外文翻译--浅析CAFD中的自动装夹规划系统

Analysis of automated modular fixture configuration design systemRONG Yi ming，LI Jie，MA Wei-dong（Worcester Polytechnic institute，MA 01609，U. S. A）Abstract To counter the fixture planning of computer-aided fixture design (CAFD) an automated modular fixture configuration design system is developed. Having fixture surface accessibility analysis as the core, the optimum selection of fixturing surfaces and points on workpiece is fullfilled. Firstly, several basic criteria for evaluating the eligibility of a surface being a preliminary candidate fixturing surface are discussed. Secondly, by applying a discretization technique, an accessibility model of fixturing surfaces is established based on an overall evaluation of the accessibility of discrete points on the surface. Finally, the implementation issue and an analysis/design example are presented.Key words computer-aided fixture design; fixture planning; fixturing surface1 INTRODUCTIONImportant manufacturing activity in the production cycle. Computer-aided fixture design (CAFD) technique has been developed and become part CAD/CAM integration. The development of CAFD contributes to the reduction manufacturing lead time, optimization of manufacturing operations, and verification of manufacturing process designs. CAFD plays an important role in flexible manufacturing system(FMS) and computer-integrated manufacturing system (CIMS) Figure one outlines the activities of fixture design in manufacturing systems which basically include three major aspects: setup planning, fixture planning, and fixture configuration design. The objective of setup planning is to determine the number of setups needed, the position and orientation of workpiece in each setup, and the machining surfaces in each setup. Fixture planning is to determine the locating, supporting, and clamping points on workpiece surfaces. The tasks of fixture configuration design is to select fixture components and place them into a final configuration to fulfill the functions of locating and clamping the workpiece. An automated modular fixture configuration design system has been developed where when fixturing surfaces and points are selected on the workpiece model fixture units are automatically generated and placed into position with the assistant of fixture component assembly relationships. In the development of automated fixture planning, it is desired that the fixturing surfaces and positions on workpiece be selected automatically. As shown in Figure2, several factors which attribute influences on fixture planning should be taken in- to consideration, ie., workpiece geometric information and operational information need to be extracted and retrieved， accuracy relationships and surface accessibility of workpiece need to be analyzed, fixturing stability and easiness of workpiece loading/unloading operation need to be verified. In this paper, the research focus is on resolving the problem of fixturing surface accessibility analysis.2 BASIC REQUIREMENTS ON FIXTURING SURFACESFixturing surfaces are the surfaces on workpiece used to locate and clamp the workpiece where functional fixture components (locators and clamps) are in contact with these surface. As the focus of this research is on analyzing the accessibility of fixturing surface on workpiece, first of all, the study is started with the discussion on the basic requirements for a surface on workpiece to be eligible as a preliminary candidate fixturing surface. In automated fixture planning, once the primary locating direction is determined in setup planning, the accessibility property of each candidate fixturing surface should be assessed so as to help the fixture planning fulfill the optimum selection of fixturing surfaces and point distributions. In this research, the