JPH1130551A - Vane vibration-measuring apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vane vibration-measuring apparatus that can measure vibrations highly accurately in a short time. SOLUTION: In the vane vibration-measuring apparatus, time series data 22 based on a rotary vane showing time series vibration information in a rotational direction for every vane are obtained on the basis of output results of a plurality of first electromagnetic sensors set via an equal distance on the stationary side opposite to a leading end part of the rotary vane and an output result of a second electromagnetic sensor set on the stationary side opposite to a base part of the rotary vane. The time series data 22 based on the vane for every rotary vane is processed by fast Fourier transform at an FFT operator 12. All resulting data 23-1-23-M are added at an adder 13.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a blade vibration measuring device for analyzing the vibration of a rotor blade incorporated in various turbines, fans, pumps and the like.

[0002]

2. Description of the Related Art In research, design, development, and manufacture of various turbomachines such as steam turbines, gas turbines, axial fans, water turbines, and pumps, in order to improve the performance and reliability of products, rotating bodies and the like are used. It is necessary to understand the vibration of the internal structure in detail. In response to such demands,
Use electromagnetic sensors with magnets, laser, infrared,
Non-contact vibration measurement technology that measures the vibration of a rotating body without directly contacting the rotating body by using an optical method that uses visible light etc. has been developed, and is applied to the vibration measurement of many turbomachines. ing. For example, a non-contact vibration measurement technology using an electromagnetic sensor is disclosed in Japanese Patent Application No. 5-229.
No. 477 and Japanese Patent Application No. 6-217045.

FIG. 8 shows a blade vibration measuring device disclosed in Japanese Patent Application No. 6-217045, in which a rotary pulse detector 3 is mounted on a stationary side of the rotor 1 opposite to the shaft, and a tip of a rotary blade 52 is mounted. 2 ⁿ (n: positive integer) electromagnetic pickups 55 are attached at regular intervals on the stationary side opposite to
An electromagnetic pickup 54 is attached to the stationary side opposite to the base of the rotor, and air is injected toward the rotating blades 52 to rotate the rotor 1 to excite the rotating blades 52. 52 is to measure the vibration.

In this case, the output signals of the rotation pulse detector 53 and the electromagnetic pickups 54 and 55 are input to a primary processing device 56, respectively. The temporary processing device 56 calculates and displays the number of rotations of the rotor 52 from the detection result of the rotation pulse detector 53, and calculates the time during which the rotor 52 rotates by the blade pitch from the detection result of the electromagnetic pickup 54, From the detection results of the electromagnetic pickups 55 and 54, the transit time difference between the tip and the base of the rotary wing 52 is determined, and based on these results, the vibration data in the rotational direction of the rotary wing 52 is determined and output.

At this time, 2 ⁿ electromagnetic pickups 55
Output the time-series vibrations in the direction of the rotation axis when the rotor 52 on the rotor 51 sequentially passes in the vicinity of itself, so that the primary processing unit 56 2 ⁿ sets of time-series vibration data (hereinafter, referred to as sensor-based time-series data) in the rotation direction when each rotor 52 sequentially passes in the vicinity are output in accordance with the number of electromagnetic pickups 55. .

The 2 _n sets of sensor-based time-series data are input to the secondary processing unit 57 via an interface for performing DMA (Direct Memory Access) transfer. The secondary processing device 57 is for performing various types of vibration analysis online, and is configured as shown in FIG.

In the secondary processing device 57, first, the data reordering device 61 rearranges the vibration data included in the 2 ⁿ sets of sensor-based time-series data 71 input from the primary processing device 56, and reorders the data by the number of rotor blades 52. Is output as time-series data 72 based on the wing. The wing-based time-series data 72 is
The time series shows vibration data when a certain rotor 52 sequentially passes near 2 ⁿ electromagnetic pickups 55.

Thereafter, a suitable one of the rotary blades 52 is monitored, and the FFT calculator 62 performs a fast Fourier transform on the blade-based time-series data 72 corresponding to the rotary blade 52. Then, the data 73 obtained as a result is analyzed by the analyzer 63, and the analysis result is represented by a Campbell diagram,
A natural frequency of the rotor 52 is obtained by showing a tracking analysis diagram, a mode diagram, and the like. FIG. 10 is an example of a Campbell diagram obtained in this manner, where the horizontal axis represents the number of rotations,
The right vertical axis represents harmonics, and the left vertical axis represents vibration frequency.

[0009]

As described above, in the conventional blade vibration measuring apparatus, when a vibration test is performed, one rotor is monitored and analyzed. At the position, the vibration amplitude becomes zero and the natural frequency of the rotor may not be known. In addition, since several tens of rotors are mounted on the rotor in one row, it takes time to perform analysis while searching for individual blades having a large vibration amplitude.

In addition, the rotor is vibrated by injecting air with an air nozzle. However, if the amount of air to be supplied is increased in order to increase the exciting force, the frictional heat between the rotor and air causes a vacuum. There was a risk that the equipment would be destroyed due to an increase in the indoor atmosphere temperature. Therefore, while the theoretical resolution of the conventional blade vibration measuring device is about 2 μm, the vibration amplitude of the rotating blade is about 10 μm at the maximum, and there is a problem that only about five times the measurement accuracy can be obtained. An object of the present invention is to provide a blade vibration measuring device capable of performing high-precision vibration measurement in a short time.

[0011]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention relates to a blade vibration measuring device for measuring the vibration of a plurality of rotors in a non-contact manner. A plurality of first vibration measuring means for measuring vibration in the direction of the rotation axis of the tip of the rotor, and a plurality of first vibration measuring means attached to a stationary side opposite to the base of the rotor, and Second vibration measuring means for measuring vibration, means for obtaining time-series vibration information in the rotation direction for each rotor based on the measurement results of the first and second vibration measuring means, And a means for performing a fast Fourier transform on all of the obtained vibration information for each rotor, and a means for adding all the Fourier transform results obtained by the means.

[0012]

FIG. 1 is a diagram showing a configuration of a blade vibration measuring device according to one embodiment of the present invention. This blade vibration measuring device is for measuring the vibration of the rotor blade 2 of the rotor 1 in a non-contact manner, and includes a rotating pulse detector 3, electromagnetic pickups 4, 5, a primary processing device 6, and a secondary processing device 7. It is configured.

The rotor 1 is provided with a plurality of stages of cascades along the direction of the rotation axis, and the cascade of each stage is composed of M rotating blades 2 at equal intervals along the circumferential direction of the rotor 1. Is attached.

On the stationary side facing the shaft of the rotor 1,
The rotation pulse detectors 3 are attached. Also,
On the stationary side opposite to the base of the rotor 52, two electromagnetic pickups 4 are arranged along the circumferential direction of the rotor 1 by (k + 0.5) times the blade pitch of the rotor 2 (k is an arbitrary positive constant). ), And N (N = 2 ⁿ : n is a positive integer) electromagnetic pickups 5 are provided along the circumferential direction of the rotor 1 on the stationary side facing the tip of the rotor 2. Installed at intervals.

When a vibration test is performed using this blade vibration measuring device, air is injected toward the rotating blades 2 as in the prior art, the rotor 1 is rotated, and the rotating blades 52 are vibrated. The vibration of the rotor 52 at that time is measured.

In this case, the rotating pulse detector 3 is a rotating blade 2
One pulse is output to the primary processing device 6 for one rotation of the. Each of the two electromagnetic pickups 4 detects the vibration in the rotation axis direction at the base of each of the rotors 2 when the M rotors 2 sequentially pass in the vicinity of itself, and compares the result with the magnitude of the vibration. With the amplitude thus set, it is output as a time series signal having a frequency corresponding to the number M of the rotating blades 2 and the number of rotations. Also, N
Similarly, when the M rotor blades 2 sequentially pass in the vicinity of themselves, each of the electromagnetic pickups 5 detects the vibration in the direction of the rotation axis at the tip end of each rotor blade 2 and determines the magnitude of the vibration. And outputs a time-series signal having a frequency corresponding to the number M of the rotating blades 2 and the number of rotations with an amplitude proportional to the number of rotating blades 2.

The output signals of the rotation pulse detector 3, the electromagnetic pickup 4, and the electromagnetic pickup 5 are input to the primary processing device 6, respectively. Primary processing unit 6
Calculates vibration data in the rotational direction of the rotor 2 based on these signals, and outputs the data as time-series data based on N sets of sensors described later.

The time series data based on these sensors is D
The data is input to the secondary processing device 7 via an interface that performs MA (Direct Memory Access) transfer, and various vibration analyzes are performed online in the secondary processing device 7. Hereinafter, the primary processing device 6 and the secondary processing device 7 will be described in detail.

FIG. 2 is a block diagram showing a specific configuration of the primary processing device 6. As shown in the figure, the primary processing device 6 includes a counter X, N counters Y, N counters Z, a counter 30, a DMA controller 31, and a buffer 32. It is also assumed that a clock signal generator (not shown) supplies reference clock signals of several tens to several hundreds of megahertz to the counters X, Y, and Z, respectively.

Here, the output signal A of the rotation pulse detector 2, the output signals B and B 'of the two electromagnetic pickups 4 and the output signals C ₁ to C of the N electromagnetic pickups 5 are supplied to the primary processing device 6. Consider that C _N has been input according to a timing chart as shown in FIG. The output signal _{B, B ', C 1 ~C} N is assumed to be shaped in advance rectangular wave through a waveform shaping circuit, not shown.

In this case, the counter 30 determines the number of revolutions T ₀ of the rotor 2 based on the output signal A, and outputs the result to the DMA controller 31. Further, this rotation speed T ₀ is displayed on an appropriate display device.

The counter X counts a clock signal by using one of the output signals B and B '(here, the output signal B) as a gate signal, thereby generating pulses B ₁ and B ₂ of the output signals B and B'. , ..., determine the period T ₁ of the B _M, and outputs the result to the counter Z and the DMA controller 31. A cycle T _{1 as} a counting result of the counter X indicates a time during which the rotary blade 2 rotates by the blade pitch. Further, since the fluctuation in the axial direction is small at the base of the rotor 2 to which the electromagnetic pickup 4 is opposed, this period T ₁ is obtained as a stable value.

The counter Y has output signals C _{1 to} C _N , respectively.
And the output signal B, by counting the clock signal as one and the gate signal B ', the output signal C ₁ ~
C _N Each pulse of the output signal B of, obtains a phase difference ΔT _C1 ~ΔT _CN with any one of each pulse of B ', and sequentially outputs the result to the counter Z. For example, from the top counter Y in the figure to which the output signal C ₁ is input, the counting result for the phase difference ΔT _C1 between each pulse of the output signal C ₁ and each pulse of the output signal B ′ is time-series. Is output to

Here, the phase differences ΔT _{C1 to} ΔT _CN , which are the counting results of the counter Y, indicate the transit time difference between the tip and the base when the rotor 2 passes near the electromagnetic pickup 4. This corresponds to the magnitude of vibration in the rotation direction of the rotary wing 2. Each counter Y selects one of the output signals B and B 'so that the transit time difference does not become zero or approaches the time difference corresponding to the blade pitch.

The counter Z sequentially divides the phase differences ΔT _{C1 to} ΔT _CN input in time series from the corresponding counter Y by the period T ₁ input from the counter X, thereby forming the rotor 2. N pieces of vibration data Δθ ₁ ~Δθ _N indicating the vibration in the rotational direction of the sequentially calculated output as a percentage of the blade pitch. The obtained vibration data Δθ _{1 to} Δθ _N are temporarily stored in the buffer 32, and then DMA-transferred to the secondary processing device 7 with the timing controlled by the DMA controller 31.

In this case, from the secondary processing device 6, time-series vibration data in the rotation direction when the M rotor blades 2 pass sequentially in the vicinity of a certain electromagnetic pickup 5 (hereinafter referred to as a sensor-based time series). Data) is an electromagnetic pickup 5
N sets are output in accordance with the number N of.

These N sets of sensor-based time series data are input to a secondary processing device 7 as a vibration analysis device. The secondary processing device 7 has a configuration as shown in FIG. 4 and includes a data rearranger 11, an FFT calculator 12, an adder 13, a divider 14, and an analyzer 15.

In the secondary processor 7, first, in the data reordering unit 11, a certain rotor 2 sequentially passes vibration data constituting the N sets of sensor-based time-series data 21 in the vicinity of the N electromagnetic pickups 5. The data is rearranged as time-series vibration data (hereinafter, referred to as blade-based time-series data), and the FFT calculator 12 generates M sets of blade-based time-series data 22 corresponding to each of the M rotating blades 2. Output to

The FFT operation unit 12 includes the data reordering unit 1
The fast Fourier transform is performed on all of the M sets of wing-based time-series data 22 input from 1 and M data 23-1 to 23 -M obtained as a result are output to the adder 13. Further, the FFT calculator 12 outputs the data 23-1 to 23 -M obtained as a result of the fast Fourier transform to the analyzer 15 as a set of analysis data 26.

The adder 13 adds together the data 23-1 to 23-M output from the FFT operation unit 12,
The divider 14 converts the data 24 obtained as a result into the number of data M
By the analyzer 15 as analysis data 25.
Output to That is, the average value of the data 23-1 to 23 -M is obtained as the analysis data 25 by the adder 13 and the divider 14.

Finally, the analyzer 15 analyzes the vibration of the rotor 2 based on the analysis data 25 input from the divider 14 and the analysis data 26 input from the FFT calculator 12, and analyzes the analysis result. It is shown by a mode diagram, a Campbell diagram, a tracking analysis diagram, etc., displayed on an appropriate display device, and the natural frequency of the rotor 2 is obtained.

In this case, the analyzer 15 analyzes the time-series data 22 based on the fast Fourier transform of all the wing-based data. In particular, since the analysis data 25 is an average value of the data 23-1 to 23 -M corresponding to the M rotor blades 2, the data indicating the vibration itself of the rotor blade 2 is emphasized, and the noise is relatively high. Because the component is reduced, unlike the conventional blade vibration measurement device, the vibration analysis is performed based on the data obtained by performing Fourier transform on the blade-based time series data corresponding to one rotor blade. , High-precision analysis can be performed.

