Moisture content determination of oilseeds based on dielectric measurement

Abstract

Oilseeds have an important role in edible oil production.  Moisture content measurement of oilseed is an inevitable operation in harvesting and almost all postharvest processing such as handling, storage, milling and oil extraction.  In this paper, a cylindrical capacitive sensor was used to predict the moisture content of sesame, soybean and canola seed as a simple, low cost, rapid and reliable method.  Two varieties of each oilseed were selected and extracted equation from a variety was evaluated for another variety.  The hyperbolic regression and paired t-test were utilized to extract the calibration equations and perform a comparison between predicted moisture with actual values.  The R² of calibration for Dashtestan and Ultan sesame were 0.998 and 0.999, respectively, for L17 and Sahar soybean were 0.972 and 0.965, respectively and for Okapi and Talaiyeh canola were 0.993 and 0.994, respectively.  The R² of prediction for Dashtestan and Ultan sesame were 0.966 and 0.932, respectively, for L17 and Sahar soybean were 0.963 and 0.952, respectively and for Okapi and Talaiyeh canola were 0.993 and 0.994, respectively.  Results of paired t-test confirmed that the measured and predicted moisture content of all oilseeds were not statistically different at the 5% level (p > 0.05).  Based on obtained results the designed system using capacitive sensor is valid and reliable for moisture measurement of the studied oilseeds.

Keywords: oilseed, sesame, soybean, canola, moisture content, capacitive sensor