Prediction of copper grade at flotation column concentrate using Artificial Neural Network

The flotation column is a multivariable process whose main control objective is to guarantee the metallurgical yield set for the process operation, expressed by the grade and the recovery of the valuable mineral in the concentrate. The on-line estimation of grade usually requires a significant amount of work in maintenance and calibration of on-stream analyzers, in order to maintain good accuracy and high availability. These difficulties and the high cost of investment and maintenance of these devices have encouraged the approach of prediction of metal grade as an index of control performance. Therefore, advanced new methods such as Artificial Neural Network (ANN) must be employed. In this work, it was used from Feed-Forward ANNs (FFANNs) method for the prediction of concentrate copper grade in flotation column using the real collected data, with 3-13-6-1 structure. The wash water and the non-floated flowrates and froth height were used as inputs to the network. The output of the model was percentage of Cu grade. It was achieved quite satisfactory correlations; so that R is equal 0.943 and 0.93 in training and testing stages for Cu grade prediction, respectively. The proposed NN model accurately estimates the effects of operational variables in column flotation plants and can be used in order to optimize the process parameters without having to conduct the new experiments in laboratory.