JP6511573B1 - Method and apparatus for diagnosing abnormality of rolling bearing, abnormality diagnosis program - Google Patents

Abstract

An object of the present invention is to calculate a robust evaluation index in consideration of the effects of rotational frequency and transfer function, and to make it possible to judge the presence or absence of abnormality with high accuracy. A vibration frequency is changed to a rotational frequency to be measured in S1, and a vibration acceleration is measured and stored in S2. A corrected vibration value is calculated from the rotational frequency and vibration acceleration at the time of vibration measurement in S3, Fourier transform of the corrected vibration value is performed in S4, and the amplitude for each specific frequency is recorded. If vibration measurement at a plurality of rotational frequencies for measuring all vibrations defined in S5 has been completed, an average is calculated for the value of the corrected vibration value for each specific frequency recorded in S6, and the feature is determined in S7 Extract the value of the corrected vibration value of the specific frequency. In the case of diagnosis based on the growth rate of vibration, vibration measurement at the rotational frequency at which all vibrations are measured is repeated twice, then an evaluation index is calculated from the bearing-induced vibration value in S11, and the evaluation index exceeds the threshold in S12. Determine if you [Selected figure] Figure 5

Description

  The present invention relates to a method, apparatus, and program for diagnosing an abnormality in a rolling bearing that is used in a machine tool or the like and supports a rotating body such as a main shaft.

When an abnormality such as damage to the inner ring occurs in the rolling bearing that supports the rotating body, vibration occurs. Since the force generated by the abnormality of the bearing is not a simple sine wave, vibrations of harmonic frequency components are simultaneously observed. The frequency (characteristic frequency) of the vibration generated at this time is proportional to the rotational speed, and can be calculated from the rotational speed of the rotating body and the bearing specifications.
If the magnitudes of the oscillations of the two feature frequencies separated by the fundamental frequency are both large, it can be interpreted that the amplitude fluctuates at the fundamental frequency when the waveform of the measured oscillation is viewed. Therefore, there is known a method of quantifying the degree of fluctuation of the amplitude and performing diagnosis by performing frequency processing after performing envelope processing on the waveform of the measured vibration.
For example, in Patent Document 1, vibration or sound is measured, envelope processing and frequency analysis are performed, and the magnitude of the calculated value obtained by dividing the magnitude of the vibration of the fundamental frequency component by the overall value which is the integral value of all spectrum components. Shows a method of determining the presence or absence of an abnormality.
In Patent Document 2, the vibration is measured, envelope processing and frequency analysis are performed, and the value of the characteristic frequency component is individually set for each characteristic frequency in consideration of the level difference of the vibration response and the rotation frequency previously measured by the impact test. A technique for diagnosing in comparison to a threshold is shown.

Patent No. 4120099 Patent No. 5146008

In the case of a machine such as a machine tool in which manual adjustment of accuracy is performed, there are machine variations to the extent that there is no functional problem. The machine's operating environment, such as factory ground and room temperature, also changes the machine's transfer function. Even in the case of vibration measurement at the same rotational frequency at which the same type of vibration force of the same magnitude is generated due to the same abnormality in the same model, it is affected by the vibration mode (transfer function) of the machine. Can not be measured identically. In the case described later, when the magnitudes of transfer functions of three identical models are compared, the machine with the smallest magnitude of the transfer function at a certain frequency is about 10% or less of the largest machine. Even if the distribution of vibration magnitudes is investigated in many machines of the same model, and an attempt is made to determine the threshold value for discriminating between normal and abnormal, compared to the difference in transfer function magnitude due to machine dispersion, If the difference in magnitude between normal and abnormal vibrations is small, it is not possible to provide a threshold value for reliably identifying normal / abnormal, in consideration of machine variations.
When diagnosis is performed on a rotating body having a complicated vibration mode such as the spindle of a machine tool, even if envelope processing is performed, the rotational frequency of the rotating body is changed by 10% and the feature frequency is changed by 10%. Even if there is, the magnitude of the vibration of the characteristic frequency may change several times. This is because the magnitude of the transfer function is largely different for each frequency. The envelope processing is processing for collectively capturing vibrations of a plurality of frequencies without considering that the magnitudes of transfer functions are different at each frequency. Since this is a processing method in which information of a plurality of frequencies is irreversibly mixed, it is theoretically impossible to perform processing taking into account the magnitude of the transfer function after envelope processing.

Therefore, the process of dividing the magnitude of the vibration of the feature frequency after envelope processing by the overall as in Patent Document 1 is not a method of removing the influence of the transfer function or the rotational frequency, so the calculation is performed by the method of Patent Document 1 Although the calculated value can be used to determine the presence or absence of a flaw, there is a problem that quantitative comparison can not be performed.
Further, as proposed in Patent Document 2, the threshold in consideration of the influence of the transfer function and the rotational frequency on the magnitude of the vibration of the feature frequency after the envelope processing is rational from the transfer function and the rotational frequency There is a problem that it can not be decided
For this reason, in the bearing diagnosis apparatus generally marketed, setting of the threshold value to be judged as abnormal is left to the user, and it can not easily be used for abnormality diagnosis, or with the threshold value to be judged as abnormal. However, there is a problem that even if the rotational frequency at which the measurement is performed is slightly changed, the determination result greatly changes, and it can not be determined whether it is really abnormal.
On the other hand, in the case of a type of abnormality in which the magnitude of the force generated by the bearing abnormality changes due to the rotational frequency, there is a problem that appropriate diagnosis can not be performed without considering the influence of the rotational frequency.

  Therefore, the present invention has been made in view of the above problems, and it is possible to calculate a robust evaluation index in consideration of the influence of rotational frequency and transfer function, and diagnose the abnormality of the rolling bearing which can judge the presence or absence of abnormality with high accuracy. It aims at providing a method and apparatus, and an abnormality diagnosis program.

In order to achieve the above object, the invention according to claim 1 is a method of diagnosing an abnormality in a rolling bearing supporting a rotating body,
Measuring a vibration of the rotating body at a plurality of rotational frequencies of the rotating body;
A frequency analysis step of calculating the magnitude of vibration at each frequency by dividing the vibration by the power of an index larger than zero of a predetermined physical quantity proportional to the rotational frequency of the rotating body at the time of vibration measurement and performing frequency analysis When,
A vibration average value calculating step of calculating a vibration average value obtained by averaging the magnitudes of the vibrations having the same ratio of frequency to rotational frequency of the rotating body at the time of vibration measurement in the vibration measuring step;
A bearing-induced vibration value extraction step of extracting a bearing-induced vibration value at a feature ratio frequency that is a ratio of the frequency at which the vibration caused by the bearing occurs to the rotational frequency among the vibration average values;
An evaluation index calculation step of calculating an evaluation index based on the bearing-induced vibration value;
Performing a determination step of determining the presence or absence of an abnormality based on the evaluation index.
The invention according to claim 2 is that in the configuration according to claim 1, any one of the rotation frequency, the reciprocal of the rotation frequency, and the logarithm of the rotation frequency is equally spaced in the vibration average value calculation step. The vibration average value may be calculated using the magnitude of the vibration measured at the combination of the rotational frequencies.
The invention according to claim 3 is the configuration according to claim 1 or 2, wherein, in the vibration measurement step, vibration measurement of the combination having the same rotational frequency is performed a plurality of times,
The evaluation index calculating step is characterized in that the evaluation index is calculated based on the degree of change of the bearing-induced vibration value.
The invention according to claim 4 is that, in the configuration according to any one of claims 1 to 3, in the vibration average value calculating step, an average is made using only the magnitude of the vibration whose frequency is within a predetermined range. It is characterized by taking.
In order to achieve the above object, the invention according to claim 5 is an apparatus for diagnosing an abnormality in a rolling bearing supporting a rotating body,
Vibration measuring means for measuring the vibration of the rotating body at a plurality of rotational frequencies of the rotating body;
Frequency analysis means for calculating the magnitude of vibration for each frequency by dividing the vibration by the power of an index larger than zero of a predetermined physical quantity proportional to the rotational frequency of the rotating body at the time of vibration measurement and performing frequency analysis When,
Vibration average value calculation means for calculating a vibration average value obtained by averaging the magnitudes of the vibrations having the same ratio of frequency to rotational frequency of the rotating body at the time of vibration measurement by the vibration measurement means;
A bearing-induced vibration value extraction unit that extracts a bearing-induced vibration value at a feature ratio frequency that is a ratio of the frequency at which the vibration caused by the bearing occurs to the rotational frequency among the vibration average values;
Evaluation index calculation means for calculating an evaluation index based on the bearing-induced vibration value;
And determining means for determining the presence or absence of an abnormality based on the evaluation index.
In order to achieve the above object, the invention according to claim 6 is a program for diagnosing an abnormality of a rolling bearing supporting a rotating body, wherein each vibration of the rotating body measured at a plurality of rotational frequencies is each rotation The frequency analysis step, the vibration average value calculation step, the bearing-induced vibration value extraction step, the evaluation index calculation step, and the evaluation method for a rolling bearing abnormality diagnosis method according to any one of claims 1 to 4 And a determination step.

