Method for Monitoring the Status of an Apparatus and Assembly

Abstract

A method for monitoring the status of an apparatus, wherein an analog signal is converted into a digital signal with an analog-digital converter operating within a measuring range, signal portions of the analog signal extending beyond the measuring range are cut in the digital signal, a spectral analysis is applied to the digital signal to determine which frequency potions the analog signal possesses in a frequency spectrum and conclude a malfunction of the apparatus when the analog signal exceeds the measuring range, where when the analog signal extending beyond the measuring range is in the digital signal, this event is detected and determined as a number, where a signal quality is provided which is used to assess whether known damage frequencies can still be identified from determined frequency portions of the frequency spectrum, although additional overcontrol portions in the frequency spectrum occur as a result of possibly cut signal portions.

Claims

1. A method for monitoring status of an apparatus by measuring an analog signal which is continuous in terms of time and value and which represents a status variable of the apparatus, the method comprising: converting the analog signal, at scanning times within a measuring interval, into a digital signal via an analog-digital converter which is operated within a measuring range; cutting signal portions of the analog signal which extend beyond the measuring range in the digital signal in an event the analog signal exceeds the measuring range; and applying a spectral analysis to the digital signal to determine which frequency portions the analog signal comprises in a frequency spectrum, and to conclude a malfunction of the apparatus based on a comparison with the occurring frequency portions and known damage frequencies; wherein in an event that, at a scanning time, the analog signal which extends beyond the measuring range is cut in the digital signal, this event is detected and is determined as a number; and wherein a signal quality is provided as a quotient from the number relating to an overall scanning number within the measuring interval, and the signal quality is utilized to assess whether the known damage frequencies can still be identified from the determined frequency portions of the frequency spectrum, although additional overcontrol portions occur in the frequency spectrum as a result of possibly cut signal portions.

2. The method as claimed in claim 1, wherein the digital signal is corrected or estimated with a signal correction method and additional overcontrol portions produced by said cutting are as a result eliminated or at least reduced in an estimate signal; wherein the known damage frequencies are still identified and mapped in a transfer function; and wherein the transfer function is further utilized as a function of the signal quality up to a predeterminable limit value (Vt) of the signal quality.

3. The method as claimed in claim 2, wherein the transfer function is modified as a function of the signal quality.

4. The method as claimed in claim 2, wherein a series of transfer functions are formed and archived in a learning phase in a quality state of the apparatus (1) as a function of the signal quality; and wherein the transfer function is subsequently utilized as a reference in the original quality state of a monitored apparatus.

5. The method as claimed in claim 3, wherein a series of transfer functions are formed and archived in a learning phase in a quality state of the apparatus as a function of the signal quality; and wherein the transfer function is subsequently utilized as a reference in the original quality state of a monitored apparatus.

6. The method as claimed in claim 2, wherein a Wiener filter is utilized as a signal correction method.

7. The method as claimed in claim 3, wherein a Wiener filter is utilized as a signal correction method.

8. The method as claimed in claim 4, wherein a Wiener filter is utilized as a signal correction method.

9. An electronic assembly which monitors a status of an apparatus, the assembly comprising: an input circuit for measuring an analog signal which is continuous in terms of time and value and which represents a status variable of the apparatus; an analog-digital converter which scans the analog signal at scanning times within a measuring interval and which converts the scanned analog signal into a digital signal, the analog-digital converter being configured to, within a measuring range, in an event the analog signal exceeds the measuring range, cut signal portions, extending beyond the measuring range, of the analog signal in the digital signal; a spectral analyzer for applying a spectral analysis to the digital signal to determine which frequency portions the analog signal comprises in a frequency spectrum and based on a comparison with the occurring frequency portions and known damage frequencies to conclude a malfunction of the apparatus exists; a detector which is configured to, in an event the analog signal which exceeds the measuring range at a scanning point is cut in the digital signal, detect this event and to increment the event in a counter as a number; a signal quality evaluation facility which provides a signal quality as a quotient from the number relating to an overall scanning number within the measuring interval; and a decision means which is configured to assess, based on the signal quality, whether the known damage frequencies can still be identified from the determined frequency portions of the frequency spectrum, although additional overcontrol portions result in the frequency spectrum as result of possible cut signal portions.

10. The assembly as claimed in claim 9, further comprising: a filter in which a signal correction method executes, which corrects or estimates the digital signal and which are eliminated or at least reduced by the cutting of developed additional overcontrol portions as a result in an estimate signal; wherein the filter further includes a transfer function which still maps the known damage frequencies, the filter having an input for the signal quality (Q) and the filter being configured to further utilize the transfer function up to a predeterminable limit value as a function of the signal quality.

11. The assembly as claimed in claim 10, further comprising: an adjustment means which is configured to modify the transfer function as a function of the signal quality.

12. The assembly as claimed in claim 10, further comprising: a learning means which, as a function of the signal quality, is configured to form and archive a series of transfer functions in a learning phase in a quality state of the apparatus, and which is further configured to take as reference the transfer function in the original quality state of a monitored apparatus.

