中国地质调查局西安地质调查中心主办
高级检索

几种聚类优化的机器学习方法在灵台县滑坡易发性评价中的应用

,  ,  , 

文章导航 > 西北地质 > 2020 > 53(1): 222-233

上一篇

下一篇

邱维蓉, 吴帮玉, 潘学树, 唐亚明. 2020. 几种聚类优化的机器学习方法在灵台县滑坡易发性评价中的应用. 西北地质, 53(1): 222-233. doi: 10.19751/j.cnki.61-1149/p.2020.01.021
引用本文: 邱维蓉, 吴帮玉, 潘学树, 唐亚明. 2020. 几种聚类优化的机器学习方法在灵台县滑坡易发性评价中的应用. 西北地质, 53(1): 222-233. doi: 10.19751/j.cnki.61-1149/p.2020.01.021
shu
QIU Weirong, WU Bangyu, PAN Xueshu, TANG Yaming. 2020. Application of Several Cluster-optimization-based Machine Learning Methods in Evaluation of Landslide Susceptibility in Lingtai County. Northwestern Geology, 53(1): 222-233. doi: 10.19751/j.cnki.61-1149/p.2020.01.021
Citation: QIU Weirong, WU Bangyu, PAN Xueshu, TANG Yaming. 2020. Application of Several Cluster-optimization-based Machine Learning Methods in Evaluation of Landslide Susceptibility in Lingtai County. Northwestern Geology, 53(1): 222-233. doi: 10.19751/j.cnki.61-1149/p.2020.01.021
shu

几种聚类优化的机器学习方法在灵台县滑坡易发性评价中的应用

  • 1. 西安交通大学 数学与统计学院, 陕西 西安 710049;

  • 2. 西安芯数启理信息科技有限公司, 陕西 西安 710065;

  • 3. 中国地质调查局西安地质调查中心/西北地质科技创新中心, 陕西 西安 710054

  • 基金项目:

    中国博士后科学基金（2016M600780）及中央高校基本科研业务费专项资金（xjj2018260）资助

详细信息
    作者简介: 邱维蓉(1994-),女,甘肃白银人,硕士,主要从事地球大数据挖掘研究。E-mail:18840840529@163.com

  • 中图分类号: P642.22

Application of Several Cluster-optimization-based Machine Learning Methods in Evaluation of Landslide Susceptibility in Lingtai County

  • 1. School of Mathematics and Statistics, Xi'an Jiaotong University, Xi'an 710049, Shaanxi, China;

  • 2. Xi'an Chip-Digit information Technology Co. Ltd., Xi'an 710065, Shaanxi, China;

  • 3. Xi'an Center of China Geological Survey/Northwest China Center for Geoscience Innovation, Xi'an 710054, Shaanxi, China

