Laser Peening for Reliable Fatigue Life

Project parameters: validated situations, optimized process parameters, designed for risk qualification, fatigue life estimation Project Parameters

Advanced mechanical surface enhancement techniques have been successfully used to increase the fatigue life of metallic components. These techniques impart compressive residual stresses in the surface and sub-surface regions to counter the potentially damage-inducing tensile stresses generated under service loading. Laser Peening (LP) is an advanced mechanical surface enhancement technique used in aircraft, automobile, and medical industries. To reduce costs and make the technique available on a large scale for industrial applications, simulation of the LP process is needed. Such simulation facilitates process optimization and demonstrates the benefits for wider application. LP can also be used to engineer the residual stress distribution in critical regions of a component to achieve an improved fatigue life.

Current research work includes:

  • validate simulations with experimental data
  • developing novel optimization methodologies to obtain the best design
  • developing risk quantification techniques to obtain a robust design
  • fatigue life experiments and estimations incorporating the LP residual stresses

Material Model Validation for Laser Peening Simulations

Accurate simulation of the Laser Peening (LP) process is challenging because the process has many variables such as constitutive material model parameters and LP process parameters. An integration framework is implemented (shown in Figure 1) for efficient data management between ABAQUS/Explicit and ABAQUS/Standard. Validation metrics are developed to simulate the high strain rate material behavior using plasticity models, including Johnson-Cook and Zerilli-Armstrong. Finite and infinite elements are used to capture the localized effect efficiently. A representative finite element mesh of the axisymmetric model is shown in Figure 2, which has been used to simulate the residual stresses induced by the LP process.

Flow chart: Preprocessing ABAQUS/CAE | Dynamics Solver ABAQUS/Explicit | Transient Data | ABAQUS/Standard | Equilibrium | Post-processing ABAQUS/CAE | Residual Stress
Figure 1. Laser Peening Simulation Procedure

Finite Element Mesh Diagram
Figure 2. Representative Finite Element Mesh of the Coupon


Fatigue Life Estimations and Experiments Incorporating LP Residual Stresses

Bending Coupon diagram
Figure 3. Early four-point bending coupon with fixed
LP shot pattern


The Laser Peening (LP) process induces extremely complicated residual stress fields in the materials to which it is applied. These stress fields can drastically modify a component's fatigue life; and these modifications can be either positive or negative depending on the loading condition seen and the geometry being investigated. Parameters typically considered for LP simulation include peak pressure, pressure pulse mid-span duration, the number of shots seen at a given location, spot shape, and spot size. An elastic-plastic finite element analysis of the LP process is combined with fatigue analysis software (fe-Safe), allowing fatigue life estimations to be made. The estimations of the fatigue life are then used as the 'cost' function or response metric for an optimization routine, while LP parameters are used as the variables. Validation of the fatigue life simulations is provided by a real world replication of the model using the final LP parameters in the FEA. An early model of a four-point bending coupon with a fixed shot pattern can be seen in Figure 3.

Quantification of Multiple Types of Uncertainty in Simulation Model

As engineering structural systems become more complex and computer technology advances, the dependence of structural analysis on multi-physics simulation increases. Simulation models can be generated in many different ways. For instance, they can be generated by altering the representation of the model's geometry or the interaction of components comprising the model. This implies that two or more different simulation models can be assumed to analyze an engineering system. Model form uncertainty—the uncertainty arising from the use of different mathematical model forms—unavoidably accompanies the generation of different simulation models for an engineering system. The degree of model form uncertainty may be considerably large in problems for which the predictions of competing simulation models are significantly different. In this research, an effective and efficient algorithm is developed to incorporate model form, parametric and predictive uncertainties involved in the generation of a simulation model set into the prediction of a system response. The proposed algorithm is shown in Figure 4.

Algorithm: Parametric Uncertainty + Predictive Uncertainty (measured error and model probability) + Model Form Uncertainty (combine model predictions)
Figure 4. Algorithm to incorporate multiple types of uncertainty into response prediction

Fatigue Life Prediction of a Laser Peened Aircraft Lug using Stress-life Approach

Many surface treatments have been developed to improve the fatigue strength of structural components. Treatment techniques induce compressive residual stresses at the surface region of components, and the compressive residual stresses resist initial fatigue fractures. Laser Peening (LP) is one of the most effective surface treatment methods to improve the fatigue strength of metallic components. LP can produce higher magnitude compressive residual stresses than traditional treatments like shot peening. In this research, a fatigue analysis process for a laser peened aircraft lug is proposed using the stress-life approach. The stress-life approach requires three pieces of information: fatigue material property (such as S-N data), fatigue load spectrum, and stress results. To consider residual stress effects induced by LP, a modified Goodman equation is adopted, and S-N data is obtained from experiments on non-peened components. FALSTAFF (Fighter Aircraft Loading STAndard For Fatigue evaluation) is adopted to generate the fatigue load spectrum. To obtain stress results from an aircraft lug, finite element analysis is performed.

Process: Static analysis of pure lug + LP simulation using 3D model | Fatigue Analysis | Predict Fatigue Life
Figure 5. Fatigue life process using S-N approach and LP simulation for a laser peened aircraft lug

Progressive Modeling and Optimization

In this research, parametric investigations of the quarter symmetric 3D model are used to investigate temporal variations of pressure pulse, pressure magnitude, and shot shape and size in the LP process. A novel LP optimization methodology is developed. The LP optimization problem is divided into two parts: single- and multiple-location peening optimization. The single-location peening optimization problems have mixed design variables and multiple optimal solutions. A mixed-variable Niche Particle Swarm Optimization (MNPSO) is developed that incorporates a mixed-variable handling technique and a niching technique to solve the problem. Designing an optimal residual stress profile for multiple-location peening is a challenging task due to the computational cost and the nonlinear behavior of LP. A Progressive Multi-fidelity Optimization Strategy (PMOS) is proposed to solve the problem. The three-stage PMOS combines low- and high- fidelity simulations, respective surrogate models, and a mixed-variable handling strategy. This strategy employs comparatively low computational-intensity models in the first two stages to locate the design space that may contain the optimal solution. The third stage employs high fidelity simulation and surrogate models to determine the optimal solution. The overall objective of this research is to employ finite element simulations and effective optimization techniques to achieve optimal residual stress fields.

Formulate the optimization problem | Eliminate part of infeasible space using symmetric 3D model | Exhaustive search using coarse mesh model | Localize the potential region | Mixed-variable PSO using fine mesh model | Optimal solution
Figure 6. Progressive Optimization Methodology

3640 Colonel Glenn Highway - Dayton, Ohio - 45435
Wright State University biplane logo Copyright Information © 2010 | Accessibility Information
Last updated: Wed. May-26-10, 15:47
Please send comments to: webmaster@wright.edu