JPH032633A - Method and apparatus for analyzing vibration of rotary machine - Google Patents

Abstract

PURPOSE:To obtain an accurate, highly reliable, nth-order component vector spectrum with a simple apparatus by operating an amplitude from an amplitude spectrum based on the nth-order frequency of the number of rotations of a rotary machine, and operating a phase from a mutual spectrum. CONSTITUTION:The vibration of the rotary shaft of a rotary machine and a rotation-synchronized pulse 1 undergo fast Fourier transformation. An amplitude spectrum 9 and a mutual spectrum of each signal are computed. The number of rotations of the rotary shaft is computed 4 from the synchronization of the rotation-synchronized pulse. Then, a frequency which becomes the maximum value in the vicinity of a frequency that is an integer multiple of the number of rotations is computed from the computed number of rotations and the amplitude spectrum of the rotation-synchronized pulse. The frequency is made to be the nth frequency 12. The amplitude is operated from the amplitude spectrum based on said frequency. The phase is operated from the mutual spectrum. Thus an nth-order component vector 15 of the vibration of the rotary shaft is obtained. In this way, the nth component vector characterized by high accuracy and high reliability is obtained with a simple apparatus.

Description

Detailed Description of the Invention [Objective of the Invention] (Industrial Application Field) The present invention provides a method for analyzing vibration components for diagnosing the state of rotating equipment such as steam turbines, gas turbines, generators, compressors, and blowers. H7 method and apparatus.

(Prior Art) Generally, in order to diagnose the condition of rotating equipment, it is necessary to calculate the n-th blow component vector of the vibration of the rotating shaft of the rotating equipment. A method of calculation is adopted using the following.

That is, a rotation synchronization pulse of a rotating device is detected, a vibration signal is input to a vector filter whose cutoff frequency changes depending on the frequency of this pulse, and the cutoff frequency of the vector filter is set to the zero-order frequency with respect to the number of rotations. From the output of the vector filter when set to
A method is used to calculate the blow component vector.

(Problems to be Solved by the Invention) The conventional vibration analysis method for rotating equipment is not sufficient to calculate an accurate 0th-order frequency vector, and it is calculated based only on the rotational speed of the rotating equipment. Therefore, when the rotational speed and the n-th component of vibration differ due to various factors, there is a problem in that it is not possible to accurately calculate the n-th component.

Furthermore, since vector calculation devices for each order are required for each vibration signal, there is also the problem that the cost of the vibration analysis device increases.

The present invention has been made in consideration of these points, and provides a vibration analysis method for rotating equipment that can easily extract an accurate and reliable n-th blowing component vector from the vibration signal of the rotating equipment, and The purpose is to provide such equipment.

[Structure of the invention]

(Means for Solving the Problems) A vibration analysis method for a rotating device according to the present invention, as a means for achieving the above-mentioned object, detects the vibration of the rotating shaft of the rotating device and the rotation synchronization pulse, and Fast Fourier transform processing (hereinafter referred to as FFT processing) is performed to extract the amplitude spectrum of each signal and the mutual spectrum of both signals, and the rotation speed of the rotating equipment is extracted from the period of the rotation synchronization pulse. In the amplitude spectrum of the rotation synchronized pulse, calculate the maximum frequency near the frequency that is a positive multiple of the rotation speed, set this as the 0th frequency of the rotation speed of the rotating equipment, and calculate from the amplitude spectrum based on this frequency. The present invention is characterized in that the amplitude is calculated and the phase is calculated from the mutual spectrum to obtain the n-th blowing component vector of the vibration of the rotating shaft.

Further, the vibration analyzer for rotating equipment according to the present invention includes, as means for achieving the above object, a vibration detector that detects vibrations of a rotating shaft of the rotating equipment, and a reference position provided on the rotating shaft. a rotation pulse detector for detecting a rotation synchronization pulse; a filter for preventing aliasing of the signals detected by both of the detectors; and an A/D converter for converting an analog signal from the filter into a digital signal; /
a buffer memory that temporarily stores the output of the D converter;
An FFT processor that reads rotation synchronization pulses and vibrations from a buffer memory and performs FFT calculations, and this FFT
A data processing device that calculates the n-th blow component vector of the vibration of the rotating shaft based on the amplitude spectrum and the t-axis spectrum from the processor, and a display and recording device that displays and records the calculation results of the data processing device. It is characterized by the presence of

(Function) In the vibration analysis method for rotating equipment according to the present invention, the FF
T processing is performed to extract the amplitude spectrum of each signal and the mutual spectrum of both signals, and the rotation speed of the rotating equipment is extracted by synchronizing the rotation synchronization pulse. and,
In the amplitude spectrum of the extracted rotation synchronization pulse,
The maximum frequency near a frequency that is a positive multiple of the rotational speed is calculated and used as the 0th frequency of the rotational speed of the rotating equipment, and based on this frequency, the amplitude is calculated from the amplitude spectrum, and the phase is calculated from the mutual spectrum. Then, the n-th blowing component vector of the vibration of the rotating shaft is obtained. Therefore, an accurate and reliable n-th blowing component vector can be obtained from the vibration signal.

