| Wheat is one of the important grain crops in China,the total annual production accounts for 22%of the total national grain production,and is an important guarantee for the supply of grain crops in China,so it is extremely important to ensure the quality and safety of wheat.The moisture content of wheat is one of the important factors affecting its quality and safety,and is also an essential testing indicator in wheat acquisition,storage,processing,transportation and other aspects.The current national standard test method for wheat moisture content is the oven drying method,which has high detection accuracy but has problems such as slow detection speed,poor sample representativeness and grain loss,and is not suitable for real-time online measurement of wheat moisture content.With wheat production and storage remaining high,there are still challenges for high accuracy,non-contact,non-destructive and fast online testing technology for wheat moisture content.Based on the principle of microwave free space detection,this paper introduces a meta-surface planar wave antenna and an error parameter de embedding algorithm to establish a wheat moisture content detection system.The dielectric and attenuation characteristics of wheat with different moisture content are detected at microwave frequencies of 5.8-5.9 GHz and 23.5-24.5 GHz,respectively,and a density independent wheat moisture content prediction model is established.The main research work is as follows:(1)Two kinds of meta-surface plane wave antennas with low profile,low loss and easy machining are designed.The resonance frequencies of the two kinds of meta-surface plane wave antennas are 5.8GHz and 24GHz respectively,which can convert the spherical wave emitted by the feed antenna into plane wave.The 5.8GHz meta-surface plane wave antenna is composed of square meta-surface transmission units,with a peak gain of 21.55d Bi,which is 9.95d B higher than the peak gain of the feed antenna.The 3d B beam width is reduced by74.3 degrees in the xoz plane and 69.2 degrees in the yoz plane.The 24GHz meta-surface plane wave antenna is composed of hexagonal meta-surface transmission unit,with a peak gain of 21.35d Bi,which is 15.5d B higher than the peak gain of the feed antenna.The 3d B beam width is 71 degrees lower in the xoz plane and 76 degrees lower in the yoz plane.The application of meta-surface plane wave antenna in the microwave free space method detection system can effectively reduce the scattering and diffraction effects of microwave transmission at the edge of the container,and improve the accuracy of water content detection.(2)A microwave moisture content detection system for wheat based on a 24GHz meta-surface lens antenna was designed.The system measures the attenuation variation characteristics of wheat with different moisture contents in the frequency range of 23.5-24.5GHz.The attenuation variation characteristics are determined by the attenuation differenceΔA of microwave propagation in the detection system with and without wheat samples.The linear regression equation(coefficient of determination R~2=0.9946)was established as the prediction model of wheat moisture content by linearly fitting the attenuation differenceΔA to the standard moisture content of wheat.Seventy-two wheat samples were selected for moisture content prediction,and the root mean square error RMSE,mean absolute error MAE and maximum relative error MRE were 0.193%,0.16%and 5.25%,respectively,compared with the standard moisture testing method.(3)A wheat moisture content detection system based on a 5.8 GHz meta-surface planar wave antenna and an error de embedding algorithm was proposed.The system is based on the propagation characteristics of spatial plane waves perpendicular to the conductor surface and the multi-layer dielectric interface.An antenna error and container error de embedding algorithm is proposed to calibrate and compensate the detection errors caused by the dielectric properties of the meta-surface plane wave antenna and the container itself during the measurement process.The accurate complex relative dielectric constant of wheat is obtained by combining the relationship between scattering parameters and dielectric properties.The test results show that the dielectric constant and loss factor of wheat exhibit a good linear relationship with the standard moisture content of wheat.By analyzing and selecting the ratio of the dielectric constant and loss factor of wheat to the moisture content for linear fitting,the impact of the bulk density of wheat in the container on the test accuracy is reduced.A linear regression prediction model for moisture content is established(determination coefficient R~2=0.993).Compared with the standard moisture detection method,the prediction results show a root mean square error The average absolute error and the maximum relative error are 0.19%,0.16%,and 4.0%. |
PDF Full Text Request