【作者】 王斌；

【导师】 杨盘洪；

【作者基本信息】 太原理工大学 ， 电路与系统， 2007， 硕士

【摘要】 步入21世纪，高度的信息化成了当今社会先进生产力的特点。而在我国广大的农村地区，由于经济条件和地域的限制，知识信息的获取和更新十分困难。广大农村地区对知识和信息的渴求，同我们的农业方面专家人才相对较少的矛盾，已经严重阻碍了当地的经济发展，进而制约了整个国民经济的发展。建立智能化农业示范区成为推动农村经济发展的迫切需求。在国家863计划的大力支持下，针对当前我省农业实践活中遇到的实际困难，我们研制了山西省智能化农业专家系统。本文以小麦为研究对象，详细的阐述了该系统的设计方案和实现细节，同时就设计涉及到的技术作了一些探讨。本系统由智能决策子系统和多媒体信息子系统两部分组成，综合集成了人工智能、现代多媒体技术和农业专门知识，并通过采用国际上流行的Browser／WebServer／Database三层分布网络体系结构，实现了高效的专家系统框架，这样的体系结构增强了系统的稳定性和可靠性，便于各种资源的管理和维护。系统采用Windows 2003 Server所提供的完全的IIS6.0中的WWW服务器，利用Dreamweaver 8、Photoshop 7、Flash 8、RealProducer和InterDev等诸多程序和多媒体开发工具对文本、图像、动画、音频、视频等信息进行有机集成，形成图、文、声并茂的Web页面，并充分运用了ASP.NET、VBScript、JavaScript、XML、ActiveX构件等新技术来强化系统的功能；采用基于模糊产生式规则的知识表示，混合推理控制策略以及应用不精确推理的可信度模型，较好地解决了农业领域知识的复杂性和不精确性，实现了网络化农业信息处理系统，解决了农业专家系统数据量庞大，多媒体信息甚多，数据的不确定性等难点，集实用性、方便性和科学性于一体，突出系统的本地化，在屯留、五台、忻州等试验点的推广应用中，得到了用户的广泛好评。此外，针对农业专家系统升级数据库时遇到的数据冗余、更改、添加以及删除数据异常的问题，本文深入研究了数据库模式层规范化理论。并根据专家系统数据库的特殊需求，总结出适合中小规模数据库使用的模式层规范方法，成功解决了农业专家系统更新知识库时的问题。最后，根据各试验点的反馈情况，对多媒体信息子系统进行了改良，并融合了智能决策子系统的部分功能。以期更方便、高效的指导农民对智能农业专家系统的使用。更多还原

【Abstract】 In the 21st century,the high information became the characteristic of the advanced productivity of our society.In the great deal rural areas of our country, receiving and renewing the knowledge information is very difficulty, because the restriction of the economic and region condition.One hand the rural areas need the knowledge and the information,and an other hand our agriculture experts are not enough.The contradiction has already obstructed the economy development seriously in the local place and restricted the whole development of our national economy. It is urgent to build up an intelligence agriculture demonstration area for the development of the village economic.In the strongly supporting of the nation 863 plan and aiming at the difficulty of our agriculture fulfillment in the Shanxi province, we developed the Intelligent Agriculture Expert System of Shanxi.This article takes wheat as a researching object, elaborating the design project and the detail of the system. At the same time, the article probes into some techniques.This system is made up by the intelligence decision sub-system and the multi-media information sub-system. It integrates the artificial intelligence, modern multi-media technique and specialized agriculture knowledge.We use three distributions computations system:Browser/WebServer/Database to design efficient system construction which improves the stability and credibility of the system,and makes easier to manage the various resources.The system uses WWW Server in IIS applied by Windows 2003 Server and adopted some exploiting tools such as Dreamweaver 8, Photoshop CS, Flash 8, RealProducer and InterDev to integrate the texts, images, animations, audios and videos to form very perfect web pages.In addition,we sufficiently used some new technology such as ASP.NET, VBScript, JavaScript, XML, ActiveX etc to strengthen the function of system. On the other hand, we used the expression of knowledge based on the RBR system, mixed reasoning control strategy and the reliable model of imprecise reasoning which resolved the complexity and imprecision in the domain of agriculture and constructed the information management system of naked oats. The system resolved many problems of agriculture expert system such as too huge data quantity, too large multimedia information and incertitude of data. So, the system was very popular when it was used in actual produce. Such as Tunliu, Wutai, Xinzhou.In addition, aiming at the abnormal problems of the data redundancy, changing, increasing and deleting when the Agriculture Expert System updates its database, this article goes deep into researching the theoretic of the database mode layer standardization. According to the special requirement of the expert system database, this article finds a way which is fit for the medium database to form the database mode layer standardization, and solves the problems successfully when it renews the knowledge base.At last, according to the feedback of each circumstance, we carried on improvement to the multi-media information sub-system and added part of the function of the intelligence decision sub-system,with expect to be more conveniently and efficiently to guide the farmers using the agriculture expert system.In the end, carried on improvement to the multi-media information subsystem according to each feedback circumstance that experiment a point, and blend parts of functions of the intelligence decision sub-system.With expect to be more convenient and guide farmer a friend this special community efficiently to the usage of the intelligence agriculture expert system.更多还原