首页    期刊浏览 2025年06月29日 星期日
登录注册

文章基本信息

  • 标题:Modelling of complex-shaped components in the scope of reverse engineering.
  • 作者:Monkova, Katarina ; Monka, Peter
  • 期刊名称:Annals of DAAAM & Proceedings
  • 印刷版ISSN:1726-9679
  • 出版年度:2008
  • 期号:January
  • 语种:English
  • 出版社:DAAAM International Vienna
  • 摘要:A considerable number of companies in Slovakia produce components on the basis of component drawings which can be drawn manually but nowadays most often by a computer technology. The software using 3D technique notably facilitates the work of constructors and designers. At the design stage of a new product, a 3D model is created at first then the analyses and simulations of the production process are conducted. After the fault elimination, drawing and technological documentation is completed and only on its basis is the component being produced. (Hatala et al., 2008)

Modelling of complex-shaped components in the scope of reverse engineering.


Monkova, Katarina ; Monka, Peter


1. INTRODUCTION

A considerable number of companies in Slovakia produce components on the basis of component drawings which can be drawn manually but nowadays most often by a computer technology. The software using 3D technique notably facilitates the work of constructors and designers. At the design stage of a new product, a 3D model is created at first then the analyses and simulations of the production process are conducted. After the fault elimination, drawing and technological documentation is completed and only on its basis is the component being produced. (Hatala et al., 2008)

The problems occur when the actual component exists, e.g. in the form of a hand-made model, is complex-shaped and its geometric characteristics are unknown. The production of other (exactly the same) components is in this case problematic, especially where a mass production is concerned. The dimension characteristics are essential in order to create a 3D model, drawing or to produce the component itself. Acquiring further data about the component, processing of this data, e.g. by a computer becomes a necessity, and only on the basis of acquired data is the drawing preparation and direct generation of CL data for the component production on NC machine possible.

Nowadays the research regarding the data acquisition on the shapes and dimensions of the real (three-dimensional) objects is in the stage of rapid development. Precise, fast and non-contact methods are significant in many industrial applications including the quality control, surface control or visual systems on assembly lines. (Valicek & Hloch 2006) They are no less important in recognizing 3D objects, in securing the area or in navigation. With respect to the required precision and quality of the subsequent production in mechanical engineering, the most precise and advantageous method of acquiring information on the topography of the real product appears to be a 3D digitization of the examined object, using modern technical devices within the scope of the so-called technological 3D scanning. This method is known as Reverse Engineering.

2. MODELLING OF COMPLEX-SHAPED PART

"Reverse Engineering is the process of the existing component, composition or product duplication without having the drawings, documentation or a computer model at the disposal." Reverse Engineering includes all the activities which enable the determination of the function principle of the product, idea or technology originally used at the product's development. It is possible to use it in order to master the process of design or to use it as a foundation for the process of redesign in the way to make the following feasible (Babjak, 2006):

* monitoring and evaluation of the mechanisms which enable the product's functionality,

* analysis and research into internal processes of the mechanical product,

* a comparison of the existing solutions with own ideas allowing improvement proposals.

Within the scope of Reverse Engineering several techniques of acquiring data on the solid geometry are being used, for instance triangulation (both active and passive triangulation, measuring systems with theodolite, focusing techniques, techniques of "shapes from shading"), optical interferometry, time of flight measurement of modulated light and others. (Hrabovsky, 2002)

The digitization of real objects is possible due to scanning equipment which enables the conversion of the real three dimensional objects into a digital form. The principle of the majority of these equipments is based on the scanning the object's surface in its discrete points and it follows that the digitized object is presented on a computer as a large number of points in space, i.e. the so-called point cloud. Scanners differ from one another especially in the way the scanning of the object's surface points is implemented. The scanning equipment can be divided according to whether the scanning technology is contact or a non-contact one. The former concerns 3D scanners and stationary co-ordinate measurement systems CMM (Control Measuring Machine). This category offers the digitization equipment ranging from 3D desktop equipment to the systems used for measurement of large objects of several meters in size. The latter, non-contact systems of measurement, i.e. scanners, generally operate on laser or optical principle. For the most part, a choice of the scanner type depends on the requirements posed on the accuracy of homogeneity between a real and a digitized model. Further, another important factor when selecting a scanner is the scanning time. The fastest scanners are the laser ones. Also, a significant factor is the size of the scanned component or, possibly the mobility of the scanning equipment. The majority of scanners are limited by the scanning space in which it is possible to scan. 3D scanners are usually constructed to be able to scan objects as large as 50 cm. For more sizeable objects the larger scanning equipments are produced.

Apart from hardware devices a substantial role at digitization of 3D objects is played by software equipment. Individual scanning equipments use own software for processing of the scanned data, however these need to be transformed several times and eventually transferred into a neutral format (IGES, STEP, ...) which CAD/CAM systems can operate with. (Dubravcik, 2005)

An example for the processing of a complex shaped component in Reverse Engineering can be a template for winding of the stator of electromotor (See fig. 1).

