| In the context of modern agriculture,precision agriculture has become a symbol of modern agricultural information.It provides accurate management and accurate analysis of large-scale structured or unstructured agricultural data,which provides a scientific basis for intelligent aided decision making.The agricultural information system as a data carrier can help people acquire and store relevant information about plants,agricultural information system,as a tool,can serve agricultural information and precision agriculture to a large extent.Based on agricultural informatization,precision agriculture and data mining,the system has developed an agricultural information system based on SSH framework.The system is divided into six functional modules,including the environment information of the greenhouse,the plant growth information,the statistical and prediction analysis,the real-time monitoring of the greenhouse,the information of the greenhouse equipment and the system management.The environment information of the greenhouse is mainly responsible for managing the daily temperature and humidity,lighting conditions and soil conditions,etc.The plant growth information is mainly responsible for the management of the information of irrigation,fertilization,pest control,etc,in the plant growth process of the greenhouse.The statistical and prediction analysis is mainly in the form of combination figure display historical data,such as temperature,humidity,light,etc in the greenhouse,and showed in the form of statistical figure after GA-BP algorithm to predict classification level of plant diseases and insect pests.The real-time monitoring of the greenhouse is mainly used in real-time display of various indicators in the greenhouse in the form of dynamic diagrams,and plays the role of monitoring and early warning.The information of the greenhouse equipment is mainly used for boarding and management of related equipment and tools in the greenhouse.The system management is mainly responsible for managing personal information and permission distribution of users,etc.The cloud database adopts open source relational database MySQL,and the statistics and analysis of key data are visualized through Hightcharts.Data acquisition and transmission are based on sensors and wireless sensor networks.The cloud platform plays the role of "intermediary" in the Web system and data acquisition and transmission.It can not only store and manage the data collected by the sensor,but also can extract data for analysis.The main work of this paper is as follows:1?Integrate the struts2 framework,Spring framework,and Hibernate framework.Each framework plays a different role in a Web system,so the three are coupled together to achieve the basic information management functions of a Web system.2?The prediction classification algorithm takes the cucumber downy mildew and gray mould as an example,aiming at environmental factors and the level of plant diseases and insect pests in the greenhouse is nonlinear relationship,this paper uses the BP neural network to classify the grade of the disease and insect pests,and revises the weights of the neural network through the error,so that the forecast value of the output is close to the real value,to achieve the purpose of predicting classification.3?In view of the disadvantages of BP neural network,including slow convergence speed and easy to get into local minimum,this paper uses genetic algorithm with strong global searching ability to optimize BP neural network.Due to the population "precocious phenomenon" of genetic algorithm,this paper uses the adaptive crossover probability and adaptive mutation probability to optimize genetic operators.The experiment shows that the predicted value output by BP neural network can basically fit the real value of training samples,but the accuracy is not high.After the weight optimization of BP neural network by genetic algorithm,the prediction precision of BP neural network is improved obviously. |
PDF Full Text Request