Abstract: To enhance the machining quality of thin-walled aerospace aluminum alloy components, reduce machining deformation, and minimize surface defects, this paper investigates the distribution characteristics of the temperature field inside the workpiece during milling and summarizes the influence of workpiece temperature on surface defects. The milling process is simulated using finite-element analysis software to obtain the instantaneous deformation in the material removal area and identify regions with severe elastic deformation of the workpiece. Based on this, to achieve accurate machining deformation, which is essential for deformation error compensation, an IPO-CNN-based regression prediction model for workpiece elastic deformation is established. This model is trained using simulation experimental data, and its accuracy is evaluated. The NSGA-II algorithm is employed to optimize the machining parameters with the aim of improving workpiece shape accuracy and machining efficiency while reducing surface defects. The optimal parameters are then input into the regression prediction network to obtain the elastic deformation values at different locations, and compensation machining is carried out accordingly. Experimental results demonstrate that the proposed method of optimal parameter selection and compensation machining can effectively reduce workpiece deformation while ensuring surface integrity and machining efficiency.