JP7444556B2 - Misalignment determination device, misalignment determination method, and rotating machine system - Google Patents

Description

The present invention relates to a technique for determining misalignment of a rotating machine.

If misalignment occurs in a rotating machine (for example, the rotating shaft of a motor and the rotating shaft of a pump), it will cause the rotating machine to malfunction. Therefore, techniques for determining misalignment of rotating machines have been proposed. For example, Patent Document 1 discloses, in a method for diagnosing an abnormality in a rotating mechanical system driven by an electric motor, a current signal during operation of the electric motor is sampled, fast Fourier transform is performed, and then the level of the spectral peak of the power supply frequency of the electric motor is determined. , the difference between the level of a spectral peak that appears at a frequency position symmetrical about the power supply frequency as the center frequency and separated by the rotational frequency of the rotor shaft of the electric motor is equal to or less than a preset value. The rotating machine is characterized in that it is determined that an alignment abnormality has occurred between the shaft of the rotor of the electric motor and the rotating shaft on the load side, which are connected by a transmission joint in the rotating mechanical system. A system abnormality diagnosis method is disclosed.

Japanese Patent Application Publication No. 2015-227889

In Patent Document 1, the difference between spectral peaks is used as an index value, and this is compared with a preset value (threshold value) to determine whether the alignment is consistent or misaligned (in other words, (determine whether misalignment has occurred). In this case, in order to prevent misjudgment, it is desirable that the difference between the index value in a state of matching alignment and the index value in a state of misalignment be as large as possible.

It is an object of the present invention to provide a misalignment determination device and a misalignment device that determine whether or not misalignment has occurred based on an index value that shows a large difference between a state in which alignment matches and a state in which misalignment occurs. An object of the present invention is to provide a determination method and a rotating mechanical system.

A misalignment determination device according to a first aspect of the present invention includes a generation unit that generates a frequency spectrum of the detection signal by performing spectrum analysis on a detection signal of an elastic wave generated from a rotating machine, and a frequency range indicated by the frequency spectrum. A processing unit that performs processing to remove the spectral intensity at a fundamental frequency and a frequency that is an integral multiple of the fundamental frequency from among the spectral intensities at , and within the frequency range indicated by the frequency spectrum subjected to the processing by the processing unit, a calculation unit that sets a predetermined range indicating a relatively high frequency range and calculates an index value based on the spectral intensity in the predetermined range; and compares the index value with a predetermined threshold; A determination unit that determines whether misalignment has occurred in the rotating machine.

Spectral intensity is, for example, voltage (amplitude) or energy. Among the frequency range shown in the frequency spectrum, the spectral intensity of the fundamental frequency and frequencies that are integral multiples of the fundamental frequency depends on the rotation speed of the rotating shaft of the rotating machine and has no correlation with misalignment. Therefore, the processing unit performs processing to remove these from the frequency spectrum.

The present inventors have determined that the state of misalignment is the state of misalignment with respect to the spectral intensity in a predetermined range indicating a relatively high frequency range among the spectral intensities in the frequency range indicated by the frequency spectrum processed by the processing unit. It was found that the spectral intensity tends to be higher than that under normal conditions. Therefore, the index value based on the spectral intensity in a predetermined range (for example, the average value of the spectral intensity) can have a large difference between a state of matching alignment and a state of misalignment.

As described above, the misalignment determination device according to the first aspect of the present invention determines whether the alignments match based on the index value that increases the difference between the aligned state and the misaligned state. In other words, it is possible to determine whether misalignment is occurring or not.

In the above configuration, the calculation unit sets the predetermined range to be a range from a maximum value of the frequency range to a half of the maximum value.

This configuration is an example of how to determine the predetermined range.

In the above configuration, the calculation unit calculates a total value of the spectral intensities in the entire frequency range, and in the frequency range, the value obtained by adding the spectral intensities from the lowest frequency side is the total value of the total value. The frequency that reaches 70% is specified, and the range from the specified frequency to the maximum value of the frequency range is defined as the predetermined range.

This configuration is another example of how to determine the predetermined range.

In the above configuration, the index value is a value obtained by dividing the average value of the spectral intensity in the predetermined range by the maximum value of the spectral intensity in the frequency spectrum before the processing by the processing unit.

The present inventors found that when comparing the misaligned state and the aligned state with respect to the maximum value of spectral intensity (spectral peak) in the frequency spectrum before being processed by the processing unit, the former is better. was found to be smaller than the latter. Therefore, with this configuration, it is possible to further increase the difference between the index value in a misaligned state and the index value in a matched alignment state.

A misalignment determination method according to a second aspect of the present invention includes a generation step of generating a frequency spectrum of the detection signal by performing spectrum analysis on a detection signal of an elastic wave generated from a rotating machine, and a frequency range indicated by the frequency spectrum. A processing step of removing the spectral intensity at a fundamental frequency and a frequency that is an integral multiple of the fundamental frequency from among the spectral intensities at, and within the frequency range indicated by the frequency spectrum subjected to the processing in the processing step, a calculation step of setting a predetermined range indicating a relatively high frequency range and calculating an index value based on the spectral intensity in the predetermined range; and comparing the index value with a predetermined threshold; and a determination step of determining whether misalignment has occurred in the rotating machine.

The misalignment determining method according to the second aspect of the present invention defines the misalignment determining device according to the first aspect of the present invention from a method perspective, and is similar to the misalignment determining device according to the first aspect of the present invention. It has the following effects.

A rotating machine system according to a third aspect of the present invention includes the misalignment determination device according to the first aspect of the present invention, the rotating machine, and a sensor that detects elastic waves generated from the rotating machine and outputs the detection signal. and.

A rotating machine system according to a third aspect of the present invention defines the misalignment determining device according to the first aspect of the present invention from the viewpoint of a rotating mechanical system, and the misalignment determining device according to the first aspect of the present invention. It has similar effects.

According to the present invention, it is possible to determine whether or not there is misalignment based on an index value that shows a large difference between a state in which alignment matches and a state in which misalignment occurs.

FIG. 1 is a block diagram of a misalignment determination device according to an embodiment. FIG. 2 is a schematic diagram of an experimental rotating machine. It is a graph showing an example of a detection signal detected during centering when the rotation speed of the rotating shaft is 500 rpm. It is a graph showing an example of a detection signal detected in a state of misalignment when the rotation speed of the rotating shaft is 500 rpm. 3B is a graph showing a frequency spectrum generated by spectrum analysis of the detection signal shown in FIG. 3A. FIG. 3B is a graph showing a frequency spectrum generated by spectral analysis of the detection signal shown in FIG. 3B. FIG. 4A is a graph illustrating a frequency spectrum that has been processed to exclude spectral intensities at the fundamental frequency and frequencies that are integral multiples of the fundamental frequency from among the spectral intensities in the frequency range indicated by the frequency spectrum shown in FIG. 4A. FIG. 4B is a graph showing a frequency spectrum that has been processed to remove the fundamental frequency and the spectrum intensities at frequencies that are integral multiples of the fundamental frequency among the spectrum intensities in the frequency range shown by the frequency spectrum shown in FIG. 4B. FIG. It is a graph showing an example of a detection signal detected during centering when the rotation speed of the rotating shaft is 100 rpm. It is a graph showing an example of a detection signal detected in a state of misalignment when the rotation speed of the rotating shaft is 100 rpm. 6A is a graph showing a frequency spectrum generated by spectral analysis of the detection signal shown in FIG. 6A. FIG. 6B is a graph showing a frequency spectrum generated by spectral analysis of the detection signal shown in FIG. 6B. FIG. 7A is a graph illustrating a frequency spectrum that has been processed to exclude spectral intensities at the fundamental frequency and frequencies that are integral multiples of the fundamental frequency from among the spectral intensities in the frequency range indicated by the frequency spectrum shown in FIG. 7A. FIG. 7B is a graph showing a frequency spectrum that has been processed to exclude the fundamental frequency and the spectrum intensities at frequencies that are integral multiples of the fundamental frequency among the spectrum intensities in the frequency range shown by the frequency spectrum shown in FIG. 7B. FIG. It is a graph which shows the example of the detection signal detected at the time of centering when the rotation speed of a rotating shaft is 1000 rpm. It is a graph showing an example of a detection signal detected in a state of misalignment when the rotation speed of the rotating shaft is 1000 rpm. 9A is a graph showing a frequency spectrum generated by spectral analysis of the detection signal shown in FIG. 9A. FIG. 9B is a graph showing a frequency spectrum generated by spectral analysis of the detection signal shown in FIG. 9B. FIG. 10A is a graph showing a frequency spectrum that has been processed to remove the spectral intensities at the fundamental frequency and frequencies that are integral multiples of the fundamental frequency among the spectral intensities in the frequency range shown by the frequency spectrum shown in FIG. 10A. FIG. 10B is a graph showing a frequency spectrum that has been processed to remove the spectral intensities at the fundamental frequency and frequencies that are integral multiples of the fundamental frequency from among the spectral intensities in the frequency range shown by the frequency spectrum shown in FIG. 10B. FIG. It is a bar graph showing the relationship between the rotation speed of a rotating shaft and the average value of spectrum intensity in a frequency range indicated by a frequency spectrum. It is a bar graph showing the relationship between the rotation speed of a rotating shaft and the average value of spectral intensity in a predetermined range. FIG. 2 is a schematic diagram of a rotating machine system including a rotating machine to be determined as to whether or not misalignment has occurred. 3 is a flowchart illustrating the operation of the misalignment determination device according to the embodiment. It is a bar graph showing the relationship between the rotation speed of the rotating shaft and the value obtained by dividing the average value of the spectral intensity in a predetermined range by the spectral peak. It is a graph showing the relationship between the rotation speed, the spectral peak, and the average value of the spectral intensity, which was generated using data at the time of centering.

