Relative density is one of the important parameters to evaluate the performance of additive manufacturing components, but the corresponding process parameters of high relative density components need a large number of experimental studies. This paper has established a prediction model of the relative density based on the process parameters, which can effectively predict the pores caused by lack of fusion, and provide more reference for the selection and optimization of laser selective melting process parameters. The finite element analysis and numerical calculation software were used to obtain the sizes of the melting pool and simulate the splicing morphology of the melting pool with multiple layers and traces, and predict the corresponding density of the process parameters.At the same time, the fluctuation coefficient, the deflection Angle and the angle of scan vectors rotation between layers are introduced into the prediction model of melting pool to account for the influence of the size fluctuation of the melting pool, the inclination of the melting pool, and the angle of scan vectors rotation between layers on the simulation results of the molten pool. Finally, the feasibility of the prediction model has been confirmed by the selective laser melting experiments of HR-2 stainless steel. The results suggest that the predicted results of the model are in good agreement with the experimental results of the melting pool of the samples, and the deviation between the predicted results of the relative density and the experimental results is within 2%.