| dc.description.abstract | In this project, IMPETUS Afea Solver was used to simulate hypervelocity impact on Whipple shields. The micrometeoroid and orbital debris environment around the International Space Station was studied in order to understand what spacecraft shielding must withstand. Typical impact angles and velocities were determined and the size and origin of orbital debris studied. Simple spacecraft shielding was looked at and hypervelocity impacts between aluminum spheres and plates and the resulting debris cloud studied.
IMPETUS Afea Solver combines the finite element method with a discrete particle module to capture the transition from a solid projectile to particles. The goal of this project was to investigate the possibilities of this method. A model was developed and a sensitivity study performed in order to investigate how the numerical model responds to changes in input parameters. The model was sensitive to changes in plate thickness, impact velocity, material density and equation of state. The numerical results were compared to physical experiments when available and it was shown that the trends and general behaviour were similar. The shape and size of the debris cloud, however, did differ. The numerical model produced a larger debris cloud than the experimental results showed, and there were no clear borders between a front, center and rear element. The projectile fragmented for lower velocities for the numerical model compared to the physical experiments.
As this is a new technique for modeling hypervelocity impact, one must expect some weaknesses in the code. The system lost a large part of its energy when the solid elements were converted into particles. The problem was presented to the program developers and some necessary changes in the software code were performed. The updated solver gave a much improved energy balance, and the debris cloud formed after impact with a plate was much more powerful than for the old solver. The debris cloud was, however, larger after the update was performed, making it far larger than the experimental results showed.
Lastly, ballistic limit curves for a monolithic and a Whipple shield were produced numerically and compared to existing data. The numerical model for the Whipple shield captured the trends of the existing curve, but over predicted the ballistic limit for all velocities. A possible explanation for this may be that the projectile is eroded too early, and when each element reaches the failure criterion of the material, it is converted straight into particles instead of fracturing to more powerful smaller pieces. A suggestion to how this fragmentation process may be captured is presented at the end of this thesis. | en |
