[1]李金湘,廖家乐,宿文姬.基于SMOTETomek和机器学习的粤东山区边坡崩滑地质灾害危险性评价研究[J].华南地震,2025,(02):50-56.[doi:10.13512/j.hndz.2025.02.07]
　LI Jinxiang,LIAO Jiale,SU Wenji.Risk Assessment of Slope Collapse Hazards in Mountainous Areas of Eastern Guangdong Based on SMOTETomek and Machine Learning[J].,2025,(02):50-56.[doi:10.13512/j.hndz.2025.02.07]
点击复制

基于SMOTETomek和机器学习的粤东山区边坡崩滑地质灾害危险性评价研究()

华南地震[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2025年02期
页码:
50-56
栏目:
地震与地质灾害
出版日期:
2025-06-30

文章信息/Info

Title:
Risk Assessment of Slope Collapse Hazards in Mountainous Areas of Eastern Guangdong Based on SMOTETomek and Machine Learning
文章编号:
1001-8662（2025）02-0050-07
作者:
李金湘1; 2; 廖家乐2; 宿文姬2
1.广东省地质环境监测总站，广州 510510;2.华南理工大学，广州 510641
Author(s):
LI Jinxiang1; 2; LIAO Jiale2; SU Wenji2
1.Guangdong Geological Environment Monitoring Station ， Guangzhou 510510， China;2.South China Universi?ty of Technology ， Guangzhou 510641， China
关键词:
地质灾害; 危险性评价; SMOTETomek; 机器学习; 深度学习
Keywords:
Geological disaster; Risk assessment; SMOTETomek; Machine learning; Deep learning
分类号:
U416.14
DOI:
10.13512/j.hndz.2025.02.07
文献标志码:
A
摘要:
为开展粤东山区边坡崩滑地质灾害危险性评价研究，以梅西镇作为代表性研究区，选取坡度、坡向、平面曲率、剖面曲率、地层岩性、断裂距、水系距、NDVI、建筑物距、道路距、土地利用类型等构建评价指标体系，采用SMOTETomek混合采样结合随机森林（RF）、梯度提升决策树（GBDT）、3D卷积神经网络（3DCNN）、图注意力网络（GAT）等机器学习、深度学习模型开展易发性评价，并在最优模型的基础上叠加降雨量分析，完成危险性评价。结果得出：基于SMOTETomek混合采样的模型精度普遍更高，其中SMOTETomek-GAT模型精度最高，AUC值为0.93。
Abstract:
To assess the geological disaster hazard of slope collapse in mountainous areas of eastern Guangdong， this study selected Meixi Town as the representative research area. An evaluation index system was constructed com?prising the following factors：slope，aspect，plan curvature，profile curvature，stratigraphic lithology，distance from fault，distance from water system，normalized difference vegetation index（NDVI），distance from building， distance from road，and land use types. Then，a susceptibility evaluation was conducted by SMOTETomek hybrid sampling integrated with machine learning and deep learning models， including random forest（RF）， gradient boosting decision tree（GBDT），3D convolutional neural network（3D CNN），and graph attention network（GAT）.Based on the optimal model， precipitation analysis was incorporated to complete the risk assessment. The results show that models based on SMOTETomek hybrid sampling generally achieve higher accuracy， and the SMOTE?Tomek-GAT model has the highest precision，with an AUC value of 0.93.

参考文献/References:

[1]罗元华，张梁，张业成.地质灾害风险评估方法[M].北京：地质出版社，1998：32-58.
[2]江见鲸，徐志胜.防灾减灾工程学[M].北京：机械工业出版社，2005：259-265.
[3]赵富萌.中巴公路（中国段）地质灾害早期识别和滑坡易发性评价研究[D].兰州：兰州大学，2021.
[4]朱文慧，邹浩，何明明，等.基于BP神经网络的地质灾害易发性分区方法研究——以蕲春县为例[J].资源环境与工程，2021，35（06）：840-844.
[5] Marjanovic M，Kovacevic M，Bajat B，et al. Landslide suscep?tibility assessment using SVM machine learning algorithm[J]. Engineering Geology，2011，123（03）：225-234.
[6] Spanos A. Near-collinearity in linear regression revisited：The numerical vs the statistical perspective[J]. Communications in Statistics-Theory and Methods，2019，48（22）：5492-5516.
[7] Asadi M，Mokhtari L G，Shirzadi A，et al. A comparison study on the quantitative statistical methods for spatial prediction of shallow landslides （case study：Yozidar-Degaga Route in Kurdistan Province，Iran）[J]. Environmental Earth Sciences， 2022，81（02）：51.

