JP3449194B2 - Method and apparatus for diagnosing abnormalities in rotating equipment - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for diagnosing abnormalities in rotating equipment.

[0002]

2. Description of the Related Art Generally, there are provided various types of rotating devices which are rotationally driven by using an electric motor or an internal combustion engine as a drive source of rotational force. It is known that when some kind of abnormality occurs in this type of rotating equipment, vibrations and sounds that are different from usual are generated, and if the drive is continued in such a state, the rotating equipment may be damaged in some cases. .

Therefore, conventionally, waveform data such as vibrations and sounds generated by a rotating device has been subjected to fast Fourier transform (FF).
T) the spectral pattern obtained by applying
It has been considered to compare a preset spectrum pattern at the time of abnormality and to diagnose the presence or absence of abnormality of the rotating device based on the comparison result (Japanese Patent Laid-Open No. 62-93620).
Issue).

[0004]

By the way, since the technique described in the above publication determines anomalies by using only spectral patterns, when the anomalies of different kinds of anomalies are the same (for example, the spectra are different from each other). It is not possible to discriminate the type of anomaly for which the patterns are the same but the occurrence time intervals are different.

The present invention has been made in view of the above circumstances, and an object thereof is a method for diagnosing an abnormality in a rotating machine, which can determine even the type of abnormality by extracting more information from waveform data. And to provide the device.

[0006]

According to a first aspect of the present invention, waveform data such as vibration or sound generated during rotation of a rotating device is detected, a time change of the spectrum of the waveform data is obtained, and a time change of the spectrum is obtained. From the frequency characteristic amount-time characteristic amount obtained from the waveform data during rotation of the rotating device, the frequency generating a peak from the frequency characteristic amount is obtained as the frequency characteristic amount, and the time interval at which the peak occurs for each frequency characteristic amount is obtained as the time characteristic amount. By comparing the set with the reference data in which the set of the frequency characteristic amount and the time characteristic amount when the rotating device is abnormal is registered in advance for each cause of abnormality, the presence or absence and the cause of the abnormality of the rotating device are specified.

According to a second aspect of the present invention, in the first aspect of the invention, the time change of the spectrum of the waveform data is obtained by subjecting the waveform data to a short-time Fourier transform. According to a third aspect of the invention, in the first aspect of the invention, the time change of the spectrum of the waveform data is obtained by performing the wavelet transform on the waveform data. The invention of claim 4 is the same as the invention of claims 1 to 3,
After obtaining the frequency feature quantity, the autocorrelation calculation is performed on the time-series data of each frequency feature quantity, and the time interval at which the autocorrelation value peaks is used as the time feature quantity.

According to a fifth aspect of the present invention, in the first to third aspects of the present invention, after obtaining the frequency feature quantity, fast Fourier transform is performed on the time series data of each frequency feature quantity to obtain the maximum value. The reciprocal of the generated frequency is used as the time feature quantity. According to a sixth aspect of the present invention, in the inventions of the first to third aspects, after obtaining the frequency feature amount, a cepstrum operation is performed in which the fast Fourier transform is performed twice for the time series data of each frequency feature amount, and the maximum The kefency value that gives a value is used as the time feature amount.

According to a seventh aspect of the present invention, in the first aspect of the present invention, time characteristic amounts are obtained for a plurality of frequency characteristic amounts, and each set of frequency characteristic amount-time characteristic amount is collated with reference data. According to an eighth aspect of the present invention, in the first aspect of the invention, the rotation speed of the rotating device is detected, and the time feature amount is corrected according to the rotation speed.

According to a ninth aspect of the present invention, waveform data such as vibration or sound generated when the rotating device is rotated is detected, the time change of the spectrum of this waveform data is obtained, and the frequency at which a peak is generated from the time change of the spectrum is determined. In addition to the frequency characteristic amount, the time interval at which a peak is generated for each frequency characteristic amount is obtained as the time characteristic amount, and the set of the frequency characteristic amount-time characteristic amount obtained from the waveform data of the rotating device during rotation is set to The presence / absence of an abnormality in the rotating device and the cause of the abnormality are specified by applying to the neural network that has been learned in advance so as to associate the set of the frequency feature amount-time feature amount at the time of abnormality and each cause of abnormality.

According to a tenth aspect of the present invention, in the first or ninth aspect of the present invention, a reference value weighted for each frequency feature amount is set, and the time series data for each frequency feature amount is the reference value. The time interval at the time point exceeding is used as the time feature amount for each frequency feature amount. According to the invention of claim 11, in the invention of claim 1, power is transmitted from the drive source to the rotating device via a belt, and the drive source is vibration-proof.

According to a twelfth aspect of the present invention, in the first aspect of the present invention, power is transmitted from the drive source to the rotating device via the belt, and the tension of the belt is kept constant while the rotating device is being driven. It is a thing. The invention of claim 13 is
According to the twelfth aspect of the present invention, the elongation amount of the belt is obtained by obtaining the axial distance between the drive source and the rotating device when the tension of the belt is a constant value.

According to a fourteenth aspect of the present invention, according to the first aspect of the present invention, power is transmitted from a drive source to a rotating device through a belt, and a waveform component is obtained after removing a frequency component of a vibration frequency of the belt from the waveform data. This is to obtain the time change of the spectrum of the data. According to a fifteenth aspect of the invention, in the fourteenth aspect of the invention, the vibration frequency of the belt is calculated based on the axial distance between the drive source and the rotating device and the belt tension.

According to a sixteenth aspect of the present invention, in the invention of the first aspect, the power is transmitted from the drive source to the rotating device through the belt, and the belt is provided in a plane orthogonal to the axes of the drive source and the rotating device. It is to be positioned. According to a seventeenth aspect of the present invention, a vibration sensor that detects waveform data such as vibration generated when a rotating device rotates, a time change of the spectrum of the waveform data, and a frequency that causes a peak from the time change of the spectrum are frequency characteristics. A feature amount extraction unit that obtains a time interval that produces a peak for each frequency feature amount as a time feature amount, and a set of frequency feature amount-time feature amount obtained from waveform data when the rotating device is rotating, Equipped with an abnormality diagnosis unit that identifies the presence / absence of an abnormality and the cause of an abnormality in a rotating device by matching the set of frequency feature-time feature when an abnormality occurs in the rotating device with reference data registered in advance for each cause of abnormality. Is.

According to an eighteenth aspect of the invention, in the seventeenth aspect of the invention, a belt for transmitting power from the drive source to the rotating device and a vibration for supporting the drive source so as not to transmit the vibration of the drive source to the rotating device. And a cushioning material. According to a nineteenth aspect of the invention, in the eighteenth aspect, a pair of sensors for detecting vibrations on both sides of the vibration damping material are provided,
It is provided with means for obtaining the transfer function of the vibration damping material based on the outputs of both sensors and detecting the deterioration of the vibration damping material by the change over time of the transfer function.

According to a twentieth aspect of the present invention, in the eighteenth aspect of the present invention, the drive source is movable with respect to the rotating device, and a linear guide for restricting the movement range of the drive source within a plane including the belt is provided. is there. According to a twenty-first aspect of the present invention, in the eighteenth aspect of the present invention, the drive source is movable with respect to the rotating device, and the air bearing is provided to limit the movement range of the drive source within a plane including the belt.

