Real-time sugar content monitoring system of sugarcane on an elevator conveyor using near-infrared spectroscopy
This research aims to propose a prototype of an online detection system based on the visible and near-infrared spectroscopic (vis/SW-NIR) technique and the result of the performance evaluation of the system for real-time evaluating the soluble solids content (SSC) of sugarcane billets on the elevator conveyor. The system consisted of two main parts, a cane billet elevator and a spectra acquisition part. The elevator speed was set to 2 m/s for transferring sugarcane billets. For the spectra acquisition, four 50 W tungsten halogen lamps were used as the light source while a vis/SW-NIR spectrometer was used for collecting the reflectance spectra of the cane billets. Partial least squares regression (PLSR) correlating the spectra with the SSC values was applied for model development. Test set validation was used to evaluate the performance of the system, showing the coefficient of determination of prediction (R¬2), root mean square error of prediction (RMSEP), and residual predictive deviation (RPD) of 78.5%, 0.301 ºBrix and 2.16, respectively. The result shows that the proposed system is feasible for online SSC measurement of the sugarcane billets on the elevator conveyor.