Based on microwave dielectric properties of wheat, corn, barley, oats, grain sorghum, soybeans, canola, shelled peanuts and pod peanuts measured over ranges of frequency and moisture content, models for predicting the dielectric constant and loss factor of these commodities are described. Nearly linear relationships between dielectric properties and log of frequency or frequency and moisture content permitted multiple linear regression models to be developed for predicting dielectric constants and loss factors for each type of grain and seed at 23degreesC. Models predicted dielectric constants with s.e. of 1 to 2% with some up to approx. 4% for grain and seed at frequencies in the range 5 to 15 GHz over given moisture ranges. Loss factors are predicted with s.e. of a few percent, but much larger errors can naturally result when loss factors are very small.
