JP6017649B2 - Abnormal diagnosis method for rotating machinery - Google Patents

Description

The present invention relates to a method for diagnosing abnormalities in a rotating machine system by comparing the state of current during normal operation and the state of current during operation (inspection).

Conventionally, for diagnosis of the state of a rotating machine system consisting of a motor or a mechanical system such as a pump driven by an electric motor or a reduction gear (gear device), a method using vibration as a measurement parameter because of high diagnostic accuracy and The device is used. However, depending on the installation position of the rotating machine system, a person may approach and cannot install a vibration sensor or the like.
Therefore, in such a case, a state monitoring method based on dimensional feature parameters such as the effective value and peak value of the current of the electric motor is generally used (see, for example, Patent Document 1).

Japanese Patent Laid-Open No. 11-83686

However, when an abnormality occurs in a rotating machine system, there is almost no significant change in the effective value or peak value of the current. Conversely, when a change such as a load occurs, the effective value or peak value of the current changes greatly. Therefore, it is impossible to determine the state using these parameters. In particular, the characteristic parameter is obtained by quantifying a part of information of the current waveform, and it cannot be said that the obtained information is effectively used.
In other words, in the diagnosis of a rotating machine system using characteristic parameters, it is necessary to change the determination of abnormality and the identification standard when the machine load and operating conditions change, so it is difficult to set unified judgment standards and identification methods. There was a problem.

The present invention has been made in view of such circumstances, and an object of the present invention is to provide a method for diagnosing abnormalities in a rotating machine system that is capable of setting an abnormality determination standard and identification and is more sensitive than the conventional one.

An abnormality diagnosis method for a rotating machine system according to the present invention that meets the above-mentioned object is an abnormality diagnosis method for a rotating machine system driven by an electric motor.
The current signal during operation of the motor is sampled, subjected to fast Fourier transform, and then the spectrum is logarithmically converted. The level of the spectrum peak of the power frequency of the motor is symmetrical with respect to the power frequency as the center frequency. The rotational mechanical system is connected by a transmission joint on the condition that the difference from the level of the spectrum peak appearing at the frequency position separated by the rotational frequency of the child shaft is within the range of 30 to 50 decibels. It is determined that an alignment error has occurred between the rotor shaft of the motor and the rotating shaft on the load side.

The abnormality diagnosis method for a rotating machine system according to the present invention is such that the current signal during operation of the motor is sampled, subjected to fast Fourier transform, and then the spectrum is logarithmically converted to symmetrically about the power supply frequency of the motor as the center frequency. A spectral peak appears at a frequency position separated by the rotational frequency of the rotor axis. The level of this spectral peak changes due to an abnormality in alignment between the rotor shaft of the motor connected by a transmission joint in the rotating mechanical system and the rotating shaft on the load side. It can be determined that an abnormality has occurred on the condition that the difference between the spectrum peak level and the above-described spectrum peak level is within the range of 30 to 50 decibels .

It is a flowchart of the abnormality diagnosis method of the rotating machine system which concerns on one embodiment of this invention. It is explanatory drawing of the abnormality diagnosis apparatus of the rotating machine system which applies the abnormality diagnosis method of the same rotating machine system. (A) is a graph of a standard sine wave signal waveform, (B) to (E) are graphs of current waveforms during operation of the motor, (F) is a graph of a reference amplitude probability density function of (A), and (G). -(J) is a graph of the amplitude probability density function at the time of inspection of (B)-(E), respectively. (A)-(C) are the graphs which compared the current waveform at the time of the normal of each phase of current of a three-phase induction motor, and the current waveform at the time of inspection. (A)-(C) are graphs when a gear bearing is judged to be normal by performing high-frequency component analysis of the current waveform of each phase of the current of the three-phase induction motor. (A)-(C) are graphs when the high frequency component analysis of the current waveform of each phase of the current of the three-phase induction motor is performed to determine that the gear bearing or the meshing is abnormal. (A)-(C) are the graphs of the result of having performed the sideband analysis of the current waveform of each phase of the current of a three-phase induction motor. (A) is a graph which shows the current waveform at the time of the normal of one phase of the electric current of a three-phase induction motor, (B) is a graph which shows the current waveform at the time of abnormality. (A)-(C) is a graph of the result of having performed the pattern analysis of the starting current which is an excessive current value of each current of each phase of a three-phase induction motor, and (D)-(F) are the transient current of each current phase. It is a graph of the result of having performed the pattern analysis of the interruption current which is a value. (A)-(C) are each a graph of the result of having performed distortion analysis of the current waveform.

