Turn:

This paper introduces how to use UG to carry out reverse design in reverse engineering. Generally, the measured data point cloud is input first, and then the surface is constructed according to the data points. Finally, the process and method of constructing the plate into a solid are introduced.

I. Introduction

Traditional product design is generally a "from scratch" process. The designer first conceives the appearance, performance and approximate technical parameters of the product, then establishes the three-dimensional digital model of the product by using CAD, and finally transfers the model into the manufacturing process to complete the whole design and manufacturing cycle of the product. This process can be called "forward design". Reverse engineering is a process from scratch, that is, according to the existing product model, the design data of the product, including design drawings and digital models, are derived reversely.

The professional software of reverse engineering includes Surfacer, ICEM, CopyCAD and RapidForm, which are very suitable for processing a large number of scanned point cloud data. For example, when a car is scanned by laser, about 300,000 measuring points can be obtained. Through the construction of professional Surfacer software, a digital model is obtained, and the expected effect is achieved. At the same time, we also explored the application of UG in reverse engineering, and gained some experience in the process, which is described as follows.

Second, the input of data points

Using UG software to do reverse engineering, the measuring equipment used is mostly contact manual three-coordinate marking machine, which mainly measures cross section, outline and characteristic line. There are not many measurement data points, and UG is easy to handle.

However, the car model in this paper has as many as 300 thousand data points measured by laser scanning, so it is very difficult to input so many data points into UG, so we preprocess the point cloud data in Surfacer software. In order to accurately retain the original feature points and contour points, we usually construct contour lines and feature lines and import them into UG together with point cloud data, as shown in figure 1.

Figure 1 input data

Thirdly, the curve is constructed by points.

1. In the process of connecting lines, the characteristic line points are usually connected first, and then the section points are connected. Have a reasonable plan before connecting. It is necessary to determine how to divide the plane according to the shape and characteristics of the car, so as to determine which points to connect, and to have a clear idea of the future construction methods. The connection error is generally controlled below 0.4 mm

2. There are lines, arcs and splines in common use, among which splines are the most commonly used. Generally choose "Pass"

Point "

Mode, the best order is 3, because the higher the order, the worse the flexibility, that is, the deformation is difficult, the subsequent processing speed is slow, and the data exchange is difficult.

3. Because of the measurement error and the unsmooth surface of the model, it is necessary to adjust the connected spline curve, otherwise the constructed surface will be unsmooth. A common adjustment method is editing.

Spline, commonly used editing

Pole options, including moving, adding control points and controlling the pole to move in a certain direction, are convenient for editing splines. In addition, the curves are separated, bridged and smooth.

Spline) is also often used.

In short, a lot of line adjustment work needs to be done before generating the surface, and curvature comb can be used for analysis during line adjustment to ensure the quality of the curve, as shown in Figure 2.

Fig. 2 Structural curve

Fourth, build the surface.

Because the car body requires smooth appearance and smooth outer surface, when constructing a surface, it should be divided into several surfaces, especially to ensure that the tangency or curvature between the surfaces is continuous, so as to form a surface without seam traces. In addition, in the surface construction, we should choose the appropriate construction method according to the specific situation.

1. Method of constructing surface

The most commonly used construction method of (1) is the through curve.

Mesh can not only ensure the continuity of curvature of surface boundary, but also control the curvature (tangency) of four peripheral boundaries, although

A curve can only guarantee the curvature of both sides.

(2) nxn command is widely used, which can dynamically display the surface being created, or increase or decrease the defined curve string at any time, and the surface will change accordingly. Similarly, it can keep continuity with G0, G 1 and G2 of adjacent faces.

(3) Trilateral surfaces and pentagonal surfaces are often encountered when constructing surfaces. Generally, a curve is made to transform a three-sided surface into a four-sided surface, or the boundary line is extended to transform a five-sided surface into a four-sided surface to reconstruct the surface. Among them, on the surface, spline (curve) is made.

"On the surface" and "Trim" are two commonly used commands, as shown in Figure 3.

The triangular surface constructed in Figure 3

(4) After the external surface structure is completed, mirror reflection shall be carried out. Protrusions often appear in the center of the surface, which makes the surface appear unsmooth. Generally, the center of the surface is cut off and then bridged to make it smooth.

(5) When constructing a surface, a "crease" often appears between two surfaces, and the surface is not smooth, mainly because the tangents of the two surfaces are discontinuous. We can solve this problem completely.

Curve mesh sets the option of boundary tangent continuity, or you can choose matching after constructing the surface.

Edge(NX3) option can match the boundaries of two surfaces, thus making the curvature continuous. In addition, even if two surfaces are not connected, they should be matched.

The Edge command can also automatically extend the boundary of one surface and overlap the boundary of another surface.

(6) When constructing a single and relatively flat surface, point clouds are directly used to construct the surface (from points

Cloud) will be more convenient and accurate. Sometimes the gaps between surfaces need to be bridged to ensure the smooth transition of surfaces. When the surfaces intersect, fillet processing will also make the two surfaces smoothly transition.

