RU2598983C1 - Diagnostic technique for type of oscillations of working blades of axial turbomachine - Google Patents

Abstract

FIELD: engines and pumps.

SUBSTANCE: invention relates to propulsion engineering and power engineering and can be used in development of gas turbine engines, as well as for creation of oscillations diagnostics systems. Before measurement signal to various operating conditions of turbomachine amplitude-frequency characteristics are made at diagnostic frequencies of auto-oscillations and rotating failure, which are loaded into memory of the control system of turbomachine, narrow-band tracking filters are selected according to them and adjusted to diagnostic frequency of auto-oscillations and rotating failure; body vibration is measured by transducer, damping parameters are determined in advance built for preset operating conditions of turbomachine amplitude-frequency characteristics, subharmonic thresholds are determined using amplitude-frequency characteristics and parameters of damping, when the threshold level is achieved by signal amplitude, which is supplied in the bandwidth of filter tuned to diagnostics frequency of auto-oscillations, the presence of auto-oscillations is stated, when the threshold level is achieved by signal amplitude, which is supplied in the bandwidth of filter tuned to diagnostics frequency of rotating failure, presence of a rotating failure is stated.

EFFECT: increase of efficiency and reliability of diagnostics of type of hazardous vibrations of turbomachine impeller.

3 cl, 4 tbl, 6 dwg

Description

The present invention relates to the field of engine building and power engineering and may find application in the development of gas turbine engines (GTE), as well as for creating systems for diagnosing vibrations.

Diagnostics of dynamic processes can be performed both in the time and in the frequency domains. Both of these approaches have their own characteristics. Diagnostics in the frequency domain is most widely used due to the fact that it is performed at a previously known diagnostic frequency and each source can be associated with spectral components.

Diagnostics in the frequency domain of non-stationary processes, such as self-oscillations and rotational stall, is carried out at frequencies that are not a multiple of the rotor speed, which complicates the diagnostic process, forces us to look for other sources of information and develop new criteria, including those taking into account damping parameters.

It is known that the damping parameters of the system carry useful information in the study of the causes and effects of vibration during operation of a turbomachine.

The closest technical solution to the proposed one is a method for diagnosing the type of vibrations of the working blades of an axial turbomachine, in which the signals from the sensor mounted on the turbomachine body are measured, the damping parameters are determined at predefined diagnostic frequencies of self-oscillations and rotating stall, and the type of vibrations of the working blades is judged (RF patent for the invention No. 2402751, IPC G01M 15/14, publ. 10/27/2010 Bull. No. 30).

In this method, signals are measured, recorded and amplified from strain gauges mounted on the working blades, and from a flow pressure pulsation sensor mounted on the turbomachine body. Convert these signals to frequency spectra. Determine the numbers of diametrical vibrational modes and the diagnostic vibrational frequencies of the blades in the spectrum of flow pressure pulsations. As a damping parameter characterizing the loss of stability, an excitation coefficient is used in an unsteady signal at the vibration frequencies of the blades and at the diagnostic vibration frequencies in the spectrum of flow pulsations. The dependences of the values of the excitation coefficients on time for the signals recorded from the sensors are constructed. The time moments are determined at which the values of the excitation coefficients from negative become positive and are used to judge the type of vibrations of the working blades (self-oscillations or rotating stall). Assign measures to eliminate them.

The method allows diagnostics according to already registered information, i.e. not in real time, but after stopping the turbomachine, which is one of its drawbacks.

Its main disadvantage is that when diagnosing fluctuations in this way, a false diagnosis is possible. This is due to the fact that when determining the excitation coefficient for a real signal with an amplitude that grows in time at certain points in time, the value of the signal amplitude can decrease while maintaining the general growth trend. When the signal amplitude decreases, the excitation coefficient in accordance with this method will change its sign to the opposite, which can lead to a false diagnosis.

When determining the excitation coefficient by this method, the amplitudes at adjacent half-waves of the signal are determined, therefore, with a relatively high frequency of the signal and its growing character, these amplitudes differ slightly from each other, as a result, their ratio is close to unity, and the natural logarithm of unity is zero. To offset the signal values relative to the zero level, the signal frequency is introduced into the formula for determining the excitation coefficient as a scale factor. However, simultaneously with the amplification of the signal and its removal from the zero level, a proportional amplification of all errors occurs. As a result, there is a constant change of sign, which can lead to a false diagnosis. In radio electronics, the term “contact bounce” is used for a similar phenomenon. Therefore, this method is applicable only for idealized signals (smooth, artificially modeled curves).

The presence of self-oscillations is judged by the simultaneous passage through the zero level of the excitation coefficients of two signals. At the same time, the previous and subsequent amplitudes for these signals are determined at different points in time, i.e., the method contains a methodological error that reduces the reliability of diagnostics caused by the need to perform analysis at two different frequencies, and, therefore, at different points in time, therefore, there is no reason to conclude about “simultaneity”.

Due to the fact that, when approaching the boundary of self-oscillations, the process is obviously unsteady in this method, the excitation coefficients for such processes are determined by the Prony method (O. B. Balakshin, B. G. Kukharenko, A. A. Khorikov. Study of dynamic processes with flutter blades using the Prony method). The complex technical implementation of the Proni method and the lack of proven algorithms makes it difficult to use the known diagnostic method in practice.

The technical result, the achievement of which the proposed technical solution is aimed at, is to increase the efficiency and reliability of the diagnosis of the type of dangerous vibrations of the turbomachine impeller by diagnosing it at an early stage of their development and eliminating the false diagnosis. The early stage of diagnosis is provided by predicting the threshold level of the signal depending on the operating conditions of the turbomachine, determined by the pre-built amplitude-frequency characteristics (AFC) (from experimental studies) at known diagnostic frequencies and damping parameters. Upon reaching the appropriate threshold level, the appearance of one of the dangerous types of vibrations is diagnosed (their simultaneous occurrence is impossible), which eliminates the false diagnosis.

The technical result is achieved by the fact that in the method for diagnosing the type of vibrations of the working blades of an axial turbomachine, in which the signals from the sensor mounted on the turbomachine body are measured, the damping parameters are determined at predefined diagnostic frequencies of self-oscillations and rotating stall, and the type of vibrations of the working blades is judged, in contrast from the known, before measuring the signal for various operating conditions of the turbomachine, the amplitude-frequency characteristics are built at the diagnostic frequencies of the self-oscillation ni and rotating stall, which are stored in the memory of the turbomachine control system, select narrow-band servo filters from them and tune them to the diagnostic frequencies of self-oscillations and rotating stall; measure the body vibration from the vibration transducer, determine the damping parameters according to the amplitude-frequency characteristics pre-built for the given operating conditions of the turbomachine, determine the threshold levels of the body vibration using the amplitude-frequency characteristics and damping parameters, when the threshold level is reached by the amplitude of the signal falling into the filter passband, tuned to the diagnostic frequency of self-oscillations, conclude that there are self-oscillations when reaching the threshold ur vnya signal amplitude falls within the band pass filter tuned to the frequency of the diagnostic rotating stall, conclude that there rotating stall.

The amplitude-frequency characteristics are built on the diagnostic frequencies of self-oscillations and rotating stall according to previously obtained data from experimental studies of a turbomachine or conduct the necessary studies to obtain them.

As the damping parameter, a logarithmic decrement of vibrations is used, while the threshold level of case vibration is determined by the formula:

where A _max - the amplitude of the maximum oscillations in frequency response,

δ is the logarithmic decrement of oscillations.

As a damping parameter, a damping coefficient is used, while the threshold level of case vibration is determined by the formula:

where A _max - the amplitude of the maximum oscillations in frequency response,

Δω is the frequency difference corresponding to equal amplitudes A on both branches of the frequency response;

ω ₀ is the resonant frequency;

γ is the damping coefficient.

The proposed diagnostic method is illustrated by drawings, which depict:

FIG. 1 - frequency response for various conditions of experimental studies;

FIG. 2 - arrangement of a vibration transducer on a turbomachine body;

FIG. 3 - AFC - the dependence of the amplitude of the vibration velocity on the diagnostic frequency of self-oscillations, used when choosing and setting up one of the filters;

FIG. 4 - AFC - the dependence of the amplitude of the vibration velocity on the diagnostic frequency of a rotating stall, used when choosing and setting up another filter;

FIG. 5 - dependence of the amplitude of the vibration stresses on the frequency of rotation of the impeller during self-oscillations;

FIG. 6 - dependence of the amplitude of the vibration velocity on the frequency of rotation of the impeller during self-oscillations.

The method is as follows.

Diagnostic frequencies of self-oscillations and rotating stall are determined.

The diagnostic self-oscillation frequency f _{∂ AK} is determined, for example, by the formula (Kurkov A., Dicus J. Synthesis of blade flutter vibratory patterns using stationary transducers / - ASME Paper N78-GT-160 / - Apr / 1978):

where f _m is the frequency of natural vibrations of the blades at different frequencies of rotation of the impeller, determined by calculation and / or experimentally (Dynamics of aircraft gas turbine engines. Edited by IA Birger, BF Shorra. M. Machine building, 1981);

m is the number of nodal dimeters of the natural waveform;

f _P - frequency of rotation of the impeller of the turbomachine.

