Research On Ultrasonic Detection Technology Of Soil Moisture

The vehicle mounted information acquisition platform is suitable for the large-scale and high-throughput measurement of farmland soil parameters in precision agriculture.It is an important part of fine agricultural management to acquire soil moisture information timely and accurately and to understand the dynamic change of soil moisture.At present,the methods of soil moisture content detection include drying weighing method,ray method,dielectric property method and remote sensing method.However,in practice,these detection methods have disadvantages such as complex analysis process,long cycle,high cost and difficult to be popularized on a large scale.As a good information carrier,ultrasonic wave interacts with soil when it propagates in the soil medium,and carries a lot of information related to the change of soil moisture content in the receiving wave.Based on this,this paper takes the cultivated soil as the research object,and uses the speed change of ultrasonic wave passing through the soil sample to detect the soil moisture content.The specific work and research content of this paper are as follows:⑴ Through the analysis of the existing forms of soil water and the influence of different forms of soil water on soil structure and properties,combined with the propagation theory of ultrasound in porous particle media,the propagation characteristics and energy dissipation factors of ultrasound in soil were studied.This paper analyzes the advantages and disadvantages of common ultrasonic testing methods and the differences of the characteristics of different ultrasonic parameters on the tested parts.Considering the testing conditions and the sensitivity of ultrasonic parameters to soil moisture,the best ultrasonic parameters that can represent soil moisture are selected.⑵ Set up an ultrasonic testing platform that can realize the in-situ measurement of soil moisture content.By using the outdoor simulated rainfall method,the soil samples of three different textures were treated with six times of dry and wet alternation.the soil bulk density and ultrasonic velocity values in the process of soil samples from wet to dry were obtained respectively.The adaptive weighted fusion algorithm is used to fuse the ultrasonic velocity obtained by six dry and wet alternation processes.The results showed that the ultrasonic speed and soil bulk density had a non-linear increasing relationship,soil bulk density had different effects on the ultrasonic speed with the soil of different texture.The change of silt soil bulk density had the greatest impact on the ultrasonic speed,followed by loam and clay loam;with the continuous decline of soil moisture content,the ultrasonic speed showed a fast and then slow down growth trend,among which the silt loam,loam and sandy clay loam the corresponding soil moisture content is about 20%,16% and 17%respectively when the ultrasonic speed increases and slows down.The relationship between the ultrasonic speed and the times of dry and wet alternation is analyzed by using the polynomial model.When the times of dry and wet alternation reaches the fourth time,the amplitude of the ultrasonic speed change is stable.⑶ The cultivated soil was selected as the research object.Four kinds of soil samples with different initial moisture content were treated alternately to obtain the ultrasonic velocity values of soil samples during the process from wet to dry.Through the analysis of the influence of soil moisture content changes on soil bulk density and ultrasonic velocity under the four kinds of dry and wet alternation treatment,the polynomial and support vector machine were used to analyze the influence of 4 kinds of soil moisture content changes on the soil bulk density and ultrasonic velocity.The relationship between soil moisture content and soil bulk density,ultrasonic velocity was analyzed under 4 different alternative treatments.The fitting analysis and regression training of ultrasonic velocity and soil moisture content were carried out by polynomial and support vector machine.A prediction model of soil moisture content based on ultrasonic velocity was established.The average relative errors of the prediction results were about 6% and 1.5% respectively.It shows that the prediction model of soil moisture content based on ultrasonic velocity constructed by support vector machine model can better describe the soil moisture content measurement of different soil moisture demand in the region.In order to provide a reference for the development of vehicle mounted soil moisture monitoring field by using ultrasonic detection technology.