Furthermore, in the vibration analyzer for rotating equipment according to the present invention, the vibration detector detects the vibration of the rotating shaft, and the rotation pulse detector detects the rotation synchronization pulse. Both of these signals are filtered to prevent aliasing, converted to digital signals by an A/D converter, and temporarily stored in a buffer memory. Both signals from the buffer memory are read into a fast Fourier transform processor (referred to as an FFT processor) and subjected to a fast Fourier transform operation (FFT processor).
FT operation) is performed, and the data processing device performs this F
Based on the amplitude spectrum and mutual spectrum from the FT processor, the n-th blowing component vector of the vibration of the rotating shaft is calculated. The results are displayed on the display/recording device.
d is recorded. Therefore, it is possible to obtain an accurate and reliable n-th blowing component vector from the vibration signal using a simple device.

(Example) Hereinafter, a first example of the present invention will be described with reference to FIGS. 1 and 2.

FIG. 1 shows an example of a vibration analyzer for rotating equipment according to the present invention, and in the figure, reference numeral 1 denotes a vibration detector for detecting vibrations of a rotating shaft 2 of the rotating equipment. teeth,
It is disposed close to and opposite to the rotating shaft 2. A rotation pulse detector 3 is also disposed facing the rotation shaft 2, and the rotation pulse detector 3 detects a reference position installed on the rotation shaft 2 to detect rotation synchronization pulses. ing.

Each of the detection signals from the two detectors 1.3 is input to a filter 5 as shown in FIG. High frequency component signals are removed.

As shown in FIG. 1, the analog signal from this filter 5 is converted into a digital signal by an A/D converter 6 and then temporarily stored in a buffer memory. The rotation synchronization pulse and vibration signal from 7 are read by the data processing device 8 and are subjected to FFT.
The number of data necessary to perform the calculation is transferred to the FFT processor 9, and the rotation speed of the rotating shaft 2 is calculated from the rotation synchronization pulse.

Previous fc! The FFT processor 9 performs amplitude spectrum calculation and mutual spectrum calculation between the rotation synchronization pulse and each vibration signal on the data transferred from the data processing device 8, and calculates the amplitude spectrum of the rotation synchronization pulse and the vibration signal, and the rotation synchronization pulse. The coherence and phase spectra of the vibration signals are calculated respectively.

Further, the data processing device 8 is configured to input the calculation result of the FFT processor 9 and calculate the zero-order frequency and the n-th blowing component vector of the vibration based on the zero-order frequency. The zero-order frequency and the n-th vibration component vector calculated by the data processing device 8 are displayed and recorded on the display/recording device IO.

Next, the vibration analysis method for rotating equipment according to the embodiment will be explained with reference to the analysis flow diagram shown in FIG.

The vibration signal of the rotating shaft 2 detected from the rotating equipment and the rotation synchronization pulse are input, and the amplitude spectrum of each manually generated signal is calculated by FFT calculation, and the mutual spectrum calculation of the rotation synchronization pulse and the vibration signal is performed. , the amplitude spectrum, the rotation-synchronized pulse amplitude spectrum, the phase spectrum, and the coherence. Furthermore, the rotation speed of the rotating shaft 2 of the rotating device is calculated from the synchronization of the rotation synchronization pulse.

Next, from the calculated rotation speed and rotation synchronization pulse amplitude spectrum, a rotation synchronization pulse peak frequency at which the rotation synchronization pulse amplitude spectrum is maximized near a frequency that is a positive multiple of the rotation speed is detected. Further, from the rotation speed and coherence, a coherence peak frequency at which the coherence is maximum near a frequency that is a positive multiple of the rotation speed is detected.

Then, the frequency calculated by the arithmetic mean of the rotation synchronization pulse peak frequency and the coherence peak frequency is set as the 0th-order frequency of the rotating device.

Next, based on this zero-order frequency, calculate the amplitude of the n-th blowing component from the amplitude spectrum calculated for each vibration signal, and calculate the phase of the n-th blowing component from the phase spectrum calculated from the mutual spectrum. do.

And with this, rotation? [Obtain the n-th component vector of the vibration of I12.

In this way, the accurate n-th blowing component edge I can be extracted from the vibration signal of the rotating equipment using a simple device.

FIG. 3 shows a second embodiment of the present invention, which will be described below.

FIG. 3(a) is an amplitude spectrum obtained from FFT calculation of the rotation synchronization pulse. Since this spectrum is an FFT calculation on the pulse waveform, the nth frequency fIR, '21? , f
3R, f4R... have maximum values in the vicinity.

Figure 3(b) shows the FF of the rotation synchronization pulse and each vibration signal.
The nth frequency fIC, f2C,'3 which is the coherence obtained from the T calculation and is a positive multiple of the rotation speed of the rotating shaft 2.
C, f4C, has a value close to 1 in the vicinity of """, and has a value close to O at other frequencies. Therefore, for coherence, set a value L that is -r, and then The range of frequencies exceeding this value L is calculated for each 0th order frequency as follows.