The diagnostic frequency of the rotating stall f _{∂ BC} for compressor stages in which it may occur (for example, the first) is determined, for example, by a formula that takes into account the geometric characteristics and parameters of the air flow (Firsov A.V., Posadov V.V. Experience in identifying malfunctions gas turbine engines using narrow-band spectral analysis of vibration // Control. Diagnostics. 2011. No. 12. P. 51-59):

where Z _pk - the number of impeller blades of the impeller (PK) in the stage;

Z _on - the number of vanes of the guide apparatus (ON) in the stage;

- относительный диаметр втулки у лопаток в ступени. $$\overline{d}$$

- the relative diameter of the sleeve at the blades in the stage.

The diagnostic frequencies f _{∂ AK} and f _{∂ BC} change with a change in the frequency of rotation of the impeller f _{p of the} turbomachine and the following dependence is preserved in all modes of its operation: f _{∂ BC} <f _p <f _{∂ AK} Diagnostics is performed at known diagnostic frequencies, the values of which lie on opposite sides with respect to the impeller rotation frequency (and do not intersect).

For various operating conditions of the turbomachine (for example, the degree of opening of the nozzle, the presence of a throttling grate at the inlet, work with dampers in the locks of the fan blades, etc.), amplitude-frequency characteristics (AFC) are built at the diagnostic self-oscillation frequencies f _{∂ AK} and rotating stall f _{∂ BC} data obtained in advance by experimental studies turbomachine (e.g., experimental trials) or in the absence thereof is carried out experimental research required for their preparation, wherein the turbomachine dissected TENSO resistors or at least one vibrator.

The frequency response obtained for different test conditions (Fig. 1) is stored in the memory of the turbomachine control system, for example, it is flashed in a read-only memory (ROM), separately for self-oscillations and rotating stall.

Based on these frequency response, narrow-band servo filters are selected and tuned to the diagnostic self-oscillation frequencies f _{∂ AK} and the rotating stall f _{∂ BC} . Tunable narrow-band servo filters can be implemented, for example, on the basis of specialized microcircuits - functionally complete links of second-order filters, well approximating classical analog filters (Furmakov E.F., Yu.G. Stolyarov, V.V. Kabanov, V.N. Kharitonov, Equivalence of narrow-band filtering methods and digital harmonic Fourier analysis in vibration control equipment for multi-shaft gas turbine rotors / Aerospace Engineering and Technology, Materials of the 9th Congress of Engine Builders, Kharkov. KhAI, 2005 No. 10, pp. 118-121):

- ME10, MAX7490 / MAX7491 (Maxim company);

- LMF100 (company "National Semiconductors);

- LTC1064 (company "Linear Technology").

The number of filters for each type of diagnosed oscillation can be determined by the number of diagnostic frequencies according to the interesting (usually the most dangerous) vibration modes.

Frequency response is characterized by Q factor, the higher the Q factor, the narrower the filter bandwidth (D. Johnson, J. Johnson, G. Moore. Guide to Active Filters. Transl. From English. M .: Energoatomizdat, 1983, 128 p.) And the higher the diagnostic efficiency, as less noise gets into the filter passband, which makes diagnosis difficult. On the contrary, with high damping in the system, which is expressed by the damping parameter, the quality factor is low, and the filter passband is wide. This must be taken into account when choosing filter parameters (filter type, filter order, etc.). The quality factor Q is associated with the logarithmic decrement of oscillations δ, which characterizes the damping of the oscillatory system, by the ratio:

Case vibration is measured with a vibration transducer mounted on the turbomachine body near the studied impeller stage (Fig. 2). As a vibration parameter, “vibration velocity” is used.

Depending on the operating conditions of the turbomachine, the frequency response previously built for self-oscillations and rotating stall is selected, which determines the damping parameter, for which, for example, the logarithmic decrement of oscillations δ, which can be defined as the natural logarithm of the ratio of the amplitudes of the subsequent and previous oscillations, is used (Skubachevsky G . S. Aircraft gas turbine engines. Design and calculation of parts. M.: Mechanical Engineering, 1981, S. 288).

In addition to the logarithmic decrement of oscillations δ, a damping coefficient γ defined as γ = δ / π can be used as a damping parameter.

Based on the selected frequency response and damping parameter values, threshold signal levels are determined for self-oscillations and rotating stall. To determine the threshold level, it is sufficient to use one ascending branch of the frequency response obtained by typing the speed of the turbomachine.

