基于加工特征的B-rep模型几何变换算法

为了真实反映加工过程中零件的几何变换情况,对边界表示法模型的特征尺寸的改变和特征的删除方法进行了研究。基于拓扑结构保持不变的前提下,提出了通过延长特征邻接边实现面特征在法向方向上尺寸改变的几何更新算法和几何形状规则的特征的删除算法,针对几何形状不规则的非独立特征,提出了通过面对齐和合并面处理来实现特征删除的算法。经过几何变换后,确保零件模型面壳封闭,保证模型的合理性和完整性,能够通过模型重构映射零件的实际加工情况。最后通过软件实例验证了方法的可行性。     

基于Bezier曲线的焊缝成形计算机仿真技术

针对车间数字化制造的应用需要,提出了一种基于Bezier曲线的焊缝成形过程的计算机仿真建模技术. 通过对电弧焊焊缝成形规律的研究,建立了焊接工艺参数与焊缝几何特征之间的映射关系,并运用B-rep法开展了电弧焊焊缝的几何特征建模技术研究,采用OpenGL技术开发了电弧焊缝成形的计算机仿真原型. 仿真结果表明文中所提出的焊缝成形过程建模技术与计算机仿真系统可有效指导焊接工艺的数字化设计.     

快速数控编程系统的制造特征构建研究

针对快速数控编程系统中不同CAD模型的特征识别和构建,论述了基于STEP文件的特征识别技术及其实现过程:首先利用词法分析器解析STEP中性文件,按照STEP的文件拓扑结构生成属性邻接图(AAG);在总结典型结构件拓扑特征基础上,结合数控编程切削逻辑,以切削级为基础进行特征识别和特征构建;最后以XML形式构造制造特征森林以供CAM系统使用.实例证明了文中方法的有效性,提高了结构件数控编程的效率和质量.     

Extendible Classification of Design and Manufacturing Features

This paper discusses three different issues in feature classification. It proposes a complete taxonomy to cater for prismatic and rotational features which can be positive or negative. It also suggests a feature definition that makes it feasible for the feature library to be extendible without additional programming effort. A unique code is given to each feature so as to avoid redundancy and ensure consistency of labelling the feature faces. This labelling is achieved by a novel formalisation method based on a newly found graph-theoretical property of a B-rep model. Results for one of the several components investigated are included.     

An image detection approach to NC rough-cut milling from solid models

This paper presents a method of converting information from a CAD solid model into a form suitable for CAM operation. The rough machining of a cavity into a block is used to illustrate the working of the system. This example was chosen because the machining of cavities presents difficulties as outlined below and rough machining involves the removal of more material than finish machining and therefore is of more economic significance in a production process. A three-axis NC miller is used to cut the cavity which has depth, requiring material to be removed in layers. A computer-based image detection method is used for cutter-path generation and models that contain planar surfaces, general quadratic surfaces, B-spline surfaces and compound surfaces can be treated. The B-rep solid object is transformed into an image as a grid-height model, allowing a three-dimensional object to be approximated by a two-dimensional spatial array. The cutter location (CL) data file is automatically generated from this spatial array. The height change of stock material in each grid is recorded in a two-dimensional array during the machining process and is utilized as an image for further roughing and verification. Efficient machining procedures are obtained by an analysis of cutting simulation, utilizing large cutter sizes for simple shaped portions in the first stage of roughing and then using small cutter sizes for complex portions and to remove uncut material left by the large cutters in a fine roughing operation. This approach allows a three-dimensional cutter path problem to be reduced to one of two dimensions which is solved by image detection of cutting attributes.     