Hereinafter, embodiments of the present invention will be described in detail based on the drawings. In each figure, components given the same reference numerals indicate the same components, and descriptions of the components that have already been described will be omitted.

FIG. 1 is a block diagram of a misalignment determination device 1 according to an embodiment. The misalignment determination device 1 includes a control processing section 10, an IF section 11, an input section 13, and an output section 14.

The control processing unit 10 controls the entire misalignment determination device 1 and performs control and processing necessary for the operation of the misalignment determination device 1. The control processing unit 10 includes hardware such as a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and an HDD (Hard Disk Drive) to execute the functions of the control processing unit 10. It is realized by programs and data, etc.

The IF unit 11 is connected to the control processing unit 10 and inputs and outputs signals and the like to and from external devices under the control of the control processing unit 10. For example, the IF unit 11 receives the detection signal S output by the AE sensor 12 and sends it to the control processing unit 10. The IF section 11 is realized by an input/output interface circuit.

An AE (Acoustic Emission) sensor 12 detects elastic waves generated from a rotating machine and outputs a detection signal S. The detection signal S is a time series signal. A vibration sensor may be used instead of the AE sensor 12.

The input unit 13 is connected to the control processing unit 10 and is a device for an operator to input various information, data, commands, etc. The input unit 13 is realized by a mouse, a keyboard, a touch panel, etc. The output unit 14 is a device that is connected to the control processing unit 10 and outputs commands, data, misalignment occurrence notification, etc. input from the input unit 13 under the control of the control processing unit 10. The output unit 14 is realized by a liquid crystal display, an organic EL display (Organic Light Emitting Diode display), or the like.

The control processing unit 10 includes a storage unit 101, a generation unit 102, a processing unit 103, a calculation unit 104, and a determination unit 105 as functional blocks. The control processing unit 10 performs AD conversion (analog-to-digital conversion) on the detection signal S sent from the IF unit 11 and stores it in the storage unit 101 .

The storage unit 101 stores the AD-converted detection signal S and the like.

The generation unit 102 reads the detection signal S from the storage unit 101 and generates a frequency spectrum of the detection signal S by performing spectrum analysis on the detection signal S. The spectral analysis is, for example, a fast Fourier transform. A frequency spectrum shows the relationship between frequency and spectral intensity. Spectral intensity is, for example, voltage (amplitude) or energy.

The frequency range indicated by the frequency spectrum is determined by which range of frequency components, among the frequency components included in the detection signal S, are targeted for spectrum analysis. For example, when 0 to 500 kHz is targeted for spectrum analysis, the frequency range indicated by the frequency spectrum is set to 0 to 500 kHz.

The processing unit 103 performs processing to remove the fundamental frequency and the spectral intensities at frequencies that are integral multiples of the fundamental frequency from among the spectral intensities in the frequency range indicated by the frequency spectrum. The fundamental frequency is a value calculated based on the rotating machine, and for example, the fundamental frequency in this embodiment is 10 kHz regardless of the rotation speed.

Among the frequency range shown in the frequency spectrum (for example, 0 to 500 kHz), for the fundamental frequency and frequencies that are integral multiples of the fundamental frequency, the spectral intensity depends on the rotation speed of the rotating machine and is correlated with misalignment. does not have. Therefore, the processing unit 103 performs processing to remove these from the frequency spectrum.

The calculation unit 104 sets a predetermined range indicating a relatively high frequency range within the frequency range indicated by the frequency spectrum processed by the processing unit 103, and calculates an index value based on the spectrum intensity in the predetermined range. do.

The present inventors have determined that, among the spectral intensities in the frequency range (for example, 0 to 500 kHz) indicated by the frequency spectrum processed by the processing unit 103, the spectral intensities in a predetermined range indicating a relatively high frequency range are It has been found that the spectral intensity tends to be larger in the aligned state than in the matched alignment state. Therefore, the index value based on the spectral intensity in a predetermined range (for example, the average value, median value, mode value of the spectral intensity) should have a large difference between the state of misalignment and the state of matching alignment. I can do it.

The predetermined range is, for example, a range from the maximum value of the frequency range to a value half of the maximum value. In this case, when the frequency range is 0 to 500 kHz, the predetermined range is 250 to 500 kHz.

The determination unit 105 compares the index value with a predetermined threshold value and determines whether misalignment has occurred in the rotating machine.

FIG. 2 is a schematic diagram of the experimental rotating machine 3. The experimental rotating machine 3 includes a servo motor 32, a shaft 33, housings 35a and 35b, and a coupling 34.

The servo motor 32 is fixed onto the base 31 with bolts. A shaft 33 is disposed above the base 31 along the central axis AX of the rotating shaft 321 of the servo motor 32 . One end of the shaft 33 is rotatably supported by a bearing (not shown) built into the housing 35a, and the other end is rotatably supported by a bearing (not shown) built into the housing 35b. ing. The housing 35a is fixed onto the base 31 with bolts via shims 36. Housing 35b is fixed onto base 31 with bolts.

