CALIBRATION METHODS FOR MULTIPARAMETER MICROWAVE SENSORS

Authors: TRABELSI, SAMIR - UNIV OF GEORGIA; Nelson, Stuart

Submitted to: IEEE Transactions on Instrumentation and Measurement
Publication Type: Proceedings
Publication Acceptance Date: 5/1/2003
Publication Date: 5/19/2003
Citation: TRABELSI, S., NELSON, S.O. CALIBRATION METHODS FOR MULTIPARAMETER MICROWAVE SENSORS. INSTITUTE OF ELECTRICAL AND ELECTRONICS ENGINEERS INSTRUMENTAITON AND MEASUREMENT/2003 TUTORIAL PROGRAM, MOISTURE AND DENSITY SENSING IN GRANULAR SOLIDS THROUGH MICROWAVE DIELECTRIC MEASUREMENTS, IEEE CONFERENCE VAIL, COLORADO. 2003. p. 23-30.

Interpretive Summary: New techniques are needed in many industries for rapidly sensing the characteristics of granular materials. In agriculture, there is need for such techniques to sense the moisture content of grain and oilseeds reliably while they are being harvested and at many other points in the handling, marketing, storage, and processing of these commodities. New techniques have been discovered for sensing both moisture content and bulk density of such materials through microwave measurement methods. This paper describes principles for three different approaches to calibrating microwave measurement systems to determine moisture content and density of grain and oilseeds. One approach determines the moisture content and bulk density from the measured dielectric properties of the grain or seed independent of temperature or the bulk density and temperature independent of moisture content. A second approach uses the measured dielectric properties to determine bulk density without knowing either moisture content or temperature. The third approach determines moisture content independent of bulk density and independent of the kind of grain or seed, thus showing promise for a universal moisture calibration for grain and seed. These findings should stimulate development of new microwave moisture meters that will have more simple and more economical calibration requirements and provide better moisture information for management of commodities with consequent improvement in quality of such products for consumers.

Technical Abstract: Three dielectric-based calibration approaches are presented for predicting physical properties of grain and oilseed from measurement of their dielectric properties. In the first approach, regression analysis provided explicit direct relationships between the dielectric properties of these materials and their physical properties, i.e., bulk density, moisture content, and temperature. In the second approach, the complex-plane representation provides an expression for bulk density in terms of the dielectric properties that is independent of moisture content and temperature. In the third approach, a single dielectric-based moisture calibration equation is used to determine moisture content in granular materials presenting significant structural and compositional differences. All of these approaches and relationships established in this study may constitute the basis for development of dielectric-based multiparameter sensors for indirect, nondestructive, and instantaneous determination of physical properties of granular materials from measurement of their dielectric properties. Reliability and stability of these calibration methods is mainly dependent on the accuracy with which the dielectric properties are measured. These findings along with the availability of low-cost microwave components and information technology capabilities provide more incentives for the commercialization of multiparameter microwave sensors and their widespread use in different industries.