DE102017002158B4 - Procedure for the target-actual evaluation of a component against CAD data - Google Patents

Abstract

Method for evaluating (target-actual comparison) or detecting deviations of an actual contour (9), which is the contour of a workpiece (1) to be produced, from a nominal contour (10), which is the contour of the CAD model The method for developing an electrode (2) for electrochemical removal is characterized in that the contours (9 and 10) overlap, that an error vector (8) is always perpendicular to the inner (smaller) contour (9, 10) is viewed from and points in the direction of the outer contour (9, 10), the error vector (8) standing vertically on a surface at a position of a measuring point or a nominal point.

Description

The invention relates to a method for target / actual evaluation of a component against CAD data, in particular for evaluating components such as turbine blades from aviation.

The measurement of components with a high proportion of free-form surfaces such as turbine blades is increasingly being measured optically today. In addition to the increasing availability of devices with the corresponding accuracy, the main reasons for this are the cross-component, area-wide and fast detection of a large number of measuring points (point cloud), which enables a holistic assessment. The visualization of the measurement results usually provides a TARGET-ACTUAL comparison of this measurement data (ACTUAL) against a CAD model (TARGET), in which the deviation of the measurement results is visualized by means of a false color representation, in which a deviation value is assigned to each color value.

According to the state of the art, there are different variants for calculating the deviations. One of the best known is the principle of shortest distances, i.e. Corresponding to each measuring point of the point cloud, the shortest distant point on the CAD component is searched for and assigned to it. The connection vector of both points defines the direction of error (= error vector) and its length the deviation.

Another variant is based on the CAD data and searches for the corresponding point on the ACTUAL data perpendicular to the nominal surface. The special cases of both variants - especially if there is no mathematical solution, for example because no measurement data is available perpendicular to the CAD surface due to component deviations, provides for an expansion of the search range of the corresponding points, e.g. by defining a valid search area in the form of a cone within which corresponding points can be detected.

If one considers the direction vectors determined according to the above methods and their amounts, these do not work in the direction that would be required to carry out a correction of the component with a deviation as a whole. Therefore, the CAD model cannot be traced solely on the basis of the information from the measurement point cloud (ACTUAL), the direction vectors of the deviation and its amount, which would be necessary for the calculation of a corrective measure, for example in order to correct a shaping tool according to the deviations on the component, by transferring the deviations in the negative direction to the tool.

The correction of tools in shaping manufacturing processes, such as the electrochemical machining of turbine blades, requires an efficient possibility of returning component deviations to the shaping tool electrode due to their process-inhomogeneous working gaps.

The high demands on measurement accuracy when testing these parts can be met with the use of optical scanners. Using the optical scanner, detailed and meaningful measurements and evaluations can be achieved, since they automatically capture, measure and assess the entire blade shape with leading edge, leading edge, blade head and blade root. The part to be measured, in particular the turbine part, is advantageously clamped in a stable measuring frame or fixed on a measuring table which is provided with pre-measured reference points.

Using electrochemical machining (ECM), particularly hard materials in the turbine industry, such as Nickel super alloys can be processed very precisely. An electrical voltage is applied between a workpiece and an electrode. This creates a current in an electrolyte within a gap between the workpiece and the electrode. Since the electric field is often non-uniform, the gap size can vary over the area to be processed. As a result, the workpiece does not take the exact desired shape.

To avoid image errors in the workpieces, the DE 10 2015 204 798 A1 a method for designing and / or checking an electrode and an electrode for electrochemical removal. A corresponding electrode shape is calculated on the basis of a 3D model of a component to be manufactured. According to the method, points defined on the 3D model are shifted in the direction of their normal vectors by a corresponding gap dimension. When calculating the gap dimension, the solid angles which the normal vectors assume relative to the feed direction are taken into account. In practice, the electrodes determined in this way have to be modified several times in order to compensate for imaging errors. Depending on the complexity of the workpiece, several iteration steps are required. These iteration steps can be summarized in an optimized iteration process for the design of electrodes for electrochemical removal as described below.

The object of the present invention is an optimized method for evaluating (target-actual comparison) or for detecting deviations specify an actual contour from a nominal contour, in particular for the design of the contour of turbine parts and / or electrodes for electrochemical removal. This object is achieved with a method according to claim 1. Advantageous further developments are the subject of the subclaims.

