JP7367535B2 - Diagnostic method and device for rotating bearings - Google Patents

Description

The present invention relates to a method and apparatus for measuring vibration waves and AE waves caused by vibrations of a rotating bearing and diagnosing the rotating bearing based on the measurement results.

AE (acoustic emission) is an elastic vibration that occurs when a material deforms or breaks, as a result of releasing the strain energy that had been stored inside. Conventionally, diagnosis of rotating bearings has been performed using vibration sensors or AE sensors, but in most cases, due to cost considerations, a sensor that can only measure either vibration waves or AE waves is used for rotating bearings. This type of device is known, for example, as described in Patent Document 1.

This conventional device uses a vibration sensor and an AE sensor depending on the purpose and processes the signals of each sensor to diagnose abnormal conditions. Therefore, in this device, if an attempt is made to measure both vibration waves and AE waves, a plurality of sensors will be installed on the bearing. Furthermore, a method is used in which a sensor signal is subjected to envelope processing using a low-pass filter, and the generated envelope waveform is further processed with a high-pass filter to remove a DC component therefrom.

Japanese Patent Application Publication No. 2011-154020

However, since AE waves are generated at a high frequency around 100 kHz, high-speed sampling is required, and when measuring equipment that requires long-term measurements such as low-speed rotating machinery, the sampled signal becomes large in volume, making it difficult to use conventional methods. The problem with this device is that recording and processing costs are high. Furthermore, since conventional devices perform envelope processing on signals sampled at high speed, there is a problem in that minute frequency changes cannot be captured.

Furthermore, when measuring both AE waves and vibration waves, with conventional equipment it is necessary to install both the AE sensor and the vibration sensor on a bearing, and since there is not enough space for installation with a small bearing, it is necessary to install both. The problem was that it was not possible. In addition, two sensors, an AE sensor and a vibration sensor, and cables for them are required, resulting in an increase in cost.

By the way, the present inventor measured the AE waves generated by the AE generator using a vibration sensor and an AE sensor, and found that the vibration sensor also measured not only low-frequency vibration waves but also high-frequency AE waves. We found that this is possible.

Therefore, an object of the present invention is to provide a diagnostic method and device for a rotary bearing that enables low-speed sampling of AE waves and capture of minute signal changes, and furthermore, based on the above knowledge, An object of the present invention is to provide a method and device for diagnosing a rotating bearing, in which both vibration waves and vibration waves are measured with one sensor.

The rotating bearing diagnosis method of the present invention, which aims to advantageously solve the above problems, includes:
A method for diagnosing the condition of a rotating bearing,
A sensor directly attached to the rotary bearing measures the AE wave of the rotary bearing and outputs a sensor signal indicating the AE wave,
extracting high frequency components in a predetermined frequency range from the sensor signal;
Frequency modulating the high frequency component at a predetermined reference frequency and converting it into an AE low frequency signal,
sampling and recording the AE low frequency signal at predetermined time intervals;
The apparatus is characterized in that the state of the rotary bearing is determined by reading and analyzing the recorded AE low frequency signal, and the determination result is output.

In the rotary bearing diagnosing method of the present invention, further:
The sensor measures vibration waves of the rotation bearing and also outputs a sensor signal indicating the vibration waves;
extracting a low frequency component in a predetermined frequency range from the sensor signal as a vibration low frequency signal;
sampling and recording the vibration low frequency signal at predetermined time intervals;
In addition to reading and analyzing the recorded AE low frequency signal, the state of the rotary bearing is determined by reading and analyzing the recorded vibration low frequency signal, and the determination result is output.

Further, the rotating bearing diagnostic device of the present invention, which aims to advantageously solve the above problems, includes:
A device for diagnosing the condition of a rotating bearing,
a sensor that is directly attached to the rotary bearing and measures the AE wave of the rotary bearing and outputs a sensor signal indicative of the AE wave;
a high-pass filter that extracts high frequency components in a predetermined frequency range from the sensor signal;
an AE converter that frequency modulates the high frequency component at a predetermined reference frequency and converts it into an AE low frequency signal;
an AE recording unit that samples and records the AE low frequency signal at predetermined time intervals;
a diagnosis unit that determines the state of the rotary bearing by reading and analyzing the AE low frequency signal recorded in the AE recording unit, and outputs the determination result;
It is characterized by having the following.

In the rotary bearing diagnostic device of the present invention, further:
the sensor that is directly attached to the rotary bearing and also measures vibration waves of the rotary bearing and outputs a sensor signal indicative of the vibration waves;
a low-pass filter that extracts low-frequency components in a predetermined frequency range from the sensor signal to generate a vibration low-frequency signal;
a vibration recording unit that samples and records the vibration low frequency signal at predetermined time intervals;
In addition to reading and analyzing the AE low frequency signal recorded in the AE recording unit, the state of the rotation bearing is determined by reading and analyzing the vibration low frequency signal recorded in the vibration recording unit, and the determination result is output. Diagnostic department and
It is equipped with

According to the rotary bearing diagnostic method and device of the present invention, a high frequency component in a predetermined frequency range is extracted from a sensor signal indicating an AE wave measured and output by a sensor directly attached to the rotary bearing, and the high frequency component is converted into a predetermined frequency component. Frequency modulation is performed using a reference frequency and converted to an AE low frequency signal, the AE low frequency signal is sampled and recorded, and the recorded AE low frequency signal is read out and analyzed to detect fatigue wear, scratches, and foreign objects in the rotating bearing. Since abnormal conditions such as contamination and/or poor lubrication are determined and the determination results are output, the low-speed sampling of the AE wave allows for the measurement of sampled signals even when measuring equipment that requires long-term measurement, such as low-speed rotating machinery. Since the capacity is suppressed, recording and processing costs can be reduced, and since signals sampled at low speed can be subjected to envelope processing, even minute frequency changes can be captured.

Furthermore, according to the method and device for diagnosing a rotating bearing of the present invention, the sensor also measures vibration waves of the rotating bearing and outputs a sensor signal indicating the vibration wave. A frequency component is extracted to produce a vibration low frequency signal, the vibration low frequency signal is sampled and recorded at a predetermined time interval, and in addition to reading and analyzing the recorded AE low frequency signal, the recorded vibration low frequency signal is The condition of the rotary bearing is determined by reading and analyzing the information, and the determination result is output. Therefore , the condition can be diagnosed from the AE waves and vibration waves measured by one common sensor directly attached to the rotary bearing. It is possible to improve the accuracy of condition diagnosis, and since it is not necessary to install both the vibration sensor and the AE sensor on the bearing, there is no shortage of installation space even with a small bearing. Since two sensors and their cables are not required, installation costs can be kept low.

In the rotary bearing diagnosing method and apparatus of the present invention, the frequency modulation is performed by heterodyne conversion at a predetermined reference frequency . Here, the predetermined reference frequency is one or more frequencies, for example, 100 kHz or more, which are included in the AE wave generation frequency band. The predetermined reference frequency may preferably be at least one of 100kHz, 150kHz, 200kHz and 300kHz. Further, the predetermined reference frequency may be set depending on the sensitivity of the sensor.

Furthermore, the analysis of at least one of the AE low frequency signal and the vibration low frequency signal is performed using conventional vibration analysis methods, such as frequency analysis using amplitude change, FFT (Fast Fourier Transform), and envelope processing. You may perform one or more of these.

1 is a block diagram showing the configuration of a rotary bearing diagnosing apparatus according to an embodiment of the present invention used in a rotary bearing diagnosing method according to an embodiment of the present invention. (a) and (b) are explanatory diagrams showing a method of heterodyne conversion performed by the modulation processing section of the rotary bearing diagnostic device of the above embodiment. It is a graph showing the relationship between the S/N ratio and the frequency of the AE wave at the initial stage and the final stage of wear of the rotary bearing measured by the shared sensor of the rotary bearing diagnostic device of the above embodiment.

Hereinafter, embodiments of the present invention will be described in detail based on the drawings. FIG. 1 is a block diagram showing the configuration of a rotary bearing diagnosing apparatus according to an embodiment of the present invention used in a rotary bearing diagnosing method according to an embodiment of the present invention. The diagnostic device of this embodiment diagnoses the condition of a rotating bearing from AE waves and vibration waves measured by one common sensor, and as shown in FIG. 1, a common sensor 1, a charge amplifier 2, It includes a high-pass filter 3, a modulation processing section 4, and an AE recording section 5.