The rotating shaft 321 and the shaft 33 are connected by a coupling 34. The center axis AX of the rotating shaft 321 and the center axis (not shown) of the shaft 33 are forcibly shifted by the shim 36. This causes a misalignment state.

The housing 35a is located closer to the coupling 34 than the housing 35b, and the AE sensor 12 is installed in the housing 35a.

The experimental conditions are as follows. The outer diameter of the shaft 33 was 40 mm, the thickness of the shim 36 was 0.6 mm, and the sampling frequency of the AE sensor 12 was 1 MHz.

FIG. 3A is a graph showing an example of a detection signal S detected during centering when the rotational speed of the rotating shaft 321 is 500 rpm. Centering is matching the alignment. Therefore, during centering, the alignment is consistent. FIG. 3B is a graph showing an example of a detection signal S detected in a misaligned state when the rotational speed of the rotating shaft 321 is 500 rpm. In FIGS. 3A and 3B, the vertical axis indicates the voltage of the detection signal S, and the horizontal axis indicates time.

FIG. 4A is a graph showing a frequency spectrum generated by spectral analysis of the detection signal S shown in FIG. 3A. FIG. 4B is a graph showing a frequency spectrum generated by spectral analysis of the detection signal S shown in FIG. 3B. All spectral analyzes are fast Fourier transforms. In FIGS. 4A and 4B, the vertical axis indicates spectral intensity (voltage), and the horizontal axis indicates frequency.

FIG. 5A is a graph showing a frequency spectrum in which the fundamental frequency and the spectrum intensities at frequencies that are integral multiples of the fundamental frequency are removed from among the spectrum intensities in the frequency range shown by the frequency spectrum shown in FIG. 4A. FIG. 5B is a graph showing a frequency spectrum in which the fundamental frequency and the spectral intensities at frequencies that are integral multiples of the fundamental frequency are removed from among the spectral intensities in the frequency range shown by the frequency spectrum shown in FIG. 4B. In FIGS. 5A and 5B, the vertical axis indicates spectral intensity (voltage), and the horizontal axis indicates frequency.

The fundamental frequency is here 10kHz. The process for removing the spectral intensity is, for example, within a range of ±1 kHz around each frequency. That is, processing is performed to remove the spectral intensity of 9 to 11 kHz, 19 to 21 kHz, 29 to 31 kHz, and so on. In FIGS. 5A and 5B, there are gaps at regular intervals along the horizontal axis. These gaps indicate locations where spectral intensity has been removed.

In FIG. 4A, FIG. 4B, FIG. 5A, and FIG. 5B, the spectral peak (maximum value of voltage) in the frequency range (0 to 500 kHz) was significantly larger than 0.1 mV. If the upper limit of the vertical axis of the graph is increased to show the spectral peak, the graph becomes difficult to understand, so the upper limit of the vertical axis is set to 0.1 mV.

Although not shown in FIGS. 4A and 4B, the spectral peak in the frequency spectrum is It was found that the state of misalignment (FIG. 4B) is smaller than the state of centering (FIG. 4A) (first event). This is considered to be because the amplitude of the voltage is smaller in the misaligned state (FIG. 3B) than in the centering state (FIG. 3A). Note that during centering, the alignment is consistent.

Referring to FIGS. 5A and 5B, after processing to remove the spectral intensity at the fundamental frequency and frequencies that are integral multiples of the fundamental frequency (after processing in processing unit 103), the frequency range (0 to Focusing on 250 to 500 kHz (predetermined range), it was found that the spectral intensity was higher in the misaligned state (Figure 5B) than in the centering state (Figure 5A) (second event). . This is considered to be because the detection signal S fluctuates more finely in the misalignment state (FIG. 3B) than in the centering state (FIG. 3A).

The first event and the second event occurred even when the rotational speed of the rotating shaft 321 was 100 rpm. A detailed explanation will be omitted, but a graph will be shown below. FIG. 6A is a graph showing an example of a detection signal S detected during centering when the rotational speed of the rotating shaft 321 is 100 rpm, and corresponds to FIG. 3A. FIG. 6B is a graph showing an example of a detection signal S detected in a misaligned state when the rotational speed of the rotating shaft 321 is 100 rpm, and corresponds to FIG. 3B.

FIG. 7A is a graph showing a frequency spectrum generated by spectrum analysis of the detection signal S shown in FIG. 6A, and corresponds to FIG. 4A. FIG. 7B is a graph showing a frequency spectrum generated by spectrum analysis of the detection signal S shown in FIG. 6B, and corresponds to FIG. 4B.

FIG. 8A is a graph showing a frequency spectrum that has been processed to remove the fundamental frequency and the spectral intensities at frequencies that are integral multiples of the fundamental frequency among the spectrum intensities in the frequency range shown by the frequency spectrum shown in FIG. 7A. handle. FIG. 8B is a graph showing a frequency spectrum that has been processed to remove the spectral intensities at the fundamental frequency and frequencies that are integral multiples of the fundamental frequency among the spectral intensities in the frequency range shown by the frequency spectrum shown in FIG. 7B. handle.

