Improving the Morrell C-model's accuracy in predicting the ball mills’ power draw based on calculating the dynamic voidage of grinding media

Abstract

This paper investigates grinding media's dynamic voidage to improve the Morrell-C model's accuracy in predicting the ball mills’ power draw. Using a three-level factorial design, we provided an empirical model to determine the voidage of each ball size distribution proposed by Bond for ball mills’ first filling in various ranges of fractional mill filling and mill rotating speed. Moreover, by employing the multiple regression method, a general model was developed to predict the balls’ dynamic voidage by calculating the mean absolute deviation (MAD) of the balls’ diameter. Results revealed that the balls’ dynamic voidage increases with increasing the rotating speed and decreasing the fractional filling and MAD of the balls’ diameter. The evaluation of the grinding media's voidage prediction models’ performance in improving the Morrell-C model's accuracy indicated an increase in the model's predictive power by calculating the ball mills’ load bulk density based on the dynamic voidage of grinding media.