RU2559566C1 - Measurement of pulsating flow parameters - Google Patents

Abstract

FIELD: transport.

SUBSTANCE: invention relates to aircraft engineering, particularly, to definition of gas-dynamic flow parameters variation dynamics in bladed machines and channels, compressors, pipelines and diffusers in preset flow areas both in boundary areas and in gas flow nucleus and can be used for diagnostics of has turbine engines, analysis of flows in pipelines and channels with flow separation. Claimed method consists in measurement and registration of instant magnitudes of three flow arte constants (axial, radial and circumferential), pulsations of complete and static pressures in whatever plane relative to the nozzle. Note here that nozzle intake incorporates at least four pressure pulsation transducers.

EFFECT: higher accuracy and validity of measurements and flow structure analysis.

11 dwg

Description

The invention relates to aeronautical engineering, and in particular to methods for determining the dynamics of changes in the gas-dynamic parameters of a flow in blade machines and channels, for example, in compressors, pipelines and diffusers in predetermined flow areas, both in the boundary zones and in the core of the gas flow, and can be used for diagnosing the technical condition of gas turbine engines, studying the flow in pipelines and channels with flow separation.

The need to measure various parameters of the pulsating flow in order to determine the characteristics of gas-dynamic paths for their further improvement requires the creation of special measuring tools that meet the conditions of the task. For example, to measure the non-stationary parameters of a pulsating flow, it is necessary to have devices that can measure the pulsations of the flow velocity, total and static pressure. The parameters of the pulsating flow are determined during research in the flow part of the compressors, where the flow is unsteady. The pulsating flow parameters measured in this case, such as velocity, total and static pressure, have the form of pulsations with a broadband polyharmonic spectrum, to which the main contribution is made by pulsations at the blade repetition rate and at multiple frequencies. For example, at the exit from the impeller (RK) of a vane machine, due to the formation of traces of the blades, full pressure pulsations are observed associated with the repetition rate of the blades, and due to the presence of circumferential flow heterogeneity, full pressure pulsations at lower frequencies appear in the spectrum as rotary harmonics in the frequency band from 0 to f _mouth × z _pk . It can be assumed that the pulsations of the flow velocity will have a similar spectrum. At the same time, knowledge of the spectral composition of the pulsations of the total pressure, suppose on the average radius beyond the RC, is not enough. To determine areas with increased losses, it is necessary to know the radial field of the total pressure, the field of static pressure, as well as the pulsation field of these parameters and the field of flow angles. For the most accurate determination of these parameters, it is necessary to have devices that allow them to be measured synchronously at each point in the field. Based on these measurements, it is possible to reconstruct the flow structure at the exit of each interscapular channel in the form of isolines of the total pressure loss and the velocity of the pulsating flow in absolute motion. It is also possible to determine the structure of disturbances propagating against the flow, for example, from the rotating wheels of a scapular machine. Similar problems arise in the study of the structure and parameters of the flow in the channels of various devices, such as transitional channels of engines, channels of pipelines. It is especially important to know the flow structure in the channels containing devices with a positive pressure gradient in which flow separation is possible. As is known, flow separation can have a different structure and is a spatial unsteady flow. In this case, to determine the spatial structure, it is necessary to know not only the axial component of the velocity, but also the change in velocity in other directions. The authors are aware of methods for determining the three velocity components using resistance thermometers and laser anemometry. These methods can not always be used in research due to their bulkiness and fragility of the receivers. In addition, they are not informative enough and are limited in the frequency range.

An urgent task is to study the temporal changes in the pattern of propagation of perturbations in a gas stream that arise, for example, when a stationary or moving compressor structural element flows around a stream.

To determine the pattern of disturbances in the gas stream, the structure of the stream is determined in the form of lines of constancy of one or another parameter of the flow — pressure isolines, velocity, reduced velocity, compression ratio, and total pressure loss coefficient. This makes it possible to identify the flow zones in which increased energy losses occur, and to establish the causes that cause them, which can include flow separations of a different nature and the occurrence of pressure surges at a supersonic flow velocity in relative motion in the compressor.

In blade machines, under conditions of propagating perturbations from rotating structural elements, the flow velocity pulsation has a complex polyharmonic broadband spectrum, the frequency composition of which in the conditions of changing compressor operating modes can be in the frequency band from 5 Hz to 10 kHz or more.

