“阶跃式”滑坡突变预测与核心因子提取的平衡集成树模型
|
摘要
"阶跃式"滑坡在复杂多变的地质环境作用下呈现突变与稳定交替、不平衡的演化形态,根据此特点提出了边坡突变与稳定分类的平衡集成树模型,建立高维地质环境影响因子与致变之间的关联,并应用于三峡库区26个具有"阶跃"特征的滑坡。考虑到直接使用分类器难以捕捉有效信息,模型利用了合成边界少数类过采样原理适当提升突变样本比率,再分别组合随机森林和梯度提升树进行优化和训练,并对测试集和预测集进行评估校核。同数据平衡前后的不同模型进行对比实验,平衡集成树模型能够有效提高突变预测的整体精度,并量化得出了所有地质环境因子的特征重要性指标,最终应用于样本外的王爷庙滑坡结果表明,模型取得了较高的预报水平。该方法能够实现突变的有效预测,并发现核心影响因子及其数据分布特征,为研究"阶跃式"滑坡机理和预警提供了新的思路。
Step-like landslides show abrupt change,stable alternation and unbalanced evolution under the complex and changeable geological environment. Based on this feature,a balanced decision tree ensemble model of slope catastrophe and stable classification is proposed to establish a correlation between highdimensional geological environmental impact factors and mutations,and it is applied to 26 typical step-like landslides in the Three Gorges Reservoir Area. Considering that it is difficult to capture effective information by directly using the classifier,the model uses the synthetic borderline minority over-sampling principle to appropriately increase the mutation sample ratio,and then combines the random forest and the gradient boosted decision tree for optimization and training,and evaluates and verifies the test set and prediction set. Compared with different models before and after data balance,the balanced decision tree ensemble model can effectively improve the overall accuracy of the catastrophe prediction,also the feature importance index of all geological environmental factors is obtained. Finally,the results applied to the Wangyemiao Landslide show that the optimization model has achieved a higher level of forecasting. This method can achieve effective catastrophe prediction,and finds the core influence factors and their data distribution characteristics,which provides the ideas for studying the step-like landslide mechanism and early warning.
Step-like landslides show abrupt change,stable alternation and unbalanced evolution under the complex and changeable geological environment. Based on this feature,a balanced decision tree ensemble model of slope catastrophe and stable classification is proposed to establish a correlation between highdimensional geological environmental impact factors and mutations,and it is applied to 26 typical step-like landslides in the Three Gorges Reservoir Area. Considering that it is difficult to capture effective information by directly using the classifier,the model uses the synthetic borderline minority over-sampling principle to appropriately increase the mutation sample ratio,and then combines the random forest and the gradient boosted decision tree for optimization and training,and evaluates and verifies the test set and prediction set. Compared with different models before and after data balance,the balanced decision tree ensemble model can effectively improve the overall accuracy of the catastrophe prediction,also the feature importance index of all geological environmental factors is obtained. Finally,the results applied to the Wangyemiao Landslide show that the optimization model has achieved a higher level of forecasting. This method can achieve effective catastrophe prediction,and finds the core influence factors and their data distribution characteristics,which provides the ideas for studying the step-like landslide mechanism and early warning.
引文
[1]许强，汤明高，徐开祥，等．滑坡时空演化规律及预警预报研究[J]．岩石力学与工程学报，2008,27(6):1104-1112.XU Qiang,TANG Minggao,XU Kaixiang,et al.Research on space time evolution laws and early warning prediction of landslides[J]. Chinese Journal of Rock Mechanics and Engineering,2008,27(6):1104-1112.
[2]易庆林，张明玉，文凯，等．三峡库区白水河滑坡变形特征及影响因素的阶段分析[J]．三峡大学学报(自然科学版)，2017,39(1):38-42.YI Qinglin,ZHANG Mingyu,WEN Kai,et al. Periodic analysis of deformation characteristics and influential factors of Baishuihe Landslide in Three Gorges Reservoir area[J]. Journal of China Three Gorges University(Natural Sciences),2017,39(1):38-42.
[3]黄达，顾东明，陈智强，等．三峡库区塔坪H2古滑坡台阶状复活变形的库水-降雨耦合作用机制[J]．岩土工程学报，2017,39(12):2203-2211.HUANG Da,GU Dongming,CHEN Zhiqiang,et al.Hybrid effects of rainfall and reservoir level fluctuation on old Taping H2 landslide in Wushan County in Three Gorges Reservoir area[J]. Chinese Journal of Geotechnical Engineering,2017,39(12):2203-2211.
