Adaptive motion control using neural network approximations

In this paper, we present a new adaptive technique for tracking control of mechanical systems in the presence of friction and periodic disturbances. Radial Basis Functions (RBFs) are used to compensate for the effects of nonlinearly occurring parameters in the friction and periodic disturbance model. Theoretical analysis, such as stability and transient performance, is provided. Furthermore, the performance of the adaptive RBF controller and its non-adaptive counterpart are compared.     

Image processing using cellular neural networks based on multi-valued and universal binary neurons

Multi-valued and universal binary neurons (MVN and UBN) are the neural processing elements with the complex-valued weights and high functionality. It is possible to implement an arbitrary mapping described by partially defined multiple-valued function on the single MVN. An arbitrary mapping described by partially defined or fully defined Boolean function, which can be non-threshold, may be implemented on the single UBN. The quickly converging learning algorithms exist for both types of neurons. Such features of the MVN and UBN may be used for solving the different problems. One of the most successful applications of the MVN and UBN is their usage as basic neurons in the Cellular Neural Networks (CNN). It opens the new effective opportunities in nonlinear image filtering and its applications to noise reduction, edge detection and solving of the super resolution problem. A number of experimental results are presented to illustrate the performance of the proposed algorithms.An erratum to this article can be found at     

Network topology and the theory of rubber elasticity

The earliest investigations on rubber elasticity, commencing in the 19th century, were necessarily limited to phenomenological interpretations. The realisation that polymers consist of very long molecular chains. commencing c. 1930, gave impetus to the molecular theory of rubber elasticity (1932-). according to which the high deformability of an elastomer, and the elastic force generated by deformation, stem from the configurations accessible to long molecular chains. Theories of rubber elasticity put forward from 1934-1946 relied on the assumption that the junctions of the rubber network undergo displacements that are affine in macroscopic strain. The theory of James and Guth (1947) dispensed with this premise, and demonstrated instead that the mean positions of the junctions of a ‘phantom’ network consisting of Gaussian chains devoid of material properties are affine in the strain. The vital significance of the distinction between the actual distribution of chain vectors in a network and their distribution if the junctions would be fixed at their mean positions went unnoticed for nearly 30 years. Experimental investigations, commencing with the incisive work of Gee in 1946. revealed large departures from the relationship of stress to strain predicted by the theories cited. This discrepancy prompted extensive studies, theoretical and experimental, during succeeding years. Inquiry into the fundamentals of polymer networks, formed for example by interlinking very long polymer molecules, exposed the need to take account of network imperfections, typically consisting of chains attached at only one end to a network junction. Various means were advocated to make corrections for these imperfections. The cycle rank ζ of the network has been shown (1976) to be the fundamental measure of its connectivity, regardless of the junction functionality and pattern of imperfections. Often overlooked is the copious interpenetration of the chains comprising typical elastomeric networks. Theories that attempt to represent such networks on a lattice are incompatible with this universal feature. Moreover, the dense interpenetration of chains may limit the ability of junctions in real networks to accommodate the fluctuations envisaged in the theory of phantom networks. It was suggested in 1975 that departures from the form predicted for the elastic equation of state are due to constraints on the fluctuations of junctions whose effect diminishes with deformation and with dilation. Formulation of a self-consistent theory based on this suggestion required recognition of the non-affine connection between the chain vector distribution function and the macroscopic strain in a real network, which may partake of characteristics of a phantom network in some degree. Implementation of the idea was achieved through postulation of domains of constraint affecting the equilibrium distribution of fluctuations of network junctions from their mean positions. This led in due course to a theory that accounts for the relationship of stress to strain virtually throughout the ranges of strain accessible to measurement. The theory establishes connections between structure and elastic properties. This is achieved with utmost frugality in arbitrary parameters.     