The first event and the second event occurred even when the rotational speed of the rotating shaft 321 was 1000 rpm. A detailed explanation will be omitted, but a graph will be shown below. FIG. 9A is a graph showing an example of the detection signal S detected during centering when the rotational speed of the rotating shaft 321 is 1000 rpm, and corresponds to FIG. 3A. FIG. 9B is a graph showing an example of a detection signal S detected in a misaligned state when the rotational speed of the rotating shaft 321 is 1000 rpm, and corresponds to FIG. 3B.

FIG. 10A is a graph showing a frequency spectrum generated by spectrum analysis of the detection signal S shown in FIG. 9A, and corresponds to FIG. 4A. FIG. 10B is a graph showing a frequency spectrum generated by spectrum analysis of the detection signal S shown in FIG. 9B, and corresponds to FIG. 4B.

FIG. 11A is a graph showing a frequency spectrum that has been processed to exclude the fundamental frequency and the spectrum intensities at frequencies that are integral multiples of the fundamental frequency among the spectrum intensities in the frequency range shown by the frequency spectrum shown in FIG. 10A. handle. FIG. 11B is a graph showing a frequency spectrum that has been processed to remove the spectral intensities at the fundamental frequency and frequencies that are integral multiples of the fundamental frequency among the spectral intensities in the frequency range shown by the frequency spectrum shown in FIG. 10B. handle.

FIG. 12 is a bar graph showing the relationship between the rotation speed of the rotating shaft 321 and the average value of the spectrum intensity in the frequency range indicated by the frequency spectrum. The frequency range represented by the frequency spectrum is 0 to 500 kHz. The average value here is not the value obtained by summing the spectral intensities from 0 to 500 kHz and dividing this value by 500 (denominator). Frequency bands (9 to 11 kHz, 19 to 21 kHz, 29 to 31 kHz, . . . ) whose spectral intensity has been removed are not included in the denominator.

(8A) shows the average value of the spectral intensity from 0 to 500 kHz in FIG. 8A (at the time of centering). (8B) shows the average value of the spectral intensity from 0 to 500 kHz in FIG. 8B (misalignment state).

(5A) shows the average value of the spectral intensity from 0 to 500 kHz in FIG. 5A (at the time of centering). (5B) shows the average value of the spectral intensity from 0 to 500 kHz in FIG. 5B (misalignment state). (11A) shows the average value of the spectral intensity from 0 to 500 kHz in FIG. 11A (at the time of centering). (11B) shows the average value of the spectral intensity from 0 to 500 kHz in FIG. 11B (misalignment state).

At any rotation speed, the average value of the spectral intensity is larger in the misaligned state than in the centering state. However, when the rotation speed is 100 rpm or 500 rpm, the difference between the average value of the spectral intensity in the misaligned state and the average value of the spectral intensity at the time of centering is small. Therefore, it is difficult to use a threshold value to determine whether misalignment has occurred.

FIG. 13 is a bar graph showing the relationship between the rotation speed of the rotating shaft 321 and the average value of the spectral intensity in a predetermined range. The predetermined range is 250 to 500 kHz, indicating a relatively high frequency range. The average value here is not the value obtained by summing the spectral intensities from 250 to 500 kHz and dividing this value by 250 (denominator). Frequency bands (249 to 251 kHz, 259 to 261 kHz, 269 to 271 kHz, . . . ) whose spectral intensity has been removed are not included in the denominator.

(8A) shows the average value of the spectral intensity at 250 to 500 kHz in FIG. 8A (at the time of centering). (8B) shows the average value of the spectral intensity from 250 to 500 kHz in FIG. 8B (misalignment state).

(5A) shows the average value of the spectral intensity at 250 to 500 kHz in FIG. 5A (at the time of centering). (5B) shows the average value of the spectral intensity from 250 to 500 kHz in FIG. 5B (misalignment state). (11A) shows the average value of the spectral intensity at 250 to 500 kHz in FIG. 11A (at the time of centering). (11B) shows the average value of the spectral intensity from 250 to 500 kHz in FIG. 11B (misalignment state).

At any rotation speed, the average value of the spectral intensity is larger in the misaligned state than in the centering state. Moreover, at any rotation speed, there is a large difference between the average value of the spectral intensity in the state of misalignment and the average value of the spectral intensity at the time of centering (the average value of the spectral intensity in the state of misalignment is (approximately 1.3 to 1.8 times the average value of the spectral intensity at centering). Therefore, it is possible to determine whether misalignment has occurred using a threshold value.

In the embodiment, the average value of the spectral intensity in the range of 250 to 500 kHz (predetermined range) is used as the index value. The determination unit 105 can determine whether or not misalignment has occurred at each rotation speed using a value obtained by multiplying the index value at centering by 1.3, for example, as a threshold value. As described above, according to the embodiment, it is determined whether misalignment has occurred based on the index value that shows a large difference between centering (a state in which the alignments match) and a misalignment state. can be determined.