相似文献/References:

[1]马长玲,李 科.地震诱发滑坡地质灾害的地貌因子敏感性评价模型[J].华南地震,2020,40(02):92.[doi:10.13512/j.hndz.2020.02.014]
　MA Changling,LI Ke.Sensitivity Evaluation Model of Geomorphological Factors for Earthquake-Induced Landslide Geological Disasters[J].,2020,40(02):92.[doi:10.13512/j.hndz.2020.02.014]
[2]尉德新,曾旻,吴荣,等.东非大裂谷山岭隧道地质灾害熵权可拓物元法评价及风险控制[J].华南地震,2021,41(03):18.[doi:10.13512/j.hndz.2021.03.03 ]
　WEI Dexin,ZENG Min,WU Rong,et al.Study on Assessment and Risk Control for Geological Disasters of Mountain Tunnel in East African Great Rift Valley Based on Entropy Weight Extension Matter Element Method[J].,2021,41(02):18.[doi:10.13512/j.hndz.2021.03.03 ]
[3]王 斌,周立鹏,李 琴.顾及InSAR形变特征的斜坡灾害隐患早期识别遥感方法[J].华南地震,2023,(01):105.[doi:10.13512/j.hndz.2023.01.13]
　WANG Bin,ZHOU Lipeng,LI Qin.Remote Sensing Method for Early Identification of Slope Hazards Considering InSAR Deformation Characteristics[J].,2023,(02):105.[doi:10.13512/j.hndz.2023.01.13]
[4]邱锦安,邓志德,刘燕辉,等.集镇地质灾害易发性分区研究——以梅州市大埔县银江镇为例[J].华南地震,2023,(04):80.[doi:10.13512/j.hndz.2023.04.11]
　QIU Jinan,DENG Zhide,LIU Yanhui,et al.Study on Susceptibility Zoning of Geological Hazards in Town Area—Taking Yinjiang Town，Dapu County， Meizhou City as an Example[J].,2023,(02):80.[doi:10.13512/j.hndz.2023.04.11]
[5]薛洪文,苏俊武,孙银锁,等.基于多源数据地质灾害监测预警应用研究[J].华南地震,2026,46(01):53.[doi:10.13512/j.hndz.2026.01.07]
　XUE Hongwen,SU Junwu,SUN Yinsuo,et al.Research on the Application of Geological Hazard Monitoring and Early Warning Based on Multi Source Data[J].,2026,46(02):53.[doi:10.13512/j.hndz.2026.01.07]
[6]杨兴明,宋骥衍,李 琛.尼泊尔“4·25”震后地质灾害发育特征及诱发因素分析——以日喀则萨嘎县为例[J].华南地震,2026,46(01):62.[doi:10.13512/j.hndz.2026.01.08]
　YANG Xingming,SONG Jiyan,LI Chen.Development Characteristics and Inducing Factors of Geological Disasters After the Nepal“4·25”Earthquake：A Case Study of Saga County，Shigatse[J].,2026,46(02):62.[doi:10.13512/j.hndz.2026.01.08]
[7]姚小月,宿文姬,李秀娟,等.基于多种组合模型的低山丘陵区地质灾害危险性评价及精度分析[J].华南地震,2023,(03):95.[doi:10.13512/j.hndz.2023.03.12]
　YAO Xiaoyue,SU Wenji,LI Xiujuan,et al.Risk Assessment of Geological Disasters in Low Mountain and Hilly Regions Based on Multiple Combined Models and Its Accuracy Analysis[J].,2023,(02):95.[doi:10.13512/j.hndz.2023.03.12]

备注/Memo

备注/Memo:
收稿日期：2024-10-12
基金项目：广东省自然资源厅科技项目（GDZRZYKJ2024008，GDZRYKJ2020002）
作者简介：李金湘（1980-），男，高级工程师，主要从事地质灾害防灾减灾研究。E-mail：2319058045@qq.com
通信作者：宿文姬（1969-），女，博士，副教授，主要从事地质灾害防灾减灾与智能监测预警预测研究。E-mail：wjsu@scut.edu.cn
更新日期/Last Update: 2025-06-30