9th frequency frequency range fIc
ICIC f'f'f'f''2C2C2Cf'f' f3C3C3C f'f''4C4C4G Then, the 9th frequency [11? , f2Rf31? , f4
F”..., the ninth calculated from the coherence
Range f' ~ M f' ~ I at the next frequency
c Ic, 20 f'f'~f'f'~f'
...2G, 3C3C, 4C4C.

Those that deviate from the above are excluded, and the 9th frequency is determined.

Frequency range f' ~ f''f' ~ f'' ICIC, 20 corresponding to the 9th frequency determined in this way
2C.

f' to f'f' to f'..., 3C3C, 4C4C.

The effective value of the amplitude spectrum is calculated to determine the amplitude, and the arithmetic mean value of the phase spectrum is calculated to determine the phase.

Based on these, the n-th component vector is calculated.

As described above, according to this vibration analysis method, the same effects as in the first embodiment can be obtained, and a more accurate ninth embodiment can be obtained.
It is possible to obtain the order frequency and a highly reliable nth order component vector.

〔Effect of the invention〕

As explained above, the vibration analysis method for rotating equipment according to the present invention detects the vibration of the rotating shaft and the rotation synchronization pulse, and performs FFT processing on these two signals to determine the amplitude spectrum of each signal and the rotation synchronization pulse. In addition to extracting the mutual spectrum, the rotation speed of the rotating equipment is extracted from the synchronization of the rotation synchronization pulse, and in the amplitude spectrum of the extracted rotation synchronization pulse, the maximum frequency is calculated near a frequency that is a positive multiple of the rotation speed. This is set as the 9th frequency of the rotation speed of the rotating equipment, and based on this frequency, the amplitude is calculated from the amplitude spectrum, and the phase is calculated from the mutual spectrum to obtain the nth-order component vector of the vibration of the rotating shaft. Therefore, an accurate and reliable n-th component vector can be easily obtained from the vibration signal of the rotating equipment.

Further, the vibration analyzer for rotating equipment according to the present invention includes a vibration detector that detects vibrations of a rotating shaft, and a reference position provided on the rotating shaft that detects 1. a rotation pulse detector that detects rotation synchronization pulses, a filter that prevents aliasing of the signals detected by both of the detectors, and an A/D converter that converts the analog signal from the filter into a digital signal; A buffer memory that temporarily stores the output of the A/D converter, an FFT processor that reads rotation synchronized pulses and vibrations from this buffer memory and performs FFT calculations, and an amplitude spectrum and mutual spectrum from this FFT processor. Based on this, it is equipped with a data processing device that calculates the n-th blowing component vector of the vibration of the rotating shaft, and a display/recording device that displays and records the calculation results of this data processing device, making it possible to use an analog analysis device that would otherwise be expensive. An accurate and reliable n-th component vector can be obtained without using

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram showing a vibration analysis apparatus for rotating equipment according to a first embodiment of the present invention, FIG. 2 is an analysis flow diagram showing a vibration analysis method thereof, and FIG. A graph showing the vibration spectrum of the rotation synchronization pulse in the vibration analysis method for rotating equipment according to the embodiment, and FIG. 3(b) is a graph showing the coherence between the similar rotation synchronization pulse and the vibration signal. DESCRIPTION OF SYMBOLS 1... Vibration detector, 2... Rotation axis, 3... Rotation pulse detector, 4... Reference position, 5... Filter, 6
... A/D converter, 7... Buffer memory, 8...
・Data processing device, 9...FFT processor, 10.
...Display and recording device. Applicant's agent Mr. Sato

Claims (1)

[Claims] 1. In a rotating device, the vibration of the rotating shaft and the rotation synchronization pulse are respectively detected, and these two signals are subjected to fast Fourier transform processing to obtain the amplitude spectrum of each signal and the mutual spectrum of both signals. At the same time, extracting the rotation speed of the rotating equipment according to the period of the rotation synchronization pulse,
In the amplitude spectrum of the extracted rotation synchronization pulse,
Calculate the maximum frequency near a frequency that is a positive multiple of the number of rotations, and use this as the nth frequency of the number of rotations of the rotating equipment. Based on this frequency, calculate the amplitude from the amplitude spectrum, and calculate the phase from the mutual spectrum. 1. A vibration analysis method for rotating equipment, characterized in that the nth-order component vector of the vibration of a rotating shaft is obtained by calculating. 2. A vibration detector that detects vibrations of the rotating shaft of rotating equipment;
a rotation pulse detector that detects a rotation synchronization pulse by detecting a reference position provided on the rotation axis; a filter that prevents aliasing of the signals detected by both of the detectors; and an analog signal from this filter. An A/D converter that converts into a digital signal, a buffer memory that temporarily stores the output of the A/D converter, and a fast Fourier transform processor that reads rotation synchronization pulses and vibrations from the buffer memory and performs fast Fourier transform operations. , a data processing device that calculates the n-th component vector of vibration of the rotating shaft based on the amplitude spectrum and mutual spectrum from this fast Fourier transform processor, and a display/recording device that displays and records the calculation results of this data processing device. A vibration analyzer for rotating equipment, characterized by comprising: