JP2011133362A - Shaft system stability measuring method and operation method of rotary machine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately evaluate the stability of a rotary shaft system, by extracting a characteristic vibration component of the rotary shaft system at operation. <P>SOLUTION: A method includes a first step of measuring vibration generated during operation of a rotary machine by a rotating section or a non-rotating section; a second step of performing frequency analysis by paying attention to a characteristic frequency of the rotary machine relative to a vibration signal measured in the first step; a third step of acquiring a frequency response, by executing equalization processing continuously for a prescribed time so that a frequency component of the characteristic frequency subjected to frequency analysis in the second step can be extracted; and a fourth step of evaluating stability of the rotary shaft system, from the frequency response obtained in the third step. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a shaft system stability measurement method and an operation method of a rotating machine that detects a natural vibration component of a rotation shaft system during operation and evaluates the stability of the rotation shaft system.

  There are various destabilizing factors in the rotating shaft system of a rotating machine, and if the stability of the rotating shaft is impaired depending on the operating state, vibration may increase. Factors that impair the stability of the rotating shaft include, for example, oil whip and oil whirl due to insufficient bearing load, and steam whirl caused by the swirling flow of steam in the steam turbine.

  All of these phenomena are caused by a decrease in the attenuation rate of the rotating shaft system, and the stability evaluation of the rotating shaft system is a very effective means for ensuring the safety of the rotating machine.

  However, the damping rate obtained from the natural vibration measurement at rest is a value of the structure, and does not include the effects of bearings and steam flow, which are unstable factors, and cannot be used for direct evaluation.

  On the other hand, during the operation of the rotating machine, the vibration amplitude at the rotation speed and its integral multiple is dominant, so the normal vibration measurement extracts the free vibration component necessary for evaluating the stability of the rotating shaft system. Is difficult.

  Conventionally, as a method for evaluating the stability of a rotating shaft system, an electromagnet is used to forcibly apply vibration to a rotating shaft that is in operation, thereby causing free vibration in the rotating shaft and measuring the attenuation rate. (For example, patent document 1) and the vibration characteristic evaluation method (for example, patent document 2) using the dark vibration which generate | occur | produces during a driving | running | working of a rotary machine without performing vibration.

Japanese Patent Publication No. 5-5057 US Patent No. 7078434

  However, the attenuation factor measuring method described in Patent Document 1 requires a large-scale vibration device and generates a vibration as large as an unbalanced vibration. Is not suitable.

  In addition, in the vibration characteristic evaluation method described in Patent Document 2, since the vibration level is low near the natural frequency of dark vibration during normal operation and a random response is obtained, the damping ratio is accurately evaluated from a real-time response curve. It is difficult.

  An object of the present invention is to provide a shaft system stability measurement method and an operation method of a rotating machine that can extract the natural vibration component of the rotation shaft system during operation and accurately evaluate the stability of the rotation shaft system. There is.

  In order to achieve the above object, the present invention measures the shaft system stability of a rotating machine by the following method.

(1) Focusing on the natural frequency of the rotating machine with respect to the vibration signal measured in the first step of measuring the vibration generated during the operation of the rotating machine at the rotating part or the non-rotating part. And a second step of performing frequency analysis, and a frequency response is obtained by performing averaging processing continuously for a predetermined time so that the frequency component of the natural frequency frequency-analyzed by the second step can be extracted. And a fourth step of evaluating the stability of the rotating shaft system from the frequency response obtained by the third step, and measuring the shaft system stability of the rotating machine.

(2) In the shaft system stability measurement method of the rotary machine, a fifth step is provided for obtaining an attenuation factor at the frequency from the frequency response obtained by the third step, and the fourth step is the fifth step. The stability of the rotating shaft system is evaluated from the magnitude of the attenuation rate obtained in the above step.

(3) In the shaft system stability measurement method of the rotating machine, from the change in stability of the rotating shaft system evaluated in the fourth step with respect to the change in operating conditions of the rotating machine up to the present, a future rotating shaft A sixth step for estimating the stability of the system is provided.

(4) Operate by adjusting the operating conditions of the rotating machine from the result of the rotating shaft system stability evaluated by the shaft system stability measuring method of the rotating machine.

  According to the present invention, it is possible to accurately evaluate the stability of the rotating shaft system by extracting the natural vibration component of the rotating shaft system during operation.

The block diagram which shows an example in 1st Embodiment for demonstrating the shaft-system stability measurement method of the rotary machine by this invention. The block diagram which shows the other example in the embodiment. The graph which shows the analysis result of the vibration of the rotating shaft by the conventional method. The graph which shows the frequency analysis result of the vibration of the rotating shaft in the 1st Embodiment of this invention. 4 is a graph showing an attenuation ratio estimation error when the averaging processing time is 1 minute in the embodiment. Similarly, a graph showing an attenuation ratio estimation error when the averaging processing time is 5 minutes. 4 is a graph showing the relationship between the averaging processing time and the maximum error in the embodiment. In the same embodiment, the curve figure which shows the relationship between the frequency and amplitude for demonstrating stability evaluation of a rotating shaft. In the same embodiment, the curve figure which shows the relationship between the turbine output and attenuation factor for demonstrating stability evaluation of a rotary machine. The figure which shows the relationship between the frequency and amplitude by the filter process in 2nd Embodiment and 3rd Embodiment for demonstrating the shaft system stability measuring method of the rotary machine by this invention. The figure which shows the relationship between the turbine output and attenuation factor in embodiment for demonstrating the operating condition determination method of the rotary machine by this invention.

  Embodiments of the present invention will be described below with reference to the drawings.

(First embodiment)
FIG. 1 is a configuration diagram showing an example in the first embodiment for explaining a shaft system stability measuring method of a rotating machine according to the present invention.

  In FIG. 1, reference numeral 1 denotes a rotary machine, and a rotary shaft 2 of the rotary machine 1 is rotatably supported by bearings 3 provided at both ends thereof. In such a rotating machine 1, the non-contact displacement meter 4 is provided on a support body (not shown) so as to correspond to the rotating shaft 2 outside the bearing 3 of the rotating shaft 2.

  This non-contact type displacement meter 4 detects the displacement of the rotating shaft 2, and the detected value is taken into the converter 5 and converted into a voltage signal. The voltage signal output from the converter 5 is input to the FFT analyzer 7 through the filter 6 and subjected to frequency analysis.

  When the frequency analysis result obtained by the FFT analyzer 7 is taken into the arithmetic device 8, the arithmetic device 8 evaluates the stability of the rotating shaft 2 based on the frequency analysis result by arithmetic processing described later in detail.

  FIG. 2 is a configuration diagram showing another example of the embodiment.

  In FIG. 1, a non-contact displacement meter 4 is provided on both ends of the rotating shaft 2 of the rotating machine 1, and the displacement of the rotating shaft 2 detected by the non-contact displacement meter 4 is converted into a voltage signal by the converter 5. However, since the vibration of the rotating shaft 2 is propagated to the bearing 3, the acceleration is applied to the bearing 3 supporting both ends of the rotating shaft 2 in FIG. A pickup 9 is provided in contact, and the acceleration extracted by the acceleration pickup 9 is amplified as a voltage signal by the vibration meter 10 and then input to the FFT analyzer 7 via the filter 6.

  The shaft system stability measuring method of the rotating machine in the first embodiment of the present invention configured as described above will be described.

  First, the case where the frequency of the vibration of the rotating shaft 2 is analyzed by the FFT analyzer 7 by a conventional method will be described.

  FIG. 3 shows a frequency response waveform when the vibration of the rotating shaft 2 generated during the operation of the rotating machine is subjected to frequency analysis by the FFT analyzer 7 by a conventional method. The horizontal axis represents frequency and the vertical axis represents amplitude. ing.

Usually, since the vibration of the rotating shaft 2 has a large unbalanced vibration, the amplitude at the rotation frequency f _R of the rotating shaft 2 is remarkably large as seen in FIG. 3, and the other frequency components are hardly distinguished. There wasn't.

  Further, no attention has been paid to frequencies other than a specific frequency at which the amplitude grows at the time of abnormality such as a rotation synchronization frequency or an integer multiple of the rotation frequency.

  Furthermore, the natural frequency of the rotating shaft 2 is noted when passing through a critical speed, but this is because it coincides with the rotational frequency, and is not considered unless a resonance problem occurs during steady operation. It was.

However, focusing on the vicinity of the natural frequency f _n of the rotating shaft 2, the presence of an amplitude peak can be recognized although it is at a very low level compared to the rotational synchronization component. This is smaller than the unbalanced excitation force of the rotating shaft 2, but the rotation of the rotating shaft 2 randomly generates vibrations in a wide band frequency, and free vibration is generated in the rotating shaft 2, so that the natural vibration of the rotating shaft 2 is generated. It is thought that the number was excited.

  Next, in the first embodiment of the present invention, the case where the frequency of the vibration of the rotating shaft 2 is analyzed by the FFT analyzer 7 will be described.

