A CAE based procedure to predict the low velocity impact response of a composite CAI specimen

Rosario Borrelli, S. Franchitti, F. Di Caprio, U. Mercurio - CIRA
Vito Primavera, Marco Perillo - EnginSoft

The widespread use of Composite Fiber Reinforced Plastics (CFRP) laminates in many industrial fields and especially in the aerospace industry, introduces a large number of issues regarding their high vulnerability to low velocity impact damages. Compression After Impact (CAI) tests are usually carried out on rectangular plate specimen to measure the damage resistance of composite materials to a drop-weight impact. Indeed, a low velocity impact event can significantly reduce the residual strength of a composite structural since several kinds of internal damages can be generated.
There are different types of damages interesting composite structure under low velocity impact conditions. They are usually classified in two categories: intralaminar damages (fibers breaking, matrix cracking and debonding) and interlaminar damages (delaminations). The damage mechanisms can evolve independently, or may interact with each other, making very complex the numerical prediction of their onset and evolution.
In this work, a CAE-based procedure aimed at simulating the CAI test on a CFRP laminate and at  assessing the residual damage distribution, is proposed. Firstly, a LS-DYNA FE model (Figure 1) was developed to simulate the impact test. The intralaminar and interlaminar damages were taken into account by using the MAT_54 material model and several tiebreak contact definitions, respectively. The damage distribution envelope, resulting from the impact analysis and summing the intralaminar contribution to the interlaminar one, was computed by a macro written by using the Ansys Parametric Design Language (APDL).
The LS-DYNA FE model was then coupled with modeFRONTIER which is a process integration and design optimization tool exploring the design space (i.e. the free parameters dominions) and finding the configurations which satisfy several objective functions. Indeed, setting some numerical LS-DYNA parameters (degradation factors, damping coefficient, interlaminar strength, etc.) is a challenge since no reliable data is available in literature  and they are usually chosen on the basis of the analyst’s experience. The LS-DYNA – modeFRONTIER integrated procedure developed in this work, allowed to obtain both a better understanding of the influence of such parameters on the simulation results (sensitivity analysis) and the configuration which provided the best agreement with the experimental data (optimization).
An experimental test campaign, according to the ASTM D7136 ( American Standard Test Method for Measuring the Damage Resistance of a Fiber –Reinforced Polymer Matrix Composite to a Drop-Weight Impact ) regulations, was performed in order to provide experimental data to validate the proposed integrated procedure. The standard CAI specimen was impacted at 50J by using an hemispherical steel impactor and the following experimental data were used for validation purpose:

  • Contact force vs. time curve
  • Deflection vs. time curve
  • Absorbed Energy
  • Damaged area (provided by ultrasonic C-Scan)

For each of these quantities, an objective function to be minimized (4 relative errors) was taken into account by the modeFRONTIER optimization algorithm. The integrated procedure allowed to find the best setting of the free parameters with respect to the above mentioned objective functions. The best design was found in excellent agreement with the experimental data.