Method and device for predictive diagnosis of a device including a rotating body and a bearing

Abstract

To provide a sign diagnosis method and apparatus which can monitor the tendency of degradation of a bearing even from a measurement value with a different rotation speed of the bearing.SOLUTION: A sign diagnosis method causes a control unit 11 to perform a fast Fourier transform treatment on a vibration waveform of acceleration data measured from a diagnosis object device by a sensor unit 10, detects a peak frequency of a frequency component with limitation to equal to or less than 60 Hz being a normal rotation speed range of an object bearing, calculates the measurement rotation speed and measurement acceleration speed from the peak frequency, calculates a relative ratio of the measurement acceleration speed difference to the relative comparison reference value as the relative determination value with the difference between the excellent reference value and the abnormal reference value determined in the relation between the bearing peripheral speed and the bearing acceleration speed of the bearing inner peripheral surface of the object bearing as the relative comparison reference value and with the difference between the excellent reference value and the measurement acceleration speed at each bearing peripheral speed as the measurement acceleration speed difference, and displays a bearing acceleration speed relative determination criterion graph on an output unit 12.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method and apparatus for predictive diagnosis of an apparatus including a rotating body and a bearing, and relates to a technique for diagnosing deterioration of a rotating body such as a bearing. [Background technology]

[0002] Conventionally, for example, Patent Document 1 describes a condition monitoring method for rotating parts incorporated in mechanical equipment, which identifies abnormal parts, determines the degree of damage or the progress of the damage, and predicts the remaining life of the abnormal parts.

[0003] In this method, a vibration waveform is detected by a vibration sensor fixed to the rolling bearing or housing, the detected waveform is divided into multiple damage filter frequency bands by a filter processing unit, and extracted, and spectrum data is calculated from the filtered waveform by an arithmetic processing unit.

[0004] The precision diagnosis unit then compares the bearing damage frequency calculated based on the rotational speed of the rolling bearing with the spectrum data obtained by the calculation processing unit to identify abnormal parts of the rolling bearing. The damage level diagnosis unit diagnoses the level of damage in the abnormal parts based on the vibration effective value calculated for each damage filter frequency band. The remaining life prediction unit predicts the remaining life of the abnormal parts based on the abnormal parts, the level of damage in the abnormal parts, and the operating environment of the rotating parts.

[0005] Furthermore, Patent Document 2 describes a method for determining the absolute value of a rolling bearing. This involves measuring acceleration G and determining the measured acceleration G based on a certain standard. The graph used for the determination shows the peripheral speed of the rolling elements of the rolling bearing, with the horizontal axis representing the dN value, which is the product of the bearing bore diameter and rotation speed. For example, if a certain type of bearing has an inner diameter of 60 mm, the dN value will be 9.0 x 10E4 at a rotation speed of 1500 rpm. If the measured acceleration G value is 1.0 G at this time, it will be determined to be "Caution." [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent Publication No. 2017-219469 [Patent Document 2] Patent Publication No. 2016-116251 Summary of the Invention [Problem to be solved by the invention]

[0007] General bearing deterioration diagnosis is performed using trend monitoring. This involves measuring the vibration of the bearing at a set measurement interval, calculating acceleration from the vibration data, and monitoring the progress of changes in acceleration. If the distance traveled by a vibrating object is considered to be displacement, the rate of change of displacement over time is velocity, and the rate of change of velocity over time is acceleration.

[0008] This vibration measurement must be performed under the same operating conditions and the measured values ​​must be compared.

[0009] However, the operating conditions of an air conditioner vary depending on the outside temperature and the number of people in the room, and the fan speed changes as the operating conditions change. For this reason, it is difficult to achieve the same operating conditions (speed) within a predetermined measurement cycle, so it is often not possible to compare the readings with the previous measurement.

[0010] Furthermore, to diagnose bearing deterioration through trend monitoring, it is necessary to determine a normal value for acceleration that will serve as the basis for diagnosis. For this reason, deterioration diagnosis is performed by comparing the measured value with the value during rated operation as the normal value, and making a judgment using the value obtained by adding an allowable value to the reference value as the threshold value.

[0011] However, when the design conditions and actual operating conditions of an air conditioner differ, it is necessary to determine a normal value of acceleration that matches the actual rated operating condition, and it is difficult to determine the normal value of acceleration for each operating condition.

[0012] Furthermore, in a diagnosis using the product of the bearing inner diameter and the rotation speed, i.e., the absolute value standard in the relationship between peripheral speed and acceleration, deterioration can be determined even if the reference value for comparison is unknown, and since the condition is diagnosed on a case-by-case basis during periodic measurements, the current result will be a judgment such as good, caution, or abnormal.

