Optimization of Regrinding Ball Mills at Carol Pellet Plant of the Iron Ore Company of Canada

This study was concerned with the application of a computer simulation methodology to optimize the performance of mills grinding iron ore concentrate prior to pelletizing. The methodology involves batch and plant testing, modeling and computer simulation. The analysis of grinding performance was carried out and capacity increases were predicted in the context of detailed population balance models developed for the plant ball mills. Model developnent and plant confirmation were performed using data obtained in a series of batch tests and three plant tests. From experimental data and computer simulation it was found that ball size, percent solids and classification all had a significant effect on mill throughput. Based on the findings of this study, a 12-152 increase in capacity was predicted to be available by a proper combination of ball size charge and optimum percent solids. These predictions have been confirmed by plant tests in which the recommended operating changes have been made.