WO2020157818A1

WO2020157818A1 - Diagnostic device, equipment comprising same, and diagnostic method

Info

Publication number: WO2020157818A1
Application number: PCT/JP2019/002883
Authority: WO
Inventors: 陽一松井; 優太小田原
Original assignee: ＰｒｉｍｅｔａｌｓＴｅｃｈｎｏｌｏｇｉｅｓＪａｐａｎ株式会社
Priority date: 2019-01-29
Filing date: 2019-01-29
Publication date: 2020-08-06
Also published as: JPWO2020157818A1; JP7077426B2

Abstract

This diagnostic device for diagnosing a piece of machinery that includes a driving part and a driven part rotationally driven by the driving part comprises: a plurality of acoustic sensors that are provided to different locations and that are for detecting sounds generated by the machinery; a sound data acquisition unit that is configured so as to acquire a plurality of sound data sets including the sounds detected by each of the plurality of acoustic sensors during each of a plurality of rotational speed conditions differing in the rotational speed of the driving part; a frequency analysis unit that is configured so as to analyze the frequencies of the plurality of sound data sets respectively acquired for the plurality of rotational speed conditions; a peak frequency identification unit that identifies, on the basis of the frequency analysis results from the frequency analysis unit, one or more rotational speed conditions in which the plurality of sound data sets share a peak frequency, and said peak frequency; and a sound source information acquisition unit that, for each of the one or more rotational speed conditions, obtains information related to a sound generation source, within the machinery, having a natural frequency corresponding to the peak frequency, the information being obtained on the basis of the installed location, relative to the machinery, of the one acoustic sensor having the largest amplitude in the sound data corresponding to the peak frequency, among the plurality of acoustic sensors.

Description

Diagnostic device, equipment including the same, and diagnostic method

The present disclosure relates to a diagnostic device for diagnosing a mechanical device that includes a drive unit and a driven unit that is rotationally driven by the drive unit, a facility including the diagnostic device, and a diagnostic method.

An acoustic sensor may be used to diagnose a mechanical device having a driven part that is rotationally driven by a driving part such as a motor.

For example, Patent Document 1 describes a device for detecting a phenomenon (so-called chattering) in which a striped pattern occurs on a steel plate due to vibration of a roll in a rolling mill including a rolling roll driven by a motor. This apparatus includes a microphone (acoustic sensor) installed in the vicinity of the rolling roll, and the microphone detects the acoustic waveform of the sound from the rolling mill. Then, the chattering described above is detected by monitoring the sound of the natural frequency of the rolling roll acquired in advance based on the acoustic waveform thus detected.

Japanese Patent Laid-Open No. 2000-158044

By the way, as for the phenomenon that the occurrence frequency is relatively high, such as the chattering described above, it occurred by using a test device prepared for a mechanical device (rolling machine, etc.) in which the phenomenon occurs, or actually occurred in an actual machine. By analyzing the phenomenon, the natural frequency of the mechanical device can be specified.
However, for an event that may cause a problem in a mechanical device but occurs infrequently (for example, damage to a driving part or a driven part), it is costly to fabricate a test device to investigate the natural frequency of a target part or the like. However, it is usually difficult to specify the natural frequency by analysis based on an actual machine because the frequency of occurrence of an event is low. Therefore, it is desired to specify the natural frequency of the mechanical device with a simple configuration.

In view of the above circumstances, at least one embodiment of the present invention has an object to provide a diagnostic device capable of identifying the natural frequency of a mechanical device with a simple configuration, a facility including the diagnostic device, and a diagnostic method.

(1) The diagnostic device according to at least one embodiment of the present invention is
A diagnostic device for diagnosing a mechanical device including a drive part and a driven part rotationally driven by the drive part,
A plurality of acoustic sensors provided at different positions, respectively for detecting sounds generated from the mechanical device,
A sound data acquisition unit configured to acquire a plurality of sound data respectively including sounds detected by each of the plurality of acoustic sensors in each of a plurality of rotation speed conditions in which the rotation speed of the drive unit is different,
A frequency analysis unit configured to perform frequency analysis on each of the plurality of sound data acquired for each of the plurality of rotation speed conditions,
A peak frequency specifying unit that specifies the peak frequency and at least one rotation speed condition in which a peak frequency common to the plurality of sound data exists, based on a frequency analysis result by the frequency analyzing unit;
For each of the one or more rotation speed conditions, the machine based on the installation position of the one acoustic sensor having the largest amplitude of the sound data corresponding to the peak frequency among the plurality of acoustic sensors with respect to the mechanical device. Of the device, a sound source information acquisition unit that obtains information about a source of sound having a natural frequency corresponding to the peak frequency,
Equipped with.

According to at least one embodiment of the present invention, there is provided a diagnostic device capable of specifying the natural frequency of a mechanical device with a simple configuration, a facility including the diagnostic device, and a diagnostic method.

It is a schematic structure figure showing rolling equipment (installation) concerning one embodiment. It is a schematic structure figure of a diagnostic device concerning one embodiment. It is a flow chart which shows the outline of the diagnostic method concerning one embodiment. It is a figure showing an example of a frequency analysis result of sound data acquired using a plurality of acoustic sensors under a specific motor rotation number condition. It is a figure showing an example of a frequency analysis result of sound data acquired using a plurality of acoustic sensors under a specific motor rotation number condition. It is a figure showing an example of a frequency analysis result of sound data acquired using a plurality of acoustic sensors under a specific motor rotation number condition. It is a figure which shows an example of the information memorize|stored in the memory|storage part which concerns on one Embodiment. It is a flow chart which shows the outline of the diagnostic method concerning one embodiment.

Hereinafter, some embodiments of the present invention will be described with reference to the accompanying drawings. However, the dimensions, materials, shapes, relative positions, and the like of the components described as the embodiments or shown in the drawings are not intended to limit the scope of the present invention thereto, but are merely illustrative examples. Absent.

First, with reference to FIG. 1, rolling equipment, which is an example of equipment including a diagnostic device according to some embodiments, will be described. FIG. 1 is a schematic configuration diagram showing rolling equipment (equipment) according to an embodiment.

As shown in FIG. 1, rolling equipment (equipment) includes a rolling mill 1 and a diagnostic device 30 for diagnosing the rolling mill 1.

The rolling apparatus 1 includes a motor 2 (driving unit), a rolling roll 8 (driven unit) that is rotationally driven by the motor 2, and a power transmission unit 3 that transmits the power of the motor 2 to the rolling roll 8. Including.

The rolling roll 8 is configured to roll the metal strip plate 9, and includes a pair of

work rolls

12A and 12B for sandwiching the metal strip plate 9 from above and below to apply a load to the metal strip plate 9 and a pair of works. The metal strip 9 includes the pair of

intermediate rolls

14A and 14B and the pair of

backup rolls

16A and 16B provided on opposite sides of the

rolls

12A and 12B, respectively. The

intermediate rolls

14A and 14B are provided between the

work rolls

12A and 12B and the

backup rolls

16A and 16B, respectively.

