DE102018119206A1 - Method for detecting the geometry of an area of an object using ultrasound - Google Patents

Abstract

The invention relates to a method for detecting the geometry of a region (2c) of an object (2) by means of ultrasound, in which individual sound waves are generated with a time delay by probe elements (1a to 11) and which are generated by a reflection surface (3) of the region (2c). of the object (2) reflected echoes of the sound waves are detected by one or more of the probe elements (1a to 11), the total transit times for the period from the generation of a sound wave to the detection of its echo being measured. The following measures are proposed so that this is simplified overall and can be carried out from one side in particular for complex workpieces with a non-parallel surface of their walls: 1a) measuring the individual total run times for a multiplicity of probe elements (1a to 11), which detect the echo of the sound wave generated by one of the probe elements (1a to 11), 1b) setting up one Associated transit time function for each measured total transit time, based on the location positions of the transmitting and receiving elements, the individual transit time curves (4) consisting of possible points on the reflection surface (3), and 1c) at least sectionally reconstructing the geometry of the area (2c ) of the object (2) based on the family of transit time curves (4) by means of analytical or numerical calculations.

Description

The invention relates to a method for detecting the geometry of a region of an object by means of ultrasound, in which individual sound waves are generated with a time delay by probe elements and the echoes of the sound waves reflected by a reflection surface of the region of the object are recorded by at least some of the probe elements. the transit times are measured for the duration from the generation of a sound wave until the detection of its echo.

The use of sound waves has long been established for imaging processes in connection with the non-destructive examination of objects, for example in objects such as components, workpieces and / or pipes. For this purpose, the sound waves with their frequency in the ultrasound range are introduced into the object to be examined by means of a test head. On the basis of the transit time of a sound wave reflected as an echo on a reflection surface of the object, knowledge of its speed makes it possible to make a statement about the distance of the reflection surface relative to the test head.

Systems which can generate and / or detect ultrasonic waves in an object (for example the workpiece to be tested) are referred to below as ultrasonic transducers (here also briefly “transducers”). Ultrasound signals (i.e. acoustic vibrations) are converted into electrical signals and vice versa. A basic distinction is made between configurations in which the transmitter and the receiver are represented by the same ultrasound transducer and those in which the transmitter and the receiver are realized by two spatially separate ultrasound transducers (this is then referred to as SE technology). In many cases, the ultrasound transducer used to generate the ultrasound pulse is identical to the ultrasound transducer used for detection.

In the conventional ultrasound test, the ultrasound transducer usually consists of a piezo crystal which is excited to vibrate by applying an electrical voltage or converts detected vibrations into electrical signals. Other types of ultrasonic transducers can generate and / or generate vibrations, for example, electromagnetic (also called “EMUS” or “EMAT”) or laser-excited. Combinations of different techniques in the receiver and transmitter are also known. An ultrasonic transducer used for non-destructive material testing is also called a test head.

A special type of probe (also in the case of SE technology) is group emitter technology (also known as phased array technology), in which the acoustically active surface facing the workpiece is divided into a large number of individual ultrasonic transducers, which are separated by a suitable one Electronics and electrical interconnection can be controlled individually. A phase-shifted control of a subset of the plurality of ultrasonic transducers can now generate a sound beam with an adjustable insonification angle or focus position. Thanks to the electronically controlled sound generation, many different insonification angles and focal distances can be realized with a single test head using group emitter technology (also referred to as group emitter or group emitter test head) (“electronic scanning”). The individual ultrasonic transducers of a group radiator are also called probe element or element only in the following.

In non-destructive material testing, ultrasonic pulses are generated, for example, using the pulse-echo method with ultrasonic transducers, which are introduced into the workpiece via a coupling medium (for example water). If the ultrasound pulse generated in this way encounters an obstacle (for example a discrepancy), an ultrasound signal is generated there by reflection or diffraction, which in turn propagates in the workpiece, runs through the coupling medium and is subsequently detected by an ultrasound transducer.

The amplitude of the electrical signal detected by the ultrasound transducer is generally represented as a function of time (transit time of the ultrasound signal). This is known as an A-scan.

