Abstract: Optimization technique was an important factor to displacement back analysis,it influenced the precision and efficiency.A new intelligent displacement back analysis which combined particle swarm optimization with support vector machine was proposed.The approach not only used the excellent characteristics of support vector machine which presented nonlinear relationship and avoided the complex computation,but also used the global optimization performance of particle swarm optimization with quick convergence.The approach was applied to a example,and compared with other approaches.It was shown that this approaches was an excellent algorithm which was scientific and feasible with good convergence.