An interpolation method for filling in missing data of arch dam deformation using a spatial panel data model

The absence of deformation monitoring data in arch dams can lead to difficulties and inaccuracies in analyzing their safety behavior. Therefore, it is crucial to employ a scientific and reasonable method to fill in missing data and obtain complete and reliable monitoring data. Traditional approaches for handling missing values only consider the local spatial correlation among measurement points in arch dams. However, it is necessary to consider the spatiotemporal correlation across all measurement points, considering that an arch dam functions as an inseparable entity. In this paper, we introduce a spatial weight matrix that takes into account the integrity of the arch dam and the spatial correlation among measurement points. We demonstrate a significant positive spatial autocorrelation in the deformation data of arch dams. Based on this, we propose a method to fill in missing deformation data using a spatial panel data model, which considers the spatiotemporal correlation of the entire arch dam. Finally, we validate the proposed model by applying it to an arch dam project. The results indicate that the interpolation residual error of the proposed method is lower than the error limit specified in The Code for Dam Safety Monitoring (SL 601—2013). Consequently, the proposed method effectively handles missing deformation monitoring data.