Predicting the terminal ballistics of kinetic energy projectiles using artificial neural networks
| dc.contributor.advisor | Hammell, Robert J., II | |
| dc.contributor.advisor | Lu, Chao | |
| dc.contributor.author | Auten, John Robert, Sr. | |
| dc.contributor.department | Towson University. Department of Computer and Information Sciences | en_US |
| dc.date.accessioned | 2017-04-10T20:25:58Z | |
| dc.date.available | 2017-04-10T20:25:58Z | |
| dc.date.issued | 2017-04-10 | |
| dc.date.submitted | 2016-12 | |
| dc.description | (D. Sc.) -- Towson University, 2016 | en_US |
| dc.description.abstract | The U.S. Army requires the evaluation of new weapon and vehicle systems through the use of experimental testing and Modeling & Simulation (M&S). Traditional M&S has worked well over the years but can be a lengthy process and often cannot provide quick results for studies involving new threats encountered in theater. So, there is increased focus on rapid M&S efforts that can provide accurate and fast results . Accurately modeling the penetration and residual properties of a ballistic threat as it progresses through a target is an extremely important part of determining the effectiveness of the threat against that target . This dissertation presents research on the application of Artificial Neural Networks (ANNs) to the prediction of the terminal ballistics of Kinetic Energy Projectiles (KEPs). By shifting the computational complexity of the problem to the fitting (regression) phase of the methodology, performance during analyses are improved when compared to other terminal ballistic models for KEPs. Another improvement in performance can be realized by removing the need for input preparation by a Subject Matter Expert (SME) prior to using the methodology for an analysis . This research shows that ANNs can be used to model the terminal ballistics of KEPs and that they are capable of being used for single element and multiple element targets. It is also shown that the runtimes of an ANN are drastically faster than the current state-of-the-art model. | en_US |
| dc.description.uri | http://library.towson.edu/cdm/ref/collection/etd/id/55908 | en_US |
| dc.format | application/pdf | |
| dc.format.extent | xii, 260 pages | en_US |
| dc.genre | dissertations | en_US |
| dc.identifier | doi:10.13016/M2HR99 | |
| dc.identifier.other | DF2016Auten | |
| dc.identifier.uri | http://hdl.handle.net/11603/3867 | |
| dc.language.iso | en_US | en_US |
| dc.title | Predicting the terminal ballistics of kinetic energy projectiles using artificial neural networks | en_US |
| dc.type | Text | en_US |