The flow in the compressor has a spatial, periodic character. The periodicity of the flow is manifested in pulsations of pressure and absolute velocity (including the values of the three components of the velocity vector and its direction) and in the form of changes in the spectral composition of pulsations.

Of particular interest is the change in the velocity pulsations and its structure when the compressor operating mode changes. In this case, a change in structure means a change in the axial, circumferential, and radial velocity components and their directions.

Under the conditions of spatiality of an unsteady flow, it is necessary to know the variable values of not only the changes in the stationary relations of the flow velocity vectors, but also their pulsations.

There is a method of determining the averaged (stationary) value of the reduced flow velocity in a section perpendicular to the direction of flow, with the measurement in one section of the total and static pressure using probes or nozzles installed in the bore of the pipeline, recording the measurements and processing the measurement results (C. M. Gorlin, Experimental Aeromechanics, Moscow: Higher School, 1970, pp. 173-178).

The disadvantage of this method is the low accuracy of the measurements and the lack of completeness of information to determine the structure of the stream.

These measurements are not enough to observe the change in the flow structure and the velocity vector components over time, since this method does not provide the necessary information about the instantaneous values of the absolute flow velocity and its three components: axial velocity, radial velocity, and peripheral velocity and their directions.

A known method of measuring the flow rate of a gas or liquid (SU 1647408, G01P 5/12, 1989), in which the accuracy of measuring the flow rate of gas or liquid depends on the heat-sensitive surface of the measuring device.

The main disadvantage of this method is that it does not provide the accuracy of measuring the total and static pressures and does not provide a flow structure that varies over time.

Closest to the claimed technical solution is a method for determining the structure of a gas stream in a compressor (RU 2227919, G01P 5/14, 2004). The method consists in the fact that continuous synchronous measurements of static and total pressures in the incoming flow are made, measurements of the pulsation of static and total pressures at the measurement point during the passage in the measurement zones of each blade and each interscapular channel of the compressor rotor with subsequent processing of the measurement results, which allows to build isolines total and static pressures and reduced flow velocity in absolute motion, build contour lines of increase and / or loss of total pressure, determine unified values of the above parameters during the multiple passage of all compressor rotor blades. The technical result is to increase the accuracy of measurements of the gas-dynamic parameters of the flow, the reliability and information content of methods for studying the structure of the flow in the compressor.

The disadvantage of this method is the inability to simultaneously determine all three components of the velocity (axial velocity, radial velocity and peripheral velocity), static and total pressure of the pulsating flow with sufficient accuracy. To do this, it is necessary to orient the nozzles in the flow, and, in addition, a necessary condition is a small range of fluctuations in the velocity components (approximately no more than 10-15% of the average value of the U-axial velocity). The proposed method allows you to remove these restrictions.

The invention is based on the following tasks:

- determination of the three components of the velocity of the pulsating flow: U - axial speed, V - radial speed, W - peripheral speed;

- simultaneous determination of three velocity components, static and total pulsating flow pressures;

- determination of the static and total pulsating flow pressures in a wide range of flow angles (up to ± 85 °).

A method of measuring the parameters of a pulsating flow, which consists in measuring and recording the instantaneous values of the three components of the flow velocity, the pulsations of the total and static pressures in any plane relative to the nozzle. In this case, a nozzle receiving device with at least four pressure pulsation sensors is used. They collect, digitally convert and register analog data from sensors, process sensor readings using calibration curves. Produce visual observation of the operation of each of the sensors is carried out spectral analysis of the measured data, determines the instantaneous direction of flow, Mach number (M), values of angle of attack (α) and slip angle (φ) in the flow and pressure ratio (π _λ) using approximating coefficients a _ijk , b _ijk and coefficients C _α , C _φ , C _M determined from the measured pressures.

To obtain the values of the three components of speed (U, V, W), static (P) and total pressure (P ^* ), the minimum number of sensors is three, however, to increase accuracy, it is necessary to use a larger number of sensors - four or more.

