Optimizing Schmidt Hammer Performance in Rock Testing: Integration of Kriging Surrogate Model and PSO-GWO Algorithm.
Saved in:
| Title: | Optimizing Schmidt Hammer Performance in Rock Testing: Integration of Kriging Surrogate Model and PSO-GWO Algorithm. |
|---|---|
| Authors: | Piao, Shenghao1 (AUTHOR) piaoshh@mail2.sysu.edu.cn, Huang, Sheng1 (AUTHOR) huangsh98@mail.sysu.edu.cn, Tan, Jianhui1 (AUTHOR), Wei, Yingjie1 (AUTHOR), Zheng, Chaowen1 (AUTHOR), Su, Xinhui1 (AUTHOR), Ma, Baosong1 (AUTHOR) mabaos@mail.sysu.edu.cn |
| Source: | Rock Mechanics & Rock Engineering. May2025, Vol. 58 Issue 5, p5207-5233. 27p. |
| Subjects: | Multi-objective optimization, Rock testing, Particle swarm optimization, Concrete testing, Kriging, Civil engineering |
| Abstract: | Schmidt hammer rebound method, a rapid technique for predicting uniaxial compressive strength (UCS), is extensively utilized in rock engineering. However, initially designed for concrete testing, this method encounters limitations when adapted for rock testing due to pronounced differences in density, mineral matrix strength, and anisotropy between concrete and rocks. Consequently, this paper introduces an innovative integrated optimization design approach that combines a Kriging surrogate model with a hybrid particle swarm optimization and gray wolf optimization (PSO-GWO) algorithm. The optimization objectives focus on minimizing damage to samples post-testing, enhancing sensitivity to different lithologies, and strengthening the correlation between rebound height (RH) and UCS. After identifying the relevant design parameters, the proposed method obtains responses and deviations within the range of design parameters, even with limited experimental data. Recognizing that rock is a damaged material, this study establishes a method to quantify the initial damage (Di) of rock samples. The sensitivity analysis elucidates the actual impact of Di on the objective functions and the interplay among design factors. Finally, accounting for the trade-offs between optimization objectives and practical applications, a set of Pareto optimal solutions and uncertainty predictions are generated for laboratory and in-situ testing conditions, respectively. This enables decision-makers to select solutions that align best with their specific research requirements and priorities. This research not only offers innovative perspectives on the application of the Schmidt hammer in rock mechanics but also presents a feasible solution for optimizing measurement equipment for damaged materials, such as rocks. Highlights: Proposed a new method integrating Kriging and PSO-GWO for multi-objective optimization design despite limited data. Applied an improved rebound testing method to significantly reduce manual operation errors. Developed a method to quantify the initial damage of rock using porosity and wave impedance. Introduced a rolling cross-validation technique to optimize training sets and minimize biases. PSO-GWO algorithm guides multi-objective optimization, enhancing real-world applicability through uncertainty analysis and preference intervals. [ABSTRACT FROM AUTHOR] |
| Copyright of Rock Mechanics & Rock Engineering is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Schmidt hammer rebound method, a rapid technique for predicting uniaxial compressive strength (UCS), is extensively utilized in rock engineering. However, initially designed for concrete testing, this method encounters limitations when adapted for rock testing due to pronounced differences in density, mineral matrix strength, and anisotropy between concrete and rocks. Consequently, this paper introduces an innovative integrated optimization design approach that combines a Kriging surrogate model with a hybrid particle swarm optimization and gray wolf optimization (PSO-GWO) algorithm. The optimization objectives focus on minimizing damage to samples post-testing, enhancing sensitivity to different lithologies, and strengthening the correlation between rebound height (RH) and UCS. After identifying the relevant design parameters, the proposed method obtains responses and deviations within the range of design parameters, even with limited experimental data. Recognizing that rock is a damaged material, this study establishes a method to quantify the initial damage (Di) of rock samples. The sensitivity analysis elucidates the actual impact of Di on the objective functions and the interplay among design factors. Finally, accounting for the trade-offs between optimization objectives and practical applications, a set of Pareto optimal solutions and uncertainty predictions are generated for laboratory and in-situ testing conditions, respectively. This enables decision-makers to select solutions that align best with their specific research requirements and priorities. This research not only offers innovative perspectives on the application of the Schmidt hammer in rock mechanics but also presents a feasible solution for optimizing measurement equipment for damaged materials, such as rocks. Highlights: Proposed a new method integrating Kriging and PSO-GWO for multi-objective optimization design despite limited data. Applied an improved rebound testing method to significantly reduce manual operation errors. Developed a method to quantify the initial damage of rock using porosity and wave impedance. Introduced a rolling cross-validation technique to optimize training sets and minimize biases. PSO-GWO algorithm guides multi-objective optimization, enhancing real-world applicability through uncertainty analysis and preference intervals. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 07232632 |
| DOI: | 10.1007/s00603-025-04425-8 |
