With the deepening of oil exploration and development, wellbore conditions have become increasingly complex, and casing wear has become one of the major threats to well integrity. Wear leads to a reduction in the residual strength of the casing, directly affecting the safety and development efficiency of oil and gas wells. Existing methods for evaluating casing residual strength based on physical models or finite element numerical simulations struggle to balance accuracy and efficiency. To overcome this bottleneck, a data-driven approach, combined with artificial intelligence theory, an intelligent prediction model integrating BPNN-Optuna-XGBoost was proposed. This model first constructed a multi-size, multi-defect casing residual strength database through numerical simulation. It then introduced a back propagation neural network (BPNN) to achieve precise mapping of the residual strength distribution across the entire well section of a single well. Furthermore, by combining Optuna hyperparameter optimization and the XGBoost(extreme gradient boosting)algorithm, an integrated learning framework incorporating multi-well data was constructed to enable transfer prediction of residual strength for target wells in the same block. Experimental results show that the BPNN model has a prediction error of less than 5%, and the Optuna-XGBoost model achieves a goodness-of-fit of up to 0.996 4. The result provides a more efficient and accurate theoretical model and tool for evaluating casing residual strength under complex working conditions.
WAN Zhi-guo, LIU Jun-lin, WANG Jian, QIN Yan-bin, DOU Yi-hua.
Casing Residual Strength Intelligent Prediction Model Based on BPNN-Optuna-XGBoost[J].
Science Technology and Engineering, 2026, 26(15): 6400-6411 DOI:10.12404/j.issn.1671-1815.2503395
式(3)中:θ为模型的全部参数;l(yi,${\stackrel{\wedge }{\mathit{y}}}_{\mathit{i}}$)为损失函数;n为训练样本的总数;yi、${\stackrel{\wedge }{\mathit{y}}}_{\mathit{i}}$分别为第 i 个样本的真实标签与预测值;Ω(fk)为正则化项,控制模型复杂度,避免过拟合;fk第 k 棵决策树;T为模型中决策树的总数量。
XingXuesong, WuZhiqiang, ZhangXingkun, et al. Study on the remaining strength of composite crescent-shaped worn casings[J]. Petroleum Machinery, 2022, 50(11): 126-132.
ChiMing, LiuTao, XueChengwen, et al. Research on wear prediction and remaining strength check of high-temperature high-pressure deep well casings[J]. Drilling and Production Technology, 2021, 44(6): 1-6.
TanLeichuan, GaoDeli, AdeebSamer, et al. Casing wear prediction model based on drill string sine buckling[J]. Journal of China University of Petroleum (Natural Science Edition), 2019, 43(4): 69-74.
MaWenhai, WangXiaozeng, CaoYinping, et al. Experiment on the abrasive mechanism of drill pipe and casing in water-based mud in horizontal well of qingshen gas field[J]. Science Technology and Engineering, 2014, 14(34): 163-168.
XiaoRongge, DaiZheng, JinWenbo, et al. Comparative study on evaluation methods for remaining strength of oil and gas pipelines[J]. Hot Working Technology, 2021, 50(24): 61-64, 73.
CaoTianbao, LiZhonghui, LiuJian, et al. Prediction of casing wear and calculation of residual strength in large displacement wells[J]. Science Technology and Engineering, 2024, 24(24): 10227-10234.
[15]
KuriyamaY, TsukanoY, MimakiT, et al. Effect of wear and bending on casing collapse strength[C]// SPE Annual Technical Conference and Exhibition. Richardson: SPE, 1992: 24597.
DouYihua, ZhangFuxiang, WangWeijun, et al. Analysis of downhole casing wear depth and remaining strength[J]. Petroleum Drilling & Production Technology, 2007, 29(4): 36-39.
[18]
ZhangQ, LianZ, LinT. Prediction of residual burst strength of worn casing by theoretical and numerical modelling[J]. International Journal of Pressure Vessels and Piping, 2020, 188: 104195.
[19]
ShiJ, ZhangQ, LianZ, et al. A calculation method for ultimate and allowable loads in high-pressure thick-walled worn casing[J]. Ocean Engineering, 2024, 313: 119448.
LiuJian, LiZhonghui, HuangYi, et al. Residual strength variation of casing with outer wall wear in extended reach wells[J]. China Petroleum Machinery, 2025, 53(4): 132-138.
HuangWenjun, ZhangXingkun, GaoDeli. Casing failure risk prediction method under composite wear mode—a case study of φ193.68 mm×12.7 mm casing[J]. Natural Gas Industry, 2022, 42(7): 85-94.
[24]
MencagliaX, CazierO, FeutryV, et al. Towards a novel approach to consider casing wear in well designs[C]// Offshore Technology Conference. Houston: OTC, 2022: 31966.
[25]
ChenJ, EustesA. Simulation of burst resistance on casing with damage and deformation using finite element analysis[C]// SPE/IADC Drilling Conference and Exhibition. Houston: SPE, 2022: 208800.
[26]
MaoL, MaM, CaiM, et al. Nonlinear analysis on the impact wear characteristics of casing based on drillstring dynamics[J]. Geoenergy Science and Engineering, 2023, 229: 212090.
ZhouLijian, ZhuDonglin, DongShouye, et al. Interaction criteria and remaining strength of oil and gas pipelines with multiple corrosion defects[J]. Corrosion and Protection, 2023, 44(9): 22-31, 64.
[29]
AbrahamJ J, AmaniM. Field evaluation and assessment of wellbore, casing and cement integrity during critical high-pressure operations in existing and abandoned wells[C]// SPE Gas & Oil Technology Showcase and Conference. Dubai: SPE, 2023: 214035.
[30]
WangJ L, HuangW J, GaoD L. Prediction models of burst strength degradation for casing with considerations of both wear and corrosion[J]. Petroleum Science, 2024, 21(1): 458-474.
HouFuheng, ChenYanfei, HeGuoyan, et al. Study on ultimate bending moment capacity of submarine pipelines with combined corrosiondent defects[J]. Shipbuilding of China, 2023, 64(6): 24-34.
[35]
CaoH, WangT, ChenP, et al. Solar energy forecasting in short term based on the ASO-BPNN model[J]. Frontiers in Energy Research, 2022, 10: 902486.
[36]
HeB, ArmaghaniD J, TsoukalasM Z, et al. A case study of resilient modulus prediction leveraging an explainable metaheuristic-based XGBoost[J]. Transportation Geotechnics, 2024, 45: 101216.
[37]
SauerJ, MarianiV C, dos SantosC L, et al. Extreme gradient boosting model based on improved Jaya optimizer applied to forecasting energy consumption in residential buildings[J]. Evolving Systems, 2022, 13: 577-588.
YangFeng, XiaoFeng, ShiBowei, et al. Deterioration trend assessment of hydroelectric units based on Optuna-CatBoost and CRITIC evaluation method[J]. Hydropower Energy Science, 2025(2): 206-210.