复杂几何模型扫描数据的孔、槽及表面类型的检测

逆向工程就是从模型样件的扫描点云中提取物理特征进而重建该实体的过程。本文介绍了一种通过人工神经网络从三维扫描数据点云识别棱柱特征并进行CAD模型重建的新算法。该算法先从扫描点云中抽取四大基本几何属性：方向、凹／凸性、圆／直性、开／闭性,接着输入点的几何属性到人工神经网络（ANN）进行特征识别。本文探讨了如何直接从扫描点云中通过上述四大几何属性识别物理特征的算法。实验结果证明：逆求产品和样件在尺寸和结构上高度吻合。     

非枝晶固相组分计算机分析系统

采用计算机图像识别技术，开发了用于非枝晶半固态金相组织成分中固相组分计算的软件处理系统。应用表明，该软件界面清晰、操作简单，使用效果理想。经对比验证，该系统误差小于1％。     

基于混合特征和堆栈稀疏自编码器的齿轮箱故障诊断

针对复杂工况下齿轮箱多故障信号诊断准确率低的问题,提出了一种基于混合特征和堆栈稀疏自编码器的齿轮箱故障诊断方法。从微观信号特征角度提取奇异值特征和小波分解后的样本熵特征;从宏观角度提取故障信号时域特征,将3种特征进行融合,并输入到由稀疏自编码和Softmax堆栈得到的深度神经网络中进行特征优化和分类识别。实验结果表明:在2种不同工况下,对6种齿轮箱故障数据进行诊断均表现出较高分类识别精度,且所构建的分类模型综合性能上均高于文中其他对比模型,因此本文作者所提出的方法能有效地进行齿轮箱故障诊断。     

Perceptually Grounded Language Learning: Part 2—DETE: A Neural/Procedural Model

In Part 1 of this two-part series, we introduced Katamic memory—a neural network architecture capable of robust sequence learning and recognition. In Part 2, we introduce the Blobs World taskjdomain for language learning and describe the DETE language learning system, which is composed of over 50 Katamic memory modules. DETE currently learns small subsets of English and Spanish via association with perceptual! motor inputs. In addition to Kaiamic memory, DETE employs several other novel features: (1) use of feature planes, to encode visual shapes, spatial relationships and the motions of objects, (2) phase-locking of neural firing, in order to represent focus of atention and to bind objects across multiple feature planes, and (3) a method for encoding temporal relationships, so that DETE can learn utterances involving the immediate past and future. We compare DETE to related models and discuss the implications of this approach for language-learning research.     

基于B-rep边界约束的Delaunay四面体剖分

为实现B-rep的三维实体的四面体剖分,研究了网格划分过程中的关键问题.基于B-rep数据结构的拓扑关系,提出一种在网格划分过程中约束实体边界的策略.研究了包含插入点的包含域的合法性检验方法,并对其改进扩展以保证生成单元的有效性.在对点和面的拓扑关系进行预处理的基础上,提出了判断四面体单元是否被实体边界包围的面法向法和基于单元边的加密点生成算法.最后使用Delaunay逐点插入法结合以上算法实现对以B-rep为模型的三维实体的四面体单元网格划分,并通过实例得到验证.     

面向模型重构的冲压零件特征识别技术

从面向特征模型重构的角度,对冲压零件的特征进行了分类,并采用图结构描述分类特征。在此基础上,将特征识别的过程分解为模型有效性检查和基于图匹配的特征提取,所识别出的特征,可根据后续应用的需要重构出新的特征模型。     

Intelligent pattern recognition and diagnosis of ultrasonic inspection of welding defects based on neural network and information fusion

Abstract

The detection of defects in real manual metal arc welds using ultrasonic non-destructive testing has been investigated. Twenty-six features, extracted from three domains, were applied for recognition of defect type. To increase the reliability and accuracy of identification and classification, statistical analysis was used to evaluate the features extracted from ultrasonic defect echoes. The subset of optimum feature was then selected using the method of discriminant analysis. An intelligent defect evaluation method derived from the study is presented. The results show that statistical analysis is an effective method for feature evaluation. The uncertainty of defect diagnosis can be decreased by the information fusion method, and for three specific defect types, defects were correctly identified in approximately 93% of cases.     

基于空间位置和轮廓线距离的船舶焊缝特征参数提取

为了实现船舶焊接件数字模型中焊缝特征参数的精确提取,进而完成机器人数据库系统中焊接工艺的自适应快速匹配和快速选择,提出了基于空间位置和轮廓线距离的船舶焊缝特征参数提取算法.首先基于海伦公式识别待判定面来确定接头空间位置关系,并结合最小轮廓线距离完成焊缝特征识别;然后基于轮廓线总条数和最小轮廓线距离的两端点,识别出焊缝坡口处特征点及线;最后基于三类焊接接头所建的数学模型提取出与焊接工艺相关的焊缝特征参数.测试结果表明,文中焊缝特征参数提取算法能准确识别4类接头形式和10种坡口类型,以及准确提取焊缝间隙、坡口夹角和焊接件板厚等参数,具有焊缝特征识别广且信息提取齐全的优点.与其他相关识别算法相比,文中算法的识别率达到100%,而识别效率提升了16.06%,从而进一步验证了算法的有效性.     

