Abstract
Dams are the main water retaining structures in the hydraulic engineering field. Safe operations of dams are important foundations to ensure the hydraulic functionalities of these engineering structures. Deformation, as the most intuitive feature of the dams’ operation behaviors, can comprehensively reflect the dam structural states. In this case, the analysis of the dam prototype deformation data and the establishment of a real-time prediction model become frontier research contents in the field of dam safety monitoring. Considering the multi-nonlinear relationships between dam deformation and relative influential factors as well as the time lag effect of these influential factors, this article adopts long-short-term memory (LSTM) network algorithm in deep learning to deal with the long-term dependence existing in dam deformation and explore the deformation law. The method proposed in this work can effectively avoid the gradient disappearance and gradient explosion problems by using the recurrent neural network (RNN). In addition, this work adopts the Attention mechanism to screen the information that has significant influence on deformation, combining the Adam optimization algorithm that has high calculation efficiency and low memory requirement to improves the learning accuracy and speed of the LSTM. The model overfitting is avoided by applying the Dropout mechanism. The effectiveness of this proposed model in studing the long time series deformation prediction of concrete dams is confirmed by case studies, whose MSE (mean square error) and other 4 error indexes can be reduced.
Highlights
As the main water retaining structure in the hydraulic engineering field, the safe operation of the concrete dam is an important foundation to achieve many hydraulic functions
CONSTRUCTION OF THE PREDICTION MODEL On the premise of knowing the measured deformation data and environmental impact factors of concrete dam measurement points, the long-short-term memory (LSTM) network is used to extract the long-term and short-term features contained in the data; the Adam conducting optimization, Attention mechanism highlight the influence of important features on prediction; and the Dropout mechanism prevents model from overfitting
After the LSTM network updating the cell state and the Attention mechanism adjusting the weight, the calculated eigenvalue is decoded by the activation function at the full connection layer to obtain the final output – deformation value
Summary
As the main water retaining structure in the hydraulic engineering field, the safe operation of the concrete dam is an important foundation to achieve many hydraulic functions. In traditional deformation prediction models, most of the deformation values are expressed as polynomials of the main influential factors (water pressure, temperature, aging, etc.) through statistical calculation. Such methods lack the ability to express the multiple nonlinear relationship between dam deformation and influential factors, so their prediction results are not good enough [5], [6]. The data processing method, through continuous integration of advanced deep learning (neural network), optimization theory and mechanism into the construction of dam deformation prediction model, combined with the existing dam engineering theory and engineering experience, will have important guiding significance for the engineering practice. Where, c1, c2 are the regression coefficients of aging factor; θ is t/100
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
