Research On Detection Technology Of Sludge Moisture Content Based On Microwave Transmission

In the process of sludge drying and dewatering,the moisture content can meet the requirements for subsequent treatment and disposal of sludge.However,at present,manual detection of moisture content is used in industry in China.This method has many instruments,poor portability and long detection time.In addition,the research on the detection of sludge moisture content is basically blank,so the detection technology of sludge moisture content based on microwave transmission is studied.The specific contents are as follows:The research status and existing problems of microwave moisture content detection technology at home and abroad were investigated.The mathematical model of sludge moisture content and dielectric constant was deduced,and the effect of the model was discussed.When the sludge moisture content is more than 40%,the model prediction effect is not accurate.The measurement model of sludge moisture content based on microwave transmission is analyzed.It is proved theoretically that the moisture content of sludge can be obtained by detecting the microwave attenuation,which establishes a theoretical basis for the following experiments.The horn antenna was designed and optimized,and the sludge moisture content detection experiment was carried out.It has been proved that the initial dehydration pressures of mud cakes are 2MPa,4MPa and 6MPa,and the drying temperatures are80°C,100°C and 120°C,which have no effect on the test results.The S₂₁ parameter fluctuates greatly when the moisture content of sludge cake exceeds 40%,and the fluctuation disappears after slicing the sludge.When the initial volume of the mud is16cm³,6cm³ and 4cm³,the S₂₁ parameters have three trends with the change of water content.The S₂₁ parameters of the three kinds of mud are fitted with the water content,and the detection effect is poor when the net water content of the mud is less than 1.5g.Therefore,the test sample requirements:for the mud cake of mechanical filtration,the thickness needs to be reduced;the net water content of the sludge treated by the cutting machine cannot be less than 1.5g.Multiple linear regression model,Support Vector Machine model and BP Neural Network model were used to establish S₂₁-quality measurement model for sludge blocks with initial volume of 16cm³ and 6cm³,respectively.The model R² were 0.94,0.996 and 0.987,respectively.The root mean square errors were 7.47,4.512 and 5.518,respectively.Due to the nonlinear problem in the modeling process and the difficulty of sample preparation for modeling,support vector machine model is the optimal model.A moisture content detection system based on voltage-quality sludge was designed.The hardware construction,software preparation and host computer design of the device were completed and the system was tested.The moisture content of the sludge block with the mass of 5g～21g and the moisture content of 20%～70%was detected.The average relative error was 1.5%,and the detection time of a single sample was 30s and the detection effect was good.