-
摘要:
在使用机器学习模型对滑坡进行易发性评价时，通常会在滑坡影响范围之外随机选取非滑坡样本点，具有一定的误差。为了提高滑坡易发性评价的精度，将自组织映射（self-organizing map，SOM）神经网络、信息量模型（information，I）以及支持向量机模型（support vector machine，SVM）进行耦合，提出一种基于SOM-I-SVM模型的滑坡易发性评价方法，并将SOM神经网络与K均值聚类算法进行对比，验证模型的可靠性。以十堰市茅箭区为例，首先通过对环境因子的相关性及重要性分析，筛选出距水系距离、坡度、降雨量、距构造距离、相对高差、距道路距离、地层岩性等7个因子，建立滑坡易发性评价指标体系，在此基础上计算出各因子的分级信息量值，并作为模型的输入变量进行滑坡易发性评价。分别采用SOM神经网络和K均值聚类算法选取非滑坡样本，然后将样本数据集代入I-SVM模型预测滑坡易发性。将SVM、I-SVM、KMeans-I-SVM、SOM-I-SVM等4种模型预测精度进行对比，其ROC曲线下面积（AUC）分别为0.82，0.88，0.90，0.91，说明SOM-I-SVM模型能有效提高滑坡易发性预测准确率。
Abstract:When using machine learning models for landslide susceptibility evaluation, the non-landslide sample points are usually selected randomly outside the landslide influence area, leading to a certain error. To improve the accuracy of landslide susceptibility evaluation, this paper couples the self-organizing map (SOM) neural network, information (I) model, and support vector machine (SVM) model, and proposes a SOM-I-SVM model-based method of landslide susceptibility evaluation, comparing with K-means clustering to verify the reliability of this model. The Maojian District of the city of Shiyan is taken as an example, and seven factors of the distance from water system, slope, rainfall, distance from structure, relative height difference, distance from road, stratigraphic lithology are selected by correlation and importance analyses of environmental factors to establish a landslide susceptibility evaluation system. Based on these, the graded information values of each factor are calculated and used as input variables for landslide susceptibility evaluation. The SOM neural network and K-means clustering are used to select non-landslide samples, and the sample data set is substituted into the I-SVM model to predict landslide susceptibility. The prediction accuracies of the four models, SVM, I-SVM, KMeans-I-SVM and SOM-I-SVM, are compared, and the area under the ROC curve (AUC values) are 0.82, 0.88, 0.90 and 0.91, indicating that the SOM-I-SVM model can effectively improve the accuracy of landslide susceptibility prediction.
-
Key words:
- landslide; susceptibility assessment /
- information model /
- SVM /
- SOM
-
-
表 1 各评价因子分级信息量值
Table 1. Information values of each evaluation factor
因子 分段 灾点数 灾点栅格数 信息量 排序 相对高差/m ≤10 23 289 0.085905143 15 11～30 25 593 0.234929441 8 31～50 28 792 0.095149584 13 >50 2 75 −1.368920571 28 坡度/（°） ≤10 27 398 0.090011487 14 11～20 11 228 0.19867783 9 21～30 24 733 0.440412967 7 31～40 12 322 −0.467576242 23 41～50 4 68 −0.995249727 27 >50 0 0.00001 −14.86423487 29 工程地质岩组 坚硬块状变辉绿岩岩组 4 98 −0.573082265 25 较坚硬中-厚层状变粒岩岩组 50 1113.75 −0.087934243 18 较坚硬-较软弱薄-厚层状变粒岩、石英片岩互层岩组 24 537.25 0.518090502 6 松散土体 0 0.00001 −15.17742068 30 距构造距离/m ≤200 14 340 0.648102911 4 201～600 9 273 −0.232119776 21 601～1000 15 255 −0.168793518 19 >1000 40 881 −0.048684404 17 距水系距离/m ≤200 42 738 0.783625576 3 201～400 12 225 −0.214301889 20 401～600 6 282 0.127347427 10 >600 18 504 −0.562859754 24 年降雨量/mm 785～855 5 109 −0.63353 26 856～891 27 525 0.10945 12 892～925 24 526 −0.04113 16 926～1010 22 589 0.11510 11 距道路距离/m ≤50 3 54 0.555417 5 51～100 9 279 1.05839 2 101～150 20 351 1.36602 1 >150 45 1065 −0.36334 22 