    摘要
  • 笔者以甘肃省平凉市灵台县为目标研究区域，基于地理空间和历史滑坡数据，利用混合高斯聚类（GMM）优化的逻辑回归（LR）、支持向量机（SVM）、BP神经网络（BP Neural Network）、随机森林（RF）4种机器学习模型构建滑坡易发性评价分析模型。选取高程、坡度、坡向、曲率、黄土侵蚀强度、归一化植被指数、地质构造7个环境因子作为滑坡易发性影响因子，以30 m栅格建立影响因子地理空间数据库，将研究区域划分为180万栅格单元。利用混合高斯聚类模型对整个研究区域的栅格单元进行聚类，得出初步的滑坡易发分区，选择易发程度最低类别中的栅格单元作为非滑坡区域，每次随机选择500个单元作为非滑坡单元，并根据历史滑坡数据将203个已知滑坡栅格单元作为滑坡单元，建立4种机器学习分类模型。利用训练好的模型对整个研究区域进行预测，绘制各算法的受试者工作曲线（ROC曲线），对各个算法的预测结果进行对比。分析结果表明，在本目标研究区域，各模型的滑坡易发区划图与实际的滑坡分布情况总体相吻合。随机森林模型的ROC曲线下面积（AUC）最大为0.96，测试集准确率最高为0.93；BP神经网络模型的ROC曲线下面积和测试集准确率次之，为0.90和0.87；支持向量机模型和逻辑回归模型的ROC曲线下面积和测试集准确率分别为0.86、0.81和0.85、0.80，均低于随机森林和BP神经网络模型。
    This paper takes Lingtai County, Pingliang City of Gansu Province as target research area. Based on the geospatial and historical landslide data, four machine learning models were used to construct the landslide susceptibility evaluation model. The four models are BP neural networks model, Random Forest classification model, support vector machine model, and logistic regression model which were optimized by GMM cluster model. In this paper, seven factors are selected as the landslide susceptibility influence factors, including elevation, slope, aspect, loess erosion intensity, vegetation coverage and geological structure. The influence factor of the geospatial database is established with 30m grid. The target area is divided into 1.8 million grid cells, and the grid cells of the whole area are clustered by the GMM model to obtain the preliminary subarea of landslide susceptibility map. 500 grid cells in the lowest-susceptibility category are selected as non-landslide units randomly, and 203 landslide grid units were used as landslide units according to historical landslide data. trained model is used to simulate and predict the whole research area, and to draw the ROC curve of each algorithm. Then compare the prediction results of each algorithm. The results of the analysis showed that the landslide susceptibility map of each algorithm is consistent with the actual landslide development. The random forest model has the largest area of 0.96 under the ROC curve, and the highest prediction accuracy of 0.93. It is followed by the BP-neural-network model with 0.89 under the ROC curve and 0.87 of the prediction accuracy. The area under the ROC curve and prediction accuracy of the support-vector-machine-model is 0.86, 0.81; and the logistic regression model is and 0.85, 0.80 respectively. The latter are lower than the first two models.
    • HTML全文
  • 加载中
    • 参考文献(55)

  • 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]. Catena,2018, 37(1):156-167.

    许冲,戴福初,姚鑫,等. 基于GIS的汶川地震滑坡灾害影响因子确定性系数分析[J]. 岩石力学与工程学报,2010,29(增刊1):2972-2981.

    XU Chong, DAI Fuchu, YAO Xin, et al. GIS-based certainty factor analysis of landslide triggering factors in wenchuan earthquake[J]. Chinese Journal of Rock Mechanics and Engineering, 2010,29(Supp.1):2972-2981.(in Chinese)

    王佳佳. 三峡库区万州区滑坡灾害风险评估研究[D].武汉:中国地质大学(武汉),2015.

    WANG Jiajia. Landslide risk assessment in Wanzhou County,Three Gorges Reservoir[Ph. D. Thesis] [D]. Wuhan:China University of Geosciences,2015.(in Chinese)

    刘艺梁, 殷坤龙, 刘斌. 逻辑回归和人工神经网络模型在滑坡灾害空间预测中的应用[J]. 水文地质工程地质, 2010,37(05):97-101.

    LIU Yiliang, YIN Kunlong, LIU Bing. Application of logistic regression and artificial neural networks in spatial assessment of landslide hazards[J]. Hydrogeology and Engineering Geology,2010,37(5):97-101.(in Chinese)

    吴益平, 张秋霞, 唐辉明, 等. 基于有效降雨强度的滑坡灾害危险性预警[J]. 地球科学-中国地质大学学报, 2014,39(7):889-895.

    WU Yiping,ZHANG Qiuxia,TANG Huiming,et al. Landslide Hazard Warning Based on Effective Rainfall Intensity[J]. Earth Science,2014,39(7):889-895. (in Chinese).

    石菊松, 徐瑞春, 石玲, 等. 基于RS和GIS技术的清江隔河岩库区滑坡易发性评价与制图[J]. 地学前缘, 2007(6):119-128.

