IGWO-MSVR model for predicting stress in coal seam during drilling process
Due to the complex and changeable distribution of stress in rock or coal, the existing stress monitoring means can hardly accurately measure the stress concentration location and range, which brings out a strong risk in underground mining. Thus, we firstly attempted to use the Grey Model (GM) to ide...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-09-01
|
| Series: | Results in Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590123025026878 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849393489045880832 |
|---|---|
| author | Jian Tan Yanfeng Geng Liangke Xu |
| author_facet | Jian Tan Yanfeng Geng Liangke Xu |
| author_sort | Jian Tan |
| collection | DOAJ |
| description | Due to the complex and changeable distribution of stress in rock or coal, the existing stress monitoring means can hardly accurately measure the stress concentration location and range, which brings out a strong risk in underground mining. Thus, we firstly attempted to use the Grey Model (GM) to identify the main influence factors related to coal stress during borehole drilling process, through correlation analysis based on the numerical simulations. Then, utilized Grey Wolf Optimizer (GWO) model to optimize the relevant factors which appear in Modified Support Vector Regression model (MSVR). After then, finite element method was used to simulate drilling process under different conditions, and part of the obtained data by simulations was taken as training samples to train MSVR model, a part of the remaining samples was used to test the prediction accuracy of MSVR model. Furthermore, Back Propagation Neural network model (BP), Spatial Autoregressive model (SAR) and MSVR model were adopted to perform the stress prediction, and the stress prediction accuracy from these models was analyzed. Finally, MSVR was used for stress prediction by in-situ trials in three underground coal mines in China. The results showed: drilling torque was most significant, followed by coal elastic modulus and drilling depth. The MSVR model has the highest prediction, traditional SAR model has the second highest prediction accuracy, and BP model has the worst prediction accuracy, relatively. All models are strongly sensitive to the physical and mechanical properties of coal or rock. The higher the integrity and hardness of coal or rock, the higher the accuracy of prediction stress by borehole drilling. In-situ trials in three underground coal mines in China were performed, the stress prediction was satisfactorily verified by the amount of drilling cuttings. However, the applicability of stress prediction by borehole drilling in broken and weak coal or rock need to further research in the future. |
| format | Article |
| id | doaj-art-ae7579f852f34ac2a97f73be4971f78f |
| institution | Kabale University |
| issn | 2590-1230 |
| language | English |
| publishDate | 2025-09-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Results in Engineering |
| spelling | doaj-art-ae7579f852f34ac2a97f73be4971f78f2025-08-20T03:40:24ZengElsevierResults in Engineering2590-12302025-09-012710661810.1016/j.rineng.2025.106618IGWO-MSVR model for predicting stress in coal seam during drilling processJian Tan0Yanfeng Geng1Liangke Xu2College of Control Science and Engineering, China University of Petroleum, Qingdao, China; Corresponding author.College of Control Science and Engineering, China University of Petroleum, Qingdao, ChinaCollege of Energy and Mining Engineering, Shandong University of Science and Technology, Qingdao, ChinaDue to the complex and changeable distribution of stress in rock or coal, the existing stress monitoring means can hardly accurately measure the stress concentration location and range, which brings out a strong risk in underground mining. Thus, we firstly attempted to use the Grey Model (GM) to identify the main influence factors related to coal stress during borehole drilling process, through correlation analysis based on the numerical simulations. Then, utilized Grey Wolf Optimizer (GWO) model to optimize the relevant factors which appear in Modified Support Vector Regression model (MSVR). After then, finite element method was used to simulate drilling process under different conditions, and part of the obtained data by simulations was taken as training samples to train MSVR model, a part of the remaining samples was used to test the prediction accuracy of MSVR model. Furthermore, Back Propagation Neural network model (BP), Spatial Autoregressive model (SAR) and MSVR model were adopted to perform the stress prediction, and the stress prediction accuracy from these models was analyzed. Finally, MSVR was used for stress prediction by in-situ trials in three underground coal mines in China. The results showed: drilling torque was most significant, followed by coal elastic modulus and drilling depth. The MSVR model has the highest prediction, traditional SAR model has the second highest prediction accuracy, and BP model has the worst prediction accuracy, relatively. All models are strongly sensitive to the physical and mechanical properties of coal or rock. The higher the integrity and hardness of coal or rock, the higher the accuracy of prediction stress by borehole drilling. In-situ trials in three underground coal mines in China were performed, the stress prediction was satisfactorily verified by the amount of drilling cuttings. However, the applicability of stress prediction by borehole drilling in broken and weak coal or rock need to further research in the future.http://www.sciencedirect.com/science/article/pii/S2590123025026878Improved grey wolf optimization algorithmModified support vector regressionStress prediction by borehole drillingNumerical simulation |
| spellingShingle | Jian Tan Yanfeng Geng Liangke Xu IGWO-MSVR model for predicting stress in coal seam during drilling process Results in Engineering Improved grey wolf optimization algorithm Modified support vector regression Stress prediction by borehole drilling Numerical simulation |
| title | IGWO-MSVR model for predicting stress in coal seam during drilling process |
| title_full | IGWO-MSVR model for predicting stress in coal seam during drilling process |
| title_fullStr | IGWO-MSVR model for predicting stress in coal seam during drilling process |
| title_full_unstemmed | IGWO-MSVR model for predicting stress in coal seam during drilling process |
| title_short | IGWO-MSVR model for predicting stress in coal seam during drilling process |
| title_sort | igwo msvr model for predicting stress in coal seam during drilling process |
| topic | Improved grey wolf optimization algorithm Modified support vector regression Stress prediction by borehole drilling Numerical simulation |
| url | http://www.sciencedirect.com/science/article/pii/S2590123025026878 |
| work_keys_str_mv | AT jiantan igwomsvrmodelforpredictingstressincoalseamduringdrillingprocess AT yanfenggeng igwomsvrmodelforpredictingstressincoalseamduringdrillingprocess AT liangkexu igwomsvrmodelforpredictingstressincoalseamduringdrillingprocess |