The operation of the misalignment determination device 1 according to the embodiment will be explained. FIG. 14 is a schematic diagram of a rotating machine system 5000 including a rotating machine 5 that is a target for determining whether misalignment has occurred. The rotating machine system 5000 includes a rotating machine 5, a misalignment determination device 1, and an AE sensor 12.

The rotating machine 5 includes a base 51, a motor 52, a pump 53, and a coupling 54. The motor 52 and the pump 53 are fixed to the base 51 with the rotating shaft 521 of the motor 52 and the rotating shaft 531 of the pump 53 facing each other. The rotating shaft 521 and the rotating shaft 531 are connected by a coupling 54.

The AE sensor 12 outputs a detection signal S used to determine whether misalignment has occurred between the rotating shaft 521 and the rotating shaft 531. The AE sensor 12 is installed at a suitable location to achieve this purpose. For example, the AE sensor 12 is arranged in a portion of the housing of the pump 53 that is close to the rotation shaft 531.

The detection signal S output by the AE sensor 12 is sent to the misalignment determination device 1.

The storage unit 101 (FIG. 1) stores in advance a threshold value determined according to the rotation speed of the rotating shaft 521 of the motor 52. For example, when the number of rotations is n1, the threshold value is th1, and when the number of rotations is n2, the threshold value is th2.

FIG. 15 is a flowchart illustrating the operation of the misalignment determination device 1 according to the embodiment. Referring to FIGS. 1 and 15, an operator operates input unit 13 to input the rotation speed of rotating shaft 521 (FIG. 14) of motor 52, and inputs a misalignment monitoring command (S1). . The determination unit 105 reads the threshold value assigned to the input rotation speed from the storage unit 101. For example, when the input rotation speed is n1, the determination unit 105 reads the threshold value th1 assigned to the rotation speed n1 from the storage unit 101.

The control processing unit 10 transmits a signal for activating the AE sensor 12 to the AE sensor 12 via the IF unit 11 in response to the input of the misalignment monitoring command. As a result, the AE sensor 12 starts operating and outputs a detection signal S of an elastic wave generated from the rotating machine 5 (FIG. 14). The IF section 11 receives the detection signal S output by the AE sensor 12 (S2) and sends it to the control processing section 10.

The control processing unit 10 performs AD conversion on the detection signal S and stores it in the storage unit 101. The generation unit 102 reads the detection signal S stored in the storage unit 101 and generates a frequency spectrum of the detection signal S (S3).

The processing unit 103 performs processing to remove the fundamental frequency and the spectral intensities at frequencies that are integral multiples of the fundamental frequency from among the spectral intensities in the frequency range indicated by the frequency spectrum generated in processing S3 (S4).

The frequency range indicated by the frequency spectrum is stored in the storage unit 101 in advance. For example, it is 0 to 500kHz. The calculation unit 104 reads the frequency range from the storage unit 101, and calculates a predetermined range based on this frequency range (for example, 250 to 500 kHz). The calculation unit 104 sets a predetermined range within the frequency range indicated by the frequency spectrum processed in step S4, and calculates an index value based on the spectrum intensity in the predetermined range (S5). The index value is, for example, an average value of spectral intensity.

The determining unit 105 determines whether the index value exceeds a threshold value (S6). When determining that the index value does not exceed the threshold (No in S6), the determining unit 105 determines that misalignment has not occurred between the rotating shaft 521 and the rotating shaft 531 (FIG. 14) (S7). ). Then, the control processing unit 10 determines whether a command to stop misalignment monitoring has been input (S8). The stop command is input by the operator operating the input unit 13.

When the control processing unit 10 determines that a command to stop misalignment monitoring has been input (Yes in S8), the misalignment determination device 1 ends the misalignment monitoring. When the control processing unit 10 determines that a command to stop misalignment monitoring has not been input (No in S8), the misalignment determination device 1 returns to processing S2.

When determining that the index value exceeds the threshold value (Yes in S6), the determination unit 105 determines that misalignment has occurred between the rotation axis 521 and the rotation axis 531 (FIG. 14) (S9). ). The determination unit 105 uses the output unit 14 to notify the operator that misalignment has occurred (S10). Then, the control processing unit 10 performs processing S8.

Another example of how to determine the predetermined range will be explained. For example, referring to FIGS. 8A and 8B, even if the frequency is lower than 250 kHz, there is a range (240 to 250 kHz) in which the spectral intensity is higher in the misalignment state than in the centering state. In other examples, a predetermined range is determined that may also include this range. The calculation unit 104 calculates a total value of the spectral intensities in the entire frequency range (0 to 500 kHz) indicated by the frequency spectrum, and adds the spectral intensities from the side with the smallest frequency (frequency 0 side) in this frequency range. The frequency at which the value reaches 70% of the total value is specified, and the range from the specified frequency to the maximum value (500 kHz) of the frequency range is defined as a predetermined range.

