Non-destructive prediction of moisture content of lime (Citrus aurantifolia Swingle 'Paan') by multiple regression analysis of its electrical and physical properties
Large quantity of juice is an important index of lime quality that consumers seek for. Therefore, a non-destructive technique for prediction of lime juice quantity is needed. In this study, moisture content (MC) of lime which is an indicator of its juice quantity was predicted by multiple regression analysis of its electrical properties - capacitance (C), inductance (L) and impedance (Z) at various frequencies (0.012, 0.05, 0.1, 0.2, 5, 10, 20, 50, 100 and 200 kHz) - and physical parameters - weight and geometric mean diameter (GMD). Samples (n=82) were divided into a calibration set (n=55) and a prediction set (n=27). A calibration model for moisture content of lime was established and cross-validated by partial least squares regression (PLSR). Prediction results achieved a coefficient of determination (R2) of 0.934 and a root mean square error of prediction (RMSEP) of 1.822% wet basic, demonstrating that this technique has a real potential for development into a practical non-destructive lime screening method.
Huong, H.T. and Teerachaichayut, S. 2017. Non-destructive prediction of moisture content of lime (Citrus aurantifolia Swingle 'Paan') by multiple regression analysis of its electrical and physical properties. Acta Hort. (ISHS) 1152:299-306
lime, impedance, capacitance, inductance, GMD, weight, moisture content