下载: 导出CSV
表 2 SVM、I-SVM模型参数表
Table 2. Parameters of the SVM and I-SVM models
模型类型 超参数 调试范围 步长 最佳参数 模型评分 SVM C 0.1~10 0.1 1.1 0.773 γ 0.1~10 0.1 0.12328 I-SVM C 0.1~10 0.1 0.8 0.848 γ 0.1~10 0.1 10.48113
下载: 导出CSV
表 3 SVM及I-SVM模型分区结果
Table 3. Results of the SVM and I-SVM models
模型类型 易发性等级 分区面积
/km2所占比例
/%滑坡数
/个占滑坡总数
比例/%SVM 低易发区 199.875 40.7 10 12.8 中易发区 129.476 26.4 16 20.5 高易发区 89.737 18.3 21 26.9 极高易发区 71.622 14.6 31 39.7 I-SVM 低易发区 221.594 45.2 11 14.1 中易发区 121.872 24.8 13 16.7 高易发区 83.549 17.0 15 19.2 极高易发区 63.695 13.0 39 50.0
下载: 导出CSV
表 4 SOM-I-SVM、KMeans-I-SVM模型参数表
Table 4. Parameters of the SOM-I-SVM and KMeans-I-SVM models
模型类型 超参数 调试范围 步长 最佳参数 模型评分 SOM-I-SVM C 0.1~10 0.1 0.7 0.879 γ 0.1~10 0.1 0.39442 KMeans-I-SVM C 0.1~10 0.1 6.0 0.856 γ 0.1~10 0.1 0.76396
下载: 导出CSV
表 5 KMeans-I-SVM、SOM-I-SVM模型分区结果
Table 5. Results of the KMeans-I-SVM and SOM-I-SVM models
模型类型 易发性等级 分区面积
/km2所占比例
/%滑坡数
/个占滑坡总数
比例/%KMeans-
I-SVM低易发区 235.54 48.0 14 18.0 中易发区 101.93 20.8 10 12.8 高易发区 88.31 18.0 17 21.8 极高易发区 64.93 13.2 37 47.4 SOM-I-SVM 低易发区 223.59 45.6 9 11.5 中易发区 121.10 24.7 10 12.8 高易发区 83.54 17.0 14 18.0 极高易发区 62.48 12.7 45 57.7
下载: 导出CSV
-
[1] GUZZETTI F,REICHENBACH P,CARDINALI M,et al. Probabilistic landslide hazard assessment at the basin scale[J]. Geomorphology,2005,72(1/2/3/4):272 − 299.
[2] CUI Kai,LU Dong,LI Wei. Comparison of landslide susceptibility mapping based on statistical index,certainty factors,weights of evidence and evidential belief function models[J]. Geocarto International,2017,32(9):935 − 955. doi: 10.1080/10106049.2016.1195886
[3] 盛逸凡,李远耀,徐勇,等. 基于有效降雨强度和逻辑回归的降雨型滑坡预测模型[J]. 水文地质工程地质,2019,46(1):156 − 162. [SHENG Yifan,LI Yuanyao,XU Yong,et al. Prediction of rainfall-type landslides based on effective rainfall intensity and logistic regression[J]. Hydrogeology & Engineering Geology,2019,46(1):156 − 162. (in Chinese with English abstract)
[4] SCHICKER R,MOON V. Comparison of bivariate and multivariate statistical approaches in landslide susceptibility mapping at a regional scale[J]. Geomorphology,2012,161/162:40 − 57. doi: 10.1016/j.geomorph.2012.03.036
[5] 金朝,费雯丽,丁卫,等. 基于信息量模型和Logistic回归模型的地质灾害易发性评价—以十堰市郧阳区为例[J]. 资源环境与工程,2021,35(6):845 − 850. [JIN Zhao,FEI Wenli,DING Wei,et al. Evaluation of geological disaster susceptibility based on information model and logistic regression model[J]. Resources Environment & Engineering,2021,35(6):845 − 850. (in Chinese with English abstract)
[6] 李利峰,张晓虎,邓慧琳,等. 基于熵指数与逻辑回归耦合模型的滑坡灾害易发性评价—以蓝田县为例[J]. 科学技术与工程,2020,20(14):5536 − 5543. [LI Lifeng,ZHANG Xiaohu,DENG Huilin,et al. Assessment of landslide susceptibility based on coupling model of index of entropy and logistic regression:A case study of Lantian County[J]. Science Technology and Engineering,2020,20(14):5536 − 5543. (in Chinese with English abstract)