  In recent years, as the performance of measuring instruments has been improved, the dynamic range can be increased, so that even such small vibrations can be analyzed with relatively high accuracy.

FIG. 4 is an enlarged view of the vicinity of the natural frequency f _n in FIG. The broken line shown in FIG. 4 is a response waveform when the frequency of the vibration of the rotating shaft 2 is analyzed in a normal analysis time.

  In normal vibration measurement, the analysis time of the FFT analyzer 7 is about several seconds, and is generally several minutes at the longest. Even when monitoring for a long time, the above-mentioned short-time measurement is repeated, and the measurement time does not coincide with the analysis time.

Although an increase in amplitude is recognized in the vicinity of the natural frequency f _n as described above, the excitation force is weak and random, so that the peak is clear enough to accurately identify the natural frequency from the response waveform in this band. I can't figure out.

  This is also considered to be one of the factors for which attention has not been paid to the natural frequency during operation so far, but the response waveform by random excitation improves measurement accuracy by repeating the averaging process many times. Therefore, when the analysis by the FFT analyzer 7 was continuously continued for a long time, a clear peak waveform could be obtained as shown by the solid line in FIG.

Since this waveform shows the frequency response of the free vibration of the rotating shaft 2 by random excitation, the damping rate ζ is calculated by obtaining the natural frequency f _n from the maximum point and applying the half power method to the response waveform. can do.

That is, since the value obtained by dividing the vibration amplitude Δf of the response waveform at the half amplitude of the peak by the natural frequency f _n is equal to 1 / 2ζ, the damping rate ζ can be calculated. In addition, the natural frequency f _n and the damping rate ζ can be obtained even by applying the curve fitting method.

  In order to accurately evaluate the attenuation rate of the shaft system by the above-described method, it is necessary to appropriately set the time for performing the averaging process.

  Fig. 5 shows (1) frequency analysis, (2) averaging analysis results for 1 minute, and (3) half power for shaft vibration data obtained by simulation calculation for a shaft / bearing system simulating a steam turbine. Attenuation ratio estimation error is repeatedly performed, and attenuation ratio estimation errors are plotted.

  In the simulation calculation, it was assumed that a random excitation force with white noise-like frequency characteristics was acting on the axis. It can be seen that an error of about 0.4% occurs at the maximum.

  Similarly, FIG. 6 shows an error plotted by performing an averaging process for 5 minutes. In this case, the maximum error is an accuracy of 0.2% or less.

  FIG. 7 is a plot of the relationship between the averaging process time and the maximum error. The longer the averaging process time is, the smaller the maximum error is. However, in 20 minutes, the error is 0.1% or less. You can see that

  In this simulation, since data for 4 seconds is required for one frequency analysis, averaging processing is performed 75 times in 5 minutes (300 seconds) and 300 times in 20 minutes (1200 seconds). . Therefore, it is desirable that the averaging processing time is at least 5 minutes or more, preferably 20 minutes or more. However, when the frequency of analysis and the natural frequency of the machine to be evaluated are greatly different from those of the steam turbine and the frequency analysis time per time changes, it is practical to perform the averaging process for a time equivalent to 75 times of averaging. An accuracy with no problem can be obtained, and a more preferable accuracy can be obtained if the time is equal to or more than 300 times of averaging.

When the dynamic system of the rotating shaft 2 changes due to changes in the load / lubricating oil temperature of the bearing 3 and the pressure / temperature of the fluid flowing inside, such as steam, the damping ratio ζ changes. Unstable vibration is also called self-excited vibration, and is caused by negative damping of the shaft system. Mathematically, the equation of motion is stable if the sign of the real part of the complex eigenvalue λ _i is negative, and unstable if it is positive.

If the angular natural frequency of the rotating shaft 2 and _{ω i, λ i = -ζ i} ω i relation holds for, omega _i> 0 the stability of the rotary shaft 2 from the value of the attenuation factor zeta _i since it is the Can be evaluated. That is, during normal operation, the value of the damping rate ζ _i is several percent, but when an unstable factor occurs due to an increase in the output of the turbine, the value gradually decreases.

FIG. 8 shows a process in which the attenuation of the rotating shaft 2 is lowered by the steam whirl and the stability is impaired. In FIG. 8, the solid line is a waveform in a stable state, and the peak amplitude is low and the base is wide, but when approaching an unstable state, the peak frequency (natural vibration) is shown as indicated by a broken line. The number f _n ′) slightly decreases and the amplitude increases. On the other hand, the bottom of the waveform changes to a narrow waveform, indicating that the attenuation factor is reduced (in this example, it is reduced to 1/2).