[4]张振华，钱明明，位伟．基于改进破坏接近度的千江坪岸坡失稳机制分析[J]．岩石力学与工程学报，2018,37(6):1371-1384.ZHANG Zhenhua, QIAN Mingming, WEI Wei.Analysis of failure mechanism of Qianjiangping slope based on improved failure approach index[J].Chinese Journal of Rock Mechanics and Engineering,2018,37(6):1371-1384.
[5]陈卫兵，郑颖人，冯夏庭，等．考虑岩土体流变特性的强度折减法研究[J]．岩土力学，2008,29(1):101-105.CHEN Weibing,ZHENG Yingren,FENG Xiating,et al. Study on strength reduction technique considering rheological property of rock and soil medium[J].Rock and Soil Mechanics,2008,29(1):101-105.
[6]赵久彬，刘元雪，宋林波，等．大数据关键技术在滑坡监测预警系统中的应用[J]．重庆理工大学学报(自然科学)，2018,32(2):182-190.ZHAO Jiubin,LIU Yuanxue,SONG Linbo,et al.Application of big data processing and monitoring technologies in landslide monitoring and warning system[J]. Journal of Chongqing University of Technology(Natural Science),2018,32(2):182-190.
[7]彭令，牛瑞卿，吴婷．时间序列分析于支持向量机的滑坡位移预测[J]．浙江大学学报(工学版)，2013,47(9):1672-1679.PENG Ling,NIU Ruiqing,WU Ting. Time series analysis and support vector machine for landslide displacement prediction[J]. Journal of Zhejiang University(Engineering Science),2013,47(9):1672-1679.
[8] MIAO F S,WU Y P,XIE Y H,et al. Prediction of landslide displacement with step like behavior based on multialgorithm optimization and a support vector regression model[J]. Landslides,2017,15(3):475-488.
[9]李骅锦，许强，何雨森，等．“阶跃式”滑坡位移预测及阈值分析的ARMA-(LASSO-ELM)-Copula模型[J]．岩石力学与工程学报．2017,36(增刊2):4075-4084.LI Huajin,XU Qiang,HE Yusen,et al. An ARMA-(LASSO-ELM)-Copula framework for landslide displacement prediction and threshold computing of the displacement of step-like landslides[J]. Chinese Journal of Rock Mechanics and Engineering,2017,36(S2):4075-4084.
[10]杨背背，殷坤龙，杜娟．基于时间序列与长短期记忆网络的滑坡位移动态预测模型[J]．岩石力学与工程学报，2018,37(10):2334-2343.YANG Beibei,YIN Kunlong,DU Juan. A model for predicting landslide displacement based on time series and long and short term memory neural network[J].Chinese Journal of Rock Mechanics and Engineering,2018,37(10):2334-2343.
[11] CHAWLA N V,BOWYER K W,HALL L O,et al.SMOTE:synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research,2002,16(1):321-357.
[12] HAN H,WANG W Y,MAO B H. BorderlineSMOTE:a new over-sampling method in imbalanced data sets learning[J]. Advances in Intelligent Computing,2005,3644(1):878-887.
[13]林威伟，钟登华，胡炜，等．基于随机森林算法的土石坝压实质量动态评价研究[J]．水利学报，2018,49(8):945-955.LIN Weiwei,ZHONG Denghua,HU Wei,et al. Study on dynamic evaluation of compaction quality of earth rock dam based on Random Forest[J]. Journal of Hydraulic Engineering,2018,49(8):945-955.
[14] ZHANG C S,LIU C C,ZHANG X L,et al. An up-todate comparison of state-of-the-art classification algorithms[J]. Expert Systems with Applications,2017,82:128-150.
[15] POWERS D M W. Evaluation:from Precision,Recall and F-Factor to ROC,informedness,markedness&correlation[J]. Journal of Machine Learning Technologies,2011,2(1):37-63.
[16] FRIEDMAN JEROME H. Stochastic gradient boosting[J]. Computational Statistics&Data Analysis,2002,38(4):367-378.
[17]乔建平．长江三峡库区蓄水后滑坡危险性预测研究———以重庆市万州区库岸为例[M]．北京:科学出版社，2012.QIAO Jianping. Study on the risk prediction of landslide after impoundment in the Three Gorges reservoir area of the Yangtze River:taking the bank of Wanzhou District in Chongqing as an example[M].Beijing:Science Press,2012.
