基于实例推理及其在农业病虫害诊断与防治中的应用研究

规则推理等传统推理方法的推理性能严重依赖于知识的完备程度。在农业领域中。知识获取和知识表示已成为推理的主要瓶颈之一，因此利用传统推理方法建立的专家系统效率低、鲁棒性不好。基于案例推理的专家系统能充分利用已有的实例经验，具有自学习功能，从而推理效率高，鲁棒性能较好。本研究将案例推理专家系统应用于作物的病虫害防治和诊断．得出了基于案例推理农业病虫害诊断与防治专家系统的知识表示和推理机制，并在实际应用中获得了较好的效果。     

车辆变速箱故障诊断的发展现状与展望

车辆在运输过程中,一旦变速箱发生异常或故障,则导致变速箱失效而发生事故。以车辆变速箱为对象的故障诊断技术属于旋转机械故障诊断技术的应用领域,是基于机械故障诊断技术的发展而发展起来的产物。为此,简要介绍了国内外车辆变速箱故障诊断的发展现状和存在的问题,并对未来的车辆变速箱故障诊断研究与应用的发展方向进行了探讨,从而为我国的农用车辆故障诊断的发展提供一定的参考。     

冀北沙荒地退化程度诊断及其短期封育效果评价

以冀北退化沙荒地为研究对象,以恢复生态学为研究理论基础,对冀北沙荒地各项指标进行权重赋值,诊断沙荒地的退化程度,并通过分析封育区、未封育区及其地带植被演替顶级群落的土壤理化性质和植被数量特征,评价5年封育效果。结果表明,沙荒地封育区及未封育区均属中度退化程度,沙荒地土壤状况尚未因为5年的封育得到明显改善;植被状况虽有明显改善,但还处在演替初期的草本植被阶段,与地带植被演替顶级群落仍存在很大差距,封育措施需要持续下去。     

栗属分子生物学研究进展

壳斗科(Fagaceae)栗属(Castanea)植物的7个种分布广泛,不仅可用于木材生产,而且在坚果生产上也占有独特地位。基于形态学、同工酶、子叶储藏蛋白和RAPD数据,通过对分布于亚洲、欧洲和北美的栗属植物的遗传多样性研究表明中国板栗(Castaneamollissima)是世界栗属植物的原始种,长江流域是中国板栗的遗传多样性中心,土耳其是欧洲板栗的起源中心之一。在分子标记辅助育种方面,已发表了基于形态学、同工酶、RAPD和ISSR数据的两张栗属植物的遗传连锁图谱,其中一个是用欧洲板栗种内杂交后代F1全部单株构建的,另一张是用美洲板栗与中国板栗种间杂交后代F2单株构建的,并已经定位了3个假定的栗疫病抗性位点;并且证明在多年生木本果树上同工酶基因可通过连锁关系分析与形态基因整合为1个单一基因图而无需另外的杂交。从栗属植物中已分离纯化了包括可抑制HIV-1反转录酶活性的Mollisin在内的几丁质酶等抗真菌蛋白、胱氨酸蛋白酶抑制剂、热击蛋白、红血球凝集素、脱水素、花粉过敏原等功能蛋白质,并已经克隆了包括伤害应答基因在内的部分功能蛋白质相关基因。因此,作为具有独特性质的栗属植物,有必要开展更多的研究,就此本文对栗属植物遗传多样性、分子标记辅助育种和功能蛋白纯化和有关基因克隆等几方面的研究进展进行了总结,希望能位同行提供参考。     

水稻抗条纹叶枯病基因Stv-bi的分子标记辅助选择

水稻条纹叶枯病是我国黄淮及长江流域粳稻区重要的病害。由于水稻条纹叶枯病的发病受外界条件影响较大,人工接种抗性鉴定比较困难,利用与抗病基因紧密连锁的分子标记进行标记辅助选择对提高抗性育种效率具有重要意义。来自籼稻抗源Modan的Stv-bⁱ是水稻育种中广泛应用的条纹叶枯病抗性基因。本研究设计了与Stv-bⁱ紧密连锁的SSR及STS分子标记,用3个抗条纹叶枯病混合群体F30718(圣稻13/镇稻88)、F50701(武优34/T022//圣稻806)、F60702 (V6/T022//镇稻88)进行分子标记检测和田间条纹叶枯病抗性鉴定,其结果的符合率分别为99.3%、87.7%和91.8%。表明这些分子标记可以用于条纹叶枯病抗性基因Stv-bⁱ分子标记辅助选择。     

Characterization of dinucleotide and trinucleotide EST-derived microsatellites in the wheat genome

