Repository logo

Browsing by Subject "permeability prediction"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • thumbnail.journal.alt
    Item type:Thesis, Access status: Restricted ,
    Wykorzystanie sieci neuronowych do interpretacji danych geofizyki otworowej i badań laboratoryjnych
    (Data obrony: 2014-10-03) Laszczak, Mateusz
    Wydział Geologii, Geofizyki i Ochrony Środowiska
    Main goal of this masters’ thesis was using artificial neural networks (ANN) for petrophysical interpretation of well logs and laboratory data. Author used Statistica®11 software to build a tool which can determine permeability in dolomite layers located in NW part of Fore-Sudetic Monocline. Permeability is classified as a fuzzy data and its determining is frequently not unambiguous. Analysis of gathered data and depth matching was made. Author has created several types of neural networks (such as MLP, RBF). ANNs were made with different input parameters, number of hidden neurons, activation and error functions. Neural network of best quality was used in permeability prediction in well with dolomite layers.