AUTOMATED RESONANCE TEST ON MULTICOMPONENT COMPONENTS BY MEANS OF PATTERN RECOGNITION

Abstract

A fast and simple classification of the state of the component is ensured by carrying out the resonance test in an automated manner on blade assemblies, in which frequency images of new and used components are compared with each other.

Claims

1. A method for the automated performance of a resonance test, in which beforehand, by a direct mechanical excitation of a new multicomponent component, in particular a blade row, relevant acoustic parameters, in particular frequency pictures and/or frequency profiles and/or decay behavior or other acoustic characteristics are measured, or the relevant acoustic parameters such as the frequency pictures, frequency the profiles, and/or acoustic behaviors are numerically computed, wherein these have been deposited in a database and performing the direct mechanical excitation of a used component, acquiring the relevant acoustic parameters, in particular the frequency pictures and/or the frequency profiles and/or the decay behavior, wherein this is compared to the frequency picture and/or the frequency profiles and/or the decay behavior of the new component, which is stored in the database, and deviations are detected and evaluated.

2. A device for performing the method as claimed in claim 1, which comprises means for recording acoustic parameters such as the frequency pictures and/or the frequency profiles and/or the decay behavior, which are assigned to a component, or the relevant acoustic parameters such as the frequency pictures, the frequency profiles, and/or the acoustic behavior are numerically computed, the database, in which these data are stored, and in which the same excitation in particular mechanical excitation on the same component after use, are performed, and the acoustic parameters, in particular the frequency pictures and/or the frequency profiles and/or the decay behavior are also recordable, wherein these are also stored and can be compared to the existing acoustic parameters, in particular the frequency pictures and/or the frequency profiles of the new component.

3. The method as claimed in claim 1, in which the deviations are classified between acceptable and to be replaced.

4. The method as claimed in claim 1, in which methods of artificial intelligence are applied to perform a pattern recognition.

Description

BRIEF DESCRIPTION

[0010] Some of the embodiments will be described in detail, with reference to the following figures, wherein like designations denote like members, wherein:

[0011] FIG. 1 shows a frequency picture of a new component;

[0012] FIG. 2 shows a frequency picture of a used component; and

[0013] FIG. 3 shows a decay behavior for new components and a decay behavior for a used component

DETAILED DESCRIPTION

[0014] FIG. 1 shows a frequency picture 1 of one or more components in the new state or before the first use. The intensity I is plotted in relation to the frequency f.

Various frequencies, which are not necessarily discrete, having various intensities are recognizable, which are typical for a new component. This is only one example of an acoustic parameter.

[0015] A frequency picture 2 of a used component according to FIG. 1 can be seen in FIG. 2.

Both the intensity I and also the location of the frequencies f have at least partially changed and/or shifted.

[0016] The decay behavior of the intensity I over the time t has a similar appearance, wherein a decay behavior 4 for new components is shown in FIG. 3 and the curve 7, shown by a dashed line here, represents the decay behavior of a used component. The decay behavior 4, 7 is only one example of an acoustic parameter.

[0017] This makes it clear that differences are provided which can be analyzed.

[0018] The pattern recognition recognizes in this case the deviation from the target state and assigns the blade rows as a component of a further classification such as “acceptable” or “to be replaced”. These classifications are established beforehand on the basis of preliminary studies and existing measurements.

[0019] To carry out the pattern recognition, inter alia, methods of artificial intelligence are applied.

[0020] The advantages are:

a) unambiguous assignment of defective blade rows by means of objective methods.
b) avoidance of the disassembly of the component, which means a savings in costs and time and results in availability improvement.

[0021] Although the present invention has been disclosed in the form of preferred embodiments and variations thereon, it will be understood that numerous additional modifications and variations could be made thereto without departing from the scope of the invention.

[0022] For the sake of clarity, it is to be understood that the use of “a” or “an” throughout this application does not exclude a plurality, and “comprising” does not exclude other steps or elements.