Research And Implementation Of Smart Irrigation Management System Based On Big Data Analysis

Water shortage is one of the most serious problems in the world.Water resources are mainly consumed in the agricultural production process,of which farmland irrigation accounts for a large proportion.Therefore,the key to solve the problem is to develop a smart irrigation system to realize the reasonable water resources utilization.Based on the project of modern demonstration irrigation district in Dujiangyan,this thesis combines big data analysis technology and machine learning algorithms to develop a smart irrigation system that can achieve reasonable water resources utilization.In this thesis,the collected soil moisture data and meteorological data are analyzed by machine learning algorithms to predict soil moisture in the next three hours.The soil moisture prediction model is established by classification and regression tree,auto regressive integrated moving average model and multiple linear regression.The error analysis results show that using classification and regression tree algorithm is better,and it is applied to the smart irrigation system.The device automatically adjusts its state based on predicted soil moisture,and the system automatically generates an irrigation water distribution plan to achieve intelligent irrigation.In this thesis,the system is divided into six modules: field equipment management,pipeline water metering monitoring,irrigation data display,real-time video monitoring,comprehensive management and smart irrigation analysis.The field equipment management module includes real-time equipment control,timing control,automatic control and equipment status monitoring.Pipeline water metering monitoring module includes remote operation of pipeline valves and real-time monitoring of operating parameters.The irrigation data display module mainly displays the collected data.Real-time video monitoring module monitors the status of irrigation areas in real time.The comprehensive management module mainly manages the users,permissions,basic data,logs and other basic information in system.Smart irrigation analysis module includes soil moisture prediction and automatic generation of irrigation decisions.These modules work together to enable smart irrigation.The smart irrigation system uses the B/S architecture,and information is released through multiple channels such as websites,mobile applications,WeChat official accounts and so on.The back-end used Spring Boot framework,and the front-end uses Vue.js framework.HBase,MySQL and Redis are used as database storage in the system.The HBase database stores historical big data for big data analysis in smart irrigation analysis module.MySQL database stores real-time monitoring data and business data.Redis acts as a cache to ensure system response speed.The research in this thesis can improve the utilization of irrigation water and achieve effective control of irrigation water.The smart irrigation system is already in use.It can predict soil moisture in real time and plan irrigation water consumption rationally based on soil conditions.It can save water resources while maintaining the output of subsidiary agricultural products,which has certain practical significance.