The power transmission unit 3 includes a gear 4 (driven unit) and a spindle 6 (driven unit) that are rotationally driven by the motor 2. That is, the gear 4 is connected to the motor 2, and the spindle 6 is connected to the motor 2 via the gear 4. The rolling roll 8 is connected to the motor 2 via the gear 4 and the spindle 6. Therefore, when the gear 4 and the spindle 6 are rotationally driven by the motor 2, the rolling roll 8 is driven by the gear 4 and the spindle 6. That is, the rotational movement of the motor 2 is transmitted to the gear 4, the spindle 6 and the rolling roll 8, and the gear 4, the spindle 6 and the rolling roll 8 rotate at the number of rotations corresponding to the output (rotation number) of the motor 2, respectively. It is supposed to do.

Next, the configuration of the diagnostic device 30 according to the embodiment will be described with reference to FIGS. 1 and 2. FIG. 2 is a schematic configuration diagram of the diagnostic device 30 according to the embodiment.

As shown in FIG. 1, a diagnostic device 30 according to an embodiment includes a plurality of

acoustic sensors

32, 34A, 34B for detecting sounds generated from the rolling mill 1, and a plurality of

acoustic sensors

32, 34A, 34B. A processing unit 36 for processing the sound data detected by, a storage unit 37 for storing the processing result of the processing unit 36, and a display unit 38 for displaying the processing result of the processing unit 36. , Is included.

The plurality of

acoustic sensors

32, 34A, 34B are provided at different positions.
In the illustrated embodiment, the plurality of acoustic sensors are provided corresponding to the first acoustic sensor 32 provided corresponding to the motor 2, the second acoustic sensor 34A provided corresponding to the gear 4, and the spindle 6. The second acoustic sensor 34B is included. That is, the first acoustic sensor 32 is provided near the motor 2, the second acoustic sensor 34A is provided near the gear 4, and the second acoustic sensor 34B is provided near the spindle 6. In the following description, the first acoustic sensor 32 and the second

acoustic sensors

34A, 34B may be collectively referred to as

acoustic sensors

32, 34A, 34B.
In some embodiments, the

acoustic sensors

32, 34A, 34B may include sound collection microphones.

The processing unit 36 may include a CPU, a memory (RAM), an auxiliary storage unit, an interface, and the like. The processing unit 36 may receive sound data indicating sounds detected by the

acoustic sensors

32, 34A, 34B via the interface. The CPU is configured to process the sound data thus received. In addition, the CPU is configured to process the program loaded in the memory.

The processing content of the processing unit 36 may be implemented as a program executed by the CPU and stored in the auxiliary storage unit. When the programs are executed, these programs are expanded in the memory. The CPU reads the program from the memory and executes the instructions included in the program.

The storage unit 37 may include an external storage unit provided outside the device (computer or the like) configuring the processing unit 36. Alternatively, the storage unit 37 may be a storage unit provided inside the device configuring the processing unit 36, and may be, for example, the above-described memory or auxiliary storage unit.

As shown in FIG. 2, the processing unit 36 includes a sound data acquisition unit 40 for acquiring sound data including sounds from the

acoustic sensors

32, 34A, 34B, and a frequency analysis unit for performing frequency analysis of the sound data. 42, and a frequency analysis result, a peak frequency specifying unit 48 for specifying a peak frequency common to a plurality of sound data, and information about a sound source having a natural frequency corresponding to the specified peak frequency. Sound source information acquisition unit 50 for Further, the processing unit 36 removes, from the frequency analysis result of the sound data, a frequency component that is not related to the natural frequency of the rolling mill 1 (mechanical device) to be identified, and the second removing unit 44 and the second removing unit. Including 46. Further, the processing unit 36 diagnoses the rolling mill 1 based on the natural frequency (peak frequency) specified by the peak frequency specifying unit 48 and the sound source information acquiring unit 50 and the information about the source of the sound of the natural frequency. The diagnostic unit 54 for performing the operation is included.

The sound data acquisition unit 40 outputs a plurality of sound data including sound detected by each of the plurality of

acoustic sensors

32, 34A, 34B under each of a plurality of rotation speed conditions in which the rotation speed of the motor 2 (driving unit) is different. Configured to get.

The frequency analysis unit 42 is configured to frequency analyze each of the plurality of sound data acquired by the sound data acquisition unit 40 for each of the plurality of rotation speed conditions. In one embodiment, the frequency analysis unit 42 is configured to obtain the frequency spectrum showing the correlation between the frequency and the amplitude for each of the plurality of sound data by performing the frequency analysis described above.

The peak frequency identification unit 48 is configured to identify one or more rotation speed conditions in which a peak frequency common to a plurality of sound data exists, based on the frequency analysis result by the frequency analysis unit 42.

The sound source information acquisition unit 50, for each of the above-described one or more rotation speed conditions identified by the peak frequency identification unit 48, a sound corresponding to the above-described common peak frequency among the plurality of

acoustic sensors

32, 34A, 34B. One acoustic sensor (any of the

acoustic sensors

32, 34A, 34B) having the largest data amplitude is specified. Then, based on the installation position of the specified acoustic sensor with respect to the rolling mill 1, the rolling mill 1 is configured to obtain information regarding the sound source of the sound of the natural frequency corresponding to the above-mentioned common peak frequency. There is.

Here, the information relating to the sound source is information indicating the position or part where the sound is generated in the rolling mill 1, that is, which of the plurality of acoustic sensors is the position or part close to which. It is information to show. For example, when the amplitude at the above-mentioned common peak frequency is the largest in the sound data acquired by the acoustic sensor 32 among the sound data acquired by the

acoustic sensors

32, 34A, 34B, the common peak frequency is the rolling device 1 Among these, it can be estimated that it is the natural frequency of a portion located closer to the acoustic sensor 32 than the

acoustic sensors

34A and 34B. By inputting to the sound source information acquisition unit 50 that the mechanical device closest to the acoustic sensor 32 among the motor 2, the gear 4, and the spindle 6 is input to the sound source information acquisition unit 50, not the gear 4 or the spindle 6, but the natural vibration of the motor 2. It can be estimated to be a number. Alternatively, even if the specific portion having the natural frequency cannot be specified, the approximate position of the portion having the natural frequency can be grasped. The processing unit 36 (sound source information acquisition unit 50) may be configured to obtain information about which mechanical device or position in the rolling mill 1 is closest to which acoustic sensor in a process before and after acquiring sound data. The information may be stored in advance.

The above-mentioned common peak frequency corresponding to the rotation speed condition specified by the peak frequency specifying unit 48 and the information regarding the sound source specified by the sound source information acquiring unit 50 are stored in the storage unit 37. It is like this. The storage unit 37 is configured to store information in which the above-described common peak frequency and the above-described information regarding the sound source are associated with each other.

The first removal unit 44 is configured to remove a component having a common frequency from the frequency analysis results of the plurality of sound data by the frequency analysis unit 42 regardless of the rotation speed condition of the motor 2 (driving unit). When the above-mentioned frequency component is removed by the first removal unit 44, the sound source information acquisition unit 50 regards the frequency component thus removed as not the natural frequency of the rolling apparatus 1 and sets the frequency component as described above. It is arranged to obtain information about the source of the sound, excluding it from the peak frequencies.

When amplitude peaks appear at the same frequency in a plurality of sound data acquired under different rotation speed conditions, the frequency is derived from the background sound that occurs regardless of the rotation speed of the motor 2 (driving unit). Presumed to be a thing. Therefore, as described above, regarding the frequency analysis result of a plurality of sound data, the frequency analysis of the plurality of sound data is performed when the component having the common frequency is removed regardless of the rotation speed condition to obtain the information about the sound source. By excluding the above-mentioned frequency component related to the background sound from the common peak frequency in the result, the natural frequency of the rolling mill 1 can be accurately specified based on the frequency analysis result of the plurality of sound data. ..