Next, embodiments of the present invention will be described with reference to the accompanying drawings for understanding of the present invention.
First, a rotating machine system abnormality diagnosis device (hereinafter also simply referred to as an abnormality diagnosis device) 10 to which a rotating machine system abnormality diagnosis method according to an embodiment of the present invention is applied will be described with reference to FIG.
The abnormality diagnosis apparatus 10 diagnoses a rotating machine system including 1) an AC electric motor 11 to be subjected to abnormality diagnosis and / or 2) a mechanical system 12 (load side) driven by the electric motor 11. Device. The mechanical system 12 includes, for example, a pump, a reduction gear (gear device), and the like. An electric motor 11 is connected to the mechanical system 12, and the electric motor 11 is connected to a power source (control panel) 14 through a power line 13. Yes.

For example, the abnormality diagnosis apparatus 10 connects the electric motor 11 and the power source 14 and is disposed in the vicinity of the power line 13 located in an electric room, a control panel (electric panel) or the like, and detects a load current when the electric motor 11 is operating. A current detector 15; an A / D converter 16 that converts an analog current waveform measured by the current detector 15 into digital current data; and a processing unit that processes the current data obtained by the A / D converter 16. 17. Note that transmission of current data from the A / D converter 16 to the processing unit 17 is performed by a LAN or a USB cable.
Here, for example, a galvanometer (CT: Current Transducer) or the like can be used as the current detector 15. The processing unit 17 includes a RAM, a CPU, a ROM, an I / O, and a conventionally known arithmetic unit (that is, a computer) having a bus connecting these elements, and is connected to the electric motor 11 (and / or). It also has a monitor that displays the diagnosis result of the rotating machine system composed of the machine system 12. Note that the processing in the processing unit 17 is realized by the CPU executing a predetermined program.

Subsequently, a rotating machine system abnormality diagnosis method (hereinafter also simply referred to as an abnormality diagnosis method) according to an embodiment of the present invention will be described.
When performing an abnormality diagnosis, first, as shown in FIG. 2, a current detector 15 is installed on the power line 13 connecting the motor 11 and the power source 14, and the current of the motor 11 at the time of inspection (operation) is measured. Since the motor 11 is a three-phase induction motor, the current detector 15 is installed on each of the three power lines 13 here. However, for example, a current detector is installed on one or two power lines. Then, the current may be measured for one phase of the three-phase power source supplied to the electric motor 11. Further, the current measurement may be performed by sequentially switching each phase of the three-phase power supplied to the electric motor 11.

The number of installed current detectors and the measurement of current are preferably determined as appropriate according to the importance of the object to be diagnosed. For example, if it is difficult to lead to a big problem even if an abnormality occurs, it is better to install a current detector in a part (for example, one or two) of power lines, and it is necessary to improve the detection accuracy of the abnormality. In this case, it is preferable to install current detectors on all (three in this case) power lines and switch the current measurement simultaneously or sequentially.
Then, the simple diagnosis (abnormality detection) of Step 1 (ST1) to Step 6 (ST6) shown in FIG. 1 is performed, and Step 7 (ST7) to Step 14 ( The precise diagnosis (abnormality identification) of ST14) is performed. This will be described in detail below.

First, in step 1 for performing a simple diagnosis, a reference amplitude probability density is obtained from a standardized current signal obtained by dividing a standard sine wave signal waveform of a rated current (60 Hz in this case) that is a power supply frequency of the motor 11 by its effective value. The function fr (x) is obtained and stored in the storage means (RAM or ROM) of the processing unit 17. The reference sine wave signal waveform is a waveform without distortion, as shown in FIG. Note that this waveform may be obtained by calculation, or a normal waveform may be used.
Thus, by using the current signal, the diagnostic accuracy can be improved as compared with the diagnosis using the vibration signal. In the case of diagnosis using vibration signals, it is necessary to measure the vibration signal in the state that the device seems to be normal and use it as a reference signal, but in fact, there is a possibility that abnormal vibration is included in it There is a difficulty in detecting the abnormality. On the other hand, in the case of diagnosis using a current signal, since the reference sine wave signal can be obtained by a reference sine wave signal generator or calculation, an abnormal signal is not included in the reference signal.
Then, the reference sine wave signal waveform is A / D converted into a plurality of point data obtained at a predetermined sampling time, and the reference amplitude probability density function fr (x) shown in FIG. First step).