• 1 zeigt ein Werkstück mit einer zugehörigen Elektrode im Querschnitt
• 2 zeigt das Spaltmaß am Werkstück in vergrößerter Darstellung
• 3 zeigt ein Auswerteverfahren des Stand der Technik unter Verwendung eines Vektors mit geringstem Abstand von der Ist-Kontur auf die Nominal-Kontur
• 4 zeigt ein Auswerteverfahren unter Verwendung eines starren Vektors von der Nominal-Kontur auf die Ist-Kontur
• 5 zeigt ein Auswerteverfahren mit Vektoren mit rechten Winkeln von der Ist-Kontur auf die Nominal-Kontur
• 6 zeigt eine erfindungsgemäße Variante von 5 bei der jeweils von der innenliegenden Kontur betrachtet lotrecht zur äußeren Kontur gesucht wird. Die Richtung der Vektoren kann dabei auf die SOLL- oder IST-Kontur bezogen sein.
• 7 zeigt eine erfindungsgemäße Variante der 5 und 6 mit einem kegelförmigen Suchbereich um den theoretisch lotrechten Vektor im Sonderfall, wenn rechnerisch aufgrund der geometrischen Verhältnisse kein Schnittpunkt mit der äußeren Kontur erfolgen kann.

The invention is described in more detail below using an exemplary embodiment.

• 1 shows a workpiece with an associated electrode in cross section
• 2 shows the gap size on the workpiece in an enlarged view
• 3 shows an evaluation method of the prior art using a vector with the smallest distance from the actual contour to the nominal contour
• 4 shows an evaluation method using a rigid vector from the nominal contour to the actual contour
• 5 shows an evaluation method with vectors with right angles from the actual contour to the nominal contour
• 6 shows a variant of 5 looking from the inner contour perpendicular to the outer contour. The direction of the vectors can be related to the TARGET or ACTUAL contour.
• 7 shows a variant of the invention 5 and 6 with a conical search area around the theoretically perpendicular vector in special cases if, due to the geometric conditions, no intersection with the outer contour can take place.

In 1 is a workpiece to be manufactured 1 with an associated electrode 2 shown. A three-dimensional model of a workpiece is produced according to the method 1 measured by optically scanning the generally complex outer contour and a multitude of points Point _I with associated coordinates on the workpiece 1 are defined or measured. The coordinates determined in this way are transferred to a calculation module. Then for each of the points Point _I a gap 4 calculated. By moving the points Point _L The points are obtained along their normal to the workpiece surface by the amount of the associated gap dimension Point _S1 , Based on the points Point _S1 is the contour of the electrode using the reverse engineering method 2 for example calculated in the form of a facet or grid model. Then a corresponding electrode 2 manufactured. The calculated grid model can advantageously be integrated directly into the CAD environment of a milling machine. One with the electrode 2 workpiece produced by electrochemical processing 1 can deviate from the desired shape due to fluctuations in the electrical field and / or the gap dimension and a workpiece contour 3 exhibit.

Modification of the electrode 2 to correct the image error is based on the 2 explained in more detail. This shows the workpiece 1 , the electrode 2 , the gap dimension 4 and the deviation 6 excerpts in an enlarged view. On all points Point _L become normals 7 on the surface of the workpiece 1 built and along these normals 7 the distances 6 between the target contour 5 and the workpiece contour 3 calculated. Because the workpiece 3 in point Point _L has an oversize, the gap dimension 4 be made smaller. To do this, the points Point _S1 along the normals 7 by the negative amount of the deviation 6 postponed. The points result from this Point _S2 , Based on the points Point _S2 is also a corrected electrode by reverse engineering 2 manufactured. Then use this electrode 2 another workpiece 1 manufactured. As described above, this is measured and the workpiece contour 3 is with the target contour 5 to compare. According to the procedure, the corrections are repeated until the deviations 6 are within predetermined target values. An advantage of the method is that the workpiece contour 3 with a variety of points Point _L determined very precisely and the subsequent correction can therefore be carried out particularly effectively. The manufacturing tolerances caused by the manufacture of the electrode, possibly also manufacturing errors, can advantageously be recognized and also compensated for.

3 shows an evaluation method using a vector 8th or error vector 8th with the smallest distance from the actual contour 9 on the nominal contour 10 , The problem with troubleshooting vectors 8th with the smallest distance is that an edge area 11 in the case of larger deviations it is not completely covered as the distance 14 to the nominal contour 10 in a side area 13 is less than the distance 15 to the edge 12 , This will make the edge's actual error situation 12 not reproduced correctly.

4 shows an evaluation method using a rigid vector 8th from the nominal contour 10 to the actual contour 9 , The problem with troubleshooting with rigid vertical alignment of the vectors 8th on the nominal contour 10 is that the edge area 11 it is not completely covered as there are larger deviations 16 the error vectors 8a the actual contour 9 do not cut. The error vectors 8th . 8a represent the deviation situation of the actual contour 9 or the actual profile 9 from the nominal contour 10 or from the CAD contour 10 limited.

