Modeling method for assembly variation propagation taking account of form error

The propagation of variations, such as fixture errors and datum errors resulting from assembly and machining processes, has been extensively studied. However, only a few studies that focus on form error propagation in assembly systems have been implemented. Machining errors, especially form errors, have great impact on assembly accuracy and accuracy stability of precision mechanical systems. With form errors being the research object, a method for calculating mating variation and specifying mating coordinate is proposed to improve the accuracy of the variation propagation model. Taking into account the form error of mating surfaces, the assembly variation propagation of a precision mechanical system is analyzed, and the brief derivation procedure of the variation propagation model is introduced afterwards. The variation propagation model involves a new concept of mating variation specified by the two mating surfaces. An innovative method, the difference surface search based method, is proposed to calculate the mating variation amongst the mating surfaces. The obtained mating variation is then utilized to specify the mating coordinate in the variation propagation model. Moreover, FEM is employed to simulate the contact state of the two mating surfaces to demonstrate effectiveness of the proposed method. Meanwhile, the mating variation and mating coordinate obtained are incorporated into the assembly variation propagation model, which is then verified by a following case study through a comparison between the calculated results and the experimental results. The comparing results indicate that the established model improves the prediction of assembly accuracy. The developed model enables the investigation of various fundamental issues in variation reduction, including variation analysis, process monitoring, accuracy prediction, and accuracy control.     

Technological form defects identification using discrete cosine transform method

Manufacturers need precise tools to simulate, validate, or improve a process plan for given tolerances. Some simulation methods calculating position and orientation defect of manufactured surfaces have already been developed. A lack in these methods is the integration of form defect of surfaces. Indeed, many methods do not study manufactured surfaces, but nominal models associated to these surfaces. The method developed in this article proposes a tool describing precisely form error in order to take it into account. The work is based on a method of the literature, using discrete cosine transformation, completed by a method for identification of classical defects composing global form error and quantification of their contribution to this defect. The method is validated on simulation examples and then applied on a milled plane.     

Directional textures auto-inspection using principal component analysis

This paper describes a global image restoration scheme using a principal component analysis that can be used to inspect defects in directional textured surfaces automatically. Decomposing the gray level of image pixels into an ensemble of row vectors, the input spatial domain image is transformed into principal component space so that the directional textures are well approximated by first k major components and their corresponding weight vectors, named truncated component solution (TCS). Then the local defects will be revealed by applying image subtraction between the original image and the TCS. This procedure blurs all directional textures and preserves only the local defects that were initially embedded in the input image. These defects, if any, are finally extracted by thresholding. Experiments on a variety of product surfaces with directional textures such as straight, slanted, orthogonal, slanted orthogonal, and oblique linear primitives were conducted to demonstrate the effectiveness and robustness of the proposed method. Furthermore, some preliminary experiments were also conducted to demonstrate the proposed scheme was insensitive to horizontal and vertical shifting, changes in illumination, and image rotation.     

Control by virtual gauge of skew surfaces for correction process

The automation of production has been considerably developed in recent years in all its phases from design and manufacturing to control. The design and manufacture of free form surfaces are a current practice in the industry; thus, the conformity problem of complex geometry parts is felt more and more. The co-ordinate measurement machine (CMM) has largely reduced the acquisition and processing time in the take up measurement operation, which explains the wide use of the CMM in the mechanical industry over the last decade. The modelling and controlling procedure proposed within this paper enables correction of complex surfaces during the parts manufacturing process. The method is based on real surfaces modeled by finite elements starting from cloud points obtained by a digitalizing procedure in the co-ordinate measurement machine. The procedure was applied on a tooth gear which equips the gear box manufactured in the Algeria Engines and Tractors Factory. A comparison between the real and ideal model has been established showing the defects of form in order to correct the grinding process. This comparison is carried out by associating a surface of perfect geometry to a group of palpated dots. The selected criterion of optimisation is the least squares method.     

