科研之友
微信
新浪微博
Facebook
分享链接
摘要
This study aims at evaluating the potential of a multipoint NIR spectrophotometer system based on a Fabry-Perot interferometer, which incorporates a beam splitter, combined with a four point photodiode array detector and collimated light for performing on-line analysis of beef composition. Additionally it also aims at analysing the spectral information obtained with the baseline correction setting and compares it to the information obtained using a conventional approach. The system was employed to spatially predict the composition of fat trimmings of mixed minced beef samples at varied path lengths (1, 1.5 and 4 cm). The system's performance was tested in two optional modes; (a) without baseline correction adjustment and (b) with baseline correction adjustment. Both measuring modes of the spectrophotometer demonstrated individual advantages in particular situations; the first situation evaluated the simultaneous analysis of different samples set at different path lengths per probe, whereas a second situation focused on sudden changes of path lengths of the same sample per probe during a single analysis. Partial least squares regression (PLSR) was employed to establish the quantitative relationship between the spectral data and its proximate analysis values. Overall results showed good predictions for all the chemical attributes. It was concluded from the results that a baseline correction adjustment setting is needed when sudden changes in the effective path length occur, while in a case where similar path lengths are encountered, both modes could be used, obtaining good and similar performances.
- 出版日期2016