The threshold level is the amplitude of the signal at point A (Fig. 1), determined, for example, by the formulas:

or

where A _max - the amplitude of the maximum oscillations in frequency response,

γ is the damping coefficient;

Δω is the frequency difference corresponding to equal amplitudes A on both branches of the frequency response;

ω ₀ is the resonant frequency.

δ is the logarithmic decrement of oscillations.

Upon reaching threshold level A with the amplitude of the signal falling into the passband of the filter tuned to the diagnostic frequency of auto-oscillations f _{∂ AK} , the conclusion is made about the presence of self-oscillations (Fig. 3), while the level of the signal falling into the pass-band of the filter tuned to the diagnostic frequency of the rotating disruption f _{∂ BC} , does not exceed the noise level of the measuring equipment (there is no useful signal).

Upon reaching threshold level A with the amplitude of the signal falling into the passband of the filter tuned to the diagnostic frequency of the rotating stall f _{∂ BC} , it is concluded that there is a rotating stall (Fig. 4), while the amplitude level of the signal falling into the passband of the filter tuned to the diagnostic self-oscillation frequency f _{∂ AK} does not exceed the noise level of the measuring equipment (there is no useful signal).

When diagnosing one of the types of vibrations, the operating mode of the turbomachine is changed in order to prevent damage to parts and components of the turbomachine.

An example implementation of the method.

Carried out diagnostics of self-oscillations and rotating stall of the axial compressor of the aircraft engine.

The diagnostic frequencies f _{∂ AK} and f _{∂ BC were previously} determined using formulas (1) and (2), respectively.

If there was a database of the results of experimental studies of a turbomachine under various conditions, they would use the available information to construct the frequency response.

In the absence of a database for constructing the frequency response, preliminary experimental studies of the turbomachine (for example, experimental tests) were conducted under various conditions of its operation, during which the vibrational stresses in the blades from the strain gauges and the vibrations of the turbomachine body from the vibration transducer were simultaneously measured.

For dynamic strain gauging, the blades and fan disk were prepared with wire strain gauges (5 mm base), for which information was taken from a current collector. Vibrography of the turbomachine body was carried out using bench vibration equipment, including a standard vibration transducer MV43-5B installed in the vertical direction on the separation housing (PK) (Fig. 2).

As a vibration parameter, “vibration velocity” was chosen, the amplitude of which is practically unchanged over the entire range of frequencies of rotation of the turbomachine, which makes it possible to measure the amplitude of the signals without correction for the frequency value, i.e. over the entire range of diagnostic frequencies. The signals from the strain gauges and the vibration transducer were recorded synchronously to the digital signal recorder MIC-300 M.

Predefined initial data and the results of experimental studies obtained by synchronously recording signals from vibration and strain gauges when self-oscillations occur are summarized in table 1.

Predefined initial data and the results of experimental studies obtained from vibration sensors, when a rotating stall occurs, are summarized in table 2.

The results of experimental studies used in performing diagnostics can be obtained, for example, using spectral analysis.

We constructed the dependences of the amplitudes of vibrational stresses on the frequency of rotation of the turbomachine for various test conditions necessary for the subsequent diagnosis of self-oscillations. During the tests, a correspondence was established between the vibrational stresses σ in the rotor blades at a frequency f _m and the vibration velocity V at the diagnostic frequency f _{∂ AK} , on the basis of which the dependence of the amplitudes of the vibration velocity on the frequency of rotation of the impeller of the turbomachine was constructed.

They built the frequency response (Fig. 1) in units of vibration velocity at the diagnostic frequencies of self-oscillations and rotating stall and flashed them into ROM. According to the frequency response, narrow-band servo filters were selected and tuned to the diagnostic frequencies f _{∂ AK} and f _{∂ BC} (tables 1 and 2).

To diagnose the type of vibrations, the case vibration was measured with a vibration transducer mounted on the turbomachine body near the studied impeller stage (Fig. 2). As a vibration parameter, “vibration velocity” was used.

An example of the diagnosis of self-oscillations during the operation of a gas turbine engine with dampers installed in the locks of the fan blades.

During preliminary experimental studies, it was found that in order to ensure a safe level of stresses in the blades, it is necessary to reduce the operation mode of the turbomachine by approximately 200 rpm, i.e. reduce to f _p = 63 Hz (table 1). The frequency difference corresponding to equal amplitudes A on both branches of the frequency response was determined, Δω = 7.33 Hz. We switched from the frequency response constructed by the tensogram (Fig. 5) to the construction of the frequency response, on which the ordinate axis is digitized in units of vibration velocity (Fig. 6) (we redrawn the axis taking into account the correspondence of the maximum amplitude of the vibration stresses A _maxG = 13.5 kgf / mm ^{2 to the} maximum the amplitude of the vibration velocity A _maxV = 12.5 mm / s).

