|本期目录/Table of Contents|

[1]邝泽民,曾 卫*,文 凯.基于自适应高斯过程响应面法的边坡可靠度分析方法[J].武汉工程大学学报,2025,47(03):349-354.[doi:10.19843/j.cnki.CN42-1779/TQ.202308013]
 KUANG Zemin,ZENG Wei*,WEN Kai.Slope reliability analysis method based on adaptive Gaussian process response surface method[J].Journal of Wuhan Institute of Technology,2025,47(03):349-354.[doi:10.19843/j.cnki.CN42-1779/TQ.202308013]
点击复制

基于自适应高斯过程响应面法的边坡可靠度分析方法
(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
47
期数:
2025年03期
页码:
349-354
栏目:
智能制造
出版日期:
2025-06-30

文章信息/Info

Title:
Slope reliability analysis method based on adaptive Gaussian process response surface method
文章编号:
1674 - 2869(2025)03 - 0349 - 06
作者:
佛山大学土木与交通学院,广东 佛山 528000
Author(s):
School of Civil Engineering and Transportation, Foshan University,Foshan 528000, China
关键词:
Keywords:
分类号:
TU457
DOI:
10.19843/j.cnki.CN42-1779/TQ.202308013
文献标志码:
A
摘要:
在基于代理模型的响应面法中,训练点的选择策略对准确评估边坡可靠度极为重要。为了进一步优化训练点的选择策略,提出了一种新的自适应搜索引导函数(ASGF),该函数根据代理模型的信息以及真实功能函数的评价次数生成搜索引导因子,并结合改进期望函数动态地调整算法的搜索过程,实现已知区域的开发和未知区域的探索。基于ASGF,提出一种新的可靠度分析方法,该方法利用高斯过程回归拟合边坡真实的功能函数,通过ASGF在极限状态面附近加入训练点以动态更新代理模型,最后结合蒙特卡罗模拟计算边坡的失效概率。数值算例和工程案例计算结果表明:该方法能够有效逼近极限状态面,并且保持新加训练点分布均匀。
Abstract:
The selection strategy of training points is extremely important for accurate assessment of slope reliability in the response surface method based on surrogate models. To further optimize this strategy, a new adaptive search guidance function (ASGF) was proposed, which generates a search guidance factor based on the information of the agent model and the number of evaluations of the true performance function, and dynamically adjusts the search process in combination with the improved expectation function. The search process of the algorithm ensures both the development of known areas and the exploration of unknown areas. Based on ASGF, a new reliability analysis method was proposed. This method uses Gaussian process regression (GPR) to fit the true performance function of the slope, adds training points near the limit state surface through ASGF to dynamically update the agent model, and finally combines Monte Carlo simulation to calculate the failure probability of slopes. Numerical examples and engineering cases were calculated respectively, and the results showed that this method can effectively approximate the limit state surface and maintain a uniform distribution of new training points.

参考文献/References:

[1] 李元松,王玉,朱冬林,等.边坡稳定性评价方法研究现状与发展趋势[J].武汉工程大学学报,2021,43(4):428-435.
[2] JIANG S H, HUANG J S ,GRIFFITHS D V, et al. Advances in reliability and risk analyses of slopes in spatially variable soils:a state-of-the-art review [J]. Computers and Geotechnics,2022,141:104498.
[3] CHAKRABORTY R, DEY A. Probabilistic slope stability analysis:state-of-the-art review and future prospects [J]. Innovative Infrastructure Solutions,2022,7(2):177.
[4] 蒋辽,喻兴,刘林洁,等.基于蒙特卡罗模拟的填方边坡可靠度分析[J].地下空间与工程学报,2017,13(增刊2):693-697.
[5] 朱彦鹏,刘鑫,马孝瑞,等.基于改进一次二阶矩法的可靠度理论在高填方边坡加固中的应用[J].科学技术与工程,2019,19(27):351-357.
[6] 朱彬,裴华富,杨庆.基于高斯过程回归的响应面法及边坡可靠度分析[J].岩土工程学报,2019,41(增刊1):209-212.
[7] 李静萍,程勇刚,李典庆,等. 基于多重响应面法的空间变异土坡系统可靠度分析[J]. 岩土力学,2016,37(1):147-155,165.
[8] 苏国韶,赵伟,彭立锋,等.边坡失效概率估计的高斯过程动态响应面法[J].岩土力学,2014,35(12):3592-3601.
[9] HU B,SU G S,JIANG J Q,et al. Gaussian process-based response surface method for slope reliability analysis [J]. Advances in Civil Engineering,2019,2019:9185756.
[10] 谭晓慧,宋传中,侯晓亮. 基于径向基函数响应面法的岩质边坡可靠度分析[J]. 地下空间与工程学报,2016,12(1):250-255.
[11] ZHU B,PEI H F,YANG Q. An intelligent response surface method for analyzing slope reliability based on Gaussian process regression [J]. International Journal for Numerical and Analytical Methods in Geomechanics,2019,43(15):2431-2448.
[12] SONG X D, LI M Y, LI Z T, et al. Global optimization algorithm based on Kriging using multi-point infill sampling criterion and its application in transportation system [J]. Sustainability,2021,13(19):10645.
[13] ZHAN D W,XING H L. Expected improvement for expensive optimization:a review [J]. Journal of Global Optimization,2020,78(3):507-544.
[14] KIM S H,NA S W. Response surface method using vector projected sampling points [J]. Structural Safety,1997,19(1):3-19.
[15] HOEK E. Practical rock engineering [M]. North Vancouver:Evert Hoek Consulting Engineer Inc.,2006:110-124.


相似文献/References:

备注/Memo

备注/Memo:
收稿日期:2023-08-20
基金项目:广东省基础与应用基础研究基金(2020A1515111137,2023A1515011772)
作者简介:邝泽民,硕士研究生。Email:zmkuang2017@163.com
*通信作者:曾 卫,博士,讲师。Email:zeng-wei@126.com
引文格式:邝泽民,曾卫,文凯. 基于自适应高斯过程响应面法的边坡可靠度分析方法[J]. 武汉工程大学学报,2025,47(3):349-354.
更新日期/Last Update: 2025-07-09