Next, in step 2 (ST2), the three-phase current during operation of the electric motor 11 is measured, A / D converted by the A / D converter 16, and the processing unit 17 (monitoring computer) via a LAN or USB cable. To the storage means.
As shown in FIGS. 3B to 3E, the current waveform during operation of the electric motor 11 has various shapes depending on the state of the rotating machine system. In addition, the measurement time of the electric current of the motor is performed, for example, at intervals of 1 to 10 seconds (preferably, the lower limit is 3 seconds and the upper limit is 7 seconds).
In step 3 (ST3), the inspection amplitude probability density function ft (x) shown in FIGS. 3G to 3J is obtained from a plurality of point data of the current waveform during operation of the electric motor 11, and the processing unit The data is stored in 17 storage means. The inspection amplitude probability density function ft (x) is a normalized current signal obtained by dividing the inspection motor current signal by the calculated effective value (the second step).

In step 4 (ST4), the reference amplitude probability density function fr (x) obtained in step 1 and the inspection amplitude probability density function ft (x) obtained in steps 2 and 3 are set to ID (Information Divergence) and KI (Kullback-Leibler Information) is calculated.
Generally, when the equipment state changes, the amplitude probability density function of the current waveform also changes. Therefore, ID and KI are used to obtain the difference between the two amplitude probability density functions, that is, the “information distance”. In particular, in abnormality diagnosis, since ID and KI can accurately represent the difference between two amplitude probability density functions, it is expected to have better abnormality detection accuracy.

In step 4, the values of both ID and KI are calculated. However, if necessary, only the value of either ID or KI may be calculated.
Here, when both the reference amplitude probability density function fr (x) and the inspection amplitude probability density function ft (x) are normal distribution functions, assuming that the respective variances are σr ² and σt ² , σr ² > sensitivity of ID if .sigma.t ² may be detected accurately. On the other hand, if σr ² <σt ² , the sensitivity of KI is good and the detection accuracy is also good.
If either the reference amplitude probability density function fr (x) or the inspection amplitude probability density function ft (x) does not become a normal distribution function, or if it is unknown whether or not it is a normal distribution function, ID and By calculating both values of KI, detection accuracy is increased.
Here, with respect to the normalized current signal of amplitude x, if fr (x) is a reference distribution and ft (x) is a test distribution, the definition of ID is expressed by equation (1), and KI Is defined by the formula (2) (the third step).

In Step 5 (ST5), the state deterioration tendency of the rotating machine system is managed based on the above-described change with time of ID and KI.
Specifically, according to the definition of ID and KI, when the two amplitude probability density distributions are exactly the same, that is, when fr (x) = ft (x), ID = 0 and KI = 0. . Further, when the deviation of fr (x) and ft (x) becomes large, ID and KI also become large.

Therefore, in Step 6 (ST6), whether the information distance ID or KI of the two distributions, ie, fr (x) and ft (x), is an amount (value) by which the abnormality of the rotating machine system is estimated by the information test theory. Judging. Let X be a sample space obtained from n independent observations, and divide X into sets E ₀ and E ₁ that do not cross each other, that is, E ₀ ∩E ₁ = 0 and X = E ₀ ∪E ₁ . If a sample point is x,
Null hypothesis H ₀ : x∈E ₀
Alternative hypothesis H ₁ : x∈E ₁
On the other hand, the first type of error α (overdetection rate) is Prob (xεE ₁ | H ₀ ), and the second type of error β (missing rate) is Prob (xεE ₀ | H ₁ ). .

If ID (5) is satisfied, H ₁ is selected. Here, _{O n} represents one of the samples of n times independent observations, the _{R ID} (alpha, beta) is called ID criteria. Note that R _ID (α, β) is expressed by Equation (3).

Further, with respect to KI, if the equation (6), to select the _{H 1.} This R _KI (α, β) is referred to as a KI criterion. Note that R _KI (α, β) is expressed by Equation (4).

Changes in R _ID (α, β) and R _KI (α, β) with respect to α and β are as follows.
When α ≧ 1-β, R _ID (α, β) and R _KI (α, β) are monotonically increasing functions.
When α is increased with β being constant, R _ID (α, β) and R _KI (α, β) are decreased, and when β is increased with α being constant, R _ID (α, β) and R _KI ( Since (α, β) becomes smaller, the criterion becomes stricter.
Therefore, using the ID or KI obtained from the reference amplitude probability density function fr (x) and the inspection amplitude probability density function ft (x), the above-described equations (5) and (3), and The presence or absence of abnormality is determined by (6) and formula (4).