2. Several problems that should be paid attention to when constructing surfaces.

(1) The key to constructing a surface is to grasp the characteristics of the sample, and at the same time, it needs to be concise, and the surface area should be as large as possible, not too many. In addition, in order to improve the efficiency of modeling, reasonable facets are needed.

(2) In the process of constructing a surface, it is sometimes necessary to add some lines in order to construct a surface. Connecting lines and constructing surfaces need to be done alternately. After the surface is created, it is necessary to check the error of the surface. Generally, the error from point to surface should not exceed1mm.

(3) The order of structural planes should be as small as possible, and it is generally recommended to be the third order. Because high-order slicing reduces the possibility of successfully exchanging data with other CAD systems, other CAD systems may not support high-order surfaces. The higher the order, the greater the rigidity of the plate, and the farther the surface deviates from the pole, which makes it difficult to edit the surface at the pole. In addition, the low order is also conducive to adding some features such as rounded corners, slopes, thickening, etc., which is conducive to the next programming and processing, and improves the speed of subsequent generation of NC machining tool paths.

Verb (abbreviation of verb) structural entity

After constructing the outer surface, it is necessary to construct a solid digital model. If the model is simple and the curvature changes little, we can sew them into a whole, and then use the thickening instruction to establish the entity, but in most cases it is impossible, especially for the model in this paper.

If the outer surface is stitched into a whole, and then the car bottom sheet is turned into a closed sheet, thus becoming a solid, but because the curvature of the car bottom surface changes greatly, it is often impossible to realize the shell extraction command. Therefore, it is necessary to offset the sheet on the outer surface first, then construct the cross sections of the inner and outer surfaces, and finally sew the cross sections with the inner and outer surfaces to make them into closed sheets, thus automatically transforming them into solids. This process generally includes the following four aspects.

1. Surface deviation value

Use the Offset command to select multiple faces or all faces at the same time to improve efficiency. The deviation of each workpiece on the outer surface of the trolley is shown in Figure 4.

Figure 4 Surface deviation value

The arrow in Figure 4 indicates the direction of the offset value. If the arrow is reversed, just enter a negative value.

Not any surface can realize the offset value, and there are several reasons why the offset value cannot be realized:

(1) Because the curvature of the surface itself is too large, the normal of the basic surface suddenly changes;

(2) The deviation distance is too large, which leads to self-intersection after deviation, leading to deviation failure (in this paper, the car body deviates by 2 mm);

(3) The quality of the partial surface is not good, and some areas have ripples, so it can only be modified before the partial surface;

(4) Some surfaces look good, but they just can't be biased. In this case, an extract can be used.

Before offset, the geometry is converted into a B-surface.

The above four situations may be encountered when constructing surfaces, so you should learn to analyze the reasons.

2. Surface stitching

It is necessary to trim or repair the offset surface, construct the inner surface by various surface editing methods, and then sew the inner surface and the outer surface. When sewing, sewing often fails. Generally, there are the following possibilities.

(1) When sewing, there is too much deviation. You can only sew a few pieces at a time, and you need more stitches.

(2) The stitching tolerance is less than the distance between adjacent edges of two stitched surfaces. When encountering such problems, it is generally necessary to increase the sewing tolerance before sewing.

(3) After extension, the two faces cannot intersect, and the shapes of the edges do not match. If the sheet is not a B-surface, it needs to be converted into a B-surface before sewing to match the edge of the corresponding other sheet.

(4) The edges are slightly deformed or have other geometric defects. Can be enlarged locally, surface analysis, inspection of geometric defects. If there are geometric defects, the patch can be modified or reconstructed and then stitched again.

3. The effectiveness of stitching

Finally, it should be noted that although the sewing command was executed, the computer did not give an error prompt. It seems that the sewing is successful, but it may not be. Some films will have bright spots or lines or even gaps when enlarged after splicing. Therefore, it is necessary to check the effectiveness of stitching immediately after the stitching is completed. If there are bright spots or bright lines on the suture line, it means that this part has not been stitched successfully, and the stitching operation must be cancelled and stitched again, otherwise it will bring difficulties to the subsequent solid modeling work, but if only the periphery is highlighted, it means that the stitching is successful, as shown in Figure 5.

Fig. 5 stitching of inner and outer surfaces and cross sections

4. Generate entities

Sew the inner and outer surfaces and cross section into a closed sheet, which will automatically turn into a solid, as shown in Figure 6.

Figure 6 Entities

Abstract of intransitive verbs

In a word, it is time-consuming to do reverse design with UG software, which is not as convenient as professional reverse software (such as Surfacer, Rapidform, etc.). ) in dealing with point clouds and building tiles. However, some ideas and methods of UG in reverse design, such as how to construct lines and surfaces, how to plan according to product characteristics and how to divide products, will also help to use professional reverse software in reverse design. In fact, this is also the key to reverse design.