A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Displacement prediction model for high arch dams using long short-term memory based encoder-decoder with dual-stage attention considering measured dam temperature
Highlights A novel deep learning model using encoder-decoder based on long short-term memory network with dual-stage attention for dam displacement prediction. The proposed approach is verified on a high arch dam during the initial impoundment period. Measured dam temperatures are adopted to represent the thermal effect on the dam displacement. The prediction performance of the proposed model is compared with the traditional statistical models, shallow network and other deep learning models. The method proposed in this study is effective and stable for various situations and outperforms other models.
Abstract Structural health monitoring method can provide important information to evaluate operational status of concrete dams, by establishing accurate models to predict concrete dam behavior with monitored data. This study proposed a model using encoder-decoder based on long short-term memory network with dual-stage attention mechanism (DALSTM) to predict the displacement of concrete arch dams. Encoder-decoder based on long short-term memory network is a deep learning technique that can perform time series prediction, and dual-stage attention mechanism focuses on the key information in the dam displacement series to improve the performance. The effectiveness and accuracy of the proposed prediction model are analyzed on a high arch dam using measured temperature in the dam body instead of the seasonal functions to represent the thermal effect. Compared with traditional stepwise regression, multiple linear regression models, radial basis function networks, and other deep learning models, results show that the proposed approach performance is more accurate and robust for dam health monitoring.
Displacement prediction model for high arch dams using long short-term memory based encoder-decoder with dual-stage attention considering measured dam temperature
Highlights A novel deep learning model using encoder-decoder based on long short-term memory network with dual-stage attention for dam displacement prediction. The proposed approach is verified on a high arch dam during the initial impoundment period. Measured dam temperatures are adopted to represent the thermal effect on the dam displacement. The prediction performance of the proposed model is compared with the traditional statistical models, shallow network and other deep learning models. The method proposed in this study is effective and stable for various situations and outperforms other models.
Abstract Structural health monitoring method can provide important information to evaluate operational status of concrete dams, by establishing accurate models to predict concrete dam behavior with monitored data. This study proposed a model using encoder-decoder based on long short-term memory network with dual-stage attention mechanism (DALSTM) to predict the displacement of concrete arch dams. Encoder-decoder based on long short-term memory network is a deep learning technique that can perform time series prediction, and dual-stage attention mechanism focuses on the key information in the dam displacement series to improve the performance. The effectiveness and accuracy of the proposed prediction model are analyzed on a high arch dam using measured temperature in the dam body instead of the seasonal functions to represent the thermal effect. Compared with traditional stepwise regression, multiple linear regression models, radial basis function networks, and other deep learning models, results show that the proposed approach performance is more accurate and robust for dam health monitoring.
Displacement prediction model for high arch dams using long short-term memory based encoder-decoder with dual-stage attention considering measured dam temperature
Huang, Ben (author) / Kang, Fei (author) / Li, Junjie (author) / Wang, Feng (author)
Engineering Structures ; 280
2023-01-20
Article (Journal)
Electronic Resource
English
Dual-stage attention-based long-short-term memory neural networks for energy demand prediction
| BASE | 2021
Short-term Inland Vessel Trajectory Prediction with Encoder-Decoder Models
| HENRY – Federal Waterways Engineering and Research Institute (BAW) | 2022
Self-Attention based encoder-Decoder for multistep human density prediction
| DOAJ | 2022