13. The assembly as claimed in claim 11, further comprising: a learning means which, as a function of the signal quality, is configured to form and archive a series of transfer functions in a learning phase in a quality state of the apparatus, and which is further configured to take as reference the transfer function in the original quality state of a monitored apparatus.

14. The assembly as claimed in claim 10, wherein the filter comprises a Wiener filter.

15. The assembly as claimed in claim 11, wherein the filter comprises a Wiener filter.

16. The assembly as claimed in claim 12, wherein the filter comprises a Wiener filter.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The drawing shows an exemplary embodiment of the invention, in which:

[0029] FIG. 1 shows an assembly for status monitoring in accordance with the invention;

[0030] FIG. 2 shows alternative embodiment of the assembly of FIG. 1;

[0031] FIG. 3 shows graphical plots of the principle of a status monitoring of an apparatus in accordance with the invention;

[0032] FIG. 4 shows an exemplary graphical plot of a spectrum with interference frequencies in accordance with the invention;

[0033] FIG. 5 shows a graphical plot of a signal to be detected, which passes into a measuring range limit, in accordance with the invention;

[0034] FIG. 6 shows a graphical plot of diagram for a spectral power density;

[0035] FIG. 7 shows a graphical plot of a transfer function derived from the spectral power density;

[0036] FIG. 8 shows a graphical plot of signal estimated with a Wiener filter in contrast to a source signal; and

[0037] FIG. 9 is a flowchart of the method in accordance with the invention.

DETAILED DESCRIPTION OF THE EXEMPLARY EMBODIMENTS

[0038] FIG. 1 shows an assembly 10 configured to monitor the status of an apparatus 1. The assembly 10 comprises an input circuit 11 for measuring an analog signal xa(t) which is continuous in terms of time and value, and represents a status variable of the apparatus 1, an analog-digital converter ADU, which scans the analog signal xa(t) at scanning times ts within a measuring interval Tn and converts the same into a digital signal xd[k], where the analog-digital converter ADU is configured to operate within a measuring range MB, where in the event that the analog signal xa(t) exceeds the measuring range MB, to cut signal portions of the analog signal xa(t), which extend beyond the measuring range MB, in the digital signal xd[k] (see FIG. 5). Furthermore, the assembly 10 has a means 12 for spectral analysis (e.g., a spectral analyzer) in order to apply a spectral analysis to the digital signal xd[k]. It is possible to read out from the determined spectral analysis or from the determined frequency spectrum FFT, which frequency portions fi the analog signal xa(t) contains and it is possible to conclude a malfunction of the apparatus 1 has occurred based on a comparison with the occurring frequency portions fi and known damage frequencies fs in a comparison device 21.

[0039] In accordance with the invention, a detection device 13 is now available in the assembly 10, which is configured, in the event the analog signal xa(t) that extends beyond the measuring range MB at a scanning time ts is cut in the digital signal xd[k], to detect this event and to increment the same in a counter 14 as a number Nsat.

[0040] A signal quality Q as a quotient from the number Nsat relating to an overall scanning number N within the measuring interval Tn can now be provided via a signal quality evaluation device 15. Furthermore, a decision device 16 is available in the assembly 10, which is configured to assess, based on the signal quality Q, whether the known damage frequencies fs can still be identified from the determined frequency portions fi of the frequency spectrum FFT, although additional overcontrol portions fk result in the frequency spectrum FFT as a result of possibly cut signal portions.

[0041] If the decision is made in the decision device 16 with the aid of the signal quality Q that the damage frequencies fs can still be determined from the determined frequency portions fi, then a valid signal 30 is therefore forwarded to an evaluation unit.

[0042] FIG. 2 shows the assembly 10 shown in FIG. 1 in one embodiment. The assembly 10 now additionally has a filter device 17, in which a signal correction method WF executes, which corrects or estimates the digital signal xd[k] and the additional overcontrol portions fk produced by the cutting are as a result eliminated or at least reduced in an estimate signal y[k]. The filter device 17 further has a transfer function G(w), which still maps the known damage frequencies fs, where the filter device 17 has an input 18 for the signal quality Q and the filter device 17 is configured to further use the transfer function G(w) up to a predeterminable limit value Vt as a function of the signal quality Q.

[0043] With additional reference to FIG. 2, the assembly 10 further has an adjustment device 19, which is configured to modify the transfer function G(w) as a function of the signal quality Q. In the embodiment in accordance with FIG. 2, the assembly 10 is also provided with a learning device 20, which is configured to form and archive a series of transfer functions Gj(w) in a learning phase in a quality state of the apparatus 1 as a function of the signal quality Q. Furthermore, the learning device 20 is configured to take as reference the transfer function P(w) in the original quality state of a monitored apparatus 1.

[0044] A Wiener filter is implemented in the filter device 17.

