Design Optimization of Composite Panels in Aerospace Applications

By

To meet certification requirements set up by federal aviation regulatory bodies such as the Federal Aviation Administration (FAA) and the European Aviation Safety Agency (EASA), aircrafts need to be designed for damage tolerance under different categories of potential damages such as Bird Strike, Hail Damage, Blade-off and Barely Visible Impact Damage (BVID), etc. Physical testing for various combinations of composite ply lay-ups, geometries, material variability and damage scenarios is cost prohibitive and unsustainable. SIMULIA has partnered with an aircraft OEM to address the design and optimization of composites skin stringer panels under BVID.

We asked Deepak Goyal, SIMULIA Senior Technical Specialist, who developed this workflow, to provide answers to some common questions.

Q. Why does this workflow need to be analyzed?

A. In the past two decades, air traffic growth has been impressive and is expected to strongly increase in the coming years. Today, composite materials are being increasingly deployed in primary structures of large commercial aircrafts, small business jets and military airplanes. The increased use of composite materials is due to their superior strength to weight ratio resulting in potentially lower operational costs, as compared to the metal materials traditionally used, and without compromising flight worthiness. However, multiphase material complexity, anisotropy, damage initiation and progression under different modes make it harder to predict the behavior of composites. The automated workflow being developed can potentially save millions of dollars in testing, engineering time and can also be extended to address other types of aircraft structural design needs.

Q. Describe the workflow.

A. A tool drop on the primary composites structures (skin-stringer panels) of aircrafts is a common situation and causes BVID. The load carrying capacity of the structure could decrease and needs to be measured and predicted. A composite model of skin-stringer panel was built. Hat shape stringers were used as they are most common in aircraft structural parts. The model contained Hashin’s damage initiation and progression criteria with calibrated material properties. First a compression load was applied to an Intact model. Intact model has no BVID. Load carrying capacity of this intact model was predicted. Then a tool drop from certain height was simulated to cause BVID. Now a compression load was applied to this damaged model. Load carrying capacity of this damaged model was predicted and compared with the Intact model. Then Abaqus simulation results were compared with experimental data. Comparisons were made on two panel types: Thin and Thick Panels. The only difference between the panels was in terms of number of plies. Different energies were imparted on thin and thick panels. Finally, results are compared with original experimental data obtained from an aircraft OEM partner. Abaqus simulation results were compared with thick and thin panels. A very good comparison between simulations and experiments validated the analysis approach. An Isight Design Of Experiments (DOE) was used to automate the BVID analysis to find best and worst designs. The primary focus industry is Aerospace and Defense, but the methodology is applicable to any industry concerned with damage in composite structures.

Q. What are the key simulation goals? What are you trying to learn from the simulation?

A. Below are the key simulation goals for this workflow:

  • Load carrying capacity of aerospace skin-stringer structural panels made of composites is compared before and after barely visible impact damage. This information can potentially be used for regulatory certification.
  • Panel buckling loads, failure loads, as well as, panel buckling strains and failure strains, are predicted. This information can be used for efficient designs.
  • Whether BVID also shifts the panel failure modes from symmetric to unsymmetric is also investigated.
  • Both thick and thin panels are simulated to understand if there are any differences in their failure behaviors. This can help designers to rule out some panel thicknesses from design considerations.
  • A design of experiments approach using simulations was used to identify the best and worst designs.

Q. Which SIMULIA solutions did you use?

A. Abaqus/CAE was used to build models and easily parameterize them for subsequent Design of Experiments (DOE) studies. Abaqus/Explicit was used for analyzing BVID caused by a tool drop and comparing the collapse loads before and after BVID. Abaqus is ideally suited for such problems because the following features are needed:

  • Non-linear solver.
  • Advanced composite modeling capabilities.
  • Contact modeling.
  • Damage initiation and progression modeling capabilities.
  • Impact damage modeling with an explicit solver.
  • Ability to easily parameterize geometry for Design of Experiments.

Isight was used to calibrate certain material properties. For example, for Hashin’s damage evolution law, energy release rates for matrix damage under tension, such as Gmt, is hard to test directly. It is an intralaminar property rather than an interlaminar fracture toughness. The Gmt also has an in-situ effect and was calibrated using Isight. Isight was also exploited to conduct a DOE to show how the BVID analysis can be automated to find the best and worst designs.

Q. What were the advantages of using simulation?

A. Below are the key advantages of using the simulation:

  • Aerospace manufactures need FAA Certification. Many times, when they design to get certification for Barely Visible Impact Damage (BVID), they have to use empirical data from similar structures if they have that data. Otherwise, they use open-hole certification to account for the effect of BVID. This is typically too conservative a design, since they end up with higher allowables with open hole analysis. So, a long term goal is to get away from this conservative design approach.
  • A short term goal is to simulate the tests to have a higher degree of confidence in certification article testing, which can have multiple impact locations and more complex loading.
  • A design of experiments approach using simulations can identify better designs.
  • Non-intuitive designs can be found with Design of Experiments and Optimization studies.
  • New materials can be rapidly incorporated when suppliers change the materials.
  • Physical testing for various combinations of composite ply lay-ups, geometries, material variability and damage scenarios is cost prohibitive and unsustainable. The automated workflow being developed can potentially save millions of dollars in testing, engineering time and can also be extended to address other types of aircraft structural design needs.
  • As of today, the proposed workflow approach in not being exploited by many Aerospace manufacturers. So, this workflow is a powerful addition to their simulation toolkit.
0

Deepak Goyal

Dr. Goyal is a Senior Technical Specialist, currently playing the role of simulation consultant in SIMULIA's Aerospace & Defense Industry growth team. He received his M.S. & Ph.D. degrees in Aerospace Engineering at Texas A&M University in August 2003 & December 2007 respectively. He is a recipient of American Institute of Aeronautics and Astronautics (AIAA) open topic graduate award in the year 2006. In academia, he has conducted extensive FAA, NASA & AFOSR funded research in tape laminated composites, textile composites & materials with complex microstructures. Since January 2008, he has been working with Dassault Systèmes SIMULIA Corp. where he has provided training & consulting services to SIMULIA’ s customers in various industries. Dr. Goyal has published several journal papers in highly reputed peer-reviewed international journals and has also reviewed scholarly articles of other researchers for leading journals in the area of Composites Materials.