According to a twenty-second aspect of the present invention, in the eighteenth aspect of the present invention, there is provided a support means for movably supporting the drive source with respect to the rotating device, and a moving means for moving the drive source. According to a twenty-third aspect of the present invention, in the eighteenth to twenty-second aspects of the present invention, a distance sensor for detecting a distance between the drive source and the rotating device is provided.

According to a twenty-fourth aspect of the invention, in the seventeenth aspect of the invention, a vibrating means for vibrating the rotating device when the drive source is not operating is provided, and the abnormality diagnosis section vibrates the rotating device by the vibrating means. Sometimes, the presence or absence of abnormality in the mechanical system including the rotating device and the vibration sensor is determined based on the waveform data detected by the vibration sensor. The invention of claim 25 is claim 2
In the invention of 4, the vibration means is a voice coil motor.

In a twenty-sixth aspect of the present invention based on the twenty-fourth aspect, the vibrating means is an actuator using a piezoelectric effect. In the invention of claim 27, in the invention of claim 25 or claim 26, the vibrating means also serves as a vibration sensor. According to a twenty-eighth aspect of the present invention, in the twenty-fourth aspect of the present invention, the vibrating means is a hammer that applies an impact force to the rotating device.

According to a twenty-ninth aspect of the invention, in the abnormality diagnosing device for a rotating device according to any one of the twenty-fourth to twenty-eighth aspects, the rotating device is vibrated by the vibrating means and the output of the vibration sensor is output before the drive source is operated. Based on this, the abnormality diagnosing unit obtains an input / output gain and compares the gain with a preset normal range to check the presence or absence of abnormality in the mechanical system including the rotating device and the vibration sensor.

According to a thirtieth aspect of the invention, in the abnormality diagnosing device for a rotating device according to the twenty-fourth to twenty-eighth aspects, the rotating device is vibrated by the vibrating means and the output of the vibration sensor is output before the drive source is operated. Based on the abnormality diagnosis section, the input / output gains at multiple frequencies are obtained, and the correlation value between the obtained gain and the preset inspection data is obtained.
By comparing the obtained correlation value with a preset threshold value, the presence or absence of abnormality in the mechanical system including the rotating device and the vibration sensor is inspected.

[0022]

DETAILED DESCRIPTION OF THE INVENTION

(Embodiment 1) In the present embodiment, as shown in FIG. 2, the rotating device which is the target of abnormality diagnosis is an electromagnetic clutch 2 which is rotationally driven by a motor 1 as a drive source via a belt 21. Has become.

A more detailed description will be given. As shown in FIG. 3, a motor plate 13 is slidable in a straight line on a base 10 fixed to a fixed position such as a floor via a linear guide 12 attached to a motor base 11 fixed on the base 10. Are arranged so that The holder 1 mounted on the motor plate 13 via the vibration damping material 14.
A motor 1 as a drive source is fixed to 5. The sliding direction of the motor plate 13 is restricted in the horizontal direction within the plane orthogonal to the output shaft of the motor 1. In addition, the motor base 11 and the holding base 15 have rising pieces 11a and 15a.
Is provided, and the vibration damping material 14a is also interposed between both rising pieces 11a and 15a.

The rotating force of the motor 1 is transmitted to the electromagnetic clutch 2 which is a rotating device via a belt 21 wound around an output shaft. As shown in FIG. 4, the electromagnetic clutch 2 includes a clutch pulley 2a around which a belt 21 is wound, and first and second bosses 2b that rotate together with a rotary shaft 23 whose both ends are supported by bearings 22a and 22b. , 2c and a coil 25 fixed to the bearing 22a. The first boss 2b is fixed to the rotary shaft 23, while the second boss 2c is connected to the first boss 2b by a spline to rotate the rotary shaft 23.
It is movable in the axial direction. Clutch pulley 2a
Is rotatable with respect to the rotary shaft 23 via a bearing 23c. Then, when an exciting current is passed through the coil 25,
The second boss 2c is magnetically attracted to the clutch pulley 2a, and the rotational force can be transmitted from the clutch pulley 2a to the boss 2b.

The bearings 22a and 22b are respectively bearing mounts 26.
The vibration sensor 3 is held on the bearing base 26a as shown in FIG. The vibration sensor 3 may be of any type as long as it can convert the vibration of the bearing base 26a into electric vibration. By the way, when the vibration sensor 3 detects the vibration of the electromagnetic clutch 2 as a rotating device, the motor 1 and the electromagnetic clutch 2 are coupled in the following procedure. First, after attaching the electromagnetic clutch 2, the belt 21 is stretched over the output shaft of the motor 1 and the clutch pulley 2a. At this stage, as shown in FIG.
The belt 21 is loosened by moving the motor plate 13 toward the bearing bases 26a and 26b, and then the motor plate 13 is separated from the bearing bases 26a and 26b as shown in FIG. Is applied with a prescribed tension, and the motor base 11 and the motor plate 13 are fixed at this position. Here, since the motor plate 13 is moved using the linear guide 12, the motor plate 13 is moved with respect to the motor base 11.
Can be smoothly moved, a horizontal force does not act on the vibration damping material 14, and a reduction in damping performance can be prevented.

Next, when the motor 1 is rotated, the electromagnetic clutch 2 is rotated. Therefore, based on the output of the vibration sensor 3 in this state, a vibration analysis described later is performed to detect the presence or absence of a defect, and if there is a defect, Identify the cause. Since the vibration buffers 14 and 14a are provided between the base 10 and the motor 1, the vibration transmitted from the motor 1 to the base 10 can be significantly attenuated, and the vibration of the motor 1 is detected by the vibration sensor 3. It is possible to suppress the detection as noise and detect the vibration of the electromagnetic clutch 2 as the rotating device separately from the noise. After the presence or absence of a defect and its cause are identified by vibration analysis, the motor 1 is stopped, the motor plate 13 is moved in the direction opposite to the above-described direction, the belt 21 is removed, and the electromagnetic clutch 2 is removed. With such a procedure, it is possible to detect the presence or absence of a defect in the electromagnetic clutch 2 and to identify the cause of the defect.

Instead of the linear guide 12, an air bearing 16 may be used as shown in FIG. The air bearing 16 is fixed to the motor base 11, and the motor plate 13
A slider 16a fixed to the lower surface side of the slider 16a, the slider 16a is accommodated in a bearing tube 16b so as to be able to float, and compressed air introduced through a channel 16c provided in the bearing tube 16b is blown to the slider 16a. The slider 16a is designed to float without contacting the bearing tube 16b. Therefore, it becomes possible to move the motor plate 13 relative to the motor base 11 more smoothly than the linear guide 12.

Since noise increases as the vibration damping materials 14, 14a deteriorate over time, the vibration damping materials 14, 1
It is necessary to detect the presence or absence of deterioration of 4a as well.
Therefore, as shown in FIG. 6, the vibration sensors 3 are provided at appropriate positions of the motor plate 13 and the base plate 10, respectively.
a and 3b are arranged and the motor 1 is rotated without the belt 21 hung, and the vibration of the motor plate 13 and the vibration of the base plate 10 are detected. If the vibration is analyzed in this way, the motor plate 13 to the base plate 10
Since it is possible to obtain the transfer function to the vibration damping material 1, by periodically obtaining the secular change of the transfer function,
4, it is possible to check the change over time of 14a.
For example, the transfer function under normal conditions is as shown in FIG. 7, and the transfer function when deteriorated due to aging is as shown in FIG.

By the way, in the apparatus shown in FIG. 2, the motor plate 13 is manually moved.
An air cylinder 17 as shown in FIG. 8 can be provided as a drive means for moving the motor plate 13. The air cylinder 17 has a rod 17a that expands and contracts in the sliding direction of the motor plate 13 and is connected to the motor plate 13, and the expansion and contraction amount of the rod is adjusted by air pressure. Therefore, similarly to the configuration shown in FIG. 2, before the belt 21 is hung, the motor plate 13 is mounted on the bearing bases 26a and 26b.
It may be arranged such that the belt 21 is placed on the belt 21 and the motor plate 13 is moved by the air cylinder 17 after the belt 21 is applied so that a predetermined tension is obtained on the belt 21. Of course, when the belt 21 is removed, the rod 17a of the air cylinder 17 is moved in the protruding direction so that the belt 21
Loosen.

As a driving means for moving the motor plate 13, a voice coil motor 18 can be used instead of the air cylinder 17, as shown in FIG.
In the boil coil motor 18, permanent magnets 18b are arranged on the inner side surfaces of the upper and lower pieces of the substantially E-shaped yoke 18a in the figure, and the central piece is inserted into the air-core coil 18c. The coil 18c moves along the longitudinal direction of the central piece by passing an electric current. Since the position of the coil 18c is determined according to the exciting current, the position of the coil 18c can be controlled only by the current command value by controlling the exciting current by giving an appropriate current command value to the driver 18d. The coil 18c is coupled to the motor plate 13 via the rod 18e and can move the motor plate 13 according to the current command value.

By the way, since the vibration damping materials 14, 14a are present between the motor 1 and the vibration sensor 3, the vibration itself of the motor 1 is hardly transmitted to the vibration sensor 3, but the vibration of the belt 21 does not occur. It will be detected by the vibration sensor 3. Since the vibration frequency of the belt 21 can be regarded as substantially constant while the vibration analysis of the electromagnetic clutch 2 as the inspection object is being performed, this vibration frequency is measured in advance and a filter for removing this vibration frequency is used. If the noise due to the vibration of the belt 21 is removed from the output of the vibration sensor 3 provided, the noise other than the vibration component generated by the electromagnetic clutch 2 is removed from the output of the vibration sensor 3, and the accuracy of the vibration analysis of the electromagnetic clutch 2 is improved. Can be increased.

The vibration frequency f _B (Hz) of the belt 21 is the tension T _B (N) of the belt 21, and the axial distance L (m) between the output shaft of the motor 1 and the rotary shaft 23 of the electromagnetic clutch 2.
And the characteristic value a (k
g / m) and is expressed by the following equation. f _B = T _B / (4 · L ² · a) When the vibration frequency f _B of the belt 21 is obtained using the above equation, the characteristic value a may be regarded as substantially constant, and the tension T
_B can be known by the magnitude of the load on the air cylinder 17 and the voice coil motor 18 (which can be known by the air pressure in the air cylinder 17 and the current value in the voice coil motor 18). By providing it, the above formula can be applied. Therefore, as shown in FIG. 10, a distance sensor 27 for measuring the distance between the motor plate 13 and the fixed position of the base plate 10 is provided. The distance sensor 27 is of an optical type, for example, and the distance L between the axes is obtained based on the distance between the distance sensors 27.

Since the tension T _B and the axial distance L can be adjusted, it becomes possible to cope with the change in the vibration frequency f _B due to the change in the amount of extension of the belt 21. For example, if the vibration frequency f _B is controlled to be constant, the noise can be removed without changing the pass band of the filter. If the distance is measured by the distance sensor 27 with the tension of the belt 21 being the same, the belt 21
You can know the change in length. Therefore, the belt 21
If it is noted that the elongation amount of the belt 21 increases as the belt 21 deteriorates, the change in the length of the belt 21 causes the belt 21 to change.
The deterioration state of can be checked. That is, the length of the belt 21 at the time of normal operation is measured, and the length of the belt 21 is regularly inspected so that the elongation amount is a predetermined value (for example, 5
%), It is possible to prompt replacement of the belt 21.

Needless to say, the position of the electromagnetic clutch 2 in the longitudinal direction of the rotary shaft 23 is adjusted so that the belt 21 is included in the plane orthogonal to the output shaft of the motor 1 and the rotary shaft 23. Absent. As a result, slippage does not occur between the output shaft of the motor 1 and the clutch pulley 2a and the belt 21, and it is possible to suppress the generation of unnecessary vibration components.

The vibration analysis based on the output of the vibration sensor 3 is
Do the following: That is, as shown in FIG.
The feature amount extraction unit 4 extracts the frequency feature amount and the time feature amount from the output of the vibration sensor 3 and inputs them to the abnormality diagnosis unit 5. The abnormality diagnosing unit 5 determines the presence or absence of an abnormality and the type of abnormality by collating the frequency characteristic amount, the time characteristic amount, and reference data described later, and displays the result on the display unit 6.

The basic procedure of vibration analysis is as shown in FIG. First, the output of the vibration sensor 3 is captured (S
1). Next, since the output of the vibration sensor 3 is waveform data, the time change of the spectrum of this waveform data is obtained (S2). That is, the relationship between the frequency, the time, and the power of the waveform data is obtained. The temporal change of the spectrum is obtained by performing a short time Fourier transform (STFT) or a wavelet transform on the waveform data. These calculations will be described later.

When the relationship between the frequency of the output of the vibration sensor 3 and time and power is obtained as described above, characteristic information on the relationship between frequency, time and power generated when an abnormality occurs in the rotating device is obtained. It is extracted as a feature amount (S3).
As the characteristic amount, a frequency having a peak when an abnormality occurs (hereinafter referred to as frequency characteristic amount) and a time interval (hereinafter referred to as time characteristic amount) when the power exceeds the reference value Ps for each frequency characteristic amount are used. If such a frequency feature quantity and a time feature quantity are paired, the pair of frequency feature quantity-time feature quantity becomes almost the same value for the same kind of abnormality of the rotating equipment, and the frequency is set for different abnormality. The feature amount-time feature amount pairs have different values. The reference value Ps can be set as appropriate, but can be set as Ps = μ + 3σ Ps = μ + 2σ using the average value μ and the standard deviation σ of the power in the waveform data of the non-defective product. If the latter reference value Ps is used, the detection probability of the frequency feature amount increases, so that it is possible to identify a large number of abnormalities by using more information, but the increase in the information amount increases the processing time. From
When there are few types of abnormalities to be identified, the former reference value P
It is sufficient to use s. Further, since the peak size does not become the same for each frequency feature quantity, it is desirable to set the reference value Ps to a different value for each frequency feature quantity. That is, this corresponds to weighting for each frequency feature amount.

