An Approach to Studying Surface Quality Based on Automatic Generation of Nonparametric Roughness Models.

Autor: Bez"yazychnyi, V. F., Palamar', I. N., Azikov, N. S., Gagarina, A. I., Nazarenko, V. A.
Zdroj: Journal of Machinery Manufacture & Reliability; Aug2022, Vol. 51 Issue 4, p306-312, 7p
Abstrakt: An approach to solving the problem of surface roughness simulation with the use of different methods for processing critical parts is considered. A nonparametric model of roughness based on a generative adversarial neural network is proposed, and algorithms for its training are developed. Studies on the roughness profile generated based on model signals and real profilograms are performed. The roughness parameters are estimated, and the permissible error of the model is obtained upon providing increased efficiency in the studies on the surface quality. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index
Externí odkaz: