Back analysis of geomechanical parameters based on a data augmentation algorithm and machine learning technique
文献类型：期刊论文
| 作者 | Li, Hui; Chen, Weizhong; Tan, Xianjun |
| 刊名 | UNDERGROUND SPACE
 |
| 出版日期 | 2025-04-01 |
| 卷号 | 21页码:215-231 |
| 关键词 | Back analysis Machine learning Data augmentation Geomechanical parameters |
| ISSN号 | 2096-2754 |
| DOI | 10.1016/j.undsp.2024.08.002 |
| 英文摘要 | Accurate geomechanical parameters are key factors for stabili optimization. The intelligent back analysis method based on the cost-effective technique for inverting parameters. To address the lo sets, this study proposes an innovative back analysis framework ta ysis approach that combines data augmentation with advanced o generative adversarial network (ACGAN)-based data augmentati ples that adhere to the underlying probability distribution of the o of small sample sizes. Subsequently, we harness the power of opt vector machine (SVM) to mine the intricate nonlinear relationshi the validity of the augmented data and the performance of the developed OPSO-SVM algorithms based on two different sample sizes are studied. Results show that the new datasets generated by ACGAN correlation coefficient exceeding 0.86. Furthermore, the superiority of the OPSO-SVM algorithm is also demonstrated by comparing the displacement prediction capability of various algorithms through four indices. It is also indicated that the relative error of the predicted displacement values reduces from almost 20% to 5% for the OPS ples. Finally, the inversed parameters and corresponding convergences predicted by the two OPSO-SVM models trained with different samples are discussed, indicating the feasibility of the combinatio chanical parameters. This endeavor not only facilitates the progr data, but also serves as a pivotal reference for both researchers a ty evaluation, disaster forecasting, structural design, and supporting monitored information is widely recognized as the most efficient and w accuracy of measured data, and the scarcity of comprehensive datailored for small sample sizes. We introduce a multi-faceted back analptimization and machine learning techniques. The auxiliary classifier on algorithm is first employed to generate synthetic yet realistic samriginal data, thereby expanding the dataset and mitigating the impact imized particle swarm optimization (OPSO) integrated with support ps between input and output variables. Then, relying on a case study, almost coincide with the actual monitored convergences, exhibiting a O-SVM model trained with 25 samples and that trained with 625 samn application of ACGAN and OPSO-SVM in back analysis of geomeession of underground engineering analysis in scenarios with limited nd practitioners alike. |
| 资助项目 | National Natural Science Foundation of China[51991392] ; National Natural Science Foundation of China[51922104] |
| WOS研究方向 | Engineering |
| 语种 | 英语 |
| WOS记录号 | WOS:001364453400001 |
| 出版者 | KEAI PUBLISHING LTD |
| 源URL | [http://119.78.100.198/handle/2S6PX9GI/43322]  |
| 专题 | 中科院武汉岩土力学所 |
| 通讯作者 | Chen, Weizhong |
| 作者单位 | Inst Rock & Soil Mech, Chinese Acad Sci, State Key Lab Geomech & Geotech Engn, Wuhan 430071, Peoples R China |
| 推荐引用方式 GB/T 7714 | Li, Hui,Chen, Weizhong,Tan, Xianjun. Back analysis of geomechanical parameters based on a data augmentation algorithm and machine learning technique[J]. UNDERGROUND SPACE,2025,21:215-231. |
| APA | Li, Hui,Chen, Weizhong,&Tan, Xianjun.(2025).Back analysis of geomechanical parameters based on a data augmentation algorithm and machine learning technique.UNDERGROUND SPACE,21,215-231. |
| MLA | Li, Hui,et al."Back analysis of geomechanical parameters based on a data augmentation algorithm and machine learning technique".UNDERGROUND SPACE 21(2025):215-231. |
入库方式： OAI收割
来源：武汉岩土力学研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。