Skip to main content
ARS Home » Southeast Area » Athens, Georgia » U.S. National Poultry Research Center » Quality Safety and Assessment Research » Research » Publications at this Location » Publication #357729

Research Project: Assessment and Improvement of Poultry Meat, Egg, and Feed Quality

Location: Quality Safety and Assessment Research

Title: Density-independent calibration functions for nondestructive moisture sensing in flowing grain

item Trabelsi, Samir
item Lewis, Micah
item NELSON, STUART - US Department Of Agriculture (USDA)

Submitted to: Journal of Microwave Power and Electromagnetic Energy
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/21/2019
Publication Date: 5/26/2019
Citation: Trabelsi, S., Lewis, M.A., Nelson, S.O. 2019. Density-independent calibration functions for nondestructive moisture sensing in flowing grain. Journal of Microwave Power and Electromagnetic Energy. 53(2):69-80.

Interpretive Summary: Nondestructive moisture sensing is important in many industries including food and agriculture, pharmaceutical, ore and minerals, construction materials and civil engineering applications. In many instances, moisture content is required in real time while the material is moving. For granular and particulate materials, bulk density changes affect the moisture sensor reading. Therefore, density-independent calibration functions were developed to determine moisture content from microwave measurements without knowledge of bulk density. These calibration functions are particularly useful for on-line moisture sensing for materials moving on a conveyer belt or flowing through a pipe or chute. In this paper, investigation of the effectiveness of three of these calibration functions is described for sensing moisture content in three different agricultural commodities, wheat, soybeans, and corn, while flowing through a chute at different flow rates. Sublots of the three commodities were conditioned to different moisture contents, and the attenuation and phase shift of microwaves transmitted through the grain were measured for both static and flowing conditions. Calibrations for each of the three calibration functions were determined on static grain, and then used for predicting moisture content in the flowing grain at three different flowing rates for each. Experimental results showed that moisture contents over a wide range for each of the three commodities were determined with good accuracy by all three calibration functions at all three of the flow rates studied. Standard errors of performance of moisture content determination of less than about 0.6% moisture content were achieved by the microwave sensing for wheat and soybeans and less than 1% moisture content for corn. These accuracies are well within requirements for many practical applications of nondestructive on-line sensing of moisture content that, if implemented, can improve management of grain handling and storage with improved quality maintenance and reduction of losses evident with current industrial practices, thus benefiting handlers, processors and consumers as well.

Technical Abstract: — In many agricultural and industrial applications, moisture content needs to be determined in real time while the material is moving. For granular and particulate materials, density-independent calibration functions were used in microwave moisture sensors to minimize the effect of bulk density changes. Three of these calibration functions expressed in term of the attenuation, phase shift and the dielectric properties were investigated for moisture determination in flowing grain and seed samples. To simulate in the laboratory measurements on flowing grain, a flowing system was designed and assembled. The flowing-grain system consisted of a conical stainless steel hopper connected to a polycarbonate chute and an iris diaphragm-type valve placed at the bottom. The iris can be manually adjusted to provide different flow rates. The flowing system was calibrated by using static samples which were placed in the polycarbonate chute between two patch antennas operating at 5.8 GHz and connected to a vector network analyzer. Measurements at room temperature on grain and seed samples flowing at varying flow rates showed that moisture content can be predicted with density independent calibration functions with a standard error of performance (SEP) of less than 1% moisture content.