In this way, by recording the time-series change of the ID and KI values obtained (calculated) during operation, the deterioration tendency of the state of the rotating machine system can be managed, and at the same time, a predetermined criterion R _ID ( A simple diagnosis of the rotating machine system is performed by comparing with α, β) and R _KI (α, β). That is, in step 6, when both ID and KI satisfy the conditions (ID <R _ID (α, β) and KI <R _KI (α, β)), no sign of abnormality appears in the rotating machine system. If either or both of ID and KI do not satisfy the above conditions (ID ≧ R _ID (α, β) and KI ≧ R _KI (α, β) or both) If satisfied, it is determined that a sign of abnormality in the rotating machine system has appeared. Α and β are an overdiagnosis rate and an oversight rate for abnormality diagnosis, respectively, and can be set in advance in accordance with a request for facility abnormality detection. For example, α is 0.05 to 0.3 (preferably, lower limit is 0.1, upper limit is 0.2), β is 0.05 to 0.3 (preferably, lower limit is 0.1, upper limit 0.2).
In this step 6, when the value of only ID is calculated in the above-mentioned step 4, whether or not the condition of ID ≧ R _ID (α, β) is satisfied, and the value of only KI is calculated in step 4 , KI ≧ R _KI (α, β) is determined (4th step).

If it is determined in step 6 that there is no sign of abnormality in the rotating machine system, steps 2 to 6 described above are sequentially repeated. If it is determined that a sign of abnormality appears in the rotating machine system, an abnormality message is displayed on the monitor in step 7 (ST7), and the following precision analysis is performed.
Specifically, 1) Abnormal detection and identification of rotating machine system by high-frequency component analysis of current waveform, 2) Abnormal detection and identification of rotating machine system by sideband analysis of current waveform, and 3) Pattern analysis of excessive current value Abnormality detection and process diagnosis, 4) One or more (preferably all) processing of motor power quality monitoring analysis by current waveform distortion analysis is performed. Hereinafter, each process will be described.

First, in step 8 (ST8), 1) rotating machine system abnormality detection and identification by high-frequency component analysis of the current waveform is, for example, a gear device, a shaft system misalignment (shaft position misalignment), This is a method for identifying an abnormality such as a balance (an axial balance deviation).
Here, with respect to the current waveform at normal time (current signal: thick line in FIG. 4) and the current waveform of the motor at the time of inspection (current signal: thin line in FIG. 4) shown in FIGS. , After performing high-pass filter processing with a high-pass filter of 1000 Hz (for example, in the range of 500 to 1500 Hz), further fast Fourier transform (FFT transform: FFT) on the current time-series signal subjected to envelope processing by envelope detection Fast Fourier Transform). The fast Fourier transform is a method for converting a signal into the frequency domain, and is a conventionally known method used for observing a frequency component and a phase (the same applies hereinafter).

Thereby, for example, the high-frequency current spectrum shown in FIGS. 5 (A) to (C) and FIGS. 6 (A) to (C) is obtained.
When the gear bearing is normal, as shown in FIGS. 5A to 5C, the high-frequency current spectrum has peaks only at frequencies that are almost even times the power supply frequency (120 Hz and 240 Hz in this case). On the other hand, if an abnormality occurs in the gear bearing or gear meshing, the region surrounded by the ellipse shown in FIGS. 6A to 6C, that is, the rotational frequency of the gear shaft is 29.7 Hz, A peak group of a high-frequency current spectrum appears in the vicinity of the even-numbered frequency (in this case, 59.4 Hz, 118.8 Hz, 178.2 Hz, 237.6 Hz, and 297 Hz). Note that a peak cannot be confirmed at a frequency that overlaps with an even-numbered peak of the power supply frequency that appears even in normal operation. Further, the vicinity of the even multiple frequency is, for example, a range of ± 10 Hz (preferably ± 5 Hz), for example, centering on the frequency of the even multiple of the rotation frequency of the gear shaft (29.7 Hz in this case) considering slippage. Means inside.

