Publication Date

2010

Document Type

Dissertation

Committee Members

Ramana Grandhi (Advisor), Ravi Penmetsa (Committee Member), Dong Qian (Committee Member), Raghavan Srinivasan (Committee Member), Thaddeus Tarpey (Committee Member)

Degree Name

Doctor of Philosophy (PhD)

Abstract

Advanced mechanical surface enhancement techniques have been used successfully to increase the fatigue life of metallic components. These techniques impart deep compressive residual stresses into the component to counter potentially damage-inducing tensile stresses generated under service loading. Laser Peening (LP) is an advanced mechanical surface enhancement technique used predominantly in the aircraft industry. To reduce costs and make the technique available on a large-scale basis for industrial applications, simulation of the LP process is required. Accurate simulation of the LP process is a challenging task, because the process has many parameters such as laser spot size, pressure profile, and material model that must be precisely determined. In the LP process material is subjected up to strain rates of 106 per second, which is very high compared to conventional strain rates. The importance of an accurate material model increases because the material behaves significantly different at such high strain rates. One of the objectives of this research is to make advancements in the simulation of residual stresses induced by laser peening. Validation of various material models under investigation that could be used in simulation and design is performed. Inverse optimization-based methodology is developed for simulation of residual stresses for materials such as Inconel718. The procedure involves optimizing the model constants for one load case and using the same constants for other load cases. The second aspect of this research is to develop a framework for uncertainty quantification of the residual stress field induced by the LP process by propagation of regression uncertainty. Development methodology includes identification of regression uncertainty as a source of input uncertainty and using the bootstrap method to verify the multivariate normality assumption of the model constant estimates. The propagation of the input uncertainty is performed using Taylor series expansion and sensitivity analysis. A confidence band for the entire residual stress field is obtained and validated using the Monte Carlo analysis.

Page Count

201

Department or Program

Ph.D. in Engineering

Year Degree Awarded

2010


Included in

Engineering Commons

Share

COinS