Comparative Studies of Response Surface Methodology (RSM) and Predictive Capacity of Artificial Neural Network (ANN) on Mild Steel Corrosion Inhibition using Water Hyacinth as an Inhibitor

Abstract

Abstract Response surface methodology (RSM) and artificial neural network (ANN) on modeling and optimization of corrosion inhibition efficiencies of mild steel using water hyacinth as an inhibitor was carried out in this work. The optimization of the process was done using generic algorithm (GA) and RSM which were subsequently compared. The optimum inhibition efficiency predicted were 87.675924% and 82.89% by ANN and RSM respectively. The value of R 2 obtained were 0.9695 and 0.85118 for ANN and RSM models respectively while RMSE values of 3.90 and 4.3089 were gotten for RSM and ANN models respectively. The model regression indicated that RSM best fit the experimental data thus perform better on mild steel corrosion inhibition.