Selected pre-built for self-oscillation frequency response (Fig. 3). The logarithmic decrement of oscillations δ = 0.314 was determined for a rotation frequency of 63 Hz and the threshold level A = 8.82 mm / s was determined by formula (5).

The calculation results used for the diagnosis of self-oscillations are shown in table 3.

Thus, when a signal reaching the passband of the filter tuned to the diagnostic self-oscillation frequency reaches a vibration velocity level of 8.82 mm / s corresponding to the set threshold level A, the presence of self-oscillations was diagnosed (Fig. 3).

At the same time, the level of the signal falling into the passband of the filter tuned to the diagnostic frequency of the rotating stall f _∂BC did not exceed the noise level of the measuring equipment (there is no signal at the diagnostic frequency).

To prevent damage to the parts of the turbomachine, the operation mode of the turbomachine was reduced by 200 rpm.

An example of diagnostics of a rotating stall when working with an interceptor at the compressor inlet.

According to the _{vibration program,} A _maxV = 15.6 mm / s was determined at the resonant frequency ω ₀ = 135 Hz. The diagnostic frequency f _∂BC = 130 Hz was adopted for further calculation. The frequency difference corresponding to equal amplitudes A on both branches of the frequency response was determined: Δω = 10 Hz.

The frequency response previously built for rotating stall was selected (Fig. 4), by which the logarithmic decrement of oscillations δ = 0.168 was determined for a frequency of 130 Hz and the threshold level A = 10.7 mm / s by formula (5).

The calculation results used to diagnose a rotating stall are shown in table 4:

Thus, when a signal reaching the passband of the filter tuned to the diagnostic frequency f _{∂ BC} reaches a vibrational level of 10.7 mm / s corresponding to the established threshold level A, a rotating stall is diagnosed.

In this case, the level of the signal falling into the passband of the filter tuned to the diagnostic self-oscillation frequency f _{∂ AK} did not exceed the noise level of the measuring equipment (there is no signal at the diagnostic frequency).

The operation mode of the turbomachine was reduced in order to prevent damage to parts and assemblies of the turbomachine when dangerous vibrations occurred.

The proposed method for diagnosing the type of oscillation allows you to uniquely determine the type of oscillation, exclude the false diagnosis and increase the timeliness of diagnosis by performing it at an early stage due to the implemented forecasting elements.

Claims (4)

1. A method for diagnosing the type of oscillation of the working blades of an axial turbomachine, in which the signals from the sensor mounted on the turbomachine body are measured, the damping parameters are determined at predefined diagnostic frequencies of self-oscillations and rotating stall, the type of vibrations of the working blades is judged, characterized in that before measuring the signal for various operating conditions of the turbomachine, the amplitude-frequency characteristics are built at the diagnostic frequencies of self-oscillations and rotating stall, which are recorded in the memory be the turbomachine control system is selected for narrowband tracking filters them and set them at the diagnostic self-oscillation frequency and rotating stall; measure the body vibration from the vibration transducer, determine the damping parameters according to the amplitude-frequency characteristics pre-built for the given operating conditions of the turbomachine, determine the threshold levels of the body vibration using the amplitude-frequency characteristics and damping parameters, when the threshold level is reached by the amplitude of the signal falling into the filter passband, tuned to the diagnostic frequency of self-oscillations, conclude that there are self-oscillations when reaching the threshold ur vnya signal amplitude falls within the band pass filter tuned to the frequency of the diagnostic rotating stall, conclude that there rotating stall. 2. The method according to p. 1, characterized in that the amplitude-frequency characteristics are built on the diagnostic frequencies of self-oscillations and rotating stall according to previously obtained data from experimental studies of a turbomachine or carry out the necessary studies to obtain them. 3. The method according to p. 1, characterized in that the logarithmic decrement of vibrations is used as the damping parameter, while the threshold level of the body vibration is determined by the formula:

where A _max - the amplitude of the maximum oscillations in frequency response,
δ is the logarithmic decrement of oscillations. 4. The method according to p. 1, characterized in that the damping coefficient is used as a damping parameter, while the threshold level of the body vibration is determined by the formula:

where A _max - the amplitude of the maximum oscillations in frequency response,
Δω is the frequency difference corresponding to equal amplitudes A on both branches of the frequency response;
ω ₀ is the resonant frequency;
γ is the damping coefficient.

Images