目的 探究喷丸对铝合金薄壁件残余应力场的影响和应力释放后尺寸稳定性的影响。方法 基于ABAQUS有限元软件建立多弹丸随机撞击薄壁件模型和喷丸薄壁件残余应力释放模型，并对模型预测7075铝合金薄壁件残余应力场的准确性进行验证。通过FEM-DEM耦合薄壁件喷丸+残余应力释放一体化模型，探究喷丸速度、喷丸直径、喷丸质量流工艺参数对7075铝合金薄壁件残余应力场和尺寸稳定性的影响规律。结果 沿层深分布的残余应力试验值和仿真值最大误差在±16%以内，平均误差为6%，证实了模型良好的残余应力预测准确性。随弹丸速度增大，铝合金薄壁件表面残余应力和约束释放后的变形增大，速度为100 m/s时最大残余压应力约为?323 MPa。随弹丸直径增大，表面残余应力缓慢增大，残余压应力层深增加，弹丸直径为0.8 mm时残余压应力层深约为0.42 mm。随弹丸质量流增大，表面残余应力和约束释放后的形变也会小幅增加。增大弹丸质量流对增大最大残余应力和约束释放后的变形的增大效率小于增大弹丸速度和弹丸直径。根据仿真结果建立了喷丸效能评估参数和喷丸工艺参数相联系的预测模型。结论 研究验证了多弹丸随机撞击薄壁件模型和喷丸薄壁件应力释放模型具有良好的预测喷丸薄壁件残余应力场和约束释放后变形的能力，为铝合金薄壁件喷丸强化工艺和成形工艺提供了一种数字化研究思路和实际数据支撑。

In order to investigate the effect of shot peening on the residual stress field of aluminum alloy thin-walled parts and the dimensional stability after constraint release, a multi-shots random impact model of thin-walled part and a residual stress release model of peened thin-walled part were established based on ABAQUS finite element software. The accuracy of multi-shots random impact model of thin-walled part in predicting the residual stress field of 7075 aluminum alloy thin-walled parts was verified. The effect of shot velocity, shot diameter, and shot mass flow rate on the residual stress field and dimensional stability of 7075 aluminum alloy thin-walled parts was investigated through the coupled FEM-DEM shot peening and residual stress release integrated model for thin-walled parts. The maximum error between the experimental and simulated residual stresses distribution in depth was within ±16% and the average error was 6%, which verified the good accuracy of residual stress prediction of this model. As the shot velocity increased, the residual stress on the surface of the aluminum alloy thin-walled parts and the deformation of thin-walled parts after constraint release increased. The maximum compressive residual stress was about ?323 MPa when the shot velocity was 100 m/s. As the shot diameter increased, the surface residual stress increased slowly and the depth of the compressive residual stress layer increased obviously. When the shot diameter was 0.8 mm, the depth of the compressive residual stress layer was about 0.42 mm. When the shot diameter was 1.0 mm, the depth of the compressive residual stress layer was 0.64 mm. As the shot mass flow rate increased, the surface residual stress and the deformation of thin-walled parts after constraint release also increased slightly. However, the increasing efficiency of shot mass flow rate decreased gradually, the shot mass flow rate increased from 1.2 kg/min by 150% to 3 kg/min, the maximum residual stress increased by only 9.7%, and the displacement in the Z direction increased by only 13.5% after constraint release. The increment efficiency of increasing the shot mass flow rate on the maximum residual stress and deformation increment after constraint release was lower than that of increasing the shot velocity and shot diameter. Specifically, the shot velocity had the highest efficiency in increasing the values of compressive residual stress and displacement change in Z direction, followed by the increase in shot diameter, and the lowest efficiency in increasing the shot mass flow rate. Generally speaking, with the uniform increase of the three process parameters which are shot velocity, shot diameter and shot mass flow rate, the increasing efficiency of the compressive residual stress caused by shot peening and the displacement change in the Z direction decreases after constraint release. Based on the simulation results, a prediction model linking shot peening effectiveness evaluation parameters and shot peening process parameters is established. A good predictive ability of a multi-shots random impact thin-walled part model and a residual stress release model of peened thin-walled part on the residual stress field before constraint release and the deformation of thin-walled parts after constraint release has been verified. This research provides a digital research approach and practical data support for shot peening strengthening and forming processes of aluminum alloy thin-walled parts.