    SHI Jusong, XU Ruichun, SHI Ling, et al. ETM+ imagery and GIS-based landslide susceptibility mapping for the regional area of Geheyan reservoir on the Qingjiang River,Hubei Province,China[J]. Earth Science Frontiers,2007,14(6):119-128.(in Chinese)

    朱莉, 卢毅敏, 罗建平. 基于灰色-Elman神经网络的区域滑坡易发性模型[J]. 自然灾害学报, 2013, 22(5):120-126.

    ZHU Li, LU Yimin, LUO Jianping. Regional landslide susceptibility model based on gray and Elman neural networks[J]. Journal of Natural Disasters, 2013,22(5):120-126.

    赵衡, 宋二祥. 滑坡灾害空间预测结果的评价方法[J]. 自然灾害学报,2010,19(05):36-41.

    ZHAO Heng,SONG Erxiang. Assessment method for results of spatial prediction of regional landslides[J]. Journal of Natural Disasters, 2010,19(05):36-41.

    赵艳南, 牛瑞卿. 基于证据权法的滑坡危险性区划探索[J]. 地理与地理信息科学, 2010, 26(6):19-23.

    ZHAO Yannan, NIU Ruiqin. Exploration of Landslide Hazard Zonation Based on the Weights of Evidence Method[J]. Geography and Geo-information Science, 2010, 26(6):19-23.

    张军, 刘祖强, 张正禄, 等. 基于神经网络和模糊评判的滑坡敏感性分析[J]. 测绘科学, 2012, 37(3):63-66.

    ZHANG Jun, LIU Zuqiang, ZHANG Zhenglu, et al.Susceptibility of landslide based on Artificial Neural Networks and fuzzy evaluating model[J]. Science of Surveying and Mapping,2012, 37(3):63-66.

    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.

    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 Ruiqin, 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.

    方匡南, 吴见彬, 朱建平, 等. 随机森林方法研究综述[J]. 统计与信息论坛, 2011, 26(3):32-38.

    FANG Kuangnan, WU Jianbin, ZHU Jianping, et al. A Review of Technologies onRandom Forests[J]. Statistics & Information Forum, 2011, 26(3):32-38.

    赵向辉, 付忠良, 谢会云, 等. 神经网络和集成学习在地质灾害危险度区划中的应用研究[J]. 工程科学与技术, 2010, 42(增刊1):50-55.

    ZHAO Xianghui, FU Zhongliang, XIE Huiyun, et al. Application Research of Ensemble Learning and Neural Network on Geological Hazard Fatalness Zoning[J]. Journal of Sichuan University(Engineering Science Edition),2010, 42(Supp.1):50-55.

    乔建平, 石莉莉. 滑坡危险度区划方法及其应用[J]. 地质通报, 2009, 28(8):1031-1038.

    QIAO Jianping, SHI Lili. Landslide fatalness zoning method and its application[J]. Geological Bulletin of China, 2009, 28(8):1031-1038.

    王涛, 吴树仁, 石菊松. 国际滑坡风险评估与管理指南研究综述[J]. 地质通报, 2009,28(8):1006-1019.

    WANG Tao, WU Shuren, SHI Jusong. A review of international landslide risk assessment and management guidelines[J].Geological Bulletin of China, 2009, 28(8):1006-1019.

    谭龙, 陈冠, 曾润强, 等. 人工神经网络在滑坡敏感性评价中的应用[J]. 兰州大学学报(自科版), 2014, 50(1):15-20.

    TAN Long, CHEN Guang, ZENG Runqiang, et al. Application of artificial neural network in landslide susceptibility assessment[J].Journal of Lanzhou University (Natural Sciences), 2014, 50(1):15-20.

    文海家, 李洋, 薛靖元, 等. 基于大数据挖掘的山区公路沿线滑坡易发性小区划[J].自然灾害学报,2018,27(04):161-167.

    WEN Haijia, LI Yang, XUE Jingyuan, et al. Landslides susceptibility microzoning along highway in mountainous region based on mining the big data[J].Journal of Natural Disasters, 2018, 27(04):161-167.

