Petrophysical characterization of comminution behavior

摘要: Comminution or feed size reduction is typically the first stage of ore processing at mines. Comminution tests are commonly conducted to assess behavior and to aid in process design and equipment selection. Testing for Bond mill work index (BMWi), a measure grindability, A*b, crushability, is common this regard. These destructive expensive time consuming and, hence, on limited number large volume samples which most cases are not representative entire orebody. Therefore alternative means desirable for efficiently characterizing comminution behavior. Petrophysical properties have potential effective characterization ore comminution truly suite samples. Petrophysical measurements quick, non-destructive, relatively cheap. Petrophysical data can be recorded either downhole core. If calibrated against measures crushability and petrophysically-based models could provide virtually continuous downhole prediction attributes intervals drill holes where these parameters available. This thesis presents new approach of ore based petrophysical measurements. As an geophysical logging, Geotek multi-sensor core logger (MSCL) was evaluated. Density, P-wave velocity, amplitude magnetic susceptibility, as well imagery, were measured cores from two Australian copper-gold deposits, namely Cadia-East, NSW, Ernest Henry, QLD. The Geotek system had never previously been used metalliferous mines. It provides with acceptable accuracy if carefully systematically but quality is adversely affected by small condition achieved in production logging approximately ±1.35% density, ±6.5% velocity, and ±1% magnetic susceptibility. The relationships between (A*b and BMWi) directly investigated, since small-scale been performed selected 2m same At Cadia East, hard in terms both crushing grinding. Henry more variable but generally softer.In cases relationship comminution parameters dependent type. Hence class-based approaches comminution modeling devised implemented. Crushability (A*b) be related to petrophysical reliably than grindability (BMWi). consistent with the fact that crushability whole rock while BMWi crushed composite Prediction high BMWi materials (>10 kWh/t) proved difficult, perhaps because particles competent crushed size. An important outcome susceptibility good indicator A*b both sites define different domains. as susceptibility increases (samples easier crush) magnetite acts crack initiator. becomes harder crush susceptibility increases; association feldspar with magnetite probably reason the low values case. At developed in depth measurements available test data not. Four classes defined P-wave amplitude, density using cluster analysis. Regression were developed each class. overall root mean square (RMS) error 1.39 kWh/t 27.3 respectively. Comminution modeling East difficult due variability of comminution parameters. and BMWi around their respective values. then linked assays neural network approach. performance of neural networks tested successively treating each hole independent hole. ranged 51% 77%. A novel attributes from images also investigated Henry. Estimates mineral abundance from classified adjusted achieve compatibility assay data. Bulk predicted volumes densities relative error of 3.5%. coefficients estimated for each phase via least squares optimization. method depth imagery but comminution test RMS errors 33.3 and 1.68 respectively. The case studies geological environments show data can useful information hence prediction throughput. Petrophysics-based limitations but they adequate use during planning. such can be improved reducing uncertainties measurements, refining classification techniques, increasing petrophysical properties recorded, incorporating other including