The embodiment includes a first modification and a second modification. Regarding these, the points that are different from the embodiment will be mainly explained. The first modification will be explained first. In the first modification, the index value is a value obtained by dividing the average value of the spectral intensity in a predetermined range by the spectral peak. The spectral peak is the maximum value of the spectral intensity in the frequency spectrum (FIG. 4A, FIG. 4B, FIG. 7A, FIG. 7B, FIG. 10A, FIG. 10B) before being processed by the processing unit 103. Note that, as mentioned above, for drawing reasons, the spectral peaks are not shown in those figures.

The present inventors have found that the spectral peak becomes smaller in the misaligned state than in the centered state. Therefore, according to the first modification, it is possible to further increase the difference between the index value in the state of misalignment and the index value at the time of centering. explain in detail.

FIG. 16 is a bar graph showing the relationship between the rotation speed of the rotating shaft and the value obtained by dividing the average value of the spectral intensity in a predetermined range by the spectral peak. The predetermined range is 250 to 500 kHz, indicating a relatively high frequency range. (8A) shows the value obtained by dividing the average value of the spectral intensity at 250 to 500 kHz by the spectral peak in FIG. 8A (at the time of centering). (8B) shows the value obtained by dividing the average value of the spectral intensity at 250 to 500 kHz by the spectral peak in FIG. 8B (misalignment state).

(5A) shows the value obtained by dividing the average value of the spectral intensity at 250 to 500 kHz by the spectral peak in FIG. 5A (at the time of centering). (5B) shows the value obtained by dividing the average value of the spectral intensity at 250 to 500 kHz by the spectral peak in FIG. 5B (misalignment state). (11A) shows the value obtained by dividing the average value of the spectral intensity at 250 to 500 kHz by the spectral peak in FIG. 11A (at the time of centering). (11B) shows the value obtained by dividing the average value of the spectral intensity at 250 to 500 kHz by the spectral peak in FIG. 11B (misalignment state).

Hereinafter, the value obtained by dividing the average value of the spectral intensity by the spectral peak will be referred to as "average value/peak". At any rotation speed, there is a large difference between the "average value/peak" in the misaligned state and the "average value/peak" during centering (the "average value/peak" in the misaligned state is (approximately 1.8 to 2.7 times the "average value/peak" at the time of release).

Referring to FIGS. 13 and 16, the difference between the value at centering and the value in the misaligned state is greater in the case of "average value/peak" than in the case of the average value of spectral intensity. big. Therefore, by using "average value/peak" as an index value, it is possible to further improve the accuracy of determining whether misalignment has occurred.

A second modification will be explained. The threshold value to be compared with the index value is determined according to the rotation speed of the rotating shaft 521 (FIG. 14), and is stored in advance in the storage unit 101 (FIG. 1). When there are many types of rotational speeds of the rotating shaft 521 (for example, 80rpm, 90rpm, 100rpm, 110rpm, 120rpm, etc.), it is troublesome to prepare threshold values for all the rotational speeds in advance. It takes. In the second modification, a threshold value is calculated by interpolation or extrapolation for a rotation speed for which a threshold value is not stored in the storage unit 101.

FIG. 17 is a graph showing the relationship between the rotation speed, the spectral peak, and the average value of the spectral intensity, which was generated using the data at the time of centering. The horizontal axis of the graph indicates the rotation speed of the rotation shaft 321 (FIG. 2). The vertical axis on one side of the graph shows the spectral peaks. The vertical axis on the other side of the graph indicates the average value of the spectral intensity in a predetermined range. From the experimental results using the rotating machine 3 shown in FIG. 2, the average value of the spectrum peak and the spectrum intensity in a predetermined range is determined for each of the rotational speeds of 100 rpm, 500 rpm, and 1000 rpm during centering. In FIG. 17, these are plotted, and a line L1 indicating the spectral peak and a line L2 indicating the average value of the spectral intensity in a predetermined range are shown.

As can be seen from the line L1 indicating the spectral peak and the line L2 indicating the average value of the spectral intensity in a predetermined range, as the rotation speed of the rotating shaft 321 (FIG. 2) increases, the spectral peak and the average value of the spectral intensity in the predetermined range increase. Both are increasing linearly. Therefore, even at rotational speeds other than 100 rpm, 500 rpm, and 1000 rpm, the average value of the spectral peak and the spectral intensity in a predetermined range at the time of centering can be determined by interpolation processing and extrapolation processing.

The storage unit 101 stores in advance data indicating spectral peaks and average values of spectral intensities in predetermined ranges for each of a plurality of types of rotation speeds (for example, 100 rpm, 500 rpm, 1000 rpm) at the time of centering.

When the rotational speed of the rotating shaft 521 is a rotational speed other than the plurality of rotational speeds, the calculation unit 104 reads the data from the storage unit 101 and calculates the spectrum peak and the spectrum in a predetermined range by interpolation processing and extrapolation processing. The average value of the intensity is determined, and a threshold value is calculated based on this.