Inversion of nonperiodic wavefields to determine localized defect structure

A methodology for obtaining high-resolution crystal structure potential distributions from electron microscope images of defects is examined using image simulation methods. The technique is based on the inversion of the multislice procedure when extended to nonperiodic potentials. The sensitivity of the method to the defect model and position in the crystal is examined briefly with the preliminary conclusion that this should not severely limit the applicability of the method. The major impediment for the successful experimental application of the method to the detection and structure determination of single, isolated self-interstitial defects in silicon is likely to be the preparation and fixation of the surfaces of the specimen.     

Development of Automated Fixture Planning Systems

Fixturing is an important manufacturing activity. The computeraided fixture design technique is being rapidly developed to reduce the lead time involved in manufacturing planning. An automated fixture configuration design system has been developed to select automatically modular fixture components and place them in position with satisfactory assembly relationships In this paper, an automated fixturing planning system is presented in which fixturing surfaces and points are automatically determined based on workpiece geometry and operational information. Fixturing surface accessibility, feature accuracy, and fixturing stability are the main concerns in the fixture planning. The system development, the fixture planning decision procedure, and an implementation example are presented in the paper.     

基于装配结构相似的零件三维模型检索方法

提出一种基于装配结构相似的零件三维模型检索方法.将零件功能表面类型、相对位置、朝向以及面与面之间拓扑关系抽象成定性符号集合,并分别用功能面邻接图和定性几何约束图来定义零件功能表面间的拓扑关系和定性空间关系,建立零件装配结构定性模型.功能面邻接图中的环结构和分支结构分别通过断环和圆括弧描述,这样,零件装配结构就可以用字母和数字的线性序列来表示.借鉴化学数据库中化合物的线性符号表示法,对装配结构进行编码,通过装配结构码,建立零件装配结构的检索机制.并以轴承端盖为例验证方法的有效性.     

碳纤维螺旋桨自动铺放成形轨迹规划方法

螺旋桨曲面为高阶自由曲面，故以四丝束碳纤维螺旋桨曲面铺放装置为平台，进行了碳纤维铺放成形轨迹规划的研究。依据船用定距桨桨叶曲面轨迹规划的要求，针对传统曲面网格和轨迹铺放法的低精度缺陷，提出了基于拉普拉斯网格优化、等弧长曲面分层的轨迹优化铺放算法，并完成了算法仿真和实物验证。结果表明，该轨迹规划方法能满足碳纤维螺旋桨的铺放要求。     

基于SDT的公差建模方法及三维公差分析技术研究

探讨了各种公差区域和各种连接的小位移旋量（SD T）的描述方法;研究如何用表面模型来实现对装配体的描述,并且借助于小位移旋量确定装配体任意表面之间的相对位置;根据各个功能要素对于产品的装配功能要求的影响,创建三维公差分析尺寸链,实现面向装配的三维尺寸和几何公差分析;最后给出公差分析实例对所提出的方法加以验证。     

Evaluation of surfaces complementarity based on high definition metrology

The gap between two mating surfaces has a direct influence on sealing performance. However, traditional surface metrology rarely characterizes the gap between two mating surfaces. To solve this problem, a novel concept of surfaces complementarity is proposed in this paper. Surfaces complementarity measures how well two rough surfaces fit into each other. To make this concept applicable in engineering practices, a virtual assembly algorithm is developed. The automatic virtual assembly algorithm aligns the mating surfaces by maximizing the overlap ratio of the surface masks. Then, a sum surface which is complementary to the surface gap is constructed to represent the mating states. The top surface of a cylinder block and corresponding cylinder head surface measured by high definition metrology is mated by the virtual assembly algorithm. The differences of functional parameters between the mating surfaces and the sum surface are discussed thoroughly. Due to surfaces complementarity, parameters of the sum surface has a certain deviation from expected combined parameters from two individual surfaces. A case of square surface shows the practical application potential of the virtual assembly algorithm to optimize the sealing performance of the mating surfaces.     

