Back Analysis Algorithm Based on Particle Swarm Optimization

Abstract

In geotechnical engineering, back analyses are generally used to investigate uncertain parameters. Back analyses can be done by considering known conditions, such as failure surfaces, displacement, and structure performance. In fact, a complex nonlinear optimization function is typically required for most geotechnical problems in order to gain better understandings of these uncertainties. Therefore, Particle Swarm Optimization is utilized in this study to assist in back analyses based on finite element upper bound limit analysis method. This technique is an evolutionary computational approach, which is suitable for solving nonlinear global optimization problems. By using Particle Swarm Optimization with numerical upper bound method, a simple case study showed that the obtained results are reasonable and reliable. The usability of the proposed method is demonstrated. It could be seen as a new potential approach for geotechnical back calculations.