JP4608564B2 - Abnormality diagnosis method for low-speed rotating machinery - Google Patents

Description

The present invention relates to an abnormality diagnosis method for a low-speed rotating machine by composite sensing for diagnosing structural system abnormalities such as unbalance, misalignment, and axial cracks of a low-speed rotating machine driven at a low speed.

Conventionally, the following method has been disclosed as an abnormality diagnosis method for a low-speed rotating machine.
For example, in Patent Document 1, an acoustic emission generated by a bearing is detected as an AE signal, the AE signal is compared with a reference value, a time when the AE signal exceeds the reference value is calculated, and the duration time is calculated. It describes a method of diagnosing that an abnormality such as damage has occurred in the bearing when the reference value is exceeded.
Further, in Patent Document 2, as shown in FIG. 7, the vibration of the rolling bearing portion that supports the rotating machine is detected by an acceleration vibration sensor, and the vibration signal is individually passed through a plurality of bandpass filters, and the output of these filters is output. The envelope signal is processed by the envelope circuit, and the acceleration signal detected by the acceleration sensor is integrated twice to convert it to vibration displacement. The vibration displacement and the output of the envelope circuit are compared. A method for diagnosing the presence / absence and the cause and type of abnormality is described.
In Patent Document 3, vibration of a rotating machine is detected by a vibration acceleration sensor, the vibration detection signal is subjected to bandpass filter processing, and a crest factor of an eigenband component representing an abnormal state of a diagnosis target is calculated and calculated. A method is described that further calculates a crest factor of the crest factor, compares the crest factor of the calculated crest factor with a preset threshold value, and diagnoses an abnormality when a peak exceeding the threshold value occurs. .

JP-A-3-245504 JP-A-3-221818 JP-A-6-323899

However, the conventional method is applied only to an abnormality diagnosis of a bearing which is an example of a low-speed rotating machine.
Further, in the method of Patent Document 1, acoustic emission generated by a bearing is detected as an AE signal, but the AE sensor itself is expensive. For example, the sensor mounting position is similar to a roller apron of a continuous casting machine. In many cases, the cost increases and it is not economical. In addition, there is a problem that high expertise is required for performing the diagnosis.
In the methods of Patent Documents 2 and 3, the vibration of the rotating machine is detected by the vibration acceleration sensor, and the abnormality diagnosis is performed based on the vibration detection signal. Therefore, the cost of the entire diagnosis apparatus is low. However, in the low-speed rotary bearing with weak excitation force, such as the roller apron of the continuous casting machine described above, the vibration detection signal is weak, so it is difficult to distinguish between the effective signal component based on noise and abnormality. There was a problem that it was difficult to make a proper abnormality diagnosis.
Furthermore, vibration of the rotating machine is detected by a vibration acceleration sensor, and it is integrated once or twice and converted into vibration speed and vibration displacement to diagnose structural abnormalities of the low-speed rotating machine. The vibration frequency range that can be detected is limited by (acceleration acceleration sensor specification: generally 3 Hz or more). In addition, if the detected electrical signal of vibration acceleration is integrated twice by the electric circuit, the signal component in the lower frequency range is further cut, the effective signal component based on the abnormality is lost, and it is difficult to accurately perform the abnormality diagnosis. was there.

The present invention has been made in view of such circumstances, and uses a commercially available general-purpose and inexpensive strain gauge and displacement meter without using an expensive sensor such as an AE sensor, and is determined to be a complicated signal processing means. An object of the present invention is to provide an abnormality diagnosis method for a low-speed rotating machine that can accurately perform abnormality diagnosis for a low-speed rotating machine without using the method.

The abnormality diagnosis method for a low-speed rotating machine according to the present invention in accordance with the above object is to detect a dynamic strain of a non-rotating part of a low-speed rotating machine to be diagnosed with a strain gauge, and to rotate the rotating part or the non-rotating part of the low-speed rotating machine. Detection step of detecting the dynamic displacement of the part by a displacement meter;
A component extraction step of filtering each electrical signal obtained from the dynamic strain and the dynamic displacement and extracting a natural band component of a frequency representing an abnormal state of the low-speed rotating machine;
Each natural band component is decomposed into a plurality of different frequencies, and the signal levels A and B of the electric signals obtained for the respective frequencies are compared with the signal levels in the normal state of the low-speed rotating machine. A diagnostic process for diagnosing abnormalities in the rotating machine.