5 shows an evaluation method with vectors 8th with right angles RW from the actual contour 9 on the nominal contour 10 , The right angle RW is on the actual contour 9 fixed, so the problem arises in the case when the actual contour 9 over the nominal contour 10 protrudes, ie if the actual contour 9 is larger than the nominal or nominal contour 10 , so there are no intersections of the error vectors 8a with the nominal contour 10 ,

6 shows a variant of 5 , the right angle RW of the vectors 8th not on the actual contour 9 is fixed, but a flexible vector base 17 with the right angle RW in places with a smaller nominal contour 10 , ie is fixed on the inside contour. This results in a flexible vector basis 17 in places with a smaller nominal contour 10 , ie if this is compared to the actual contour 9 is inside. The right angle RW is always on the inside contour 9 . 10 starting from the actual contour 9 ,

7 shows variant of 5 and 6 with a conical search area 18 for a point 19 with a vector base 21 , This results, for example, from open contour areas in the direction of the vector base 21 arithmetically no intersection, the closest point will be replaced 19 recorded within the valid conical search area, in this case the deviation vector is 20 not perpendicular to the contour. The conical search area 18 is through a vector 22 with a right angle on the place of the smaller contour 10 pinned and around this vector 22 spread out conically. The conical search area 18 serves as an extension for areas with gaps or if there is no arithmetic intersection. In the case of deliberate deviations, permissible conical widening of the search area, this is the definition reference of the respective normal in the valid point from the previous definition. For example, a search can be carried out using the shortest distance to the comparison surface within the cone 18 with a 90 degree opening angle or opening angle of 90 degrees.

The procedure for the target-actual comparison or the recording of deviations of an actual contour 9 of a nominal contour 10 , where the error vector 8th always perpendicular to the inner (smaller) contour 9 . 10 viewed from stands and towards the outer contour 9 . 10 shows, and being the error vector 8th is perpendicular to a surface at a position of a measuring point or a nominal point, is preferably used for the measurement of bodies or three-dimensional surfaces, especially in the case of strong curvatures and large deviations. In the case of curvatures, there can be highly inhomogeneous error profiles due to the curvatures, which can lead to large error interpretations that lead to misunderstandings or misinterpretations in the recording of the deviations.

LIST OF REFERENCE NUMBERS

1

workpiece

2

electrode

3

Workpiece contour

4

clearance

5

target contour

6

deviation

7

normal

8th

Vector / error vector

9

Actual contour / profile

10

Nominal contour / CAD contour

11

edge region

12

edge

13

page range

14

distance

15

distance

16

deviations

17

vector basis

18

Conical search area

19

Point

20

distance

21

RW

right angle

8a

Vector / error vector

Claims (5)

Method for evaluating (target-actual comparison) or detecting deviations of an actual contour (9), which is the contour of a workpiece to be produced (1), from a nominal contour (10), which is the contour of the CAD model The method for developing an electrode (2) for electrochemical removal is characterized in that the contours (9 and 10) overlap, that an error vector (8) is always perpendicular to the inner (smaller) contour (9, 10) is viewed from and points in the direction of the outer contour (9, 10), the error vector (8) standing vertically on a surface at a position of a measuring point or a nominal point. Procedure according to Claim 1 , characterized in that the method comprises a completely digital process chain for developing an electrode (2) for electrochemical removal with the following method steps: a) providing a three-dimensional model of workpieces to be produced (1) b) defining a plurality of points Pkt _L with associated coordinates the model c) transferring the coordinates to a calculation module d) calculating an amount for a gap dimension (4) greater than zero for each of the points point _L e) shifting the points point _I by the respectively calculated gap dimension (4) along an associated normal (7 ) to points Pt _S1 f) Determine the electrode geometry using the points Pt _S1 g calculated according to e)) Manufacture of an electrode (2) with the geometry determined under f) h) Machining a workpiece (1) with the previously produced electrode ( 2) i) defines a plurality of points Pkt _L with associated coordinates on the previously produced workpiece (1) k) the coordinates determined in method steps b) and i) are compared and deviations (6) are determined I) the electrode (2) produced in method step g) is corrected and the deviations (6) determined in k) are compensated and that m ) the process steps h), i), I), k) are repeated until the deviations (6) in process step k) lie within predetermined target values. Procedure according to Claim 4 , characterized in that in method step I) all points _S1 are shifted in the negative direction along the normal (7) by the amounts of the deviations (6) to points _S2 . Procedure according to Claim 2 or 3 , characterized in that the electrode geometry in method step f) is calculated using the reverse engineering method. Procedure according to one of the Claims 2 to 4 , characterized in that the electrode geometry determined in method step f) is integrated directly into the CAD environment of a milling machine.

Images