Nik Petrinic

Prediction of impact damage in composite plates

This paper gives details of the implementation of improved failure criteria for laminated composite structures into LS-DYNA3D. Out-of-plane stresses have been taken into consideration for damage initiation. It is suggested, for the first time, that delamination is constrained by through-thickness compression stress. Interactions between different damage mechanisms have been considered. Damage predictions in good agreement with experimental ones have been achieved.

A delamination criterion for laminated composites under low-velocity impact

Abstract This paper describes details of an improved delamination criterion for laminated composite structures and its implementation to LS-DYNA3D. Out-of-plane stresses have been taken into consideration for damage initiation. The new delamination criterion, verified by experimental results for low-velocity impact, takes into consideration both the interlaminar shear and through-thickness compression stresses. Improved from the criterion proposed previously [Hou JP, Petrinic N, Ruiz C. Prediction of impact damage in composite plates. Composites Science and Technology 2000; 60(2): 273–81], delamination is not fully constrained by the out-of-plane compression. The influences of high local interlaminar shear stress induced by matrix cracking and fibre failure on delamination have also been taken into account. Predicted results from different criteria have been discussed, in comparison with experiment work.

A combined DEM-FEM numerical method for Shot Peening parameter optimisation

A novel combined DEM-FEM method for Shot Peening optimisation is presented.The model provides a rapid method to investigate the effect of the peening parameters.A new algorithm was implemented accounting for impact dependent coefficients of restitution.The minimum area required in the analyses to obtain true residual stresses was determined. A numerical modelling approach capable of simulating Shot Peening (SP) processes of industrial interest was developed by combining the Discrete Element Method (DEM) with the Finite Element Method (FEM).In this approach, shot-shot and shot-target interactions as well as the overall shot flow were simulated efficiently using rigid body dynamics. A new algorithm to dynamically adapt the coefficient of restitution (CoR) for repeated impacts of shots on the same spot was implemented in the DEM code to take into account the effect of material hardening. Then, a parametric study was conducted using the Finite Element Method (FEM) to investigate the influence of the SP parameters on the development of residual stresses.Finally, a two-step coupling method is presented to combine the output of DEM simulation with FEM analyses to retrieve the Compressive Residual Stresses (CRS) after multiple impacts with the aim to evaluate the minimum area required to be modelled to realistically capture the field of residual stresses. A series of such coupled analyses were performed to determine the effect of peening angle and the combination of initial velocity and mass flow rate on CRS.

Cross-section reconstruction during uniaxial loading

The inelastic response of materials to applied uniaxial loading is typically measured using tensile or compressive specimens of an initially circular cross-section. Under deformation, this cross-section may become elliptical due to anisotropic material behaviour. An optical technique for measuring the elliptical deformation of anisotropic, homogeneous cylindrical specimens undergoing uniaxial deformation is presented. It enables the quantification of anisotropic deformation in situ and provides data for material characterization. Three or more silhouette views of a specimen are obtained using multiple cameras or mirrored views. The positions of the edges are computed using a sub-pixel edge detection method, and 3D tangent rays from the camera through these positions are calculated. These bounding tangents are used as the basis for an elliptical fit by least squares at cross-sections along the length of the specimen. Stochastic error estimates are performed by simulation of the experiment. Error estimates, for the experimental set-up used, are also calculated by reconstructing elliptical prisms of precisely measured dimensions. Example reconstructions from specimens of rolled titanium deformed plastically in tension at quasi-static (7 × 10-4 s-1) and high strain rates (3 × 103 s-1) are presented.

The Response of Rigid Plates to Deep Water Blast: Analytical Models and Finite Element Predictions

One-dimensional analytical models and finite element calculations are employed to predict the response of a rigid plate, supported by a linear spring, to loading by a planar pressure shock wave traveling in water or in a similar inviscid liquid. Two problems are considered: (i) a spring-supported rigid plate in contact with fluid on one side and (ii) a spring-supported rigid plate in contact with fluid on both sides; for both problems, plates are loaded by an exponentially decaying shock wave from one side. Cavitation phenomena in the liquid, as well as the effect of the initial static fluid pressure, are explicitly included in the analytical models and their predictions are found to be in excellent agreement with those from FE calculations. The validated analytical models are used to determine the sensitivity of the plate’s response to mass, spring stiffness and initial static pressure. [DOI: 10.1115/1.4006458]