accessibility analysis is investigated briefly on the basis of pure geometric information of the workpiece and its surfaces which can be extracted from the CAD solid model. Other information such as surface finish and tolerance are excluded from concern because they are the factors considered in the accuracy analysis of fixture planning.On a complex workpiece, some surfaces might be obviously ineligible to be the candidate fixturing surfaces and should be filtered out at first. In our current research, only the surfaces which satisfy the following basic requirements can be selected as preliminary candidate fixture surfaces, imachining surfaces,planar surfaces,surfaces with accessible normal directions, and surfaces which are large enough. The main purpose of identifying these requirements is to filtez out those obviously ineligible surfaces on workpiece and assume all re-maining surfaces as preliminary candidate .fixturing surfaces.2.1 Non-machining SurfacesIn a real fixture design, it is well known that the surfaces to be machined at one setup should not be used as fixturing surfaces and hence are definitely inaccessible to any fixture component Therefore a candidate fixturing surface must be a non-machining surface.2.2 Planar SurfacesThe fixturing surface types are commonly divided into the planar surface type and the cylindrical surface type. However, the accessibility analysis approach developed in this research is limited to planar surfaces on workpiece, though the method presented may be applicable to cylindrical surfaces. In many cases, planar surfaces are selected as fixturing surfaces with high priority in fixture design.2.3 Surfaces with Accessible Normal DirectionIn most fixtures, the primary locating surfaces is perpendicular to other locating surfaces, which can be defined as the bottom-locating and side-locating form while the common clamping forms are top-clamping and side-clamping. This assumption is especially true when modular fixtures are employed in production. For these locating and clamping forms, a constraint is valid that the side-locating direction n S and side-clamping direction nsc are perpendicular to the bottom-locating direction nL, and the top-clampin direction nTC is negative to the bottom-locating direction, and nsc are all normalized vectors and could be regarded as the accessible directions in one setup. Generally, nBL is always set equal to the primary locating direction which is obtained from setup planning. Unlike curved surfaces, a planar surface on workpiece has a unique normal direction, which is written as n.r. If n.t is not coincident with any accessible directions mentioned above, the surface is regarded inaccessible to fixture components and ineligible to be the candidate fuxturing suface. In another word, the fixturing surface should have an accessible normal direction.2.4 Surfaces Which are Large EnoughIt is a common sense of fixture design that the surfaces with too small size or a too slim shape are also inelegible to be the candidate fixutring surface. To roughly determine whether the size of a surface is large enough for fixturing, a simple rule can be applied. The eule states that a surface is eligible in size if the smaller edge length of its bounding rectangle is bigger than a threshold lT. The value of lT is set based on the sized of fixture components used in fixture de-sign, which can be specified and modified by user.After filtering out apparently ineligible surfaces according to above requirements, the remaining surfaces on the workpiece can be regarded as the preliminary candidates of fixturing surface and their accessibility properties needs to be evaluated.3 ACCESSIBILITY ANALYSISFixturing surface accessibility is a vague concept, which is associated with the fixture components used in fixture