摘要
土壤液化是一种常见于地震中的次生灾害,对建筑物和构筑物等生产生活都造成过严重破坏。国内外液化判别方法都是源于地震液化灾害现场数据,受到原位测试方法和场地区域性等条件的影响,存在一定的局限性。近年来随着机器学习等人工智能计算算法的兴起和发展,其在土壤液化研究方面显示出了特有的优势。以土壤液化判别相关研究成果及第18届世界地震工程大会部分报告为基础,本文阐述了研究土壤液化判别方法的机器学习算法,分析了相关机器学习算法的基本原理,列举了机器学习算法在土壤液化研究中的应用,展示了机器学习算法相对于传统方法具备灵活、稳定和泛化性强等诸多优异性能,提出了当前机器学习算法存在的缺陷,展望了机器学习算法在土壤液化研究领域未来发展趋势。
Soil liquefaction is a secondary disaster commonly seen in earthquakes,which has caused serious damage to buildings,structures,and other infrastructure.The methods for soil liquefaction assessment used both domestically and internationally are based on field data from earthquake-induced liquefaction disasters,and are subject to limitations due to in-situ testing methods and regional specificity.In recent years,with the rise and development of machine learning and other artificial intelligence computational algorithms,they have shown unique advantages in soil liquefaction research.Based on relevant research findings and the 18th World Conference on Earthquake Engineering report,this paper discusses the machine learning algorithms used for soil liquefaction assessment,analyzes the basic principles of related machine learning algorithms,showcases the applications of machine learning algorithms in soil liquefaction research.The rewiew demonstrates the superior performance of machine learning algorithms compared to traditional methods,such as flexibility,stability,and strong generalization ability,identifies the shortcomings of current machine learning algorithms,and outlines the future trends of machine learning algorithms in soil liquefaction research.
作者
耿铭屿
李兆焱
张升
袁晓铭
GENG Mingyu;LI Zhaoyan;ZHANG Sheng;YUAN Xiaoming(Key Laboratory of Earthquake Engineering and Engineering Vibration,Institute of Engineering Mechanics,China Earthquake Administration,Harbin 150080,China;Key Laboratory of Earthquake Disaster Mitigation,Ministry of Emergency Management,Harbin 150080,China)
出处
《世界地震工程》
北大核心
2025年第1期99-109,共11页
World Earthquake Engineering
基金
国家自然科学基金面上项目(52378543)
国家重点研发计划项目(2023YFC3805203)。
关键词
岩土工程
机器学习算法
土壤液化
液化判别
第18届世界地震工程大会
geotechnical engineering
machine learning algorithms
soil liquefaction
liquefaction discrimination
18th World Conference on Earthquake Engineering
