Performance Comparison of Three Density-Independent Calibration Functions for Microwave Moisture Sensing in Unshelled Peanuts during Drying

Abstract

HighlightsReal-time, free-space transmission measurements of relative complex permittivity in unshelled peanuts during dryingDynamic application: temperature, density and moisture content changing during dryingThree density-independent calibration functions evaluated for accuracy in real-time moisture content determinationReal-time moisture content determination with standard error of performance (SEP) = 0.55% moisture content for all calibration functionsCalibration function most commonly used with microwave moisture sensor was determined to be most accurate; SEP = 0.448% moisture contentAbstract. A microwave moisture sensor, developed within USDA ARS, has been used to determine moisture content in unshelled peanuts during drying. Relative complex permittivities of the peanuts obtained from free-space transmission measurements at 5.8 GHz are used for the moisture determination. Due to variations in density caused by drying, it is advantageous to estimate moisture content independent of bulk density. Therefore, moisture content was estimated with three density-independent calibration functions to assess which one provided optimal accuracy. One of the functions is based on the measured attenuation and phase shift, and the other two are permittivity based (one of which is commonly used with the microwave moisture sensor). The sensor was calibrated for peanut pod moisture content determination over a temperature range of 10°C to 40°C and a moisture content range of 6.5% to 19% wet basis (w.b.). Statistical analysis showed high coefficients of determination (r2), = 0.97 for the calibration with each function. Peanut pod moisture content was determined with the sensor in real-time as peanuts dried, and estimated moisture content was compared to the reference oven drying method. While the standard error of performance (SEP) for the three functions was = 0.55% moisture content, the calibration function most commonly used with the microwave sensor was observed to be the most accurate (SEP = 0.448% moisture content). Microwave sensing is a viable solution for nondestructive, real-time determination of moisture content in peanuts in dynamic situations such as drying. Keywords: Complex permittivity, Dielectric properties, Free-space measurements, Microwave sensing, Moisture content, Peanut drying.