Figure 51 - Schematic of CCRD experimental structure for 3 process variables. An application of CCRD in a full-scale production environment is reported in Vizcarra et al. (2018) where it was used to improve the performance of a fully autogenous (AG) mill in the Boyne smelter plant in Queensland, Australia. The product of the AG mill was being used as anode cover, acting as an insulating barrier and allowing the thermal balance of the underlying reduction cells to be controlled. Anode cover was also the most important barrier mitigating fluoride emissions into the environment. The Boyne AG mill was discharging a product that was excessively fine, making it ineffective at dissipating heat from the underlying reduction cells. There was a desire to limit any capital spending that would normally be considered to prevent the overgrinding of feed (e.g. pre-screens), and so a CCRD was designed that would enable the effect of two key controllable variables, feed blend and mill power draw, upon product sizing to be established, and optimum settings identified. The corresponding design employed for the Boyne trial is shown in Table 4. Key features are the mill power and feed blend set-point combinations (a third feed source comprised the balance of each blend). Also critical was the incorporation of ‘time-blocks’ which act to guard against the effect of systematic background changes in time that would otherwise contaminate the measured effects of the trial variables. Both the order of the time-blocks, and the runs within each timeblock, were randomised. Variable B Variable A Variable C
RkJQdWJsaXNoZXIy MTM0Mzk2