[7] 覃乙根,杨根兰,江兴元,等. 基于确定性系数模型与逻辑回归模型耦合的地质灾害易发性评价—以贵州省开阳县为例[J]. 科学技术与工程,2020,20(1):96 − 103. [QIN Yigen,YANG Genlan,JIANG Xingyuan,et al. Geohazard susceptibility assessment based on integrated certainty factor model and logistic regression model for Kaiyang,China[J]. Science Technology and Engineering,2020,20(1):96 − 103. (in Chinese with English abstract)
[8] 吴雨辰,周晗旭,车爱兰. 基于粗糙集-神经网络的IBURI地震滑坡易发性研究[J]. 岩石力学与工程学报,2021,40(6):1226 − 1235. [WU Yuchen,ZHOU Hanxu,CHE Ailan. Susceptibility of landslides caused by IBURI earthquake based on rough set-neural network[J]. Chinese Journal of Rock Mechanics and Engineering,2021,40(6):1226 − 1235. (in Chinese with English abstract)
[9] 丁茜,赵晓东,吴鑫俊,等. 基于RBF核的多分类SVM滑塌易发性评价模型[J]. 中国安全科学学报,2022,32(3):194 − 200. [DING Xi,ZHAO Xiaodong,WU Xinjun,et al. Landslide susceptibility assessment model based on multi-class SVM with RBF kernel[J]. China Safety Science Journal,2022,32(3):194 − 200. (in Chinese with English abstract)
[10] 刘坚,李树林,陈涛. 基于优化随机森林模型的滑坡易发性评价[J]. 武汉大学学报·信息科学版,2018,43(7):1085 − 1091. [LIU Jian,LI Shulin,CHEN Tao. Landslide susceptibility assesment based on optimized random forest model[J]. Geomatics and Information Science of Wuhan University,2018,43(7):1085 − 1091. (in Chinese with English abstract)
[11] YESILNACAR E,TOPAL T. Landslide susceptibility mapping:A comparison of logistic regression and neural networks methods in a medium scale study,Hendek region (Turkey)[J]. Engineering Geology,2005,79(3/4):251 − 266.
[12] 万洋,郭捷,马凤山,等. 基于最大熵模型的中尼交通廊道滑坡易发性分析[J]. 中国地质灾害与防治学报,2022,33(2):88 − 95. [WAN Yang,GUO Jie,MA Fengshan,et al. Landslide susceptibility assessment based on MaxEnt model of along Sino-Nepal traffic corridor[J]. The Chinese Journal of Geological Hazard and Control,2022,33(2):88 − 95. (in Chinese with English abstract)
[13] 黄武彪, 丁明涛, 王栋, 等. 基于层数自适应加权卷积神经网络的川藏铁路沿线滑坡易发性评价[J]. 地球科学, 2022, 47(6): 2015−2030.
HUANG Wubiao, DING Mingtao, WANG Dong, et al. Evaluation of landslide susceptibility based on layer adaptive weighted convolutional neural network model along Sichuan-Tibet railway[J]. Earth Science, 2022, 47(6): 2015−2030. （in Chinese with English abstract）
[14] 安凯强,牛瑞卿. 信息量支持下SVM模型滑坡灾害易发性评价[J]. 长江科学院院报,2016,33(8):47 − 51. [AN Kaiqiang,NIU Ruiqing. Landslide susceptibility assessment using support vector machine based on weighted-information model[J]. Journal of Yangtze River Scientific Research Institute,2016,33(8):47 − 51. (in Chinese with English abstract) doi: 10.11988/ckyyb.20150311
[15] 杜国梁,张永双,吕文明,等. 基于加权信息量模型的藏东南地区滑坡易发性评价[J]. 灾害学,2016,31(2):226 − 234. [DU Guoliang,ZHANG Yongshuang,LV Wenming,et al. Landslide susceptibility assessment based on weighted information value model in southeast Tibet[J]. Journal of Catastrophology,2016,31(2):226 − 234. (in Chinese with English abstract)
[16] 陈飞, 蔡超, 李小双, 等. 基于信息量与神经网络模型的滑坡易发性评价[J]. 岩石力学与工程学报, 2020, 39（增刊1）: 2859 − 2870
CHEN Fei, CAI Chao, LI Xiaoshuang, et al. Evaluation of landslide susceptibility based on information volume and neural network model[J]. Chinese Journal of Rock Mechanics and Engineering, 2020, 39（Sup 1）: 2859 − 2870. （in Chinese with English abstract）