According to the inventions of claims 1 5 and 6, since the influence of the rotational frequency is removed from the measured vibration value to calculate the average, the influence of the machine variation is reduced to reduce the bearing-induced vibration value. You can get it. Since the evaluation index calculated from the bearing-induced vibration value is also less affected by machine fluctuation, it is easy to set the threshold value to be judged as abnormal. That is, it is possible to determine the presence or absence of abnormality with high accuracy by calculating a robust evaluation index in consideration of the influence of the rotational frequency and the transfer function.
According to the second aspect of the invention, in addition to the above effects, the effect of reducing machine-to-machine variation in transfer function magnitude and variation on specific frequency of the magnitude of the transfer function are improved.
According to the third aspect of the invention, in addition to the above effects, the value of the evaluation index is a value that does not depend on the value of the magnitude of the transfer function. Even when measuring at the sensor position, diagnosis using the same threshold value is possible.
According to the fourth aspect of the present invention, in addition to the above effects, the reduction effect of the variation of the transfer function magnitude and the variation of the transfer function magnitude for each specific frequency can be further improved.

It is a functional block diagram of an abnormality diagnosis device for rolling bearings. It is a graph which shows the relationship between the excitation force and angular velocity by bearing abnormality. It is a graph which shows the dispersion | variation in the magnitude | size of the transfer function in the same model. It is a graph which shows the dispersion | variation in the average of the magnitude | size of the transfer function in the same model. It is a flowchart of the abnormality diagnosis method. It is an explanatory view of a diagnostic result display in a case of using a growth rate of vibration as an evaluation index. It is explanatory drawing of a diagnostic result display in the case of using the maximum value of a bearing origin vibration value as an evaluation index.

Hereinafter, embodiments of the present invention will be described based on the drawings.
FIG. 1 is a functional block diagram showing a configuration in the case where a rolling bearing abnormality diagnosis device is applied to a main shaft of a machine tool, which will be specifically described based on this drawing.
The spindle 1 is rotatably attached to the spindle housing 2 via a bearing 7 which is a rolling bearing, and a tool 3 for processing is fixed. The motor 4 drives the spindle 1. The motor 4 is provided with a speed detector 5 so that the measured rotational frequency of the motor 4 is input to the controller 6. At the time of processing, the control device 6 controls the current supplied to the motor 4 so as to maintain the rotational frequency of the motor 4 measured by the speed detector 5 at the command rotational frequency.

  A vibration sensor 8 as vibration measuring means is attached to the spindle housing 2, and the vibration acceleration measured by the vibration sensor 8 is converted into a digital value by the A / D converter 9, and stored together with the rotational frequency at the time of vibration measurement. It is stored in the unit 10. The storage unit 10 also stores a threshold set in advance. The calculation unit 11 which is a computer calculates the feature ratio frequency of the bearing abnormality from the bearing specifications stored in the storage unit 10 according to the abnormality diagnosis program stored in the storage unit 10, and the vibration stored in the storage unit 10 Performs multiplication, Fourier transform, calculation of absolute value, and interpolation processing from rotational frequency and vibration acceleration at the time of measurement to calculate a corrected vibration value to be described later for the relative frequency, and also calculates the magnitude of vibration for each frequency. . In addition, the calculation unit 11 calculates a vibration average value obtained by averaging the magnitudes of vibrations having the same ratio of the frequency to the rotational frequency of the main shaft 1 at the time of vibration measurement, and vibrations due to the bearings of the vibration average value occur. A bearing-induced vibration value at a feature ratio frequency which is a ratio of the frequency to the rotational frequency is extracted. Furthermore, the calculation unit 11 calculates an abnormality degree, which is an evaluation index, based on the bearing-induced vibration value, and determines whether the bearing 7 is normal. That is, the calculation unit 11 functions as frequency analysis means, vibration average value calculation means, bearing induced vibration value extraction means, evaluation index calculation means, and judgment means. The diagnosis result by the calculation unit 11 is displayed on the display unit 12.