Here, the determination of abnormality may be made on condition that the peak group of the spectrum that appears in the vicinity of the frequency that is an even multiple of the rotational frequency of the gear shaft exceeds a preset value. The preset value is, for example, 10% or more, preferably 20% or more, more preferably 30% or more of the peak level (highest) that is almost an even multiple of the power supply frequency.
Therefore, by detecting a peak group of the high-frequency current spectrum in the vicinity of a frequency that is an even multiple of the rotational frequency of the gear shaft, it is possible to perform a precise diagnosis of abnormalities in the rotating machine system.
The determination as to whether or not an abnormality has occurred in the above-described gear bearing or gear meshing may be performed without performing the simple diagnosis (abnormality detection) in Steps 1 to 6 described above. That is, after sampling the current signal when the motor is in operation and performing the filter processing, the time series signal of the current subjected to the envelope processing is further subjected to a fast Fourier transform to be an even multiple of the rotational frequency of the gear shaft. A peak group of spectrum appearing near the frequency of is detected.

Also, in step 9 (ST9), 2) rotating machine system abnormality detection and identification by sideband analysis of the current waveform is, for example, stator abnormality, winding abnormality, rotor eccentricity, rotor This is a method for identifying abnormalities such as damage to the magnetic pole part, misalignment and unbalance of the shaft system.
Here, a normal current waveform (current signal) and a motor current waveform (current signal) at the time of inspection shown in FIGS. 4A to 4C are subjected to fast Fourier transform and then logarithmic conversion (log 10Z). I do. Thus, as shown in FIGS. 7A to 7C, a large value can be a small peak spectrum, while a small value can be a large peak spectrum.
Then, symmetrically with the power supply frequency as the center frequency, it is confirmed whether or not a sideband of this power supply frequency exists and what the spectrum level is. The center frequency may be, for example, the stator slot passing frequency or the rotor bar magnetic pole passing frequency. In this case, instead of the power source frequency sideband, the center frequency is equal to the rotational frequency or pole passing frequency. Check the peak level of the spectrum of the remote frequency.

Thereby, the state of a stator, a rotor, and a rotating shaft system can be diagnosed, for example.
For example, as shown in FIGS. 7A to 7C, a frequency that is shifted by ± 1 to 5 Hz centered on the power supply frequency f _L (here, 60 Hz) and pole passing frequency f _p (here, centered on 60 Hz). ) Sidebands and the difference between the peak level and the peak level of the center frequency exceeds a preset difference, it is determined that an abnormality has occurred in the rotor bar.
Further, the symmetrical power frequency as the center frequency (in this case, 30.7Hz and 89.3Hz) rotational frequency partial but only distant frequency axis of the rotor of the motor at the position of, connected by a coupling (transmission joint) A peak appears when the axis of the rotor shaft of the motor being connected (connected) and the axis of the rotation shaft on the load side are deviated.

Therefore, as shown in FIGS. 8A and 8B, whichever of the difference between the level of the spectrum peak of the power supply frequency of the motor and the level of the spectrum peak appearing at a frequency symmetrically away from the center frequency is selected. On the other hand, it is determined that an alignment error has occurred, provided that one of the values is equal to or less than a preset value. Here, the difference in the level of each spectral peak is within a range of 30 to 50 decibels (hereinafter also referred to as dB. Here, 40 dB) (preferably, the lower limit is 36 dB, depending on the required detection accuracy. The upper limit can be set at 48 dB).
Here, dB is represented by 20 log (I _line / I _shift ). In addition, I _line is a current component of the current spectrum power supply frequency part, and I _shift is a current component of the axial rotation frequency sideband part centering on the current spectrum power supply frequency.
Therefore, it is possible to carry out precise diagnosis of abnormalities in the rotating machine system.
The determination as to whether or not the above-described alignment abnormality has occurred may be performed without performing the simple diagnosis (abnormality detection) in Steps 1 to 6 described above. That is, after sampling the current signal during operation of the motor, performing fast Fourier transform, logarithmically transforming the spectrum, the level of the spectrum peak of the power frequency of the motor and the rotation frequency of the shaft of the rotor of the motor from the power frequency The difference from the level of the spectrum peak appearing at the position of the remote frequency is obtained.

3) Abnormality detection and process diagnosis by pattern analysis of excessive current values performed in step 10 (ST10) are methods for diagnosing the state and process of the driven rotating machine.
Here, the starting current analysis and the operating current analysis shown in FIGS. 9A to 9C, and the breaking current analysis and the operating current analysis shown in FIGS. 9D to 9F are performed. This starting current analysis is a method of comparing (comparing) the starting current pattern (current effective value waveform, the same applies hereinafter) in a normal state with no abnormality in the rotating machine system and the current effective value waveform during operation of the motor. Yes, interrupting current analysis is a method of collating normal interrupting current pattern with motor RMS current waveform during motor operation, and operating current analysis is normal operating current pattern and motor operation This is a method of collating with the current RMS value waveform at the time.
Here, when the current waveform of the motor at the time of inspection exceeds ± 10% (preferably ± 5%) with respect to the current pattern in the normal state, it can be precisely diagnosed that an abnormality has occurred in the rotating machine system.

