Simultaneous modeling of efficiency and filtration rate of zinc extraction from concentrate by operating condition and use of Artificial Neural networks

Efficiency and filtration rate of zinc extraction from concentrate is a function of its operating condition as pH, temperature and time of extraction of each stage, agitation speed and delay time between stages. Capacity of zinc extraction depends on efficiency of extraction and rate of filtration, simultaneously. There are some mathematical models to predict efficiency of zinc extraction but in narrow range of operating condition and there is not any mathematical model to predict filtration rate of extraction as operating condition. For optimization of zinc extraction, these two parameters (Efficiency and filtration rate) are needed, simultaneously. In this research, efficiency and filtration rate of zinc extraction were modeled as operating condition by artificial neural networks (ANN). ANN with different neurons in two hidden layers was employed and the regression coefficient of normalized experimental data and normalized predicting data show that the networks with eight and six neurons in first and second hidden layer respectively have good agreement with experimental data. Regression coefficient of normalized experimental data and normalized predicting data is 8799.0 in this research.

Keywords:

Zinc extraction , Efficiency , Filtration rate , Neural networks , Operating condition , leaching