Determinationof critical failure surface in slope stability analysisusing optimization algorithms (PSO & GA)

The use of limit equilibrium methodsare the common approaches for determining of soil slope stability ingeotechnical engineering project. In these methods many researchers tend to find a circular criticalfailure surface for homogeneous soils, but failure surfaces can be non-circular especially for layeredslopes. Finding of critical failure surface of a general soil slope is a difficult problem as the non-convexand multiple minima exist of the safety factor objective function. In this paper simple genetic algorithmand particle swarm optimization methods are employed to find the critical non-circular failuresurface.For this purpose the Morgenstern–Price method is used to find the factor of safety. A computerprogram was coded by MATLAB for searching the minimum factor of safety.The result of a exampleiscompared withstress-deformation method using PLAXISsoftware for one example. It shows thedetermined safety factors are very close to each other for two methods

Keywords:

Critical failure surface , Factor of safety , Slope stability , Particle swarm optimization , Genetic algorithm