Steps 8 to 10 described above are performed simultaneously or sequentially, and if an abnormal pattern is not recognized in step 11 (ST11), the process returns to step 2 and a simple diagnosis is repeated to determine that an abnormal pattern is recognized. If so, the abnormality identification result is displayed on the monitor of the processing unit 17 in step 14 (ST14).

In addition, the monitoring analysis of the power quality of the motor by the distortion analysis of the current waveform, which is performed in step 12 (ST12), includes the harmonic component of the current signal during the operation of the motor and the monotone ratio of the power frequency component and the total harmonics. This analysis is performed by calculating the wave ratio and the unbalance rate of the interphase current.
Here, the current waveform (current signal) at normal time and the current waveform (current signal) during operation of the motor shown in FIGS. 4 (A) to 4 (C) are shown in FIGS. 10 (A) to 10 (C). Thus, harmonic (distortion) analysis up to the 50th order (50 times) of the power supply frequency of the electric motor is performed. In addition, the vertical axis | shaft of FIG. 10 (A)-(C) is a ratio of each harmonic with respect to a fundamental wave.
The ideal power supply quality is such that the power supply frequency f _L component of the current (voltage) signal is overwhelmingly strong and the harmonic component of the power supply frequency hardly appears.

For this reason, the ratio between the harmonic component up to the 50th order and the power supply frequency component, that is, the monotone ratio i _dis and total harmonic ratio (total current distortion factor) i _{dis of the} current signal, and the imbalance of the interphase current, that is, the current The unbalance rate I _ub is obtained. The monoharmonic ratio i _dis , the total harmonic ratio i _dis , and the current imbalance ratio I _ub are expressed by equations (7) to (9), respectively.
i _dis = {effective value of n-th harmonic current (In)} / {basic current (I ₁ )} × 100 (%)
... (7)
i _dis = {effective value of harmonic current} / {basic current (I ₁ )} × 100 (%) (8)

The effective value of the harmonic currents, when the effective value of the n-th harmonic current and I _n, are revealed by the formula (10).

Then, in step 13, monotonic wave ratio _{i dis} and total harmonic ratio _{i dis} and current imbalance factor _I reference values all previously set the _ub (e.g., monotonic wave ratio _{i dis} 3%, the total harmonic ratio i _{If the dis} and the current imbalance rate I _ub do not exceed 5%), it is diagnosed that no abnormality has occurred in the quality of the power supplied to the motor and the inverter, and the process returns to step 2 to perform simple diagnosis repeatedly. On the other hand, if any one of the monoharmonic ratio i _dis , the total harmonic ratio i _dis , or the current unbalance ratio I _ub is greater than the reference value, it is diagnosed that an abnormality has occurred in the power quality or the inverter, 14 (ST14), the abnormality identification result is displayed on the monitor of the processing unit 17.
Note that step 12 can also be performed simultaneously or sequentially with steps 8-10.
By the above method, it is possible to set the criterion and identification of abnormality determination, and it is possible to carry out abnormality diagnosis of a rotating machine system with higher sensitivity than before.

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, a case where the rotating machine system abnormality diagnosis method 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.

10: rotating machine system abnormality diagnosis device, 11: electric motor, 12: mechanical system, 13: power line, 14: power supply, 15: current detector, 16: A / D converter, 17: processing unit

Claims (2)

In an abnormality diagnosis method for a rotating machine system driven by an electric motor,
The current signal during operation of the motor is sampled, subjected to fast Fourier transform, and then the spectrum is logarithmically converted. The level of the spectrum peak of the power frequency of the motor is symmetrical with respect to the power frequency as the center frequency. The rotational mechanical system is connected by a transmission joint on the condition that the difference from the level of the spectrum peak appearing at the frequency position separated by the rotational frequency of the child shaft is within the range of 30 to 50 decibels. An abnormality diagnosis method for a rotating machine system, characterized in that it is determined that an alignment abnormality has occurred between the axis of the rotor of the motor and the rotation axis on the load side. The abnormality diagnosis method for a rotating machine system according to claim 1, wherein the rotating machine system includes a gear device on the load side of the electric motor.

Images