基于长短期记忆网络的甘肃舟曲立节北山滑坡变形预测

高子雁; 李瑞冬; 石鹏卿; 周小龙; 张娟

doi:10.16031/j.cnki.issn.1003-8035.202303062

基于长短期记忆网络的甘肃舟曲立节北山滑坡变形预测

doi: 10.16031/j.cnki.issn.1003-8035.202303062

高子雁^{1, 2, 3,},
李瑞冬^{1, 3},
石鹏卿^{1, 2, 3},
周小龙^{1, 2, 3},
张娟^{1, 2, 3}

1.
甘肃省地下水工程及地热资源重点实验室，甘肃兰州　730050
2.
甘肃省自然资源厅地质灾害防治技术指导中心，甘肃兰州　730050
3.
甘肃省地质环境监测院，甘肃兰州　730050

基金项目: 甘肃省自然资源厅科技创新项目（202257）；甘肃省科技重大专项（19ZD2FA002）

详细信息

作者简介:
高子雁（1999-），女，甘肃兰州人，本科，助理工程师，主要从事地质灾害早期识别工作。E-mail：1269782387@qq.com

中图分类号: P642.22
计量
- 文章访问数: 18
- HTML全文浏览量: 1
- PDF下载量: 3
- 被引次数: 0
出版历程
- 收稿日期: 2023-03-27
- 修回日期: 2023-09-27
- 网络出版日期: 2023-11-08

Deformation prediction of the Northern Mountain landslide in Lijie Town of Zhouqu, Gansu Province based on long-short term memory network

Ziyan GAO^{1, 2, 3
,},
Ruidong LI^{1, 3},
Pengqing SHI^{1, 2, 3},
Xiaolong ZHOU^{1, 2, 3},
Juan ZHANG^{1, 2, 3}

1.
Key Laboratory of Groundwater Engineering and Geothermal Resources in Gansu Province, Lanzhou, Gansu　730050, China
2.
Geological Disaster Prevention and Control Technology Guidance Center, Gansu Provincial Department of Natural Resources, Lanzhou, Gansu　730050, China
3.
Gansu Institute of Geological Environment Monitoring, Lanzhou, Gansu　730050, China

摘要

摘要: 立节镇北山滑坡长期处于蠕动变形状态，已多次发生过滑坡、泥石流灾害。监测滑坡的地表形变，以掌握灾害体地表形变规律，是实现地质灾害预警预报的可靠依据。文中引入一种机器学习的模型进行相关数据预测，通过立节北山监测点位移数据，运用长短期记忆网络（LSTM）对立节北山滑坡变形进行预测，并且将预测结果与实际数据进行比对和分析。文章预测结果评价指标选用均方根误差（RMSE）、平均绝对误差（MAE）、决定系数（R²）以及可解释方差，其中决定系数（R²）和可解释方差均达到0.99，预测值和真实值的拟合均方根误差（RMSE）和平均绝对误差（MAE）也表现较低，说明本文采用的长短期记忆网络（LSTM）在立节北山滑坡变形的预测中达到了良好的预测性能。
- 滑坡 /
- 长短期记忆网络 /
- 预测分析 /
- 立节北山 /
- 机器学习
Abstract: The North Mountain landslide in Lijie Town has been in a long-term creeping deformation state and has experienced multiple landslide and debris flow disasters. Monitoring the surface deformation of landslide to grasp the surface deformation pattern of disaster body is a reliable basis for realizing early warning prediction of geological disaster. In this paper, a machine learning model is introduced to predict the relevant data, and a long and short-term memory network (LSTM) is used to predict the landslide deformation by monitoring the displacement data of North Mountain in Lijie, and the prediction results are compared with the actual data and analyzed. In this paper, root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R²) and explainable variance are used to evaluate the prediction results, among which the coefficient of determination (R²) and explainable variance reach 0.99. It shows that the long short-term memory network (LSTM) used in this paper achieves good prediction performance in the prediction of landslide deformation in the North Mountain of Lijie.
- landslide /
- LSTM neural network /
- predictive analysis /
- North Mountain of Lijie /
- machine learning

