A Multi-Point Correlation Model to Predict and Impute Earth-Rock Dam Displacement Data for Deformation Monitoring
Abstract
1. Introduction
2. Theory and Methodology
2.1. Convolutional Neural Network (CNN)
2.2. Bidirectional LSTM Network (BiLSTM)
2.3. Attention Mechanism (AM)
3. Proposed CNN–BiLSTM–AM Hybrid Model and Evaluation Metrics
3.1. Framework of the CNN–BiLSTM–AM Hybrid Model
3.2. Multi-Point Modeling of Dam Displacement Based on the CNN–BiLSTM–AM Model
3.2.1. Basic Idea of Multi-Point Displacement Modeling
3.2.2. Technical Roadmap for the Multi-Point Displacement Model Proposed in This Paper
- Data cleaning: The goal of data cleaning is to improve the quality of the deformation monitoring data by eliminating outliers and noise. Data cleaning is typically achieved using scatter plots.
- Data standardization: Standardization reduces systematic error caused by data being collected at different monitoring points over different time periods. In addition, standardization helps mitigate discrepancies between datasets.
- Normalization: Normalization plays a significant role in accelerating the training process of neural networks and preventing issues such as gradient explosion [29]. Normalization ensures that the original characteristics of the data are preserved while improving the convergence speed of computations. In our model, input variables are normalized to the [0,1] range, as shown in Equation (11):
3.3. Evaluation Metrics
4. Experimental Study
4.1. Study Area
4.2. Dataset
4.3. Model Experiments
4.3.1. Deformation Prediction Experiments
4.3.2. Data Filling Experiment
5. Results and Analyses
5.1. Analysis of the Deformation Experiment Results
5.2. Analysis of Data Filling Experimental Results
5.3. The Limitations of the Model and Considerations in Practical Applications
5.3.1. The Constraints of Model Experiments
5.3.2. Considerations for Practical Applications of the Model
6. Conclusions
- ▪
- Improved Capture of Spatial Relationships: Unlike traditional single-point displacement models, the multi-point model can capture the spatial relationships between measurement points, overcoming the limitations resulting from relying solely on data from individual points. The experimental results confirm the superior predictive performance of the proposed model as well as its high reliability in filling missing values in monitoring datasets.
- ▪
- Superiority of Deep Learning Models: Compared with machine learning models, deep learning models show greater accuracy in both dam deformation prediction and data filling. This is largely attributed to the deeper, more complex network layers of deep learning models, which enable better feature extraction and improved prediction capabilities.
- ▪
- Hybrid Model Advantages: Achieving high prediction accuracy with a single network is challenging. However, hybrid models—such as the combination of CNNs with other models—can improve accuracy by leveraging a CNN’s ability to capture spatiotemporal features in the monitoring data. Although hybrid models increase complexity, the experimental results demonstrate that they outperform single models in predictive accuracy.
- ▪
- Applications in Future Work: For researchers and engineers aiming to enhance prediction accuracy, the CNN–BiLSTM–AM model is a robust choice. Compared with existing models, the proposed model achieves higher prediction accuracy, making it a valuable tool for dam monitoring data analysis and decision-making.
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Dai, B.; Gu, C.; Zhao, E.; Qin, X. Statistical Model Optimized Random Forest Regression Model for Concrete Dam Deformation Monitoring. Struct. Control. Health Monit. 2018, 25, e2170. [Google Scholar] [CrossRef]
