UNI-MB - logo
UMNIK - logo
 
Fakulteta za gradbeništvo in geodezijo, Ljubljana (FGGLJ)
  • Modelling of the oedometer test by neural networks = Modélisation d'un essai oedométrique avec réseaux de neurones = Modellierung von Oedometerversuchen mit neuronalen Netzen
    Logar, Janko, 1962- ; Turk, Goran
    Constitutive modelling of non-linear material such as soil is a very difficult task. The development of artificial intelligence offers a new possibility in this field. A feed-forward neural network ... used in predicting the oedometer loading curve was trained by 40 oedometer curves obtained from tests made on marshland soil. The neural network was tested on 6 oedometer curves which were not included in the learning process. The oedometer curves and the clay parameter were evaluated. Good agreement between measured and predicted values was obtained, which implies that the neural networks can serve as an effective constitutive model as well as predicting sample behaviour
    Vir: Knowledge support systems in civil engineering : report (Str. 273-281)
    Vrsta gradiva - prispevek na konferenci
    Leto - 1995
    Jezik - angleški
    COBISS.SI-ID - 257121

Vnos na polico