For example, if there is a plurality of common peak frequencies described above under a certain rotation speed condition, and one of the common peak frequencies is a frequency component removed by the first removing unit 44, another common peak frequency is used. Then, the information regarding the source of the sound of the natural frequency corresponding to the "other" peak frequency is obtained.

The second removing unit 46 is configured to remove a component whose frequency is proportional to the rotation speed of the motor 2 (driving unit) in the frequency analysis result of the plurality of sound data by the frequency analyzing unit 42. When the above-mentioned frequency component is removed by the second removal unit 46, the sound source information acquisition unit 50 regards the frequency component thus removed as not the natural frequency of the rolling apparatus 1 and sets the frequency component as described above. It is arranged to obtain information about the source of the sound, excluding it from the peak frequencies.

When a peak of a frequency component proportional to the rotation speed of the motor 2 (driving unit) appears in a plurality of sound data acquired under different rotation speed conditions, the frequency is caused by the magnitude of the rotational movement of the motor 2. It is presumed to be a noise that is generated and does not relate to the natural frequency of the rolling mill 1. Therefore, as described above, in the frequency analysis result of the plurality of sound data, the frequency analysis of the plurality of sound data is performed when the component whose frequency is proportional to the rotation speed of the motor 2 is removed to obtain the information about the sound source. By excluding the above-mentioned frequency components that are not related to the natural frequency of the rolling mill 1 from the common peak frequency in the results, the natural frequency of the rolling mill 1 can be accurately determined based on the frequency analysis results of a plurality of sound data. Can be specified.

The diagnostic unit 54 determines the amplitude of the above-described peak frequency component (that is, the frequency component corresponding to the natural frequency of the identified sound source) in the frequency analysis result of the diagnostic sound data by the frequency analysis unit 42. On the basis of the above, the sound source is determined to be abnormal.

The diagnostic sound data is acquired by the diagnostic acoustic sensor 35 near the rolling mill 1. The diagnostic acoustic sensor 35 detects the natural frequency and the sound for identifying the source of the sound of the natural frequency (that is, the peak frequency specifying unit 48 and the sound source information acquiring unit 50 process the sound. Any of the

acoustic sensors

32, 34A, 34B used for acquiring the sound data to be reproduced may be used, or an acoustic sensor different from these

acoustic sensors

32, 34A, 34B may be used.

In some embodiments, the sound detected by the diagnostic acoustic sensor 35 is acquired as diagnostic sound data by the above-described sound data acquisition unit 40, and the frequency analysis unit 42 performs a frequency analysis on this diagnostic sound data. Is done. Then, the above-mentioned diagnosis unit 54 uses the frequency analysis result to make an abnormality determination of the sound generation source as described above.

In some embodiments, the diagnosis unit 54 determines the peak frequency corresponding to the sound source identified by the peak frequency identification unit 48 and the sound source information acquisition unit 50 in the frequency analysis result of the diagnostic sound data described above. When the amplitude of the component (that is, the frequency component of the natural frequency corresponding to the peak frequency) is equal to or larger than the threshold value, it is configured to determine that the sound source has an abnormality.

Hereinafter, diagnostic methods according to some embodiments will be described.
First, a method of identifying the natural frequency of the rolling mill 1 by the above-described diagnostic device 30 (see FIG. 2) in the diagnostic method according to the embodiment will be described.

FIG. 3 is a flowchart showing an outline of the diagnostic method according to the embodiment.
4 to 6 are

acoustic sensors

32, 34A, 34B under specific motor rotation speed conditions (n [rpm] in FIG. 4, 2n [rpm] in FIG. 5, 4n [rpm] in FIG. 6). It is a figure which shows an example of the frequency analysis result (frequency spectrum) of the sound data acquired using. In FIGS. 4 to 6, the

acoustic sensors

32, 34A, 34B are shown as a microphone A, a microphone B, and a microphone C, respectively. Further, on the vertical axis of the graphs of FIGS. 4 to 6, the amplitude (intensity) of the component at the frequency f [Hz] regarding the sound data acquired by each acoustic sensor (microphone) is represented by I (symbol indicating f, microphone). Is displayed. For example, the amplitude of the sound data acquired by the acoustic sensor 32 (microphone A) at the frequency f1 is I(f1, A). It should be noted that, when frequency analysis is performed on actual sound data, many peaks generally appear over a wide frequency band in addition to the frequency components shown in FIGS. 4 to 6, but in FIGS. In order to make it easy to understand, only the frequency component with a large peak value is shown.
FIG. 7 is a diagram showing an example of information in which a sound generation source stored in the storage unit 37 and a peak frequency (natural frequency) corresponding thereto are associated with each other.

In the following description, the plurality of rotation speed conditions are three conditions in which the rotation speed of the motor 2 is n, 2n, and 4n, and the three

sound sensors

32, 34A, and 34B (microphones A to C) are used as the plurality of sound sensors. However, the number of rotation speed conditions and the number of acoustic sensors are not limited to this, and any number of rotation speed conditions (two or more) and plural (two or more) acoustic sensors can be used to The diagnostic method according to the embodiment of the invention can be implemented.

As shown in FIG. 3, in the method for diagnosing the rolling mill 1 according to the embodiment, first, a plurality of

acoustic sensors

32, 34A, 34B (that is, microphones A to C) installed at different positions are used. The sound generated from the rolling mill 1 is detected (sound detection step; step S102).

Then, the sound data acquisition unit 40 (see FIG. 2) includes sounds detected by the

acoustic sensors

32, 34A, and 34B under each of a plurality of rotation speed conditions in which the rotation speed of the motor 2 (driving unit) is different. The sound data acquisition unit 40 of the processing unit 36 acquires a plurality of sound data (sound data acquisition step; step S104). Here, under each of three types of rotation speed conditions in which the rotation speed of the motor 2 is n, 2n (twice n), and 4n (four times n), the sound data is obtained using each acoustic sensor. get. That is, since sound data is acquired by the three

acoustic sensors

32, 34A, 34B for each of the three types of rotation speed conditions, nine kinds of sound data are acquired.

Next, the frequency analysis unit 42 performs a frequency analysis on each of the plurality of sound data acquired for each of the plurality of rotation speed conditions (three conditions of the motor rotation speed n, 2n, and 4n) (frequency analysis step; Step 106). The frequency spectrum obtained as a result of this frequency analysis is shown in FIG. 4 (when the rotation speed condition is n [rpm]), FIG. 5 (when the rotation speed condition is 2n [rpm]), and FIG. 6 (rotation speed condition: 4n [ rpm])).

Next, the peak frequency specifying unit 48 specifies one or more rotation speed conditions in which a peak frequency common to a plurality of sound data exists and the peak frequency based on the result of the above-described frequency analysis (peak frequency specifying step). ; Step S108). Specifically, as described below, the above-described rotation speed condition and the common peak frequency are specified.

In the frequency spectrum of the sound data relating to the acoustic sensors (microphones A to C) shown in FIG. 4, the

acoustic sensors

32, 34A, and 34B (microphones A to C) respectively have three frequencies fa, f1, and fb. A peak appears in the sound data. Therefore, under the rotational speed condition that the rotational speed of the motor 2 is n [rpm], the peak frequencies common to these sound data are specified as fa, f1 and fb.