[0013] However, assuming a relationship between peripheral speed and acceleration, the measured value changes depending on the rotation speed at the time of measurement. Therefore, while it is possible to compare trends if the rotation speed is the same, if the rotation speed fluctuates, the acceleration changes, making it impossible to compare trends and making it difficult to monitor trends.

[0014] Precision diagnostic equipment that uses FFT (Fast Fourier Transform) analysis, envelope analysis, etc. requires high-precision vibration sensors and analyzers. Precision diagnostic equipment is not suitable for daily trend monitoring because the vibration sensors and analyzers are expensive and measurement is time-consuming, so it is mainly used to identify the cause when an abnormal trend is observed.

[0015] The present invention is devised to solve the above-mentioned problems, and aims to provide a method and device for predictive diagnosis of a device including a rotating body and a bearing, which is capable of monitoring the tendency of bearing deterioration even from measured values ​​that vary in the bearing rotation speed at the time of measurement. [Means for solving the problem]

[0016] In order to solve the above problems, the present invention provides a method for predictive diagnosis of a device including a rotating body and a bearing, in a data measurement step, measuring the acceleration occurring in the device to be diagnosed using an acceleration sensor, performing fast Fourier transform processing on the vibration waveform of the measured acceleration data to obtain frequency components, detecting peak frequencies of frequency components below 60 Hz, which is the normal rotation speed range of a target bearing installed in the device to be diagnosed, calculating the measured rotation speed of the target bearing from the detected peak frequency, and using the value of the acceleration data measured by the acceleration sensor as the measured acceleration, and in a judgment value calculation step, The bearing circumferential speed is the circumferential speed of the inner circumferential surface of the target bearing around the axis.and bearing acceleration, a good reference value and an abnormal reference value of the bearing acceleration are determined for each bearing circumferential speed, the difference between the good reference value and the abnormal reference value is set as a relative comparison reference value, the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measured rotation speed is set as a measured acceleration difference, the relative ratio of the measured acceleration difference to the relative comparison reference value is calculated as a relative judgment value, and in the data evaluation step, it is judged that the closer the relative judgment value is to a relative abnormal reference value corresponding to the abnormal reference value, the greater the sign of the occurrence of an abnormal event.

[0017] In the predictive diagnosis method for a device including a rotating body and a bearing of the present invention, in the judgment value calculation step, a bearing acceleration absolute value judgment reference graph is created with the bearing circumferential speed of the inner circumferential surface of the bearing of interest on the horizontal axis and the bearing acceleration on the vertical axis, a good reference value and an abnormal reference value of bearing acceleration are determined for each bearing circumferential speed on the bearing acceleration absolute value judgment reference graph, the measured acceleration at the bearing circumferential speed corresponding to the measured rotational speed is plotted on the bearing acceleration absolute value judgment reference graph, the difference between the good reference value and the abnormal reference value on the bearing acceleration absolute value judgment reference graph is found as a relative comparison reference value, and the good reference value at the bearing circumferential speed corresponding to the measured rotational speed is calculated. The difference between the acceleration and the measured acceleration is obtained as the measured acceleration difference, and the relative ratio of the measured acceleration difference to the relative comparison reference value is calculated as the relative judgment value. In the data evaluation step, a bearing acceleration relative judgment criterion graph is created with the bearing circumferential speed of the inner circumferential surface of the target bearing on the horizontal axis and the relative judgment value on the vertical axis, and a relative good reference value corresponding to the good reference value and a relative abnormal reference value corresponding to the abnormal reference value are set on the bearing acceleration relative judgment criterion graph, and the relative judgment values ​​are plotted on the bearing acceleration relative judgment criterion graph. It is determined that the closer the plotted relative judgment value is to the relative abnormal reference value, the greater the sign of an abnormal event occurring.

[0018] The predictive diagnosis method for a device including a rotating body and a bearing of the present invention includes, in a data measurement step, measuring the acceleration occurring in the device to be diagnosed using an acceleration sensor, performing fast Fourier transform processing on the vibration waveform of the measured acceleration data to obtain frequency components, detecting peak frequencies of frequency components of 60 Hz or less, which is the normal rotation speed range of a target bearing installed in the device to be diagnosed, calculating the measured rotation speed of the target bearing from the detected peak frequency, and using the value of the acceleration data measured by the acceleration sensor as the measured acceleration, and in a judgment value calculation step, The bearing circumferential speed is the circumferential speed of the inner circumferential surface of the target bearing around the axis. and bearing acceleration, a good reference value and an abnormal reference value of the bearing acceleration are determined for each bearing circumferential speed, the difference between the good reference value and the abnormal reference value is set as a relative comparison reference value, the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measured rotation speed is set as a measured acceleration difference, and the relative ratio of the measured acceleration difference to the relative comparison reference value is calculated as a relative judgment value, and in the data evaluation step, a time series of the relative judgment values ​​is formed, and it is judged that the closer the trend of fluctuation in the relative judgment values ​​in the time series approaches the relative abnormal reference value corresponding to the abnormal reference value, the greater the sign of the occurrence of an abnormal event.