[0045] FIG. 3 shows a schematic representation of a status monitoring of an apparatus 1. The apparatus 1 comprises a motor with a transmission and a drive shaft and an electrical mass. An imbalance 31 sits on the electrical mass, for instance. A measurement 33 is recorded in the quality state via a sensor 32. On account of the imbalance 31 on the electrical mass, a measurement 34 with an increased amplitude of the machine vibration can be determined as a malfunction. The associated spectrum 35 of the vibration speeds shows three different frequency portions for three different rotation frequencies.

[0046] FIG. 4 shows by way of example, for instance, which errors can be determined with the aid of damage frequencies on a frequency spectrum. Therefore, the spectrum 40 with interference frequencies shows at points 41 a belt error, at points 42 a rolling bearing damage, at points 43 a blade pass frequency and at points 44 a meshing defect.

[0047] FIG. 5 shows from top down a raw signal x′a(t) that reaches the limit, where the measuring range MB is shown in the signal course x′a(t) and as soon as the signal portions of the raw signal x′a(t) leave the measuring range, these are cut. The signal quality Q is applied therebeneath, which reduces from 100% to 0% with an increasing clipping. A number Nsat of clipped signal portions mapped therebelow behaves in a conversely proportional manner.

[0048] FIG. 6 shows a spectral power density PSD, where the diagram shows the power density 61 for x with x, on the one hand, and the power density 62 for x with the clipped signal, on the other.

[0049] FIG. 7 shows the filters constructed from the courses in FIG. 6 in its transfer function according to Wiener. Clipping produces above all at high frequencies spectral lines with amplitudes up to 10%. This behavior, transferred to bearing damage analyses, means that spectral lines, which are available for a bearing damage, could be faded in. It either results in a misinterpretation (false positive) or a bearing damage is graded as fatal or as it actually is. This is because the existing spectral lines and harmonics are overlaid by the interference portions produced by clipping.

[0050] The estimated signal formed from the original signal using the Wiener filter now has almost exclusively the three main components of the original signal. The false portions of the estimated signal are in the range <1%, i.e., spectral lines produced by clipping are suppressed by 10.sup.−2 and existing spectral lines are broadened. The Wiener filter can be optimized for so long by continuous calculation of the spectral power densities PSD of PSDxx and PSDXS using a transfer function resulting therefrom, provided the inventive measure Q, or a similar number of saturated measured values exceeds a threshold.

[0051] FIG. 8 shows in an FFT spectrum 80 an original signal Xa[k], an observed signal Xd[k], which contains clipping portions and the signal y[k] estimated by the Wiener filter.

[0052] The signal Y estimated from X with the Wiener filter has almost exclusively the three main components of the original signal. The false portions of the estimated signal are in the range <1%, i.e., spectral lines produced by clipping are suppressed by 10.sup.−2 and existing spectral lines are shown broadened.

[0053] FIG. 9 is a flowchart the method for monitoring the status of an apparatus 1 by measuring an analog signal xa(t) that is continuous in terms of time and value and that represents a status variable of the apparatus 1.

[0054] The method comprises converting the analog signal xa(t), at scanning times is within a measuring interval TN, into a digital signal xd[k] via an analog-digital converter ADU which is operated within a measuring range MB, as indicated in step 910.

[0055] Next, signal portions of the analog signal xa(t) that which extend beyond the measuring range MB are cut in the digital signal xd[k] in an event the analog signal xa(t) exceeds the measuring range (MB), as indicated in step 920.

[0056] Next, a spectral analysis is applied to the digital signal xd[k] to determine which frequency portions fi the analog signal xa(t) comprises in a frequency spectrum FFT, and to conclude a malfunction of the apparatus 1 based on a comparison with the occurring frequency portions fi and known damage frequencies fs, as indicated in step 930.

[0057] In accordance with the invention, in an event that, at a scanning time ts, the analog signal xa(t) that extends beyond the measuring range MB is cut in the digital signal xd[k], this event is detected and is determined as a number NSat. In addition, a signal quality Q is provided as a quotient from the number NSat relating to an overall scanning number N within the measuring interval TN, and the signal quality Q is utilized to assess whether the known damage frequencies fS can still be identified from the determined frequency portions fi of the frequency spectrum FFT, although additional overcontrol portions fk occur in the frequency spectrum as a result of possibly cut signal portions.

[0058] Thus, while there have been shown, described and pointed out fundamental novel features of the invention as applied to a preferred embodiment thereof, it will be understood that various omissions and substitutions and changes in the form and details of the methods described and the devices illustrated, and in their operation, may be made by those skilled in the art without departing from the spirit of the invention. For example, it is expressly intended that all combinations of those elements and/or method steps which perform substantially the same function in substantially the same way to achieve the same results are within the scope of the invention. Moreover, it should be recognized that structures and/or elements and/or method steps shown and/or described in connection with any disclosed form or embodiment of the invention may be incorporated in any other disclosed or described or suggested form or embodiment as a general matter of design choice. It is the intention, therefore, to be limited only as indicated by the scope of the claims appended hereto.