Design of Blast Resistant Panels
During my undergraduate studies I performed research into the design and optimization of blast-resistant panels (BRPs). This work arose from a problem posed by the Army Research Laboratory through the participation of the Georgia Tech’s Systems Realization Laboratory in the Penn State – Georgia Tech I/UCRC for Computational Materials Design. I was advised by Dr Janet Allen of the Systems Realization Lab. This research also fed into the larger study on managing complexity in embodiment design processes being performed by the lab.
Here I describe the research project and methodology.
Blast resistant panels are sandwich structures that consist of two outer face sheets bonded by a core structure. They are specially designed to dissipate the forces generated by an explosion through plastic deformation. They absorb large amounts of energy per unit mass compared to solid plates through core crushing. They also deflect less under the same impulse loading compared to a solid plate of equal mass. One application of BRPs is that they can be attached on the outside of military vehicles to protect them from landmines.
The forces impacting on a BRP will be unevenly distributed over its surface. Also these impulse forces will not be of constant magnitude, but will vary slightly depending on the radial distance of each point on the surface of the BRP from the explosive. In addition, the impact of the blast will not be felt over the entire BRP surface at the same time. Overall there is a lot of uncertainty associated with the manner in which a BRP will be loaded.
The primary goal of this study was to design BRPs that will be robust to changes in loading conditions and have a relatively consistent performance in a changing environment.
The secondary goal was to demonstrate a systematic strategy to embodiment design-process generation and selection (design embodiment is the design phase which takes a product from the conceptual stage to the detail design stage) as part of a larger study on how to manage complexities in a design process. This culminated in a paper "Designing Embodiment Design Processes for Blast Resistant Panels" , which is attached.
Core Shape, Thickness, & Materials
The BRP design used was that of a sandwich structure, consisting of two flat outer plates with a square honeycomb structure in the middle. The impulse force of the explosion impact on one of the outer panels, and these forces are transmitted to the honeycomb structure, which gets crushed and absorbs the energy of the blast. One of the design goals was to minimize the transmission of energy to the opposite plate.
The square topology was considered for simplicity and the fact that previous simulations predicted that square topology is very effective at absorbing energy.
Changes in the thickness of the BRP walls, especially the honeycomb core, change the way the core crushes and absorbs energy. Some designs crush more to absorb the energy, whereas others undergo larger shear to absorb and dissipate the energy. One of the design goals was to find the one that performs better, or a balance between them, so that energy can be absorbed most efficiently, and with least overall deflection of the blast resistant panel.
Changes in the materials used in a BRP also affect the performance of the BRP. Some materials are more conducive to crushing, whereas others will withstand greater shear forces. Still others will be deformed to different extents under different loading, based on their intrinsic properties. It is desirable to obtain the best combination of material for use for the plates as well as the honeycomb of the BRP.
1. Modeling assumptions
Only 1/4 of the BRP was modeled taking advantage of symmetry conditions to save on computational cost.
2. Python Scripts
Abaqus provides graphical user interfaces (GUI) known as Abaqus/CAE for pre-processing and Abaqus/Viewer for post-processing. However this research study required the generation and testing of thousands of different models, with different thicknesses, different material properties, and different loadings, and manually creating these in Abaqus/CAE would have been a daunting task.
I wrote a Python script to automate the entire process. The script handled all pre-processing related tasks, such as creating the parts, assigning materials and sections, assembling them, meshing them, creating the analysis steps, assigning loads and boundary conditions, and requesting field and history outputs. The script would then submit the task to the solver, and wait for the analysis to complete. Subsequently the script would perform the post-processing as well, extracting useful data such as maximum deflection and providing it in a tabulated format. The script also provided the ability to fill in certain holes of the core as required, to see if performance improved with certain parts of the core below hollow while other parts were solid.
The script was parameterized so it would accept a number of parameters such as model dimensions, material properties, load amplitudes, and so on.
3. Finite element solver
In order to simulate exposure of BRPs to explosions, I used the Abaqus finite element analysis (FEA) software.
The solver used was the explicit dynamic one (Abaqus/Explicit). For this type of problem, where a BRP is subject to an explosive load that lasts a very short time, it is important that the inertial effects be captured by the solution, therefore a dynamic solver is required. The explicit dynamic solver was preferred to the implicit dynamic one because it is often more efficient for solving extremely discontinuous short-term events or processes, and is also often much more computationally efficient for problems involving stress wave propagation, and even though the increment time is very small this is ok for problems like this one where the total dynamic response time is very short.
Here are some images obtained at the end of a run, showing the core crushing:
4. Design optimization with ModelCenter
I then created a workflow in ModelCenter, which is a tool that helps integrate various products and codes to create a workflow. The script was linked to ModelCenter from which the variables could be changed and the script initiated. The material plasticity data and the load amplitude vs time data were stored in separate spreadsheets and wrappers were created to extract this data from these spreadsheets and introduce them into the workflow (and inject them into the script). The Python script was modified to export the post processing results to a spreadsheet, which too was integrated into the ModelCenter workflow with a wrapper.
The ModelCenter workflow setup is displayed in the following figures
Once the workflow was set up, it was possible to run the simulations repeatedly for various combinations of input parameters. A Design of Experiments (DOE) study was run, manipulating multiple inputs at the same time to determine the effect of each parameter on the output while identifying important interactions between them. A Monte Carlo simulation was also run to find probabilities of different outcomes.
The workflow was also subsequently used by the Systems Realization Laboratory in their study on managing complexity in embodiment design processes, and those results are published in the attached paper.