Over the past decade microsatellites or simple sequence repeats (SSRs) have attracted a considerable amount of attention from researchers. The aim of the present paper was to analyse expressed sequence tag-derived SSR (EST-SSR) marker variability in wheat and to investigate the relationships between the number and type of repeat units and the level of microsatellite polymorphism. Two hundred and forty-one new EST-SSR markers available in a public database () were characterized in eight durum wheat cultivars (Svevo, Ciccio, Primadur, Duilio, Meridiano, Claudio, Latino, Messapia), two accessions of Triticum turgidum var. dicoccoides (MG4343, MG29896), one accession of T. turgidum var. dicoccum (MG5323) and in the common wheat cv. Chinese Spring. Of these, 201 primer pairs (83.4%) amplified PCR products successfully, while the remaining 40 (16.6%) failed to amplify any product. Of the EST-SSRs analysed, 45.2% of the primer pairs amplified one or two PCR products. Multiple discrete PCR products were observed among both di- and trinucleotide EST-SSR markers (31.2 and 40.5%, respectively). Markers based on dinucleotide microsatellites were more polymorphic than those based on trinucleotide SSRs in the 12 wheat genotypes tested (68.9 and 52.7%, respectively). An average of 2.5 alleles for dinucleotide and 2.0 alleles for trinucleotide SSRs was observed. The data reported in the present work indicate the presence of a significant relationship between motif sequence types and polymorphism. The primer set based on the AG repeat motif showed the lowest percentage of polymorphism (55.0%), while the primer set based on the AC repeat motif showed t he highest percentage (85.0%). Among trinucleotide SSRs, the AGG microsatellite markers showed the highest percentage of polymorphism (70.0%), and the ACG motif the lowest value (25.0%). The characterization of these new EST-SSR markers and the results of our studyon the effect of repeat number and type of motifs could have important applications in the genetic analysis of agronomically important traits, quantitative trait locus discovery and marker-assisted selection.     

Amplified fragment length polymorphism as a tool for molecular characterization of almond germplasm: genetic diversity among cultivated genotypes and related wild species of almond,and its relationships with agronomic traits

Amplified fragment length polymorphism (AFLP) analysis is a rapid and efficient method for producing DNA fingerprints and molecular characterization. Our objectives were to: estimate genetic similarities (GS), marker indices, and polymorphic information contents (PICs) for AFLP markers in almond cultivars; assess the genetic diversity of almond cultivars and wild species, using GS estimated from AFLP fingerprints and molecular characterization; and facilitate the use of markers in inter-specific introgression and cultivar improvement. The genetic diversity of 45 almond cultivars from Iran, Europe, and America, were studied assaying 19 primer combinations. In addition, several agronomic traits were evaluated, including flowering and maturity times, self-incompatibility, and kernel and fruit properties. Out of the 813 polymerase chain reaction fragments that were scored, 781 (96.23%) were polymorphic. GS ranged from 0.5 to 0.96, marker indices ranged from 51.37 to 78.79, and PICs ranged from 0.56 to 0.86. Results allowed the unique molecular identification of all assayed genotypes. However, the correlation between genetic similarity clustering as based on AFLP and clustering for agronomic traits was low. Cluster analysis based on AFLP data clearly differentiated the genotypes and wild species according to their origin and pedigree, whereas, cluster analysis based on agronomic data differentiated according the pomological characterization. Our results showed the great genetic diversity of the almond cultivars and their interest for almond breeding.     

A Multi-agent Fault Diagnosis System of Rotating Machinery Based on CBR

Recently,distributed intelligent systems based on multi-agent have been applied to many fields successfully.the multi-agent technique is introduced into the rotating machinery fault diagnosis system.With the combination of CBR technique,Basalstructure and development approach of the system are analyzed,and better cooperation among agents is realized.As a result,the system overcomes the limitation of single fault diagnosis method,and solves the contradictionbetween versatilityand adaptabilityof diagnosis software.Furthermore,a reference can be used for the research and development of multi-algorithmdiagnosis system of rotating machinery.The application in a factory's Networked Online Monitoring and Fault Diagnosis System for Turbine Fan shows that the system can diagnosefault rapidly and exactly.     

A Method for State Diagnosis of Grounding Grid

This paper discuses the method with which the corrosion of the grounding grid can be diagnosed .A grounding grid can handled as a circuit network. With the application of circuit network graph theory and the fault diagnosis theory of analog circuit and optimization, the fault diagnosis equations can be set up. This equations is figured out by Matlab, and the corrosion status of the grounding grid can be deduced from the result.