A central term in the method according to the invention described below is the term “transit time curve”, which is explained below. If an ultrasonic wave emitted by a probe element is reflected at a position and the reflected ultrasonic wave is received by a probe element, the total transit time of the ultrasonic wave can be measured from the transmitting to the receiving element. The sending and the receiving element can be identical or different. The position at which the reflection took place cannot be clearly determined from the total running time alone. The positions at which a reflection with the measured transit time can have taken place form a curve, the so-called transit time curve. After measuring the total transit time from the transmission of the signal to the reception of the echo and calculation of the transit time curve, it is only known that the reflection at some point on the transit time curve has taken place. At which point on the curve is unclear.

If the transmitting and receiving element are identical and the ultrasonic wave does not experience a transition from one medium to another medium, the transit time curve forms a circle around the sound exit point. If the transmitting and receiving element are different and the ultrasound wave does not undergo a transition from one medium to another medium, the transit time curve forms an ellipse, the positions of the transmitting and receiving ultrasound transducer precisely representing the focal points of the ellipse. In the case of a media transfer, for example from water to steel when using a water inlet section, an analytical description of the transit time curves is much more difficult.

With the disclosure US 2014/03 30 127 A1 A method for reconstructing the geometry of the surface of an object by means of an ultrasound test has become known. The method enables the non-destructive testing of mechanical components with complex shapes. In order to obtain an improved representation of the surface in addition to a high speed of reconstruction, it is proposed to initially generate sound waves staggered in time by a large number of probe elements. The sound wave generated by a probe element is then recorded as an echo by the same probe element; Interactions between different probe elements are not considered. On the basis of the transit times that can be determined in this way for each individual probe element, corrections are then made to the relative time offset of the sound waves subsequently generated, namely until they all hit the surface of the component at the same time. With the help of this step-by-step process, the necessary pauses between the otherwise overlapping sound waves, which only allow insufficient scanning, are initially minimized. Only as soon as the simultaneous impingement of the individual sound waves is achieved is a geometric reconstruction of the surface scanned in this way carried out by linear interpolation of points lying on the surface via the transit time detection. In essence, an iteratively adapted delaying of individual A-scans is taught before the reconstruction of the scanned surface takes place.

Since this is an iterative process, the time required or the accuracy that can be achieved within a preselected period of time is unknown in advance. Regardless of this, the method only allows the reconstruction of the next interface, i.e. the surface of the component to be tested, which faces the probe elements, or, if it comes into contact with its surface, the associated rear wall from a water inlet section. Only after the surface has been completely reconstructed can any delay laws for the propagation times of the sound waves be applied on the basis of the available data, in order to then focus at different angles at a constant depth of the component, in real time.

From the DE 10 2014 105 308 A1 results in a method for ultrasound geometry testing, the measurement accuracy of which is based on a previous calibration by means of a calibration body whose dimensions are known. For this purpose, the calibration body is positioned between two opposing test heads, the ultrasound waves generated by the test heads striking one of the surface areas of the calibration body facing them through a water jacket (water inlet section) and being reflected by the latter. Over several calibration steps, a measurement position-dependent distance between the calibration body and test heads is determined and stored using an ultrasonic transit time method. The calibration body is then removed and an object to be examined is positioned between the test heads, whereupon the transit times of the ultrasound waves are recorded in the same way as before. In a final evaluation, the actual dimensions of the object are ultimately calculated using the measuring position-dependent distances previously stored in relation to the known calibration body. In particular, the necessary diametrical arrangement of two test heads and their assumption of different measuring positions around the calibration body and the object to be examined, which are necessary for the calibration, go hand in hand with an increased expenditure of material and time.

