Monitoring vegetable dehydration process by aquaphotomics from lab scale to farm.

ABSTRACT

Solar dehydration processes, implemented with NIR sensors, were studied. The research plan was divided into phases to achieve specific objectives. Phase 1 laboratory tests on micro plants; phase 2 scale transposition tests on small-sized plants - pilot scale; phase 3 tests on commercial systems already in use by Italian SMEs. The realisation of the scheduled activities started with the design, programming, and positioning of NIR sensor for data collection and configuration optimization. NIR spectra were collected in reflectance mode (900-1700 nm) using the MicroNIR1700™ or the MicroNIR On-site W™ (VIAVI Solutions Italia S.r.L., Monza, Italy) portable spectrometers with the MicroNIR Pro ES 1700 software. Spectra were acquired automatically throughout the process by placing the NIR probe over a sample slice, positioned on the intermediate shelf. The probe was thermally insulated to avoid temperature variations. The spectra were first transformed by converting reflectance to absorbance; then the second derivative Savitzky-Golay filter (second order polynomial fit and 21 points) and multiplicative scatter correction were applied to remove potential scatter effects. Aquagrams were calculated from the spectral data. The experiments were carried out with a micro-drying system (45x45x45 cm), and a pilot scale plant available at CREA.IT (Milan). Then, the transfer of the selected configuration was realised using a commercial plant already active in a farm of Pavia area. Different food matrices were tested (pineapple, apple, melon, eggplant, onion). NIR data processing by multivariate analysis was made to prove the reliability of the aquaphotomics approach in detecting the actual end of the drying process. The evaluation of the potential development of dedicated App, easy to consult, to be made available through integration on virtual platform was finally computed.