Here, in FIG. 1 or FIG. 2, the arithmetic unit 8 obtains the natural frequency f _n and the damping rate ζ based on the frequency analysis result obtained by the FFT analyzer 7. The change in the attenuation rate ζ is monitored to determine the stability state.

  In conventional monitoring, it is common to make a determination from the vibration amplitude of the rotating shaft 2, but even if there is an unstable sign as described above, the natural frequency component is much higher than the rotational frequency component. Because it is small, there is almost no effect on the vibration amplitude. For this reason, an abnormality cannot be detected until an unstable phenomenon occurs.

  On the other hand, in the present invention, it is possible to quantitatively evaluate the degree of progress of the unstable phenomenon by monitoring the attenuation rate. Generation of vibration can be avoided.

  FIG. 9 is a plot of the relationship between output and attenuation rate from past operation records in a steam turbine. When an approximate curve of the attenuation rate ζ with respect to the output P is obtained based on this actual value, a curve as shown in the figure is obtained.

As described above, instability occurs when the sign of the attenuation rate ζ is reversed from positive to negative, and the intersection with the horizontal axis is the stability limit. That is, since it is predicted that an unstable phenomenon will occur when the output exceeds P _u , it is possible to limit the operation below the output P _u .

As described above, in the first embodiment of the present invention, the frequency analysis is performed by paying attention to the natural vibration f _n of the rotating machine among the vibrations generated during the operation of the rotating machine, and the frequency analysis result is continuously averaged. By calculating the attenuation rate ζ of the frequency from the frequency response obtained by performing the conversion processing and monitoring the magnitude of the attenuation rate ζ, the stability of the rotating shaft system can be accurately evaluated. In addition, by monitoring the change in the damping factor ζ of the frequency obtained from the frequency response during operation, the stability of the future rotating shaft system is estimated from the change in the stability of the rotating shaft system with respect to changes in the operating conditions up to now. By doing so, an abnormal operation state can be avoided from the stability of the rotating machine 1 and a safe operation of the rotating machine can be ensured.

(Second Embodiment)
FIG. 10 is a diagram showing the relationship between the frequency and the amplitude by the filter processing in the second embodiment for explaining the shaft system stability measuring method of the rotating machine according to the present invention.

In the second embodiment of the present invention, in the configuration shown in FIG. 1 or FIG. 2, a bandpass filter having a transmission frequency width Δf _{P with} the natural frequency f _n as shown in FIG. At this time, the transmission frequency width Δf _P is set so that the frequency to be removed does not enter the transmission region, but it is necessary to take at least the frequency width Δf required by the half power method.

As described above, in the measured vibration data, the amplitude at the rotation frequency f _R is much larger than the amplitude at the natural vibration frequency f _n . Since the dynamic range of the measuring instrument is limited, the accuracy of the natural frequency component inevitably deteriorates.

Therefore, by using a bandpass filter having a transmission frequency width Δf _P centered on the natural frequency f _n as the center of the filter 6 as in the present embodiment, only signals in the necessary frequency range are input to the FFT analyzer 7. , Greatly improve the dynamic range.

  As described above, in the second embodiment of the present invention, when frequency analysis is performed on the vibration generated during the operation of the rotating machine 1 while paying attention to the natural vibration frequency of the rotating machine, the periphery of the frequency is input to the input of the vibration signal. By performing band pass filter processing on the band to improve the gain of the frequency band, the accuracy of the frequency response is improved, so that the averaging processing time can be shortened.

(Third embodiment)
In a third embodiment of the present invention, in the configuration shown in FIG. 1 or FIG. 2, a band eliminator Ted filter removal range Delta] f _e as shown in FIG. 10 to the filter 6.

  By using such a band-eliminated filter, it is possible to perform frequency analysis by removing frequency components unnecessary for extraction of natural vibration components such as rotational frequency components.

  As described above, in the third embodiment of the present invention, when frequency analysis is performed by paying attention to the natural vibration frequency of the rotating machine for the vibration generated during the operation of the rotating machine 1, a frequency other than that frequency is input to the vibration signal. By applying band-eliminated filter processing to the frequency component and improving the gain of the frequency band, the accuracy of the frequency response is improved as in the second embodiment, so the averaging processing time is shortened. be able to.

(Fourth embodiment)
In the fourth embodiment of the present invention, when performing frequency analysis with the FFT analyzer 7 shown in FIG. 1 or 2 in the first embodiment, zooming processing with the natural vibration frequency fn as the center frequency is performed.

