结合注意力机制和Bi-LSTM的降雨型滑坡位移预测

doi:10.13474/j.cnki.11-2246.2022.0267

测绘通报 ›› 2022, Vol. 0 ›› Issue (9): 74-79,104.doi: 10.13474/j.cnki.11-2246.2022.0267

结合注意力机制和Bi-LSTM的降雨型滑坡位移预测

唐菲菲¹, 唐天俊^1,2, 朱洪洲³, 胡川¹, 马英¹, 李昕⁴

1. 重庆交通大学智慧城市学院, 重庆 400074;
2. 重庆工商职业学院城市建设工程学院, 重庆 400052;
3. 重庆交通大学交通土建工程材料国家地方联合工程实验室, 重庆 400074;
4. 长安大学地质工程与测绘学院, 陕西西安 710054

收稿日期:2022-06-22 发布日期:2022-09-30
作者简介:唐菲菲(1980—),女,博士,副教授,研究方向为地质灾害监测预警。E-mail:fftang80@126.com
基金资助:
国家重点研发计划(2021YFB2600603);重庆市教委科学技术研究项目(KJQN201900728)

Rainfall landslide deformation prediction based on attention mechanism and Bi-LSTM

TANG Feifei¹, TANG Tianjun^1,2, ZHU Hongzhou³, HU Chuan¹, MA Ying¹, LI Xin⁴

1. School of Smart City, Chongqing Jiaotong University, Chongqing 400074, China;
2. College of Urban Construction Engineering, Chongqing Technology & Business Institute, Chongqing 400052, China;
3. National & Local Joint Engineering Laboratory of Transportation and Civil Engineering Material, Chongqing Jiaotong University, Chongqing 400074, China;
4. School of Geological Engineering and Geomantics, Chang'an University, Xi'an 710054, China

Received:2022-06-22 Published:2022-09-30

摘要/Abstract

摘要： 受季节降雨波动和邻近点位的牵引作用影响,滑坡位移呈阶梯状变化趋势。为有效预测该类滑坡的位移,本文提出一种基于注意力机制的双向长短时记忆(Bi-LSTM)神经网络位移预测模型。首先,建立滑坡监测累计位移时间序列模型,将滑坡累计位移分解为趋势项和周期项;然后,分析滑坡因子与趋势项及周期项的相关性,采用多项式回归对趋势项进行拟合,通过基于注意力机制的Bi-LSTM对周期项进行预测。试验结果表明:基于注意力机制的Bi-LSTM预测模型具有稳健的泛化能力,能有效捕获不同时序数据间的相关性;预测结果精度平均绝对误差为0.088 mm,平均均方误差为0.042 mm,相比常规的长短时记忆(LSTM)神经网络模型,本文方法的预测结果精度更高。

关键词: 降雨型滑坡, 位移预测, 时间序列分解, 注意力机制, 双向长短时记忆神经网络

Abstract: Affected by seasonal rainfall fluctuations and the traction of adjacent points, the landslide displacement shows a step-like change trend. In order to predict the displacement effectively, the attention mechanism is introduced into bidirectional long-short term memory (Bi-LSTM) neural network prediction model in this paper. Firstly, a landslide monitoring cumulative displacement time series model is established to decompose the landslide cumulative displacement into a trend item and a period item. Then, the correlation coefficient among the landslide factor, the trend item and the period item are analyzed, and the polynomial regression is used to fit the trend item. For the period item prediction, the attention mechanism-based Bi-LSTM neural network prediction model is constructed. Taking a landslide data in Chongqing as an example, the experiment result shows that the proposed model has better robust generalization ability and can capture the correlation between different time series data effectively. The average absolute error of prediction accuracy is 0.088 mm, and the average mean square error is 0.042 mm. Compared with common long-short term memory neural network, the model proposed in this paper has higher prediction accuracy.

Key words: rainfall landslide, deformation prediction, time series decomposition, attention mechanism, bidirectional long-short term memory neural network

中图分类号:

P237

唐菲菲, 唐天俊, 朱洪洲, 胡川, 马英, 李昕. 结合注意力机制和Bi-LSTM的降雨型滑坡位移预测[J]. 测绘通报, 2022, 0(9): 74-79,104.

TANG Feifei, TANG Tianjun, ZHU Hongzhou, HU Chuan, MA Ying, LI Xin. Rainfall landslide deformation prediction based on attention mechanism and Bi-LSTM[J]. Bulletin of Surveying and Mapping, 2022, 0(9): 74-79,104.

