Industrial axial flow compressor last-stage stationary blade blocking flutter analysis method and equipment and medium

Abstract

The invention relates to the technical field of industrial axial flow compressors, and discloses an industrial axial flow compressor last-stage stationary blade blocking flutter analysis method and device and a medium. The last-stage stationary blade blocking flutter analysis method comprises the steps that a structural dynamics control equation and a flow control equation are rapidly solved through a fluid-solid decoupling method; aerodynamic damping is obtained based on unsteady CFD analysis of a harmonic balance method, and finally the risk of blocking and fluttering of the last-stage stationary blade is judged according to the numerical value of the aerodynamic damping. The equipment and the medium are used for storing a computer program, and the last-stage stationary blade blocking flutter analysis method is achieved when the computer program is executed. The method has the characteristics of high risk estimation accuracy, simplicity, convenience and feasibility, and can further improve the operation safety and reliability of the industrial axial flow compressor.

Description

Technical Field

[0001] The invention relates to the technical field of industrial axial flow compressors, and in particular to a method, equipment and medium for analyzing the blockage and flutter of a last-stage stator blade of an industrial axial flow compressor. Background Art

[0002] Industrial axial flow compressors are power machinery widely used in blast furnace ironmaking, oil refining, basic chemical industry, biological fermentation, large-scale air separation, energy storage, aviation wind tunnels and other fields. They are one of the core equipment of many large-scale process industrial production enterprises. Industrial axial flow compressors are widely used due to their high efficiency and wide range of operating conditions, bringing considerable economic and social benefits to many enterprises.

[0003] Compared with stall flutter, a common flutter phenomenon in aviation axial compressors, blocking flutter is a more common flutter phenomenon in industrial axial compressors during operation. Blocking flutter refers to a type of flutter of the last-stage stator blades whose flutter boundary is close to the blocking boundary of the axial compressor. Blocking flutter can cover a considerable range of stator blade angles. When crossing the blocking flutter boundary, there is no aerodynamic phenomenon of sudden drop in pressure ratio or efficiency. When blocking flutter occurs, negative aerodynamic damping LogDec appears. aero The structural vibration is unstable, usually accompanied by vibration states such as sudden increase in vibration stress and amplitude of the last-stage stationary blades. There is one or more strong shock waves in the blade row channel where blocked flutter occurs, which mostly occur in the first-order mode dominated by bending vibration type.

[0004] When an industrial axial compressor operates under low exhaust pressure conditions, its last-stage stator blades are in a blocked state with a negative angle of attack or even a high negative angle of attack. At this time, there is a risk of blocking flutter in the last-stage stator blades. In particular, the industrial axial compressor for blast furnace blasting, which operates differently from conventional industrial axial compressors, usually operates under extremely low exhaust pressure conditions with the outlet valve fully open when the blast furnace is temporarily shut down to save energy. As a result, the last-stage stator blades often experience sudden large-amplitude violent vibrations due to entering the blocking flutter condition, causing the last-stage stator blades to fail and break in a short period of time, which has a serious impact on the continuous operation and production of the steel industry and causes huge economic losses. At present, in industrial practice, anti-blocking control devices are mainly set to prevent operators from entering the blocking area of ​​the last-stage stator blades during the start-up, operation and process switching of the axial compressor. However, the anti-blocking control device increases the initial hardware investment cost of the anti-blocking valve and the like on the one hand, and narrows the operating pressure range of the axial compressor on the other hand. The increase in exhaust pressure during shutdown leads to increased energy consumption, which has an adverse effect on the economy of operating energy consumption.

[0005] At present, empirical methods such as reduced frequency are commonly used to predict flutter of aviation axial compressors. However, this type of empirical method is mainly aimed at stall flutter and is powerless to predict choking flutter. The design process of industrial axial compressors lacks qualitative and quantitative evaluation of choking flutter, and it is impossible to accurately judge the risk of choking flutter at the operating point.

