«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《地球科学与环境学报》[ISSN:1672-6561/CN:61-1423/P]

卷:

第45卷

期数:

2023年第03期

页码:

511-521

栏目:

大地测量、遥感与地学大数据

出版日期:

2023-05-15

文章信息/Info

Title:

Improved Method for Fusion of Loess Landslide Monitoring Data Based on Feature Selection and Stepwise Regression

文章编号:

1672-6561(2023)03-0511-11

作者:

王 利¹; 2; 3; 张懿恺¹; 2; 3; 舒 宝¹; 2; 3*; 许 豪¹; 2; 3; 魏 拓¹; 2; 3; 雷体俊¹; 2; 3

(1. 长安大学 地质工程与测绘学院,陕西 西安 710054; 2. 地理信息工程国家重点实验室,陕西 西安 710054; 3. 长安大学 西部矿产资源与地质工程教育部重点实验室,陕西 西安 710054)

Author(s):

WANG Li¹; 2; 3;  ZHANG Yi-kai¹; 2; 3;  SHU Bao¹; 2; 3*;  XU Hao¹; 2; 3;  WEI Tuo¹; 2; 3;  LEI Ti-jun¹; 2; 3

(1. School of Geological Engineering and Geomatics, Chang'an University, Xi'an 710054, Shaanxi, China; 2. State Key Laboratory of Geo-information Engineering, Xi'an 710054, Shaanxi, China; 3. Key Laboratory of Western China's Mineral Resources and Geological Engineering of Ministry of Education, Chang'an University, Xi'an 710054, Shaanxi, China)

关键词:

黄土滑坡;  变形监测;  多源数据融合;  多点位监测;  加权关联度;  特征优选;  逐步回归;  黑方台

Keywords:

loess landslide;  deformation monitoring;  multi-source data fusion;  multi-point monitoring;  weighted correlation degree;  feature selection;  stepwise regression;  Heifangtai

分类号:

P228; P642.22

DOI:

10.19814/j.jese.2022.09064

文献标志码:

A

摘要:

针对滑坡监测多源异构数据融合处理中存在的影响因子筛选难、结果差异大、数据处理复杂程度高等问题,提出一种基于最大互信息系数(MIC)、灰色关联分析(GRA)和逐步回归的黄土滑坡多源多点位异构监测数据融合方法。该方法首先将最大互信息系数和灰色关联分析结合起来,采用基于加权关联度的特征优选方法综合筛选滑坡变形影响因子,提取具有代表性的影响因子并剔除关联性差的影响因子; 然后,通过逐步回归方法赋予各监测点位移和优选后的影响因子对应的重要性权重系数,获取多源异构数据融合序列; 最后,采用甘肃黑方台党川滑坡监测设备所获取的全球卫星导航系统(GNSS)监测数据、裂缝位移计数据及气象数据进行实验验证。结果表明:在滑坡变形影响因子筛选性能方面,基于加权关联度的特征优选方法优于传统的Pearson相关系数法; 基于特征优选和逐步回归的多源多点位异构数据融合模型的预测精度较传统的BP神经网络有所提升,其中均方根误差(RMSE)降低了51.8%,平均绝对百分比误差(MAPE)降低了2.26%,拟合优度达到了0.964。

Abstract:

Aiming at the problems of difficult screening of influencing factors, large difference of results and high complexity of data processing in multi-source heterogeneous data fusion processing of landslide monitoring, a multi-source and multi-point heterogeneous monitoring data fusion method of loess landslide based on maximal information coefficient(MIC), grey relational analysis(GRA)and stepwise regression was proposed. Firstly, the maximum mutual information coefficient and grey correlation analysis are combined, and the feature optimization method based on weighted correlation degree is used to comprehensively screen the influencing factors of landslide deformation, extract representative features and eliminate factors with poor correlation. Secondly, the importance weight coefficient corresponding to the displacement of each monitoring point and the optimized influence factor is given by the stepwise regression method, and the multi-source heterogeneous data fusion sequence is obtained. Finally, the global navigation satellite system(GNSS)observation data, crack displacement meter data and meteorological data obtained by Dangchuan landslide monitoring equipment in Heifangtai of Gansu are used for experimental verification. The results show that the feature selection method based on weighted correlation degree is superior to the traditional Pearson correlation coefficient method in the screening performance of landslide deformation influencing factors; compared with the traditional BP neural network, the prediction accuracy of the multi-source and multi-point heterogeneous data fusion model based on feature selection and stepwise regression is improved; the root mean square error(RMSE)is reduced by 51.8%, the mean absolute percentage error(MAPE)is reduced by 2.26%, and the goodness of fit reaches 0.964.

参考文献/References:

[1] 黄润秋.20世纪以来中国的大型滑坡及其发生机制[J].岩石力学与工程学报,2007,26(3):433-454.
HUANG Run-qiu.Large-scale Landslides and Their Sliding Mechanisms in China Since the 20th Century[J].Chinese Journal of Rock Mechanics and Engineering,2007,26(3):433-454.
[2] 杨 迁,王雁林,马园园.2001～2019年中国地质灾害分布规律及引发因素分析[J].地质灾害与环境保护,2020,31(4):43-48.
YANG Qian,WANG Yan-lin,MA Yuan-yuan.Distribution Rule and Influencing Factors of Geological Di-sasters from 2001 to 2019 in China[J].Journal of Geological Hazards and Environment Preservation,2020,31(4):43-48.
[3] SHI X G,JIANG H J,ZHANG L,et al.Landslide Displacement Monitoring with Split-bandwidth Interferometry:A Case Study of the Shuping Landslide in the Three Gorges Area[J].Remote Sensing,2017,9(9):937.
[4] WOODS A,MACCIOTTA R,HENDRY M T,et al.Updated Understanding of the Deformation Characteristics of the Checkerboard Creek Rock Slope Through GB-InSAR Monitoring[J].Engineering Geo-logy,2021,281:105974.
[5] 陈明金,欧阳祖熙,范国胜.基于数据融合的滑坡综合监测信息提取方法[J].大地测量与地球动力学,2007,27(6):77-81.
CHEN Ming-jin,OUYANG Zu-xi,FAN Guo-sheng.On Information Extraction Method Based on Data Fusion for Integrated Landslide Monitoring[J].Journal of Geodesy and Geodynamics,2007,27(6):77-81.
[6] 邱丹丹.基于多源数据融合的滑坡风险分析研究[D].武汉:中国地质大学,2017.
QIU Dan-dan.Landslide Risk Analysis Based on Mul-ti-source Data Fusion[D].Wuhan:China University of Geosciences,2017.
[7] 张 勤,白正伟,黄观文,等.GNSS 滑坡监测预警技术进展[J].测绘学报,2022,51(10):1985-2000.
ZHNAG Qin,BAI Zheng-wei,HUANG Guan-wen,et al.Review of GNSS Landslide Monitoring and Early Warning[J].Acta Geodaetica et Cartographica Sinica,2022,51(10):1985-2000.
[8] 朱 武,张 勤,朱建军,等.特大滑坡实时监测预警与技术装备研发[J].岩土工程学报,2022,44(7):1341-1350.
ZHU Wu,ZHANG Qin,ZHU Jian-jun,et al.Real-time Monitoring and Early Warning Technology for Huge Landslides[J].Chinese Journal of Geotechnical Engineering,2022,44(7):1341-1350.
[9] 许 强,董秀军,李为乐.基于天-空-地一体化的重大地质灾害隐患早期识别与监测预警[J].武汉大学学报(信息科学版),2019,44(7):957-966.
XU Qiang,DONG Xiu-jun,LI Wei-le.Integrated Spa-ce-air-ground Early Detection,Monitoring and Warning System for Potential Catastrophic Geohazards[J].Geomatics and Information Science of Wuhan University,2019,44(7):957-966.
[10] 许 强,彭大雷,何朝阳,等.突发型黄土滑坡监测预警理论方法研究:以甘肃黑方台为例[J].工程地质学报,2020,28(1):111-121.
XU Qiang,PENG Da-lei,HE Chao-yang,et al.Theory and Method of Monitoring and Early Warning for Sudden Loess Landslide:A Case Study at Heifangtai Terrace[J].Journal of Engineering Geology,2020,28(1):111-121.
[11] LAN H X,ZHOU C H,WANG L J,et al.Landslide Hazard Spatial Analysis and Prediction Using GIS in the Xiaojiang Watershed,Yunnan,China[J].Engineering Geology,2004,76(1/2):109-128.
[12] 兰恒星,周成虎,伍法权,等.GIS 支持下的降雨型滑坡危险性空间分析预测[J].科学通报,2003,48(5):507-512.
LAN Heng-xing,ZHOU Cheng-hu,WU Fa-quan,et al.Spatial Hazard Analysis and Prediction on Rainfall-induced Landslide Using GIS [J].Chinese Science Bulletin,2003,48(5):507-512.
[13] LAN H X,TIAN N M,LI L P,et al.Kinematic-based Landslide Risk Management for the Sichuan-Tibet Grid Interconnection Project(STGIP)in China[J].Engineering Geology,2022,308:106823.
[14] LAN H X,MARTIN C D,ZHOU C H,et al.Rockfall Hazard Analysis Using LiDAR and Spatial Modeling[J].Geomorphology,2010,118(1/2):213-223.
[15] LI L P,LAN H X,GUO C B,et al.A Modified Frequency Ratio Method for Landslide Susceptibility Assessment[J].Landslides,2017,14(2):727-741.
[16] KURADUSENGE M,KUMARAN S,ZENNARO M.Rainfall-induced Landslide Prediction Using Machine Learning Models:The Case of Ngororero District,Rwanda[J].International Journal of Environmental Research and Public Health,2020,17(11):4147.
[17] KIM J,LEE K,JEONG S,et al.GIS-based Prediction Method of Landslide Susceptibility Using a Rainfall Infiltration-groundwater Flow Model[J].Engineering Geology,2014,182:63-78.
[18] 陆付民,蒋廷耀.基于多因子及泰勒级数的滑坡变形预测模型研究[J].大地测量与地球动力学,2017,37(10):1029-1032.
LU Fu-min,JIANG Ting-yao.The Deformation Forecast Model of Landslides Based on Multiple Factors and Taylor Series[J].Journal of Geodesy and Geodynamics,2017,37(10):1029-1032.