Effects of modeling means on properties of monitoring models of spot welding quality

Analyzing and modeling the relation between monitoring information during welding and quality information of the joints is the foundation of monitoring resistance spot welding quality. According to the means of modeling, the known models can be divided into three large categories: single linear regression models, multiple linear regression models and multiple non-linear models. By modeling the relations between dynamic resistance information and welding quality parameters with different means, this paper analyzes effects of modeling means on performances of monitoring models of resistance spot welding quality. From the test results, the following conclusions can be drawn: By comparison with two other kinds of models, artificial neural network (ANN) model can describe non-linear and high coupling relationship between monitoring information and quality information more reasonably, improve performance of monitoring model remarkably, and make the estimated values of welding quality parameters more accurate and reliable.     

宝浪油田侏罗系三工河组浅水辫状河三角洲沉积模式研究

焉耆盆地侏罗系三工河组广泛发育由辫状河进积到湖泊中形成的辫状河三角洲。这种三角洲特征显著不同于其它三角洲,其主体是辫状分流河道和水下分流河道。宝浪油田含油层段主体为辫状分流河道,以心滩发育为主要特征,是典型的浅水缓坡粗粒辫状河三角洲沉积。     

对流域水电建设档案科学管理的研究

介绍清江公司从流域水电建设的特点出发,研究流域水电建设工程档案的完整性、准确性,确保工程质量,提高工程管理水平,将大量的档案信息资源转化为科学知识,为流域水电开发服务,取得显著的经济和社会效益的情况。     

环氧化天然橡胶改性软质聚氯乙烯的研究

对环氧化度为50％的环氧化天然橡胶（ENR－50）改性软质聚氯乙烯（PVC）体系的性能进行了研究，并用动态力学和电镜分析了PVC与ENR－50的相容性。结果表明：适量的ENR－50可明显改善软质PVC的性能，而且PVC的相对分子质量越大，PVC／ENR－50共混材料的性能越好。虽然DMA测试PVC／ENR－50共混材料只有1个玻璃化转变温度（Tg），但在微观上并不完全相容。     

特大型桥梁工程地质勘察区域地壳稳定性研究

重大工程区域地壳稳定性研究，是工程前期工程地质勘察的重要内容之一，并越来越多地为工程建设所重视。大桥工程地质勘察以往对此涉及较少，也无明确的统一规定，以长江上所建特大桥为例，对此项工程作了些探索性的研究，重点研究的问题是：区域构造环境及构造变形，地壳结构和深部构造，新构造活动，断裂活动，地震活动等，为大桥建设宏观决策和抗震设计提供依据。     

黄河黑山峡河段大柳树水库比小观音水库原始库容多40亿m3的作用

一级开发大柳树高坝原始库容比二级开发小观音高坝多 4 0亿m3 ,库容大带来的好处体现在 5～ 7月可多向下游增供水量 ,缓解下游河道断流 ;延长泄放清水时间 ;增加发电效益 ;满足宁蒙河段防凌要求 ;增加汛期集中泄水 ,减缓下游河道淤积萎缩 ;满足南水北调西线工程水量调节要求等。为了更好优化配置黄河水资源 ,从全局和长远利益考虑 ,应采用一级开发方案 ,修建大柳树高坝大库     

Monoacylglycerols derived from butter oil by Penicillium roquefortii in suspension cultures

Structural isomers of monoacylglycerols (monoglycerides, MAGs) were identified and compared after degradation of butter oil by two strains of Penicillium roquefortii and a commercial lipase from P roquefortii (EC 3.1.1.3) at pH 7.0 and 10 °C. The conditions were selected as they were comparable with those used in the manufacture of blue mould‐ripened cheese. The commercial lipase was selected to compare with the fungal strains in terms of acyl migration. Results showed that the main isomers formed by lipolysis with the commercial lipase were sn‐2 MAGs (64 mol%), whilst spores and emerging mycelia of P roquefortii produced mainly sn‐1(3) MAGs (83–90 mol%). The work reported here may lead to further assessment of different MAG structural isomers as natural preservatives in foods and dairy products. © 2002 Society of Chemical Industry