1 Misalignment determination device 3 Experimental rotating machine 31 Base 32 Servo motor 321 Rotating shaft 33 Shaft 34 Couplings 35a, 35b Housing 36 Shim 5 Rotating machine 51 Base 52 Motor 521 Rotating shaft 53 Pump 531 Rotating shaft 54 Coupling 5000 Rotating machine System AX Central axis L1 Line L2 indicating the spectral peak Line S indicating the average value of the spectral intensity in a predetermined range Detection signal

Claims (5)

a generation unit that generates a frequency spectrum of the detection signal by performing spectrum analysis on a detection signal of an elastic wave generated from a rotating machine;
A processing unit that performs processing to remove the spectral intensity at a fundamental frequency and a frequency that is an integral multiple of the fundamental frequency from among spectral intensities in a frequency range indicated by the frequency spectrum;
Within the frequency range indicated by the frequency spectrum processed by the processing unit, in a frequency region in which the spectrum intensity is higher in a misaligned state than in a matched alignment state, the maximum of the frequency range is a calculation unit that calculates an index value based on the spectral intensity in the predetermined range by setting a predetermined range from the maximum value to a half of the maximum value;
a determination unit that compares the index value with a predetermined threshold value and determines whether misalignment has occurred in the rotating machine;
The index value is a value obtained by dividing the average value of the spectral intensity in the predetermined range by the maximum value of the spectral intensity in the frequency spectrum before the processing by the processing unit,
The frequency range is 0 to 500kHz ,
Misalignment determination device. a generation unit that generates a frequency spectrum of the detection signal by performing spectrum analysis on a detection signal of an elastic wave generated from a rotating machine;
A processing unit that performs processing to remove the spectral intensity at a fundamental frequency and a frequency that is an integral multiple of the fundamental frequency from among spectral intensities in a frequency range indicated by the frequency spectrum;
a calculation unit that sets a predetermined range within the frequency range indicated by the frequency spectrum processed by the processing unit, and calculates an index value based on the spectrum intensity in the predetermined range;
a determination unit that compares the index value with a predetermined threshold value and determines whether misalignment has occurred in the rotating machine;
The predetermined range is such that in a frequency range in which the spectral intensity is higher in the misaligned state than in the aligned state, the value obtained by adding the spectral intensities from the lower frequency side is the The range is from a frequency that reaches 70% of the sum of the spectral intensities in the entire frequency range to the maximum value of the frequency range,
The index value is a value obtained by dividing the average value of the spectral intensity in the predetermined range by the maximum value of the spectral intensity in the frequency spectrum before the processing by the processing unit,
The frequency range is 0 to 500kHz ,
Misalignment determination device. a generation step of generating a frequency spectrum of the detection signal by performing spectrum analysis on the detection signal of an elastic wave generated from the rotating machine;
A processing step of removing the spectral intensities at a fundamental frequency and frequencies that are integral multiples of the fundamental frequency from among the spectral intensities in the frequency range indicated by the frequency spectrum;
Within the frequency range indicated by the frequency spectrum processed in the processing step, the maximum of the frequency range is in a frequency region where the spectral intensity is higher in a misaligned state than in a matched alignment state. a calculation step of setting a range from the value to half the maximum value as a predetermined range, and calculating an index value based on the spectral intensity in the predetermined range;
a determination step of comparing the index value with a predetermined threshold value and determining whether misalignment has occurred in the rotating machine;
The index value is a value obtained by dividing the average value of the spectral intensity in the predetermined range by the maximum value of the spectral intensity in the frequency spectrum before the processing by the processing unit,
The frequency range is 0 to 500kHz ,
Misalignment determination method. a generation step of generating a frequency spectrum of the detection signal by performing spectrum analysis on the detection signal of an elastic wave generated from the rotating machine;
A processing step of removing the spectral intensities at a fundamental frequency and frequencies that are integral multiples of the fundamental frequency from among the spectral intensities in the frequency range indicated by the frequency spectrum;
a calculation step of setting a predetermined range within the frequency range indicated by the frequency spectrum processed in the processing step, and calculating an index value based on the spectral intensity in the predetermined range;
a determination step of comparing the index value with a predetermined threshold value and determining whether misalignment has occurred in the rotating machine;
The predetermined range is such that in a frequency range in which the spectral intensity is higher in the misaligned state than in the aligned state, the value obtained by adding the spectral intensities from the lower frequency side is the The range is from a frequency that reaches 70% of the sum of the spectral intensities in the entire frequency range to the maximum value of the frequency range,
The index value is a value obtained by dividing the average value of the spectral intensity in the predetermined range by the maximum value of the spectral intensity in the frequency spectrum before the processing by the processing unit,
The frequency range is 0 to 500kHz ,
Misalignment determination method. The misalignment determination device according to claim 1 or 2, and the rotating machine,
a sensor that detects elastic waves generated from the rotating machine and outputs the detection signal;
Rotating mechanical system.

Images