A reliable radial basis function neural network model (RBF-NN) for the prediction of densities of ionic liquids

Density of liquids is an important physical property which clear knowledge of its value is essential in various scientific and engineering applications such as design of equipment, liquid metering calculations and many other applications. Hence, developing accurate and reliable predictive tools for prediction of this property seems to be of great importance. In this communication, a model was developed based on radial basis function neural network (RBF-NN) for estimation of density values of 146 ionic liquids at various temperatures. The outputs of the study reveal that the new model is capable to predict density data with an acceptable accuracy. The reliability and precision of the model was validated by using various statistical and graphical approaches. Results of the developed model were put into comparison with literature correlations and it was observed that the model effectively outperforms other correlations and exhibit higher accuracy and reliability.