Greedy information acquisition algorithm: A new information theoretic approach to dynamic information acquisition in neural networks

In this paper, we proopose a new information theoretic approach to competitive learning. The new approach is called greedy information acquisition , because networks try to absorb as much information as possible in every stage of learning. In the first phase, with minimum network architecture for realizing competition, information is maximized. In the second phase, a new unit is added, and thereby information is again increased as much as possible. This proceess continues until no more increase in information is possible. Through greedy information maximization, different sets of important features in input patterns can be cumulatively discovered in successive stages. We applied our approach to three problems: a dipole problem; a language classification problem; and a phonological feature detection problem. Experimental results confirmed that information maximization can be repeatedly applied and that different features in input patterns are gradually discovered. We also compared our method with conventional competitive learning and multivariate analysis. The experimental results confirmed that our new method can detect salient features in input patterns more clearly than the other methods.     

Teacher-directed learning: information-theoretic competitive learning in supervised multi-layered networks

In this paper, we propose a new type of efficient learning method called teacher-directed learning. The method can accept training patterns and correlated teachers, and we need not back-propagate errors between targets and outputs into networks. Information flows always from an input layer to an output layer. In addition, connections to be updated are those from an input layer to the first competitive layer. All other connections can take fixed values. Learning is realized as a competitive process by maximizing information on training patterns and correlated teachers. Because information is maximized, information is compressed into networks in simple ways, which enables us to discover salient features in input patterns. We applied this method to the vertical and horizontal lines detection problem, the analysis of US–Japan trade relations and a fairly complex syntactic analysis system. Experimental results confirmed that teacher information in an input layer forces networks to produce correct answers. In addition, because of maximized information in competitive units, easily interpretable internal representations can be obtained.     

Information theoretic competitive learning in self-adaptive multi-layered networks

In this paper, we propose self-adaptive multi-layered networks in which information in each processing layer is always maximized. Using these multi-layered networks, we can solve complex problems and discover salient features that single-layered networks fail to extract. In addition, this successive information maximization enables networks gradually to extract important features. We applied the new method to the Iris data problem, the vertical-horizontal lines detection problem, a phonological data analysis problem and a medical data problem. Experimental results confirmed that information can repeatedly be maximized in multi-layered networks and that the networks can extract features that cannot be detected by single-layered networks. In addition, features extracted in successive layers are cumulatively combined to detect more macroscopic features.     

Improving feature extraction performance of greedy network-growing algorithm by inverse euclidean distance

We propose here a new computational method for the information-theoretic method, called the greedy network-growing algorithm, to facilitate a process of information acquisition. We have so far used the sigmoidal activation function for competitive unit outputs. The method can effectively suppress many competitive units by generating strongly negative connections. However, because methods with the sigmoidal activation function are not very sensitive to input patterns, we have observed that in some cases final representations obtained by the method do not necessarily faithfully describe input patterns. To remedy this shortcoming, we employ the inverse of distance between input patterns and connection weights for competitive unit outputs. As the distance becomes smaller, competitive units are more strongly activated. Thus, winning units tend to represent input patterns more faithfully than in the previous method with the sigmoidal activation function. We applied the new method to artificial data analysis and animal classification. Experimental results confirmed that more information can be acquired and more explicit features can be extracted by our new method.     

高强钢焊接缺陷磁光成像分形特征检测

研究一种基于磁光成像原理的焊接缺陷无损检测新方法.以高强钢表面微小焊接缺陷为例,采用分形维数对焊缝磁光图像进行特征识别并估计最优尺度,根据Adabost分类算法对提取的焊接缺陷特征进行分析和训练,构建焊接缺陷特征量并对高强钢表面缺陷磁光图像进行自动识别.结果表明,运用磁光成像方法可以获取高强钢焊接缺陷特征,并通过图像分形维数分析可识别焊缝缺陷的位置、形状和类别.     

An Innovative Methodology for the Performance Evaluation of Coated Cemented Carbide Inserts in Milling of Inconel 718

The performance of cemented carbide inserts coated with various PVD-films in milling Inconel 718 is evaluated by innovative analytical and experimental methods. Three PVD films with different micro and macro structures were applied. The coatings’ strength properties were detected by nanoindentations and by impact tests at various temperatures. These results were considered in FEM calculations of the material removal process to determine the mechanical and thermal loadings of the cutting wedge. Employing this innovative methodology, it is possible to capture proactively the effect of the cutting conditions on tool wear, thus reducing the effort dedicated in traditional sequential cutting experimentation.     