For a thickness measurement of the walls of an object with a complex geometry, such as pipes with a non-constant wall thickness, the previously known methods of ultrasonic measurement cannot be used or can only be used to a very limited extent. Surfaces running at the same distance from each other and, in this respect, parallel (top, bottom) allow the wall thickness to be determined based on the transit time difference between the echo reflected by the top and bottom, but such methods usually fail as soon as the surfaces are oblique to one another, i.e. do not run in parallel. This is essentially due to the fact that even when a test head is aligned parallel to the upper side of the workpiece, it is not known at what angle the sound waves strike the opposite oblique underside. Other well-known imaging methods such as TFM (Total Focussing Method) or SAFT ( Synthetic Aperture Focussing Technique) sometimes require a complex extraction of the respective workpiece geometry from the image reconstruction, for which a complete acquisition of the A-scans is first necessary. Ultimately, known optical methods also require access from both sides of a wall, which is difficult or impossible at all for objects with large dimensions or a hollow design. In view of these observations, the known ultrasonic methods for examining or testing objects therefore still offer room for improvement.

Against this background, the object of the present invention is to improve a method for detecting the geometry of the wall thickness of an object in such a way that it is simplified overall and in particular can also be carried out from one side for complex workpieces with a non-parallel surface of their walls.

This object is achieved by a method with the features of claim 1. Advantageous embodiments of the invention are the subject of the respective dependent claims.

• 1a) Messen der einzelnen Gesamtlaufzeiten für eine Vielzahl an Sonden-Elementen , die das Echo derselben von jeweils einem der Sonden-Elemente erzeugten Schallwelle erfassen,
• 1b) Aufstellen einer zugehörigen Laufzeit-Funktion für jede gemessene Gesamtlaufzeit, basierend auf den Ortspositionen des jeweils sendenden und des empfangenden Elementes, wobei die einzelnen Laufzeitkurven aus möglichen Punkten auf der Reflexionsfläche bestehen, und
• 1c) zumindest abschnittsweises Rekonstruieren der Geometrie des Bereichs des Gegenstandes auf Basis der Schar an Laufzeitkurven mittels analytischer oder numerischer Berechnungen,

eine besonders einfache und wirtschaftliche Methode zum Erfassen der Geometrie eines Bereichs eines Gegenstandes mittels Ultraschall.According to the invention, the method for detecting the geometry of a region of an object by means of ultrasound, in which individual sound waves are generated with a time delay by probe elements and which detects the echoes of the sound waves reflected by a reflective surface of the region of the object by one or more of the probe elements and wherein the Total transit times for the period from the generation of a sound wave to the detection of its echo, which is characterized by the following measures:

• 1a) measuring the individual total transit times for a multiplicity of probe elements which detect the echo of the same from each sound wave generated by one of the probe elements,
• 1b) setting up an associated transit time function for each measured total transit time, based on the location positions of the transmitting and the receiving element, the individual transit time curves consisting of possible points on the reflection surface, and
• 1c) at least sectionally reconstructing the geometry of the area of the object on the basis of the family of runtime curves by means of analytical or numerical calculations,

a particularly simple and economical method for acquiring the geometry of a region of an object using ultrasound.

According to the invention, it is now basically proposed to use, for various combinations of sound elements that transmit sound waves and receive them in the form of echoes, preferably only the values recorded for this purpose for the total transit times of the sound waves in order to use a suitable algorithm to determine the respective geometry of a region of the area to be examined To reconstruct the object directly from the total run times.

The term “total transit time” means the sum of the values of the time required for the simple distance (one way) of a sound wave from a probe element to a reflection surface of the object and the time required for the simple distance (way back) of the reflected sound wave in the form of an echo back to a probe element. In contrast to, for example, imaging reconstruction methods, it is not necessary to save the complete A-scans here, which results in a drastic reduction in the ultrasound data to be transmitted.

In essence, several elements send ultrasound signals in succession, that is to say at different times, the transmitted signals being received by one or more elements in each case. The runtime of a possible reflection is extracted for each received signal. Based on the positions of the sending and receiving elements, the associated transit time curves are determined for all transit times. This finally results in a family of curves for a test head position on which there are a large number of reflection points.

For the reconstruction of the actual geometry to be determined from this family of curves, two advantageous method variants according to the invention will now be described.

According to a first advantageous method variant, the geometry of the area of the object is extracted or reconstructed at least in sections at a test head position in the form of an envelope selected for the family of transit time curves in accordance with the test head arrangement. This means that it is first determined that the family of runtime curves clings to the geometry to be determined in a certain area for a fixed probe head position. By calculating an envelope on the family of curves, the geometry to be determined is then described in the previously determined area depending on the position of the test head.