Real time kinematic solutions of a non-contacting, three dimensional metrology frame

The forward and the inverse kinematics solutions for a three dimensional, non-contacting metrology system are presented. The solutions are a continuation of earlier works involving the derivation of an error budget for the metrology system; however, the earlier work required iterative methods. The kinematic solutions presented in this paper are closed form, providing real-time feedback and making the system practical for metrology applications. The kinematics are developed using a line geometry, resulting in an efficient representation that may describe a general assembly of the kinematic system and error parameters determined through calibration. The inverse kinematics solution is a general solution, assuming very little about the assembly of the system. The forward kinematics solution is less general and requires a few assumptions. In the development of the forward and inverse kinematics solutions, singularities of the system are identified. The resulting analyses provide the kinematics necessary to enable real-time determination of position and orientation of a target in space.     

Evaluation of the form errors of axisymmetric components

The form errors of three-dimensional surfaces have an effect on the proper functioning of an assembly. Methods for evaluating form errors of surfaces of spherical and cylindrical components are available. However, in most machines and instruments the functional components are axisymmetric, e.g. spindles, and have varying geometry in the form of steps, tapers etc. Evaluation of the form errors at different zones on the component as well as a form error characteristic of the whole component is required in such cases. To do this coordinate data or circularity traces are obtained at different horizontal sections using a coordinate measuring machine or a roundness measuring instrument respectively and the required form error is then evaluated. Different algorithms developed for fitting the axis of the component which is the first step in the evaluation of form errors are described. The method of normal least-squares fitting was found to be superior to the general least-squares method.     

Ensuring the reliability of liquid-friction bearings for rolling mills

Means of ensuring the reliability of liquid-friction bearings are considered on the basis of spectral analysis of discrepancies in the shape and mutual position of the contact surfaces and supporting pins of rollers. The amplitude and phase spectra of the discrepancies in shape and mutual position of the basic bearing surfaces within the roller assembly are shown to influence the oscillatory motion of the common axis and the reliability of the components.     

Multi scale modal decomposition of primary form, waviness and roughness of surfaces

This article introduces an innovative method for the multi-scale analysis of high value-added surfaces, which consists of applying a method based on a new parameterization. This kind of surface parameterization refers to natural modes of vibration, and is therefore named modal parameterization. It allows us to characterize the form, waviness and roughness defects of a surface. This parameterization opens up new fields of analysis, such as the appearance quality of surfaces. It is thereby possible to decompose a measured surface in a vector basis, of which vectors are represented by plane natural eigenmodes sorted by frequency and complexity. Different filtering operations can then be produced, such as extracting the primary form of the surface. To analyze the perceived quality of surfaces, these investigations focus on two approaches: that appearance defects have small periodicity, and that there is a link between curvatures and the visual impact of an anomaly. This methodology is applied to two prestige lighters, whose surfaces were measured by extended field confocal microscopy. Moreover, a prospect of this work is to develop an augmented-reality-type monitoring tool for sensory experts.     

Characterization of surface defects in fast tool servo machining of microlens array using a pattern recognition and analysis method

Microlens array (MLA) is a type of structured freeform surfaces which are widely used in advanced optical products. Fast tool servo (FTS) machining provides an indispensible solution for machining MLA with superior surface quality than traditional fabrication process for MLA. However, there are a lot of challenges in the characterization of the surface defects in FTS machining of MLA. This paper presents a pattern recognition and analysis method (PRAM) for the characterization of surface defects in FTS machining of MLA. The PRAM makes use of the Gabor filters to extract the features from the MLA. These features are used to train a Support Vector Machine (SVM) classifier for defects detection and analysis. To verify the method, a series of experiments have been conducted and the results show that the PRAM produces good accuracy of defects detection using different features and different classifiers. The successful development of PRAM throws some light on further study of surface characterization of other types of structure freeform surfaces.     

