Study of in situ stress inversion of deeply incised valleys considering the river evolution process

• Published in: Scientific reports Vol. 15; no. 1; pp. 17295 - 19
• Main Authors: Wang, Xianliang, Zhao, Changgui, Cheng, Fenyuan, Ma, Zhiwei, Qi, Wei, Dou, Yinling, Lan, Hao, Zhang, Jianhai, He, Benguo, Peng, Lingying, Du, Wei
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 19.05.2025; Nature Publishing Group; Nature Portfolio
• Subjects: 639/166/986; 704/2151; Deep V-shaped river Valley; Evolution; Excavation; Genetic algorithm; Humanities and Social Sciences; Hydroelectric power; In situ stress inversion; multidisciplinary; Multistep excavation; River beds; River erosion; River valleys; Rivers; Science; Science (multidisciplinary); Stress; Stress concentration; Tectonics; In situ stress inversion; River erosion; Deep V-shaped river Valley; Genetic algorithm; Multistep excavation
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-025-02753-x

The study of the in situ stress field is significant for the stability of large underground caverns near deep V-shaped valleys. Tectonic movements, gravity and river erosion have a great influence on the formation of the modern stress field. With the existing inversion method of the stress field in river valleys, it is assumed that the current stress field was formed in an ancient time. The inversion results are derived by first calculating the influence of tectonic movements and gravity on the ancient stress field, followed by simulating modern river erosion through multistep excavation. Thus, the influence of tectonic movements on the process of river erosion is ignored. To solve this problem, an optimized inversion method considering tectonic movements during different periods of river erosion (ORE) is proposed based on multistep excavation and a genetic algorithm (GA). In this method, unit extrusion, shear movements and gravity are carried out at different periods of valley evolution to simulate the effects of tectonic movements at different geological times on the current geostress field. The method is applied to the ground stress inversion of the Shuangjiangkou hydropower station, and the relative errors between the inversion results and the field-measured data are less than 30%. By comparing the results with those of the multiple regression method and multistep excavation method, it is found that the unloading phenomenon of slopes and the stress concentration phenomenon below the riverbed can be simulated with accuracy. This indicates that the proposed method can provide a reference for the design and construction of engineering work in similar locations. Finally, by analyzing the ground stress field of the underground main powerhouse in Shuangjiangkou, it is found that high-stress-induced disasters such as rock bursts and large deformations may occur in the area.