| With the continuous advancement of the modern agricultural construction in China,the application of the agricultural Internet of Things in facility agriculture is expanding.With the vigorous development of the country,the number of terminal devices in agricultural Io T systems has gradually increased.Agricultural Io T systems that rely more on cloud computing models are facing delays,increased computing and storage pressures.Simultaneously,the agricultural Io T systems in China have gradually matured in environmental monitoring and remote control now,but it is still not perfect in early warning and prevention of facility agricultural diseases,the early warning and prevention mechanism is poor in timeliness and automation capabilities is inadequate.Based on the above needs and problems,through the study of the edge computing model,this dissertation applies the edge computing technology in the agricultural Io T,and designs a simple and effective early warning system for facility agricultural diseases.In this dissertation,by studying the triangle theory of plant diseases and disease control technologies,taking cucumber as an example,a mechanism for early warning and prevention of agricultural diseases based on environmental parameters prediction is designed.After collecting environmental parameters,the system predicts the environmental parameters at a future time point,and gives different levels of alarms according to the range of early-warning environmental parameters.When serious alarms occurs,the ecological prevention and control decision of temperature and humidity adjustment will be made automatically.Among them,the environment prediction model uses a time series analysis method,and a PSO-SVR model is established based on a sliding window.At the same time,through the study of the edge computing model,in the aspect of agricultural Io T and edge computing applications,an application scheme that decentralizes some computing and storage tasks to the field control platform is proposed to optimize the cloud computing model of the agricultural Io T.And summarized the advantages of combining agricultural Io T with edge computing.Based on the above research,this dissertation designs a system for early warning and prevention of facility agricultural diseases based on a three-tier model of the edge computing reference architecture.The system is divided into a physical layer,an edge computing layer,and a cloud service layer.The physical layer is close to the real physical world,is mainly sensor networks and actuator,and uses Wi Fi for data exchange with edge computing layers.The field master controller in the edge computing layer is the core of the entire system,and is responsible for the main functions of system network connection,data processing,and local storage.The cloud service layer is mainly responsible for long-term storage of user information and daily average environmental data.This article details the development of a field master controller.Through research and comparison,this article uses embedded devices equipped with the Android operating system as the field master controller.Based on Android Studio and MVP development mode,it uses Java network programming,SQLite programming and other technologies,as well as libraries such as MPAndroid Chart,Libsvm to design and implement the required function module of the field control station,the control terminal APP realizes the functions of data transmission and reception,visual display,parameter prediction,local storage and so on.Finally,the communication protocol was written,and the functional test was finished by using the black box method.The test showed that each functional module meets the design requirements and can run well. |