盾构机管片拼装机振动与摩擦分析

针对盾构机管片拼装过程中,管片拼装机运行到水平或者垂直位置时发生剧烈振动和产生噪声的问题,对管片拼装机的运动机制进行分析,利用ADAMS对管片拼装机进行运动学和动力学仿真分析,得到运动过程中关键部件的受力情况。仿真结果表明:管片拼装机运行至水平和垂直位置出现振动和噪声可能的原因有2种,加工精度不够导致摩擦力过大和接触应力比较大的接触面由于空隙发生碰撞。提出在空隙间增加金属橡胶涂层来减少接触面之间接触应力的解决方案,并通过ADAMS仿真验证该方案能减少接触应力,从而能有效消除振动和噪声。     

手势识别引导机械臂的零件智能装配方法

文中利用手势识别方法,以示教编码的模式完成零件自动抓取的动作,防止装配过程中出现错装、漏装等问题,满足零件柔性装配多样化的需求。首先提出利用SSD模型实现零件抓取位置目标检测方法,根据SSD网络给出当前输入图像的零件种类与位置,缩小零件抓取位置分析范围;然后通过整合空间金字塔池化方法与传统特征提取卷积网络,利用传统的边框检测法,设计了物体角度的抓取参考矩形框方法来代替滑动窗口生成方法,以提高检测速度;最后结合YOLOv3网络进行手势识别,引导机械臂进行零件的智能装配。     

装配误差传递建模及其精度可靠性评估

研究了刚体零件的装配误差建模以及装配精度可靠性评估方法。将零件误差分成位置误差和方向误差,并采用误差矢量加以描述。综合考虑尺寸误差和形位误差,将装配过程误差分成零件内误差和配合面误差,给出主动配合面和被动配合面的定义。针对典型配合面类型,分析误差组成及其计算方法。采用位姿变换矩阵法建立装配误差模型,分析误差在零件内部以及配合面间的传递过程,计算装配总误差并据此确定装配误差的概率分布。给出装配精度可靠度的定义,通过仿真获得装配精度的可靠度指标。以某测量平台的装配过程为例,完成装配过程误差分析和装配精度可靠度评估,验证模型的有效性。     

TOOL DWELL TIME COMPUTATION IN COMPUTER-CONTROLLED SURFACING BASED ON CONSTRAINED OPTIMIZATION

Computer-controlled surfacing has been used for the improvement of the form accuracy of surfaces fabricated by precision machining. This method relies on repeated measurement and correction of the surface form error. Surface correction needs as input the dwell time of all the tool locations on the surface. Past attempts at dwell time computation involved approximations and assumptions with no guaranteed goodness of the solution. This paper proposes to formulate the problem from the perspective of constrained optimization. Practical process limitations and considerations are reflected in the formulation of cost terms and constraints. Commercial numerical packages are available for solving such optimization problem.     

3-D reconstruction of the atomic positions in a simulated gold nanocrystal based on discrete tomography: prospects of atomic resolution electron tomography

A novel reconstruction procedure is proposed to achieve atomic resolution in electron tomography. The method exploits the fact that crystals are discrete assemblies of atoms (atomicity). This constraint enables us to obtain a three-dimensional (3-D) reconstruction of test structures from less than 10 projections even in the presence of noise and defects. Phase contrast transmission electron microscopy (TEM) images of a gold nanocrystal were simulated in six different zone axes. The discrete number of atoms in every column is determined by application of the channelling theory to reconstructed electron exit waves. The procedure is experimentally validated by experiments with gold samples. Our results show that discrete tomography recovers the shape of the particle as well as the position of its 309 atoms from only three projections.

Experiments on a nanocrystal that contains several missing atoms, both on the surface and in the core of the nanocrystal, while considering a high noise level in each simulated image were performed to prove the stability of the approach to reconstruct defects. The algorithm is well capable of handling structural defects in a highly noisy environment, even if this causes atom count “errors” in the projection data.