AUTOMATION OF THICKNESS MEASUREMENTS FOR NOISY ULTRASONIC SIGNALS

Abstract

An automated method, computer program product and device for determining a thickness of an object at a specific location, such as the determination of a wall thickness of a blade or a vane of a gas turbine. The method includes performing a plurality of ultrasonic measurements around the specific location with an ultrasonic probe of the ultrasonic measurement device, recording and storing the measurement signals, determining at least one apparent thickness of the object for each measurement, sorting the apparent thicknesses in a histogram, and determining the thickness of the object by selecting the most frequently occurring apparent thickness in the histogram.

Claims

1. An automated method for determining a thickness of an object at a specific location, the method comprising the following steps which are carried out automatically by an ultrasonic measurement device: a) performing a plurality of ultrasonic measurements around the specific location by means of an ultrasonic probe of the ultrasonic measurement device, b) recording and storing the measurement signals of the ultrasonic measurements by an evaluation unit of the ultrasonic measurement device, c) determining at least one apparent thickness of the object for each measurement, d) sorting the apparent thicknesses in a histogram, and e) determining the thickness of the object by selecting the most frequently occurring apparent thickness in the histogram.

2. The method according to claim 1, wherein each peak in the ultrasonic measurement, which is above a time-varying threshold, is converted into one apparent thickness.

3. The method according to claim 2, wherein multiple peaks above the threshold lying within a predetermined timespan are converted into one single apparent thickness.

4. The method according to claim 3, wherein any peak above the threshold preceded by another peak within the predetermined timespan is discarded.

5. The method according to claim 1, wherein a time-dependent low pass filter with a cut-off frequency which decreases for increasing measurement time is applied in the ultrasonic measurements.

6. The method according to claim 1, wherein a change in the polarity of the measured signals is taken into account by applying an appropriate offset to the respective measurement signals.

7. The method according to claim 1, wherein at least twenty individual ultrasonic measurements are carried out around the specific location to determine the thickness of the object at the specific location.

8. The method according to claim 1, wherein the measurements are carried out at predetermined locations relative to the specific location.

9. The method according to claim 8, wherein the measurements are carried out in a star-like and/or concentric pattern around the specific location.

10. The method according to claim lone of the claims 1, wherein the measurements are carried out at random positions around the specific location.

11. The method according to claim 1, wherein the thickness of the object is in a range of 0.5 mm to 20 mm.

12. The method according to claim 1, wherein the material of the object comprises a grain structure at the specific location.

13. The method according to claim 1, wherein the object is a blade or a vane of a gas turbine.

14. A computer program product stored on a non-transitory computer readable medium, comprising: instructions which, when the program is executed by a computer, cause the computer to carry out the steps of the method according to claim 1.

15. An ultrasonic measurement device for automatically determining a thickness of an object at a specific location, the ultrasonic measurement device comprising: an ultrasonic probe and an evaluation unit, wherein the evaluation unit is configured to trigger the ultrasonic measurement device to perform a plurality of ultrasonic measurements at and around the specific location, record and store the measurement signals, determine at least one apparent thickness of the object for each measurement, sort the apparent thicknesses in a histogram, and determine the thickness of the object by selecting the most frequently occurring apparent thickness in the histogram.

16. The method according to claim 7, wherein at least fifty individual ultrasonic measurements are carried out around the specific location to determine the thickness of the object at the specific location.

17. The method according to claim 7, wherein at least one hundred individual ultrasonic measurements are carried out around the specific location to determine the thickness of the object at the specific location.

18. The method according to claim 12, wherein the material of the object comprises a grain structure at the specific location featuring an average grain size of about the thickness of the object at the specific location.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In the following, the invention is illustrated by the help of accompanying drawings, of which

[0048] FIG. 1 shows an ultrasonic measurement signal with several peaks above a predetermined threshold value; and

[0049] FIG. 2 shows a histogram with a plurality of apparent thickness obtained from several ultrasonic measurements.

DETAILED DESCRIPTION OF INVENTION

[0050] The FIG. 1 illustrates an ultrasonic measurement signal, commonly referred to as an A-scan, which has been obtained by sending out an ultrasonic pulse into an object and subsequently receiving and recording the reflected signal. The abscissa 11 (x-axis) relates to the time in dimensionless sample units. Note that a sample unit can be calculated into a time span. However, this is dependent on the specific ultrasonic measurement device, in particular on the specific ultrasonic measurement probe in use. The ordinate 12 (y-axis) relates to the intensity of the reflected ultrasonic measurement signal in decibels. The ultrasonic measurement signal—alternatively named as “A-scan”—is referred to with the reference sign 21; a time-varying threshold is referred to with the reference signs 221 and 222.

[0051] In the exemplary A-scan illustrated in FIG. 1, measurement artefacts (overshoots) which are to be ignored can be seen between approximately 2600 sample units and 3000 sample units. Subsequently, there exist distinct peaks with decreasing height between approximately 3000 sample units and 4800 units. In order to identify and filter out the relevant peaks, a positive threshold 221 and a negative threshold are applied to the measurement signal 21. As a consequence, only those peaks which lie above the threshold 221 and which do not belong to the measurement artefacts mentioned above are taken into account. By multiplying the time attributed to a peak with the speed of the ultrasonic wave in the investigated material, the thickness of the object can be calculated (in general, the obtained value for the thickness needs to be divided by two as the detected signal relates to an ultrasonic measurement signal travelled forth until the backwall, being reflected there and travelled back until the detector). Eventually, one apparent thickness thus results for each relevant peak of the A-scan.

[0052] In the example of FIG. 1, any peak above the threshold which is preceded by another peak within a predetermined timespan is discarded. As a result, only the peaks which are referred to by the reference signs 211, 212, 213, 214, 215, 216 and 217 are taken into account as positive intensity peaks. Likewise, only the relevant negative intensity peaks are taken into account, too. For the sake of clarity, the relevant negative intensity peaks are not referred to with reference signs in FIG. 1.

[0053] Each apparent thickness of an A-scan is listed in a histogram. As ultrasonic measurements are repeated for a plurality of different locations around the specific location for which the thickness shall be determined, many apparent thicknesses are obtained. They are all listed in the same histogram.

[0054] An example of such a resulting histogram is shown in FIG. 2. At the abscissa 31 (x-axis), the apparent thickness (again in sample units) is shown; at the ordinate 32 (y-axis), the number of related peaks to the respective apparent thickness is shown. To determine the true thickness out of the amount of obtained apparent thicknesses, simply the most frequently listed apparent thickness is taken. In the example of FIG. 2, the true thickness of the object at the specific location would correspond to approximately 1500 sample units. Note again, that these 1500 sample units still need to be calculated into a concrete thickness value. This is carried out by accounting for the specific measurement setup and for the specific ultrasonic measurement device in use.

[0055] An important advantage of the present method for determining the thickness of an object at a specific location is that the thickness can be determined with a generally satisfying reliability fully automatically—even if the material of the object features up to a certain degree of flaws and defects which normally impede or even prohibit automated ultrasonic measurements.