The shared sensor 1 is directly attached to a rotating bearing such as a ball bearing, roller bearing, or sliding bearing of rotating mechanical equipment, and measures AE waves and vibration waves caused by metal contact or bearing damage in the rotating bearing, and detects the AE waves. The sensor outputs a sensor signal including vibration waves, and may be a vibration sensor or an AE sensor. In particular, the structure of non-amplifier sensors is similar to that of vibration sensors and AE sensors, and as mentioned above, it has been confirmed that even vibration sensors can detect high-frequency AE waves.On the other hand, even AE sensors can detect low-frequency AE waves. This is because it should be possible to detect vibration waves. Such a shared sensor 1 can be configured using a piezoelectric element, for example.

The charge amplifier 2 is an amplifier circuit that amplifies the sensor signal output by the shared sensor 1, and is an amplifier tailored to the type of the shared sensor 1, i.e., a vibration amplifier if the shared sensor 1 is a vibration sensor, and an AE if the shared sensor 1 is an AE sensor. Use an amplifier. Note that the vibration amplifier is also capable of amplifying sensor signals up to the frequency of the AE wave.

The high-pass filter 3 is a filter circuit that extracts the AE wave in the sensor signal output from the charge amplifier 2, and passes high frequency components of, for example, 50 kHz or more, preferably 70 kHz or more in the sensor signal. This is to eliminate low frequency noise at the resonant frequency when a vibration sensor is used as the shared sensor 1.

The modulation processing unit 4 generates a modulation reference signal of a predetermined reference frequency, for example, 100 kHz, in order to extract high frequency components of 100 kHz or more, which is a characteristic range of the frequency of AE waves generated due to an abnormality in the rotating bearing, from the sensor signal. The high-frequency component of the sensor signal that has passed through the high-pass filter 3 is frequency-modulated by a modulation reference signal of the reference frequency, for example, by heterodyne conversion, and converted into an AE low-frequency signal. Therefore, the modulation processing section 4 functions as an AE conversion section.

The reference frequency for heterodyne conversion is set to the AE generation frequency band, for example, 100 kHz or higher, but the sensitivity changes depending on the frequency characteristics of the sensor used as shared sensor 1, so the reference frequency should be set according to the sensitivity of the sensor used. It is preferable to determine the frequency, and in this case, it is set to 100 kHz through the following preliminary experiment. Furthermore, since the AE generation frequency band is wide, a plurality of frequencies, for example, not only 100 kHz, but also 150 kHz, 200 kHz, and 300 kHz, can be switched or used in parallel, or the frequency can be varied from 100 kHz to 300 kHz.

FIGS. 2(a) and 2(b) are explanatory diagrams showing a method of heterodyne conversion performed by the modulation processing section of the rotary bearing diagnostic device of the above embodiment. As shown in Fig. 2(a), when the input signal is frequency modulated by heterodyne conversion, which multiplies the input signal by the modulation reference signal of the reference frequency f, the input signal with a frequency lower than the reference frequency f becomes a deviation from the reference frequency f. Accordingly, as shown by the arrow, the signal turns over and overlaps with the signal having a frequency higher than the reference frequency f, resulting in a low frequency signal with a modulated waveform having a reference frequency of 0 kHz, as shown in FIG. 2(b).

As a result, the high frequency component in the sensor signal is converted to a low frequency band, and it becomes possible to measure the AE wave at a low sampling rate of the sampling frequency used in conventional vibration measurement. In addition, originally, AE measurements with high-speed sampling of about 5 MHz could only be measured for a very short time due to the data logger's ability, but by switching to low-speed sampling in the low frequency band, long-term measurements are now possible. .

The AE recording unit 5 performs low-speed sampling of the frequency-modulated AE low-frequency signal at a predetermined period, for example, 30 kHz corresponding to the period of bearing damage, and records the sampled data for the predetermined period in, for example, a memory or a hard disk. Note that before this low-speed sampling, range adjustment may be performed to narrow the frequency band of the AE low-frequency signal using a bandpass filter (not shown) to remove the DC component.