铝土矿的宏观及微观特征剖析

从宏观和微观两方面提出了鉴别铝土矿矿床和矿石的依据。在宏观上，通过对铝土矿形成的古地理特征及铝土矿床在空间上分布规律的研究，提出了寻找、鉴别铝土矿床的依据；在微观上．通过对铝土矿矿石的结构、构造及铝土矿与其它类似矿石区别的论述．提出了辨别铝土矿矿石的有效方法。     

铸造合金凝固组织的计算机模拟与预测

数值模拟技术与计算机技术的迅速发展，使凝固过程的数值模拟正在由宏观向微观转变。微观模拟不仅可以得到材料的凝固组织，而且还能为宏观模拟提供准确的潜热释放信息。对宏观和微观现象的完整耦合可以使我们对铸造凝固过程作出更加准确的模拟预测。针对目前微观组织模拟的研究现状，介绍了几种主要的模拟研究方法，如确定性模拟方法、随机性模拟方法和相场方法等，阐述了其主要特征和模拟微观组织时存在的优缺点，并给出了各种模拟方法的计算实例。由于凝固组织数值模拟的复杂性，对其进行深入的研究是必要的，也是未来计算材料科学的重要发展方向。最后对微观模拟中的问题及发展方向作了分析。     

基于混合时间序列卷积神经网络的轴承故障诊断

在传统滚动轴承故障诊断中,绝大多数方法采用了从振动信号提取特征的诊断模式,但是这种模式必然会使原始信号降维进而导致故障信息的丢失。卷积神经网络(CNN)通过权重共享和稀疏连接直接对原始信号进行操作,实现自适应特征提取,最大化保留故障信息。受CNN原理启发,开发出了一种基于工业振动信号特征的新型诊断框架,称之为混合时间序列CNN(HTS-CNN)。首先,利用估计总体比例的方法自适应确定模型训练样本数目;其次,通过对时间序列片段进行随机组合的方式,使模型能够提取非相邻信号特征;最后,利用Softmax激活函数在模型输出端执行多分类任务。通过对凯斯西储大学及CUT-2平台轴承数据进行分析,实验结果表明:该方法能够准确、有效的对滚动轴承故障进行分类。     

显微偏析数值模拟的微观模型和数值方法

研究显微偏析数值模拟中的微观模型和数值方法,在微观模型中考虑了树枝晶粗化、固相溶质逆扩散、枝晶尖端过冷、随温度变化的溶质扩散系数等影响显微偏析形成的动力学因素.数值方法中采用变网格技术跟踪移动界面,通过迭代求解溶质扩散方程和溶质守恒方程计算显微偏析参数.并给出了微观计算流程图,这种数值方法可以方便地与宏观计算、相图计算相耦合.模拟算例表明模拟结果与实验结果吻合较好.     

SiC/Al-Si复合材料颗粒分布的数值模拟

介绍了颗粒分布的定量分析技术在颗粒增强复合材料性能预测方面的研究方法以及应用进展。在对搅拌铸造法制备SiC颗粒增强Al-7．0Si复合材料凝固过程的宏／微观数值模拟的基础上，将样方法以及方差分析法嵌入数值模型，对不同颗粒体积分数条件下复合材料颗粒分布均匀性进行了定量分析，并与试验结果进行了对比。结果表明：将宏／微观数值模拟与定量分析技术相结合，能够更为准确地分析不同工艺参数对复合材料微观组织以及颗粒分布的影响。在本试验条件下，随着SiC颗粒体积分数的增加，样方法中的颗粒均匀性参数值以及方差系数都逐渐减小，SiC颗粒分布更均匀。     

基于GRETA正则表达式的机器人解释器的研究

为了解决传统解释器编译过程复杂,效率低的问题,提出了一种基于GRETA正则表达式的工业机器人解释器的实现方法。利用GRETA正则表达式建立字符匹配表和语法规则后,该解释器即可方便地对指令代码进行解释处理,包括:词法分析、语法分析、语义分析和信息提取。实验表明,该解释器可实现对机器人指令集的正确解释和信息提取,并为机器人控制的开放性和高效性奠定基础。     

基于宏程序的冶金锯片数控加工方法

新型硬质合金齿圆锯片在钢铁行业中的应用越来越广泛,传统的加工方法已不能满足当前的生产效率及加工精度的需求。提出基于宏程序的卧式加工中心对冶金锯片的数控加工方法,分析了加工工艺,阐述了宏变量的定义,编制了结构化的宏程序,实现了数控宏程序与制造工艺的巧妙结合。利用台湾LEADWELL四轴卧式加工中心制造成品,实践证明:该方法控制方便、加工精度高、效率高、适应性强及程序可读性好。     

