Abstract:Aiming to explore a rapid method for detecting content of wheat content, based on microwave free-space measurement, the effects of frequency (1~4GHz), temperature (10~40℃), moisture content (11.35%~17.79%) and bulk density (low and high) on the relative dielectric constant of wheat were studied. The factors causing the effect of changes in the relative dielectric constant of wheat were analyzed, the prediction models of wheat moisture content with different frequencies were established, and the model was verified. The experimental results demonstrated that in the frequency range of 1~4GHz, the relative dielectric constant of wheat was increased with the increase of moisture content, temperature and bulk density of samples, but it was decreased when frequency was increased. Furthermore, all samples were partitioned into calibration set (60 samples) and prediction set (24 samples), the established support vector machine (SVM) model had good performance for the moisture content prediction. Compared with single-frequency and multi-frequency, the prediction model with full-frequency was the best, and the correlation coefficient of prediction, root-mean-square error of prediction and residual prediction deviation were 0.9929, 0.0513% and 18.67, respectively. The results indicated that the microwave free-space measurement could be used to detect moisture content of wheat. The research was helpful to develop a new wheat moisture detector which could be used in situ or online detection.