The diagnostic device of this embodiment also includes a low-pass filter 6 and a vibration recording section 7, as shown in FIG. The low-pass filter 6 is a filter circuit that extracts vibration waves in the sensor signal output by the charge amplifier 2, and passes low frequency components of, for example, 50 kHz or less in the sensor signal to generate vibration low frequency signals. Thereby, high frequency components that become noise are removed from the sensor signal.

The vibration recording unit 7 performs low-speed sampling of the vibration low-frequency signal at a predetermined period, for example, 30 kHz corresponding to the period of bearing damage, and records the sampled data for the predetermined period in, for example, a memory or a hard disk. The AE recording section 5 and the vibration recording section 7 may be one common recording section. Note that before this low-speed sampling, range adjustment may be performed to narrow the frequency band of the vibration low-frequency signal using a bandpass filter (not shown) to remove the DC component.

The diagnostic device of this embodiment further includes a determination processing section 8 and an alarm output section 9, as shown in FIG. In addition to reading and analyzing the AE low-frequency signal recorded in the AE recording unit 5, the determination processing unit 8 determines the state of the rotating bearing by reading and analyzing the vibration low-frequency signal recorded in the vibration recording unit 7. When the determination processing section 8 determines that the abnormality of the rotating bearing has worsened by a predetermined level or more as a result of the state determination, the alarm output section 9 outputs an alarm as the determination result. Therefore, these determination processing section 8 and alarm output section 9 function as a diagnosis section .

FIG. 3 is a graph showing the relationship between the S/N ratio and the frequency of AE waves at the initial and final stages of wear of a rotary bearing, which the inventor obtained through preliminary experiments. In experiments, it was confirmed that as the bearing abnormality worsened, the level of the high frequency band increased. This makes it possible to understand the state of wear by measuring the AE waves that occur at the initial stage of rubbing and the AE waves that occur at the final stage while changing the reference frequency.If the AE wave occurs around 100kHz, it is an early abnormality, and when it occurs around 300kHz, it is possible to understand the state of wear. It can be determined that the abnormal condition has worsened.

Therefore, the determination processing section 8 sets the reference frequencies in the conversion processing section 4 to, for example, 100kHz and 300kHz for the AE waves, and evaluates the AE low frequency signal at each reference frequency to detect abnormal wear of the rotating bearing. It is possible to determine whether the disease is in its early stages or in its final stage.

By the way, the present inventor has determined through preliminary experiments that sudden AE waves are detected in fatigue wear of rotating bearings, but vibration waves are not detected, and sudden AE waves and vibration waves are detected in the case of scratches, and foreign objects are detected. It has been confirmed that sudden AE waves are detected in the case of contamination, but vibration waves are hardly detected, and that continuous AE waves and vibration waves are detected in the case of poor lubrication. It can be seen that it is possible to determine the abnormal state of the rotation bearing from the output signal of the sensor 1.

Note that the conventional vibration analysis method can be used as is for the AE low frequency signal. Therefore, both the AE low frequency signal and the vibration low frequency signal can be analyzed by frequency analysis using amplitude changes, FFT (Fast Fourier Transform), envelope processing, and the like. Since the determination processing unit 8 analyzes low-frequency AE signals similar to vibration waves, an existing CMS (condition monitoring system) can also be used.

Although the above has been described based on the illustrated example, the present invention is not limited to the above-mentioned example . For example, instead of the alarm output unit 9, the result of the state determination of the rotating bearing is displayed regardless of the state or degree of abnormality. The device may be provided with a status output unit that outputs the status in the form of a graph or the like .