The envelope to the family of curves is formed from the opposite direction to the test head. So is the test head open, for example positioned on the top of the test piece, the lower envelope is formed.

The envelopes of the transit time curves are then advantageously reconstructed at further test head positions, and the geometry of the entire region is reconstructed by forming an overall envelope to the large number of individual envelopes. For this purpose, the test head is now moved and this method is carried out at further test head positions, so that additional envelopes are obtained which describe the geometry to be determined in other areas. A further representation of the geometry to be determined can finally be obtained by further forming an appropriate envelope on this plurality of individual envelopes.

In the second advantageous method variant of the invention it is provided that for two or more probes receiving an echo, elements from the temporal difference of the received echo calculate an average direction of the incoming sound signal and from this the intersection of the sound beam with an associated transit time curve is calculated.

These elements are preferably, but not necessarily, arranged next to one another and thus adjacent. An average direction of the incoming sound can now be calculated from the time difference of the echo reception in the receiving elements.

Furthermore, it is advantageously provided that the transit time curve used belongs to the location position of the transmitter and the location position of one of the received elements used or an averaged location position of the receiving elements.

Here, an averaged position coordinate is determined for the receiving elements, from which the sound can be traced back in the previously calculated direction on a sound beam. The intersection of this sound beam with the associated transit time curve then forms a reflection point on the geometry sought.

If there are one or more media transitions (interfaces) between the test head elements and the possible reflection points, for example in the case of a water inlet section, these can easily be taken into account when tracing the sound beam.

Carrying out this method for different transmitter-receiver combinations then results in a large number of reflection points which lie on the geometry sought. Again, moving the test head results in a number of points at further test head positions on the geometry sought, which together form a final representation of the geometry to be determined.

In both variants of the method according to the invention, it is conceivable to also take the amplitude of the received signal into account in order, for example, to obtain a weighting of the associated part of the geometry sought.

In addition, the method according to the invention makes it possible to always calculate the next interface from the knowledge of an interface using the same means, if the transit times are calculated at points behind the interface using Fermat's principle.

According to the invention, the test head elements used above can be elements of a group radiator or individual conventional test heads.

In the context of the invention, it is fundamentally considered advantageous if the geometry of the workpiece area is reconstructed exclusively on the basis of the measured transit times.

Regardless of this, it is of course conceivable that the amplitudes of the ultrasound signals associated with the detected echoes are also used. These can be used in the calculation of the envelope, for example in order to weight at least a portion of the reflection surface in relation to its echo intensity.

After this, amplitudes can be taken into account in addition to obtaining further information, whereas the actual reconstruction of the geometry of the area of the object takes place on the basis of the total running times.

With the now presented method according to the invention according to its two variants, it is possible to record the geometry and / or measure the wall thickness of an object from a single side, in particular for complex workpieces, such as, for example, pipes with a non-parallel surface of their walls perform. Depending on the design of the respective method, such as the number of probe elements used for this and corresponding runtime curves, this can contribute to a significant improvement in the detection of the geometry and / or the measurement of the wall thickness of an object. Due to the reduction in the data required for this, this can be done faster and correspondingly more dynamically compared to other known methods. Because of the proposed Algorithms can also record complex workpieces in terms of their geometry.

• 1 eine schematische Ansicht einer Vorrichtung zur zerstörungsfreien Ultraschallprüfung von Rohren,
• 2 eine schematische Ansicht eines auf eine Oberfläche eines Gegenstandes aufgelegten Prüfkopfs,
• 3 eine Simulation von Laufzeiten von Schallwellen und/oder Echos an einer schrägen Geometrie eines Gegenstandes,
• 4 eine Grafik mit Laufzeitkurven aus einer realen Messung und
• 5 eine schematische Ansicht gemäß 1 mit von der Oberfläche des Gegenstandes über einen Wassermantel beabstandeten Prüfkopf.