In the abnormality diagnosis method for a low-speed rotating machine according to the present invention, in the detecting step, a dynamic vibration of a non-rotating portion of the low-speed rotating machine is further detected by a vibration acceleration sensor, and the dynamic vibration is detected by the component extracting step. The electric signal obtained from the filter is filtered to detect the signal for each frequency band, and the signal level C of the signal extracted for each frequency band is compared with the signal level of the normal state of the low-speed rotating machine by the diagnosis step. It is preferable to do.

In the abnormality diagnosis method for a low-speed rotating machine according to the present invention, it is preferable that the type of abnormality relating to the structural system of the low-speed rotating machine is specified by the signal levels A and B obtained in the diagnosis step.
In the abnormality diagnosis method for a low-speed rotating machine according to the present invention, it is preferable to use a non-contact type laser displacement sensor for the displacement meter.

The abnormality diagnosis method for a low-speed rotating machine according to any one of claims 1 to 4, wherein an electrical signal of a dynamic strain obtained by a strain gauge and an electrical signal of a dynamic displacement obtained by a displacement meter have a detection sensitivity of a low frequency band component. Since it is excellent, the detection process, the component extraction process, and the diagnosis process are sequentially performed, so that it is possible to accurately diagnose whether or not there is an abnormality in the structural system of the low-speed rotating machine due to the strength of the frequency signal component. Since strain gauges and displacement gauges are used, there is no need to use expensive measurement equipment, and since a detection process, component extraction process, and diagnosis process are performed, no special data processing method is used, so dynamic Abnormalities in the structure of low-speed rotating machines that are difficult to detect and diagnose by vibration alone, such as unbalance, shaft breakage, misalignment, etc., can be diagnosed economically and easily with high accuracy.

In particular, the abnormality diagnosis method for a low-speed rotating machine according to the second aspect of the present invention can also perform a conventional abnormality diagnosis for a low-speed rotating machine by detecting an electric signal of dynamic vibration obtained by a vibration acceleration sensor.
In the abnormality diagnosis method for a low-speed rotating machine according to the third aspect, the type of abnormality related to the structure of the low-speed rotating machine can be specified based on the signal levels A and B obtained in the diagnosis step, so that the subsequent maintenance work becomes easy.

The abnormality diagnosis method for a low-speed rotating machine according to claim 4 uses a non-contact type laser displacement sensor for the displacement meter, so that, for example, measurement is performed in comparison with an eddy current type displacement meter that is often used conventionally. The range of the object is wide, the resolution is high, the measurement range can be further widened, and the dynamic displacement of the rotating part of the low-speed rotating machine can be measured with higher accuracy.

Next, embodiments of the present invention will be described with reference to the accompanying drawings for understanding of the present invention.
As shown in FIG. 1, the abnormality diagnosis method for a low-speed rotating machine according to an embodiment of the present invention detects an abnormality related to the structure of a stirrer (an example of a low-speed rotating machine) to be diagnosed. In this method, a component extraction step and a diagnosis step are sequentially performed. In addition, (a) shown in FIG. 1 is the flow of diagnosis by the vibration measurement diagnostic device 11 that can cope with the diagnosis of the commercially available vibration acceleration sensor 10 and the low-speed rotating machine. Moreover, (b) is a flow of diagnosis by a general-purpose strain gauge 12. And (c) is a flow of diagnosis by a commercially available non-contact type laser displacement sensor (an example of a displacement meter) 13. Reference numeral 14 in FIG. 1 denotes data processing software. Hereinafter, the stirring shaft of the stirrer will be described as a rotating portion, and the casing portion (bearing box) of the rolling bearing that supports the stirring shaft will be described as a non-rotating portion. In addition, the rotation speed of a stirring shaft is 300 rpm or less, Furthermore, 180 rpm or less (here about 20-50 rpm).

First, the detection process will be described.
A vibration acceleration sensor 10 that detects the vibration of the stirrer is attached to a casing portion of a rolling bearing that supports the stirring shaft of the stirrer. Thereby, an electric signal having an amplitude (dynamic vibration) corresponding to the vibration acceleration of the rolling bearing can be output as a vibration detection signal.
Moreover, the strain gauge 12 which detects the deformation | transformation of the casing part by external force is affixed on the casing part of an above-described rolling bearing. The place where the strain gauge 12 is affixed is a place where the stress is most greatly received by the rotational load of the rolling bearing (for example, a place obtained from past experience). Thereby, the electrical signal of the dynamic strain according to the stress can be output as the strain detection signal.
And the laser displacement sensor 13 which detects the displacement which generate | occur | produces during rotation of the stirring shaft of a stirrer is arrange | positioned in the position (The position which can detect the displacement of a casing part) which can detect the displacement of a stirring shaft. In addition, as a displacement detection location by the laser displacement sensor 13, a location (for example, a location obtained from past experience) that receives the greatest stress due to the rotational load of the stirring shaft is selected. Thereby, the electrical signal of the dynamic displacement according to the displacement generated during the rotation of the stirring shaft can be output as the displacement detection signal.

