土壤湿度传感器影响因素及补偿方法研究

发布时间：2018-04-15 23:14

  本文选题：FDR土壤湿度传感器 + 温度补偿　； 参考：《河北工业大学》2015年硕士论文

【摘要】：植物生长在土壤中,时刻需要土壤水分的供给,研究土壤的水分状况对植物的生长具有重要意义。土壤湿度传感器能实时监测土壤含水量,越来越多的土壤湿度传感器被用在农业种植领域。土壤湿度传感器的种类有多种,FDR土壤湿度传感器具有使用简单、实时性好等优点,已广泛应用在土壤含水量检测上。但是根据前人研究和用户反应,FDR土壤湿度传感器在使用过程中易受温度、土壤硬度等因素影响,如果不对这些影响因素进行补偿,势必会对最后的测量结果产生影响,影响正确判断土壤含水量的多少。在分析与研究了各种影响因素以后,本文重点研究了温度对FDR土壤湿度传感器的影响,采用二元回归分析法和BP神经网络法对其进行了温度补偿,并利用Matlab软件进行了仿真。补偿前传感器的灵敏度温度系数为3.9×10-3/℃,经二元回归分析法对传感器进行温度补偿以后,传感器的灵敏度温度系数下降为1.3×10-3/℃;经BP神经网络法对传感器进行温度补偿以后,传感器的灵敏度温度系数下降为5.85×10-4/℃。由此可以看出,经两种方法对FDR土壤湿度传感器进行温度补偿以后,其灵敏度温度系数有所降低,表明温度稳定性有了一定程度的提高。本文还对二元回归分析法和BP神经网络法的补偿效果进行了对比,经过对比发现,二元回归分析法的平均补偿误差为0.1224%,BP神经网络补偿法平均补偿误差为0.0548%,要比二元回归分析补偿法的低一个数量级,且BP神经网络法的补偿误差要比二元回归分析法的补偿误差分布要均匀。
[Abstract]:Plant growth in the soil, always need soil moisture supply, the study of soil moisture status is of great significance to plant growth.Soil moisture sensors can monitor soil moisture in real time. More and more soil moisture sensors are used in the field of agricultural planting.The kinds of soil moisture sensors have been widely used in soil moisture detection for their advantages of simple use and good real-time performance.However, according to previous research and user response, FDR soil moisture sensor is easily affected by temperature, soil hardness and other factors in the process of use. If these factors are not compensated, it will certainly have an impact on the final measurement results.The effect of water content on soil water content is correct.After analyzing and studying various influencing factors, the effect of temperature on FDR soil moisture sensor was studied in this paper. The temperature compensation was carried out by using binary regression analysis method and BP neural network method, and the simulation was carried out by Matlab software.The sensitivity temperature coefficient of the sensor is 3.9 脳 10 ~ (-3) / 鈩，

本文编号：1756243