GLER-BiGRUnet: A Surface Deformation Prediction Model Fusing Multiscale Features of InSAR Deformation Information and Environmental Factors

Accurate surface deformation (SD) predictions are critical for early warning and timely remediation of infrastructure damage. However, the current SD prediction models do not integrate the multiscale features of InSAR SD and environmental factors (EFs), which make their prediction results inaccurate. To address these limitations, we proposed a bidirectional gated recurrent unit (BiGRU) multioutput SD prediction network (GLER-BiGRUnet), which mainly included global–local feature extraction (GLFE), multifactor cross-attention residual (MCAR), and local residual module embedded in self-attention mechanism (RCSA) modules. Specifically, dense and one-dimensional convolutional layers were concatenated in the GLFE module to extract global–local SD features. The long time-series dependence between EFs and SD was learned in the MCAR module using the multihead cross-attention mechanism to obtain the corresponding attention weight feature matrix. The residual connection and self-attention mechanisms were used in the RCSA module to merge the multiscale features and enhance the model fitting ability. We chose four typical regions in the permafrost area of Qinghai–Tibet Railway as the scene for the experiment. The spatial distribution and local profile exhibited relatively small discrepancies between the prediction results of the GLER-BiGRUnet model and the InSAR SD. Meanwhile, the average root-mean-square error of the GLER-BiGRUnet model in the four typical regions was 0.19 mm, and the proposed model had the best evaluation index compared with other SD prediction models. Additionally, the prediction trend of SD of the proposed GLER-BiGRUnet model was consistent with the original InSAR SD, and the prediction results were more stable than those of the other prediction models. The SD prediction model proposed in this article contributes to early warning of SD.