A dynamic finite element simulation of the shot-peening process.

摘要

The process of shot peening has been widely used for more than 70 years to improve fatigue life as well as corrosion and wear resistance in metallic components. The basic mechanism involves repetitive stretching and unloading of the surface by bombarding it with small spherical shot media so that a uniform residual compressive stress is eventually created. Historically the implementation of shot-peering has relied on empiricism for its control and only recently have there been attempts to develop a better analytical understanding of the mechanics of the process. High-strength steel is often the material of choice for certain aircraft components, such as aircraft landing gear, that must be able to withstand infrequent occurrences of very high limit loads in addition to more frequent applications of fatigue loads of much lower amplitude. In order to obtain reasonable fatigue life it is essential that these components be shot-peered with the utmost quality assurance. High-strength steel (300M) represents a somewhat unique material, in this regard, due to its very high yield and ultimate strengths, thereby translating directly into a surface that is extremely hard. The mechanics of shot-peered 300M were investigated analytically using a 2D axisymmetric finite element model to consider material and process variation during a single shot impact. This investigation demonstrated that elasto-plastic representation of the shot particle was essential to achieving accurate results for shot that was nominally either 15% harder or softer than the 300M target. This contrasted previous finite element investigations of the shot-peered high-strength steel that utilized a rigid particle. It was also demonstrated that the effect of process variation on residual stress state response could be well characterized by a linear combination of shot diameter and velocity, target thickness and the presence of friction or lack thereof along with their first-order interactions using a deterministic adaptation of a 24 designed experiment. The experimental design approach also enabled a determination of predominant main effects or interactions to be determined. A relative yield strength parameter was used to combine the hardness tolerance bands for the shot particle and target into a single normal random variable in order to address the effect of material variation on the residual stress state probabilistically.