[17] 武雪玲,任福,牛瑞卿,等. 斜坡单元支持下的滑坡易发性评价支持向量机模型[J]. 武汉大学学报·信息科学版,2013,38(12):1499 − 1503. [WU Xueling,REN Fu,NIU Ruiqing,et al. Landslide spatial prediction based on slope units and support vector machines[J]. Geomatics and Information Science of Wuhan University,2013,38(12):1499 − 1503. (in Chinese with English abstract)
[18] BALLABIO C,STERLACCHINI S. Support vector machines for landslide susceptibility mapping:The staffora river basin case study,Italy[J]. Mathematical Geosciences,2012,44(1):47 − 70. doi: 10.1007/s11004-011-9379-9
[19] 徐胜华,刘纪平,王想红,等. 熵指数融入支持向量机的滑坡灾害易发性评价方法—以陕西省为例[J]. 武汉大学学报·信息科学版,2020,45(8):1214 − 1222. [XU Shenghua,LIU Jiping,WANG Xianghong,et al. Landslide susceptibility assessment method incorporating index of entropy based on support vector machine:A case study of Shaanxi Province[J]. Geomatics and Information Science of Wuhan University,2020,45(8):1214 − 1222. (in Chinese with English abstract)
[20] 刘福臻,王灵,肖东升. 机器学习模型在滑坡易发性评价中的应用[J]. 中国地质灾害与防治学报,2021,32(6):98 − 106. [LIU Fuzhen,WANG Ling,XIAO Dongsheng. Application of machine learning model in landslide susceptibility evaluation[J]. The Chinese Journal of Geological Hazard and Control,2021,32(6):98 − 106. (in Chinese with English abstract)
[21] 鲍帅,刘纪平,王亮. 联合DBSCAN聚类采样和SVM分类的滑坡易发性评价[J]. 震灾防御技术,2021,16(4):625 − 636. [BAO Shuai,LIU Jiping,WANG Liang. Landslide susceptibility evaluation based on combined DBSCAN cluster sampling and SVM classification[J]. Technology for Earthquake Disaster Prevention,2021,16(4):625 − 636. (in Chinese with English abstract) doi: 10.11899/j.issn.1673-5722.2021.4.zzfyjs202104003
[22] 杜国梁,杨志华,袁颖,等. 基于逻辑回归-信息量的川藏交通廊道滑坡易发性评价[J]. 水文地质工程地质,2021,48(5):102 − 111. [DU Guoliang,YANG Zhihua,YUAN Ying,et al. Landslide susceptibility mapping in the Sichuan-Tibet traffic corridor using logistic regression-information value method[J]. Hydrogeology & Engineering Geology,2021,48(5):102 − 111. (in Chinese with English abstract)
[23] LLOYD S. Least Squares quantization in PCM. IEEE Transactions on Information Theory[J]. The Chinese Journal of Geological Hazard and Control,1982,28(2):129 − 137.
[24] 黄发明,殷坤龙,蒋水华,等. 基于聚类分析和支持向量机的滑坡易发性评价[J]. 岩石力学与工程学报,2018,37(1):156 − 167. [HUANG Faming,YIN Kunlong,JIANG Shuihua,et al. Landslide susceptibility assessment based on clustering analysis and support vector machine[J]. Chinese Journal of Rock Mechanics and Engineering,2018,37(1):156 − 167. (in Chinese with English abstract)
[25] 周超,殷坤龙,曹颖,等. 基于集成学习与径向基神经网络耦合模型的三峡库区滑坡易发性评价[J]. 地球科学,2020,45(6):1865 − 1876. [ZHOU Chao,YIN Kunlong,CAO Ying,et al. Landslide susceptibility assessment by applying the coupling method of radial basis neural network and adaboost:A case study from the Three Gorges Reservoir area[J]. Earth Science,2020,45(6):1865 − 1876. (in Chinese with English abstract)
[26] WUBALEM A,METEN M. Landslide susceptibility mapping using information value and logistic regression models in Goncha Siso Eneses area,northwestern Ethiopia[J]. SN Applied Sciences,2020,2(5):1 − 19.
-
图(15)
表(5)
计量
- 文章访问数: 1281
- PDF下载数: 30
- 施引文献: 0