北方玉米专家系统知识库的构建

阐述了玉米专家系统的总体设计思路和知识库的构建方法,采用＂Web浏览器/Web服务器/数据库系统＂三层分布计算机体系结构,应用模糊产生式规则的知识表示方法和不精确推理方法,较好地解决了东北玉米生产知识的复杂性和不精确性。     

An New Hybrid Model of Qualitative Diagnostic Reasoning

Two main approaches are used in qualitative diagnostic reasoning.These approaches are the first principle approach and the heuristic knowledge approach.Systems based on the first one can handle hard problems and new situation.On the other hand,systems based on the second perform diagnostic tasks faster but can deal only with the most common problem cases.This paper provides a model using both the deep and shallow knowledges, and presents some algorithms.     

基于知识工程的玉米果穗剥皮装置设计

收获机械结构复杂多样,使用季节性强,且用户多样性、定制化需求特征明显,传统研发模式存在设计周期长、效率低和质量难以保证等问题。本文以玉米联合收获机果穗剥皮装置为研究对象,根据剥皮装置结构特征、技术参数和性能评价指标之间的关系,提出了基于知识工程的玉米果穗剥皮装置设计方法。首先明确剥皮装置设计流程,制定模块化设计方案,按照功能划分为专用件模块、通用件模块和标准件模块,其中专用件模块为剥皮装置核心组成部件,主要包括剥皮辊、压送器,通用件模块包括喂入辊、输送机构、排杂器、传动机构和果穗回收机构等,标准件模块包括传动件、连接紧固件和轴承等。然后按照标准、规范和约束范围,建立剥皮装置相关设计知识库,分析玉米品种特性、作业形式、传动方案、结构参数和工作参数之间的数学关系,同时利用框架式表示法对剥皮装置进行分解,建立自顶向下的谱系层次结构。基于果穗运动学和动力学分析,融合文献资料、试验数据和专家经验,建立了剥皮装置工作性能评价模型,包括苞叶剥净率评价模型、籽粒损失率评价模型和籽粒破碎率评价模型。基于Visual Studio平台,融合知识库、推理机、评价模型和系统人机界面,开发了基于知识工程的玉米剥皮装置设计系统,实现用户需求参数输入下设计参数的实时计算输出及参数评价。基于上述研究,以TPJ16型玉米果穗剥皮装置参数为例,在交互界面输入功率7.5kW、喂入量16.6t,计算获取剥皮装置关键结构参数和运动参数,并进行设计参数的性能评价,求解结果表明该剥皮装置的苞叶剥净率为96.01%,籽粒破碎率为1.42%,籽粒损失率为3.25%。     

基于GIS的退耕还草专家系统的研制与开发

适宜草种的选择在退耕还草决策中起着重要作用, 依据牧草生物学特性和牧草与诸气象因子之间的关系确定牧草的适宜分布区间能够为退耕还草工作提供决策支持。该研究利用地理信息系统（Geographic Information System，GIS）和牧草生长适宜度模型相结合，采用模糊逻辑推理机制开发了退耕还草决策支持系统，用户可以查询到甘肃省任意一点的气象信息、土壤信息和草原类型，咨询到适宜种植的牧草品种及其栽培管理技术，系统还将22种牧草的适应性分布成图。样本运行结果表明，系统具有较好的决策功能，可以为甘肃省的退耕还草工作提供决策支持。     

基于证据可信度不确定推理的大豆病害诊断方法

针对大豆病害专家系统中诊断特征条件的复杂性和不确定性等问题,将病害证据可信度和不确定性推理的传递算法有机结合,提出了一种基于证据可信度不确定推理的大豆病害诊断方法。由植保知识库的专家先验知识和作物病害发生时期、部位和症状等表现特征,构成可信度权重作为推理依据;利用该文提出的基于证据可信度的不确定推理方法及推理规则,实现了大豆病害的自动诊断与决策。应用实践表明：该方法简单、可靠、易于实现,对大豆病害的诊断准确率达到87.62%,为植物病害的智能远程诊断和科学管理提供了一种高效的新途径。     

基于案例推理技术的研究与应用

基于案例推理(Case-based reasoning．CBR)技术被认为是当今人工智能与专家系统设计的一种非常具有发展前景的方法。本文阐述了CBR的基本原理、工作过程及其主要技术,详细介绍了CBR的国内外的应用现状和研究进展,最后对CBR今后的发展趋势做出概括。     

面向顾客需求的在线匹配定制方法

在线定制的产品涉及到设计、生产等复杂环节,所以定制品的方案设计及其可行性等信息的快速反馈是实施在线定制的难点。提出基于CBR（case-based reasoning）的匹配定制方法,减少定制时间避免不必要的重复设计。首先运用层次分析法和质量功能展开确定需求工程设计权重;结合模糊数学修改高斯函数,构造出一种具有接近客观现实、方便处理、区分性强等优点的匹配度计算方法;利用Beta分布构建机器学习方法,得到随市场偏好变动而调整的匹配度阀值。其次针对顾客对不同产品的价格敏感度不同,引入调节因子修正最终匹配值对比匹配阀值从而获得定制产品的设计方案及可行性并反馈。最后以电冰箱定制为例,证明该方法有效且易于实施。     

中心渔场智能预报系统的设计与实现

应用人工智能技术设计中心渔场智能预报系统,包括系统的总体结构,范例库的建设,范例推理、规则库和规则修正,最后按设计路线给出试验性预报实例,预报结果与渔场实际情况比较,预报的准确性为７６．２％。研究结果表明,采用范例推理为主,规则修正为辅的技术路线使用计算机实现中心渔场智化预报是可行的。它可以为海洋渔业生产和管理部门提供快速、准确、实用的信息预报。     

A case-based method of selecting covariates for digital soil mapping

Selecting a proper set of covariates is one of the most important factors that influence the accuracy of digital soil mapping(DSM). The statistical or machine learning methods for selecting DSM covariates are not available for those situations with limited samples. To solve the problem, this paper proposed a case-based method which could formalize the covariate selection knowledge contained in practical DSM applications. The proposed method trained Random Forest(RF) classifiers with DSM cases extracted from the practical DSM applications and then used the trained classifiers to determine whether each one potential covariate should be used in a new DSM application. In this study, we took topographic covariates as examples of covariates and extracted 191 DSM cases from 56 peer-reviewed journal articles to evaluate the performance of the proposed case-based method by Leave-One-Out cross validation. Compared with a novices' commonly-used way of selecting DSM covariates, the proposed case-based method improved more than 30% accuracy according to three quantitative evaluation indices(i.e., recall, precision, and F1-score). The proposed method could be also applied to selecting the proper set of covariates for other similar geographical modeling domains, such as landslide susceptibility mapping, and species distribution modeling.     

A cloud computing based SWRL distributed reasoning framework

With the explosion of semantic web technologies, large amounts of OWL ontologies are common place. Conventional rule engines inevitably meet the bottleneck of computing performance and scalability. A cloud computing based SWRL distributed reasoning framework named CloudSWRL is proposed. Based on the Hadoop open-source framework and SWRL semantics, the storage schema for OWL ontologies is designed to implement efficient data retrieving from HBase. Some novel data models for SWRL rules and intermediate data are defined. At last, a MapReduce paradigm based distributed SWRL reasoning algorithm is proposed under DL-safe restriction. An experiment on a simulation environment shows our framework is more efficient and scalable than conventional rule engines when reasoning over large-scale of OWL data.