Therefore, regarding various abnormalities of the rotating equipment,
Frequency characteristics and time characteristics are calculated in advance, and various
A set of frequency feature-time feature for each abnormality is set as reference data.
The waveform data generated when the rotating equipment
The set of frequency feature-time feature obtained from
(S4). For example, as shown in Figure 11.
, Multiple types of frequency feature quantity f₁, F₂,, and multiple
Different in combination with different types of time feature values ta, tb, ...
Create a table with standard types as standard data
From the waveform data obtained when rotating the rotating equipment,
The extracted frequency feature-time feature pairs are stored in this table.
To match. In the example shown in the figure,
The set of traits is (f₁, Ta), there is a scratch inside the bearing.
Defective (R ₁), (F₁, Tb) contact failure 1
(R₂), (F₂, Tb) contact failure 2 (R₃)
It is supposed to be. Poor contact is due to the contact between the bearing and the rotating shaft.
Touch, the clutch pulley 2a and the boss 2c of the electromagnetic clutch 2
It is a defect related to the contact of. In this example, the bearing scratches
And contact failure 1 have the same frequency feature amount f₁have
However, if the bearing is scratched,
Time corresponding to the rotation cycle of the member in contact with the member
Frequency feature f at intervals₁Peak occurs, and contact failure is 1.
In the case of contact failure 2, when it corresponds to the rotation cycle of rotating equipment
Frequency feature quantity f at intervals₂Peak occurs. like this
In addition, the two are distinguished by the difference in the time feature amount. Also,
Similarly, contact failure 1 and contact failure 2 have the same time characteristics.
Although the amount is tb, it can be identified because the frequency feature amount is different.
Is.

As described above, the presence / absence of abnormality and the type of abnormality can be specified by comparing the set of frequency characteristic amount-time characteristic amount with the reference data and determining whether they match (S5). . Note that it is necessary to allow the frequency feature amount and the time feature amount to have a slight width in the collation. By the way, in order to obtain the temporal change of the spectrum from the waveform data, it is necessary to obtain the spectrum at relatively short time intervals. Short-time Fourier transform and wavelet transform are known as this type of technique. When these techniques are used, the vibration analysis can be performed only by the output of one vibration sensor 3 without using a large number of filters for each frequency feature in order to obtain the frequency feature.

These conversions will be briefly described. The short-time Fourier transform converts the waveform data to be measured to x
(T), the window function is w (t, Δt), and the local spectrum Xs (t ₀ , w) at time t ₀ is x (t).
It is obtained by the Fourier transform of w (t−t ₀ , Δt). That is, it becomes as shown in Formula 1.

[0041]

[Equation 1]

The time-dependent changes in the spectrum obtained by subjecting the waveform data output from the vibration sensor 3 to short-time Fourier transform for the above-mentioned various abnormalities are as shown in FIGS. 12 to 14, respectively. According to these figures, the above-mentioned frequency feature quantities f ₁ and f ₂ and time feature quantity t
Since a and tb are obtained, it can be seen that the type of abnormality can be specified by the combination of the frequency feature amount and the time feature amount.

On the other hand, the wavelet transform is a wavelet transform.
Window function that expands and contracts along the time axis
The original waveform data and wavelet
This is for finding the convolution with a function. sand
That is, wavelet function Ψ _{a, b}(T) is the mother
Wavelet function Ψ (t)
Window function is scaled a times and the origin is b
This is the one that has been shifted by
Perform wavelet transform.

[0044]

[Equation 2]

In the wavelet transform, since the window function is expanded / contracted in the time axis direction, as shown in FIG. 15, the time resolution can be increased in the high frequency region of the waveform data, and the frequency resolution can be increased in the low frequency region. Get higher In addition,
In the above example, it is assumed that the time feature amount does not change during the vibration analysis, but the rotation speed of the motor 1 may have speed unevenness. Therefore, if a unit (rotary encoder or the like) for detecting the rotation speed of the rotating device is provided and the time characteristic amount is corrected according to the rotation speed, the abnormality diagnosis can be accurately performed even if the rotation speed is uneven.

(Embodiment 2) When the time feature amount is obtained from the time change of the spectrum as in the above embodiment,
In the first embodiment, the time interval at which the power exceeds the reference value is used as the time feature amount. However, the time interval thus obtained may include an error, so it is necessary to determine the time feature amount in consideration of the error. is there. For example, as shown in FIG. 16, the time feature value ta includes the errors Δt ₁ , Δt ₂ , and Δt _3.
May occur. Therefore, time-series data regarding each frequency feature is created and autocorrelation calculation is performed. Now, when the time-series data is represented by f (t) with respect to one frequency feature, the autocorrelation function Φ _xx (τ) is represented by Expression 3. When such an autocorrelation function Φ _xx (τ) is obtained, the autocorrelation function Φ _xx (τ) is obtained as shown in FIG. 17, so that the periodicity of the time series data f (t) can be known, The time feature amount can be determined.

Instead of the autocorrelation function Φ _xx (τ) for the time series data of the frequency feature quantity, the fast Fourier transform (FF
T) may be applied. In this case, as shown in FIG.
Since a sharp peak occurs at a specific frequency, the reciprocal of this frequency can be determined as the time feature amount. However, if the peak of the waveform data that occurs at the time of abnormality is a waveform that is attenuated in a short time, a plurality of peaks due to harmonic components are generated as shown in FIG. May not be identified. Therefore, in such a case, after performing the fast Fourier transform, the data is further subjected to the fast Fourier transform to perform so-called cepstrum calculation. In this case, as shown in FIG. 20, a sharp peak occurs in the time characteristic amount, and the time characteristic amount can be determined.

(Embodiment 3) This embodiment shows a case where a plurality of types of abnormalities occur at the same time. For example, the rotating portion of the rotating device is in contact with the fixed portion at two locations. In this case, if an abnormality occurs at each location, as shown in FIG. 21, two types of frequency feature values f ₃ ,
f ₄ will be detected at the same time. In such a case, after obtaining the respective frequency feature quantity f _3, f _4, each frequency feature quantity f _3, f respectively time every _four feature quantities tc, td
By determining, it is possible to specify a set of frequency feature amount-time feature amount. Therefore, as shown in FIG. 22, the causes of abnormality R ₃ and R ₄ can be identified by collation with the reference data.

(Embodiment 4) In the present embodiment, the type of abnormality is identified by inputting a set of a frequency feature amount and a time feature amount into a neural network without collating them with reference data in a table. is there. Specifically, as shown in FIG. 23, the frequency feature amount and the time feature amount are extracted from the output of the vibration sensor 3 in the feature amount extraction unit 4 and input to the abnormality diagnosis unit 5 using a neural network. The abnormality diagnosing unit 5 gives a frequency characteristic amount and a time characteristic amount to various known abnormalities, and gives a learning signal corresponding to the cause of the abnormality from the outside so that the learning is performed by a well-known technique such as back propagation. It functions as a classifier.
Therefore, when the frequency feature amount and the time feature amount are given to the abnormality diagnosis unit 5, an output corresponding to the cause of the abnormality is generated. In this case, when there are a plurality of types of frequency feature amounts and time feature amounts as in the third embodiment, the frequency feature amount-time feature amount sets are sequentially given to the abnormality diagnosis unit 5. The diagnosis result by the abnormality diagnosis unit 5 may be displayed on an appropriate display unit 6.

(Embodiment 5) In this embodiment, as shown in FIG. 24, a voice coil motor 31 as a vibrating means is added to the structure of Embodiment 1 shown in FIG.
The voice coil motor 31 is mechanically coupled to the bearing stand 26a. The voice coil motor 31 has the same configuration as the voice coil motor 18 shown in FIG. 9, and includes a yoke 31a, a permanent magnet 31b, and an air-core coil (voice coil) 31c. The permanent magnet 31b is magnetized so that the inner surface and the outer surface have different polarities, so that the coil 31c moves straight in the longitudinal direction of the central piece of the yoke 31a in accordance with the direction of the exciting current flowing through the coil 31c. It has become. In addition, the yoke 31a and the coil 31
c is connected to the coil 31c by a return spring 32, and the coil 31c is connected to the yoke 3 when no exciting current is flowing through the coil 31c.
It is held in a fixed position with respect to 1a. For the return spring 32, a coil spring, a diaphragm-shaped leaf spring, or the like is used. Further, the coil 31c has a connecting shaft 33.
One end of the connecting shaft 33 is connected, and a weight 34 is connected to the other end of the connecting shaft 33. The connecting shaft 33 is held by a bearing stand 24 provided upright on the base 10 and provided with a thrust bearing. An appropriate current command value is set as the exciting current flowing through the coil 31c in the driver 3
It is controlled by giving 1d. A yoke 31a is coupled to the bearing stand 26a.

Thus, the bearing base 26a can be vibrated by making the current flowing through the coil 31c pulse-shaped or sinusoidal-shaped. That is, the yoke 31a and the coil 31c try to move relative to each other in response to the exciting current to the coil 31c, and the weight 34 having an appropriate mass is attached to the side of the coil 31c.
Vibration can be applied to.

In this embodiment, the bearing stand 26 of the vibration sensor 3 is used.
It is intended to detect whether or not there is any abnormality in the mounting state with respect to a or the mounting state of the electromagnetic clutch 2 with respect to the bearing base 26a, and the voice coil motor 31 is driven while the motor 1 is not driven. I am doing it. Specifically, the abnormality diagnosis unit 5 determines whether or not there is an abnormality in the following procedure.

First, similarly to the case of analyzing the vibration of the electromagnetic clutch 2, the belt 21 is wound around the motor 1 and the clutch pulley 2a of the electromagnetic clutch 2. At this time, as shown in FIG. 24, the belt 24 is loosened by moving the motor plate 13 to the bearing bases 26a and 26b. Then, the motor plate 13 is moved to fix the motor plate 13 to the base 10 with a required tension applied to the belt 21.

Thereafter, the voice coil motor 31 is driven to vibrate the bearing stand 26a, and the ratio of the electric power Pi applied to the voice coil motor 31 to the vibration power Po detected by the vibration sensor 3 (= Po / pi). For a plurality of frequencies. This ratio corresponds to the input / output gain, and its numerical value becomes relatively small. Therefore, it is desirable to obtain the gain G converted into decibels by the following equation. G = 10 · log (Po / pi) An upper limit value Gs and a lower limit value Gi as threshold values set on the basis of the gain obtained when the device is normal (generally, a constant value of the average value of the gain G at the normal time is a constant value). When the calculated gain G is between the upper limit value Gs and the lower limit value Gi (that is, Gi <G <Gs), as shown in FIG. 25A, the vibration sensor is compared. 3 is bearing base 2
It is determined that the electromagnetic clutch 2 is securely attached to the bearing 6a and the electromagnetic clutch 2 is securely held on the bearing base 26a. By making such a judgment for a plurality of frequencies,
It is possible to check whether the device is normal or abnormal. Here, if the exciting current applied to the voice coil motor 31 is pulse-like, the frequency distribution of the gain G can be known by one excitation, and if the exciting current is sinusoidal, the gain G is changed while changing the frequency. Will be asked. If there is an abnormality in the device, the obtained gain G will deviate from between the upper limit value Gs and the lower limit value Gi, as shown in FIG.

As the vibrating means, as shown in FIG.
The actuator 35 using the piezoelectric effect may be used. This type of actuator 35 can also apply vibration to the bearing stand 26a, like the voice coil motor 31. Moreover, this type of actuator 35 can be driven at a high frequency, and can have a wider measurement frequency range than the voice coil motor 31. Further, in this type of actuator 35, the force applied to the bearing base 26a can be changed according to the command value of the voltage signal given from the outside, but conversely, the actuator 35 can be changed in accordance with the force applied to the bearing base 26a. It is also possible to obtain a voltage output. Therefore, the number of parts can be reduced by also using the actuator 35, which is a vibrating means, as the vibration sensor 3. However, in this configuration, the period during which the bearing base 26a is vibrated by the actuator 35 and the actuator 35
Is shifted in timing from the period in which is used as the vibration sensor 3. To measure the gain G, a pulsed or sinusoidal voltage signal is applied.

Further, as the vibrating means, a hammer 36 may be used as shown in FIG. The hammer 36 is driven by a drive source such as a solenoid (not shown) and gives an impact force to the bearing stand 26a. Since the impact force is applied when the hammer 36 is used, vibration including a large number of frequency components can be applied to the bearing stand 26a by one impact. Further, since the hammer 36 separates from the bearing base 26a after the impact force is applied, the vibration state of the mechanical system including the vibration sensor 3 and the bearing base 26a can be detected more accurately. Here, the load cell 37 is provided at a portion of the hammer 36 where the impact force acts on the bearing stand 26a, and the impact force applied to the bearing stand 26a can be accurately measured by measuring the impact force.

In the present embodiment, the normality / abnormality is judged only by whether or not the gain G is within the range between the upper limit Gs and the lower limit Gi. For example, the frequency distribution of the gain G is shown in FIG. With such a shape, although the gain G is within the range between the upper limit Gs and the lower limit Gi, there is no similarity in the frequency distribution pattern of the gain G. In such cases, the device is often not normal. Therefore,
The average value of the upper limit value Gs and the lower limit value Gi is used as inspection data, the correlation value between the average value and the gain G is calculated, the calculated correlation value is compared with a threshold value, and the correlation value is equal to or more than the threshold value (that is, similar). It may be determined to be normal when the frequency is high.

[0058]

According to the first aspect of the present invention, waveform data such as vibration or sound generated when the rotating device is rotated is detected, the temporal change of the spectrum of the waveform data is obtained, and a peak is generated from the temporal change of the spectrum. In addition to determining the frequency as the frequency feature amount, the time interval at which a peak occurs for each frequency feature amount is also determined as the time feature amount, and the set of frequency feature amount-time feature amount obtained from the waveform data when the rotating device is rotated is rotated. This is to identify the presence or absence of abnormalities in rotating equipment and the cause of abnormalities by comparing the set of frequency feature-time feature at the time of equipment abnormality with the reference data registered in advance for each cause of abnormality. Since the cause of abnormality is identified by combining it with the time feature quantity, it is possible to compare the cause of abnormality with the one that identifies the cause of abnormality using frequency only as in the past. There is an advantage that determination accuracy is increased.

When the time change of the spectrum of the waveform data is obtained by applying the short-time Fourier transform to the waveform data as in the second aspect of the invention, the frequency characteristic can be obtained based on one waveform data without using many filters. There is an advantage that the quantity and the time feature quantity can be obtained. When the time change of the spectrum of the waveform data is obtained by performing the wavelet transform on the waveform data as in the third aspect of the invention, the time resolution becomes high in the high frequency region and the frequency resolution becomes high in the low frequency region. There is an advantage that the quantity and the time feature quantity can be efficiently extracted.

According to the invention of claim 4, after the frequency feature amount is obtained, the autocorrelation calculation is performed on the time series data of each frequency feature amount, and the time interval at which the autocorrelation value peaks is the time feature amount. There is an advantage that a good value can be obtained as the time feature amount even if there is some deviation in the time intervals at which peaks occur for each frequency feature amount.

As in the invention of claim 5, after obtaining the frequency feature quantity, the fast Fourier transform is performed on the time series data of each frequency feature quantity, and the reciprocal of the frequency at which the maximum value is obtained is set as the time feature quantity. The one used has an advantage that the generation cycle of the peak in each frequency feature amount can be easily extracted. As in the invention of claim 6, after the frequency characteristic amount is obtained, the cepstrum calculation is performed in which the fast Fourier transform is performed twice for the time series data of each frequency characteristic amount, and the kefency value that gives the maximum value is obtained as the time characteristic. The method used for the quantity has an advantage that the peak generation period can be easily extracted even when the peak decays quickly.

According to the seventh aspect of the present invention, the time feature amount is obtained for each of the plurality of frequency feature amounts, and each set of frequency feature amount-time feature amount is collated with the reference data.
Even if there are a plurality of causes of abnormality, there is an advantage that the cause of abnormality can be specified. According to the invention as claimed in claim 8, in which the rotation speed of the rotating device is detected and the time feature amount is corrected according to the rotation speed, the cause of the abnormality can be specified even when the rotation speed of the rotating device varies. become.

According to a ninth aspect of the present invention, waveform data such as vibration or sound generated when the rotating device is rotated is detected, the time change of the spectrum of this waveform data is obtained, and the frequency at which a peak is generated from the time change of the spectrum is determined. In addition to the frequency characteristic amount, the time interval at which a peak is generated for each frequency characteristic amount is obtained as the time characteristic amount, and the set of the frequency characteristic amount-time characteristic amount obtained from the waveform data of the rotating device during rotation is set to The neural network is designed to identify the presence / absence of an abnormality and the cause of an abnormality in a rotating device by giving a neural network trained in advance so as to associate a set of frequency feature amount-time feature amount at the time of abnormality with each cause of abnormality. Is used to classify the causes of anomalies, so when an anomaly occurs, you can learn the cause of anomalies simply by learning There is an advantage that it becomes possible. That is, it is not necessary to set reference data.

According to the tenth aspect of the present invention, weighted reference values are set for the respective frequency feature amounts, and the time interval at which the time series data for each frequency feature amount exceeds the reference value is set for each frequency. With the one used for the time feature amount for each feature amount, it is possible to obtain the frequency feature amount and the time feature amount that match the conditions of the rotating device, compared to the case where the reference value is uniformly set, and the accuracy of determining the cause of the abnormality is higher. improves.

According to the eleventh aspect of the present invention, in the case where power is transmitted from the drive source to the rotary device via the belt and the drive source is vibration-proof, the vibration from the drive source to the rotary device is reduced. There is an advantage that the transmission can be suppressed and the noise in the waveform data is reduced. According to the invention of claim 12, in which the power is transmitted from the drive source to the rotating device via the belt and the tension of the belt is kept constant while the rotating device is driven, the vibration measurement of the rotating device is performed. The conditions become constant, leading to improved diagnostic accuracy.

According to the thirteenth aspect of the present invention, the belt elongation amount is obtained by obtaining the axial distance between the drive source and the rotating device when the belt tension is a constant value. It has the advantage that it can be managed and the timing of belt replacement can be known. Claim 1
As in the invention of 4, the power is transmitted from the drive source to the rotating device through the belt, and the time component of the spectrum of the waveform data is obtained after removing the frequency component of the vibration frequency of the belt from the waveform data. The noise due to the vibration of the belt can be removed to perform the abnormality diagnosis.

According to the fifteenth aspect of the invention, in which the vibration frequency of the belt is calculated based on the axial distance between the drive source and the rotating device and the tension of the belt, the vibration of the belt is changed even if the tension of the belt changes. Since the frequency can be known, the noise due to the vibration of the belt can be easily removed. According to a sixteenth aspect of the present invention, in which the power is transmitted from the drive source to the rotating device via the belt and the belt is positioned in a plane orthogonal to the axes of the drive source and the rotating device, the belt is a shaft. Since it does not slip in the direction, unnecessary vibration does not occur and the accuracy of abnormality diagnosis is improved.

According to a seventeenth aspect of the present invention, a vibration sensor for detecting waveform data such as vibration generated when the rotating equipment rotates, a time change of the spectrum of the waveform data, and a frequency which produces a peak from the time change of the spectrum. Is obtained as a frequency feature amount, and a feature amount extraction unit that obtains a time interval at which a peak occurs for each frequency feature amount as a time feature amount, and a frequency feature amount-time feature amount obtained from waveform data during rotation of a rotating device. An abnormality diagnosis unit that identifies the presence or absence of an abnormality and the cause of an abnormality in a rotating device by comparing the set of frequency feature-time feature when an abnormality occurs in the rotating device with reference data that has been registered in advance for each cause of abnormality. In this case, since the cause of abnormality is specified by combining the frequency feature quantity and the time feature quantity, only the frequency is used as in the conventional method. There is an advantage that discrimination accuracy of the comparison to the abnormality cause which specifies a cause of abnormality is high.

According to the eighteenth aspect of the present invention, it is provided with a belt for transmitting power from the drive source to the rotating device and a vibration damping material for supporting the drive source so as not to transmit the vibration of the drive source to the rotating device. However, since the vibration from the drive source is hardly transmitted to the rotating device, the vibration sensor does not detect an unnecessary vibration component, and the accuracy of abnormality diagnosis is improved.

According to the nineteenth aspect of the present invention, a pair of sensors for detecting vibrations on both sides of the vibration damping material are provided, and the transfer function of the vibration damping material is obtained based on the outputs of both sensors. In a device provided with a means for detecting the deterioration of the vibration damping material by the dynamic change, the deterioration of the vibration damping material can be managed by managing the transfer function of the vibration damping material.

According to the twentieth aspect of the present invention, in the case where the drive source is movable with respect to the rotating device and the linear guide is provided for restricting the movement range of the drive source within the plane including the belt, the movement of the drive source is prevented. It is possible to prevent the generation of an unnecessary vibration component by restricting the direction, and it is possible to appropriately adjust the tension of the belt. According to the invention of claim 21, the drive source is movable with respect to the rotating device,
In addition, in the case where the air bearing is provided to restrict the moving range of the drive source within the plane including the belt, the drive source can be moved more smoothly than in the linear guide.

According to the twenty-second aspect of the present invention, the driving source can be easily moved by the moving means by the supporting means for movably supporting the driving source with respect to the rotating device and the moving means for moving the driving source. It can be moved. Claim 2
According to the third aspect of the present invention, the distance sensor that detects the distance between the drive source and the rotating device can be used to determine the amount of belt elongation or the belt tension by detecting the distance. There is an advantage that the deterioration of the belt can be detected and a guideline for the belt replacement timing can be given.

According to the twenty-fourth aspect of the present invention, there is provided a vibrating means for vibrating the rotating device when the drive source is not operating, and the abnormality diagnosis section uses a vibration sensor when the rotating device is vibrated by the vibrating means. In the one that determines the presence or absence of abnormality in the mechanical system including the rotating device and the vibration sensor based on the detected waveform data,
It becomes possible to inspect the mounting state of the vibration sensor on rotating equipment, and it is possible to perform reproducible abnormality diagnosis.

When the vibrating means is the voice coil motor as in the twenty-fifth aspect, it is easy to accurately control the vibration applied to the rotating device, and it is easy to determine whether there is an abnormality in the rotating device or the vibration sensor. According to the twenty-sixth aspect of the present invention, in the actuator using the piezoelectric effect as the vibrating means, the frequency of vibration applied to the rotating device can be increased, and the rotating device and the vibration sensor can be used over a wide frequency range. You can check whether there is any abnormality with.

As in the twenty-seventh aspect of the invention, in which the vibrating means is also used as the vibration sensor, there is an advantage that the constituent elements can be reduced while using the vibrating means. According to the twenty-eighth aspect of the invention, in the case where the vibrating means is a hammer that gives an impact force to the rotating device, there is an advantage that the configuration becomes simple as compared with a voice coil motor or an actuator as the vibrating means.

According to a twenty-ninth aspect of the invention, in the abnormality diagnosing device for a rotating device according to any one of the twenty-fourth to twenty-eighth aspects, the rotating device is vibrated by the vibrating means before the drive source is operated, and the vibration sensor In order to check the presence or absence of abnormality in the mechanical system including the rotating device and the vibration sensor by obtaining the input and output gain in the abnormality diagnosis section based on the output and comparing this gain with the preset normal range, Since the presence or absence of an abnormality in the mechanical system including the rotating device and the vibration sensor is determined by comparing the gain and the threshold value, it is possible to easily diagnose whether or not there is an abnormality in the mounting state of the rotating device and the vibration sensor. There is an advantage that you can.

According to a thirtieth aspect of the present invention, in the abnormality diagnosing device for a rotating device according to the twenty-fourth to twenty-eighth aspects, the rotating device is vibrated by the vibrating means before the drive source is operated, and the vibration sensor Based on the output, the input / output gains at multiple frequencies are obtained by the abnormality diagnosis unit, the correlation value between the obtained gain and the preset inspection data is obtained, and the obtained correlation value is the preset threshold value. In order to inspect the presence or absence of abnormality in the mechanical system including the rotating device and the vibration sensor by comparing with, the correlation value between the gain and the inspection data obtained at a plurality of frequencies is obtained, and the rotating device is calculated based on the correlation value. Since the presence / absence of an abnormality in the mechanical system including the vibration sensor and the vibration sensor is determined, it is possible to more accurately determine the presence / absence of an abnormality in comparison with the configuration of claim 29.

[Brief description of drawings]

FIG. 1 shows the first embodiment, (a) is a block diagram,
(B) is an operation explanatory view.

FIG. 2 is a plan view showing the above measuring device.

3A and 3B show the above-described measuring device, FIG. 3A is a side view when a belt is worn, and FIG. 3B is a side view in a measurement state.

FIG. 4 is a sectional view of an electromagnetic clutch used in the above.

FIG. 5 is a partially cutaway side view of the above when an air bearing is used.

FIG. 6 is a side view showing an example of measuring the transfer function of the vibration damping material in the above.

FIG. 7 is a diagram showing changes in the transfer function of the vibration damping material in the above.

FIG. 8 is a side view showing an example using an air cylinder in the above.

FIG. 9 is a plan view showing an example using a voice coil motor in the above.

FIG. 10 is a side view showing an example using a distance sensor in the above.

FIG. 11 is a diagram showing an example of reference data used in the above.

FIG. 12 is a diagram showing an example of a temporal change of a spectrum in the above.

FIG. 13 is a diagram showing an example of temporal change of spectrum in the above.

FIG. 14 is a diagram showing an example of temporal change of spectrum in the above.

FIG. 15 is a conceptual diagram of a wavelet transform used in the above.

FIG. 16 is a diagram showing an example of a temporal change of a spectrum in the second embodiment.

FIG. 17 is a diagram showing an example using an autocorrelation function in the above.

FIG. 18 is a diagram showing an example in which a fast Fourier transform is applied in the above.

FIG. 19 is a diagram showing a problem example when a fast Fourier transform is applied in the above.

FIG. 20 is a diagram showing an example of performing a cepstrum calculation in the above.

FIG. 21 is a diagram showing an example of a temporal change of a spectrum in the third embodiment.

FIG. 22 is a diagram showing an example of reference data in the above.

FIG. 23 is a block diagram showing a fourth embodiment.

24A and 24B show a fifth embodiment, wherein FIG. 24A is a plan view and FIG.
Is a side view.

FIG. 25 is an operation explanatory diagram of the above.

FIG. 26 is a plan view showing another configuration example of the above.

FIG. 27 is a plan view showing still another configuration example of the above.

FIG. 28 is an explanatory diagram of the operation of the above.

[Explanation of symbols]

1 motor 2 electromagnetic clutch 3 Vibration sensor 4 Feature extraction unit 5 Abnormality diagnosis section 31 voice coil motor 35 Actuator 36 Hammer

─────────────────────────────────────────────────── ─── Continuation of the front page (56) Reference JP-A-2-251771 (JP, A) JP-A-3-92795 (JP, A) JP-A-4-89534 (JP, A) JP-A-4- 273036 (JP, A) JP 5-248937 (JP, A) JP 6-160172 (JP, A) JP 8-95955 (JP, A) JP 8-177530 (JP, A) JP-A-8-219873 (JP, A) JP-A-9-113416 (JP, A) JP-A-10-26580 (JP, A) JP-A-10-258974 (JP, A) JP-B-7-31076 (JP, B2) Patent 2924243 (JP, B2) (58) Fields investigated (Int.Cl. ⁷ , DB name) G01H 17/00 G01M 13/00-13/04 G01M 19/00-19/02

Claims (30)