Next, the component extraction process will be described.
The vibration detection signal output from the vibration acceleration sensor 10 is first amplified by an amplifier, and then filtered by a band-pass filter, as in the prior art, to obtain a detection signal for each frequency band. Specifically, the speed obtained by integrating the high frequency signal (10 to 40 kHz) from the acceleration Hi signal, the medium to low frequency signal (3 Hz to 10 kHz) signal to the acceleration Md signal, and the low frequency signal (3 to 20 Hz) signal once. Similarly, the Lo signal, and similarly, the displacement Disp signal obtained by integrating the low frequency region (5 to 20 Hz) signal twice are obtained.
Further, the strain detection signal output from the strain gauge 12 and the displacement detection signal output from the laser displacement sensor 13 are each filtered by a band pass filter. At this time, since the frequency of the pass band is selected to be 10 to 20 times the rotation frequency of the stirring shaft, for example, more than 0 Hz and 20 Hz or less, a natural band component of a frequency representing an abnormal state of the stirrer can be extracted.
The component extraction process for processing the strain detection signal and the displacement detection signal is performed by a conventionally known data processing method.
The band-pass filter is a conventionally known filter that combines a high-pass filter with low-frequency cutoff and a low-pass filter with high-frequency cutoff.

Next, the diagnosis process will be described.
The detection signal of each band detected from the vibration acceleration sensor 10, that is, the acceleration detection signal, the velocity signal, and the displacement signal are each subjected to FFT conversion to calculate a spectrum (signal level C), and a normal state spectrum (signal) of the stirrer Compared with the level), the abnormality of the agitator is diagnosed. It should be noted that an effective value, an average value, and a peak value may be further calculated from the above-described detection signal to perform abnormality diagnosis of the stirrer.
Here, the FFT transform is an abbreviation for Fast Fourier Transform, which is a method of transforming a signal into the frequency domain, and is a conventionally known method used for observing frequency components and phases ( The same applies below).
The component extraction process and the diagnostic process for processing the vibration detection signal output from the vibration acceleration sensor 10 are performed by the diagnostic measurement diagnostic device 11.

In addition, the strain detection signal detected by the strain gauge 12 and processed by the band pass filter is decomposed into a plurality of different frequencies and subjected to FFT conversion, so that the spectrum exceeds 0 Hz and is 20 Hz or less (further 10 Hz or less). (Signal level A obtained for each frequency).
As a result, for example, any of imbalance (stirring blade misalignment), misalignment (stirring shaft misalignment when the stirrer is installed), and shaft breakage (stirring shaft damage due to use of the stirrer) If there is an abnormality related to one or more types of structures, that is, a structural system abnormality, the strain (stress) component of the rotational frequency increases, so the strain component Fr of the rotational frequency fr stands out from the strain components of other frequencies. Therefore, the abnormality can be confirmed.

Then, the ratio of the value ε (fr) of the distortion component Fr of the rotational frequency fr to the effective value of all the components Σε (fi) up to 10 times the rotational frequency, that is, the crest factor for the signal component of the rotational frequency is obtained. Whether or not there is an abnormality is determined by comparing whether or not the value exceeds a preset threshold value. Here, the preset threshold value is a value obtained from the signal level in the normal state of the stirrer.
Further, by observing the characteristics of the spectrum of the strain signal detected from each direction (for example, the frequency of the signal level A and the strain intensity thereof), the cause and part of the abnormality (kind of abnormality) can be specified.

Further, the displacement detection signal detected by the laser displacement sensor 13 and processed by the bandpass filter is decomposed into a plurality of different frequencies and subjected to FFT conversion, so that it exceeds 0 Hz and is 20 Hz or less (and further 10 Hz or less). It represents a spectrum (signal level B obtained for each frequency).
Thereby, for example, when there is one or more structural system abnormalities of imbalance, misalignment, and shaft breakage, the displacement component of the rotational frequency becomes high, and the displacement component Dr of the rotational frequency fr becomes abnormal. Can stand out from the noise.

Then, the ratio of the value D (fr) of the displacement component Dr of the rotational frequency fr to the effective value of all the components ΣD (fi) up to 10 times the rotational frequency, that is, the crest factor for the signal component of the rotational frequency is obtained. Whether or not there is an abnormality is determined by comparing whether or not the value exceeds a preset threshold value. Here, the preset threshold value is a value obtained from the signal level in the normal state of the stirrer.
Furthermore, by observing the characteristics of the spectrum of the displacement signal detected from each direction (for example, the frequency of the signal level B and its axial runout intensity), the cause and part of the abnormality (type of abnormality) can be specified.

It should be noted that the diagnosis of the stirrer abnormality in the above-described diagnosis process is preferably performed by combining each electric signal obtained from dynamic strain and dynamic displacement, but only one of the obtained electric signals is used. You may do it.
As described above, the characteristics of the frequency and amplitude of each signal level obtained from the vibration acceleration sensor 10, the strain gauge 12, and the laser displacement sensor 13 are the same signal level in the normal state of the stirrer. By comparing with the characteristics of the frequency and the magnitude of the amplitude, it is possible to diagnose the presence or absence of abnormality of the stirrer and to accurately diagnose the cause and type of the abnormality.
In the diagnosis of the stirrer, either simple diagnosis or precise diagnosis or both are performed. Here, the simple diagnosis is a diagnosis for determining whether the execution value, average value, and peak value calculated from the signal are larger than the value of the stirrer in a normal state, and the precise diagnosis is a signal level. This is a diagnosis that examines the frequency and amplitude of the signal and identifies the abnormal part.

Next, examples carried out for confirming the effects of the present invention will be described.
Here, the result of having performed the diagnosis using the stirrer which is a low-speed rotating machine is demonstrated. The specifications of the stirrer are as follows: motor: vertical type, motor rating: 0.75 kW (1780 rpm), rotation speed of the stirring shaft: over 0 and 108 rpm, number of stirring blades: 2, tank volume: about 3075 Liter, water storage capacity: 429 liters. The rotation speed of this motor is controlled by an inverter. The power is transmitted from the motor to the stirring shaft via the speed reducer. And the diagnostic object part of a stirrer is made into the stirring shaft part of the low speed rotation after a reduction gear.
Hereinafter, the unbalanced state was diagnosed by attaching a weight to one of the stirring blades.

First, a strain gauge is attached to the bearing box (below the casing), which is a non-rotating part of the stirrer, and the bearing box detected when the rotation speed of the stirring shaft of the stirrer is 20 rpm (fr = 0.33 Hz). The result of the strain detection signal will be described with reference to FIGS. 2 (A), 2 (B), 3 (A), and 3 (B). 2A and 2B show the results when the stirrer is in a normal state, and FIGS. 3A and 3B show the results when the stirrer is in an unbalanced state. 2A and 3A are time waveforms of strain detection signals, respectively, and FIGS. 2B and 3B are spectra obtained by FFT transforming the strain detection signals, respectively. .

In FIG. 2B where the stirrer is in a normal state, the passing frequency of the stirring blade (2 fr in FIG. 2A) and its harmonic components appeared, but no component of the rotational frequency was confirmed. . This means that it matches the operating characteristics of the stirrer (not the unbalanced state and the misaligned state).
On the other hand, in FIG. 3 (B) where the stirrer is in an unbalanced state, although not conspicuous, in addition to the passing frequency of the stirring blade (2fr in FIG. 3 (B)), the rotation frequency (FIG. 3 (B) The fr) peak in the middle also appeared.

Next, the result of the bearing box strain detection signal detected when the rotation speed of the stirring shaft of the stirrer is 50 rpm (fr = 0.83 Hz) will be described with reference to FIGS. To do. 4A and 4B show the results when the stirrer is in a normal state, and FIGS. 4C and 4D show the results when the stirrer is in an unbalanced state. 4A and 4C are time waveforms of strain detection signals, respectively, and FIGS. 4B and 4D are spectra obtained by FFT transforming the strain detection signals, respectively.
Even when the rotation speed of the stirring shaft was 50 rpm, as in the case of 20 rpm described above, in FIG. 4B where the stirrer is in a normal state, the rotational frequency component was not confirmed at all. In FIG. 4D showing the balance state, the peak of the rotation frequency (fr in FIG. 4D) appeared prominently.

