Abstract: In the case of weak earthquakes, a back analysis model considering interaction and radiation damping effects is proposed for dynamic parameters of earth-rock dams. In the model, the viscous-elastic artificial boundary combined with transfer function is used for achieving the wave motion input of valley topography, and the multiple output support vector machine (MSVM) and particle swarm optimization (PSO) are adopted. Using PSO to optimize the dynamic parameters of the dam can effectively get the optimal combination of parameters. The trained MSVM used to describe the mapping relationship between the model parameters and the acceleration reduces computing time of parametric inversion, avoiding the problem that the traditional single support vector machines easily ignore the correlation between monitoring data. The inversion results of Nuzhadu dam show that the calculated settlements agree well with the measured data. The proposed model is feasible in back-analysis of dynamic parameters of earth-rock dams. The study shows that the dynamic shear modulus coefficient C of the dam materials obtained from the indoor dynamic triaxial tests is small and should be revised.