Research and Application of Hydropower Dam Deformation Monitoring Model Based on HBA‐GRU

Abstract The nonlinear features of dam deformation is one of the key factors that limits the accurate prediction of dam deformation. On this promise, this paper proposes a dam deformation prediction model that is applicable to the nonlinear features of dams. In particular, the Honey Badger Optimization Algorithm (HBA), which has strong seeking ability and high convergence accuracy, is integrated with the deep learning Gated Recurrent Unit (GRU) model, which is advantageous in nonlinear prediction problems, to generate HBA‐GRU for dam deformation monitoring. Experiments were carried out to apply the proposed method in the measured deformation monitoring of panel rock fill dams at Gongboxia Hydropower Station, and the effectiveness of the proposed method was compared with the existing prediction methods, such as LSTM, GRU, and HBA‐LSTM. The results show that compared with the existing prediction methods, the proposed method performs better than the LSTM, GRU and HBA‐LSTM algorithms, HBA‐GRU algorithm, and can provide technical support for dam deformation monitoring.