Science and Engineering, 194: 107542. https://doi.org/10.1016/j.petrol.2020.107542 Li, W., Li, Y. N., Chen, S. C., et al. ., 2013. Drilling Rate Model of Cone Bit in Bottom Hole and Field Application. Journal of China University of Petroleum (Edition of Natural Science), 37(3): 74-77 (in Chinese with English abstract). Liu, J. B., Wei, H. S., Zhao, J. F., et al., 2015. A New 3D ROP Equation Considering the Rotary Speed of Bit. Petroleum Drilling Techniques, 43(1): 52-57 (in Chinese with English abstract). Lu, C. D., Gan, C., Chen, L. F., et al., 2023. Development and Prospect of Intelligent Control of Geological Drilling Process. Coal Geology & Exploration, 51(9): 31-43 (in Chinese with English abstract). Najjarpour, M., Jalalifar, H., Norouzi-Apourvari, S., 2022. Half a Century Experience in Rate of Penetration Management: Application of Machine Learning Methods and Optimization Algorithms: a Review. Journal of Petroleum Science and Engineering, 208: 109575. https://doi.org/10.1016/j.petrol.2021.109575 Pan, H. G., Su, T., Huang, X. D., et al., 2021. LSTM-Based Soft Sensor Design for Oxygen Content of Flue Gas in Coal-Fired Power Plant. Transactions of the Institute of Measurement and Control, 43(1): 78-87. https://doi.org/10.1177/0142331220932390 Yuan, Q. R., He, M., Chen, Z. C., et al. ., 2025. A Real-Time Prediction Method for Rate of Penetration Sequence in Offshore Deep Wells Drilling Based on Attention Mechanism-Enhanced BiLSTM Model. Ocean Engineering, 325: 120820. https://doi.org/10.1016/j.oceaneng.2025.120820 Zhu, Z. P., Song, X. Z., Zhang, R., et al. ., 2022. A Hybrid Neural Network Model for Predicting Bottomhole Pressure in Managed Pressure Drilling. Applied Sciences, 12(13): 6728. https://doi.org/10.3390/app12136728 杜胜, 马天宇, 黄澄, 等, 2026. 基于联邦字典学习的地质钻进过程智能监测. 地球科学, https://doi.org/10.3799/qkx.2026.002 谷亚冰, 2023. 基于井底钻压的自动送钻系统钻压实时预测与控制(博士学位论文). 西安: 西安石油大学. 管志川, 呼怀刚, 王斌, 等, 2019. 基于机械比能与滑动摩擦系数的PDC钻头破岩效率试验. 中国石油大学学报(自然科学版), 43(5): 92-100. 李玮, 李亚楠, 陈世春, 等, 2013. 井底牙轮钻头的钻速方程及现场应用. 中国石油大学学报(自然科学版), 37(3): 74-77. 刘军波, 韦红术, 赵景芳, 等, 2015. 考虑钻头转速影响的新三维钻速方程. 石油钻探技术, 43(1): 52-57. 陆承达, 甘超, 陈略峰, 等, 2023. 地质钻进过程智能控制研究进展与发展前景. 煤田地质与勘探, 51(9): 31-43. -
点击查看大图
图(9) / 表(2)
计量
- 文章访问数: 188
- HTML全文浏览量: 24
- PDF下载量: 29
- 被引次数: 0
