Computational simulation of explosively generated pulsed power devices

摘要

Technology and size constraints have limited the development of the end game mechanisms of today's modern military weapons. A smaller, more efficient means of powering these devices is needed, and explosive pulsed power devices could be that answer. Potential advancement opportunities exist with the growing field of research that surrounds explosive pulsed power devices. While most of the research to date has been in the experimental field, if these devices are going to be a viable option for use in future weapon development, there is a genuine need for more theory-based research and an accurate computer modeling capability. One of the programs that has done much experimental work with ferroelectric generators (FEG) is the US Army at Redstone Arsenal in Huntsville, Alabama. The objective of this research was to use the Redstone experimental data collected from an FEG of their own design in combination with the ALEGRA-EMMA code, a hydrodynamic code developed by Sandia National Laboratories, to develop a computer model that can accurately represent an FEG and that can be verified against existing experimental data and eventually used to predict future experiments. Three experimental scenarios were used from the existing collected data: an FEG wired into an open circuit, an FEG wired into an 8-blasting cap circuit, and an FEG wired into a 64-blasting cap circuit. The three areas of this research that had to be explored simultaneously were developing an accurate model for the ferroelectric material, developing an accurate model to represent the external circuit load, and recreating the Redstone FEG design in the ALEGRA computer environment. Once these three aspects were covered and the overall model was developed, the individual cases for each scenario were run in the simulation model. The simulation results were compared to the respective experimental data, both current and voltage, and the model was evaluated. While the ALEGRA code is not capable of simulating the b- eakdown phenomenon seen in the open circuit cases, the model can accurately reproduce the peak values for the current but has problems reproducing the peak values for the voltage for both the 8-blasting cap and 64-blasting cap scenarios. The model also fairly accurately reproduces the general shape of the current and voltage data in both scenarios as well, though the time scale of the simulation reaction is slightly shortened from the time scale seen in the experimental data. Overall, the developed model provides a good baseline simulation capability that can be used as a springboard for future development with further research.