Neural networks for Nyquist plots prediction in a nanocomposite polymer electrolyte (PEO–LiPF6–EC–CNT)

Abstract

In this study, the Nyquist plots for nanocomposite polymer electrolyte system (polyethylene oxide (PEO)–lithium hexafluorophosphate (LiPF6)–ethylene carbonate (EC)–carbon nanotube (CNT)), which was produced by using solution cast technique, were obtained using Bayesian neural network. First, to prepare the training and test set of the network, some results were experimental obtained and recorded. In the experiment, PEO, LiPF6, EC, and CNT were mixed at various ratios. The effects of the chemical composition on the impedance spectra of polymer electrolyte system were investigated. In neural network training, different chemical composition and real impedance were used as inputs and imaginary impedance in the produced polymer electrolytes was used as outputs. After the training process, the test data were used to check system accuracy. As a result, the neural network was found successful for the prediction of imaginary impedance of nanocomposite polymer electrolyte system.