《铸造三维设计及优化》学练做一体化教学改革与实践

根据铸造行业人才培养新要求,整合传统专业课程,结合三维设计软件的应用,开展《铸造三维设计及优化》课程的教学改革。课程结构分为铸造工艺设计基础,专业绘图软件的应用,铸造仿真模拟及优化三大模块,采用学练做一体化教学,灵活运用案例法、分组法、任务驱动等教学方法,取得较好的教学效果。最后对教学改革中遇到的问题和改进措施进行了梳理总结。     

电阻点焊熔核微观组织模拟

基于有限差分思想建立了电阻点焊的三维宏观传热模型,并在此模型基础上通过进一步细化网格差分得到枝晶形核和生长所需的温度信息.基于元胞自动机方法建立枝晶形核和生长的微观模型,将已建立的宏观温度场模型和微观元胞自动机模型互相耦合起来,建立了电阻点焊熔核微观组织模拟的宏微观耦合模型.利用耦合模型模拟了5754铝合金电阻点焊一定参数下熔核的微观组织形成过程,所得的模拟结果为"柱状+等轴"晶组织,与试验结果吻合较好.     

精密微塑性成形系统的研制

随着微型零件尺寸的减小,对成形设备提出了更高要求,传统的塑性成形设备难以在小行程下实现载荷和位移的精确控制。针对微成形的特点研制了精密微塑性成形系统,设计了宏动／微动相结合的驱动系统,微动部分使用压电陶瓷作为驱动器,宏动部分采用精密丝杠旋钮来实现,借助数据采集系统进行数据的实时采集和处理,使用成形工艺控制器对微成形过程进行精确控制,该系统可以对模具进行加热以实现等温成形,由温度PID控制器控制。使用该精密微塑性性成形系统进行了微成形试验,所成形的微型齿轮零件质量良好。     

基于云计算与风险评估的数字化信息发布系统设计

为了解决大数据环境下,传统数字化信息发布系统存在的响应效率低以及抗风险性能差的弊端,设计基于云计算与风险评估的数字化信息发布系统,系统由数据层、支撑层以及应用层构成,分析了系统的业务流程,并对系统的综合信息管理、风险评估交流以及信息发布共享功能进行了分析。给出了系统中多媒体播放终端模块、信息发布管理模块的设计和实现过程。分析基于云计算的信息安全风险评估流程,实现信息发布系统安全风险的有效评估。实验结果说明所设计的信息发布系统,具有较高的响应效率和抗风险性能。     

基于VMD和流形学习的滚动轴承故障诊断研究

针对轴承早期故障信号非线性、非平稳和故障特征难以提取的问题,提出一种变分模态分解(VMD)与流形学习相结合的特征提取方法。该方法应用VMD将信号分解成包含不同故障信息的固有模态分量,然后从中提取特征并构建高维的混合域特征集。最后,应用流形学习等度规映射算法将高维的特征集约简为故障区分度更好的低维混合域特征集,并利用支持向量机实现故障分类识别。滚动轴承实验结果表明该方法能准确清晰地提取故障特征信息,与传统方法相比诊断准确率更高。     

Modelling and Simulation of Micro EDM Process

In micro EDM, simple-shaped tools are being used to machine complex 3D micro features because of the cost and difficulty in fabricating multiple complex-shaped micro electrodes. However, the tool wear adversely affects the accuracy of machined micro features. Therefore, a theoretical model for surface profile generation which takes into account the effect of tool wear is needed. This paper presents such a model and a simplified simulation algorithm. The experimental results of micro and macro EDM generated surface profiles confirm the validity of the proposed mathematical model and the simulation algorithm.     

过滤控流中间包流场及夹杂物去除的数值模拟

以60 t板坯连铸中间包为研究对象,采用雷诺平均方程、组分输送以及离散相模型分别计算了高拉速下中间包内钢液流场特征演变以及夹杂物运动轨迹和去除率。结果表明,采用多孔过滤控流装置后,10 μm的小夹杂物去除率提升了10%以上。与45°倾角过滤器中间包相比,55°倾角过滤器中间包的夹杂物去除率增加约5.3%,其顶部孔为水平方向,减小了卷渣与钢液裸露的风险。通过RTD曲线获得多孔过滤控流中间包的钢液停留时间缩短了30~50 s、死区体积比增加0.03%,但夹杂物去除率明显提高。因此,采用停留时间、死区体积比等宏观流场特征参数评估中间包内夹杂物去除效率的方法已不适应过滤控流装置中间包的开发研究,宜采用速度云图、夹杂物运动轨迹等微观分析方法评估其冶金效果。