Optimization And Experimental Verification Of The On-line Detection System For The Moisture Content Of Composting Materials

Livestock manure compost is an important way to utilize agricultural waste.It can effectively solve the agricultural non-point source pollution caused by livestock manure,realize the recycling of waste,and meet the requirements of low-carbon environmental protection and sustainable development.During the composting process,the fermentation effect of the composting microorganisms will be affected when the moisture content of the material is too high or too low,which will ultimately affect the composting decompose effect.Therefore,the on-line detection of the moisture content of composting materials is very important for timely getting the composting information,optimizing composting process parameters and controlling composting process.Compared with other on-line moisture content detection methods,frequency domain reflectometry(FDR)determines the moisture content by the dielectric constant of material,which has been widely used in soil moisture content detection due to its simple operation,fast,accurate,and continuous fixed point.Compared with soil,the bulk density and temperature of livestock manure during composing vary widely.The common range of bulk density is 600-800 kg/m~3and the temperature of livestock manure during composing is 20-70?,respectively.However,the existing FDR is designed according to soil characteristics.The dielectric constant and frequency domain dielectric change range of livestock manure composting materials differ from that of soil.Therefore,the existing FDR detects the moisture content of compost materials with large error and low precision.In order to solve this problem,based on the previous research results,the following research work were performed:The pig manure and straw mixture was mixed in a fermenter for composting test,and the fermenter's inner diameter is 60 cm and the height is 40 cm.The FDR moisture sensors with probe lengths of 5cm,10cm,15cm and 20cm were used to measure the moisture content of materials with different bulk density and temperature.The optimum probe length for measusring moisture content of compost material was determined by comparing and regression analysis between measured results and drying method results.Different levels of bulk density,temperature and initial water content of compost material were designed and the single-factor test of moisture content of material measured by FDR moisture sensor was carried out under the condition of no moisture evaporation.The deviation between FDR results and drying method results was compared,and the influence of moisture content,temperature,and bulk density on the deviation of FDR results was analyzed.A three-factor,three-level orthogonal test was designed by using the response surface methodology(Design expert).And a response surface model for FDR moisture content detection of materials with different moisture content at different temperatures and different bulk density was established.Furthermore,polynomial regression was performed on the calculated results of moisture content of the model of samples and the test results of drying method,and the optimized model of FDR moisture content detection was obtained.The optimization model was embedded in the compost water content online detection system and the test of compost water content detection was carried out.The optimization model was proved correct and the compost water content online system were reliable.