The Strain Rate Dependent Material Behavior of S-GFRP Extracted from GLARE

S-Glass fiber reinforced epoxy (S-GFRP) extracted from GLARE has been experimentally characterized at three distinct strain-rates (5 × 10−4 s−1, 10 s−1, and c.a. 2000 s−1) and in four loading directions (0°, 90°, and 45° to the fiber direction and in the through-thickness direction). A novel specimen clamping mechanism was developed and full-field optical strain measurement was applied. With the aid of these techniques, a significant increase in failure-strength and apparent elastic modulus in all loading directions, particularly in fiber direction, was observed with an increased strain-rate; strain to failure increased in the fiber direction and decreased in all other loading directions.

An Error Analysis into the Use of Regular Targets and Target Detection in Image Analysis for Impact Engineering

This study concentrates on the use of corners targets for photogrammetry in impact engineering. An example of high speed experimentation is presented and the associated difficulties are discussed. The relevant corner detection methods that have been implemented and developed are investigated and their accuracy assessed. This study focuses solely upon the effect of blurring on the accuracy of the detection methods; it is part of a much wider investigation into the use and accuracy of different targets and target detection methods for photogrammetry in impact engineering. A set of tests has been performed and the errors between the true position of the corner and the detected position are compared.

Improving the GJK Algorithm for Faster and More Reliable Distance Queries Between Convex Objects

This article presents a new version of the Gilbert-Johnson-Keerthi (GJK) algorithm that circumvents the shortcomings introduced by degenerate geometries. The original Johnson algorithm and Backup procedure are replaced by a distance subalgorithm that is faster and accurate to machine precision, thus guiding the GJK algorithm toward a shorter search path in less computing time. Numerical tests demonstrate that this effectively is a more robust procedure. In particular, when the objects are found in contact, the newly proposed subalgorithm runs from 15% to 30% times faster than the original one. The improved performance has a significant impact on various applications, such as real-time simulations and collision avoidance systems. Altogether, the main contributions made to the GJK algorithm are faster convergence rate and reduced computational time. These improvements may be easily added into existing implementations; furthermore, engineering applications that require solutions of distance queries to machine precision can now be tackled using the GJK algorithm.

Increasing data from high rate characterization experiments using optical reconstruction

Uniaxial characterization experiments in tension and compression are widely used to evaluate the mechanical response of materials to applied deformation over a wide range of strain rates. By taking photographs of the specimen as it deforms, it is possible to more fully characterize materials by reconstructing the specimen shape during the deformation process. This allows us to better evaluate the stress and strain in the specimen during deformation, and also provides a more rigorous validation of material models. For many materials, specimens with initially circular cross sections will evolve to a different shape; in particular, hcp metals become elliptical. We have developed a technique whereby images of a specimen during deformation (for example, from a high speed camera) from three different angles are used to reconstruct elliptical cross sections during an experiment. The technique has been applied to tensile Hopkinson bar experiments and Taylor Impact experiments on Zirconium. This paper presents an ...

Optical surface profile tracking for high-resolution strain measurement

Characterization of the mechanical behaviour of materials often requires accurate determination of specimen strain. This paper presents an optical method whereby the surface displacements of a cylindrical specimen undergoing uniaxial loading are measured by tracking the profile of its surface texture. These profiles are obtained through sub-pixel edge detection from a digital image. High-resolution digital imaging with sub-pixel edge detection is used to measure the radial position of edges in images of the specimen. Edges are compared in consecutive images to track the positions of zero-crossings and local maxima of the surface roughness. A method for dealing with cases in which the mean wavelength of the surface texture is shorter than the displacement between frames is presented. False matches can still occur and are removed by modal average proximity filtering in sub-regions spanning short axial-lengths and several frames. The new method of finding specimen displacement is found to yield high data densities (in our case 32 independent measurements per mm). The data compare favourably with measurements made using a laser extensometer: surface roughness tracking produces a drop in accuracy of ±25 μm whilst increasing the spatial data density. A further advantage of the technique is that no specimen preparation is required. The data are also used to calculate strain, values of which are validated against strains inferred from the change in cross-sectional area.