In order to achieve the task, low-inertia pressure pulsation sensors were used as sensitive elements. Purging an existing nozzle for measuring the pulsations of the axial flow velocity in the wind tunnel showed that when using inertial pressure pulsation sensors as sensitive elements, it is possible to obtain a pulsating flow velocity meter with an accuracy of 0.7-3% in the constant component in the range of the bevel from 0 to 15-20 °. The difference between the speed determined by the constant component of the pulsations of the total and static pressures and the flow rate obtained by averaging all the data over a long period of time from the speed values obtained using the inertial reference nozzle is not more than 2%. The accuracy of determining the velocity pulsations is 1.1% at a zero bevel and 1.6-1.2% at a bevel in the range of ± 20 °. The use of high-frequency sensors and the small size of the receiving device in the form of a hemisphere due to synchronous measurement and registration allow you to get instantaneous pressure values. Reducing the size of the sensors reduces the distance between the receiving holes in the hemisphere, which allows you to expand the frequency range. In addition, an increase in the number of sensors makes it possible to expand the range of determining the flow velocity and the range of angles of attack (α) and slip (φ) due to the placement of a larger number of holes on the surface of the hemisphere. In this case, the hemisphere surface size of ⌀8 mm is due to the arrangement of four small-sized sensors in it, having the shape of cylinders with ⌀1.6 mm. In this case, a registrar having a sampling frequency of 200 kHz during synchronous registration can be used as a registrar.

In connection with the foregoing, it is assumed possible, to achieve the objectives, the use of low-inertia pressure pulsation sensors as sensitive elements in the nozzle to determine the three components of the velocity of the pulsating flow, static and total pressure of the pulsating flow in a wide range of flow angles.

The present invention is illustrated by the following detailed description of a method for measuring pulsating flow parameters with reference to FIG. 1-9, where:

in FIG. 1 shows a nozzle layout with the placement of four pressure pulsation sensors;

in FIG. 2 is a nozzle diagram showing an incident flow with Mach number (M), angle of attack (α) and slip (φ), certain pressure values P ₁ , P ₂ , P ₃ , P ₄ in the nozzle receiving holes;

in FIG. 3 shows a block diagram of the operation of a hardware unit for converting signals from pressure pulsation sensors to the values of the three components of the velocity (U, V, W) of the pulsating flow, static pressure (P) and total pressure (P *) of the flow in a wide frequency range;

in FIG. 4 shows an exemplary view of the flow structure (in the form of isolines of the axial reduced velocity λ _U ) in the wall region in front of the compressor;

in FIG. Figure 5 shows examples of pressure oscillograms obtained using a nozzle for measuring velocity pulsations: a) vortex flow, b) quasi-potential flow;

in FIG. 6 shows the time variation of the above examples an axial flow velocity (λ _U), obtained by the nozzle: a) a vortex flow, b) during the quasi-potential;

in FIG. 7 in the form of a graph shows the type of calibration dependencies;

;in FIG. Figure 8 shows a graph of oscillograms of pressure pulsations measured by each of the four sensors at a rotational speed. $$\overline{n}=0.8$$

;

.in FIG. 9 is a graphical representation of spectrograms of pressure pulsations measured by each of the four sensors at a rotational speed $$\overline{n}=0.8$$

.

When implementing the method, the following parameters of the pulsating flow are determined:

- three components of the velocity of the pulsating flow (U is the axial velocity, V is the radial velocity, W is the peripheral speed), static (P) and the total pressure (P ^* ) of the pulsating flow;

- pulsations of the three components of speed, static and total pressure;

- instantaneous flow directions;

- values of the Mach number (M);

- values of the angle of attack (α) and the slip angle (φ) in the flow;

- pressure coefficient (π _λ ) using approximating coefficients a _ijk , b _ijk and coefficients C _α , C _φ , C _M determined from the measured pressures.

A method of measuring the parameters of a pulsating flow, which consists in measuring and recording the instantaneous values of the three components of the flow velocity, pulsations of the total and static pressures in any plane relative to the nozzle 1 (see Fig. 1). In this case, a receiving device 2 nozzle 1 is used, made in the form of a hemisphere with a given radius (⌀8 mm). On the surface of the hemisphere there are receiving holes 3 connected to four pressure pulsation sensors 4, 5, 6, 7, and the distance between the receiving holes 3 is less than one third of the radius of the hemisphere.

