Prediction of Horizontal in Situ Stress in Shale Reservoirs Based on Machine Learning Models

To address the limitations of traditional methods in modeling complex nonlinear relationships in horizontal in situ stress prediction for shale reservoirs, this study proposes an integrated framework that combines well logging interpretation with machine learning to accurately predict horizontal in...

Full description

Saved in:
Bibliographic Details
Main Authors: Wenxuan Yu, Xizhe Li, Wei Guo, Hongming Zhan, Xuefeng Yang, Yongyang Liu, Xiangyang Pei, Weikang He, Longyi Wang, Yaoqiang Lin
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/12/6868
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849433070332018688
author Wenxuan Yu
Xizhe Li
Wei Guo
Hongming Zhan
Xuefeng Yang
Yongyang Liu
Xiangyang Pei
Weikang He
Longyi Wang
Yaoqiang Lin
author_facet Wenxuan Yu
Xizhe Li
Wei Guo
Hongming Zhan
Xuefeng Yang
Yongyang Liu
Xiangyang Pei
Weikang He
Longyi Wang
Yaoqiang Lin
author_sort Wenxuan Yu
collection DOAJ
description To address the limitations of traditional methods in modeling complex nonlinear relationships in horizontal in situ stress prediction for shale reservoirs, this study proposes an integrated framework that combines well logging interpretation with machine learning to accurately predict horizontal in situ stress in shale reservoirs. Based on the logging data from five wells in the Luzhou Block of the Sichuan Basin (16,000 samples), Recursive Feature Elimination (RF-RFE) was used to identify nine key factors, including Stoneley wave slowness and caliper, from 30 feature parameters. Bayesian optimization was employed to fine-tune the hyperparameters of the XGBoost model globally. Results indicate that the XGBoost model performs optimally in predicting maximum horizontal principal stress (SHmax) and minimum horizontal principal stress (SHmin). It achieves R<sup>2</sup> values of 0.978 and 0.959, respectively, on the test set. The error metrics (MAE, MSE, RMSE) of the XGBoost model are significantly lower than those of SVM and Random Forest, demonstrating its precise capture of the nonlinear relationships between logging parameters and in situ stress. This framework enhances the model’s adaptability to complex geological conditions through multi-well data training and eliminating redundant features, providing a reliable tool for hydraulic fracturing design and wellbore stability assessment in shale gas development.
format Article
id doaj-art-325a081645bd476aae48c91ec83d3f20
institution Kabale University
issn 2076-3417
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-325a081645bd476aae48c91ec83d3f202025-08-20T03:27:11ZengMDPI AGApplied Sciences2076-34172025-06-011512686810.3390/app15126868Prediction of Horizontal in Situ Stress in Shale Reservoirs Based on Machine Learning ModelsWenxuan Yu0Xizhe Li1Wei Guo2Hongming Zhan3Xuefeng Yang4Yongyang Liu5Xiangyang Pei6Weikang He7Longyi Wang8Yaoqiang Lin9Research Institute of Petroleum Exploration and Development, PetroChina, Beijing 100083, ChinaResearch Institute of Petroleum Exploration and Development, PetroChina, Beijing 100083, ChinaResearch Institute of Petroleum Exploration and Development, PetroChina, Beijing 100083, ChinaResearch Institute of Petroleum Exploration and Development, PetroChina, Beijing 100083, ChinaShale Gas Research Institute of Southwest Oil & Gas Field Branch, Chengdu 610051, ChinaShale Gas Research Institute of Southwest Oil & Gas Field Branch, Chengdu 610051, ChinaResearch Institute of Petroleum Exploration and Development, PetroChina, Beijing 100083, ChinaResearch Institute of Petroleum Exploration and Development, PetroChina, Beijing 100083, ChinaResearch Institute of Petroleum Exploration and Development, PetroChina, Beijing 100083, ChinaResearch Institute of Petroleum