Modeling for electrical impedance spectroscopy of (4E)-2-amino-3cyanobenzo[b]oxocin-6-one by artificial neural network

Abstract

The efficiency of artificial neural networks (ANNs) for modeling the electrical impedance spectroscopy of (4E)-2-amino-3-cyanobenzo[b]oxocin-6-one was investigated. The experimental data for electrical impedance and dissipation factor were used as input data for the model. The optimum network structure was obtained by testing different numbers of neurons with altered transfer functions to normalize the data. This structure simulated the experimental data with very high accuracy and predicted new values that were untested experimentally. A nonlinear equation that indicates the relation between inputs and output was introduced based on the ANN model. The performances of the optimum network are obtained. Finally, this study showed that neural networks are a very effective tool in modeling and are able to follow the patterns of the experimental data with high precision.