Development of a feedforward system based neural networks for the control of a hydrometallurgical unit of cobalt

Abstract

The application of a feedforward control system based on a model of artificial neural networks in the cobalt industry has made it possible to identify the neutralization step as a bottleneck. It results in a considerable reduction in productivity (20%) and an increase in cycle time of about 18%. It is important to note that this is mainly due to a high solid level in this neutralization step. The implementation of the FF-ANN control scheme made it possible to control the pH at its optimum value of 0.9. This regulation also made it possible to reduce the presence of a solid content of 31%, which consequently leads to control of cycle time and optimum production.