Exploration and Development, PetroChina, Beijing 100083, ChinaTo address the limitations of traditional methods in modeling complex nonlinear relationships in horizontal in situ stress prediction for shale reservoirs, this study proposes an integrated framework that combines well logging interpretation with machine learning to accurately predict horizontal in situ stress in shale reservoirs. Based on the logging data from five wells in the Luzhou Block of the Sichuan Basin (16,000 samples), Recursive Feature Elimination (RF-RFE) was used to identify nine key factors, including Stoneley wave slowness and caliper, from 30 feature parameters. Bayesian optimization was employed to fine-tune the hyperparameters of the XGBoost model globally. Results indicate that the XGBoost model performs optimally in predicting maximum horizontal principal stress (SHmax) and minimum horizontal principal stress (SHmin). It achieves R<sup>2</sup> values of 0.978 and 0.959, respectively, on the test set. The error metrics (MAE, MSE, RMSE) of the XGBoost model are significantly lower than those of SVM and Random Forest, demonstrating its precise capture of the nonlinear relationships between logging parameters and in situ stress. This framework enhances the model’s adaptability to complex geological conditions through multi-well data training and eliminating redundant features, providing a reliable tool for hydraulic fracturing design and wellbore stability assessment in shale gas development.https://www.mdpi.com/2076-3417/15/12/6868horizontal in situ stress predictionXGBoostwell logging interpretationmachine learningcomplex nonlinearityBayesian optimization
spellingShingle Wenxuan Yu
Xizhe Li
Wei Guo
Hongming Zhan
Xuefeng Yang
Yongyang Liu
Xiangyang Pei
Weikang He
Longyi Wang
Yaoqiang Lin
Prediction of Horizontal in Situ Stress in Shale Reservoirs Based on Machine Learning Models
Applied Sciences
horizontal in situ stress prediction
XGBoost
well logging interpretation
machine learning
complex nonlinearity
Bayesian optimization
title Prediction of Horizontal in Situ Stress in Shale Reservoirs Based on Machine Learning Models
title_full Prediction of Horizontal in Situ Stress in Shale Reservoirs Based on Machine Learning Models
title_fullStr Prediction of Horizontal in Situ Stress in Shale Reservoirs Based on Machine Learning Models
title_full_unstemmed Prediction of Horizontal in Situ Stress in Shale Reservoirs Based on Machine Learning Models
title_short Prediction of Horizontal in Situ Stress in Shale Reservoirs Based on Machine Learning Models
title_sort prediction of horizontal in situ stress in shale reservoirs based on machine learning models
topic horizontal in situ stress prediction
XGBoost
well logging interpretation
machine learning
complex nonlinearity
Bayesian optimization
url https://www.mdpi.com/2076-3417/15/12/6868
work_keys_str_mv AT wenxuanyu predictionofhorizontalinsitustressinshalereservoirsbasedonmachinelearningmodels
AT xizheli predictionofhorizontalinsitustressinshalereservoirsbasedonmachinelearningmodels
AT weiguo predictionofhorizontalinsitustressinshalereservoirsbasedonmachinelearningmodels
AT hongmingzhan predictionofhorizontalinsitustressinshalereservoirsbasedonmachinelearningmodels
AT xuefengyang predictionofhorizontalinsitustressinshalereservoirsbasedonmachinelearningmodels
AT yongyangliu predictionofhorizontalinsitustressinshalereservoirsbasedonmachinelearningmodels
AT xiangyangpei predictionofhorizontalinsitustressinshalereservoirsbasedonmachinelearningmodels
AT weikanghe predictionofhorizontalinsitustressinshalereservoirsbasedonmachinelearningmodels
AT longyiwang predictionofhorizontalinsitustressinshalereservoirsbasedonmachinelearningmodels
AT yaoqianglin predictionofhorizontalinsitustressinshalereservoirsbasedonmachinelearningmodels