Ultra-wideband microwave humidity detecting method based on machine learning

Abstract

Microwave detection is nondestructive and rapid, the portability is good, but a severe development bottleneck exists on the aspect of humidity range detection. Most existing microwave moisture detecting systems adopt single frequency points for humidity measurement, are not wide in measurement range and are difficult to apply in practice. A microwave attenuation principle is utilized, microwave scattered signals of measured objects different in humidity under broadband frequency are obtained by using ultrawide-band antennae and serve as humidity regression training sample sets of the measured objects, and accordingly a measured object humidity regression model is established by utilizing a supervised machine learning method. The regression-type machine learning algorithm is adopted for data modeling, a cross validation mode is utilized to obtain optimum training parameters, accordingly the obtained model is optimized, and regression errors are smallest. The humidity range of fabrics which can be detected is greatly increased, and a foundation is laid for further devolvement of the microwave detection systems towards the application field.

Description

technical field [0001] The invention relates to an ultra-wideband microwave humidity detection method based on machine learning, which is especially suitable for conveniently and quickly judging the moisture content of fabrics under the condition of no radiation, no damage and wide humidity range, and belongs to the technical field of microwave detection. Background technique [0002] At present, the commonly used test methods for moisture content of materials at home and abroad include oven method, DC resistance method, capacitance method and infrared method. Although the results of the oven method are accurate, it is troublesome to use in actual work, not real-time online, and belongs to destructive testing; the DC resistance moisture measurement method is due to the large DC resistance of the measured object and the easy polarization of the electrode plate in the DC electric field, etc. There are defects such as poor test stability, large error, and low versatility; while...

AI Technical Summary

Problems solved by technology

The method of detecting the internal moisture content of objects based on microwave signals generally adopts the microwave transmission method, and mainly uses the microwave attenuation of a single frequency point to measure the moisture content of the material. In actual use, the single frequency point is easily disturbed by the outside world, which leads to an increase in the humidity measurement error.

In addition, there is still a problem that the detection range of humidity is very narrow in the microwave detection of material moisture at present.

In short, the current moisture microwave detection technology is facing a severe development bottleneck in the detection range.

Images