Optimal number of neurons for a two layer neural network model of a process

Neural networks are known as powerful tools to represent the essential properties of nonlinear processes because of their global approximation property. However, a key problem in modeling nonlinear processes by neural networks is the determination of neuron numbers. In this paper, a data based strat...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Asadi, M. S., Fatehi, A., Hosseini, M., Sedigh, A. K.
Format:	Tagungsbericht
Sprache:	eng
Schlagworte:	Approximation methods bicoherence test Biological neural networks Complexity theory Feedforward neural networks Frequency domain analysis Inphase-quadrature demodulation neural network Neurons Nonlinearity measure Training
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	Neural networks are known as powerful tools to represent the essential properties of nonlinear processes because of their global approximation property. However, a key problem in modeling nonlinear processes by neural networks is the determination of neuron numbers. In this paper, a data based strategy for determining number of hidden layer neurons based on the Barrons work, describing function analysis and bicoherence nonlinearity measure is proposed. The proposed algorithm is evaluated for a pH neutralization process. It is shown that this algorithm has acceptable results.