In the bearing 7, when the inner ring flaw is present locally, the inner ring flaw rotates with the rotation of the main shaft 1, and the direction of the vibration generated when the rolling element passes the position of the scratch changes. for frequencies body passes, the vibration of only high frequency rotational frequency f _ROT only low frequency rotation frequency f _ROT is observed. The calculation of the feature frequency can be performed as in the following Equation 1 and Equation 2. Here, f _{I, N−} is the feature frequency on the lower side of the inner ring scratch N next, f _{I, N +} is the feature frequency on the higher side of the inner ring scratch N next, Z is the number of rolling elements of the bearing 7, D is the bearing 7 Is the diameter of the rolling element of the bearing 7, d is the diameter of the rolling element of the bearing 7, and α is the contact angle of the bearing 7.

FIG. 2 measures the vibration acceleration at a plurality of rotational frequencies for the main shaft 1 supported by the bearing 7 in which the inner ring flaw exists, calculates the vibration amplitude of the feature ratio frequency, and causes the bearing abnormality corresponding to each feature ratio frequency The angular velocity is plotted on the horizontal axis, with the vibrational force F due to a bearing abnormality calculated by dividing the transfer function to the vibration at the position of the vibration sensor 8 from the vibrational force.
However, the transfer function used here is not a transfer function from the excitation force due to the bearing abnormality to the vibration at the position of the vibration sensor 8, but measures the vibration at the position of the vibration sensor 8 when vibrating near the bearing. The transfer function obtained in is substituted. Here, in the vicinity of the bearing, when excited to the position of the vibration sensor 8 obtained by finite element analysis etc., the generation position of the exciting force due to the bearing abnormality and the direction and magnitude of the vibration at least in a frequency range , Represents the position that can be considered the same. According to the reciprocity theorem which becomes the same even if the input and output of the transfer function are interchanged, the transfer function obtained by exciting the vicinity of the bearing in a certain frequency range is from the excitation force due to the bearing abnormality to the vibration at the position of the vibration sensor 8 It is possible to substitute as a transfer function of

  From FIG. 2, it can be seen that the excitation force due to the bearing abnormality tends to increase as the angular velocity increases. In general, the excitation force due to the bearing abnormality is proportional to the square of the angular velocity. If there is an inner ring flaw, the main shaft 1 is considered to be displaced or deformed in the direction of the flaw when the rolling elements pass, but at this time, the locus (displacement) through which the center of gravity of the main spindle 1 passes is constant regardless of the rotational frequency If it can be considered, the force generated as a reaction that displaces the center of gravity of the main shaft 1 is proportional to the second derivative of the displacement of the center of gravity. Since the frequency of the vibration displacement of the center of gravity is a characteristic frequency that is proportional to the rotation frequency or the angular velocity, this force is proportional to the square of the rotation frequency when measuring the vibration. From these experimental results and hypotheses, it is presumed that the excitation force due to the inner ring injury approximates in proportion to the square of the rotational frequency.

As described above, when the relationship between the excitation force and the rotational frequency due to the bearing abnormality is clarified, it is possible to calculate a value (corrected vibration value) obtained by removing the influence of the rotational frequency from the measured value of vibration. By performing vibration measurement at a plurality of rotational frequencies, it is possible to realize processing for reducing the influence of the magnitude of the transfer function.
When measurement is performed at a plurality of rotational frequencies, it is difficult to discuss the feature frequency that changes depending on the rotational frequency, so the characteristic frequency divided by the rotational frequency (a characteristic ratio frequency) will be discussed. The feature ratio frequency k _{I, N−} of the Nth lower side of the inner ring injury and the feature ratio frequency k _{I, N +} of the next higher side of the inner ring injury can be determined as in the following Equation 3 and Equation 4 it can.

