Modeling Studies Of Semi-Commercial Flotation Column For Beneficiation Of Sillimanite Using Artificial Neural Network

The present paper discusses a three layer feed forward artificial neural network (ANN) model, trained using the error back propagation algorithm, has been established to simulate the column flotation circuit used for beneficiation of Sillimanite. Parameters such as superficial air velocity, wash water rate, froth height, % solid, feed velocity, sodium silicate and oleic cid are considered as process operating variables and % yield of Sillimanite is the output of the experiment. The results from the ANN modeling, involving the non linear relationship between inputs and outputs, indicate good agreement with experimental observations. The network model validates the experimentally observed trends. The optimal model parameters in terms of network weights have been estimated and can be used for computing parameters of the process over wide-ranging experimental conditions.