Spatiotemporal clustering analysis and zonal prediction model for deformation behavior of super-high arch dams

Abstract

Super-high arch dams are affected by similar environmental factors, and there is some spatial and temporal correlation among the deformation measurement points, while the deformation law varies in different parts of the arch dam. Using a single-point model for the deformation points or a single spatiotemporal model for the whole arch dam may ignore the spatiotemporal correlation of the measurement points or not consider the spatiotemporal variability between regions. In this study, a novel analytical model for multi-point deformation monitoring of super-arch dams is proposed. By constructing deformation similarity characteristics to characterize the deformation similarity degree, then using the clustering by fast search and find of density peaks method (CFSFDP) to obtain the measurement point regions with similar deformation laws, the deformation laws of different deformation regions are analyzed. Then, based on the spatiotemporal integrated autoregressive moving average model (STARIMA), a spatiotemporal deformation analysis model is constructed for the deformation of super-high arch dams. Through example analysis, the fitting and prediction performance of the model is better than that of the single measurement point model and the spatiotemporal hybrid model, which provides an efficient and convenient new method for deformation monitoring and analysis of super-high arch dams.