From the above results, it can be confirmed that the strain detection signal detected by the strain gauge can be used to diagnose and identify the unbalanced state of the stirrer based on the change and strength of the rotational frequency component from the spectrum. It was.
The dynamic displacement of the agitation shaft was measured with a laser displacement sensor fixed to the bearing housing.In this case, the measurement data is the dynamic displacement of the agitation shaft relative to the bearing housing. It was also confirmed that it could not be used.

Next, the result of detecting the abnormality using an experimental rotating machine simulator (hereinafter simply referred to as a simulator) that is a low-speed rotating machine will be described. The simulator has a motor rating of 0.75 kW (1780 rpm). The rotation speed of this motor is controlled by an inverter. The power is transmitted from the motor via the belt to the rotating shaft and the rotor connected by the coupling. And the diagnostic object part of a simulator is a coupling, a rotor, a rotating shaft, and a bearing box.
Hereinafter, the unbalanced state is set by attaching a weight of 107 g to one part of the rotor, the misaligned state is finely adjusted in the horizontal direction with the position of the bearing box on the non-driving side (driven side), and the coupling is adjusted. The diagnosis was made by setting the rotation axis and the drive side angle at the center by shifting by 0.5 degrees.

First, FIG. 5A shows the result of the displacement detection signal obtained by detecting the counter-drive side of the rotating shaft with the laser displacement sensor when the rotating shaft, which is the rotating portion of the simulator, is adjusted to 59 rpm (fr = 0.98 Hz). This will be described with reference to (F). 5A and 5B show the results when the simulator is in a normal state, and FIGS. 5C and 5D show the results when the simulator is in an unbalanced state. FIG. , (F) are the results in the misalignment state. 5A, 5C, and 5E are time waveforms of displacement detection signals, respectively, and FIGS. 5B, 5D, and 5F are FFT-transformed displacement detection signals, respectively. It is a spectrum.

In FIG. 5B in which the simulator is in a normal state, the rotational frequency component (fr in FIG. 5B) and its double component appear as strong as the same.
On the other hand, only the rotational frequency component (fr in FIGS. 5D and 5F) is strongest in FIG. 5D in which the simulator shows an unbalanced state and FIG. 5F in which the misaligned state is shown. In addition, with respect to the time waveforms shown in FIGS. 5C and 5E, the periodicity of the rotation frequency was also strong.

Next, with respect to the result of the bearing box strain detection signal detected when a strain gauge is attached to the side surface of the bearing box, which is a non-rotating part of the simulator, and the rotational speed of the rotating shaft is 59 rpm, FIG. This will be described with reference to (F). 6A and 6B show the results when the simulator is in a normal state, and FIGS. 6C and 6D show the results when the simulator is in an unbalanced state. FIG. , (F) are the results in the misalignment state. 6 (A), (C), and (E) are time waveforms of strain detection signals, respectively, and FIGS. 6 (B), (D), and (F) are FFT-transformed strain detection signals, respectively. It is a spectrum.

In FIG. 6B where the simulator is in a normal state, the peak of the rotational frequency component (fr in FIG. 6B) did not appear.
On the other hand, in FIG. 6D in which the simulator shows an unbalanced state, the peak of the rotation frequency component (fr in FIG. 6D) is the main component compared to FIG. 6B which shows the normal state. As it appeared prominently. Further, in FIG. 6F showing the misalignment state, the peak of the rotation frequency component (fr in FIG. 6F) clearly appeared.
The above results were consistent with the mechanical characteristics at the time of occurrence of the unbalanced state and the misaligned state, and it was confirmed that an abnormality could be diagnosed.

Therefore, when diagnosing the unbalanced state and the misalignment state of the low-speed rotating machine, the low-speed rotating machine whose rotation speed is approximately 180 rpm or less is diagnosed using the strain detection signal and the displacement detection signal rather than the vibration detection signal. Was confirmed to be effective. Furthermore, vibration detection signals are also effective for low-speed rotating machines with a rotational speed of 180 rpm or more and 300 rpm or less, but by using a strain detection signal and a displacement detection signal together, it is possible to more reliably identify and diagnose the state. It could be confirmed.

