Akademska digitalna zbirka SLovenije - logo
VSE knjižnice (vzajemna bibliografsko-kataložna baza podatkov COBIB.SI)
  • Prediction of tool condition by applying family of artificial neural networks
    Spaić, Obrad ; Krivokapić, Zdravko, strojnik ; Soković, Mirko
    Based on the previously established correlation on the real object of research, conducted between the axial cutting force, representing a function of the objective and influential parameters for ... drilling steel of high hardness and strength (tempered steel), this paper has conducted prediction of tool condition by applying a family of artificial neural networks. In establishing the correlation between a function of the objective and influential parameters (performance of the experiment), the following parameters were varied on three levels: nominal diameter, number of turns and feed length and drilling until twist drills become blunt. In addition to control neural networks, results of prediction, obtained by applying family of artificial neural networks, have been compared in the stages and points of the experiment with the results of the experiment. The previously named checks have confirmed that the family of neural networks can be applied, as a very reliable method for predicting tool condition, depending on the influential factors and duration of drilling and tools (twist drills) blunting.
    Vir: Metalurgia internaţional. - ISSN 1582-2214 (Vol. 18, no. 6, 2013, str. 87-93)
    Vrsta gradiva - članek, sestavni del
    Leto - 2013
    Jezik - angleški
    COBISS.SI-ID - 13952283

vir: Metalurgia internaţional. - ISSN 1582-2214 (Vol. 18, no. 6, 2013, str. 87-93)
loading ...
loading ...
loading ...

Vnos na polico