For example, FIG. 5 shows an example of a Campbell diagram according to the present embodiment, and the variation in data is much smaller than that of the Campbell diagram using the conventional blade vibration measuring device shown in FIG. The data corresponding to the natural frequency of the rotor is shown with high accuracy. In the figure, the horizontal axis represents the rotation speed, the right vertical axis represents the harmonics, and the left vertical axis represents the vibration frequency.

FIGS. 6A and 6B show an example of a tracking analysis diagram according to the present embodiment. FIG. 6A shows the phase of the vibration of the rotor 2 with respect to the rotation speed of the rotor 1, and FIG. The vibration value of the rotor 2 is shown. In this case, data at the time of harmonics 10H is shown,
Rotation speed 2848 rpm corresponding to the natural frequency of rotor 2
And 2880 rpm, respectively, are highlighted.

FIG. 7 shows an example of a mode diagram according to the present embodiment, in which (a) shows a case where the number of rotations is 2879.4 rpm.
m, the 10 mode diagram, and FIG.
It is a 9-mode diagram at the time of 9 rpm. These mode diagrams can be created based only on the analysis data 26.

As described above, in the blade vibration measuring apparatus of the present embodiment, the fast Fourier transform is performed on all of the M sets of blade-based time-series data corresponding to each of the M rotating blades 2, and the result is obtained. Vibration analysis is performed based on this. In particular, the analysis data 25 is the data 23- that has been subjected to the fast Fourier transform.
Since the average value of 1 to 23-M is used, the data indicating the vibration of the rotor 2 itself is emphasized, and other noise components are relatively reduced and the S / N ratio is improved. Therefore,
Vibration analysis can be performed with higher accuracy than before, and for example, the natural frequency can be accurately known.

Further, since the S / N ratio of the analysis data 25 is improved, the measurement accuracy can be improved without forcibly increasing the amount of air blown for exciting the rotor 2. There is no danger of equipment being destroyed due to an increase in the ambient temperature.

Further, since the analysis is performed at once based on the measurement results of all the rotors 2, it is not necessary to conduct the test while searching for the rotors 2 having large vibrations one by one as in the related art. The time can be significantly reduced.

[0039]

As described above, according to the present invention, time-series vibration information in the direction of rotation is obtained for each rotor from the measurement results of the first and second vibration measuring means. After applying the Fast Fourier Transform to all of the vibration information, the signals obtained as a result are emphasized on the signal component indicating the vibration of the rotor itself, and unnecessary components such as noise are reduced. Thus, highly accurate vibration measurement can be performed.

Further, since the measurement accuracy is improved, it is not necessary to forcibly increase the amount of air blown to excite the rotor, and there is no danger of damaging the equipment. Furthermore, since it is possible to perform analysis using information on all the rotors at once, it is not necessary to search for rotors having a large vibration amplitude one by one from vibration information for each rotor as in the past,
The test can be performed in a short time.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a blade vibration measuring device according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a primary processing device in FIG. 1;

FIG. 3 is a timing chart of data input to the primary processing device in FIG. 1;

FIG. 4 is a block diagram showing a configuration of a secondary processing device in FIG. 1;

FIG. 5 is an example of a Campbell diagram obtained by the embodiment.

FIG. 6 is an example of a tracking analysis diagram obtained by the embodiment.

FIG. 7 is an example of a mode diagram obtained by the embodiment.

FIG. 8 is a diagram showing a configuration of a conventional blade vibration measuring device.

9 is a block diagram showing a specific configuration of the secondary processing device in FIG.

FIG. 10 is an example of a Campbell diagram obtained by a conventional vibration analyzer.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Rotor 2 ... Rotor blade 3 ... Rotation pulse detector 4, 5 ... Electromagnetic pickup 6 ... Primary processing apparatus 7 ... Secondary processing apparatus 11 ... Data rearrangement unit 12 ... FFT arithmetic unit 13 ... Adder 14 ... Divider 15 ... Analyzer 30 ... Counter 31 ... DMA controller 32 ... Buffer X, Y, Z ... Counter

Claims (1)

[Claims] 1. A blade vibration measuring device for measuring the vibrations of a plurality of rotors in a non-contact manner, wherein the rotor is mounted on a stationary side opposite to the tip of the rotor at equal intervals, and the rotation axis of the tip of the rotor is A plurality of first vibration measuring means for measuring a vibration in a direction, and a second vibration measuring means attached to a stationary side opposed to the base of the rotor and measuring the vibration of the base of the rotor in the rotation axis direction. Means for obtaining time-series vibration information in the rotation direction for each of the rotors based on the measurement results of the first and second vibration measuring means; and vibration for each of the rotors obtained by this means. A blade vibration measuring device comprising: means for performing fast Fourier transform on all information; and means for adding all Fourier transform results obtained by the means.