Using Temperature Measurements, Energy Balancing and Inferential Parameter Estimation

"Energy balance modeling has recently been shown to be a competent tool in the estimation of SAG mill discharge rheology. Work performed by van Drunick and Moys (2001) validated the ability of an energybalance filter to estimate the slurry density from an operating SAG mill. The importance of rheological data is highlighted by several authors who show the effects on load behaviour and thus mill performance. Energy balance models were developed to account for the dynamic energies in the mill discharge sump. The models then made it possible to infer the mill discharge density. A sampling campaign was performed and the errors from the sump energy balance were within tolerable limits.Since the initial results were published, it has been decided to further develop the modeling techniques in order to focus on balancing the energies around the entire mill. The critical measurements for the energy balance will be the temperatures of all the streams entering and leaving the mill, including energy losses to the environment. Metso’s OCS© (Optimising Control System) currently controls the 3 SAG mills at Anglogold’s Mponeng Gold Plant (South Africa). The OCS has built-in Kalman filtering tools to effectively handle variances and noise in the models and the mill circuit data respectively. The new models are therefore constructed in the OCS platform, and will be run in parallel with Metso’s current modeling and filtering systems. The work is a collaborative effort between Anglogold, the University of the Witwatersrand and Metso. The benefits of energy modeling, compared to mass modeling, are that energy data are more accurately measurable than mass data. The inferred mill parameters can therefore be improved provided that a competent set of models is produced to process the (accurate) thermal measurements."