design. To determine accessible to a regular fixture component whether a candidate fuxturing surface of workpiece is and figure out a numerical value to represent the corre.A. Geometry of the fixturing surface which contains the information of surface area and shape B .Possible obsturctive workpiece geometry along the normal direction of fixturing surface or around the geometric region of fixturing surface.C. The size and shape of functional fixture components.Factor A merely refers to the geometric representation of the fixturing suface. In a feasible fixture design, the selected fixturing points usually locate inside the region of fixturing surface and the contact area between this region and the fixture component should be over half of the area of relevant functional surface of the fixture component. In fact, the accessiblility analysis result should reflect the real effective accessible area of the fixturing surface especially when there exists obstructive workpiece geometry along the normal direction of fixturing surface or around the geometric region of fixturing surface.Factor B also greatly affects the actual accessibility of the fixturing surface because possible obsturctive workpiece geometry along the normal direction of fixturing surface or around the geometric region of the fixturing surface may block the approaching of the fixture component to fixturing surface in some sub-areas of suface region and hence lead to a decrease of the effective accessible area. For a workpiece as shown in Figure4, even though the face F, is large enough in size and not complex in shape, its accessibility to a regular fixture component reduces a lot because of the inherent obstructive geometry of workpiece.It is obvious that the accessibility analysis can not be made without considering the functional sizes and features of fixture components. To obtain a more accurate evaluation of accessibility to guide the later fixture configuration design, factor must be involved into the comprehensive analysis. However, in the real circumstance, before the fixture configuration design is finished, the fixture component selected from the library is unknown at the stage of fixture planning. To circumvent this problem, a least accessing unit size, T, is applied to represent t卜e minimum functional size of fixture components, which can be specified and modified by user. It implies that if a fixturing suface is accessible, at least a fixture component with the functionally bounded surface size of Tx T can be placed in contact with the surface.To establish the accessibility model for a fixturing surface, several basic facts of evaluating the property of the accessibility are considered.1)With the same shape and no obstruction along/around the surface by the workpiece, the surface with larger area will have a higher accessibility value.2) With the same area and no obstruction, the surface with simple shape complexity will have a higher accessibility value.3) With the same area and same shape, the surface with less obstruction along or around it will have higher value.In complex gion. It practical accessibility situations, it is very possible that the planar surface of the workpiece has a shape and fully/partially obstruction along its normal direction or around its geometric reis thus required tinned above date surface that a that the accessibility model should reasonably comparable accessibility comprehensively reflect the facts men- value can be applied to every the workpiece no matter how complex the grometry of the surface might can be.A discretization modeling method is preferred since it is generic in principle and the algorithm is easy to implement on computer. The methodology is made up of three steps: 1)The surface is sampled into a set of discrete point, 2) Both