Artificial neural network to predict excess physico-chemical properties of binary mixture of TBP and CCl4.

Artificial Neural Network (ANN) is becoming a prevalent modeling tool for processes involving non-linear multivariable relationships. In order to develop a correlation between the variables, a large number of neurons are arranged in layers and interconnected. The novel feature of the current work is estimating the excess properties (VE, ɛE, PE and RE) using ANN for liquid mixture containing TBP and CCl4. The excess properties support in analyzing the type of interactions existing between the extractant (TBP) and diluent (CCl4) in order to frame an efficient system for liquid-liquid separation technique. The methodology adopted in this system can be used in the analysis of wide-ranging data sets. [ABSTRACT FROM AUTHOR]