    郭子正, 殷坤龙, 黄发明, 等. 基于滑坡分类和加权频率比模型的滑坡易发性评价[J]. 岩石力学与工程学报, 2019, 38(02):76-89.

    GUO Zizheng, YIN Kunlong, HUANG Faming, et al. Evaluation of landslide susceptibility based on landslide classification and weighted frequency ratio model[J]. Chinese Journal of Rock Mechanics and Engineering, 2019,38(02):76-89.

    于晓辉, 林玲玲, 李静, 等. 基于GIS的中秦岭地区滑坡灾害易发性研究[J]. 防灾科技学院学报, 2010(04):106-113.

    YU Xiaohui, LIN Lingling, LI Jing, et al. A Study on the Susceptibility to Landslide Hazards in Middle Qinling Mountain by Using GIS Technology[J].Journal of Institute of Disaster-Prevention Science and Technology, 2010(4):106-113.

    李航.统计学习方法[M].北京:清华大学出版社,2012.

    周志华.机器学习[M].北京:清华大学出版社,2016.

    唐亚明, 冯卫, 毕银强, 等. 基于风险评价的黄土滑坡分类及特征[J]. 地质通报, 2015, 34(11):2092-2099.

    TANG Yaming, FENG Wei, BI Yinqiang, et al. The classification and features of loess landslide based on risk assessment[J].Geological Bulletin of China, 2015, 34(11):2092-2099.

    薛强, 张茂省, 李林. 基于斜坡单元与信息量法结合的宝塔区黄土滑坡易发性评价[J]. 地质通报, 2015, 34(11):2108-2115.

    XUE Qiang, ZHANG Maosheng, LI Lin. Loess landslide susceptibility evaluation based on slope unit and information value method in Baota District, Yan'an[J].Geological Bulletin of China, 2015, 34(11):2108-2115.

    张成航, 王佳运, 高波, 等. 陕西胥家村黄土滑坡风险定量分析[J].地质通报, 2015,34(11):2138-2142.

    ZHANG Chenghang, WANG Jiayun, GAO Bo, et al. Risk quantitative evaluation for Xujia Village loess landslide, Xi'an, Shaanxi Province[J], Geological Bulletin of China, 2015, 34(11):2138-2142.

    冯立, 张茂省, 张成航, 等. 四川虹口黑泥湾滑坡风险性评估[J]. 西北地质, 2014,47(3):165-176.

    FENG Li, ZHANG Maosheng, ZHANG Chenghang, et al. Risk Assessment of the Heiniwan Landslide in Hongkou County,Sichuan Province[J]. Northwestern Geology, 2014,47(3):165-176.

    缪卫东. 西安市白鹿塬滑坡发生时间预测研究[J]. 西北地质, 2003, 36(4):90-95.

    MIAO Weidong. Time prediction study on occurring of landslides in Bailuyuan, Xi'an[J]. Northwestern Geology, 2003, 36(4):90-95.

    PRADHAM B. A comparative study on the predictive ability of the decision tree,support vector machine and neuro-fuzzy models in landslide susceptibility mapping using GIS[J]. Computers & Geosciences, 2013, 51(2):350-365.

    AYALEW L, YAMAGISHI H. The application of GIS-based logistic regression for landslide susceptibility mapping in the Kakuda-Yahiko Mountains, Central Japan[J]. Geomorphology, 2005, 65(1-2):0-31.

    TSANGARATOS P, BENARDO A. Estimating landslide susceptibility through an artificial neural network classifier[J]. Natural Hazards, 2014, 74(3):1489-1516.

    • 相关文章
  • 施引文献
  • 资源附件(0)
  • 加载中
计量
  • 文章访问数:  2019
  • PDF下载数:  675
  • 施引文献:  0
出版历程
收稿日期:  2019-10-02
修回日期:  2019-11-05

目录

    版权所有：中国地质调查局发展研究中心 copyright @2020
    返回