Automatic quality estimation in blending using a 3D stockpile management model.

Stockpile blending is always considered to be an effective method of controlling the quality and maintaining the grade consistency of delivered bulk material, such as iron ore. However, major challenges remain to predict the quality of a stockpile during blending (stacking and reclaiming) operations, because the chemical composition of the ore body is not always available during blending. Consequently, the performance of current stockpile management systems is relatively poorer than expectations. This paper details an innovative modelling approach to estimate the quality of a stockpile during the blending process. The geometric model created from laser scanning data is capable of recording the dynamic shapes of the stockpile using mathematical equations. Therefore, the quality of the stockpile is calculated with a high degree of accuracy when the chemical analysis is completed. Thus, a quality embedded geometric model is created. Furthermore, this geometric model is associated with the kinematic model of a Bucket Wheel Reclaimer (BWR) to achieve precise auto-control in reclaiming operations. The link between these two models also supports the quality of every single cut accomplished by the BWR, to be calculated in advance. Using the calculation results in conjunction with precise machine control, the output grade can be predicted, planned and controlled. This will optimise the efficiency and effectiveness of stockpile blending. [ABSTRACT FROM AUTHOR]