(57) [Claims] 1. Waveform data such as vibration or sound generated during rotation of a rotating device is detected, a temporal change of the spectrum of the waveform data is obtained, and a frequency at which a peak is generated from the temporal change of the spectrum is obtained as a frequency feature amount. At the same time, the time interval at which a peak occurs for each frequency feature amount is obtained as a time feature amount, and the set of frequency feature amount-time feature amount obtained from the waveform data during rotation of the rotating device is used as the frequency feature when the rotating device is abnormal. A method for diagnosing abnormalities in rotating equipment, characterized in that the presence / absence of an abnormality in the rotating equipment and the cause of the abnormality are specified by collating a set of quantity-time characteristic amounts with reference data registered in advance for each cause of abnormality. 2. The method for diagnosing abnormalities in a rotating machine according to claim 1, wherein the time change of the spectrum of the waveform data is obtained by applying a short-time Fourier transform to the waveform data. 3. The method for diagnosing abnormalities in a rotating machine according to claim 1, wherein the time change of the spectrum of the waveform data is obtained by performing wavelet transformation on the waveform data. 4. After obtaining the frequency feature quantity, an autocorrelation calculation is performed on the time-series data of each frequency feature quantity, and a time interval at which the autocorrelation value peaks is used as the time feature quantity. A method for diagnosing an abnormality in a rotating device according to claim 1. 5. After obtaining the frequency feature quantity, a fast Fourier transform is performed on the time series data of each frequency feature quantity, and the reciprocal of the frequency at which the maximum value is obtained is used as the time feature quantity. A method for diagnosing an abnormality in a rotating device according to any one of claims 1 to 3. 6. A cepstral operation for performing fast Fourier transform twice on the time-series data of each frequency feature amount after obtaining the frequency feature amount, and using a kefency value for obtaining the maximum value as the time feature amount. The method for diagnosing an abnormality of a rotating device according to claim 1, wherein 7. The method for diagnosing an abnormality in a rotating device according to claim 1, wherein time characteristic amounts are obtained for a plurality of frequency characteristic amounts, and each set of frequency characteristic amount-time characteristic amount is collated with reference data. . 8. The method for diagnosing an abnormality in a rotating device according to claim 1, wherein the rotating speed of the rotating device is detected, and the time feature amount is corrected according to the rotating speed. 9. Waveform data such as vibration or sound generated when a rotating device is rotated is detected, the time change of the spectrum of this waveform data is obtained, and the frequency at which a peak is generated from the time change of the spectrum is obtained as a frequency feature amount. At the same time, the time interval at which a peak occurs for each frequency feature amount is obtained as a time feature amount, and the set of frequency feature amount-time feature amount obtained from the waveform data during rotation of the rotating device is used as the frequency feature when the rotating device is abnormal. Abnormality diagnosis method for rotating equipment, characterized in that the presence or absence of abnormality in the rotating equipment and the cause of the abnormality are specified by giving to a neural network that has been learned in advance so as to associate a set of quantity-time feature quantities and each cause of abnormality. . 10. A weighted reference value is set for each frequency feature quantity, and a time interval at which time series data for each frequency feature quantity exceeds the reference value is a time feature quantity for each frequency feature quantity. 10. The method for diagnosing abnormalities of rotating equipment according to claim 1 or 9, characterized in that 11. The method for diagnosing abnormalities in a rotating device according to claim 1, wherein power is transmitted from the driving source to the rotating device via a belt, and the driving source is vibration-isolated. 12. The rotating device according to claim 1, wherein power is transmitted from the drive source to the rotating device through the belt, and the tension of the belt is kept constant while the rotating device is being driven. Abnormality diagnosis method. 13. The abnormality diagnosis of rotating equipment according to claim 12, wherein the belt elongation amount is obtained by obtaining the axial distance between the drive source and the rotating equipment when the tension of the belt is a constant value. Method. 14. A power source is transmitted from a drive source to a rotating device through a belt, the frequency component of the vibration frequency of the belt is removed from the waveform data, and then the time change of the spectrum of the waveform data is obtained. The method for diagnosing abnormalities of rotating equipment according to claim 1. 15. The method for diagnosing an abnormality in a rotating device according to claim 14, wherein the vibration frequency of the belt is calculated based on the axial distance between the drive source and the rotating device and the tension of the belt. 16. The power source is transmitted from a drive source to a rotating device via a belt, and the belt is positioned in a plane orthogonal to an axis between the drive source and the rotating device.
Abnormality diagnosis method for rotating equipment described. 17. A vibration sensor for detecting waveform data such as vibration generated during rotation of a rotating device, a time change of a spectrum of the waveform data is obtained, and a frequency which produces a peak from the time change of the spectrum is used as a frequency feature amount. In addition to obtaining the feature, a feature amount extraction unit that obtains a time interval at which a peak occurs for each frequency feature amount as a time feature amount, and a set of frequency feature amount-time feature amount obtained from waveform data during rotation of the rotating device Is provided with an abnormality diagnosing unit that identifies the presence or absence of an abnormality in a rotating device and the cause of the abnormality by checking the set of frequency feature-time feature at the time of abnormality with reference data registered in advance for each cause of abnormality. An abnormality diagnosis device for rotating equipment. 18. A belt, which transmits power from the drive source to the rotating device, and a vibration damping material, which supports the drive source so as not to transmit the vibration of the drive source to the rotating device. 17. A rotating device abnormality diagnosing device according to item 17. 19. A pair of sensors for detecting vibrations on both sides sandwiching the vibration damping material are provided, a transfer function of the vibration damping material is obtained based on the outputs of both sensors, and the vibration damping material of the vibration damping material is changed with time. The rotating machine abnormality diagnosis apparatus according to claim 18, further comprising means for detecting deterioration. 20. The rotating device abnormality according to claim 18, further comprising a linear guide that allows the drive source to move with respect to the rotating device and that limits a moving range of the drive source within a plane including a belt. Diagnostic device. 21. An abnormality of a rotating device according to claim 18, further comprising an air bearing that makes the drive source movable with respect to the rotating device and restricts a moving range of the drive source within a plane including a belt. Diagnostic device. 22. The abnormality diagnosing device for rotating equipment according to claim 18, further comprising: a supporting means for movably supporting the driving source with respect to the rotating equipment, and a moving means for moving the driving source. 23. The rotating device abnormality diagnosing device according to claim 18, further comprising a distance sensor for detecting a distance between the drive source and the rotating device. 24. A vibrating means for vibrating the rotating device when the drive source is not operating is provided, and the abnormality diagnosis section is based on waveform data detected by a vibration sensor when the rotating device is vibrated by the vibrating means. The abnormality diagnosing device for rotating equipment according to claim 17, wherein the presence or absence of abnormality of a mechanical system including the rotating equipment and the vibration sensor is determined. 25. The abnormality diagnosing device for rotating equipment according to claim 24, wherein the vibrating means is a voice coil motor. 26. The abnormality diagnosing device for rotating equipment according to claim 24, wherein the vibrating means is an actuator using a piezoelectric effect. 27. The abnormality diagnosing device for a rotating device according to claim 25, wherein the vibrating means is also used as a vibration sensor. 28. The abnormality diagnosing device for a rotating device according to claim 24, wherein the vibrating means is a hammer that applies an impact force to the rotating device. 29. The abnormality diagnosis device for a rotating device according to claim 24, wherein the rotating device is vibrated by the vibrating means before the drive source is operated, and the abnormality diagnosis unit is based on the output of the vibration sensor. Abnormality diagnosis of rotating equipment characterized by checking the presence or absence of abnormality in the mechanical system including the rotating equipment and vibration sensor by obtaining the input / output gain with and comparing this gain with the preset normal range. Method. 30. The abnormality diagnosing device for a rotating device according to claim 24, wherein the rotating device is vibrated by the vibrating means before the drive source is operated, and the abnormality diagnosing unit is based on the output of the vibration sensor. The input / output gains at multiple frequencies are calculated with, and the correlation value between the calculated gain and the preset inspection data is calculated, and the calculated correlation value is compared with the preset threshold value. And a vibration sensor for inspecting the presence or absence of an abnormality in a mechanical system.