ANN prediction of corrosion behaviour of uncoated and biopolymers coated cp-Titanium substrates

Abstract The present study focuses on biopolymer surface modification of cp-Titanium with Chitosan, Gelatin, and Sodium Alginate. The biopolymers were spin coated onto a cp-Titanium substrate and further subjected to Electrochemical Impedance Spectroscopic (EIS) characterization. Artificial Neural Network (ANN) was developed to predict the Open Circuit Potential (OCP) values and Nyquist plot for bare and biopolymer coated cp-Titanium substrate. The experimental data obtained was utilized for ANN training. Two input parameters, i.e., substrate condition (coated or uncoated) and time period were considered to predict the OCP values. Backpropagation Levenberg-Marquardt training algorithm was utilized in order to train ANN and to fit the model. For Nyquist plot, the network was trained to predict the imaginary impedance based on real impedance as a function of immersion periods using the Back Propagation Bayesian algorithm. The biopolymer coated cp-Titanium substrate shows the enhanced corrosion resistance compared to uncoated substrates. The ANN model exhibits excellent comparison with the experimental results in both the cases indicating that the developed model is very accurate and efficiently predicts the OCP values and Nyquist plot.