针对传统滑坡位移预测过程中的不足,提出了一种基于最小二乘支持向量机(LSSVM)的滑坡位移预测方法。以某流域大华滑坡为例,基于时序分析和集合经验模态分解法(EEMD)将原始序列重构为趋势项和波动项,趋势项位移受滑坡内部因素影响,采用最小二乘法与多项式方程进行拟合预测;波动项位移受库水位、降雨、地下水位等周期性因素影响,结合灰色关联度法和核主成分分析法(KPCA)对输入因子进行筛选与降维,并用粒子群算法-最小二乘支持向量机耦合模型(PSO-LSSVM)进行建模预测。最后将趋势项与周期项预测位移相加得到累计预测位移,并对模型预测精度进行定量分析。结果表明,建立的EEMD-KPCA-PSO-LSSVM组合模型预测效果良好,较传统BP神经网络、LSSVM等单一模型有着更高的预测精度,可为同类型滑坡位移预测提供新的思路。
Abstract
A method of landslide displacement prediction based on least squares support vector machine (LSSVM) is proposed to overcome the shortcomings of traditional prediction for Dahua landslide in Lancang River basin as a case study. The original sequence is decomposed and reconstructed into trend item and fluctuation item based on time series analysis and ensembled empirical mode decomposition (EEMD). The displacement of trend item is affected by internal factors of landslide,and the least square method and polynomial equation are used for fitting prediction. The displacement of fluctuation item is affected by periodic factors such as reservoir water level,rainfall and groundwater level. The grey correlation degree method and kernel principal component analysis (KPCA) are combined to screen and reduce the dimension of input factors,and then PSO-LSSVM coupling model is used for predicting the fluctuation item. Finally,the cumulative predicted displacement is obtained by adding the predicted displacement of trend item and fluctuation item,and the model prediction accuracy is quantitatively analyzed. Results demonstrate that the EEMD-KPCA-PSO-LSSVM combined model has a sound prediction performance with higher prediction accuracy than traditional BP neural network,LSSVM and other single models.
关键词
水动力型滑坡 /
位移预测 /
集合经验模态分解 /
核主成分分析 /
最小二乘支持向量机
Key words
water-induced landslide /
displacement prediction /
EEMD /
KPCA /
LSSVM
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 周家文.水动力型滑坡灾害机理与防控[J].工程科学与技术:专栏简介,2019,51(4):11-11.
[2] 杨 帆,许 强,范宣梅,等.基于时间序列与人工蜂群支持向量机的滑坡位移预测研究[J].工程地质学报,2019,27(4):880-889.
[3] 韩 斐,牛瑞卿,李士垚,等.基于变分模态分解和深度置信神经网络模型的滑坡位移预测[J].长江科学院院报,2020,37(8):61-68.
[4] 徐 峰,汪 洋,杜 娟,等.基于时间序列分析的滑坡位移预测模型研究[J].岩石力学与工程学报,2011,30(4):746-751.
[5] 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.
[6] 鄢 好,李绍红,吴礼舟.联合多种数据驱动建模方法的滑坡位移预测研究[J].工程地质学报,2019,27(2):459-465.
[7] 文 静.基于小波与灰色模型的滑坡时间预测预报[D].西安:长安大学,2013.
[8] ZHOU Z,ZHANG J,PENG J.The Application of Wavelet Analysis and Support Vector Machine Coupling Model in Displacement Prediction of Landslide[J].Electronic Journal of Geotechnical Engineering,2015,20(16):6823-6833.
[9] 姚 琦,牛瑞卿,赵金童,等.基于经验模态分解-支持向量机的滑坡位移预测方法研究[J].安全与环境工程,2017,24(1):26-32.
[10] 孟繁林.集合经验模态分解的理论及应用研究[D].镇江:江苏科技大学,2013.
[11] 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.
[12] 彭 令,牛瑞卿,赵艳南,等.基于核主成分分析和粒子群优化支持向量机的滑坡位移预测[J].武汉大学学报(信息科学版),2013,38(2):148-161.
[13] CAI Z,XU W,MENG Y,et al.Prediction of Landslide Displacement Based on GA-LSSVM with Multiple Factors[J].Bulletin of Engineering Geology & the Environment,2016,75(2):637-646.
[14] PRADHAN B,LEE S.Landslide Risk Analysis using Artificial Neural Network Model Focusing on Different Training Sites[J].International Journal of Physical Sciences,2009,4(2):1-15.
[15] 胡 军,董建华,王凯凯,等.边坡稳定性的CPSO-BP模型研究[J].岩土力学,2016,37(增刊1):577-582.
[16] SUYKENS J,VANDEWALLE J.Least Squares Support Vector Machines Classifiers[J].Neural Process Letters,1999,9(3):293-300.
[17] 刘清山,汤 俊.基于相空间重构和最小二乘支持向量机的滑坡预测[J].浙江水利水电专科学校学报,2010,22(2):58-60.
[18] 韩贺鸣,张 磊,施 斌,等.基于光纤监测和PSO-SVM模型的马家沟滑坡深部位移预测研究[J].工程地质学报,2019,27(4):853-861.
[19] 王念秦,朱文博,郭有金.基于PSO-SVM模型的滑坡易发性评价[J].长江科学院院报,2021,38(4):56-62.
[20] WU Z,HUANG N E.Ensemble Empirical Mode Decomposition:A Noise-assisted Data Analysis Method[J].Advances in Adaptive Data Analysis,2009(1):1-41.
[21] 张浩然,韩正之,李昌刚.支持向量机[J].计算机科学,2002,29(12):135-137.
[22] 廉 城.基于极限学习的重大工程灾变滑坡预测研究[D].武汉:华中科技大学,2014.
[23] 邓冬梅,梁 烨,王亮清,等.基于集合经验模态分解与支持向量机回归的位移预测方法:以三峡库区滑坡为例[J].岩土力学,2017,38(12):3660-3669.
基金
国家重点研发计划项目(2017YFC1501100);江苏省六大人才高峰项目(JZ-008);江苏高校“青蓝工程”项目(苏教师[2020]10号)
PDF(7289 KB)
