Inversion study of dam incidence angle under oblique incidence of seismic waves

• Published in: Computers and geotechnics Vol. 173; p. 106524
• Main Authors: Zhang, Hongyang, Li, Tong, Li, Xuan, Han, Liwei, Guo, Shaolei, Ding, Zelin
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.09.2024
• Subjects: Dam; Incident angle; Inversion; Neural network; Oblique incidence of seismic wave; SHAP; Incident angle; Dam; Inversion; SHAP; Neural network; Oblique incidence of seismic wave
• ISSN: 0266-352X; 1873-7633
• DOI: 10.1016/j.compgeo.2024.106524

Determination of seismic wave incidence angle is a critical step in performing seismic oblique incidence analysis of dams. However, for structures with complex dynamic responses, such as dams, it is difficult to derive the incident angle by traditional mathematical methods. For this reason, this paper proposes a method for determining the seismic wave incidence angle based on the dynamic response of the dam. The model calculates the dynamic response of the dam under different incidence angles of seismic waves, establishes the mapping relationship between the dynamic response and the incidence angle by using a hybrid particle network, and performs feature screening to obtain the optimal inversion features by the SHAP method, and then constructs the optimal inversion model. The results show that there is a significant correlation between the dynamic response of the dam body and the incident angle, the correlation coefficients of the model are all greater than 0.96, and the average absolute percentage errors are less than 5 %, which proves that the model can effectively invert the incident angle. The method can also be applied to the seismic safety analysis of other structures, providing a new way to determine the seismic wave incidence angle.