Research On Remote Monitoring System Of Soil Organic Matter Based On Artificial Olfaction

Food production has a crucial influence on China,a big agricultural country.The content of organic matter in soil is an important element affecting grain yield.How to monitor the content of organic matter in farmland soil is an important issue in current research.At present,a lot of research results have been made in the field of monitoring soil organic matter content at home and abroad,but most of them are collected by manual on-site collection of soil samples using chemical methods,or soil organic matter content data collected by testing instruments on the spot,lack of intelligent automatic Remote monitoring system for collecting data.In order to achieve the purpose of remote monitoring of soil organic matter content in large-scale farmland,this paper proposes a remote monitoring system for soil organic matter based on artificial smell.At first,the article first analyzes the research status of artificial olfactory technology and soil organic matter monitoring technology at home and abroad,and summarizes the current problems in remote monitoring of soil organic matter in China,and determines the key technologies and overall framework of the remote monitoring system for soil organic matter.According to the overall framework of the system,the monitoring system is divided into three parts: soil organic matter data collection terminal,remote server data storage terminal and system monitoring platform.Then,the hardware architecture of the data acquisition terminal was established,and the hardware parts such as the gas sensor array,core controller,remote data transmission module,and power module were selected.And use Altium designer software to design the peripheral circuits of each module in detail.The STM32F103ZET6 chip was selected as the core controller,and the "solar and lithium battery" power supply method was used to provide a continuous power support for the monitoring system.The keil5 software was used to complete the programming of the data acquisition terminal.Next,the content of soil organic matter in 40 monitoring sites in Dunhua City,Jilin Province was studied.The method of pattern recognition is used to classify and identify the soil odor data of sampling points,and the chemical method is used to detect the content of soil organic matter to complete the calibration work of soil organic matter data.Compared with the effect of MEA-BP,GA-BP and single BP pattern recognition methods on the prediction results of soil organic matter content,the simulation experiment shows that the prediction stability of MEA-BP model is the best and the average is the best.The minimum square error(RMSE)and the determination coefficient(R~2) are closest to 1,where RMSE is 7.80 and R~2 is 0.9268.Furthermore,the article designs the monitoring platform of the soil organic matter remote monitoring system.The system monitoring interface is written using LabVIEW programming software.The user login module,user management module,data query and display module and other functional modules are drawn.Data communication with the database,using MATLAB scrip nodes to call soil organic matter prediction models,to further improve the system monitoring accuracy.Finally,the system’s data acquisition terminal was built on-site,and the working stability of the hardware and software parts of the system and the accuracy of the prediction of soil organic matter content were tested under specific test environments.The whole system can run stably in the complex farmland environment,with an average monitoring accuracy of 93.2%,which can meet the actual production needs.