Development And Research Of Intelligent Water And Fertilizer Decision-making System For Maize Based On Neural Network

As an important grain crop in China,maize is widely planted in Xinjiang.At present,in the process of maize production in Xinjiang,there is a general lack of professional theoretical guidance,agricultural producers can only rely on experience to explore the production law or copy the production mode of other areas.Due to the differences of individual ability of agricultural producers and regional differences between different areas,it is difficult to maximize the benefits of maize production.Based on the problems existing in the process of maize production in the soil and water improvement experimental field of the Department of water resources of Xinjiang Uygur Autonomous Region,this paper carefully analyzed the actual demands of relevant agricultural workers in Northern Xinjiang,designed and implemented an intelligent water and fertilizer decision-making system for maize based on neural network.In the process of system development,this paper followed the software development specifications and successively completed the work in the software life cycle,completed the data management module,model management module,information interaction module,knowledge management module,account management module and system management module in the system,and put forward a reliable maize yield prediction model in the model management module.The main research contents and achievements are as follows:(1)By consulting the relevant data of agricultural intelligent decision-making system and investigating in maize production areas,this paper defined system problems,analyzed the feasibility of system development in economy,technology,operation and society,expounded the functional requirements of the system with the help of system function use case diagram,explained the non functional requirements of the system from the aspects of system performance,reliability,safety,usability,scalability and maintainability,established the architecture of the system and designed the operation flow and database structure of each functional module of the system.(2)Aiming at the defects of the Sparrow Search Algorithm,this paper put forward some improvements,and analyzed the performance of several intelligent optimization algorithms through simulation experiments,which proved that the Improved Sparrow Search Algorithm has better optimization performance.The Improved Sparrow Search Algorithm was combined with neural network technology to construct the model,and the model was used in the simulation experiment of maize yield prediction in the soil and water improvement experimental field of the Department of water resources of Xinjiang Uygur Autonomous Region.By analyzing the performance of several different models in simulation experiments,it was proved that the maize yield prediction model based on Improved Sparrow Search Algorithm and Radial Basis Function Neural Network technology has better prediction performance.(3)According to the system architecture and the operation flow design of each functional module of the system,the business logic was realized,and the maize yield prediction model proposed in this paper is embedded into the system model management module.Finally,the system test scheme was designed to test the function and non function of the system.The test results showed that the system function and performance are in line with expectations and needs of users.The intelligent water and fertilizer decision-making system for maize developed in this paper has reasonable structure design and clear business logic,which has a certain value for relevant agricultural workers in Northern Xinjiang,and provides a reference solution and model for relevant system software developers.