In the frequency spectrum of the sound data relating to the acoustic sensors (microphones A to C) shown in FIG. 5, the

acoustic sensors

32, 34A, 34B (microphones A to C) are three frequencies fa, f2, and 2fb (twice fb). A peak appears in the sound data for each of C). Therefore, under the rotational speed condition that the rotational speed of the motor 2 is 2n [rpm], the peak frequencies common to these sound data are specified as fa, f2, and 2fb.

In the frequency spectrum of the sound data related to the acoustic sensors (microphones A to C) shown in FIG. 6, the

acoustic sensors

32, 34A, and 34B (microphones A to C) at three frequencies fa, f3, and 4fb (four times fb). A peak appears in the sound data for each of C). Therefore, under the rotation speed condition that the rotation speed of the motor 2 is 4 n [rpm], the peak frequencies common to these sound data are specified as fa, f3, and 4fb.

Next, the plurality of sounds acquired by the plurality of acoustic sensors (microphones A to C) by the first removing unit 44 under a plurality of rotation speed conditions (three motor rotation speed conditions of n, 2n, and 4n). From the results of frequency analysis of the data (9 kinds of sound data) (frequency spectra shown in FIGS. 4 to 6), components having a common frequency are removed regardless of the rotation speed condition (first removal step; step S110).

In the nine frequency spectra shown in FIGS. 4 to 6, the peak of the frequency fa exists in all of these frequency spectra. Therefore, the component of the frequency fa is a component that exists in common regardless of the rotation speed condition of the motor 2 (driving unit). That is, the first removing unit 44 removes the peak having the component of the frequency fa from each frequency spectrum. 4 to 6, the peak having the component of the frequency fa to be removed is indicated by a broken line.

Next, the plurality of sounds acquired by the plurality of acoustic sensors (microphones A to C) by the second removing unit 46 under a plurality of rotation speed conditions (three conditions of motor rotation speeds of n, 2n, and 4n). From the frequency analysis result of the data (9 kinds of sound data) (frequency spectrum shown in FIGS. 4 to 6), a component whose frequency is proportional to the rotation speed of the motor 2 (driving unit) is removed (second removing step; Step S112).

Here, the frequency spectrum shown in FIG. 4, that is, the three frequency spectra acquired under the condition that the rotation speed of the motor 2 is n [rpm], has a peak of the frequency fb in common. Further, the frequency spectrum shown in FIG. 5, that is, the three frequency spectra acquired under the condition that the rotation speed of the motor 2 is 2n [rpm], has a peak of frequency 2fb in common. Further, the frequency spectrum shown in FIG. 6, that is, the three frequency spectra acquired under the condition that the rotation speed of the motor 2 is 4 n [rpm], has a peak of frequency 4fb in common.
Therefore, the components of these frequencies fb, 2fb, 4fb are components whose frequency is proportional to the rotation speed of the motor 2. That is, the second removing unit 46 removes peaks having components of frequencies fb, 2fb, and 4fb from each frequency spectrum. 4 to 6, the peaks having the components of the frequencies fb, 2fb, 4fb to be removed are shown by broken lines.

Next, the sound source information acquisition unit 50 corresponds to the common peak frequency specified in the peak frequency specifying step (S112) among the plurality of acoustic sensors (microphones A to C) for each of the one or more rotation speed conditions. Based on the installation position of the one acoustic sensor having the largest amplitude of the sound data with respect to the rolling mill 1, information on the sound source of the rolling mill 1 having the natural frequency corresponding to the common peak frequency is obtained ( Sound source information acquisition step; step S114).

In the sound source information acquisition step (S114), the frequency component removed in the above-described first removal step (S110) may be excluded from the above-mentioned common peak frequency to obtain information regarding the sound source.
Further, in the sound source information acquisition step (S114), the frequency component removed in the second removal step (S112) is excluded from the common peak frequency described above to obtain information about the sound source. Good.

More specifically, in the three frequency spectra shown in FIG. 4 (the frequency spectrum of the sound data acquired by the three acoustic sensors (microphones A to C under the rotational speed condition in which the rotational speed of the motor 2 is n)) As described above, the peak frequencies fa, f1 and fb common to these sound data are specified. Further, in the first removing step S110, the peak having the component of the frequency fa is removed, and in the second removing step S112, the peak having the component of the frequency fb is removed.

Therefore, the sound source information acquisition unit 50 excludes the components of the frequency fa and the component of the frequency fb from the peak frequencies fa, f1, and fb identified in the peak frequency identification step for these three frequency spectra, that is, For the component of the peak frequency f1, one acoustic sensor (microphones A to C) having the largest amplitude of the sound data corresponding to this peak frequency is specified.

As shown in FIG. 4, the amplitude at the peak frequency f1 in the frequency spectrum corresponding to each acoustic sensor (microphones A to C) is I(f1,A), I(f1,B), I(f1, C), and these magnitude relationships are I(f1,C)≈I(f1,A)<I(f1,B). Therefore, the acoustic sensor that has acquired the sound data having the largest amplitude at the peak frequency f1 can be identified as the acoustic sensor 34A (microphone B).

The acoustic sensor 34A (microphone B) specified in this way is provided corresponding to the gear 4, and is closer to the gear 4 in the rolling mill 1 than other acoustic sensors (

acoustic sensors

32 and 34B). Since it is provided at the position, it is possible to obtain information that the source of the sound having the natural frequency (f1) corresponding to the peak frequency f1 is the gear 4 of the rolling mill 1.

Further, in the three frequency spectra shown in FIG. 5 (frequency spectra of sound data acquired by the three acoustic sensors (microphones A to C under the rotational speed condition in which the rotational speed of the motor 2 is 2n)), it has already been described. Thus, the peak frequencies fa, f2, and 2fb common to these sound data are specified. Further, in the first removing step S110, the peak having the component of the frequency fa is removed, and in the second removing step S112, the peak having the component of the frequency 2fb is removed.

Then, the sound source information acquisition unit 50 excludes the components of the frequency fa and the component of the frequency 2fb from the peak frequencies fa, f2, and 2fb identified in the peak frequency identification step for these three frequency spectra, that is, For the component of the peak frequency f2, the one acoustic sensor (microphones A to C) having the largest amplitude of the sound data corresponding to this peak frequency is specified.

As shown in FIG. 5, the amplitude at the peak frequency f2 in the frequency spectrum corresponding to each acoustic sensor (microphones A to C) is I(f2,A), I(f2,B), I(f2,). C), and these magnitude relationships are I(f2,A)<I(f2,B)<I(f2,C). Therefore, the acoustic sensor that has acquired the sound data having the largest amplitude at the peak frequency f2 can be identified as the acoustic sensor 34B (microphone C).

The acoustic sensor 34B (microphone C) thus identified is provided corresponding to the spindle 6, and is closer to the spindle 6 in the rolling mill 1 than the other acoustic sensors (

acoustic sensors

32 and 34A). Since it is provided at the position, it is possible to obtain information that the sound source of the natural frequency (f2) corresponding to the peak frequency f2 is the spindle 6 of the rolling mill 1.

In addition, in the three frequency spectra shown in FIG. 6 (the frequency spectrum of the sound data acquired by the three acoustic sensors (microphones A to C under the rotational speed condition in which the rotational speed of the motor 2 is 4n)), it has already been described. Thus, the peak frequencies fa, f3, and 4fb common to these sound data are specified. Further, in the first removing step S110, the peak having the component of the frequency fa is removed, and in the second removing step S112, the peak having the component of the frequency 4fb is removed.

Then, the sound source information acquisition unit 50 excludes the components of the frequency fa and the component of the frequency 4fb from the peak frequencies fa, f3, and 4fb identified in the peak frequency identification step for these three frequency spectra, that is, For the component of the peak frequency f3, one acoustic sensor (microphones A to C) having the largest amplitude of the sound data corresponding to this peak frequency is specified.

As shown in FIG. 6, the amplitude at the peak frequency f3 in the frequency spectrum corresponding to each acoustic sensor (microphones A to C) is I(f3,A), I(f3,B), I(f3, respectively). C), and these magnitude relationships are I(f3,C)<I(f3,B)<I(f3,A). Therefore, the acoustic sensor that has acquired the sound data having the largest amplitude at the peak frequency f3 can be identified as the acoustic sensor 32 (microphone A).

The acoustic sensor 32 (microphone A) specified in this way is provided corresponding to the motor 2, and is closer to the motor 2 in the rolling mill 1 than the other acoustic sensors (

acoustic sensors

34A and 34B). Since it is provided at the position, it is possible to obtain information that the source of the sound having the natural frequency (f3) corresponding to the peak frequency f3 is the motor 2 of the rolling mill 1.

Next, common peak frequencies (f1, f2, f3) corresponding to the rotation speed condition (motor rotation speed: n, 2n, 4n) of the motor 2 specified in the peak frequency specifying step (step S108), and sound source information Information relating to the sound source information acquired in the acquisition step (step S114) and the information associated with each other is stored in the storage unit 37 (step S116).

Here, FIG. 7 is a diagram illustrating an example of information stored in the storage unit 37.
As described above, the common peak frequency (f1) corresponding to the rotation speed condition of the motor 2 (motor rotation speed: n, 2n, 4n) is determined by the peak frequency specifying step (step S108) and the sound source information acquiring step (step S114). , F2, f3) and a sound source (gear 4, spindle 6, motor 2) having a natural frequency (f1, f2, f3) corresponding to the common peak frequency (f1, f2, f3). Relationships are identified. Therefore, as shown in FIG. 7, the storage unit 37 stores the above-mentioned peak frequency (that is, the natural frequency) and information (a part in the rolling mill 1) regarding the source of the sound of the natural frequency.

The information in which the peak frequency (natural frequency) and the sound generation source stored in the storage unit 37 are associated with each other can be used when diagnosing the rolling mill 1, as described later.

According to the diagnostic device and the diagnostic method described above, a plurality of

acoustic sensors

32, 34A, 34B provided at different positions are used to obtain a plurality of rotational speed conditions (motor rotational speeds: n, 2n, 4n). By performing frequency analysis on each of the plural sound data, the peak frequency at which the amplitude becomes maximum can be grasped for each sound data. Then, based on the frequency analysis result of each sound data, it is possible to specify the rotation frequency condition in which the peak frequency common to the plurality of sound data exists and to specify the peak frequency. Here, under a certain rotation speed condition, the common peak frequency existing in the plurality of sound data acquired by the plurality of

acoustic sensors

32, 34A, 34B may match the natural frequency of the rolling mill 1 (machine). Therefore, the natural frequency of the rolling mill 1 can be specified.
Further, under the rotation speed condition specified as described above, among the sound data acquired by the respective

acoustic sensors

32, 34A, 34B, the peak frequency specified as described above (that is, the natural frequency of the rolling mill 1 One acoustic sensor having the largest sound amplitude at a frequency (which may match) is closest to the sound source of the above-mentioned peak frequency in the rolling mill 1 among the plurality of

acoustic sensors

32, 34A, 34B. It can be presumed to have been installed at the location. Therefore, based on the installation position of the above-mentioned one acoustic sensor, the information regarding the sound generation source corresponding to the natural frequency of the rolling mill 1 (for example, the position or part of the sound generation source in the rolling mill 1) is acquired. can do.
Therefore, according to the diagnostic device and the diagnostic method described above, the natural frequency of the rolling mill 1 can be specified with a simple configuration including the plurality of

acoustic sensors

32, 34A, 34B, and the natural frequency of the rolling mill can be specified. It is possible to acquire information indicating a position or a part having a frequency.

Next, a method of diagnosing the rolling mill 1 based on the natural frequency identified as described above will be described. Here, a case where the diagnosis target is the gear 4 will be described.

FIG. 8 is a flowchart showing an outline of the diagnostic method according to the embodiment.
As shown in FIG. 8, in the diagnostic method according to the embodiment, first, the sound from the rolling mill 1 is output by the diagnostic acoustic sensor 35 while the motor 2 (drive unit) is operating at the specified value N. The sound data acquisition unit 40 detects and acquires diagnostic sound data including the sound (step S202). Next, the frequency analysis unit 42 performs frequency analysis on the acquired diagnostic sound data to acquire a frequency spectrum (step S204).

Then, based on the amplitude of the peak frequency (natural frequency) component of the diagnosis target part (eg, gear 4) in the frequency analysis result (frequency spectrum) of the diagnostic sound data, the sound generation source (that is, the diagnosis target part). The abnormality determination of) is performed (steps S206 to S208). Here, the peak frequency (natural frequency) of the diagnosis target part is specified in the peak frequency specifying step (step S108) and the sound source information acquiring step (step S114).

More specifically, in step S206, the amplitude at the above-mentioned peak frequency (natural frequency) f1 of the gear 4 to be diagnosed in the frequency analysis result of the diagnostic sound data is acquired. The information on the peak frequency (natural frequency) corresponding to the diagnosis target part (gear 4) may be acquired from the storage unit 37. Then, the amplitude I thus obtained is compared with the threshold value Ith. This threshold value is set to an appropriate value corresponding to the specified value N of the rotation speed of the motor 2.

If the amplitude I is less than the threshold value Ith (NO in step S206), it is not determined that the gear 4 is abnormal, and the flow of the diagnostic method is ended. On the other hand, when the amplitude I is equal to or larger than the threshold value Ith (YES in step S206), it is determined that the gear 4 has an abnormality or a sign of abnormality. In this case, the fact that the abnormality of the gear 4 or the sign of the abnormality is detected may be displayed on the display unit 38, or an alarm may be sounded by a speaker or the like.

The threshold value Ith of the amplitude I may be set to a different value depending on the rotation speed of the motor 2 (driving unit). Since the amplitude of the sound at the peak frequency component varies depending on the rotation speed of the motor 2, by setting an appropriate threshold value according to the rotation speed of the motor 2 (driving unit) in this way, the abnormality of the diagnosis target site is detected. The determination can be made more appropriately.

The following is a summary of diagnostic devices and diagnostic methods according to some embodiments.

