F Nakhaei - Department of Mining & Metallurgical Engineering, Amirkabir University of Technology
M Irannajad - Department of Mining & Metallurgical Engineering, Amirkabir University of Technology

Metallurgical performance prediction is a key parameter associated with the characterization of any control process. In fact, it is not possible to have accurate control to many mineralprocessing problems without having accurate prediction value. Attempts have been made to utilize artificial neural networks (ANNs) for prediction of process performance for the completeuse of available control capabilities. Despite of the wide range of applications and flexibility of ANNs, there is still no general framework or procedure through which the appropriate network for a specific task can be designed. Design and structural optimization of neural networks is stillstrongly dependent upon the designer's experience. This is an obvious barrier to the wider applications of neural network. To mitigate this problem, a new method for the auto-design of neural networks was used, based on genetic algorithm (GA). The new proposed method wasevaluated by a case study in pilot plant flotation column at Sarcheshmeh Copper Plant. Design of topology and parameters of the neural networks as decision variables was done using genetic algorithms in order to improve the effectiveness of forecasting when ANN is applied to apermeability predicting problem from well. The chemical reagents dosage, froth height, air, wash water flow rates, gas holdup, Cu grade in the rougher feed, flotation column feed, column tail and final concentrate streams were used to the simulation by GANN. In this work, multi-layer GANNs with Back Propagation (BP) algorithm with 8-17-10-2 arrangement has been applied to predict the Cu grade and recovery, respectively. The correlation coefficient (R) values for the testing sets for Cu grade and recovery were 0.93, 0.93 respectively. The results discussed in this paper indicate that the proposed model can be used to predict the Cu grade and recovery with a reasonable error.

Artificial Neural Network, Genetic Algorithm, Flotation Column, Grade, Recovery,Prediction