[18]殷跃平．三峡库区边坡结构及失稳模式研究[J]．工程地质学报，2005,13(2):145-154.YIN Yueping. Human-cutting slope structure and failure pattern at the Three Gorges Reservoir[J].Journal of Engineering Geology,2005,13(2):145-154.
[19]乔素云，马云彪，尹思杰，等．基于岩土参数变异特性的三峡库区白水河滑坡失稳概率分析[J]．中国地质灾害与防治学报，2018,29(5):38-43.QIAO Suyun, MA Yunbiao, YIN Sijie, et al.Probability analysis of Baishuihe landslide in the Three Gorges Reservoir area based on variation characteristics of geotechnical parameters[J]. The Chinese Journal of Geological Hazard and Control,2018,29(5):38-43.
[20]简文星，许强，童龙云．三峡库区黄土坡滑坡降雨入渗模型研究[J]．岩土力学，2013,34(12):3527-3533.JIAN Wenxing,XU Qiang,TONG Longyun. Rainfall infiltration model of Huangtupo landslide in Three Gorges Reservoir area[J]. Rock and Soil Mechanics,2013,34(12):3527-3533.
[21]郑颖人，唐晓松．库水作用下的边(滑)坡稳定性分析[J]．岩土工程学报，2007,29(8):1115-1121.ZHENG Yingren,TANG Xiaosong. Stability analysis of slopes under drawdown condition of reservoirs[J].Chinese Journal of Geotechnical Engineering,2007,29(8):1115-1121.
[22]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.
[23] MOORE D S, MCCABE G P, CRAIG B A.Introduction to the practice of statistics[M]. New York:W. H. Freeman and Company,2012.
[2]易庆林，张明玉，文凯，等．三峡库区白水河滑坡变形特征及影响因素的阶段分析[J]．三峡大学学报(自然科学版)，2017,39(1):38-42.YI Qinglin,ZHANG Mingyu,WEN Kai,et al. Periodic analysis of deformation characteristics and influential factors of Baishuihe Landslide in Three Gorges Reservoir area[J]. Journal of China Three Gorges University(Natural Sciences),2017,39(1):38-42.
[3]黄达，顾东明，陈智强，等．三峡库区塔坪H2古滑坡台阶状复活变形的库水-降雨耦合作用机制[J]．岩土工程学报，2017,39(12):2203-2211.HUANG Da,GU Dongming,CHEN Zhiqiang,et al.Hybrid effects of rainfall and reservoir level fluctuation on old Taping H2 landslide in Wushan County in Three Gorges Reservoir area[J]. Chinese Journal of Geotechnical Engineering,2017,39(12):2203-2211.
[4]张振华，钱明明，位伟．基于改进破坏接近度的千江坪岸坡失稳机制分析[J]．岩石力学与工程学报，2018,37(6):1371-1384.ZHANG Zhenhua, QIAN Mingming, WEI Wei.Analysis of failure mechanism of Qianjiangping slope based on improved failure approach index[J].Chinese Journal of Rock Mechanics and Engineering,2018,37(6):1371-1384.
[5]陈卫兵，郑颖人，冯夏庭，等．考虑岩土体流变特性的强度折减法研究[J]．岩土力学，2008,29(1):101-105.CHEN Weibing,ZHENG Yingren,FENG Xiating,et al. Study on strength reduction technique considering rheological property of rock and soil medium[J].Rock and Soil Mechanics,2008,29(1):101-105.
[6]赵久彬，刘元雪，宋林波，等．大数据关键技术在滑坡监测预警系统中的应用[J]．重庆理工大学学报(自然科学)，2018,32(2):182-190.ZHAO Jiubin,LIU Yuanxue,SONG Linbo,et al.Application of big data processing and monitoring technologies in landslide monitoring and warning system[J]. Journal of Chongqing University of Technology(Natural Science),2018,32(2):182-190.
[7]彭令，牛瑞卿，吴婷．时间序列分析于支持向量机的滑坡位移预测[J]．浙江大学学报(工学版)，2013,47(9):1672-1679.PENG Ling,NIU Ruiqing,WU Ting. Time series analysis and support vector machine for landslide displacement prediction[J]. Journal of Zhejiang University(Engineering Science),2013,47(9):1672-1679.
[8] MIAO F S,WU Y P,XIE Y H,et al. Prediction of landslide displacement with step like behavior based on multialgorithm optimization and a support vector regression model[J]. Landslides,2017,15(3):475-488.
[9]李骅锦，许强，何雨森，等．“阶跃式”滑坡位移预测及阈值分析的ARMA-(LASSO-ELM)-Copula模型[J]．岩石力学与工程学报．2017,36(增刊2):4075-4084.LI Huajin,XU Qiang,HE Yusen,et al. An ARMA-(LASSO-ELM)-Copula framework for landslide displacement prediction and threshold computing of the displacement of step-like landslides[J]. Chinese Journal of Rock Mechanics and Engineering,2017,36(S2):4075-4084.
[10]杨背背，殷坤龙，杜娟．基于时间序列与长短期记忆网络的滑坡位移动态预测模型[J]．岩石力学与工程学报，2018,37(10):2334-2343.YANG Beibei,YIN Kunlong,DU Juan. A model for predicting landslide displacement based on time series and long and short term memory neural network[J].Chinese Journal of Rock Mechanics and Engineering,2018,37(10):2334-2343.
[11] CHAWLA N V,BOWYER K W,HALL L O,et al.SMOTE:synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research,2002,16(1):321-357.
[12] HAN H,WANG W Y,MAO B H. BorderlineSMOTE:a new over-sampling method in imbalanced data sets learning[J]. Advances in Intelligent Computing,2005,3644(1):878-887.
[13]林威伟，钟登华，胡炜，等．基于随机森林算法的土石坝压实质量动态评价研究[J]．水利学报，2018,49(8):945-955.LIN Weiwei,ZHONG Denghua,HU Wei,et al. Study on dynamic evaluation of compaction quality of earth rock dam based on Random Forest[J]. Journal of Hydraulic Engineering,2018,49(8):945-955.
[14] ZHANG C S,LIU C C,ZHANG X L,et al. An up-todate comparison of state-of-the-art classification algorithms[J]. Expert Systems with Applications,2017,82:128-150.
[15] POWERS D M W. Evaluation:from Precision,Recall and F-Factor to ROC,informedness,markedness&correlation[J]. Journal of Machine Learning Technologies,2011,2(1):37-63.
[16] FRIEDMAN JEROME H. Stochastic gradient boosting[J]. Computational Statistics&Data Analysis,2002,38(4):367-378.
[17]乔建平．长江三峡库区蓄水后滑坡危险性预测研究———以重庆市万州区库岸为例[M]．北京:科学出版社，2012.QIAO Jianping. Study on the risk prediction of landslide after impoundment in the Three Gorges reservoir area of the Yangtze River:taking the bank of Wanzhou District in Chongqing as an example[M].Beijing:Science Press,2012.
[18]殷跃平．三峡库区边坡结构及失稳模式研究[J]．工程地质学报，2005,13(2):145-154.YIN Yueping. Human-cutting slope structure and failure pattern at the Three Gorges Reservoir[J].Journal of Engineering Geology,2005,13(2):145-154.
[19]乔素云，马云彪，尹思杰，等．基于岩土参数变异特性的三峡库区白水河滑坡失稳概率分析[J]．中国地质灾害与防治学报，2018,29(5):38-43.QIAO Suyun, MA Yunbiao, YIN Sijie, et al.Probability analysis of Baishuihe landslide in the Three Gorges Reservoir area based on variation characteristics of geotechnical parameters[J]. The Chinese Journal of Geological Hazard and Control,2018,29(5):38-43.
[20]简文星，许强，童龙云．三峡库区黄土坡滑坡降雨入渗模型研究[J]．岩土力学，2013,34(12):3527-3533.JIAN Wenxing,XU Qiang,TONG Longyun. Rainfall infiltration model of Huangtupo landslide in Three Gorges Reservoir area[J]. Rock and Soil Mechanics,2013,34(12):3527-3533.
[21]郑颖人，唐晓松．库水作用下的边(滑)坡稳定性分析[J]．岩土工程学报，2007,29(8):1115-1121.ZHENG Yingren,TANG Xiaosong. Stability analysis of slopes under drawdown condition of reservoirs[J].Chinese Journal of Geotechnical Engineering,2007,29(8):1115-1121.
[22]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.
[23] MOORE D S, MCCABE G P, CRAIG B A.Introduction to the practice of statistics[M]. New York:W. H. Freeman and Company,2012.