Zhitao Chen, Jiayu Ou, Caixu Yue, Hong Zeng, Xianli Liu, Steven Y. Liang
Residual stress induced by machining is one of the key factors affecting the surface integrity and service performance of aerospace thin-walled components. Its evolution in multi-pass milling is governed by multiple factors, including historical stress accumulation, thermo-mechanical coupling effects, and structural stiffness variations. Existing studies still have limitations in the quantitative analysis of the regeneration mechanism of residual stress and the geometric constraint effect during multi-pass cutting of thin-walled structures. This paper focuses on 7075-T6 aluminum alloy thin-walled components and constructs a numerical prediction model for multi-pass milling residual stress that considers the stress/strain state of previous passes. Through an incremental iterative calculation mechanism coupling mechanical and thermal loads, the evolution laws of stress fields influenced by radial cutting depth allocation (single-pass depth and pass combination) and thin-walled geometric features (unidirectional milling of side walls, bidirectional milling of rib plates) are systematically analyzed. The stress iterative algorithm is adopted to take the residual stress from previous passes as the initial condition for subsequent processing. Combined with the characterization of material's elastoplastic response, the surface stress trend and the transition position of compressive-tensile stress are verified through experiments. The research shows that multi-pass cutting can reduce the amplitude of residual stress through staged stress release, while the stiffness difference of thin-walled structures leads to significant asymmetry in stress distribution. The established model explains the physical mechanism of stress regeneration in multi-pass milling, providing engineering-applicable theoretical support for the optimization of cutting parameters (such as depth allocation and pass planning) and processing deformation control of aerospace thin-walled components.