[0006] The invention with patent publication number CN 118153465 A and name “Wing flutter boundary prediction reduced-order model training method, device, equipment and medium” constructs a training set based on the displacement of interpolation points selected on the wing aerodynamic and structural interpolation interface and the aerodynamic load; constructs a load prediction reduced-order model based on a long short-term memory network, and uses the training set to train the load prediction reduced-order model to obtain a trained aerodynamic load prediction reduced-order model; couples the aerodynamic load prediction reduced-order model with finite element software to obtain a flutter boundary prediction reduced-order model, and uses the flutter boundary prediction reduced-order model to predict the structural displacement response under different dynamic pressures, and finally obtains the wing flutter boundary; however, this method only predicts the flutter boundary of aviation wings and cannot be applied to the prediction of blockage flutter of industrial axial compressors. Summary of the invention

[0007] In order to overcome the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide a method, device and medium for analyzing the blocking flutter of the last-stage stator blades of an industrial axial compressor. Since the material of the last-stage stator blades of an axial compressor is usually high-strength alloy steel, its density is relatively large, resulting in a weaker degree of fluid-solid coupling, that is, the influence of fluid-solid coupling on the frequency and mode of the vibration of the last-stage stator blade is relatively small, and the difference between the vibration of the last-stage stator blade when blocking flutter occurs and its natural vibration mode and frequency can be ignored. The present invention quickly solves the structural dynamics control equations and the flow control equations by the method of fluid-solid decoupling, and obtains the aerodynamic damping LogDec by unsteady CFD (computational fluid dynamics) analysis based on the harmonic balance method. aero , and finally according to the aerodynamic damping LogDec aero The positive and negative signs and values ​​of are used to judge the risk of blockage flutter of the last-stage stator blades; the present invention has the characteristics of high risk prediction accuracy and simplicity.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] A method for analyzing flutter of a last-stage stator blade of an industrial axial compressor, comprising the following steps:

[0010] Step 1: Set the boundary conditions and other model parameters when the industrial axial flow compressor is in a blocked condition, calculate and obtain the single-channel steady-state CFD when the industrial axial flow compressor is in a blocked condition, the boundary conditions include the inlet total pressure, the inlet total temperature, and the outlet static pressure, and the other model parameters include the turbulence model, the solid wall model, and the rotary-static interface model;

[0011] Step 2: Perform structural dynamic modal analysis on the last-stage stator blade and tenon overall model of the industrial axial compressor described in step 1 under non-blocking working conditions to obtain the first three-order natural frequencies and vibration modes of the last-stage stator blade. The parameters that need to be preset before the structural dynamic modal analysis include finite element calculation grid units, elastic modulus, Poisson's ratio, and density;

[0012] Step 3: interpolate the vibration mode of the last-stage stator blade described in step 2 from the normalized mode of the structure domain mesh quality to the fluid domain mesh, and vibrate the last-stage stator blade according to the first three-order natural frequencies and vibration modes described in step 2;

[0013] Step 4: Taking the single-channel steady-state CFD of the industrial axial compressor in the blocked condition described in step 1 as the initial field, perform the single-channel unsteady CFD analysis of the vibrating last-stage stator blade under different inter-blade phase angle conditions for each frequency of the first three natural frequencies described in step 2 to obtain the aerodynamic damping LogDec of the last-stage stator blade. aero ;

[0014] Step 5: Use the pneumatic damping LogDec described in step 4 aero The positive or negative value of determines the flutter risk of the last stage stator blades. When the aerodynamic damping LogDec aero When it is negative, there is a risk of flutter in the last stage stator blades. aero When the rotation speed is positive, the last stage stator blades will not flutter.

[0015] Furthermore, the outlet static pressure in step 1 simulates the exhaust pressure when the outlet regulating valve is fully opened, and the outlet static pressure is usually lower than 150 kPa; the turbulence model in step 1 adopts the SA turbulence model; the solid wall model adopts the slip wall and wall function to process the wall boundary, and the circumferential geometric boundary in the solid wall model adopts the periodic boundary condition; the mixing surface averaging method is used to transmit information between the rotation-static interface models.

[0016] Furthermore, the overall model of the last-stage stator blade and the tenon described in step 2 adopts a structured hexahedral unit, the hexahedral unit material is X3CrNiMo13-4, the overall model of the last-stage stator blade and the tenon adopts a cyclic symmetric model, and a contact constraint is set between the tenon and the root groove of the last-stage stator blade.

[0017] Furthermore, the harmonic balance method is used in the single-channel unsteady CFD analysis in step 4, and phase lag boundary conditions should be used on the upper and lower periodic boundaries in the single-channel unsteady CFD analysis.