  When such zooming processing is performed, the number of samplings can be reduced while maintaining the frequency resolution with respect to the original analysis conditions, so that the analysis time per time can be shortened. Therefore, if the number of times of averaging is the same, the averaging time is shortened.

  As described above, in the fourth embodiment of the present invention, when the frequency analysis is performed by paying attention to the natural vibration frequency of the rotary machine for the vibration generated during the operation of the rotary machine 1, the zooming with the peripheral band of the frequency as the analysis range is performed. By performing the processing to improve the frequency resolution, the averaging processing time can be shortened and an abnormality can be detected early.

(Fifth embodiment)
In the fifth embodiment of the present invention, in the first embodiment, the attenuation rate and the operation condition are adjusted from the operation history obtained in the past by the arithmetic device 8 shown in FIG. 1 or FIG.

FIG. 11 shows the correlation between the damping rate ζ and the turbine output P when the bearing oil supply temperature is L ₁ (×) and L ₂ (•).

In FIG. 11, approximate curves obtained from the respective plots are indicated by broken lines and solid lines. Although the stability limit of the output in the bearing oil supply temperature L ₁ is P _u1, stability limit of the output to alter the bearing oil supply temperature to the L ₂ it can be seen that can stretch to P _u2. Therefore, when operating at the bearing oil supply temperature L ₁ , the operating condition is changed so as to change the bearing oil supply temperature to L ₂ when the output increases and approaches _{Pu 1} .

  As described above, in the fifth embodiment of the present invention, the shaft system stability is evaluated, and by adjusting the bearing oil supply temperature as the operating condition of the rotating machine, that is, the operating condition for increasing the stability limit based on the result. Since the operating conditions can be changed so as to suppress the occurrence of the unstable phenomenon, it is possible to operate the rotary machine while ensuring the stability.

  DESCRIPTION OF SYMBOLS 1 ... Rotary machine, 2 ... Rotary shaft, 3 ... Bearing, 4 ... Non-contact displacement meter, 5 ... Converter, 6 ... Filter, 7 ... FFT analyzer, 8 ... Arithmetic device, 9 ... Accelerometer, 10 ... Vibrometer

Claims (8)

A first step of measuring vibration generated during operation of the rotating machine at the rotating part or the non-rotating part;
A second step of performing frequency analysis on the vibration signal measured in the first step, focusing on the natural frequency of the rotating machine;
A third step of obtaining a frequency response by performing an averaging process continuously for a predetermined time so that a frequency component of the natural frequency subjected to frequency analysis by the second step can be extracted;
A fourth step of evaluating the stability of the rotating shaft system from the frequency response obtained by the third step;
A shaft system stability measuring method for a rotating machine, comprising: In the shaft system stability measuring method of the rotary machine according to claim 1,
A fifth step of determining an attenuation factor at the frequency from the frequency response obtained by the third step is provided;
In the fourth step, the stability of the rotating shaft system is evaluated from the magnitude of the attenuation rate obtained in the fifth step. In the shaft system stability measuring method of the rotating machine according to claim 1 or 2,
When the vibration signal measured in the first step is input, a bandpass filter process is performed on the peripheral band of the frequency to improve the gain of the frequency band. System stability measurement method. In the shaft system stability measuring method of the rotating machine according to claim 1 or 2,
When the vibration signal measured in the first step is input, the rotating machine is characterized in that a band-eliminated filter process is performed on a peripheral band of the frequency to improve a gain of the frequency band. Of measuring the stability of the shaft system. In the shaft system stability measuring method of the rotary machine in any one of Claims 1 thru | or 4,
A shaft system stability measurement method for a rotating machine, wherein when performing frequency analysis in the third step, a frequency resolution is improved by performing a zooming process in which a peripheral band of the frequency is an analysis range. In the shaft system stability measurement method of the rotary machine in any one of Claims 1 thru | or 5,
A sixth step is provided for estimating the future stability of the rotating shaft system from the change in stability of the rotating shaft system evaluated in the fourth step with respect to the change in operating conditions of the rotating machine to date. A method for measuring the stability of a shaft system of a rotating machine.   It operates by adjusting the operating condition of a rotating machine from the result of the rotating shaft system stability evaluated by the shaft system stability measuring method of the rotating machine according to any one of claims 1 to 6. How to operate a rotating machine. In the operating method of the rotary machine of Claim 7,
A method for operating a rotating machine, wherein the operating condition of the rotating machine is adjusted by adjusting a bearing oil supply temperature of the rotating part.

Images