| China is a country with severe water shortage.In recent years,irrigation water consumption accounts for about 65%of the country's total water supply.At present,the water use of agriculture is inefficient,resulting in an increasing shortage of water resources.It is an urgent task for modern agriculture to develop precision agriculture through water-saving irrigation.Soil moisture sensor can reflect the change of soil moisture efficiently,which is beneficial to the implementation of water-saving irrigation.Currently,the high-precision soil moisture sensor on the market has the problem of high price generally,which cannot achieve the purpose of large-scale distribution to obtain moisture information.At the same time,the low-cost moisture sensors mostly use low-frequency capacitance measurement method.However,due to soil salt,pore density,soil quality and other disturbance factors,this method cannot meet the needs of precision.In order to complete the sensing system with low cost and meet the requirement of agricultural moisture detection,this paper introduces conductivity and temperature parameters to the modified algorithm with a large number of experimental data based on the traditional capacitive soil moisture detection method.The theoretical model of three-parameter moisture detection is established,and the corresponding sensing system is developed,which is applied to the site of Yulin Ecological Experiment Station in Shaanxi Province.So,it can carry on the multi-point continuous detection to the field soil moisture for a long time without monitoring.According to the principle that the intermediate electric constant of soil with different moisture content is different,the relationship between water content and capacitance,conductivity and temperature parameters is studied,and the measuring method of water content is defined.The structure and circuit design of the sensor probe are realized,and the optimal detection model algorithm is compared and selected.In addition,a set of water sensing system with primary and sub-node distribution is completed according to the detection requirements.The designed detection system is applied to the field experiment,and the system is further improved in allusion to the problems.The experimental results prove the feasibility of the scheme.In the research of the whole sensing system,the following work has been accomplished:1.This paper analyzes the principle of soil moisture detection with dielectric method,discusses the main interference factors of soil capacitance detection in low frequency section,such as conductivity and temperature,and deduces the calculation formula of low frequency capacitance moisture with conductivity and temperature correction.Different kinds of soil samples were tested and analyzed,and the application scope of the method was obtained.2.According to the common measuring methods of capacitance and conductivity,the structure design and peripheral circuit design of capacitance parallel coplanar electrode probes and the two-electrode conductivity probes are carried out respectively.The MPR121 touch capacitor chip is used to detect the low frequency equivalent capacitance.According to the frequency characteristics of capacitance detection,the optimal capacitance measurement parameters are obtained by experiments,such as charge current 63?A,charging time 4?s,and effective capacitance detection range from 200 to800pF.Moreover,through the calibration experiment of conductivity in low salinity soil,the formula of conductivity conversion is obtained,and the correction determining coefficient R~2 is 0.9872.3.We use the drying method to calibrate the soil sample,and obtain the empirical formulas with the different mathematical models of logistic,exponential,polynomial.The corresponding moisture measurement algorithms for three models were gotten according to the moisture detection methods.The root mean square error(RMSE)was used to compare its moisture detection performance,and logistic was selected as the best detection model.Finally,the repeatability experiment and error analysis of the logistic model algorithm are carried out.It is proved that the maximum absolute measurement error is less than 3.0%and the average absolute measurement error is less than 2.0%.4.We design and complete the sensing system with function modules such as power supply,control,storage,clock,wireless communication and so on.Through the design of the main sub-node of Zigbee,one-to-many multi-point acquisition and control functions are realized,in the hardware design and software design.It focuses on the problems of the power consumption of the sensor system,improving the efficiency of data transmission,the transmission of electricity alarm information,and so on,to improve the practicability of the sensor system on the spot.In order to facilitate the debugging of the sensing system,real-time detection and determination of the position information of all sub-nodes,a portable soil moisture collector with human-computer interface and GPS positioning function was designed by using STM32 controller to perfect the use of the sensing system.5.In order to further verify the performance of the system,the capacitive sensor developed in the Yulin Ecological Experiment Station in Shaanxi was connected to the wireless sensor network,and a four-month moisture detection experiment was carried out to analyze the initial application of the sensor.Due to the problem of power consumption,data loss in wireless transmission and debugging,we optimized the system and eliminated the data error when the packet loss rate was below 15%.In addition,the performance of the moisture sensing system is evaluated by comparing the measured data with the weather. |
PDF Full Text Request