HTML全文

图 1 立节北山滑坡GNSS分布图

Figure 1. The North Mountain of Lijie landslide GNSS distribution map

下载: 全尺寸图片幻灯片

图 2 LSTM模型结构

Figure 2. LSTM model structure

下载: 全尺寸图片幻灯片

图 3 GNSS1累计位移与雨量关系

Figure 3. GNSS1 relationship between cumulative displacement and rainfall

下载: 全尺寸图片幻灯片

图 4 不同隐藏神经元数量的RMSE变化

Figure 4. RMSE variation with different numbers of hidden neurons

下载: 全尺寸图片幻灯片

图 5 LSTM模型训练中的损失函数数值变化

Figure 5. Numerical changes in loss function during LSTM model training

下载: 全尺寸图片幻灯片

图 6 GNSS1位移预测结果

Figure 6. GNSS1 displacement prediction results

下载: 全尺寸图片幻灯片

图 7 GNSS8位移预测结果

Figure 7. GNSS8 displacement prediction results

下载: 全尺寸图片幻灯片

图 8 GNSS1水平位移未来48 d预测结果

Figure 8. Forecasted results for horizontal displacement of GNSS1 for the next 48 days

下载: 全尺寸图片幻灯片

图 9 治理工程实施GNSS三维分布图

Figure 9. The GNSS three-dimensional distribution map of the governance project implementation

下载: 全尺寸图片幻灯片

表 1 GNSS1垂直位移精度评价指标

Table 1. Evaluation metrics for vertical displacement precision of GNSS1

评价指标 RMSE/mm MAE/mm R² Evar

数值 12.88 6.56 0.99 0.99

下载: 导出CSV

表 2 GNSS8垂直位移精度评价指标

Table 2. Evaluation metrics for vertical displacement precision of GNSS8

评价指标 RMSE/mm MAE/mm R² Evar

数值 6.63 5.66 0.99 0.99

下载: 导出CSV

表 3 GNSS8水平位移精度评价指标

Table 3. Evaluation metrics for horizontal displacement precision of GNSS8

评价指标 RMSE/mm MAE/mm R² Evar

数值 4.00 3.79 0.99 0.99

下载: 导出CSV

参考文献(16)

[1]	张勤,黄观文,杨成生. 地质灾害监测预警中的精密空间对地观测技术[J]. 测绘学报,2017,46(10):1300 − 1307. [ZHANG Qin,HUANG Guanwen,YANG Chengsheng. Precision space observation technique for geological hazard monitoring and early warning[J]. Acta Geodaetica et Cartographica Sinica,2017,46(10):1300 − 1307.(in Chinese with English abstract) ZHANG Qin, HUANG Guanwen, YANG Chengsheng. Precision space observation technique for geological hazard monitoring and early warning［J］. Acta Geodaetica et Cartographica Sinica, 2017, 46（10）: 1300 − 1307.（in Chinese with English abstract）
[2]	FROUDE M J,PETLEY D N. Global fatal landslide occurrence from 2004 to 2016[J]. Natural Hazards and Earth System Sciences,2018,18(8):2161 − 2181. doi: 10.5194/nhess-18-2161-2018
[3]	宋昭富, 张勇, 佘涛, 等. 基于易发性分区的区域滑坡降雨预警阈值确定—以云南龙陵县为例[J]. 中国地质灾害与防治学报,2023,34(4):22 − 29. [SONG Zhaofu, ZHANG Yong, SHE Tao, et al. Determination of regional landslide rainfall warning threshold based on susceptibility zoning: A case study in Longling County of Yunnan Province[J]. The Chinese Journal of Geological Hazard and Control,2023,34(4):22 − 29. (in Chinese with English abstract) SONG Zhaofu, ZHANG Yong, SHE Tao, et al. Determination of regional landslide rainfall warning threshold based on susceptibility zoning: A case study in Longling County of Yunnan Province［J］. The Chinese Journal of Geological Hazard and Control, 2023, 34（4）: 22 − 29. （in Chinese with English abstract）
[4]	唐亚明,张茂省,薛强,等. 滑坡监测预警国内外研究现状及评述[J]. 地质论评,2012,58(3):533 − 541. [TANG Yaming,ZHANG Maosheng,XUE Qiang,et al. Landslide monitoring and early-warning:An overview[J]. Geological Review,2012,58(3):533 − 541. (in Chinese with English abstract) TANG Yaming, ZHANG Maosheng, XUE Qiang, et al. Landslide monitoring and early-warning: An overview［J］. Geological Review, 2012, 58（3）: 533 − 541. （in Chinese with English abstract）
[5]	GLADE T,NADIM F. Early warning systems for natural hazards and risks[J]. Natural Hazards,2014,70(3):1669 − 1671. doi: 10.1007/s11069-013-1000-8
[6]	余凯,贾磊,陈雨强,等. 深度学习的昨天、今天和明天[J]. 计算机研究与发展,2013,50(9):1799 − 1804. [YU Kai,JIA Lei,CHEN Yuqiang,et al. Deep learning:Yesterday,today,and tomorrow[J]. Journal of Computer Research and Development,2013,50(9):1799 − 1804. (in Chinese with English abstract) YU Kai, JIA Lei, CHEN Yuqiang, et al. Deep learning: Yesterday, today, and tomorrow［J］. Journal of Computer Research and Development, 2013, 50（9）: 1799 − 1804. （in Chinese with English abstract）
[7]	杜续,冯景瑜,吕少卿,等. 基于随机森林回归分析的PM_2.5浓度预测模型[J]. 电信科学,2017,33(7):66 − 75. [DU Xu,FENG Jingyu,LYU Shaoqing,et al. PM_2.5 concentration prediction model based on random forest regression analysis[J]. Telecommunications Science,2017,33(7):66 − 75(in Chinese with English abstract) DU Xu, FENG Jingyu, LYU Shaoqing, et al. PM_2.5 concentration prediction model based on random forest regression analysis［J］. Telecommunications Science, 2017, 33（7）: 66 − 75（in Chinese with English abstract）
[8]	张品一,薛京京. 多分形互联网金融市场的风险预警模型研究[J]. 数量经济技术经济研究,2022,39(8):162 − 180. [ZHANG Pinyi,XUE Jingjing. Research on risk early warning model of multi-fractal Internet financial market[J]. The Journal of Quantitative & Technical Economics,2022,39(8):162 − 180. (in Chinese with English abstract) ZHANG Pinyi, XUE Jingjing. Research on risk early warning model of multi-fractal Internet financial market［J］. The Journal of Quantitative & Technical Economics, 2022, 39（8）: 162 − 180. （in Chinese with English abstract）
[9]	程宇也. 基于人工神经网络的短期电力负荷预测研究[D]. 杭州:浙江大学, 2017. [CHENG Yuye. Research on short-term power load forecasting based on artificial neural network[D]. Hangzhou:Zhejiang University, 2017. (in Chinese with English abstract) CHENG Yuye. Research on short-term power load forecasting based on artificial neural network［D］. Hangzhou: Zhejiang University, 2017. （in Chinese with English abstract）
[10]	李小妍. 基于图神经网络的交通流量预测[D]. 成都:电子科技大学, 2020. [LI Xiaoyan. Traffic flow forecast based on graph neural network[D]. Chengdu:University of Electronic Science and Technology of China, 2020. (in Chinese with English abstract) LI Xiaoyan. Traffic flow forecast based on graph neural network［D］. Chengdu: University of Electronic Science and Technology of China, 2020. （in Chinese with English abstract）
[11]	刘新,赵宁,郭金运,等. 基于LSTM神经网络的青藏高原月降水量预测[J]. 地球信息科学学报,2020,22(8):1617 − 1629. [LIU Xin,ZHAO Ning,GUO Jinyun,et al. Prediction of monthly precipitation over the Tibetan Plateau based on LSTM neural network[J]. Journal of Geo-Information Science,2020,22(8):1617 − 1629. (in Chinese with English abstract) LIU Xin, ZHAO Ning, GUO Jinyun, et al. Prediction of monthly precipitation over the Tibetan Plateau based on LSTM neural network［J］. Journal of Geo-Information Science, 2020, 22（8）: 1617 − 1629. （in Chinese with English abstract）
[12]	冯非凡,武雪玲,牛瑞卿,等. 一种V/S和LSTM结合的滑坡变形分析方法[J]. 武汉大学学报(信息科学版),2019,44(5):784 − 790. [FENG Feifan,WU Xueling,NIU Ruiqing,et al. A landslide deformation analysis method using V/S and LSTM[J]. Geomatics and Information Science of Wuhan University,2019,44(5):784 − 790. (in Chinese with English abstract) FENG Feifan, WU Xueling, NIU Ruiqing, et al. A landslide deformation analysis method using V/S and LSTM［J］. Geomatics and Information Science of Wuhan University, 2019, 44（5）: 784 − 790. （in Chinese with English abstract）
[13]	张宇,张鹏远,颜永红. 基于注意力LSTM和多任务学习的远场语音识别[J]. 清华大学学报(自然科学版),2018,58(3):249 − 253. [ZHANG Yu,ZHANG Pengyuan,YAN Yonghong. Long short-term memory with attention and multitask learning for distant speech recognition[J]. Journal of Tsinghua University (Science and Technology),2018,58(3):249 − 253. (in Chinese with English abstract) ZHANG Yu, ZHANG Pengyuan, YAN Yonghong. Long short-term memory with attention and multitask learning for distant speech recognition［J］. Journal of Tsinghua University （Science and Technology）, 2018, 58（3）: 249 − 253. （in Chinese with English abstract）
[14]	刘宇. 深度学习在图像处理中的应用——基于深度学习的视频帧预测算法研究[D]. 成都:电子科技大学, 2020. [LIU Yu. Application of deep learning in image processing:Research on video frame prediction algorithm based on deep learning[D]. Chengdu:University of Electronic Science and Technology of China, 2020. (in Chinese with English abstract) LIU Yu. Application of deep learning in image processing: Research on video frame prediction algorithm based on deep learning［D］. Chengdu: University of Electronic Science and Technology of China, 2020. （in Chinese with English abstract）
[15]	彭燕,刘宇红,张荣芬. 基于LSTM的股票价格预测建模与分析[J]. 计算机工程与应用,2019,55(11):209 − 212. [PENG Yan,LIU Yuhong,ZHANG Rongfen. Modeling and analysis of stock price forecast based on LSTM[J]. Computer Engineering and Applications,2019,55(11):209 − 212. (in Chinese with English abstract) PENG Yan, LIU Yuhong, ZHANG Rongfen. Modeling and analysis of stock price forecast based on LSTM［J］. Computer Engineering and Applications, 2019, 55（11）: 209 − 212. （in Chinese with English abstract）
[16]	杨青,王晨蔚. 基于深度学习LSTM神经网络的全球股票指数预测研究[J]. 统计研究,2019,36(3):65 − 77. [YANG Qing,WANG Chenwei. A study on forecast of global stock indices based on deep LSTM neural network[J]. Statistical Research,2019,36(3):65 − 77. (in Chinese with English abstract) doi: 10.19343/j.cnki.11-1302/c.2019.03.006 YANG Qing, WANG Chenwei. A study on forecast of global stock indices based on deep LSTM neural network［J］. Statistical Research, 2019, 36（3）: 65 − 77. （in Chinese with English abstract） doi: 10.19343/j.cnki.11-1302/c.2019.03.006