individual and neighbor related accessibility of each sample point is assessed, and 3) The overall accessibility of the surface is evaluated based on the results of all sample points.3.1 Surface Discretization As the accessibility analysis is prior to the fixture planning, the accessibility model between an arbitrary planar surface and fixture component surface is difficult to be established if the fixturing point is undetermined. Before the final position of workpiece on the baseplate is settled down, the possible candidate fixturing points on a planar surface may be enormous in number and hard to handle by a continuous model. Thus in our approach, the surface is sampled into grid-arrayed discrete points with equal interval length T. In order to make the sampling algorithm generic, the outer-bounding rectangle of the surface is used as the sampling region instead of the surface region itself. When a certain set of fixture components are used in fixture design, T can be reasonably selected in terms of the smallest functional surface size of fixture componentsThe outer-bounding rectangle of a planar surface can succinctly information of the exterior shape limitations, and also is very helpful for provide the geometric restraining the sampling region and enabling the sampling algorithm more generic. The rule for extracting the outer bounding rectangle is very simple. For a bottom-locating/top-clamping surface which normal direction ns is identical/negative to two edges of the outer-bounding rectangle must be paralled to X axis and two other edges parallel to Y axis since the bottom-locating direction is identical to the negative vector of Z axis in the workpiece coordinate system, as illustrated in Figure. For a side-locating/clamping surface where ns_n, there must be two edges parallel to Z axis, while the other two edges should be parallel to the cross product of ns and nBL, as shown in Figure Sb. In such a way the surface can be sampled into a set of discrete points within the outer-bounding rectangle, as shown in Figure 6. Some points are sampled outside the outer-bounding rectangle. In this approach, these points are not exorbitant because they may be useful to check the possible obstructions around the fixturing surface.3.2 Point Accessibility (PA) of Sample Fixturing PointIn our model, the surface accessibility is a statistical value based on the Point Accessibility(PA) of every valid sample point. PA consists of two parts: the point Self Individual Accessibility (STA) and the point Neighbor Related Accessibility (NRA). The SIA is mainly corresponding to the isolated accessibility of the fixturing point while the NRA reflects the extended accessibility of the fixturing point. A sample point cab be regarded valid if it is tested to be at least not inaccessible to a fixture component with a functional surface size T x T. The definitions and calculation methods of SIA and NRA are given below.SIA of a virtual sample point is defined on the basis of three attribute tags which are separately assigned by s, representing the position status of a sample point on the surface, representing the obstruction status of the surface in the normal direction at the sample point, ands, representing the contact area matching extent in the text area.The tag of s, is used to indicate whether the square tset grid with a center at current sample.浅析CAFD中的自动装夹规划系统融亦鸣， 李杰， 马卫东（伍士德理工学院，美国 01609）摘要 针对计算机辅助夹具设计(CAFD)中装夹规划问题，开发了自动装夹规划系统，并以工件装夹表面的影响性分析为核心，完成工件上装夹表面及点的优化选择:讨论了几个用以评佑候选装夹表面的适合性的要点；利用离散技术，基于所有表面离散点影响性的评佑，建立了装夹表面影响性模型；并给出了系统运行要。汽和分析设计实例。关键词 计算机辅助夹具设计；装夹规划；装夹表面1. 简介生产制造工艺活动在生产周期中很重要。计算机辅助夹具设计工艺已经发展起来并且成为CAD/CAM集成中的一部分。计算机辅助夹具设计的发展主要有助于减少机械加工的时间和产生最佳的机械制造方法，还有验证最佳的制造工艺设计。计算机辅助夹具设计在柔性制造系统中和计算机集成制造系统起着很重要的作用。设备的设计在机械加工系统中，计算出主程序数据主要包括下面三方面：客观的装备设计主要决定于零件的需求，工件的定位分析，每个零件的表面粗糙度。