In the configuration of the above (1), by performing frequency analysis on each of the multiple sound data acquired under each of the multiple rotation speed conditions by using the multiple acoustic sensors provided at different positions, the amplitude of each sound data is increased. It is possible to grasp the peak frequency at which is the maximum. Then, based on the frequency analysis result of each sound data, it is possible to specify the rotation frequency condition in which the peak frequency common to the plurality of sound data exists and to specify the peak frequency. Here, under a certain rotation speed condition, a common peak frequency existing in a plurality of sound data acquired by a plurality of acoustic sensors may match the natural frequency of the mechanical device. The number can be specified.
Further, under the rotation speed condition specified as described above, among the sound data acquired by each acoustic sensor, the peak frequency specified as described above (that is, the natural frequency of the mechanical device may match. It is assumed that the one acoustic sensor with the largest sound amplitude at the frequency) is the one that is installed closest to the sound source of the above-mentioned peak frequency in the mechanical device among the plurality of acoustic sensors. You can Therefore, based on the installation position of the above-mentioned one acoustic sensor, to obtain information about the sound source corresponding to the natural frequency of the mechanical device (for example, the position or part of the sound source in the mechanical device). You can
Therefore, according to the configuration of (1) above, the natural frequency of the mechanical device can be specified with a simple configuration including a plurality of acoustic sensors, and the position or site having the natural frequency in the mechanical device can be specified. It is possible to obtain information indicating.

(2) In some embodiments, in the configuration of (1) above,
The diagnostic device is
The frequency analysis result of the plurality of sound data further comprises a first removing unit configured to remove a component having a common frequency regardless of the rotation speed condition,
The sound source information acquisition unit is configured to exclude the frequency component removed by the first removal unit from the peak frequency and obtain information about the sound source.

In a plurality of sound data acquired under different rotation speed conditions, when the peak of the amplitude appears at the same frequency, the frequency is derived from the background sound that occurs regardless of the rotation speed of the drive unit. Can be estimated. In this respect, according to the configuration of the above (2), in the frequency analysis result of a plurality of sound data acquired by a plurality of acoustic sensors under a plurality of rotation speed conditions, a component having a common frequency is removed regardless of the rotation speed condition. However, when obtaining the information about the sound source, the above-mentioned frequency component related to the background sound is excluded from the common peak frequency in the frequency analysis results of the plurality of sound data. Therefore, the natural frequency of the mechanical device can be accurately specified based on the frequency analysis results of the plurality of sound data.

(3) In some embodiments, in the configuration of (1) or (2) above,
The diagnostic device is
A second removing unit configured to remove a component whose frequency is proportional to the rotation speed of the driving unit, from the frequency analysis result of the plurality of sound data;
The sound source information acquisition unit is configured to exclude the frequency component removed by the second removal unit from the peak frequency and obtain information about the sound source.

In a plurality of sound data acquired under different rotation speed conditions, when a peak of a frequency component proportional to the rotation speed of the drive unit appears, the frequency is a sound due to the magnitude of motion of the drive unit, It can be estimated that it is not related to the natural frequency of the mechanical device. In this respect, according to the configuration of (3), the component whose frequency is proportional to the rotation speed of the drive unit is removed from the frequency analysis result of the plurality of sound data acquired by the plurality of acoustic sensors under the plurality of rotation speed conditions. However, when obtaining the information about the sound source, the above-mentioned frequency components not related to the natural frequency of the mechanical device are excluded from the common peak frequency in the frequency analysis result of the plurality of sound data. Therefore, the natural frequency of the mechanical device can be accurately specified based on the frequency analysis results of the plurality of sound data.

(4) In some embodiments, in any of the configurations of (1) to (3) above,
The diagnostic device is
A memory for storing information in which the peak frequency corresponding to the rotation speed condition specified by the peak frequency specifying unit and the information regarding the sound source specified by the sound source information acquiring unit are associated with each other. Further comprises a section.

According to the configuration of (4), the peak frequency corresponding to the rotation speed condition specified as described above (that is, the frequency that may be the natural frequency of the mechanical device) and the peak frequency specified as described above. Since the information relating to the source of the generated sound can be stored, the mechanical device can be diagnosed based on the stored information, and the mechanical device can be easily diagnosed.

(5) In some embodiments, in any of the configurations of (1) to (4) above,
The diagnostic device is
During operation of the rotation speed of the drive unit at a specified value, a frequency analysis unit configured to perform frequency analysis of diagnostic sound data acquired based on the sound detected by the diagnostic acoustic sensor,
And a diagnostic unit configured to make an abnormality determination of the sound source based on the amplitude of the component of the peak frequency in the frequency analysis result of the diagnostic sound data by the frequency analysis unit.

The amplitude of the sound of the natural frequency of the mechanical device changes according to the rotation speed of the drive unit. In this respect, according to the configuration of the above (5), since the diagnostic sound data is acquired when the drive unit is driven at a specific rotation speed (specified value), the frequency of the diagnostic sound data is increased. Based on the amplitude of the component of the above-mentioned peak frequency (the peak frequency specified by the configuration of (1) above) in the analysis result, it is possible to appropriately determine the abnormality of the sound source of the peak frequency.

(6) In some embodiments, in the configuration of (5) above,
The diagnosis unit is configured to determine that an abnormality has occurred in the sound generation source when the amplitude of the peak frequency component is equal to or larger than a threshold in the frequency analysis result of the diagnosis sound data.

According to the configuration of (6) above, by comparing the amplitude of the component of the peak frequency in the frequency analysis result of the diagnostic sound data with the threshold value, it is possible to appropriately determine the abnormality of the sound source of the peak frequency. Can be done. The above-mentioned threshold value is based on, for example, the amplitude obtained in advance by setting the rotation speed of the drive unit to a specified value for the peak frequency (natural frequency of the mechanical device) specified by the configuration of (1) above. Can be decided.

(7) In some embodiments, in any of the configurations (1) to (6) above,
The mechanical device is
A motor as the drive unit,
A gear or a spindle as the driven portion that is rotationally driven by the motor,
A rolling roll driven by the gear or the spindle,
It is a rolling mill including.

According to the above configuration (7), the natural frequency of the rolling mill including the motor as the driving unit, the gear or the spindle as the driven unit, and the rolling roll driven by the gear or the spindle is specified. In addition, it is possible to obtain information indicating the position or site having the natural frequency in the rolling device.

(8) In some embodiments, in the configuration of (7) above,
The plurality of acoustic sensors include a first acoustic sensor provided corresponding to the motor and a second acoustic sensor provided corresponding to the gear or the spindle.

According to the above configuration (8), the sound data is acquired using the first acoustic sensor provided corresponding to the motor and the second acoustic sensor provided corresponding to the gear or the spindle. The natural frequency can be specified for the motor and the gear or spindle in the rolling mill.

(9) The equipment according to at least one embodiment of the present invention,
A mechanical device including a drive unit and a driven unit that is rotationally driven by the drive unit;
The diagnostic device according to any one of (1) to (8), which is configured to diagnose the mechanical device,
Equipped with.

In the configuration of the above (9), by performing frequency analysis on each of the multiple sound data acquired under each of the multiple rotation speed conditions by the multiple acoustic sensors provided at different positions, the amplitude of each sound data is increased. It is possible to grasp the peak frequency at which is the maximum. Then, based on the frequency analysis result of each sound data, it is possible to specify the rotation frequency condition in which the peak frequency common to the plurality of sound data exists and to specify the peak frequency. Here, under a certain rotation speed condition, a common peak frequency existing in a plurality of sound data acquired by a plurality of acoustic sensors may match the natural frequency of the mechanical device. The number can be specified.
Further, under the rotation speed condition specified as described above, among the sound data acquired by each acoustic sensor, the peak frequency specified as described above (that is, the natural frequency of the mechanical device may match. It is assumed that the one acoustic sensor with the largest sound amplitude at the frequency) is the one that is installed closest to the sound source of the above-mentioned peak frequency in the mechanical device among the plurality of acoustic sensors. You can Therefore, based on the installation position of the above-mentioned one acoustic sensor, to obtain information about the sound source corresponding to the natural frequency of the mechanical device (for example, the position or part of the sound source in the mechanical device). You can
Therefore, according to the above configuration (9), the natural frequency of the mechanical device can be specified with a simple structure including a plurality of acoustic sensors, and the position or site having the natural frequency in the mechanical device can be specified. It is possible to obtain information indicating.

