An efficient double-loop reliability-based optimization with metaheuristic algorithms to design soil nail walls under uncertain condition

Abstract

One of the applicable methods for stabilizing soil walls is the nailing system consisting of tensile struts. The stability and safety of soil nail wall systems are affected by the geometrical parameters of the nailing system. Commonly, the determination of nailing parameters to achieve optimum performance of the nailing system for the safety of soil walls is described in the framework of optimization problems. Also, according to the various uncertainty in mechanical parameters of soil structures, it is essential to assess the system's reliability as a probabilistic problem. In this paper, the optimum design of the nailing system is carried out in deterministic and probabilistic cases using metaheuristic and reliability-based design optimization methods. The enhanced atomic orbital search optimization algorithm and first-order reliability method are used for optimization and reliability analysis problems. The geometrical on mechanical properties of the nailing system are selected as design vari-ables, and the mechanical properties of soil are selected as random variables. The results indicate that the reliability of the optimally designed soil nail system is sensitive to uncertainty in soil mechanical parameters. Reliability-based design optimization outcomes show that a nailing system can be designed for the expected reliability and failure probability level.