The incident flow with the Mach number (M) (see Fig. 2), the angle of attack (α) and slip (φ) creates certain pressure values P ₁ , P ₂ , P ₃ , P ₄ in the receiving holes 3 of the nozzle 1.

According to the block diagram shown in FIG. 3, the analog signals from sensors 4, 5, 6, 7 of the pressure pulsation are received in block 8 for digital conversion and registration of analog signals. Block 8 contains:

- block 9 sets the sampling frequency of the conversion of incoming analog signals;

- block 10 synchronization of readings of sensors 4, 5, 6, 7;

- block 11 accounting calibration curves;

- block 12 of the mathematical model algorithm for determining gas-dynamic functions and approximating coefficients;

- block 13 determining the angle of attack (α), the angle of slip (φ);

- block 14 determining the pressure coefficient π _λ ;

- block 15 determining the three components of the flow rate and visualizing the processing results;

- a block 16 for indicating three components of the flow velocity, angle of attack and slip, compression ratio and total pressure loss.

All blocks are interconnected by feedbacks to improve the accuracy of determining the velocity components and pressure values.

Block 8 of the digital conversion and registration of analog signals contains a block 11 for accounting for individual calibrations of sensors 4, 5, 6, 7 of pressure pulsations, as well as a block for determining the Mach number of the incoming flow (not shown), block 13 for determining the angle of attack (α), sliding angle ( φ) and block 14 for determining the pressure coefficient (π _λ ) using approximating coefficients a _ijk , b _ijk and coefficients C _α , C _φ , C _M determined from the measured pressures P ₁ , P ₂ , P ₃ , P ₄ .

An example of a mathematical model based on approximation by a three-dimensional second-order polynomial of the main flow parameters is presented below:

Block 8 of the digital conversion and registration of analog signals contains a unit for visualization of the converted analog signals, registration of the converted analog signals and a block for synchronizing analog signals (not shown).

Blocks are a set of programs with which the collection, digital conversion and registration of analog data coming from sensors 4, 5, 6, 7 to a computer for their subsequent processing is carried out.

Below is an example form of formulas for determining the coefficients C _α , C _φ , C _M and the gas-dynamic function π _λ :

With _α = (P ₂ -P ₁ ) / (P ₂ - (P ₁ + P ₃ + P ₄ ) / 3);

With _φ = (P ₃ -P ₄ ) / (P ₂ - (P ₁ + P ₃ + P ₄ ) / 3);

C _M = (P ₂ - (P ₁ + P ₃ + P ₄ ) / 3) / P ₂ ;

π _λ = (P ₁ + P ₃ + P ₄ ) / 3 / P ^* = (P ₁ + P ₃ + P ₄ ) / 3 / P · (1- (k + 1) / (k-1) · λ ² ) ^{k / (k-1)} .

In FIG. 4-6 show the results of measuring the pulsations of the flow parameters at the compressor inlet in the peripheral region under different operating conditions in terms of speed and air flow, as well as the position of the premium device 2 nozzle 1 relative to the peripheral wall. In each particular compressor, instantaneous speed values will have corresponding values.

The method of determining the structure of the gas stream in the compressor is as follows.

The pressure with the help of sensors 4, 5, 6, 7 is converted into electrical pulses, which are recorded using a digital recorder and converted to pressure, after which they are compared with the nozzle 1 calibrations, which determine the deviation of the receiving device angle 2, nozzle 1 from the flow direction angle . The digital recorder simultaneously converts the readings of sensors 4, 5, 6, 7 into analog data from all sensors 4, 5, 6, 7 and records the converted analog signals. The visualized unit of the converted analog signals (not shown) produces a visualization of the recorded values. The digital files received in block 8 for digital conversion and registration of analog signals are sent to block 15 for determining the three components of the flow rate and visualizing the processing results. Block 15 is a complex of programs, including a program for determining the reduced flow velocity λ = f (t), a program for graphical images in the form of oscillograms, and a program for constructing lines λ = Const of the reduced flow velocity that visualize the flow structure.

