基于真有效值检测的高频电容式土壤水分传感器

摘要

土壤水分的测量是精细农业中实施节水灌溉的基础.基于真有效值检测技术,利用土壤的介电特性,设计了一个高频电容式土壤水分传感器,主要由电源滤波电路、100 MHz有源晶振、XC74UL14AA、探针电极和AD8361组成.其中,探针电极由印刷电路板制成并与主印刷电路板一体化成型.传感器以直流电压输出,分别在空气和去离子水中测得其输出范围为工作电压的20％～70％.通过2-异丙氧基乙醇、二氧六环和去离子水3种溶液配制了一系列不同等效土壤体积含水率的待测溶液,在不同工作电压下,对传感器进行了标定以及在5～40℃范围内以24.8℃基准进行了温度变异性试验.试验结果表明:传感器对工作电压有明显的依赖性,在特定含水率下传感器的输出电压随工作电压的升高而增加;特定工作电压下传感器的输出电压与土壤体积含水率呈线性负相关,其决定系数R2＞0.987;温差越大则传感器的测量偏差也越大,最大偏差为4.44％.并配制土样对传感器进行了验证,最大误差为4.95％.%Measurement of soil water content is the basis of implementing water-saving irrigation in precision agriculture. A high-frequency capacitive soil water content sensor which is consist of power supply filter circuit, 100MHz oscillator, XC74UL14AA, probe electrode which is made up of PCB (Printed Circuit Board) and is part of the whole PCB, and AD8361 was developed based on True RMS detection technology, by using the dielectric properties of the soil. The output of the sensor is in form of DC voltage, and the output range is 20%-70% of supply voltage, measured in air and deionized water respectively, under different supply voltages. The sensor was calibrated, and the temperature variation of the sensor was tested in the range of 5-40 ℃ according to the basis of 24.8 ℃, in a series of tested solution with different equivalent soil volumetric water content, under different supply voltages. The tested solution was prepared through 2-isopropoxy ethanol, dioxane and deionized water three kind of solution. The results show that: there is a significant dependency of the sensor's output on the supply voltage, the output voltage of the sensor increases with the supply voltage in specific soil volumetric water content; the out voltage of the sensor has a negative linear correlation with soil volumetric water content in specific supply voltage, and the coefficient of determination R2>0.987; the greater the temperature difference, the greater the measurement error of the sensor, and the maximum error is 4.44%. Verification of the sensor was done through preparing a series of soil samples, and the maximum error is 4.95%.