[0018] Furthermore, step 4 specifically includes the following steps:

[0019] Step 4.1: For the different inter-blade phase angle conditions described in step 4, the trigonometric function interpolation method is used to perform flutter assessment, analyze multiple inter-blade phase angles and calculate the aerodynamic damping LogDec under multiple inter-blade phase angles aero , the phase angle δ is expressed by the following formula:

[0020]

[0021] In formula (1), N is the number of wave node diameters of the vibration of the last stage stator blades, and Nb is the number of the last stage stator blades;

[0022] Pneumatic damping LogDec aero It is expressed by the following formula:

[0023]

[0024] The accumulated work on the surface of the last stage stator blade W aero It is expressed by the following formula:

[0025]

[0026] In formula (3), A x is the specified vibration displacement of the last stage stator blade, A f1 is the unsteady aerodynamic force of the last stage stator blade obtained by single-channel unsteady CFD solution, It is the phase difference between the unsteady force CFD and the vibration of the last stage stator blade;

[0027] Strain Energy E strain It is expressed by the following formula:

[0028]

[0029] In formula (4), m is the mass of the last stage stator blade, ω is the natural frequency, and A is x is the specified vibration displacement of the last stage stator blade;

[0030] Step 4.2: According to the aerodynamic damping LogDec at multiple inter-blade phase angles described in step 4.1 aero , a complete aerodynamic damping curve is obtained by using trigonometric function interpolation, and the logarithmic attenuation rate is used to characterize the aerodynamic damping curve.

[0031] An electronic device comprises: a memory for storing a computer program; and a processor for executing the computer program to implement the last stage stationary blade blocking flutter analysis method.

[0032] A computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method for analyzing the blockage flutter of a last-stage stationary blade is implemented.

[0033] Compared with the prior art, the present invention has the following technical features:

[0034] 1. The present invention realizes high-precision prediction of the blocking flutter risk of industrial axial compressors, and solves the problem that the existing empirical formula for flutter assessment is only applicable to the stall flutter of aviation axial compressors, and cannot accurately assess the blocking flutter risk that often occurs in the last-stage stator blades of industrial axial compressors.

[0035] 2. In step 4 of the present invention, the fluid-solid decoupling method is used to accurately obtain the aerodynamic damping LogDec based on the energy method aero Value, via pneumatic damping LogDec aero The positive and negative signs of the values ​​qualitatively determine the risk of blocking flutter, and the pneumatic damping LogDec aero The size of the numerical value can be used to quantitatively judge the degree of harm caused by blocking flutter, and it has the characteristics of high accuracy and simplicity.

[0036] In summary, the present invention evaluates the risk of blocking flutter of the last-stage stator blades based on the independently developed blocking flutter analysis method, and further improves the operating safety and reliability of the industrial axial compressor through verification of the actual operating data and operating status of the industrial axial compressor. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flow chart of the flutter analysis method for the last stage stator blade blockage in industrial axial compressors.

[0038] Figure 2 Schematic diagram of the performance and flutter boundary of an industrial axial compressor.

[0039] Figure 3 The relative Mach number distribution diagram of the last stage stator blade at 10%, 50% and 90% blade height under deep blockage conditions in Example 1.

[0040] Figure 4 It is a comparison diagram of the axial, circumferential and radial components of the first three-order modal vibration shapes of the last-stage stator blade in Example 1 before and after interpolation.

[0041] Figure 5 It is a graph showing the aerodynamic damping characteristic curves of the first three modal states of the last-stage stator blade in Example 1.

[0042] Figure 6 This is the surface accumulated work distribution diagram when the inter-blade phase angle is -120° in the first-order mode of the last-stage stator in Example 1.

[0043] Figure 7 This is the relative Mach number distribution diagram of 90% blade height when the inter-blade phase angle is -120° under the first-order mode of the last-stage stationary blade in Example 1. DETAILED DESCRIPTION

[0044] The present invention mainly analyzes industrial axial compressors with the risk of blocking and flutter, generally blast furnace axial compressors. In order to fully verify the effectiveness of the model calibration method, the following is combined with Figures 1 to 7 The present invention is described in further detail.

[0045] Example 1 A flutter stability analysis is performed by taking an industrial axial flow compressor in which the last stage stator blade actually fails and breaks as an example. The specific analysis and calculation steps are as follows:

[0046] Step 1: First, select the deep blocking condition with low outlet exhaust pressure. Use ANSYS CFX software to perform single-channel steady-state CFD calculation for the selected condition. The inlet boundary conditions are inlet total pressure of 101.325 kPa, inlet total temperature of 288.15 K, and the outlet static pressure is given as 150 kPa. The wall boundary is processed by sliding wall and wall function, and the circumferential geometric boundary adopts periodic boundary conditions. Therefore, a single-channel calculation domain can be used. The mixing surface method is used to transmit information between the transfer-static interface model (i.e., different blade rows); the turbulence model adopts the SA turbulence model; Figure 3 The cloud diagram of the flow field relative to the Mach number at three typical blade heights (10%, 50% and 90% blade height) under deep blockage conditions is shown, and it can be clearly seen that there is a strong shock wave in the last stage stator channel.

