This paper aims to explore the influence of fixed point injection on residual stress field of GCr15 bearing steel under different process parameters of strengthening grinding. Image processing technology was used to analyze the distribution characteristics of the surface coverage along the width direction of the strengthening grinding fixed point injection under different process parameters. The two-dimensional normal distribution function was used to describe the distribution characteristics of steel ball coordinates under the fixed point injection of strengthening grinding. Python/Opencv was used to determine the number of steel balls required by the finite element model under different process parameters. Based on Abaqus/ Python, the finite element model with normal distribution of strengthening grinding was constructed. Using the established normal distribution model, the influence of different injection velocity, ball diameter and coverage rate on the residual stress field was analyzed. When the injection velocity increased from 45 m/s to 70 m/s, the surface residual compressive stress increased from ?683.5 MPa to ?902.4 MPa, the maximum residual compressive stress increased from ?981.6 MPa to ?1330.6 MPa, and the thickness of residual pressure layer increased from 89 μm to 151 μm. The maximum residual compressive stress depth was moved from 30 μm to 70 μm. When the diameter of the steel ball increased from 0.4 mm to 1.0 mm, the surface residual compressive stress increased first and then decreased, the maximum residual compressive stress increased from ?1063.5 MPa to ?1240.7 MPa, and the maximum residual compressive stress depth increased from 30 μm to 60 μm. The thickness of residual compressive stress layer increased from 103 μm to 147 μm, and the diameter of steel ball increased from 0.8 mm to 1.0 mm, and the maximum residual compressive stress almost remained unchanged. When the injection coverage rate was from 100% to 300%, the surface residual compressive stress, the maximum residual compressive stress and the maximum residual compressive stress depth increased slightly, and the residual compressive stress layer thickness almost remained unchanged. The simulation values of the normal distribution model and the random distribution model were compared with the experimental values, and it was found that the surface residual compressive stress, maximum residual compressive stress depth and residual compressive stress thickness of the three models were almost the same. The simulation values of the random distribution model and the normal distribution model were 32.1% and 18.9% higher than the experimental values of the maximum residual compressive stress. The finite element model with normal distribution can accurately predict the change process of residual stress, which can provide some guidance for the optimization of process parameters of strengthening grinding.