On the other hand, the rotational frequency of the excitation force due to bearing abnormality in characteristic ratio frequency k in the case of f _ROT magnitude F (k, f _ROT) is a constant F ^* that does not depend on the rotation frequency depends on the degree of damage of the bearing 7 It can be expressed as the following equation 5 using:

Assuming that the magnitude of the transfer function from the excitation force due to this bearing abnormality to the vibration at the position of the vibration sensor 8 is G (k, f _ROT ), the feature ratio when the rotational frequency measured by the vibration sensor 8 is f _ROT The magnitude A (k, f _ROT ) of the vibration at the frequency k can be expressed as the following equation 6. Note that the magnitude G (k, f _ROT ) of the transfer function takes other values when the rotational frequency is different even if the relative frequency is the same.

Therefore, the constant F ^* which is dependent on the magnitude A (k, f _ROT ) of the vibration measured by the vibration sensor 8 and the degree of damage to the bearing 7 has a relationship of the following equation 7.

In order to remove the influence of the change in magnitude of the excitation force due to the bearing abnormality due to the rotational frequency from the measured vibration, the magnitude A (k, f _ROT ) of the vibration at the feature ratio frequency k is set to 2 of the angular velocity 2πf _ROT The corrected vibration value A ^* (k, f _ROT ) may be obtained by the following equation 8 by dividing by the power.

FIG. 3 shows the magnitudes of the transfer functions of the three identical models, and the ratio of the magnitude to the maximum of the magnitudes of the transfer functions of the three models. Since the magnitude of the transfer function is largely different due to a slight difference in rotational frequency, it is obvious that if the feature frequency is changed by changing the rotational frequency to be measured, the magnitude of the vibration to be measured is largely changed. Furthermore, the machine with the smallest transfer function magnitude at a frequency is about 10% or less of the largest machine. As described above, the transfer function G (k, f _ROT ) varies even with machines having the same structure, and even with the same degree of abnormality, the vibration is completely different depending on the combination of the machine and the rotational frequency. It is difficult to set a threshold value to judge an abnormality because it may be measured. Also, the magnitude of the transfer function changes depending on the operating environment, operating conditions, etc. of the machine.
Then, if the magnitude A (k, f _ROT ) of the vibration at the feature ratio frequency k of Eq. 7 is replaced with the corrected oscillation value A ^* (k, f _ROT ) using Eq. 8, the following Eq. 9 is obtained.

The value obtained by averaging the rotational frequency of the N condition which is within the predetermined range for the equation (9) is obtained as the following equation (10). The left side of Expression 10 is an average of the corrected vibration values (vibration average value), and is also a bearing-induced vibration value at the feature ratio frequency k. The right side of Formula 10 is the product of the average of the transfer function magnitudes of the feature ratio frequency k corresponding to each of the plurality of rotational frequencies and the constant F ^* that does not depend on the rotational frequency depending on the degree of damage to the bearing.

FIG. 4 shows an average (average of vibration magnitudes) of transfer function magnitudes at respective fractional frequencies corresponding to each of a plurality of rotational frequencies for the three identical models described above. It can be seen that the average of the transfer function magnitudes remains almost unchanged at different fractional frequencies in a particular fractional frequency range. Further, the average of the transfer function magnitudes is small, and even if the average of the transfer function magnitudes is the smallest, the magnitude is about 70% of the largest machine.
The constant F ^{* that} is dependent on the degree of damage to the bearing and is not dependent on the rotational frequency is the same regardless of whether it is the number 9 or 10, and the proportional to the constant F ^* that is independent of the degree of damage on the bearing and independent of the rotational frequency Since the machine variation of the coefficient is small, calculating the evaluation index by the average of the corrected vibration values (value on the left side of the tens) is higher than calculating the evaluation index by the value on the left side of the equation 9 (corrected vibration value) It becomes diagnostic accuracy. Since the difference between the magnitude of the vibration between normal and abnormal is sufficiently large compared to the average machine variation of the transfer function, collecting a sufficient number of data of the same model results in the average of the corrected vibration values (10 It is easy to set a threshold for the evaluation index calculated from the value on the left side.