[FIGURE 1 OMITTED]

It is a real component, which used to be produced abroad in a way that its finite shape underwent a hand grinding into an anti-template but the drawing documentation of the resultant topography was not available. The average delivery time was longer than 3 month.

A 3D scanner Roland Picza LPX 250, in which all digitizing operations are controlled by a program Dr. PICZA, was selected for scanning. This software enables the scanning of data to be optimized, edited and transferred to NURBS surfaces. Then the data can be exported to STEP, STL or IGES formats for further processing in 3D programs. Considering the surface of the original component was too reflexive (as it was polished), it was necessary to decrease its gloss values, e.g. by spray-painting it with a gray undercoat colour. At the same time, it was essential to evenly apply the sprayed layer as this factor may also affect the approximation rate of a created model toward its original and a finite accuracy of the component created on the basis of a virtual 3D model.

However, after the scanning the virtual surfaces were not even and the transition curves not smooth. Also, the technological elements such as, for instance, holes were not scanned in a sufficient accuracy but only as the surfaces indicating the position of these elements.

For further processing it was necessary to export the acquired data from the software Dr. Picza3 and subsequently import them into Pixform software. This software allowed

* to translate one complex shaped surface through a cloud (a grid) of scanned points, whilst the accuracy of coverage depended on the number of selected checkpoints,

* to modify the polygonal meshes by means of editing control points, polygon edges and surfaces (removing, moving or adding new surfaces),

* a reduction of a polygonal meshes, i.e. a reduction of the number of polygons in the meshes, however, at the expense of the quality and display fidelity,

* to fill the cracks which arose at the scanning in polygonal meshes on the basis of the NURBS surface definition, and repair a partially insufficient representation of the scanned data,

* to partially polish the obtained model, however, not with a sufficient accuracy

Geometric data describing the established surface were neither applicable for various types of analyses, nor for a CL data generation. As a result, it was necessary to export them again in IGES, STEP or STL format and import them to a selected CAD/CAM system, in order to further process them.

Pro/Engineer system was selected as a CAD/CAM system based on the experience and in connection with the software availability. It is advantageous to utilise the surface operations in this system for work with complex shaped design. For this purpose, sections were created on an imported model and the interpolating or approximating curves, defining the profile of "top surface" in the section plane, were translated through a point set via a mathematical apparatus. Approximating curves of Bezier and Spline types were used most often at work as they best represented the imported template shape in the parallel planes. The curves were covered with a coat surface which was created as Pro/Engineer system's own element so the geometric data describing this surface were readable also for CAM system area. As with the curves, it was also possible to control, analyse and modify the curvature and "smoothness" of the selected surface.

[FIGURE 2 OMITTED]

In the process of finalising of a 3D model version, various techniques and tools were used, which a user is offered by a selected CAD/CAM system Pro/Engineer. The final version of a 3D model created without geometric and drawing definition is on the Fig. 2.

3. CONCLUSION

Created 3D model was compared by geometry with the scanned shape in Pro/Engineer. The spaces between measured points were 1 mm and the tolerance was 0.1 mm. It can be said that 98 % of the surfaces were inside tolerance. On the basis of created 3D model it was made the new physical part by Rapid Prototyping method. The cast plastic part was compared with the original real part by means of 3D measuring equipment and then it was used in real situation. It was possible to allege that the model corresponds to the real steel part in required accuracy and so CL data were generated. The obtained data were transformed and applied as NC program for controller of concrete CNC machine in company. After the creating of 3D model and after the generating of NC program, the terms of delivery were shortened about 98 % (from 180 days on 2-5 days), the number of stored templates decrease about 50 % and the price of the parts made in Slovakia derogated about 60 % compared with original foreign supplier.

The creation of a 3D model also allows conflicting situations predicting not only at the machining but also at their putting together to assemblies resulting in the reduction of the preparation time, expenses and the quality enhancement of production. One modelled component can thus become the basis for its simple modification and subsequent production of other, type similar components, e.g. within a group technology. The company uses several types of these templates and so it will be created other of them by similar manner in the future.

4. REFERENCES

Babjak, S. (2006). The planning of reverse engineering in quick product development system I., Transfer of innovation, 9/2006, p. 59-61, ISSN 1337-7094.

Dubravcik, M. (2005). The equipment of digitizing, Transfer of innovation, 8/2005, p. 52-54, ISSN 1337-7094.

Hatala et al. (2008). Application suitability of power-jet cutting technologies for constructional steel, Scientific Bulletin: Machine Manufacturing Technology, vol. 22, serie c, 5/2008, p. 211-214, ISSN 1224-3264.

Hrabovsky, M. (2002). Optical methods in experimental mechanics, Proceedings of the 40th International conference experimental stress analysis, CTU in Prague, ISBN 80-01-02547-0, Czech Republic, 6/2002, Prague

Valicek, J. & Hloch, S. (2006). The optical identification of surface topography machined by abrasive waterjet, Fine mechanics and optics, vol. 51, no. 11-12 /2006, p. 320-322, ISSN 0447-6441
联系我们|关于我们|网站声明
国家哲学社会科学文献中心版权所有