Intelligent Monitoring System Of Greenhouse Based On Internet Of Things

In this paper,in view of the fact that the intelligent and information level of the facilities agriculture in China lag behind the developed county.This paper built an overall architecture composed of the perception layer,the transport layer and the application layer according to the hierarchy of the Internet of Things,then analyzed the actual needs of the system,and determined the environmental parameters that the system needs to monitor,specified the principles and objectives of the system design,designed the greenhouse remote monitoring and intelligent management system.This paper mainly completed the following work:(1)In terms of hardware: Completed the circuit design and PCB design used the STM32 microcontroller as the core control,wrote program of data acquisition,realized remote monitoring and control of six kinds of environmental parameters: air temperature,air humidity,CO2 concentration,light intensity,soil temperature,and soil moisture content.The core collecting board sent the data to DTU through RS-232 serial port communication,DTU used GPRS network to send the data to the data center;introduced the data format in the process of transmission;designed a device that can self-adjusting the height of the test point to realize the three-dimensional mesh measurement of the monitoring system.(2)In terms of database: The paper designed and development the database of remote monitoring and intelligent management system based on MongoDB database.Based on the subsystem of greenhouse Web management information of the B / S structure,users can view real-time data and historical data of greenhouse through web browsing,and can export data according to demand;The administrator queried user information,sensor information and the working status of the control mechanism,and set the threshold of greenhouse environmental parameters to realize the remote control of the control mechanism.(3)In terms of temperature prediction: Aiming at the accuracy and efficiency of temperature prediction,this paper proposed PCA_PSO_LSSVM algorithm to predict the temperature of greenhouse.In this algorithm,Principal Components Analysis(PCA)was used to select the main factors that affect the temperature in the greenhouse.Particle Swarm Optimization(PSO)was used to optimize the parameters of the Least Squares Support Vector Machine(LSSVM)model.The nonlinear prediction model between temperature and its influencing factor was constructed by using the optimal hyperparameter combination),(r?.The experimental results showed that compared with the PCA_LSSVM prediction model and the PSO_LSSVM prediction model,the PCA_PSO_LSSVM prediction model had a good prediction effect.The evaluation indexes of the model: the root mean square error,the average absolute error and the decision coefficients were 0.5663%,0.0298 and 0.97032 respectively,which were better than other predictors methods.The greenhouse remote monitoring and intelligent management system has been successfully applied in Tianjin Binhai tea grape science and Technology Development Co.Ltd.The actual operation results showed that the system runs stably and could comprehensively collect six kinds of environmental factors with high accuracy;the system realized the regulation of environmental factors;data transmission process was reliable,and the operation of the Web interface was simple;...