In the formula feed industry, sufficient mixing is essential to produce uniformity in blended mixtures of solid ground maize or sorghum with protein supplement and other components. Excess mixing is a waste of energy and time, adding cost to the process. We use near-IR imaging to assess adequate mixing as an alternative to inorganic tracers with density and flow characteristics that differ from other commodities and do not reflect the distribution of components. Chemically dependant image contrast was used to reveal the result of mixing with a self-tracer (protein supplement) and an edible alternative protein supplement possessing a combination of spectral features not found in any other commodities. Four samples taken from a pilot scale two cycle double ribbon mixer were imaged with three replicates each. Heterogeneity revealed in the first two sample sets obtained after a few cycles gave way to excellent homogeneity in the final segment after operation of 25 cycles. This was done for ground maize/soybean meal, with the latter used as a self-tracer. The process was repeated with added blood meal, to use its distinct spectral features as an organic tracer. In each case, the tracer spectrum was characterized using a multivariate signature. Each image pixel was identified, and the concentration was detected by pixel counting. © 2010 Elsevier B.V. All rights reserved.
Mendeley saves you time finding and organizing research
Choose a citation style from the tabs below