大坝变形监测数据序列具有非平稳、非线性特征,是水压、温度和时效综合作用的结果。引入集合经验模态分解(EEMD)方法处理变形数据,在得到多尺度大坝变形分量的基础上,对于其变化复杂的高频分量,采取长短期记忆神经网络(LSTM)以获得较优预测结果;对于周期性变化的低频分量,借助多元线性回归(MLR)实现快捷且有效的预测;最终通过分量重构,得到大坝变形的预测结果。工程实例分析表明:EEMD方法避免了模态混叠现象,可以得到更为合理的多尺度变形分量;LSTM和MLR分别对高、低频分量具有良好的预测能力。将分量叠加重构的最终结果分别与多种单一预测算法、基于EMD的组合算法以及传统模型等预测效果比较表明,基于EEMD-LSTM-MLR的组合预测模型的平均绝对误差(MAE)、平均绝对百分误差(MAPE)及均方根误差(RMSE)均低于上述对比模型,有着更高的预测精度,为大坝变形预测提供了新的思路。
Abstract
Under the combined action of water pressure, temperature and aging, dam deformation monitoring data series is non-stationary and nonlinear. We introduced the method of Ensemble Empirical Mode Decomposition (EEMD) to process deformation data and obtained the multi-scale dam deformation components. For the complex high frequency components, we employed the Long and Short Term Memory network (LSTM) to achieve better prediction results; for low frequency components with periodic changes, we adopted Multiple Linear Regression (MLR) for rapid and effective prediction. Through the refactoring of components we can acquire the predicted result of dam deformation. Analysis of engineering examples demonstrated that the EEMD method avoided modal aliasing and attained more reasonable multi-scale deformation components. LSTM and MLR have good predictive ability for high and low frequency components respectively. By comparing the result of components superposition separately with those of a variety of single prediction algorithms, EMD decomposition algorithm, and traditional models, we found that the Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and the Root Mean Square Error (RMSE) of the EEMD-LSTM-MLR combinatorial model were lower than the comparative models above, indicating higher prediction precision.
关键词
大坝变形 /
组合预测模型 /
集合经验模态分解 /
长短期记忆神经网络 /
多元线性回归
Key words
dam deformation /
combinatorial prediction model /
Ensemble Empirical Mode Decomposition /
Long and Short Term Memory network /
Multiple Linear Regression
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] SHU Xiao-song, BAO Teng-fei, LI Yang-tao, et al. Dam Safety Evaluation Based on Interval-Valued Intuitionistic Fuzzy Sets and Evidence Theory[J]. Sensors, 2020, 20(9): 2648.
[2] SU Huai-zhi, WEN Zhi-ping, SUN Xiao-ran, et al. Time-varying Identification Model for Dam Behavior Considering Structural Reinforcement[J]. Structural Safety, 2015, 57: 1-7.
[3] 吴中如.混凝土坝安全监控的确定性模型及混合模型[J].水利学报,1989(5) : 64-70。
[4] 罗亦泳,张 豪,张立亭. 基于遗传支持向量机的多维灰色变形预测模型研究[J]. 浙江工业大学学报,2010,38(1):79-83.
[5] 曹茂森,邱秀梅,夏 宁. 大坝安全诊断的混沌优化神经网络模型[J]. 岩土力学,2006(8):1344-1348.
[6] 苏怀智,温志萍,吴中如. 基于SVM理论的大坝安全预警模型研究[J]. 应用基础与工程科学学报,2009,17(1):40-48.
[7] 罗德河,郑东健. 大坝变形的小波分析与ARMA 预测模型[J]. 水利水运工程学报,2016(3) : 70-75.
[8] SU Huai-zhi, LI Xing, YANG Bei-bei, et al. Wavelet Support Vector Machine-based Prediction Model of Dam Deformation[J]. Mechanical Systems and Signal Processing, 2018, 110: 412-427.
[9] 曹恩华,包腾飞,刘永涛,等. 基于EMD-RVM-ARIMA的大坝变形预测模型及其应用[J]. 水利水电技术,2018,49(12):59-64.
[10] 吴 俊,张榆锋. 经验模态分解和小波分解滤波特性的比较研究[J]. 云南大学学报(自然科学版),2012,34(3):285-290,297.
[11] HUANG N E, SHEN Z, LONG S R, et al. The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinear and Non-stationary Time Series Analysis[J]. Proceedings Mathematical Physical & Engineering Sciences, 1998, 454(1971): 903-995.
[12] 郑旭东,陈天伟,王 雷,等. 基于EEMD-PCA-ARIMA模型的大坝变形预测[J]. 长江科学院院报,2020,37(3):57-63.
[13] MA Xiao-lei, TAO Zhi-min, WANG Yin-hai, et al. Longshort-term Memory Neural Network for Traffic Speed Prediction Using Remote Microwave Sensor Data[J]. Transportation Research Part C Emerging Technologies, 2015, 54: 187-197.
[14] LI Yang-tao, BAO Teng-fei, GONG Jian, et al. The Prediction of Dam Displacement Time Series Using STL, Extra-Trees, and Stacked LSTM Neural Network[J]. IEEE Access, 2020, 8: 94440-94452.
[15] 李 瑞,姜新元,秦 涛,等. 多元线性回归在大坝变形监测数据处理中的应用[J]. 黄河水利职业技术学院学报,2017,29(1):17-54.
[16] 邓带雨,李 坚,张真源,等. 基于EEMD-GRU-MLR的短期电力负荷预测[J]. 电网技术,2020,44(2):593-602.
[17] 张 豪,许四法. 基于经验模态分解和遗传支持向量机的多尺度大坝变形预测[J]. 岩石力学与工程学报,2011,30(增刊2):3681-3688.
[18] PASCUAL S, BONAFONTE A. Multi-output RNN-LSTM for Multiple Speaker Speech Synthesis with α-interpolation Model[C]//Proceedings of the 9th ISCA Speech Synthesis Workshop. Sunnyvale, CA: International Speech Communication Association (ISCA). September 13-15, 2016: 112-117.
基金
国家重点研发计划课题项目(2019YFC1510801,2018YFC0407101);国家自然科学基金项目(51979093)
PDF(4881 KB)
