The Research And Development Of The Key Tecgnology For Public Service Platform Oriented Precision Argriculture

Recently, China is in the transformation from traditional agriculture to modern agriculture. The mechanical, chemical, electrical and other technologies based "industrial agriculture" increases the production of agricultural products, which however, ignores the negative effects on environmental, resource and ecological aspects. Therefore, the current model is not suitable for the requirement of sustainable development. In terms of the actual demands in crop plantation and production management, this paper studies and designs a growth data acquisition and control system for economic crops. At the same time, an intelligent, network-based, practical modern agricultural technology oriented public service platform is implemented for the direct guide on crop plantation management, the diagnosis of pest and disease, the improvement of crop yields and quality, and the reduction of energy consumption and agricultural pollution. The detailed study work can be summarized as follows:1) Through the study of networking technology and the comparison of the four kinds of common used wireless network technology, we propose a design scheme for agricultural greenhouses oriented environment detection system. The scheme is based on agricultural real-world environment and requirement, and considers power loss, signal transmission distance and other factors. In addition, in the scheme, ZigBee is selected as the networking protocol and the topology style which combines centralization and transmission modelis designed. Furthermore, anacquisition and control system for crops growth, which includes man-machine interface and the master program, is constructed.2) By reading a large number of farming techniques, various types of agricultural expert systems, as well as the basic knowledge on fuzzy mathematics, artificial neural network, database and software development, we establish a tomato-based four kinds of economic crops expert system with SQL Server2010. In addition, we focus on the diagnosis on the diseases and insect pests which crops suffer, and analyze/explain the characteristics of the diseases and insect pests. According to the above theoretical basis, we design a fuzzy neural network based diagnostic system, which utilizes the improved BP neural network algorithm (adding momentum factor and learning rate) for network training. 3) Taking advantages of design patterns and object-oriented technologies, we develop an intelligent public service system which uses Java as the program language, MyEclipse as the development tool and SSH as the development framework. The system efficiently combines the crop acquisition and control systems, pest diagnostic systems and integrated crop management system. The interface of the system is simple and beautiful while the operation of the system is simple. The system users can do all kinds of access/query operations and can conduct pest diagnosis without special training and professional knowledge on agriculture and computer.The achievement of the implemented system has been applied in ma ny fruit and vegetable planting companies. The trial results show that the system can work reliably and stably.