【作者】 王明霞；

【导师】 蒋文科；

【作者基本信息】 河北农业大学 ， 农业机械化工程， 2004， 硕士

【摘要】 保护地蔬菜栽培是一项技术含量很高的农业产业，尤其近几年人们对蔬菜产品质量要求的提高，促使菜农对蔬菜栽培知识的需求日益迫切。同时，保护地蔬菜栽培中的环境控制、数据采集等计算机控制系统，也由简单的以数据采集处理和检测为主，逐步转向以知识处理和应用为主，并开始向智能化、网络化方向发展。如何为这些计算机系统提供及时、可靠并能由计算机系统自动识别的专家知识也成了目前急需解决的问题。 Web服务是自包含、模块化应用程序，它能够通过网络，通常是互联网，描述、发布、定位和调用，具有封装性、松散耦合、使用标准协议规范、高度可集成能力的特点。它为异构计算机系统集成提供了可行的解决方案。 本文将Web服务技术与专家系统技术结合起来，在分析了现有蔬菜栽培专家系统现状及缺陷的基础上，设计并实现了基于Web服务的保护地蔬菜栽培专家系统。系统采用了分层设计思想，将系统分成数据层、数据访问层、业务逻辑层、Web服务层和用户交互层，通过上层对下层的调用实现系统提供的Web服务，从而提高了系统的可维护性和代码的复用率。系统根据蔬菜栽培知识的内容不同分别采用了树形层次语义网络和因果网络模型两种知识表示形式，并运用了不精确混合推理策略，使得系统推理更快更准确。同时，本文给出了保护地蔬菜栽培专家系统Web服务的封装过程及调用方法，以及Web服务的描述、测试及注册流程。最后，本文介绍了蔬菜栽培知识表示规范化和XSL样式单描述，以及系统的安全性分析。更多还原

【Abstract】 Protected horticulture vegetable planting is an agricultural industry with high technology. In recent year, it urges farmer to obtain more knowledge about vegetable planting that people want high quality vegetable products. At the same time, computer control systems in horticulture vegetable planting, such as environment monitoring and data collection, develop from simply gathering data and checking it into managing and applying knowledge with intelligence over a network. Therefore it becomes an urgent problem that how to provide timely and reliable expert knowledge for these computer systems.Web services are self-contained, modular applications that can be described, published, located, and invoked over a network, generally the web. It offers the prospect of creating distributed, loosely coupled applications, using standard protocols and infrastructure. It provides a practicable scheme for heterogeneous computer system integration.This paper combines web services with expert system, designs and implements protected horticulture vegetable planting expert system based on web services, after analyzing the status and limitation of current vegetable planting expert system. It introduces layered method, and the system is divided into data layer, data access layer, logic layer, web services layer and user interaction layer, provides web services through calling from super layer to under layer. In this way, it’s easy to maintain the system, and it also can improve the reusability of the code. Our system adopts tree hierarchy semantic network and causal network model according to different knowledge content, applys inexact mixed reasoning to enhance the veracity of the result. This paper also offers the encapsulation procedure and calling method of protected horticulture vegetable planting web services, as well as web services’ description, testing and registration. At last, this paper presents the normalization of vegetable planting knowledge representation and XSL style sheet description, as well as system security analysis.更多还原