[0019] In the predictive diagnosis method for a device including a rotating body and a bearing of the present invention, in the judgment value calculation step, a bearing acceleration absolute value judgment criterion graph is created with the bearing circumferential speed of the inner circumferential surface of the bearing of interest on the horizontal axis and the bearing acceleration on the vertical axis, a good reference value and an abnormal reference value of bearing acceleration are determined for each bearing circumferential speed on the bearing acceleration absolute value judgment criterion graph, the measured acceleration at the bearing circumferential speed corresponding to the measured rotation speed is plotted on the bearing acceleration absolute value judgment criterion graph, the difference between the good reference value and the abnormal reference value on the bearing acceleration absolute value judgment criterion graph is found as a relative comparison criterion value, and the good reference value at the bearing circumferential speed corresponding to the measured rotation speed is calculated. The difference between the acceleration and the velocity is obtained as the measured acceleration difference, and the relative ratio of the measured acceleration difference to the relative comparison reference value is calculated as the relative judgment value. In the data evaluation step, a graph of acceleration change over time is created with time on the horizontal axis and the relative judgment value on the vertical axis. A relative good reference value corresponding to the good reference value and a relative abnormal reference value corresponding to the abnormal reference value are set on the acceleration change over time graph, and the relative judgment values ​​are plotted on the acceleration change over time graph in the order of measurement to form a trajectory consisting of a time series of the relative judgment values. It is determined that the closer the trend of change in the trajectory of the relative judgment value approaches the relative abnormal reference value, the greater the sign of an abnormal event occurring.

[0020] The predictive diagnostic device for a device including a rotating body and a bearing of the present invention comprises an output section, a control section, and a sensor section, the sensor section having an acceleration sensor that measures acceleration occurring in the device to be diagnosed and transmits the measured acceleration data to the control section, the control section having a fast Fourier transform section that performs fast Fourier transform on the vibration waveform of the measured acceleration data to determine frequency components, a peak frequency detection section that detects, from the determined frequency components, peak frequencies of frequency components below 60 Hz that is the normal rotation speed range of a target bearing installed in the device to be diagnosed, a measurement value calculation section that calculates the measured rotation speed of the target bearing from the detected peak frequency and uses the value of the acceleration data measured by the acceleration sensor as the measured acceleration, and a relative judgment value calculation section, The bearing circumferential speed is the circumferential speed of the inner circumferential surface of the target bearing around the axis.and bearing acceleration, a good reference value and an abnormal reference value of the bearing acceleration are determined for each bearing circumferential speed, the difference between the good reference value and the abnormal reference value is set as a relative comparison reference value, the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measured rotation speed is set as a measured acceleration difference, and the relative ratio of the measured acceleration difference to the relative comparison reference value is calculated as a relative judgment value, and the output unit relatively displays the relative good reference value corresponding to the good reference value, the relative abnormal reference value corresponding to the abnormal reference value, and the relative judgment value in terms of their positional relationship on a graph, and the closer the relative judgment value is to the relative abnormal reference value, the greater the likelihood of an abnormal event occurring.

[0021] The predictive diagnostic device for a device including a rotating body and a bearing of the present invention comprises an output section, a control section, and a sensor section, the sensor section having an acceleration sensor that measures acceleration occurring in the device to be diagnosed and transmits the measured acceleration data to the control section, the control section having a fast Fourier transform section that performs fast Fourier transform on the vibration waveform of the measured acceleration data to determine frequency components, a peak frequency detection section that detects, from the determined frequency components, peak frequencies of frequency components below 60 Hz that is the normal rotation speed range of a target bearing installed in the device to be diagnosed, a measurement value calculation section that calculates the measured rotation speed of the target bearing from the detected peak frequency and uses the value of the acceleration data measured by the acceleration sensor as the measured acceleration, and a relative judgment value calculation section, The bearing circumferential speed is the circumferential speed of the inner circumferential surface of the target bearing around the axis. and bearing acceleration, a good reference value and an abnormal reference value of the bearing acceleration are determined for each bearing circumferential speed, the difference between the good reference value and the abnormal reference value is set as a relative comparison reference value, the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measured rotational speed is set as a measured acceleration difference, and the relative ratio of the measured acceleration difference to the relative comparison reference value is calculated as a relative judgment value, and the output unit forms a time series of the relative judgment values ​​on a graph that displays the relative good reference value corresponding to the good reference value and the relative abnormal reference value corresponding to the abnormal reference value, and the closer the trend of change in the relative judgment value in the time series is to the relative abnormal reference value corresponding to the abnormal reference value, the greater the sign of an abnormal event that will occur.

