System identification and artificial intelligent (AI) modelling of the molten salt electrolysis process for prediction of the anode effect

•Acquisition of the proper data for modeling the molten salt electrolysis process.•Evaluation of the dynamic behavior of the process.•Implementation of both linear and non-linear identification models for predicting the anode effect of the process.•Application of a Deep Neural Network (DNN) model for identification of the process.•Comparison of all models for estimating dynamic system outcomes.