Artificial Neural Network Approach For Modelling The Extraction Of Gold In The Mixer Settler

The process of liquid?liquid extraction of gold from chloride solution is complicated by the presence of many variables acting simultaneously in an unknown non-linear fashion for the determination of the final extraction efficiency. In practice, it is quite difficult to develop a model for predicting various properties of such process. Therefore, an alternative approach based on Artificial Neural Network (ANN) is considered to assist in the simulation process. Artificial neural networks (ANN) are powerful tools that can be used to model and investigate various highly complex and non-linear phenomena. This paper presents application of Multi-Layer Perceptions (MLP) to construct relationship that will be employed to develop a quantitative method for the estimation of the extraction efficiency as a function of hydrodynamical parameters such as stirrer speed, [AuCl4]- and DBC concentration, organic and aqueous flow rate. The model is shown to be in excellent agreement with the experimental data (R2 value for training and testing data was 0.9992 and 0.9915 respectively). Therefore, it can be used to optimize the parameters of liquid?liquid gold extraction with dibutyl carbitol (DBC).