[19] 柳 青,易 武.基于诱发因素响应与BP神经网络的滑坡位移预测预报[J].三峡大学学报(自然科学版),2019,41(3):41-45.
LIU Qing,YI Wu.Prediction of Landslide Displacement Based on Induced Factors Response and BP Neural Network[J].Journal of China Three Gorges University(Natural Sciences),2019,41(3):41-45.
[20] 李德营,殷坤龙.基于影响因子的GM(1,1)-BP模型在八字门滑坡变形预测中的应用[J].长江科学院院报,2013,30(2):6-11.
LI De-ying,YIN Kun-long.Application of GM(1,1)-BP Model to the Deformation Prediction of Bazimen Landslide in Consideration of Influencing Factors[J].Journal of Yangtze River Scientific Research Institute,2013,30(2):6-11.
[21] 尚 敏,廖 芬,马 锐,等.白家包滑坡变形与库水位、降雨相关性定量化分析研究[J].工程地质学报,2021,29(3):742-750.
SHANG Min,LIAO Fen,MA Rui,et al.Quantitative Correlation Analysis on Deformation of Baijiabao Landslide Between Rainfall and Reservoir Water Le-vel[J].Journal of Engineering Geology,2021,29(3):742-750.
[22] 汪 洋,殷坤龙,安关峰.滑坡敏感因子的灰色关联分析[J].岩土力学,2004,25(1):91-93.
WANG Yang,YIN Kun-long,AN Guan-feng.Grey Correlation Analysis of Sensitive Factors of Landslide[J].Rock and Soil Mechanics,2004,25(1):91-93.
[23] 王 利,许 豪,舒 宝,等.利用互信息和IPSO-LSTM进行滑坡监测多源数据融合[J].武汉大学学报(信息科学版),2021,46(10):1478-1488.
WANG Li,XU Hao,SHU Bao.et al.A Multi-source Heterogeneous Data Fusion Method for Landslide Monitoring with Mutual Information and IPSO-LSTM Neural Network[J].Geomatics and Information Science of Wuhan University,2021,46(10):1478-1488.
[24] 王 利,岳 聪,舒 宝,等.基于混沌时间序列的黄土滑坡变形预测方法及应用[J].地球科学与环境学报,2021,43(5):917-925.
WANG Li,YUE Cong,SHU Bao,et al.Chaotic Time Series Based Surface Displacement Prediction Method and Application to Loess Landslides[J].Journal of Earth Sciences and Environment,2021,43(5):917-925.
[25] 张美霞,李泰杰,杨 秀,等.基于特征优选和加权聚类的商场用电模式分析[J].电力科学与技术学报,2021,36(5):137-143.
ZHANG Mei-xia,LI Tai-jie,YANG Xiu,et al.Analysis of Power Consumption Mode for Shopping Malls Based on Feature Selection and Weighted Clustering[J].Journal of Electric Power Science and Technology,2021,36(5):137-143.
[26] 孔德兵,尚可政,王式功,等.基于逐步回归分析的西北地区东部雷暴概率预报方法研究[J].干旱气象,2016,34(1):181-187.
KONG De-bing,SHANG Ke-zheng,WANG Shi-gong,et al.Probability Forecast Method of Thunderstorm in East Region of Northwest China Based on Stepwise Regression Analysis [J].Journal of Arid Meteorology,2016,34(1):181-187.
[27] 赵超英,刘晓杰,张 勤,等.甘肃黑方台黄土滑坡InSAR识别、监测与失稳模式研究[J].武汉大学学报(信息科学版),2019,44(7):996-1007.
ZHAO Chao-ying,LIU Xiao-jie,ZHANG Qin,et al.Research on Loess Landslide Identification,Monitoring and Failure Mode with InSAR Technique in Heifangtai,Gansu[J].Geomatics and Information Science of Wuhan University,2019,44(7):996-1007.
[28] 亓 星,许 强,彭大雷,等.地下水诱发渐进后退式黄土滑坡成因机理研究:以甘肃黑方台灌溉型黄土滑坡为例[J].工程地质学报,2017,25(1):147-153.
QI Xing,XU Qiang,PENG Da-lei,et al.Mechanism of Gradual Retreat Loess Landslide Caused by Groundwater:A Case Study of the Irrigation Loess Landslide in Heifangtai,Gansu Province[J].Journal of Engineering Geology,2017,25(1):147-153.
[29] 王亚敏,冯 起,李宗省.1960～2005年西北地区低云量的时空变化及成因分析[J].地理科学,2014,34(5):635-640.
WANG Ya-min,FENG Qi,LI Zong-xing.Analysis of Low Cloud Amount Variations in the Northwest China During 1960-2005[J].Scientia Geographica Sinica,2014,34(5):635-640.
[30] 王智伟,王 利,黄观文,等.基于BP神经网络的滑坡监测多源异构数据融合算法研究[J].地质力学学报,2020,26(4):575-582.
WANG Zhi-wei,WANG Li,HUANG Guan-wen,et al.Research on Multi-source Heterogeneous Data Fusion Algorithm of Landslide Monitoring Based on BP Neural Network[J].Journal of Geomechanics,2020,26(4):575-582.
[31] 李秀珍,许 强,黄润秋,等.滑坡预报判据研究[J].中国地质灾害与防治学报,2003,14(4):5-11.
LI Xiu-zhen,XU Qiang,HUANG Run-qiu,et al.Research on Prediction Criterion for Temporary Prediction of Landslide[J].The Chinese Journal of Geologi-cal Hazard and Control,2003,14(4):5-11.
[32] 杨背背,殷坤龙,杜 娟.基于时间序列与长短时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 Bei-bei,YIN Kun-long,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.
[33] 张耀辉.多源异构滑坡监测数据融合方法研究[D].西安:长安大学,2021.
ZHANG Yao-hui.Research on Multi-source Heterogeneous Data Fusion Method for Landslide Monitoring[D].Xi'an:Chang'an University,2021.