, $${\varphi }_{н}^{о}$$

, Р^* и Р в трехмерном потоке использован насадок, разработанный для измерения аналогичных стационарных и нестационарных параметров, с помощью которого можно измерять указанные выше стационарные параметры, используя для определения угла потока $${\alpha }_{н}^{о}$$

и угла $${\varphi }_{н}^{о}$$

тарировочные кривые, и с помощью тарировок определять по значениям углов отношение истинного и измеренного давлений Р*, а также статического давления Р. Тарировочные кривые определяются предварительными продувками и системой сбора и обработки, алгоритм которой в виде графика приведен на фиг. 7, где на графике показан вид тарировочных кривых для предлагаемого устройства, полученных на одном из режимов по расходу в специальной аэродинамической трубе.As a base nozzle for measuring non-stationary parameters $${\alpha }_n^{\mathrm{about}}$$

, $${\varphi }_n^{\mathrm{about}}$$

, P ^* and P in a three-dimensional flow, nozzles are used, designed to measure similar stationary and non-stationary parameters, with which you can measure the above stationary parameters, using to determine the flow angle $${\alpha }_n^{\mathrm{about}}$$

and angle $${\varphi }_n^{\mathrm{about}}$$

calibration curves, and using calibrations, determine the ratio of the true and measured pressures P * and static pressure P. from the values of the angles. Calibration curves are determined by preliminary purges and a collection and processing system, the algorithm of which is shown in the graph in FIG. 7, where the graph shows the type of calibration curves for the proposed device, obtained at one of the flow rates in a special wind tunnel.

The waveforms shown in FIG. 4 and 5, as well as in FIG. 8, represent a view of the analog signals coming from each pressure pulsation sensor 4, 5, 6, 7 located in the receiving device 2 during synchronous continuous measurement of pressure pulsations at the measurement point over time t, allow the value of static pressure (P) on the basis of calibration curves in the form of the ratio P = f (t) and the total pressure (P *) on the basis of the ratio P * = f (t), where P and P * are functions of time.

The method of measuring the parameters of the pulsating flow allows you to determine the flow structure (using preliminary calibration curves) without correcting the position of the receiving device 2 and the measuring elements (sensors).

The method makes it possible to observe a change in the flow structure under conditions of significant longitudinal pressure gradients and to study the properties of flow breaks arising under these conditions.

Each of the three sensors 5, 6, 7 (see Fig. 1) located around the sensor 4, which detects the pulsations of the total pressure, measures the pulsations of the intermediate pressure between the total and static pressure. Similar data are received from each of the four measurement points; an approximate type of change in time of their values is shown in FIG. 8.

A visual observation is made of the operation of each of the sensors 4, 5, 6, 7. A spectral analysis of the measured data is carried out. A view of the data spectra obtained by each of the sensors 4, 5, 6, 7 is shown in FIG. 9.

To obtain accurate values of static and total pressures performs machining of the sensors 4, 5, 6, 7 by use of calibration curves, obtained by puff nozzle 1 in a special wind tunnel measurement with the incident flow velocity of λ = 0.3-0.8 _nab when changing the flow angle meridional α and circumferential φ directions. Calibration curves allow you to determine the instantaneous flow direction, the values of the Mach number (M) and the values of the angles (α) and (φ) in the stream running on the receiving device 2, which remains stationary.

A view of the change in the field of the reduced U-axial flow velocity (λ _U ) is shown in FIG. 4, at the measurement point.

With the simultaneous measurement of pressure pulsations at several points of the flow, contours of the constancy of gas-dynamic parameters in the flow can be constructed. In this case, the reduced axial flow velocity (λ _U ) is determined by the formula

Where:

λ _U is the instantaneous value of the reduced axial flow velocity;

k is the adiabatic exponent;

P is the instantaneous value of the static flow pressure;

P * is the instantaneous value of the total flow pressure.

The shape of the isolines of the reduced axial flow velocity (λ _U ) in absolute motion in the wall region, commensurate with the thickness of the boundary layer formed on the wall of the inlet cylindrical channel over a time interval, corresponds to the transit time of one interscapular channel and characterizes the influence of structural elements - impeller vanes on the structure flow at the inlet to the compressor impeller.

