Optimality search and advanced regulation method of NC milling
Owing to increasing demands and reduce of human impact on milling processes it is necessary that they are regulated and controlled by new regulation methods. In this article neural network method is described and represented on a concrete milling example. Neural network is developed and tested on me...
Shranjeno v:
| Main Authors: | , , , |
|---|---|
| Format: | Software Book Chapter |
| Jezik: | English |
| Teme: | |
| Sorodne knjige/članki: | Vsebovano v:
23rd DAAAM International symposium Conference papers |
| Oznake: |
Označite
Brez oznak, prvi označite!
|
| Izvleček: | Owing to increasing demands and reduce of human impact on milling processes it is necessary that they are regulated and controlled by new regulation methods. In this article neural network method is described and represented on a concrete milling example. Neural network is developed and tested on measured cutting forces which occur in main coordinates. Neural network is formed to predict best cutting parameters and develop new one if necessary. With this method all logical reflection belongs to computer and trained neural network. Those methods reduce human impact and give us better results than standard optimization. During milling process neural network is trained for so long, that relative error is reduced to minimum. Relative error reduction to required values give us better final tolerance results after milling process and help us to increase milling process to higher intelligent level. |
|---|---|
| Opis knjige/članka: | Nasl. pri prispevku. Soavtorji: Veza, I.; Paulic, M. & Irgolic, T. |
| Fizični opis: | Str. 943-946. |
| Bibliografija: | Bibliografija: str. 946. Abstract. |
| ISBN: | 9783901509919 |
| ISSN: | 2304-1382 |