Thus, according to the rotary bearing diagnostic method and device of the present invention, by low-speed sampling of AE waves, the capacity of the sampled signal is suppressed even when measuring equipment that requires long-time measurement, such as low-speed rotating machinery, and recording is possible. Moreover, since the processing cost can be reduced, and since the signal sampled at low speed can be subjected to envelope processing, even minute frequency changes can be captured.
Furthermore, according to the rotary bearing diagnostic method and device of the present invention, the accuracy of condition diagnosis can be improved by diagnosing the condition from the AE waves and vibration waves measured by one common sensor directly attached to the rotary bearing. Moreover, since it is not necessary to install both the vibration sensor and the AE sensor on the bearing, there is no shortage of installation space even with a small bearing, and the two sensors, the AE sensor and the vibration sensor, and their Since no cables are required, installation costs can be reduced.

1 Common sensor 2 Charge amplifier 3 High pass filter 4 Modulation processing section 5 AE recording section 6 Low pass filter 7 Vibration recording section 8 Judgment processing section 9 Alarm output section f Reference frequency

Claims (10)

A method for diagnosing the condition of a rotating bearing,
A sensor directly attached to the rotary bearing measures the AE waves and vibration waves of the rotary bearing and outputs a sensor signal indicating the AE waves and vibration waves,
extracting high frequency components in a predetermined frequency range from the sensor signal;
Frequency modulating the high frequency component by heterodyne conversion at a predetermined reference frequency that is one or more frequencies included in the generation frequency band of the AE wave and converting it into an AE low frequency signal;
sampling and recording the AE low frequency signal at predetermined time intervals;
extracting a low frequency component in a predetermined frequency range from the sensor signal as a vibration low frequency signal;
sampling and recording the vibration low frequency signal at predetermined time intervals;
A rotary bearing characterized in that, in addition to reading and analyzing the recorded AE low frequency signal, the state of the rotary bearing is determined by reading and analyzing the recorded vibration low frequency signal and the determination result is output. diagnostic method. The method for diagnosing a rotating bearing according to claim 1, wherein the predetermined reference frequency includes at least one of 100 kHz, 150 kHz, 200 kHz, and 300 kHz. 2. The method for diagnosing a rotating bearing according to claim 1, wherein the predetermined reference frequency is set according to the sensitivity of the sensor . The rotation according to any one of claims 1 to 3, characterized in that the analysis of the AE low frequency signal is performed using one or more of amplitude change, FFT, and frequency analysis using envelope processing. How to diagnose bearings. The rotation according to any one of claims 1 to 4, characterized in that the analysis of the vibration low frequency signal is performed using one or more of amplitude change, FFT, and frequency analysis using envelope processing. How to diagnose bearings. A device for diagnosing the condition of a rotating bearing,
a sensor that is directly attached to the rotary bearing, measures AE waves and vibration waves of the rotary bearing, and outputs a sensor signal indicative of the AE waves and vibration waves;
a high-pass filter that extracts high frequency components in a predetermined frequency range from the sensor signal;
an AE converter that frequency modulates the high frequency component by heterodyne conversion at a predetermined reference frequency that is one or more frequencies included in the generation frequency band of the AE wave and converts it into an AE low frequency signal;
an AE recording unit that samples and records the AE low frequency signal at predetermined time intervals;
a low-pass filter that extracts low-frequency components in a predetermined frequency range from the sensor signal to generate a vibration low-frequency signal;
a vibration recording unit that samples and records the vibration low frequency signal at predetermined time intervals;
In addition to reading and analyzing the AE low frequency signal recorded in the AE recording unit, the state of the rotation bearing is determined by reading and analyzing the vibration low frequency signal recorded in the vibration recording unit, and the determination result is output. Diagnostic department and
A diagnostic device for a rotating bearing, comprising: 7. The rotating bearing diagnostic device according to claim 6, wherein the predetermined reference frequency includes at least one of 100 kHz, 150 kHz, 200 kHz, and 300 kHz. 7. The rotating bearing diagnostic device according to claim 6, wherein the predetermined reference frequency is set according to the sensitivity of the sensor. The rotation according to any one of claims 6 to 8, characterized in that the analysis of the AE low frequency signal is performed using one or more of amplitude change, FFT, and frequency analysis using envelope processing. Bearing diagnostic equipment. The rotation according to any one of claims 6 to 9, characterized in that the vibration low frequency signal is analyzed using one or more of amplitude change, FFT, and frequency analysis using envelope processing. Bearing diagnostic equipment.

Images