Finite element modelling of shot peening and peen forming processes and characterisation of peened AA2024-T351 aluminium alloy

The main purpose of this thesis was to develop and validate finite element (FE) simulation tools for shot peening and peen forming. The specific aim was to achieve quantitatively accurate predictions for both processes and demonstrate the potential of reliable FE modelling for scientific investigation and industrial applications.;First, an improved dynamic impact model that takes into account the stochastic nature of shot peening was proposed by carefully studying its dimensions, introducing a dispersion of shot sizes and significantly reducing its computational cost. In addition, cyclic mechanical testing was conducted to define a suitable material constitutive theory for aluminium alloy (AA) 2024-T3/T351 subjected to shot peening. By combining a realistic shot peening model with an appropriate material law, fairly good residual stress predictions were achieved for three different sets of shot peening parameters.;Second, an experimental and numerical characterization of AA2024-T351 shot peened with parameters representative of fatigue life improvement applications was conducted. Multiple techniques, such as micro-indentation, residual stress determination and electron backscatter diffraction, were combined to gain a better understanding of the influence of shot peening on the material. The potential uses of finite element simulation to complement experimental data were also studied. The material heterogeneity arising from the random impact sequence was investigated and it was found that the impact modelling methodology could provide useful information on such heterogeneities.;Third, a novel peen forming simulation methodology was introduced. The impact model provided the necessary input data as part of a multiscale approach. Numerically calculated unbalanced induced stress profiles were input into shell elements and the deformed shape after peen forming was computed as a springback analysis. In addition, a simple interpolation method was proposed to model the incremental nature of peen forming in a computationally efficient manner. The process was therefore simulated as a series of springback analyses. This approach was first validated using data from small-scale experimental trials.;The potential effect of sheet orthotropy was then investigated numerically and experimentally. This factor could have a significant influence in industrial applications since peen formed components usually originate from rolled sheets and plates. The orientation of the rolling direction was found to have a significant effect on resulting curvatures for small AA2024-T3 sheets. The experimentally determined orthotropic elastic properties and initial stress state of the samples were input into forming simulations and numerical results correlated well with small-scale experimental data for one of the two sets of peening parameters under study.;The modelling methodology was improved further so as to take into account the trajectory of the peening nozzle. This led to a more realistic representation of actual peen forming procedures used for large components, which require moving peening equipment over the surfaces. The peening trajectory and boundary conditions considered in small-scale tests led to complex distributions of radius of curvature and FE simulations correctly predicted the experimentally observed trends.;Finally, the potential applications of the novel simulation strategy were demonstrated by simulating peen forming of typical wing skin panels. The modelled components had realistic features such as variable thicknesses and integral stiffeners and were subjected to multiple representative peen forming treatments using different shot types. (Abstract shortened by UMI.).