Moisture content of cereal grains is one of the most important characteristics for determining quality of grains.Objectives
The electrical properties of Indian wheat with moisture content range of 14.3% to 29.38% were characterized to develop a moisture prediction model.Methods
A specially designed coaxial cylindrical cell was used for the purpose and electrical properties characterization system was used to carry out measurements of these properties. The studies were conducted for frequency range of 100 kHz- 10 MHz and a fixed frequency was selected based on best correlation basis. Principal Component Analysis (PCA) was used to study the moisture dependence of various electrical properties such as Capacitance (C), Conductance (G), Impedance (Z), Theta (Θ) and Dissipation Factor (D) of Indian wheat. The various statistical studies were conducted using The Unscrambler Version X, a multivariate Statistical Analysis Software.Results
It was observed that Conductance (G) and capacitance (C) show the maximum variation with moisture. Partial Least Square (PLS) Regression was then used to develop the moisture prediction model based on these two electrical properties.Conclusion
The developed model is highly accurate with high correlation of 0.973.