Modelling of the anodizing process of aluminum using neural networks

@inproceedings{25741,
   author = {Kmec, Ján and Vagaská, Alena and Gombár, Miroslav and Spišák, Emil and Michal, Petr and Badida, Miroslav},
   booktitle = {Proceedings of the 2014 15th International Carpathian Control Conference, ICCC 2014, art.},
   keywords = {anodizing; neural unit; prediction model},
   language = {eng},
   isbn = {978-1-4799-3528-4},
   pages = {629-634},
   title = {Modelling of the anodizing process of aluminum using neural networks},
   year = {2014}
}

TY  - JOUR
ID  - 25741
AU  - Kmec, Ján - Vagaská, Alena - Gombár, Miroslav - Spišák, Emil - Michal, Petr - Badida, Miroslav
PY  - 2014
TI  - Modelling of the anodizing process of aluminum using neural networks
SN  - 9781479935284
KW  - anodizing
KW  - neural unit
KW  - prediction model
N2  - The aim of the research work was to present some possibilities of control and optimization of the technological process of aluminum anodic oxidation using neural networks and Design of Experiments (DoE) in order to evaluate and monitor the influence of the input factors on the resulting AAO (Anodic aluminum oxide) film thickness. Three types of neural units (first order neural unit, second order neural unit, third order neural unit) were used to create the prediction model describing the thickness of the final aluminium oxide layer formed during the process of anodic oxidation of aluminum. The paper also deals with the evaluating of minimal range of training data used for learning process, so the neural unit can produce sufficiently reliable model.
ER  - 

KMEC, Ján, Alena VAGASKÁ, Miroslav GOMBÁR, Emil SPIŠÁK, Petr MICHAL a Miroslav BADIDA. Modelling of the anodizing process of aluminum using neural networks. In \textit{Proceedings of the 2014 15th International Carpathian Control Conference, ICCC 2014, art.}. s.~629-634. ISBN~978-1-4799-3528-4. 2014.