Abstract: In view of the low efficiency of the conventional inversion method for the comprehensive elastic modulus of concrete dams and the limitation of the finite element forward analysis to the grid division,a finite element fast inversion method combined the scaled boundary finite element method (SBFEM) and optimization of least squares support vector machine by particle swarm optimization (PSO-LSSVM) are proposed.Analyzing the mapping relationship between the comprehensive elastic modulus of concrete gravity dam and the horizontal displacement,and on the basis of constructing the inverse equation of the comprehensive elastic modulus of the gravity dam,the scaled boundary finite element model of the gravity dam structure based on the MATLAB is established,and the PSO-LSSVM algorithm is used to learn and train the positive analysis results of the dam structure to obtain its inverse parameters.The case study shows that the method has improved both the accuracy of numerical results and the efficiency of calculation.At the same time,it overcomes the inherent defects of conventional back analysis of dam structure,such as the influence of singular element of finite element,and realizes the rapid identification of equivalent parameters of gravity dam structure performance.It provides a new idea for monitoring the deformation performance of concrete dam.