(10) The diagnostic method according to at least one embodiment of the present invention comprises:
A diagnostic method for diagnosing a mechanical device including a drive unit and a driven unit that is rotationally driven by the drive unit,
A sound detection step of detecting a sound generated from the mechanical device by a plurality of acoustic sensors respectively installed at different positions,
A sound data acquisition step of acquiring a plurality of sound data respectively including sounds detected by each of the plurality of acoustic sensors in each of a plurality of rotation speed conditions in which the rotation speed of the drive unit is different,
A step of frequency-analyzing each of the plurality of sound data acquired for each of the plurality of rotation speed conditions;
A peak frequency specifying step of specifying one or more rotation speed conditions in which a peak frequency common to the plurality of sound data exists and the peak frequency based on a result of the frequency analysis;
For each of the one or more rotation speed conditions, the machine based on the installation position of the one acoustic sensor having the largest amplitude of the sound data corresponding to the peak frequency among the plurality of acoustic sensors with respect to the mechanical device. Of the device, a sound source information acquisition step of obtaining information about a source of sound of a natural frequency corresponding to the peak frequency,
Equipped with.

In the above method (10), by performing frequency analysis on each of the multiple sound data acquired under each of the multiple rotation speed conditions by using the multiple acoustic sensors provided at different positions, the amplitude of each sound data is increased. It is possible to grasp the peak frequency at which is the maximum. Then, based on the frequency analysis result of each sound data, it is possible to specify the rotation frequency condition in which the peak frequency common to the plurality of sound data exists and to specify the peak frequency. Here, under a certain rotation speed condition, a common peak frequency existing in a plurality of sound data acquired by a plurality of acoustic sensors may match the natural frequency of the mechanical device. The number can be specified.
Further, under the rotation speed condition specified as described above, among the sound data acquired by each acoustic sensor, the peak frequency specified as described above (that is, the natural frequency of the mechanical device may match. It is assumed that the one acoustic sensor with the largest sound amplitude at the frequency) is the one that is installed closest to the sound source of the above-mentioned peak frequency in the mechanical device among the plurality of acoustic sensors. You can Therefore, based on the installation position of the above-mentioned one acoustic sensor, to obtain information about the sound source corresponding to the natural frequency of the mechanical device (for example, the position or part of the sound source in the mechanical device). You can
Therefore, according to the above method (10), the natural frequency of the mechanical device can be specified with a simple configuration including a plurality of acoustic sensors, and the position or site having the natural frequency in the mechanical device can be specified. It is possible to obtain information indicating.

(11) In some embodiments, the method of (10) above comprises
The result of the frequency analysis of the plurality of sound data further comprises a first removing step of removing a component having a common frequency regardless of the rotation speed condition,
In the sound source information acquisition step, the frequency component removed in the first removal step is excluded from the peak frequency to obtain information about the sound source.

According to the method of (11), the frequency analysis result of the plurality of sound data acquired by the plurality of acoustic sensors under the plurality of rotation speed conditions removes the component having the common frequency regardless of the rotation speed condition, When obtaining the information about the source of the above, the above-mentioned frequency component related to the background sound is excluded from the common peak frequency in the frequency analysis result of the plurality of sound data. Therefore, the natural frequency of the mechanical device can be accurately specified based on the frequency analysis results of the plurality of sound data.

(12) In some embodiments, the method of (10) or (11) above comprises
Further comprising a second removal step of removing a component whose frequency is proportional to the rotation speed of the drive unit, from the result of the frequency analysis of the plurality of sound data,
In the sound source information acquisition step, the frequency component removed in the second removal step is excluded from the peak frequency to obtain information about the sound source.

According to the above method (12), the frequency analysis result of the plurality of sound data acquired by the plurality of acoustic sensors under the plurality of rotation speed conditions is removed by removing the component whose frequency is proportional to the rotation speed of the drive unit. When obtaining the information about the source of the above, the above-mentioned frequency components not related to the natural frequency of the mechanical device are excluded from the common peak frequency in the frequency analysis result of the plurality of sound data. Therefore, the natural frequency of the mechanical device can be accurately specified based on the frequency analysis results of the plurality of sound data.

(13) In some embodiments, any one of the above methods (10) to (12)
The storage section stores information in which the peak frequency corresponding to the rotation speed condition specified in the peak frequency specifying step and the information regarding the sound source acquired in the sound source information acquiring step are associated with each other. The method further includes steps.

According to the above method (13), the peak frequency (that is, the frequency that may be the natural frequency of the mechanical device) corresponding to the rotation speed condition specified as described above and the peak frequency specified as described above are specified. Since the information relating to the source of the generated sound can be stored, the mechanical device can be diagnosed based on the stored information, and the mechanical device can be easily diagnosed.

(14) In some embodiments, the method according to any one of (10) to (13) above,
A step of detecting a sound from the mechanical device with a diagnostic acoustic sensor during operation of the rotation speed of the drive unit at a specified value, and obtaining diagnostic sound data including the sound;
Frequency-analyzing the diagnostic sound data,
The method further includes a diagnostic step of performing abnormality determination of the sound generation source based on the amplitude of the peak frequency component in the frequency analysis result of the diagnostic sound data.

According to the above method (14), the diagnostic sound data is acquired when the drive unit is driven at a specific rotation speed (specified value). Therefore, in the frequency analysis result of the diagnostic sound data, Based on the amplitude of the component of the above-mentioned peak frequency (the peak frequency specified by the method (10) above), it is possible to appropriately determine the abnormality of the sound source of the peak frequency. Since the rotation speed of the drive unit may be set to a different value depending on the rolling condition, the amplitude of the specified peak frequency component also changes according to each rotation speed.

(15) In some embodiments, in the method of (14) above,
In the diagnosis step, when the amplitude of the peak frequency component is equal to or larger than a threshold in the frequency analysis result of the diagnostic sound data, it is determined that an abnormality has occurred in the sound generation source.

According to the method (15), the abnormality determination of the sound source of the peak frequency sound is appropriately performed by comparing the amplitude of the peak frequency component in the frequency analysis result of the diagnostic sound data with the threshold value. Can be done. The above-mentioned threshold value is based on, for example, the amplitude obtained in advance by setting the rotation speed of the drive unit to a specified value for the peak frequency (natural frequency of the mechanical device) specified by the method (9) described above. Can be decided.

(16) In some embodiments, in any one of the methods (10) to (15) above,
The mechanical device is
A motor as the drive unit,
A gear or a spindle as the driven portion that is rotationally driven by the motor,
A rolling roll driven by the gear or the spindle,
It is a rolling mill including.

According to the above method (16), the natural frequency of the rolling mill including the motor as the driving unit, the gear or the spindle as the driven unit, and the rolling roll driven by the gear or the spindle is specified. In addition, it is possible to obtain information indicating the position or site having the natural frequency in the rolling device.

Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and includes modified forms of the above-described embodiments and combinations of these forms as appropriate.