The invention is explained in more detail by way of example with reference to the following description and associated figures. Show it:

• 1 1 shows a schematic view of a device for non-destructive ultrasonic testing of pipes,
• 2 1 shows a schematic view of a test head placed on a surface of an object,
• 3 a simulation of transit times of sound waves and / or echoes on an oblique geometry of an object,
• 4 a graph with runtime curves from a real measurement and
• 5 a schematic view according 1 with a test head spaced from the surface of the object via a water jacket.

The 1 shows a schematic view of a device for non-destructive ultrasonic testing of an object 2 in the form of a hot rolled seamless tube for defects. Errors are defined here as imperfections, such as cracks, cavities, inclusions, pores or doublings. The object 2 has in the usual way a central and in the longitudinal direction extending tube axis R on. The core component of the test device is a test head 1 using phased array technology. For the exam the subject 2 in a feed direction V moved and the test head 1 is in a circumferential direction U around the object 2 moved around so that the object 2 is examined on a helical path. Can also use a variety of probes 1 around the object 2 be arranged or the object 2 rest and only the test head 1 will proceed.

The 2 shows a schematic view of a test head 1 with a longitudinal direction L , in particular for measuring the wall thickness and / or geometry of elongated hollow profiles such as a round or rectangular tube. Things 2 can also have a complex geometry. The ultrasonic test device essentially consists of at least one test head 1 using phased array technology. In the exemplary embodiment, the test head 1 twelve consecutive and thus linearly arranged probe elements 1a to 11 , In the usual way, these are probe elements 1a to 11 designed as piezoelectric vibrating elements. During the ultrasonic test, especially after the pulse echo method, the probe elements are used 1a to 11 Ultrasonic pulses are generated with or without a coupling medium (water or oil) in the object 2 be initiated.

In 2 is the test head 1 on a top 2a of the object 2 with a trapezoidal cross section without coupling medium. The top 2a and a bottom 2c of the workpiece 2 do not run parallel but at an angle to each other. It is shown that the test head 1 with its probe element 1f generates a sound wave and into the object 2 sends while another probe element 1a one on one, especially the top 2a opposite bottom 2 B located, reflection surface 3 of the object 2 reflected echo of the sound wave recorded. For the sake of simplicity, the sound wave is here a sound beam S and the echo as an echo beam e shown. This can be seen from the sending probe element 1f outgoing sound beam S at one point P the reflective surface 3 reflected from where it came as an echo beam e the receiving probe element 1a reached.

In the usual way, before the ultrasound examination of an area 2c of the object 2 so-called A-scan apertures are defined, which can be adjusted during the test. The individual runtimes of geometry displays - in the sense of reflected echo signals - are extracted or reconstructed from these A-scan screens. The respective transit time is made up of the simple transit time for the outward travel of the sound beam S to the reflection surface 3 as well as the simple transit time for the return of the echo beam e together.

The amount of possible points that have a fixed term from the sending probe element 1a to 11 to point P and on to the receiving probe element 1a to 11 describes an associated runtime curve 4 (please refer 2 ). Without knowing the respective angle of the echo beam e the possible points therefore all form those on the reflection surface 3 could lie because the sum of the simple terms for the sound beam S and the echo beam e as a total duration remains constant. Accordingly, it is known that the point of the geometry of the object at which the sound wave or its sound beam S was reflected on this transit time curve 4 must lie.

3 is a simulation of echo propagation times on an oblique geometry, which is, for example, the reflection surface 3 of the object 2 out 1 can act. The elliptical runtime curves were available 4 for different combinations of sending and receiving probe elements 1a to 11 and their extracted terms are calculated. The resulting family of runtime curves 4 hugs at least in one section 5 evident from the contour of the reflecting surface 3 on.

4 shows another example of runtime curves from a real measurement, which conform to the geometry sought.

This describes the basic principle of the advantageous first method variant according to the invention, according to which the geometry sought is at least in sections in the form of a family of runtime curves 4 lower envelope Enveloppe (envelope) extracted or reconstructed.

