Current Proceedings on Technology
Yazarlar: Deniz Korkmaz, Ömer Faruk Alçin, Sami Ekici
Konular:-
Anahtar Kelimeler:Memristor,Radial basis artifical neural network,Back-propagation,Feed-forward,Training algorithm
Özet: In this paper, an Artificial Neural Network (ANN) model is presented for the missing memristor device by using Matlab/Neural Network Toolbox. Input and target data sets are derived based on the mathematical model of memristor. The ANN model consists of the Radial Basis Artificial Neural Network (RB-ANN) and the designed model is compared with Back-Propagation Feed-Forward Neural Network (BP-FFN). Before training, input and target data sets are normalized between ranges of -1 to 1. For the spread of radial basis coefficient, two-level grid-search is performed and chosen around 0.0241 for compatibility of the normalized data sets. Obtained simulation results are illustrated in the paper and compared with the mathematical and ANN models.