相似文献/References:

[1]李同录,徐 平,李 萍.一种用于非饱和黄土钻探的双管螺旋钻[J].地球科学与环境学报,2009,31(02):185.
　LI Tong-lu,XU Ping,LI Ping.A Dual Tube Auger for Unsaturated Loess Drilling[J].Journal of Earth Sciences and Environment,2009,31(03):185.
[2]郑亚萌,戴福初.灌溉淋滤对原状黄土物理力学性质的影响[J].地球科学与环境学报,2017,39(04):575.
　ZHENG Ya-meng,DAI Fu-chu.Effect of Irrigation Leaching on Physical and Mechanical Properties of Undisturbed Loess[J].Journal of Earth Sciences and Environment,2017,39(03):575.
[3]王 利,岳 聪,舒 宝*,等.基于混沌时间序列的黄土滑坡变形预测方法及应用[J].地球科学与环境学报,2021,43(05):917.[doi:10.19814/j.jese.2021.03037]
　WANG Li,YUE Cong,SHU Bao*,et al.Chaotic Time Series Based Surface Displacement Prediction Method and Application to Loess Landslides[J].Journal of Earth Sciences and Environment,2021,43(03):917.[doi:10.19814/j.jese.2021.03037]
[4]孟振江,曹一迪,康尘云,等.降雨促发黄土滑坡的启动机制模拟[J].地球科学与环境学报,2023,45(03):474.[doi:10.19814/j.jese.2022.11004]
　MENG Zhen-jiang,CAO Yi-di,KANG Chen-yun,et al.Simulation of the Initiation Mechanism of Loess Landslide Promoted by Rainfall[J].Journal of Earth Sciences and Environment,2023,45(03):474.[doi:10.19814/j.jese.2022.11004]
[5]瞿 伟,刘祥斌,李久元,等.改进哈里斯鹰优化算法与BP神经网络组合的滑坡位移高精度预测模型[J].地球科学与环境学报,2023,45(03):522.[doi:10.19814/j.jese.2022.11062]
　QU Wei,LIU Xiang-bin,LI Jiu-yuan,et al.High-precision Landslide Displacement Prediction Model Based on IHHO Algorithm Combined with BP Neural Network[J].Journal of Earth Sciences and Environment,2023,45(03):522.[doi:10.19814/j.jese.2022.11062]

备注/Memo

备注/Memo:

收稿日期:2022-09-28; 修回日期:2023-01-30
基金项目:国家自然科学基金项目(41877289); 国家重点研发计划项目(2021YFC3000503,2021YFC3000501)
作者简介:王 利(1975-),男,新疆奇台人,长安大学教授,博士研究生导师,工学博士,E-mail:wangli@chd.edu.cn。
*通讯作者:舒 宝(1990-),男,湖北随州人,长安大学副教授,工学博士,E-mail:bao613@163.com。

常用功能

更新日期/Last Update: 2023-05-30