The article presents the information about the usage of artificial neural networks. The automation process of neural networks of the analysed evaluation data results is highlighted. The kinds of MG type artillery fuses are described and the kinds of cartridges’ calibres, in which they are used, are also specified. The way of preparation of databases of test results to computer simulation is described. Building of neural networks determining the main technical parameters and sizes of learning, test and validation sets is characterized. The summary for chosen active neural networks for individual kinds of the analysed MG type artillery fuses is presented. Graphs of learning, values of sensibility indicators and fragments of prediction sheets for the chosen neural networks were shown.
Neural Networks for Tested MG Artillery Fuses
2019
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
Automotive engineering | 1996
|SAE Technical Papers | 1972
Fuses for spaceflight applications.
NTRS | 1968
|