夹具的设计主要取决于工件的定位方式，支撑方式，夹紧位置及工件的表面情况。夹具结构的设备选取和安装设计主要取决于工件的定位和夹紧方式。自动化的模块夹具结构设计系统已经发展到设备表面和位置能随工件的变化而变化。夹具单位自动生成和放到正确的位置和辅助夹具有很紧密的关系。在自动夹具设计发展中，夹具的表面和位置关于工件表面需要自动化。工程师必须考虑到好多影响夹具设计因素，如工件的表面几何形状，加工时候的震动等因素。准确关系和工件表面粗糙度的分析。夹具的稳定性和工件夹装方式的简单可靠都必须验证。在这篇文章中，就是研究关于自动夹具装夹表面影响的。很少能找到相应的文学报告。为了工件的装夹方式和验证夹具的设计，一种分析方法已经产生来验证工件安装在夹具上和工件在夹具上的拆卸的容易度。一种从不同角度的方式用来评价夹具设计装夹表面影响性。它限制于一个简单的表面影响性例子。夹具可包括两方面：夹具屋面的易用性和工件装卸易用性。前者是一种在混合的设计中缓解了表面性质的评估程度，如下，在表面放置一个夹具元件(定位或夹子)是很容易的。con- tact表面是一个非常重要的选择标准的固定准备表面。后者是用来判定当夹具被设计和建造好了，一个工件是否能容易得放进去。迄今为止，在文献中还没有找到一种广泛应用的夹具可分析方法。在这篇文章中， 提出了一个准确评估夹具可表面的方法。这种方法只能用在那些经常应用的夹具表面。第二部分，讨论了一些夹具表面的基本要求。第三部分，提出了一个广泛接受的夹具表面。一个离散化造型方法用来判定分析夹具表面发展的实施性问题和分布特性。最后，一个分析的例子被提出来了。2.对夹具表面的基本要求夹具表面是在工件定位和夹具工件时夹具元件(定位器的功能及夹子)所接触的这些表面。本研究的焦点是分析夹具的设计，卡具表面上的工件，首先，文章以工件表面被初步评定为“合格候选人”的夹具的表面的基本要求开始的，主要定位的方向在设置的规划中决定了。这些被评定为“合格候选人”的夹具表面的容易得到的性质应该用来作为评定满足夹具表面和点分布的最佳选择。在这项研究中，被接受的分析是在工件的简要纯几何信息基础上的简要考察和从CAD模型中被提取的表面。另外的信息例如表面的光洁和公差是不用考虑的，因为它们是作为精度分析夹具的计划要考虑在内的因素。在一个复杂的工件，一些表面明显的不合格去成为工件表面候选应首先被去除。在我们最近的研究中，只有那些符合基本要求的表面能够选作最初的候选表面:1、未加工的表面；2、平面表面；3、接近正常方向的表面；4、足够大的表面。识别这些要求的主要目的，挑选出那有些明显的资质的工件表面并且假象所有的表面作为初步的候选人夹具的表面。2.1未加工的表面在真正的夹具设计中，那些在装备中加工过的表面不能作为夹具表面是众所周知的，因此绝对无法应用于任何夹具元件。因此，一个合格的夹具表面必须是一个未加工的夹具表面。2.2平面表面夹具表面的模型备分为平面表面类型和圆柱体表面类型。然而，研究中接受的类型只有平面表面类型。不过这种放发对于圆柱体表面类型还是适用的。在一些方案中，平面表面类型选作夹具表面在夹具设计中带有高优越性。2.3接近正常方向的表面在众多的设备中， 主要定位表面与定位表面是相垂直的。定位表面是那些底和边被定位的定义。通常的夹紧形式是顶夹紧和变夹紧。当模块化的夹具被雇佣来生产时，这些设想是尤其正确的。对于那些夹紧和定位形式，一个人约束条件是有效的，边定位和边夹紧的方向垂直于底定位的方向，顶夹紧的方向与底定位的方向是相对的。国家安全委员会规定所有的归一化向量并且能够作为一种设备的可接受的方向。一般的，NBL总是设为主要的定位方向。不像弯曲的表面，平面表面有独一无二的正规方向，称为nr。如果nr与前面提到的那些被认定接受的方向不相符和，这些表面就不能成为合格的候选夹具表面。换句话说，夹具表面应该有一个被普遍接受的方向。2.4足够大的表面那些太小的和形状太细长的表面是不能作为夹具表面候选是夹具设计的常识。粗略估计一下这个表面是否足够大去做为一个夹具表面，一个简单的规则就出来了。这个规则表明了，表面的边框的边缘长度比入口门槛大那它在形状方面是合格的。 IT 的价值是建立在夹具设计的被使用者指定和修改的夹具零元件上的。根据前面的要求过滤显然无资格的表面之后，这些剩余的工件表面成为初步的合格的夹具表面。并且它们的那些容易得到的性质需要被评价。3.表面影响性分析自动夹具设计装夹表面影响性是一个模糊的理论,它与用于设计固定装置的固定装置组成成分紧密相关。要确定某普通固定装置的组成成分是否可行,要看该备选零件的表面情况并计算出表面影响性值。以下几家公司设计情况如下：A是包含装置表面面积和形状方面信息的固定装置表面的几何图；B可能的围绕固定装置表面普通方向或表面周边几何区域的阻碍性零件几何图；C有效固定装置组成成分的大小和形状。 公司A仅仅考虑到该固定装置表面的几何再现图。在一个可行的固定装置设计里,被选固定装置点通常安装在固定装置的内部区域,接触区在该区域之间,该固定装置组成成分应该在该装置相关作用表面的半上方。实际上,影响性分析的结果应该能反映固定装置表面的真正有效接触区,尤其是当存在沿着该装置表面普通方向或表面周边区域的阻碍性零件几何图时。公司B也从很大程度上考虑到了固定装置表面的实际影响性。因为沿着该装置表面普通方向或表面周边区域的可能性的阻碍性零件几何图很可能阻碍固定装置表面的一些次要区域的固定装置组成成分接触装置表面,从而导致有效接触区的减少。对于一个像数据4所展现的零件,尽管F面足够大,并且形状也不复杂,但由于零件本身的几何结构障碍,它与一般固定装置组成成分的结合能力大大降低了。很明显,表面影响性分析不可不考虑作用尺寸和固定装置组成成分的特征。为了获得一个更准确的表面影响性评估从而更好的指导接下来的固定装置结构设计,公司必须要进行综合性分析。然而,实际情况下,在固定装置结构设计图完成之前,从库存里选出来的固定装置组成成分在装置设计过程中是不知道的。为了避免这个问题,至少要通过组合尺寸,T是用来代替固定装置组成成分的最小作用尺寸,它能被使用者特殊化和修改,它暗含如果一个固定装置表面是可接受的,至少固定装置组成成分和相关的作用表面要是TX，T能被相关表面确定。 要建立一个固定装置表面影响性模型，要适当地考虑以下几个基本的评估事实:1)同样的形状，零件表面的周边没有阻碍，更大面积的表面将会有更好的接触性2)同样的面积且无周边阻碍,简单形状的表面将会有更好的接触性。3)同样的面积和形状，表面周边的阻碍越少，接触性越高。 在复杂的祗园，事实很可能是，平面工件有形状和完全或部分阻力沿其正常的方向或绕着它的几何形状，从而要求上述日期必须表面镀锡。这种适当的模型应该合理可比无障碍地全面反映人的需求,它可以适用于每个工件，无论多么复杂的金属扣眼表面都有可能。阿离散建模方法是首选，因为它是一般原则而且该算法容易在计算机上实现。这种方法包括三个步骤：1）表面取样到一套离散点；2）对单独和易于关联的相邻样点进行评估；3）无障碍环境的表面总体评价的结果基于所有采样点。3.1 表面离散作为辅助分析之前的夹具规划，如果夹具点未定，无障碍模式之间的任意平面和夹具元件表面是很难确定。在最终定位前，工件基板是固定下来的，可能的表面候选装夹点可能是数量巨大的而且难以用连续模型处理。因此，我们的做法是在表面取样到网格离散点一排平等间隔长度T。为了使采样算法通用，外一包围矩形表面被用作抽样区域，而不是表面区域本身。当某一套夹具元件用于夹具设计，T可以合理地从最小的功能表面尺寸中选择夹具元件。外一包围矩形平面是非常有益的几何约束提供的抽样地区，它可以简明扼要地提供外部形状的限制信息，使采样算法更加地通用。提取外包围矩形的规则是非常简单的。由于以自下而上定位/顶端夹紧表面法线方向是一致的纳秒/负向两个边缘外一包围矩形，所以它必须与X轴平行，而底部定位的方向是一致的消极载体的Z轴的工件坐标系，因此其他两个边与Y轴平行，如图所示。对于一方一定位/夹紧表面下ns上n，必须有两个边与Z轴平行，而其他两个边应平行于向量积的NS和NBL，以这种方式可以在表面取样到一组离散点的外一包围矩形，采样的某些点在外一包围矩形之外。在此方法中，这些问题并不过分，因为他们可能有助于检查装夹表面周围可能的障碍物。3.2 点无障碍（ PA ）的样品夹具点在我们的模型中，表面是一个无障碍的统计值点基于无障碍（ PA ）的每一个有效的采样点。PA由两部分组成：孤立的点自我无障碍（ STA ）和点周围相关无障碍（NRA）。SIA主要是对应于孤立的无障碍的装夹百分点，而NRA反映了延长无障碍的装夹点。样本点驾驶室被视为有效，如果是测试，至少不会无法获得功能表面大小为T x T的固定元件。SIA和NRA的定义和计算方法如下。新航虚拟采样点的定义的基础上，三个属性标记分别指定由S ，代表的立场地位的采样点在表面上，代表梗阻地位表面的法线方向的采样点，阿富汗国家发展战略，代表接触面积匹配程度的文本区。该标记的S ，是用来说明是否平方米测试电网的中心，目前的样本荣坐标白元组合夹具元建模和装配关系分析自动夹具配置设计。