Furthermore, the magnitude of the corrected vibration value measured under the same conditions after M rotation of the bearing is A ^* '(k, f _ROT ), the transfer function at that time is G' (k, f _ROT ), the degree of damage to the bearing Assuming that the constant dependent and not dependent on the rotational frequency is F ^* ′, the following equation 11 holds similarly. Since the average of the transfer function magnitudes is also less affected by the operating environment and operating conditions of the machine, the following approximation of Equation 12 holds.

  The ratio of the growth rate R of vibration depending on the degree of damage of the bearing per one rotation of the bearing and the increment of the constant independent of the rotational frequency to the mean of the constant independent of the degree of damage of the bearing and independent of the rotational frequency is It sets like a number 13. Since Equation 13 does not include the magnitude of the transfer function, it is a comparable value even for different models.

Although F ^* and F ^* 'included in Eq. 13 can not be measured directly, Eq. 13 is rewritten as Eq. 14 using Eq. 8, Eq. 9, Eq. 10, Eq. 11, Eq. Be

The right side of Equation 14 includes only measurable or computable variables, and it is easy to estimate the growth rate R of the vibration. If it is difficult to collect a sufficient number of data of the same model, it is possible to detect an abnormality by determining the distribution of the growth rate R of vibration that does not include a variable dependent on the model and setting a threshold.
Although the growth rate R of vibration here is calculated as the ratio of the amount of increase in vibration to the average of vibration per one rotation of the bearing, the ratio of the average value of vibration in two measurements and the number of rotations of the bearing It is also possible to use another definition equation such as the following equation 15 using M, the number of rotations M ₀ of the reference bearing.

And FIG. 5 shows the flowchart of the method of performing abnormality diagnosis of the bearing 7, It demonstrates concretely based on this flowchart.
First, at S1, the rotational frequency of the main shaft 1 is changed to a rotational frequency for measuring vibration, and at S2, vibration acceleration is measured and stored by the vibration sensor 8 (S1, S2: vibration measuring step).
Next, in S3, a corrected vibration value is calculated from the rotational frequency and vibration acceleration at the time of vibration measurement, and in S4, Fourier transformation of the corrected vibration value is performed, and for each corrected frequency, the corrected vibration value whose frequency is within the predetermined range. Record the amplitude of (S3, S4: frequency analysis step).
Then, in S5, it is determined whether or not the vibration measurement at a plurality of rotational frequencies at which all the predetermined vibrations are measured has ended, and if the vibration measurement has ended, the process proceeds to S6, and if it has not ended It returns to S1 and repeats the process to S4.
Here, it is desirable that the plurality of rotational frequencies at which the vibration is measured be a combination of rotational frequencies such that any one of the rotational frequency, the reciprocal of the rotational frequency, and the logarithm of the rotational frequency is equally spaced. If the vibration average value is calculated using the magnitude of vibration measured with such a combination of rotational frequencies, machine variation of the transfer function magnitude and variation of the transfer function magnitude for each ratio frequency The reduction effect is improved.

In S6, the vibration average value is calculated for the value of the corrected vibration value for each of the recorded ratio frequencies (vibration average value calculation step), and in S7, the value of the corrected vibration value (bearing-induced vibration value) at the feature ratio frequency is extracted (Bearing-induced vibration value extraction step).
Next, in S8, it is determined whether or not to diagnose based on the growth rate of vibration (the degree of change in bearing-induced vibration value) which is selected and set in advance by the person in charge of diagnosis. Here, if the diagnosis is made based on the growth rate of vibration, the process proceeds to S9, and if the diagnosis is not made based on the growth rate of vibration, the process proceeds to S11.
In the case of diagnosis based on the growth rate of vibration, in S9, it is determined whether vibration measurement at the rotational frequency at which all vibrations are measured has been repeated twice. Here, if the vibration measurement is repeated twice, the process proceeds to S11. On the other hand, when the said vibration measurement is not repeated twice, it transfers to S10, implements the break-in operation until the bearing 7 rotates only the time or frequency | count set beforehand, and transfers to S1.