In the present specification, expressions representing relative or absolute arrangements such as "in a certain direction", "along a certain direction", "parallel", "orthogonal", "center", "concentric", or "coaxial". Not only strictly represents such an arrangement, but also represents a relative displacement or a state in which the components are relatively displaced with an angle or distance such that the same function can be obtained.
For example, expressions such as "identical", "equal", and "homogeneous" that indicate that they are in the same state are not limited to a state in which they are exactly equal. It also represents the existing state.
In addition, in the present specification, expressions representing shapes such as a quadrangle and a cylinder are not limited to shapes such as a quadrangle and a cylinder in a geometrically strict sense, and are within a range in which the same effect can be obtained. A shape including an uneven portion and a chamfered portion is also shown.
Further, in this specification, the expressions “comprising”, “including”, or “having” one element are not exclusive expressions excluding the existence of other elements.

DESCRIPTION OF SYMBOLS 1 Rolling apparatus 2 Motor 3 Power transmission part 4 Gear 6 Spindle 8 Rolling roll 9 Metal strips 12A, 12B Work rolls 14A, 14B Intermediate rolls 16A, 16B Backup roll 30 Diagnostic device 32 1st

acoustic sensor

34A, 34B 2nd acoustic sensor 35 diagnostic acoustic sensor 36 processing unit 37 storage unit 38 display unit 40 sound data acquisition unit 42 frequency analysis unit 44 first removal unit 46 second removal unit 48 peak frequency identification unit 50 sound source information acquisition unit 54 diagnostic unit

Claims

A diagnostic device for diagnosing a mechanical device including a drive part and a driven part rotationally driven by the drive part,
A plurality of acoustic sensors provided at different positions, respectively for detecting sounds generated from the mechanical device,
A sound data acquisition unit configured to acquire a plurality of sound data respectively including sounds detected by each of the plurality of acoustic sensors in each of a plurality of rotation speed conditions in which the rotation speed of the drive unit is different,
A frequency analysis unit configured to perform frequency analysis on each of the plurality of sound data acquired for each of the plurality of rotation speed conditions,
A peak frequency specifying unit that specifies the peak frequency and at least one rotation speed condition in which a peak frequency common to the plurality of sound data exists, based on a frequency analysis result by the frequency analyzing unit;
For each of the one or more rotation speed conditions, the machine based on the installation position of the one acoustic sensor having the largest amplitude of the sound data corresponding to the peak frequency among the plurality of acoustic sensors with respect to the mechanical device. Of the device, a sound source information acquisition unit that obtains information about a source of sound having a natural frequency corresponding to the peak frequency,
A diagnostic device including.
The frequency analysis result of the plurality of sound data further comprises a first removing unit configured to remove a component having a common frequency regardless of the rotation speed condition,
The diagnostic device according to claim 1, wherein the sound source information acquisition unit is configured to exclude the frequency component removed by the first removal unit from the peak frequency and obtain information about the sound source.
A second removing unit configured to remove a component whose frequency is proportional to the rotation speed of the driving unit, from the frequency analysis result of the plurality of sound data;
The diagnosis according to claim 1 or 2, wherein the sound source information acquisition unit is configured to exclude the frequency component removed by the second removal unit from the peak frequency and obtain information about the sound source. apparatus.
A memory for storing information in which the peak frequency corresponding to the rotation speed condition specified by the peak frequency specifying unit and the information regarding the sound source specified by the sound source information acquiring unit are associated with each other. The diagnostic device according to claim 1, further comprising a unit.
During operation of the rotation speed of the drive unit at a specified value, a frequency analysis unit configured to perform frequency analysis of diagnostic sound data acquired based on the sound detected by the diagnostic acoustic sensor,
A diagnostic unit configured to determine an abnormality in the sound source based on the amplitude of the component of the peak frequency in the frequency analysis result of the diagnostic sound data by the frequency analysis unit. Item 5. The diagnostic device according to any one of items 1 to 4.
In the frequency analysis result of the diagnostic sound data, the diagnosis unit is configured to determine that an abnormality has occurred in the sound source when the amplitude of the peak frequency component is equal to or greater than a threshold value. Item 6. The diagnostic device according to item 5.
The mechanical device is
A motor as the drive unit,
A gear or a spindle as the driven portion that is rotationally driven by the motor,
A rolling roll driven by the gear or the spindle,
The diagnostic device according to any one of claims 1 to 6, which is a rolling mill including a.
The diagnostic device according to claim 7, wherein the plurality of acoustic sensors include a first acoustic sensor provided corresponding to the motor and a second acoustic sensor provided corresponding to the gear or the spindle.
A mechanical device including a drive unit and a driven unit that is rotationally driven by the drive unit;
The diagnostic device according to claim 1, wherein the diagnostic device is configured to diagnose the mechanical device.
Equipment equipped with.
A diagnostic method for diagnosing a mechanical device including a drive unit and a driven unit that is rotationally driven by the drive unit,
A sound detection step of detecting a sound generated from the mechanical device by a plurality of acoustic sensors respectively installed at different positions,
A sound data acquisition step of acquiring a plurality of sound data respectively including sounds detected by each of the plurality of acoustic sensors in each of a plurality of rotation speed conditions in which the rotation speed of the drive unit is different,
A step of frequency-analyzing each of the plurality of sound data acquired for each of the plurality of rotation speed conditions;
A peak frequency specifying step of specifying one or more rotation speed conditions in which a peak frequency common to the plurality of sound data exists and the peak frequency based on a result of the frequency analysis;
For each of the one or more rotation speed conditions, the machine based on the installation position of the one acoustic sensor having the largest amplitude of the sound data corresponding to the peak frequency among the plurality of acoustic sensors with respect to the mechanical device. Of the device, a sound source information acquisition step of obtaining information about a source of sound of a natural frequency corresponding to the peak frequency,
A diagnostic method comprising.
The result of the frequency analysis of the plurality of sound data further comprises a first removing step of removing a component having a common frequency regardless of the rotation speed condition,
The diagnostic method according to claim 10, wherein in the sound source information acquisition step, the frequency component removed in the first removal step is excluded from the peak frequency to obtain information about the sound source.
Further comprising a second removal step of removing a component whose frequency is proportional to the rotation speed of the drive unit, from the result of the frequency analysis of the plurality of sound data,
The diagnostic method according to claim 10 or 11, wherein in the sound source information acquisition step, the frequency component removed in the second removal step is excluded from the peak frequency to obtain information about the sound source.
The storage section stores information in which the peak frequency corresponding to the rotation speed condition specified in the peak frequency specifying step and the information regarding the sound source acquired in the sound source information acquiring step are associated with each other. The diagnostic method according to claim 10, further comprising a step.
A step of detecting a sound from the mechanical device with a diagnostic acoustic sensor during operation of the rotation speed of the drive unit at a specified value, and obtaining diagnostic sound data including the sound;
Frequency-analyzing the diagnostic sound data,
14. The diagnostic step according to claim 10, further comprising a diagnostic step of performing an abnormality determination of a sound generation source based on an amplitude of a component of the peak frequency in a frequency analysis result of the diagnostic sound data. Diagnostic method.
The said diagnostic step WHEREIN: When the amplitude of the component of the said peak frequency is more than a threshold value in the said frequency analysis result about the said diagnostic sound data, it determines with the abnormality having generate|occur|produced the said sound source. Diagnostic method.