As described above, the present invention has been described with reference to the embodiment. However, the present invention is not limited to the configuration described in the above embodiment, and the matters described in the scope of claims. Other embodiments and modifications conceivable within the scope are also included. For example, the case where the abnormality diagnosis method for a low-speed rotating machine of the present invention is configured by combining some or all of the above-described embodiments and modifications is also included in the scope of the right of the present invention.
In the above embodiment, the case where the abnormality of the stirrer is diagnosed as the low-speed rotating machine has been described. However, the present invention is not limited to this, and an abnormality of another low-speed rotating machine, for example, a continuous casting machine is diagnosed. You can also
In the above embodiment, the case where the vibration acceleration sensor is used has been described. However, in order to diagnose an abnormality in a low-speed rotating machine having a very low rotational speed and a large load on the rotation of the machine, the vibration acceleration sensor is used. Without using it, the presence or absence of abnormality may be diagnosed using two strain gauges and a laser displacement sensor.

It is explanatory drawing of the abnormality diagnosis method of the low speed rotary machine which concerns on one embodiment of this invention. (A) is a graph showing a time waveform in a normal state of a strain detection signal of a bearing box when the rotation speed of the stirring shaft of the stirrer is 20 rpm, and (B) is a graph showing a spectrum obtained by subjecting (A) to FFT conversion. It is. (A) is a graph showing a time waveform in an unbalanced state of a bearing box strain detection signal when the rotation speed of the stirring shaft of the stirrer is 20 rpm, and (B) shows a spectrum obtained by subjecting (A) to FFT conversion. It is a graph. (A) is a graph showing a time waveform in a normal state of a strain detection signal of a bearing box when the rotation speed of the stirring shaft of the stirrer is 50 rpm, and (B) is a graph showing a spectrum obtained by subjecting (A) to FFT conversion. (C) is a graph which shows the time waveform in an unbalanced state, (D) is a graph which shows the spectrum which carried out FFT conversion of (C). (A) is a graph showing a time waveform in a normal state of the displacement detection signal of the rotating shaft when the number of rotations of the stirring shaft of the simulator is 59 rpm, (B) is a graph showing a spectrum obtained by FFT conversion of (A), (C) is a graph showing a time waveform in an unbalanced state, (D) is a graph showing a spectrum obtained by FFT transforming (C), (E) is a graph showing a time waveform in a misaligned state, and (F) is It is a graph which shows the spectrum which FFT-transformed (E). (A) is a graph showing the time waveform in the normal state of the strain detection signal of the bearing box when the rotation speed of the stirring shaft of the simulator is 59 rpm, (B) is a graph showing the spectrum obtained by FFT conversion of (A), (C) is a graph showing a time waveform in an unbalanced state, (D) is a graph showing a spectrum obtained by FFT transforming (C), (E) is a graph showing a time waveform in a misaligned state, and (F) is It is a graph which shows the spectrum which FFT-transformed (E). It is explanatory drawing of the abnormality diagnosis method of the low speed rotary machine which concerns on a prior art example.

Explanation of symbols

10: vibration acceleration sensor, 11: vibration measurement diagnostic device, 12: strain gauge, 13: laser displacement sensor (displacement meter), 14: data processing software

Claims (4)

Detecting the dynamic strain of the non-rotating part of the low-speed rotating machine to be diagnosed with a strain gauge, and detecting the dynamic displacement of the rotating part of the low-speed rotating machine or the non-rotating part with a displacement meter;
A component extraction step of filtering each electrical signal obtained from the dynamic strain and the dynamic displacement and extracting a natural band component of a frequency representing an abnormal state of the low-speed rotating machine;
Each natural band component is decomposed into a plurality of different frequencies, and the signal levels A and B of the electric signals obtained for the respective frequencies are compared with the signal levels in the normal state of the low-speed rotating machine. And a diagnostic process for diagnosing abnormality of the rotating machine. 2. The abnormality diagnosis method for a low-speed rotating machine according to claim 1, wherein in the detecting step, a dynamic vibration of a non-rotating portion of the low-speed rotating machine is further detected by a vibration acceleration sensor, and the dynamic extraction is performed by the component extracting step. The electric signal obtained from the vibration is filtered to detect the signal for each frequency band, and the signal level C of the signal extracted for each frequency band in the diagnosis step is set as the signal level of the normal state of the low-speed rotating machine. An abnormality diagnosis method for a low-speed rotating machine, characterized by comparing. 3. The abnormality diagnosis method for a low-speed rotating machine according to claim 1, wherein the type of abnormality relating to the structural system of the low-speed rotating machine is specified based on the signal levels A and B obtained in the diagnosis step. An abnormality diagnosis method for a low-speed rotating machine. The abnormality diagnosis method for a low-speed rotating machine according to any one of claims 1 to 3, wherein a non-contact type laser displacement sensor is used for the displacement meter.

Images