# US 2020/0003736 A1 · Method for Early Damage Recognition, and Program and Control Unit for Executing the Method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2020/0003736 A1
- Document ID: US20200003736A1
- Status: Published application
- Kind: A1
- Application number: 16/392,712
- Filed: 2019-04-24
- Published: 2020-01-02
- Earliest claimed: 2018-06-27
- Assignees: Robert Bosch GmbH
- Applicants: Robert Bosch GmbH
- Inventors: Ulrich Bittner; Wilfried Eichner
- Web page: https://patentgrep.com/patent/US20200003736A1

## CPC

- G01N 29/449: {Statistical methods not provided for in G01N29/4409, e.g. averaging, smoothing and interpolation}
- G01N 29/46: by spectral analysis, e.g. Fourier analysis {or wavelet analysis (spectral signal processing per se G06F17/14)}
- G01N 29/14: using acoustic emission techniques {(echo of particles G01N29/046; measuring mechanical vibrations or acoustic waves in solids in general G01H1/00)}
- G01N 29/4427: {with stored values, e.g. threshold values}
- G01N 29/4436: {with a reference signal (amplitude comparison G01N29/48)}
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)

## Abstract

A method for early damage recognition, and program and control unit for executing the method are disclosed. A method is disclosed for early damage recognition, wherein a frequency-transformed signal, filtered of dominant excitations, is supplied to a comparative early damage recognition.

## Claims

### Claim 1

A method for early damage recognition of a machine, the method comprising:
detecting at least one oscillation signal of the machine;
transforming, after the detecting, the at least one oscillation signal from a time range into an image range;
filtering, after the transforming, the at least one oscillation signal to reduce, at least in absolute value, at least one dominant excitation, the at least one dominant excitation being determined according to at least one operating variable; and
recognizing, after the filtering, damage of the machine by comparing at least one signal value of the at least one oscillation signal to a comparison value.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the at least one oscillation signal of the machine is at least one of a pivot angle, a fluid-borne sound, an airborne sound, a structure-borne sound, and a combination thereof.

### Claim 3 (depends on 1)

The method according to claim 1, the transforming further comprising:
integrally transforming the at least one oscillation signal.

### Claim 4 (depends on 1)

The method according to claim 1 further comprising:
detecting, before the filtering, the at least one operating variable, the detecting of at least one operating variable being performed one of (i) once, (ii) irregularly, (iii) periodically, and (iv) continuously.

### Claim 5 (depends on 1)

The method according to claim 1 further comprising:
scaling, after the transforming, a spectrum into an order spectrum based on the at least one dominant excitation.

### Claim 6 (depends on 1)

The method according to claim 1 further comprising:
further filtering the at least one oscillation signal,
wherein the further filtering is at least one of (i) after the filtering, (ii) integrated into the filtering, (iii) before the filtering, (iv) before the transforming, and (v) before the detecting.

### Claim 7 (depends on 6)

The method according to claim 6 further comprising:
outputting an envelope curve by rectifying the at least one oscillation signal,
wherein the outputting is at least one of (i) after the filtering, (ii) after the further filter, (iii) integrated into the filtering, (iv) integrated into the further filtering, (v) before the filtering, (vi) before the further filtering, (vii) before the transforming, and (viii) before the detecting.

### Claim 8 (depends on 1)

The method according to claim 1, furthermore comprising:
allocating, before the recognizing, the at least one oscillation signal into at least two bands.

### Claim 9 (depends on 1)

The method according to claim 1, wherein:
at least one characteristic value, based on the at least one signal value, is used for the comparing, and
the at least one characteristic value is at least one of an arithmetic mean value, a root mean square value, a median value, a focal point frequency (spectral centroid), a frequency of the maximum (maximum amplitude, maximum sum of the amplitudes), a spectral flatness (quotient of the root mean square value and the arithmetic mean value, Wiener entropy), an inharmoniousness, a standard deviation, mel frequency cepstrum coefficients (MFCC), a variance, a kurtosis (bulge), a skew (skewness), a variation breadth, a loudness, a spectral flux, a tonality, a variation breadth (peak2peak and/or span), and a combination thereof.

### Claim 10 (depends on 1)

The method according to claim 1, wherein the comparison value is at least one of a target value, an extrapolated value, and a combination thereof.

### Claim 11 (depends on 1)

The method according to claim 1, wherein the detecting, transforming, the filtering, and the recognizing are performed at least one of (i) periodically, (ii) in an event-controlled manner, (iii) overlapping, and (iv) at least partially in parallel.

### Claim 12

A non-transitory computer program for early damage recognition of a machine, the computer program configured to, when executed:
detect at least one oscillation signal of the machine;
transform, after the detecting, the at least one oscillation signal from a time range into an image range;
filter, after the transforming, the at least one oscillation signal to reduce, at least in absolute value, at least one dominant excitation, the at least one dominant excitation being determined according to at least one operating variable; and
recognize, after the filtering, damage of the machine by comparing at least one signal value of the at least one oscillation signal to a comparison value.

### Claim 13

A control unit for early damage recognition of a machine, the control unit configured to execute a non-transitory computer program to:
detect at least one oscillation signal of a machine;
transform, after the detecting, the at least one oscillation signal from a time range into an image range;
filter, after the transforming, the at least one oscillation signal to reduce, at least in absolute value, at least one dominant excitation, the at least one dominant excitation being determined according to at least one operating variable; and
recognize, after the filtering, damage of the machine by comparing at least one signal value of the at least one oscillation signal to a comparison value.

### Claim 14 (depends on 3)

The method according to claim 3, the transforming further comprising:
Fourier transforming the at least one oscillation signal.

### Claim 15 (depends on 14)

The method according to claim 14, the transforming further comprising:
discrete Fourier transforming the at least one oscillation signal.

### Claim 16 (depends on 14)

The method according to claim 14, the transforming further comprising:
fast Fourier transforming the at least one oscillation signal.