- Kang, F.; Li, J.; Dai, J. Prediction of Long-Term Temperature Effect in Structural Health Monitoring of Concrete Dams Using Support Vector Machines with Jaya Optimizer and Salp Swarm Algorithms. Adv. Eng. Softw. 2019, 131, 60–76. [Google Scholar] [CrossRef]
- Cai, S.; Gao, H.; Zhang, J.; Peng, M. A Self-Attention-LSTM Method for Dam Deformation Prediction Based on CEEMDAN Optimization. Appl. Soft Comput. 2024, 159, 111615. [Google Scholar] [CrossRef]
- Mao, Y.; Li, J.; Qi, Z.; Yuan, J.; Xu, X.; Jin, X.; Du, X. Research on Outlier Detection Methods for Dam Monitoring Data Based on Post-Data Classification. Buildings 2024, 14, 2758. [Google Scholar] [CrossRef]
- Li, F.; Wang, Z.; Liu, G.; Fu, C.; Wang, J. Hydrostatic Seasonal State Model for Monitoring Data Analysis of Concrete Dams. Struct. Infrastruct. Eng. 2015, 11, 1616–1631. [Google Scholar] [CrossRef]
- Kang, F.; Liu, J.; Li, J.; Li, S. Concrete Dam Deformation Prediction Model for Health Monitoring Based on Extreme Learning Machine. Struct. Control Health Monit. 2017, 24, e1997. [Google Scholar] [CrossRef]
- Liu, M.; Feng, Y.; Yang, S.; Su, H. Dam Deformation Prediction Considering the Seasonal Fluctuations Using Ensemble Learning Algorithm. Buildings 2024, 14, 2163. [Google Scholar] [CrossRef]
- Li, Z.; Jia, H.; Zhang, Y.; Liang, J.; Abdelhafiz, A.E.; Cao, Q.; Lu, W. Deflection Statistical Monitoring Model Identification of the Concrete Gravity Dam Based on Uncertainty Analysis. Struct. Control Health Monit. 2022, 29, e3026. [Google Scholar] [CrossRef]
- Zhou, W.; Hua, J.; Chang, X.; Zhou, C. Settlement Analysis of the Shuibuya Concrete-Face Rockfill Dam. Comput. Geotech. 2011, 38, 269–280. [Google Scholar] [CrossRef]
- Szymanowski, M.; Kryza, M.; Spallek, W. Regression-Based Air Temperature Spatial Prediction Models: An Example from Poland. Meteorol. Z. 2013, 22, 577–585. [Google Scholar] [CrossRef]
- Alalade, M.; Nguyen-Tuan, L.; Wuttke, F.; Lahmer, T. Damage Identification in Gravity Dams Using Dynamic Coupled Hydro-Mechanical XFEM. Int. J. Mech. Mater. Des. 2018, 14, 157–175. [Google Scholar] [CrossRef]
- Li, R.; Jie, Y.; Pengli, Z.; Jiaming, W.; Chunhui, M.; Chao, C.; Lin, C.; Jian’e, W.; Mingjuan, Z. A Hybrid Monitoring Model of Rockfill Dams Considering the Spatial Variability of Rockfill Materials and a Method for Determining the Monitoring Indexes. J. Civ. Struct. Health Monit. 2022, 12, 817–832. [Google Scholar] [CrossRef]
- Zhang, K.; Gu, C.; Zhu, Y.; Li, Y.; Shu, X. A Mathematical-Mechanical Hybrid Driven Approach for Determining the Deformation Monitoring Indexes of Concrete Dam. Eng. Struct. 2023, 277, 115353. [Google Scholar] [CrossRef]
- Xu, C.; Yue, D.; Deng, C. Hybrid GA/SIMPLS as Alternative Regression Model in Dam Deformation Analysis. Eng. Appl. Artif. Intell. 2012, 25, 468–475. [Google Scholar] [CrossRef]
- Shao, C.; Gu, C.; Yang, M.; Xu, Y.; Su, H. A Novel Model of Dam Displacement Based on Panel Data. Struct. Control Health Monit. 2018, 25, e2037. [Google Scholar] [CrossRef]
- Liu, H.-F.; Ren, C.; Zheng, Z.-T.; Liang, Y.-J.; Lu, X.-J. Study of a Gray Genetic BP Neural Network Model in Fault Monitoring and a Diagnosis System for Dam Safety. ISPRS Int. J. GEO-Inf. 2018, 7, 4. [Google Scholar] [CrossRef]
- Ren, F.; Wu, X.; Zhang, K.; Niu, R. 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. Environ. Earth Sci. 2015, 73, 4791–4804. [Google Scholar] [CrossRef]
- Liu, W.; Pan, J.; Ren, Y.; Wu, Z.; Wang, J. Coupling Prediction Model for Long-Term Displacements of Arch Dams Based on Long Short-Term Memory Network. Struct. Control Health Monit. 2020, 27, e2548. [Google Scholar] [CrossRef]
- Wei, Y.; Li, Q.; Hu, Y.; Wang, Y.; Zhu, X.; Tan, Y.; Liu, C.; Pei, L. Deformation Prediction Model Based on an Improved CNN plus LSTM Model for the First Impoundment of Super-High Arch Dams. J. Civ. Struct. Health Monit. 2023, 13, 431–442. [Google Scholar] [CrossRef]
- Jiedeerbieke, M.; Li, T.; Chao, Y.; Qi, H.; Lin, C. Gravity Dam Deformation Prediction Model Based on I-KShape and ZOA-BiLSTM. IEEE Access 2024, 12, 50710–50722. [Google Scholar] [CrossRef]
- Luo, S.; Wei, B.; Chen, L. Multi-Point Deformation Monitoring Model of Concrete Arch Dam Based on MVMD and 3D-CNN. Appl. Math. Model. 2024, 125, 812–826. [Google Scholar] [CrossRef]
- Wei, B.; Liu, B.; Yuan, D.; Mao, Y.; Yao, S. Spatiotemporal Hybrid Model for Concrete Arch Dam Deformation Monitoring Considering Chaotic Effect of Residual Series. Eng. Struct. 2021, 228, 111488. [Google Scholar] [CrossRef]
- Yang, G. Deformation Similarity Characteristics-Considered Hybrid Panel Model for Multi-Point Deformation Monitoring of Super-High Arch Dams in Operating Conditions. Measurement 2022, 192, 110908. [Google Scholar] [CrossRef]
- LeCun, Y.; Boser, B.; Denker, J.S.; Henderson, D.; Howard, R.E.; Hubbard, W.; Jackel, L.D. Backpropagation Applied to Handwritten Zip Code Recognition. Neural Comput. 1989, 1, 541–551. [Google Scholar] [CrossRef]
- Pan, M.; Zhou, H.; Cao, J.; Liu, Y.; Hao, J.; Li, S.; Chen, C.-H. Water Level Prediction Model Based on GRU and CNN. IEEE Access 2020, 8, 60090–60100. [Google Scholar] [CrossRef]
- Bengio, Y.; Simard, P.; Frasconi, P. Learning Long-Term Dependencies with Gradient Descent Is Difficult. IEEE Trans. Neural Netw. 1994, 5, 157–166. [Google Scholar] [CrossRef]
- Graves, A.; Schmidhuber, J. Framewise Phoneme Classification with Bidirectional LSTM and Other Neural Network Architectures. Neural Netw. 2005, 18, 602–610. [Google Scholar] [CrossRef]
- Kavianpour, P.; Kavianpour, M.; Jahani, E.; Ramezani, A. A CNN-BiLSTM Model with Attention Mechanism for Earthquake Prediction. J. Supercomput. 2023, 79, 19194–19226. [Google Scholar] [CrossRef]
- Hrynaszkiewicz, I. A Call for BMC Research Notes Contributions Promoting Best Practice in Data Standardization, Sharing and Publication. BMC Res. Notes 2010, 3, 235. [Google Scholar] [CrossRef]
- Lu, W.; Li, J.; Wang, J.; Qin, L. A CNN-BiLSTM-AM Method for Stock Price Prediction. Neural Comput. Appl. 2021, 33, 4741–4753. [Google Scholar] [CrossRef]
- Kang, X.; Li, Y.; Zhang, Y.; Wen, L.; Sun, X.; Wang, J. PCA-IEM-DARNN: An Enhanced Dual-Stage Deep Learning Prediction Model for Concrete Dam Deformation Based on Feature Decomposition. Measurement 2025, 242, 115664. [Google Scholar] [CrossRef]
Model Type | Model Name | Parameters |
---|---|---|
Deep learning | CNN–BiLSTM–AM | Epoch = 70, batch size = 7, lr = 0.001, head = 1, keys = 2, convolution kernel size = 64, Nh = 64 |
BiLSTM–AM | Epoch = 70, batch size = 7, lr = 0.001, head = 1, keys = 2, Nh = 64 | |
BiLSTM | Epoch = 70, batch size = 7, lr = 0.001, Nh = 64 | |
LSTM | Epoch = 70, batch size = 7, lr = 0.001, Nh = 64 | |
CNN | Epoch = 70, batch size = 7, lr = 0.001, convolution kernel size = 64 | |
Machine learning | MLP | NI = 1, Nh = 64 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Pi, L.; Yue, C.; Shi, J. A Multi-Point Correlation Model to Predict and Impute Earth-Rock Dam Displacement Data for Deformation Monitoring. Buildings 2024, 14, 3780. https://doi.org/10.3390/buildings14123780
Pi L, Yue C, Shi J. A Multi-Point Correlation Model to Predict and Impute Earth-Rock Dam Displacement Data for Deformation Monitoring. Buildings. 2024; 14(12):3780. https://doi.org/10.3390/buildings14123780
Chicago/Turabian StylePi, Lilang, Chunfang Yue, and Jiachen Shi. 2024. "A Multi-Point Correlation Model to Predict and Impute Earth-Rock Dam Displacement Data for Deformation Monitoring" Buildings 14, no. 12: 3780. https://doi.org/10.3390/buildings14123780
APA StylePi, L., Yue, C., & Shi, J. (2024). A Multi-Point Correlation Model to Predict and Impute Earth-Rock Dam Displacement Data for Deformation Monitoring. Buildings, 14(12), 3780. https://doi.org/10.3390/buildings14123780