[0022] In the predictive diagnostic device for an apparatus including a rotating body and a bearing of the present invention, the output unit issues an alarm when the relative determination value exceeds the set threshold value and approaches the relative abnormality reference value. [Effects of the Invention]

[0023] As described above, according to the present invention, the rotation speed of the bearing of the equipment to be diagnosed is estimated in the data measurement process. That is, the acquired acceleration data is subjected to fast Fourier transform to obtain frequency components, and by detecting the peak frequency within the range of rotation speeds commonly used in general air conditioners, which is 60 Hz or less, the measurement value is no longer affected by the value of high-frequency components that become noise, and the rotation speed can be estimated appropriately.

[0024] Then, in the judgment value calculation step, the difference between the good reference value and the abnormal reference value determined in relation to the bearing circumferential speed of the inner peripheral surface of the target bearing and the bearing acceleration is used as a relative comparison reference value, and a relative judgment value of a relative ratio is calculated. Thus, the absolute value of the measured bearing acceleration can be expressed as a relative numerical level, i.e., as a relative positional relationship with the good reference value and the abnormal reference value, which are judgment standards in absolute values.

[0025] For this reason, in the data evaluation process, it is possible to monitor the degree of deterioration for each measurement using only the relative judgment value, regardless of the bearing rotation speed at each measurement. Also, by comparing relative judgment values ​​for different bearing rotation speeds at each measurement, it is possible to grasp signs of failure such as an upward trend, and it is possible to determine that the closer the relative judgment value is to a relative abnormality reference value corresponding to the abnormality reference value, or the closer the fluctuation trend of the relative judgment value is to the relative abnormality reference value when a time series of relative judgment values ​​is formed, the greater the likelihood of an abnormal event occurring. [Brief explanation of the drawings]

[0026] [Figure 1] FIG. 1 is a schematic diagram showing a predictive diagnostics device for a device including a rotating body and a bearing according to an embodiment of the present invention; [Figure 2] FIG. 10 is a graph showing a bearing acceleration absolute value judgment criterion in the method for predictive diagnosis of a device including a rotating body and a bearing according to an embodiment of the present invention. [Figure 3] FIG. 10 is a graph showing a bearing acceleration relative criteria graph in the predictive diagnosis method for a device including a rotating body and a bearing according to the embodiment; [Figure 4] FIG. 10 is a graph showing a change in acceleration over time in the predictive diagnosis method for a device including a rotating body and a bearing according to the embodiment; DETAILED DESCRIPTION OF THE INVENTION

[0027] Hereinafter, an embodiment of the present invention will be described.

[0028] First, the method for predictive diagnosis of a device including a rotating body and a bearing according to the present invention will be described using an air conditioner as the device to be diagnosed. Example 1 1. Data measurement process The acceleration occurring in the air conditioner of the equipment to be diagnosed is measured using an acceleration sensor. The vibration waveform of the measured acceleration data is subjected to fast Fourier transform to determine the frequency components. Of the frequency components obtained, the peak frequency of frequency components below 60 Hz, which is the normal rotation speed range of the target bearing installed in the equipment to be diagnosed, is detected. The measured rotation speed of the target bearing of the air conditioner is calculated from the detected peak frequency, and the value of the acceleration data measured by the acceleration sensor is taken as the measured acceleration. In other words, if the total amplitude in the vibration waveform of the measured acceleration data is considered to be displacement, the rate of change of displacement over time is velocity, and the rate of change of velocity over time is acceleration.

[0029] In this data measurement process, the frequency components obtained by fast Fourier transform processing are limited to those below 60 Hz, which is the normal rotation speed range for a typical air conditioner, and peak frequencies are detected.This prevents the measured values ​​from being affected by the values ​​of high-frequency components that become noise, allowing for an appropriate estimation of the rotation speed. 2. Judgment value calculation process As shown in Figure 2, a bearing acceleration absolute value judgment criterion graph is created with the product of the bearing bore diameter and the bearing rotation speed of the target bearing, i.e., the bearing circumferential speed of the inner circumferential surface of the target bearing, on the horizontal axis and bearing acceleration G on the vertical axis. In this bearing acceleration absolute value judgment criterion graph, good reference values, abnormal reference values, and caution reference values ​​for bearing acceleration are set for each bearing circumferential speed in the relationship between the bearing circumferential speed of the inner circumferential surface of the target bearing and the bearing acceleration. Here, the locus of each good reference value is displayed as good level A1, the locus of abnormal reference values ​​as abnormal level B1, and the locus of caution reference values ​​as caution level C1.

[0030] On this bearing acceleration absolute value judgment reference graph, the measured acceleration is plotted at the bearing peripheral speed corresponding to the measured rotational speed calculated in the data measurement step. Here, as an example, the plot of an abnormal bearing determined to be in an abnormal state is indicated by a cross, the plot of a bearing determined to be in a state requiring attention is indicated by a square, the plot of a good bearing determined to be in a good state is indicated by a triangle, and the plot of a new bearing is indicated by a circle.

[0031] In this bearing acceleration absolute value judgment criterion graph, it is meaningless to compare the values ​​of each plot displayed with each other. That is, the bearing rotation speed is different for each plot, and the plots are not arranged in measurement order. Therefore, the bearing acceleration absolute value judgment criterion graph cannot be used to monitor the trend of bearing deterioration.

[0032] For this reason, the difference between the good reference value and the abnormal reference value on the bearing acceleration absolute value judgment reference graph is obtained as the relative comparison reference value, the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measured rotation speed is obtained as the measured acceleration difference, and the relative proportion of the measured acceleration difference to the relative comparison reference value is calculated as a percentage as the relative judgment value.

[0033] Therefore, the absolute value of the measured bearing acceleration can be expressed as a relative numerical level, that is, as a relative positional relationship to the good reference value and the abnormal reference value, which are the judgment criteria in absolute values. 3. Data evaluation process As shown in Figure 3, a bearing acceleration relative judgment criterion graph is created with the bearing circumferential speed of the inner circumferential surface of the target bearing (the product of the bearing bore diameter and the bearing rotation speed of the target bearing) on ​​the horizontal axis and the relative judgment value on the vertical axis. On this bearing acceleration relative judgment criterion graph, the locus of the relative good reference value corresponding to the good reference value is set as good level A2 with a relative judgment value of 0%, the locus of the relative abnormal reference value corresponding to the abnormal reference value is set as abnormal level B2 with a relative judgment value of 100%, and the locus of the relative caution reference value corresponding to the caution reference value is set as caution level C2 with a relative judgment value of 50%.

[0034] The relative judgment values ​​calculated in the judgment value calculation step are plotted on this bearing acceleration relative judgment criterion graph. The bearing acceleration relative judgment criterion graph does not display the measured acceleration as an absolute value, but as a relative judgment value expressed as a percentage, so it is possible to monitor the degree of deterioration for each measurement using only the relative judgment value, regardless of the bearing rotation speed at each measurement.

[0035] Furthermore, by comparing the relative judgment values ​​for which the bearing rotation speeds differ during each measurement, it is possible to identify signs of failure, such as an upward trend, and the closer the relative judgment value is to the relative abnormality reference value corresponding to the abnormality reference value, or the closer a time series of relative judgment values ​​is formed and the tendency of the relative judgment value to fluctuate, the greater the likelihood of an abnormal event occurring. Example 2 The data measurement step and the judgment value calculation step are the same as those in the first embodiment, and therefore the description thereof will be omitted. Here, the data evaluation step will be described.

[0036] Data evaluation process As shown in Figure 4, a graph of acceleration change over time is created with time on the horizontal axis and the relative judgment value on the vertical axis. In this graph of acceleration change over time, the trajectory of the relative good reference value corresponding to the good reference value is set as good level A3 with a relative judgment value of 0%, the trajectory of the relative abnormal reference value corresponding to the abnormal reference value is set as abnormal level B3 with a relative judgment value of 100%, and the trajectory of the relative caution reference value corresponding to the caution reference value is set as caution level C3 with a relative judgment value of 50%.

[0037] The relative judgment values ​​are plotted in the order of measurement on this graph of acceleration change over time to form a trajectory T consisting of a time series of relative judgment values, and it is determined that the closer the trend of fluctuation in the trajectory T of the relative judgment values ​​approaches the abnormal level B3 of the relative abnormality reference value, the greater the signs of an abnormal event occurring.

[0038] The predictive diagnostic device for a device including a rotating body and a bearing according to the present invention will be described below using an air conditioner as the device to be diagnosed. Example 3 As shown in Figure 1, air conditioner 1 has multiple chambers inside casing 2, and a blower 3 is placed in one of these chambers. Blower 3 consists of a drive motor 5 and a fan unit 6 installed on a stand 4, and a rotating shaft 7 of fan unit 6 is supported by a bearing (not shown) installed inside the exterior 8 of drive motor 5.

[0039] The predictive diagnosis device 9 according to this embodiment comprises a sensor unit 10, a control unit 11, and an output 12. The sensor unit 10, the control unit 11, and the output 12 are connected by wireless communication, wired communication, or the like, and the control unit 11 and the output 12 can also be configured as an integrated unit.

[0040] The sensor unit 10 is attached to the exterior 8 of the drive motor 5 and is installed at a position corresponding to the internal bearing. In this example, the sensor unit 10 is made up of an acceleration sensor that measures the acceleration occurring in the air conditioner 1, and transmits the acceleration data occurring in the exterior 8 of the drive motor 5 measured by the acceleration sensor to the control unit.

[0041] Here, control unit 11 is mounted on casing 2 of air conditioner 1, but there are no restrictions on where it can be installed. Control unit 11 has a fast Fourier transform unit that performs fast Fourier transform on the vibration waveform of the acceleration data received from sensor unit 10 to determine the frequency components, a peak frequency detection unit that detects, from the determined frequency components, the peak frequencies of frequency components below 60 Hz, which is the normal rotation speed range of the target bearing installed in the device to be diagnosed, a measurement value calculation unit that calculates the measured rotation speed of the target bearing from the detected peak frequency and uses the value of the acceleration data measured by the acceleration sensor as the measured acceleration, and a relative judgment value calculation unit.

[0042] The relative judgment value calculation unit determines a good reference value and an abnormal reference value for bearing acceleration for each bearing circumferential speed in the relationship between the bearing circumferential speed and bearing acceleration of the bearing inner circumferential surface of the target bearing, sets the difference between the good reference value and the abnormal reference value as the relative comparison reference value, sets the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measured rotational speed as the measured acceleration difference, and calculates the relative ratio of the measured acceleration difference to the relative comparison reference value as the relative judgment value. This calculation follows the same procedure as the judgment value calculation step described above in Example 1.

[0043] The output unit 12 displays the relative good reference value corresponding to the good reference value, the relative abnormal reference value corresponding to the abnormal reference value, and the relative judgment value relative to each other in terms of their positional relationship on a graph displayed on the display. This display follows the same procedure as the data evaluation step described in the first embodiment, and displays the bearing acceleration relative judgment reference graph shown in Figure 3 on the display.

[0044] This bearing acceleration relative judgment standard graph does not display the measured acceleration as an absolute value, but as a relative judgment value expressed as a percentage, so that the degree of deterioration can be monitored for each measurement using only the relative judgment value, regardless of the bearing rotation speed at each measurement.

[0045] Furthermore, by comparing the relative judgment values ​​for which the bearing rotation speeds differ during each measurement, it is possible to identify signs of failure, such as an upward trend, and the closer the relative judgment value is to the relative abnormality reference value corresponding to the abnormality reference value, or the closer a time series of relative judgment values ​​is formed and the tendency of the relative judgment value to fluctuate, the greater the likelihood of an abnormal event occurring.

[0046] Then, the output unit 12 issues an alarm when the relative determination value exceeds the set threshold value and approaches the relative abnormality reference value. Example 4 The sensor unit 10 and the control unit 11 have the same configuration as in the previous third embodiment, and therefore a description thereof will be omitted.

[0047] In the fourth embodiment, the content displayed on the output unit 12 is different. That is, the output unit 12 forms and displays a time series of relative judgment values ​​on a graph that displays a relative good reference value corresponding to the good reference value and a relative abnormal reference value corresponding to the abnormal reference value. This display follows the same procedure as the data evaluation step described in the second embodiment, and displays the graph of acceleration change over time shown in Fig. 4. The relative judgment values ​​are plotted in the order of measurement on this graph of acceleration change over time to form a trajectory T consisting of a time series of relative judgment values, and indicates that the closer the trend of fluctuation in the relative judgment values ​​in the time series is to the relative abnormal reference value corresponding to the abnormal reference value, the greater the sign of the occurrence of an abnormal event.

[0048] Furthermore, the output unit 12 issues an alarm when the relative determination value exceeds the set threshold value and approaches the relative abnormality reference value. [Explanation of symbols]

[0049] 1. Air conditioner 2 Casing 3. Blower 4 Mounting stand 5 Drive motor 6 Fan unit 7 Rotation Axis 8 Exterior 9. Predictive diagnostic equipment 10 Sensor section 11 Control section 12 Output

Claims

1. In the data measurement step, the acceleration occurring in the device to be diagnosed is measured by an acceleration sensor, the vibration waveform of the measured acceleration data is subjected to fast Fourier transform processing to determine frequency components, the peak frequency of frequency components below 60 Hz, which is the normal rotation speed range of the target bearing installed in the device to be diagnosed, is detected from the determined frequency components, the measured rotation speed of the target bearing is calculated from the detected peak frequency, and the value of the acceleration data measured by the acceleration sensor is used as the measured acceleration, In the judgment value calculation step, a good reference value and an abnormal reference value of the bearing acceleration are determined for each bearing circumferential speed in relation to the bearing acceleration and the bearing circumferential speed, which is the circumferential speed around the axis of the inner circumferential surface of the target bearing, the difference between the good reference value and the abnormal reference value is set as a relative comparison reference value, the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measurement rotation speed is set as a measured acceleration difference, and the relative ratio of the measured acceleration difference to the relative comparison reference value is calculated as a relative judgment value, A method for predictive diagnosis of an apparatus including a rotating body and a bearing, characterized in that in a data evaluation step, it is determined that the closer the relative judgment value is to a relative abnormality reference value corresponding to the abnormality reference value, the greater the likelihood of an abnormal event occurring.

2. In the judgment value calculation step, a bearing acceleration absolute value judgment reference graph is created with the bearing circumferential speed of the inner circumferential surface of the target bearing as the horizontal axis and the bearing acceleration as the vertical axis, a good reference value and an abnormal reference value of the bearing acceleration are determined for each bearing circumferential speed in the bearing acceleration absolute value judgment reference graph, and the measured acceleration at the bearing circumferential speed corresponding to the measured rotation speed is plotted on the bearing acceleration absolute value judgment reference graph, The difference between the good reference value and the abnormal reference value on the bearing acceleration absolute value judgment reference graph is obtained as a relative comparison reference value, the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measured rotation speed is obtained as a measured acceleration difference, and the relative ratio of the measured acceleration difference to the relative comparison reference value is calculated as a relative judgment value, In the data evaluation step, a bearing acceleration relative judgment criterion graph is created with the bearing circumferential speed of the inner peripheral surface of the target bearing as the horizontal axis and the relative judgment value as the vertical axis, and a relative good reference value corresponding to the good reference value and a relative abnormal reference value corresponding to the abnormal reference value are set in the bearing acceleration relative judgment criterion graph, 2. The method for predictive diagnosis of an apparatus including a rotating body and a bearing according to claim 1, wherein the relative judgment value is plotted on a bearing acceleration relative judgment criterion graph, and the closer the plotted relative judgment value is to the relative abnormality criterion value, the greater the likelihood of an abnormal event occurring.

3. In the data measurement step, the acceleration occurring in the device to be diagnosed is measured by an acceleration sensor, the vibration waveform of the measured acceleration data is subjected to fast Fourier transform processing to determine frequency components, the peak frequency of frequency components below 60 Hz, which is the normal rotation speed range of the target bearing installed in the device to be diagnosed, is detected from the determined frequency components, the measured rotation speed of the target bearing is calculated from the detected peak frequency, and the value of the acceleration data measured by the acceleration sensor is used as the measured acceleration, In the judgment value calculation step, a good reference value and an abnormal reference value of the bearing acceleration are determined for each bearing circumferential speed in relation to the bearing acceleration and the bearing circumferential speed, which is the circumferential speed around the axis of the inner circumferential surface of the target bearing, the difference between the good reference value and the abnormal reference value is set as a relative comparison reference value, the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measurement rotation speed is set as a measured acceleration difference, and the relative ratio of the measured acceleration difference to the relative comparison reference value is calculated as a relative judgment value, A method for predictive diagnosis of an apparatus including a rotating body and a bearing, characterized in that in a data evaluation step, a time series of relative judgment values ​​is formed, and it is determined that the greater the tendency of fluctuations in the time series of the relative judgment values ​​approaches a relative abnormality reference value corresponding to the abnormality reference value, the greater the likelihood of an abnormal event occurring.

4. In the judgment value calculation step, a bearing acceleration absolute value judgment reference graph is created with the bearing circumferential speed of the inner circumferential surface of the target bearing as the horizontal axis and the bearing acceleration as the vertical axis, a good reference value and an abnormal reference value of the bearing acceleration are determined for each bearing circumferential speed in the bearing acceleration absolute value judgment reference graph, and the measured acceleration at the bearing circumferential speed corresponding to the measured rotation speed is plotted on the bearing acceleration absolute value judgment reference graph, The difference between the good reference value and the abnormal reference value on the bearing acceleration absolute value judgment reference graph is obtained as a relative comparison reference value, the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measured rotation speed is obtained as a measured acceleration difference, and the relative ratio of the measured acceleration difference to the relative comparison reference value is calculated as a relative judgment value, In the data evaluation step, a graph of acceleration change over time is created with time as the horizontal axis and the relative judgment value as the vertical axis, and a relative good reference value corresponding to the good reference value and a relative abnormal reference value corresponding to the abnormal reference value are set in the graph of acceleration change over time, 4. The method for predictive diagnosis of an apparatus including a rotating body and a bearing according to claim 3, further comprising: plotting the relative judgment values ​​in the order of measurement on a graph of change in acceleration over time to form a trajectory consisting of a time series of the relative judgment values; and determining that the closer the tendency of fluctuations in the trajectory of the relative judgment values ​​is to the relative abnormality reference value, the greater the likelihood of an abnormal event occurring.

5. An output unit, a control unit, and a sensor unit are provided, the sensor unit has an acceleration sensor that measures acceleration occurring in the diagnosis target device, and transmits the measured acceleration data to the control unit; The control unit has a fast Fourier transform unit that performs fast Fourier transform on the vibration waveform of the measured acceleration data to obtain frequency components, a peak frequency detection unit that detects, from the obtained frequency components, peak frequencies of frequency components below 60 Hz that is the normal rotation speed range of the target bearing installed in the device to be diagnosed, a measurement value calculation unit that calculates the measured rotation speed of the target bearing from the detected peak frequency and sets the value of the acceleration data measured by the acceleration sensor as the measured acceleration, and a relative judgment value calculation unit. the relative judgment value calculation unit determines a good reference value and an abnormal reference value for bearing acceleration for each bearing circumferential speed in the relationship between the bearing circumferential speed, which is the circumferential speed around the axis of the inner circumferential surface of the target bearing, and the bearing acceleration, sets the difference between the good reference value and the abnormal reference value as a relative comparison reference value, sets the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measurement rotation speed as a measured acceleration difference, and calculates the relative ratio of the measured acceleration difference to the relative comparison reference value as a relative judgment value; The output unit displays a relative good reference value corresponding to the good reference value, a relative abnormal reference value corresponding to the abnormal reference value, and the relative judgment value relative to each other in terms of their positional relationship on a graph, and indicates that the closer the relative judgment value is to the relative abnormal reference value, the greater the likelihood of an abnormal event occurring.

6. An output unit, a control unit, and a sensor unit are provided, the sensor unit has an acceleration sensor that measures acceleration occurring in the diagnosis target device, and transmits the measured acceleration data to the control unit; The control unit has a fast Fourier transform unit that performs fast Fourier transform on the vibration waveform of the measured acceleration data to obtain frequency components, a peak frequency detection unit that detects, from the obtained frequency components, peak frequencies of frequency components below 60 Hz that is the normal rotation speed range of the target bearing installed in the device to be diagnosed, a measurement value calculation unit that calculates the measured rotation speed of the target bearing from the detected peak frequency and sets the value of the acceleration data measured by the acceleration sensor as the measured acceleration, and a relative judgment value calculation unit. the relative judgment value calculation unit determines a good reference value and an abnormal reference value for bearing acceleration for each bearing circumferential speed in the relationship between the bearing circumferential speed, which is the circumferential speed around the axis of the inner circumferential surface of the target bearing, and the bearing acceleration, sets the difference between the good reference value and the abnormal reference value as a relative comparison reference value, sets the difference between the good reference value and the measured acceleration at the bearing circumferential speed corresponding to the measurement rotation speed as a measured acceleration difference, and calculates the relative ratio of the measured acceleration difference to the relative comparison reference value as a relative judgment value; The output unit forms a time series of relative judgment values ​​on a graph that displays a relative good reference value corresponding to the good reference value and a relative abnormal reference value corresponding to the abnormal reference value, and indicates that the closer the trend of fluctuations in the relative judgment values ​​over time approaches the relative abnormal reference value corresponding to the abnormal reference value, the greater the likelihood of an abnormal event occurring.

7. 7. The predictive diagnostic device for a device including a rotating body and a bearing according to claim 5, wherein the output unit issues an alarm when the relative determination value exceeds a set threshold value and approaches a relative abnormality reference value.

Images