[0047] Step 2: ANSYS Mechanical software is used to calculate the first three natural frequencies and vibration modes of the last-stage stator blades, and the blocking flutter analysis is performed on the first three modes of the last-stage stator blades. In order to ensure the calculation accuracy and period, the overall model of the last-stage stator blades and tenons in step 2 adopts structured hexahedral units, and the hexahedral unit material is X3CrNiMo13-4. The overall model of the last-stage stator blades and tenons adopts a cyclic symmetric model, and a contact constraint is set between the tenon and the root groove of the last-stage stator blade. The stress stiffening and rotation softening effects are considered during the analysis.

[0048] Step 3: After obtaining the first three modes of the last-stage stator blade described in step 2, the vibration mode is interpolated from the finite element mesh (i.e., the structural domain mesh) to the fluid domain mesh required for steady-state CFD calculation. The vibration mode on the CFD mesh after interpolation is consistent with the vibration mode data of the finite element mesh before interpolation. Figure 4 The comparison of the axial, circumferential and radial components of the first three modal vibration shapes of the last stage stator blade before and after interpolation is shown.

[0049] Step 4: Perform a single-channel unsteady CFD analysis to calculate the aerodynamic damping LogDec aeroSpecifically, the harmonic balance method is used in the single-channel unsteady CFD analysis, and the phase lag boundary conditions should be used on the upper and lower cycle boundaries; the flutter stability of the last-stage static blade often varies due to the change of the inter-blade phase angle. When the frequency domain method is used for single-channel unsteady CFD analysis, it is often necessary to consider the situation under different inter-blade phase angles. In this embodiment, the trigonometric function interpolation method is used for flutter evaluation. In order to reduce the calculation time, the aerodynamic damping LogDec is analyzed and calculated for 9 inter-blade phase angles. aero .

[0050] Since the mechanical damping of the last stage stator structure is much smaller than the aerodynamic damping LogDec aero , so the mechanical damping is ignored and only the aerodynamic damping LogDec is considered aero When the logarithmic decay rate is negative, it indicates that the aerodynamic damping of the last stage stator blade LogDec aero When the logarithmic decay rate is positive, it indicates that the aerodynamic damping of the last stage stator blades is too high. aero If it is positive, there is no risk of blocking flutter. The higher the positive value, the better the stability of the last stage stator blade blocking flutter.

[0051] Figure 5 The aerodynamic damping curves obtained by calculating the first three modes are shown. When the last-stage stator blade vibrates in the first-order mode (M1), the aerodynamic damping LogDec of the last-stage stator blade at 30 inter-blade phase angles is aero If it is negative, there is a greater risk of flutter. Taking the inter-blade phase angle of -120° as an example, the flow field is analyzed. The aerodynamic damping LogDec under this inter-blade phase angle is aero is -1.42%, Figure 6 The accumulated work distribution on the surface of the last-stage stator blade is shown. The accumulated work is the work done by the fluid on the last-stage stator blade. If the accumulated work is negative, it means that there is a risk of blockage and flutter of the last-stage stator blade. By observing the accumulated work distribution, it can be seen that the accumulated work in most areas of the last-stage stator blade surface is negative, especially near the blade tip and the trailing edge, where the accumulated work increases significantly. Figure 7 The relative Mach number distribution at 90% blade height is shown. According to the relative Mach number distribution diagram of the flow field (i.e., the relative Mach number cloud diagram), it can be seen that there is a strong shock wave in the last-stage stator channel.

[0052] In summary, it can be considered that the last-stage stator has a high risk of first-order mode blocking flutter; secondly, the aerodynamic damping LogDec calculated for the second-order mode (M2) and the third-order mode (M3) is aero The minimum value of is greater than 0, indicating that there is no risk of blocking flutter in the second and third order modes of the stator blades.

[0053] An electronic device includes: a memory for storing a computer program, the memory being a carrier for storing resources, and may be a read-only memory, a random access memory, a disk or an optical disk, etc., the resources stored thereon may include an operating system, a computer program, etc., and the storage method may be temporary storage or permanent storage; a processor, being used to execute the computer program to implement the last-stage stator blade blockage flutter analysis method.

[0054] A computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method for analyzing the blockage flutter of a last-stage stationary blade is implemented.