Fuzzy similarity-based rough set method for case-based reasoning and its application in tool selection

Case-based reasoning (CBR) embodied in die and mold NC machining will extend the application of knowledge-based system by utilizing previous cases and experience. However, redundant features may not only dramatically increase the case memory, but also make the case retrieval algorithm more complicated. Additionally, traditional methods of feature weighting limit the development of CBR methodology. This paper presents a novel methodology to apply fuzzy similarity-based Rough Set algorithm in feature weighting and reduction for CBR system. The algorithm is used in tool selection for die and mold NC machining. The proposed method does not need to discretize continuous or real-valued features included in cases, from which can effectively reduce information loss. The weight of feature a_i is computed based on the difference of its dependency defined as γ_A−γ_A−{ai}, which also represents the significance of the corresponding feature. If the difference is equal to 0, the feature is considered to be redundant and should be removed. Finally, a case study is also implemented to prove the proposed method.     

Magnetic leakage field due to sub-surface defects in ferromagnetic specimens

This paper describes analytical expressions for the magnetic leakage field of two types of internal defects: two dimensional rectangular and elliptical defects as functions of the applied magnetic field strength, the defect size and the distance below the surface. In this study, the magnetic image effects from the spatial boundary and the defect boundary are taken into consideration. That is, the leakage field of rectangular-like defects or elliptically shaped defects ‘below the surface’ are derived by using the modified dipole model and image theory. The profiles and strengths of leakage fields calculated from the proposed analytical forms are presented under various conditions and compared with experimentally measured ones.     

An inverse estimation scheme to measure steady-state tool–chip interface temperatures using an infrared camera

A new technique is developed to estimate the average steady-state chip–tool interface temperature during turning. An infrared (IR) video camera attached on the carriage of the lathe measures the transient cooling behavior on the rake surface of an insert after the feed motion is halted. This allows the zero heat flux boundary condition, where the transient Laplace heat conduction problem can be solved numerically to obtain the temporal and spatial temperature distribution. With the experimentally determined transient temperature distribution, the one-dimensional ellipsoidal model is used to estimate the average steady-state chip–tool interface temperature during machining. The results on turning gray cast iron (GCI) and AISI 1045 steels with various coated and uncoated K313 carbide inserts are presented.     

基于规则的汽车覆盖件孔和翻边特征识别方法研究

特征识别技术是提高汽车覆盖件制造水平,实现CAD/CAPP/CAM集成的关键.在研究了汽车覆盖件自由曲面特征识别技术的基础上.提出了孔类和翻边类特征的表示方法,并采用基于启发式规则的特征识别算法,实现了汽车覆盖件自由曲面上的孔类和翻边类特征的自动识别.并以UGⅡ软件为平台,开发出汽车覆盖件零件模型自动特征识别功能.     

基于SolidWorks的工程图转为三维模型的实用方法研究

简介SolidWorks中将二维工程图生成三维模型的实现方法，提出一种以匹配线框为形状特征的视图识别和建模方法，并对在转换过程中出现的问题进行了分析。     

Ultrasonic non-destructive evaluation with spatial combination of Wigner–Ville Transforms

This paper introduces a novel ultrasonic signal combination technique to be applied in detection systems based on multiple transducers. The technique uses a spatial combination approach that considers the specimen inspection from several apertures located in different planes. Information received from transducers is fused in a common integrated pattern with a signal to noise ratio (SNR) improvement. The result of the combination is a high quality image of the inspected material obtained from simple A-scans.The method is based on digital signal processing techniques, more concretely time–frequency analysis. Combination is performed by means of the Wigner–Ville Transform preserving temporal and frequencial information. Temporal techniques for combination are presented and the results obtained from both techniques are compared using the SNR.