5 shows a case for the second method variant in which the test head 1 incorporating a water supply section W or a water jacket to the top 2a of the object 2 is spaced. The water supply route W is to be understood as representative of all possible media within the scope of the invention. Due to the linear arrangement of the probe elements 1a to 11 can be used for two receiving probe elements directly next to each other 1a . 1b be approximated that the respective associated points P on the captured echo beam e reflective reflective surface 3 are almost identical. From the difference in the total running times of the two probe elements 1a . 1b it is possible to change the direction of the detected echo e than on the reflective surface 3 to calculate pointing vector. Via the subsequent calculation of an intersection of the vector with the associated runtime curve 4 can the point P on the reflection surface 3 ultimately determine directly. Any distractions from the media transition between the item 2 and the water supply section W can be done in a known manner by taking into account the different angles on both sides of the interface α , β are included.

LIST OF REFERENCE NUMBERS

1

probe

1a

first probe element

1b

second probe element

1c

third probe element

1d

fourth probe element

1e

fifth probe element

1f

sixth probe element

1g

seventh probe element

1h

eighth probe element

1i

ninth probe element

1j

tenth probe element

1k

eleventh probe element

11

twelfth probe element

2

object

2a

top

2 B

bottom

2c

Area

3

reflecting surface

4

Maturity curve

5

section

α

angle

B

angle

e

echo beam

L

longitudinal direction

P

Point on a reflection surface of the object

R

pipe axis

S

supersonic jet

U

circumferentially

V

feed direction

W

Water path

QUOTES INCLUDE IN THE DESCRIPTION

This list of documents listed by the applicant has been generated automatically and is only included for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• US 2014/0330127 A1 [0010]
• DE 102014105308 A1 [0012]

Claims (11)

Method for detecting the geometry of a region (2c) of an object (2) by means of ultrasound, in which individual sound waves are generated with a time delay by probe elements (1a to 11) and which are generated by a reflection surface (3) of the region (2c) of the object (2 ) reflected echoes of the sound waves are detected by one or more of the probe elements (1a to 11), the total transit times for the duration from the generation of a sound wave to the detection of its echo being measured, characterized by the following measures: 1a) measuring the individual Total transit times for a multiplicity of probe elements (1a to 11), which detect the echo of the sound wave generated by one of the probe elements (1a to 11), 1b) setting up an associated transit time function for each measured total transit time, based on the Local positions of the respective sending and receiving element, the individual transit time curves (4) from possible points on the reflection surface surface (3), and 1c) at least section-wise reconstruction of the geometry of the area (2c) of the object (2) on the basis of the family of transit time curves (4) by means of analytical or numerical calculations. Procedure according to Claim 1 , characterized in that the geometry of the area (2c) of the object (2) is reconstructed at least in sections at a test head position in the form of an envelope chosen for the family of runtime curves (4) in accordance with the test head arrangement. Procedure according to Claim 2 , characterized in that the envelopes of the transit time curves are reconstructed at further test head positions and the geometry of the entire area (2c) is reconstructed by forming an overall envelope to the large number of individual envelopes. Procedure according to Claim 1 , characterized in that for two or more probe elements (1 a to 1i) receiving an echo, an average direction of the incoming sound signal is calculated from the time difference of the receiving echo and the intersection of the sound beam with an associated transit time curve (4) is calculated therefrom , Procedure according to Claim 4 , characterized in that the at least two probe elements (1a to 1i) are arranged side by side. Procedure according to Claim 4 and 5 , characterized in that the runtime curve used belongs to the location of the transmitter and the location of one of the received elements used or an averaged location of the receiving elements. Procedure according to the Claims 4 to 6 , characterized in that for different location positions of the transmitter and the location of one of the received elements used, intersection points of the sound beam with an associated transit time curve are calculated, which together form a final representation of the geometry to be determined. Method according to one of the preceding claims, characterized in that probe elements (1a to 1i) in the form of elements of a conventional group radiator or in the form of individual conventional test heads are used to generate the sound waves and to measure the transit times. Method according to one of the preceding claims, characterized in that in order to determine the direction of a sound wave and / or its echo, a possible media transition, in particular water flow path (W), and the associated deflection of the sound wave is taken into account. Method according to one of the preceding claims, characterized in that the geometry of the area (2c) of the object (2) is reconstructed exclusively on the basis of the measured transit times. Method according to one of the preceding claims, characterized in that the amplitudes associated with the detected echoes of the sound waves are used.

Images