  Then, in S11, an evaluation index is calculated from the bearing-induced vibration value (evaluation index calculation step), and in S12, it is determined whether the evaluation index exceeds a threshold (determination step). Here, if the evaluation index exceeds the threshold, it is judged as abnormal in S13, and if the evaluation index does not exceed the threshold, it is judged as normal in S14 and shown in FIG. 6 or 7 in S15. Display diagnostic results like FIG. 6 is a diagnostic result screen when the growth rate of vibration is used as the evaluation index, and FIG. 7 is a diagnostic result screen when the maximum value of the bearing-induced vibration value is used as the evaluation index.

  As described above, according to the abnormality diagnosis method and the abnormality diagnosis apparatus for the rolling bearing of the above embodiment, and the abnormality diagnosis program, the vibration measurement step of measuring the vibration of the spindle 1 at a plurality of rotational frequencies of the spindle 1 Dividing by the square of the angular velocity proportional to the rotational frequency of the main spindle 1 and performing frequency analysis to calculate the magnitude of vibration for each frequency, and the main spindle 1 at the time of vibration measurement in the vibration measurement step The ratio of the frequency to the rotational frequency of the vibration average value calculation step of calculating the vibration average value obtained by averaging the magnitudes of vibrations having the same frequency ratio, and the ratio of the vibration average value to the rotational frequency of the frequency at which the vibration caused by the bearing 7 occurs A bearing-induced vibration value extraction step of extracting a bearing-induced vibration value at the feature ratio frequency, and an evaluation index calculation of calculating an evaluation index based on the bearing-induced vibration value In order to calculate the average after removing the influence of the rotational frequency from the measured vibration value by executing the step and the judgment step of judging the presence or absence of abnormality based on the evaluation index, the influence of machine variation To reduce the bearing-induced vibration value. Since the evaluation index calculated from the bearing-induced vibration value is also less affected by machine fluctuation, it is easy to set the threshold value to be judged as abnormal. That is, it is possible to determine the presence or absence of abnormality with high accuracy by calculating a robust evaluation index in consideration of the influence of the rotational frequency and the transfer function.

In particular, here, in the vibration measurement step, vibration measurement of the same combination of rotational frequencies is performed multiple times at intervals of the bearing rotation for a predetermined time or predetermined rotation so that the growth of the vibration can be clearly captured for measurement variation. In the evaluation index calculation step, the evaluation index is calculated based on the degree of change in bearing-induced vibration value (growth rate of vibration), so the value of the evaluation index depends on the value of the transfer function size Since the value is not set, it is possible to make a diagnosis using the same threshold even when diagnosing the main shaft 1 of a different structure or when measuring at different positions of the vibration sensor 8.
In addition, in the vibration average value calculation step, since the average is taken using only the vibration magnitude that is within the predetermined range of the vibration frequency, machine variation of the transfer function magnitude, and the transfer function The reduction effect of the variation of each ratio frequency of the size of is further improved.

In the above embodiment, in the calculation of the vibration value (corrected vibration value) considering the rotational frequency, division is performed by the square of the angular velocity, but any physical quantity such as the rotational frequency itself that is proportional to the rotational frequency is equivalent. It may be replaced because the effect of is obtained. Although the case where the value of the exponent of the power is 2 has been shown, replacing it with another numerical value (the value of the exponent is greater than 0) capable of achieving the purpose of reducing the effect of changing the excitation force by the rotation frequency Is possible.
Moreover, since the calculation of dividing by the square of the angular velocity is a process of dividing only by the scalar quantity of a constant, the frequency analysis step of the above embodiment is an example of calculating the corrected vibration value before the frequency analysis. Since the calculation result does not affect the calculation result until the vibration average value calculation step is started, for example, after calculating the acceleration in the frequency domain by Fourier transform, dividing by the square of the angular velocity, etc. Good.
On the other hand, when the evaluation index is calculated from the bearing-induced vibration value, the average value of the bearing-induced vibration values of each feature ratio frequency may be adopted, or the growth rate of the vibration of each feature ratio frequency is determined It may be a value or an average value. Furthermore, the value of the vibration average value other than the bearing-induced vibration value may also be determined by a complex function or the like that is referred to.