Similarly, contours of other gas-dynamic flow parameters can be constructed. In this case, when determining the flow structure in the compressor, the degree of increase ( ^π *) and the total pressure loss coefficient ( ^σ *) during the passage of each interscapular channel are calculated using the formulas

;

;

,

,

Where:

^π * is the degree of increase in total pressure;

^σ * is the total pressure loss coefficient;

- величина полного давления на входе в компрессор; $$P_{\mathrm{at}x\mathrm{about}d}^{*}$$

- the value of the total pressure at the inlet to the compressor;

- величина полного давления на выходе из компрессора; $$P_{\mathrm{at}sx}^{*}$$

- the value of the total pressure at the outlet of the compressor;

At ₀ - barometric pressure.

Based on the calculation of gas-dynamic functions, the pulsations of gas-dynamic parameters in time are determined.

The type of pulsations of the total and static pressures at the measurement point in the parietal layer of the inlet channel shows the possibility of obtaining an undistorted flow structure in the parietal layer (see Fig. 5).

While measuring pressure pulsations by each of the sensors 4, 5, 6, 7 in the receiving holes on the surface of the hemisphere, located at several points located radially, for example, in the wall layer of the inlet channel, a picture of the distribution of flow parameters, both inside the wall layer, and in the core of the stream. Similarly, flow fields and changes in it of the above parameters can be obtained in any given part of the flow, both at the input and at the output of the compressor channel.

The proposed method is implemented on test benches having a set of equipment, using well-known personal computers, using both known and special devices created for the best implementation of the method.

The application of this method allows to determine the structure of the gas flow in blade machines, for example, in a compressor, and makes it possible to increase the accuracy of measurements of the gas-dynamic parameters of the flow, the reliability and informativeness of the methods for studying the flow structure in the compressor in order to improve calculation methods in the design and search of reserves for increasing the efficiency of compressors.

To select the most favorable places for the location of the receiving holes 3 in the hemisphere, a computational study was conducted of the turbulent flow around the receiving device 2 nozzle 1.

, направление среднего вектора скорости потока в плоскости, перпендикулярной оси насадка, может измениться на 15° при уменьшении расхода воздуха с Gмакс=15.2 кг/с на 20% относительно осевого направления.A prototype was made, and tests were conducted to measure the velocity field in the compressor. Using a prototype sample, calibration dependences of the readings of four sensors 4, 5, 6, 7 on the direction of flow in the perpendicular axis of the nozzle in the wind tunnel and at the inlet to the compressor stage, perpendicular to the axis of the nozzle in the wind tunnel and in the stream at the entrance to the stage were obtained. The determination of the flow direction from the readings of the receiving device showed that in the stage at a low speed and reduced axial flow velocity λ _U = 0.6, corresponding to the speed of the compressor $$\overline{n}=0.6$$

, the direction of the average flow velocity vector in the plane perpendicular to the axis of the nozzle can change by 15 ° with a decrease in air flow from Gmax = 15.2 kg / s by 20% relative to the axial direction.

The proposed method for measuring the parameters of the pulsating flow allows you to measure and record the instantaneous values of the axial, radial and peripheral speeds, static and total pressure of the pulsating flow and can be used to determine the structure and flow parameters in the blade machines, to diagnose the technical condition of the gas turbine engine.

Claims (1)

A method of measuring the parameters of a pulsating flow, which consists in measuring and recording the instantaneous values of the three components of the flow velocity, pulsations of the total and static pressures in any plane relative to the nozzle, using a nozzle receiving device with at least four pressure pulsation sensors, collecting, digital conversion and registration of analog data from sensors, process the sensor readings using calibration curves, produce a visual observation denie the performance of each of the sensors is carried out spectral analysis of the measured data, determines the instantaneous direction of flow, Mach number (M), the value of the angle of attack (α) and slip angle (φ) in the flow and pressure ratio (π _λ) using the approximating coefficients a _ijk , b _ijk and coefficients C _α , C _φ , C _M , determined by measured pressures

;

;

;

^;
With _α = (P ₂ -P ₁ ) / (P ₂ - (P ₁ + P ₃ + P ₄ ) / 3);
With _φ = (P ₃ -P ₄ ) / (P ₂ - (P ₁ + P ₃ + P ₄ ) / 3);
C _M = (P ₂ - (P ₁ + P ₃ + P ₄ ) / 3) / P _2;
π _λ = (P ₁ + P ₃ + P ₄ ) / 3 / P * = (P ₁ + P ₃ + P ₄ ) / 3 / P · (1- (k + 1) / (k-1) · λ ^{2) k / (k-1).}

Images