A Multi-Point Correlation Model to Predict and Impute Earth-Rock Dam Displacement Data for Deformation Monitoring

Deformation is a critical indicator of structural integrity, and monitoring deformation is essential for ensuring the long-term safety of dams. However, characterizing the spatial correlations among dam deformation sequences and the similarity between displacements at various measurement points poses significant challenges when using single-point measurement models. Considering the limitations inherent in conventional models for processing spatiotemporal data, this paper introduces a novel model for predicting and imputing multi-point displacement monitoring data from earth-rock dams. The model integrates a convolutional neural network (CNN) with a bidirectional long short_term memory neural network (BiLSTM) while also incorporating an attention mechanism (AM). The CNN captures the spatial features of the displacement data, while the BiLSTM extracts temporal features. The AM assigns varying weights to input features, thereby enhancing the predictive accuracy of the model. The proposed model was experimentally validated, demonstrating its robust capabilities in data prediction and the imputation of missing data. The model provides a new strategy for forecasting dam deformation and addressing issues related to incomplete data.

This creative work has been published under the Creative Commons Attribution 4.0 International (CC-BY 4.0) license which allows copying, and redistribution as well as adaptation of the original work provided appropriate credit is given to the original author and the conditions of the license are met.