In addition, in the above-mentioned form, although bearing abnormalities by inner ring scratches were mentioned as an example and explained, it may apply to different bearing abnormalities, such as damage to a rolling element.
The present invention is also applicable to rolling bearings used in machines other than machine tools.
Furthermore, in addition to the form incorporated into a machine tool as an abnormality diagnosis apparatus, at least the storage unit, the calculation unit, and the display unit can communicate with the machine tool in a wired or wireless manner as a separate device from the machine tool. The abnormality diagnosis method may be executed based on the abnormality diagnosis program while measuring data and acquiring data. In this way, centralized control can be performed on a plurality of machine tools.

  1 · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · speed control device · · · · control device 7 · · · bearing 8 · · · vibration sensor 9 · · A / D converter, 10 · · · storage unit, 11 · · · operation unit, 12 · · display unit.

Claims (6)

A method of diagnosing an abnormality in a rolling bearing supporting a rotating body, the method comprising:
Measuring a vibration of the rotating body at a plurality of rotational frequencies of the rotating body;
A frequency analysis step of calculating the magnitude of vibration at each frequency by dividing the vibration by the power of an index larger than zero of a predetermined physical quantity proportional to the rotational frequency of the rotating body at the time of vibration measurement and performing frequency analysis When,
A vibration average value calculating step of calculating a vibration average value obtained by averaging the magnitudes of the vibrations having the same ratio of frequency to rotational frequency of the rotating body at the time of vibration measurement in the vibration measuring step;
A bearing-induced vibration value extraction step of extracting a bearing-induced vibration value at a feature ratio frequency that is a ratio of the frequency at which the vibration caused by the bearing occurs to the rotational frequency among the vibration average values;
An evaluation index calculation step of calculating an evaluation index based on the bearing-induced vibration value;
And a judging step of judging the presence or absence of an abnormality based on the evaluation index.   In the vibration average value calculation step, using the magnitude of the vibration measured at a combination of the rotation frequency such that any one of the rotation frequency, the reciprocal of the rotation frequency, and the logarithm of the rotation frequency is equally spaced. The method according to claim 1, wherein the vibration average value is calculated. In the vibration measurement step, vibration measurement of the combination having the same rotational frequency is performed a plurality of times,
The method according to claim 1 or 2, wherein in the evaluation index calculation step, the evaluation index is calculated based on the degree of change in the bearing-induced vibration value.   The abnormality diagnosis of the rolling bearing according to any one of claims 1 to 3, wherein in the vibration average value calculating step, only the magnitude of the vibration which is within a predetermined range is used to calculate the average. Method. An apparatus for diagnosing an abnormality in a rolling bearing supporting a rotating body, the apparatus comprising:
Vibration measuring means for measuring the vibration of the rotating body at a plurality of rotational frequencies of the rotating body;
Frequency analysis means for calculating the magnitude of vibration for each frequency by dividing the vibration by the power of an index larger than zero of a predetermined physical quantity proportional to the rotational frequency of the rotating body at the time of vibration measurement and performing frequency analysis When,
Vibration average value calculation means for calculating a vibration average value obtained by averaging the magnitudes of the vibrations having the same ratio of frequency to rotational frequency of the rotating body at the time of vibration measurement by the vibration measurement means;
A bearing-induced vibration value extraction unit that extracts a bearing-induced vibration value at a feature ratio frequency that is a ratio of the frequency at which the vibration caused by the bearing occurs to the rotational frequency among the vibration average values;
Evaluation index calculation means for calculating an evaluation index based on the bearing-induced vibration value;
And a judging means for judging the presence or absence of an abnormality based on the evaluation index.   A frequency analysis step in a method for diagnosing a rolling bearing abnormality according to any one of claims 1 to 4, in a computer to which vibration of a rotating body measured at each of a plurality of rotational frequencies is input together with each of the rotational frequencies, and vibration average A rolling bearing abnormality diagnosis program, characterized by executing a value calculation step, a bearing-induced vibration value extraction step, an evaluation index calculation step, and a determination step.

Images