Font Size: a A A

Design And Experimental Study Of Grain Humidity Sensor

Posted on:2020-01-05Degree:MasterType:Thesis
Country:ChinaCandidate:R Y WangFull Text:PDF
GTID:2393330623456547Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
Food is the main resource for national survival and national development,and it occupies a very important position in the development of the national economy.Cereal crops are a large category of food.The water content is an important indicator to measure the quality of cereals.The humidity of cereals is too high,which may cause problems such as mildew and insects.If the humidity is too low,the quality of food will be affected.Therefore,it is very important to realize rapid and non-destructive testing of humidity in the transportation,storage,trading and processing of grain.At present,grain moisture measurement based on dielectric properties is widely used in production,but the dielectric parameters are not only closely related to humidity,but also the frequency of excitation signals,grain packing density and external temperature will affect it.The influence of factors on measurement accuracy and repeatability is a key issue that must be addressed in the rapid detection of grain moisture.Based on the principle of dielectric property measurement,a fast grain moisture measuring instrument with sweep impedance measurement function was developed.Based on this,a two-channel one-dimensional convolutional neural network was proposed to reconstruct the grain moisture to reduce the grain moisture.Measurement error due to uneven packing density.Firstly,in order to reduce the influence of system error on measurement accuracy,the structural optimization design of the coaxial probe for grain moisture measurement was carried out,and the sweep amplitude and phase angle sweep measurement were realized based on the high precision impedance conversion chip AD5933.The grain moisture sensor is used to study the effect of grain packing density on humidity measurement in a laboratory environment.A two-channel convolutional neural network is proposed to model the grain moisture,and the sensitive frequency segment is selected through data analysis.Electrical data is used as a data source for training and testing.The results of the performance test show that the designed sensor can effectively reduce the influence of grain packing density change and uneven filling on the grain moisture measurement result,and realize the rapid non-destructive detection of grain moisture.
Keywords/Search Tags:grain moisture, dielectric properties, sweep impedance measurement, convolutional neural network
PDF Full Text Request
Related items