Claims

1. A method for analyzing the blockage and flutter of the last-stage stator blades of an industrial axial compressor, characterized in that: The following steps are involved: Step 1: Set the boundary conditions and other model parameters when the industrial axial flow compressor is in a blocked condition, calculate and obtain the single-channel steady-state CFD when the industrial axial flow compressor is in a blocked condition, the boundary conditions include the inlet total pressure, the inlet total temperature, and the outlet static pressure, and the other model parameters include the turbulence model, the solid wall model, and the rotary-static interface model; Step 2: Perform structural dynamic modal analysis on the last-stage stator blade and tenon overall model of the industrial axial compressor described in step 1 under non-blocking working conditions to obtain the first three-order natural frequencies and vibration modes of the last-stage stator blade. The parameters that need to be preset before the structural dynamic modal analysis include finite element calculation grid units, elastic modulus, Poisson's ratio, and density; Step 3: interpolate the vibration mode of the last-stage stator blade described in step 2 from the normalized mode of the structure domain mesh quality to the fluid domain mesh, and vibrate the last-stage stator blade according to the first three-order natural frequencies and vibration modes described in step 2; Step 4: Taking the single-channel steady-state CFD of the industrial axial compressor in the blocked condition described in step 1 as the initial field, perform the single-channel unsteady CFD analysis of the vibrating last-stage stator blade under different inter-blade phase angle conditions for each frequency of the first three natural frequencies described in step 2 to obtain the aerodynamic damping LogDec of the last-stage stator blade. aero ; Step 5: Use the pneumatic damping LogDec described in step 4 aero The positive or negative value of determines the flutter risk of the last stage stator blades. When the aerodynamic damping LogDec aero When it is negative, there is a risk of flutter in the last stage stator blades. aero When the rotation speed is positive, the last stage stator blades will not flutter.

2. The method for analyzing the blocked flutter of the last stage stator blades according to claim 1, characterized in that: The outlet static pressure described in step 1 simulates the exhaust pressure when the outlet regulating valve is fully opened, and the outlet static pressure is usually lower than 150 kPa; the turbulence model described in step 1 adopts the SA turbulence model; the solid wall model adopts the sliding wall and wall function to process the wall boundary, and the circumferential geometric boundary in the solid wall model adopts the periodic boundary condition; the mixing surface averaging method is used to transmit information between the rotation-static interface models.

3. The method for analyzing the blocked flutter of the last stage stator blades according to claim 1, characterized in that: The overall model of the last-stage stator blade and the tenon described in step 2 adopts a structured hexahedral unit, the hexahedral unit material is X3CrNiMo13-4, the overall model of the last-stage stator blade and the tenon adopts a cyclic symmetric model, and a contact constraint is set between the tenon and the root groove of the last-stage stator blade.

4. The method for analyzing the blocked flutter of the last stage stator blades according to claim 1, characterized in that: In step 4, the harmonic balance method is used in the single-channel unsteady CFD analysis, and phase lag boundary conditions should be used on the upper and lower periodic boundaries in the single-channel unsteady CFD analysis.

5. The method for analyzing the blocked flutter of the last stage stator blades according to claim 1 or 4, characterized in that: Step 4 specifically includes the following steps: Step 4.1: For the different inter-blade phase angle conditions described in step 4, the trigonometric function interpolation method is used to perform flutter assessment, analyze multiple inter-blade phase angles and calculate the aerodynamic damping LogDec under multiple inter-blade phase angles aero , the phase angle δ is expressed by the following formula: In formula (1), N is the number of wave node diameters of the vibration of the last stage stator blades, and Nb is the number of the last stage stator blades; Pneumatic damping LogDec aero It is expressed by the following formula: The accumulated work on the surface of the last stage stator blade W aero It is expressed by the following formula: In formula (3), A x is the specified vibration displacement of the last stage stator blade, A f1 is the unsteady aerodynamic force of the last-stage stator blade obtained by single-channel unsteady CFD solution, It is the phase difference between the unsteady force CFD and the vibration of the last stage stator blade; Strain Energy E strain It is expressed by the following formula: In formula (4), m is the mass of the last stage stator blade, ω is the natural frequency, and A is x is the specified vibration displacement of the last stage stator blade; Step 4.2: According to the aerodynamic damping LogDec at multiple inter-blade phase angles described in step 4.1 aero , a complete aerodynamic damping curve is obtained by using trigonometric function interpolation, and the logarithmic attenuation rate is used to characterize the aerodynamic damping curve.

6. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the last-stage stator blade blockage flutter analysis method according to any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that: Used to store a computer program, which, when executed by a processor, implements the last-stage stator blade blockage flutter analysis method as described in any one of claims 1 to 5.

Images