参考文献

[1] 雷德鑫, 易武.三峡库区王家坡滑坡降雨阈值分析[J].中国地质灾害与防治学报,2018,29(5):95-101.
[2] 杨攀, 杨军. 考虑前期降雨的边坡稳定降雨阈值曲面[J]. 岩土力学, 2015, 36(S1):169-174.
[3] 陈丽霞, 殷坤龙, 刘长春. 降雨重现期及其用于滑坡概率分析的探讨[J]. 工程地质学报,2012,20(5):745-750.
[4] 胡涛,樊鑫,王硕.考虑不同降雨概率和雨型的滑坡渗流场及稳定性变化率特征分析[J].科学技术与工程,2020,20(14):5521-5529.
[5] 张夏冉,殷坤龙,李烨,等.库水位下降和降雨对重庆万州下坪滑坡稳定性的影响[J].中国地质灾害与防治学报,2017,28(2):22-29.
[6] 唐扬,殷坤龙,唐子珺.基于HYDRUS的三舟溪滑坡降雨入渗规律研究[J].水文地质工程地质,2017,44(1):152-156.
[7] 周超, 殷坤龙, 黄发明. 混沌序列WA-ELM耦合模型在滑坡位移预测中的应用[J]. 岩土力学,2015, 36(9):2674-2680.
[8] 黄晓虎, 雷德鑫, 夏俊宝,等. 降雨诱发滑坡阶跃型变形的预测分析及应用[J]. 岩土力学, 2019, 40(9):3585-3592.
[9] 赵艳南,牛瑞卿,彭令,等. 基于粗糙集和粒子群优化支持向量机的滑坡变形预测[J]. 中南大学学报(自然科学版),2015, 46(6):2324-2332.
[10] ZHANG F Y, HUANG X W. Trend and spatiotemporal distribution of fatal landslides triggered by non-seismic effects in China[J]. Landslides, 2018, 15(8):1663-1674.
[11] CHEN H Q, ZENG Z G. Deformation prediction of landslide based on improved back-propagation neural network[J].Cognitive Computation,2013,5(1):56-62.
[12] GERS F A, SCHMIDHUBER J, CUMMINS F. Learning to forget:continual prediction with LSTM[J]. Neural Computation, 2000, 12(10):2451-2471.
[13] XU S L, NIU R Q. Displacement prediction of Baijiabao landslide based on empirical mode decomposition and long short-term memory neural network in Three Gorges area, China[J].Computers & Geosciences, 2018,111(2):87-96.
[14] CHO K,COURVILLE A,BENGIO Y.Describing multimedia content using attention-based encoder-decoder networks[J]. IEEE Transactions on Multimedia, 2015, 17(11):1875-1886.
[15] REN F, WU X L, ZHANG K X, et al. Application of wavelet analysis and a particle swarm-optimized support vector machine to predict the displacement of the Shuping landslide in the Three Gorges, China[J].Environmental Earth Sciences,2015,73(8):4791-4804.
[16] 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,2018,15(3):475-488.
[17] NIU J Y, CHEN J, XU Y T. Twin support vector regression with Huber loss[J]. Journal of Intelligent & Fuzzy Systems, 2017, 32(6):4247-4258.
[18] 许强, 曾裕平. 具有蠕变特点滑坡的加速度变化特征及临滑预警指标研究[J]. 岩石力学与工程学报, 2009, 28(6):1099-1106.
[19] 周美玲, 高春艳. 山区灾害滑坡风险非线性预测研究[J]. 灾害学, 2018, 33(4):23-27.

结合注意力机制和Bi-LSTM的降雨型滑坡位移预测

Rainfall landslide deformation prediction based on attention mechanism and Bi-LSTM

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

编辑推荐

Metrics

本文评价

[1]	潘建平, 李鑫, 孙博文, 胡勇, 李明明. 基于注意力密集连接金字塔网络的新增建设用地变化检测[J]. 测绘通报, 2022, 0(3): 41-46,59.
[2]	郁汀, 王铎, 陈钦. 基于伪语义相似度模型的中文地址匹配方法[J]. 测绘通报, 2022, 0(3): 101-106.
[3]	杨勇, 崔丽珍, 郭倩倩, 薛中浩. 煤矿井下基于Attention机制的LSTM地磁定位算法[J]. 测绘通报, 2022, 0(1): 72-78.
[4]	吴永静, 吴锦超, 林超, 窦宝成, 黎珂. 基于深度学习的高分辨率遥感影像光伏用地提取[J]. 测绘通报, 2021, 0(5): 96-101.
[5]	刘广进, 王光辉, 毕卫华, 刘慧杰, 杨化超. 改进D-LinkNet模型在国产卫星影像云检测中的应用[J]. 测绘通报, 2021, 0(11): 54-58,64.
[6]	马林飞, 倪欢, 周子涵. 基于注意力机制的高分遥感图像采石场识别[J]. 测绘通报, 2021, 0(11): 96-100.
[7]	胡建朗, 罗亚荣, 郭迟. 一种用于运动姿态评估的视觉伺服机器人[J]. 测绘通报, 2020, 0(8): 33-38.
[8]	张朕通, 单玉刚, 袁杰. 改进注意力机制的遥感地貌识别算法[J]. 测绘通报, 2020, 0(10): 93-96,100.