A turbine blade life performance analysis model

Abstract

The present invention discloses a turbine blade life performance analysis model, which relates to the technical field of turbine blade performance analysis. The method comprises: performing laser scanning and point cloud modeling on a blade to be analyzed to obtain a finite element model; simulating the finite element model to obtain an equivalent stress distribution diagram and an equivalent strain distribution diagram, dividing the blade area and performing a primary screening process to obtain target points; measuring the target points to obtain target vibration data, calculating the mutual correlation coefficient of adjacent points and performing a secondary screening process to obtain a set of high-risk points; calculating a frequency response ratio function between adjacent high-risk points, and calculating an interaction damage factor based on the frequency response ratio function; calculating a nonlinear coupling total loss value based on the finite element model, the equivalent stress distribution diagram, the equivalent strain distribution diagram, and the interaction damage factor, and calculating the remaining life of the blade based on the nonlinear coupling total loss value.

Description

Technical Field

[0001] The present invention relates to the technical field of turbine blade performance analysis, and in particular to a turbine blade life performance analysis model. Background Art

[0002] Turbine blades are core components of industrial drive systems such as gas turbines, steam turbines, hydroturbines, and aircraft turbofan engines, and are widely used in power grid peak shaving and distributed energy applications. In gas turbine applications, high-temperature, high-pressure gas generated in the combustion chamber drives the turbine blades to rotate, causing the gas to gradually expand and generate work within the blades at different levels, converting the gas's thermal and pressure energy into mechanical energy, enabling large-scale power generation. Turbine blades must not only withstand complex forces such as gas erosion and thermal stress, but also control the gas turbine's power output by adjusting the blade angle and number. Therefore, only high-performance and reliable performance can ensure the long-term, stable operation of gas turbines. Life performance analysis can proactively identify potential blade safety hazards, preventing serious accidents such as gas turbine downtime and blade breakage caused by blade failure. It also facilitates the development of more rational maintenance plans, determining appropriate overhaul times and replacement cycles, and ultimately improving equipment availability. Given the widespread use of turbines in my country's industrial sector, scientific analysis of turbine blade life is a promising research direction.

[0003] Currently, a Chinese invention patent application with application number CN202010830110.6 discloses a life assessment model for gas turbine blades based on fatigue-creep interaction damage. This application includes a blade temperature model, a blade stress model, and a creep-fatigue interaction damage model. The blade temperature model accounts for the effects of thermal barrier coatings, film cooling, and metal wall thickness. The blade stress model considers the effects of aerodynamic and centrifugal forces on the blade, and also considers the equivalent superposition of blade stress due to blade pitch. The creep-fatigue interaction damage model is based on the SN curve and Larson-Miller parameters. Through implementation of this application, creep and fatigue damage of blade metal can be assessed in real time, providing a basis for gas turbine blade life prediction and optimized operation. However, this application focuses on thermal temperature analysis and insufficient consideration of high-cycle damage from a physical perspective. Creep damage typically manifests as bulging on the blade surface, which is easy to detect. However, turbine blade damage caused by high-pressure shock and centrifugal loads is difficult to detect and is a major cause of serious blade accidents. Summary of the Invention

[0004] The technical problem addressed by this invention is that conventional methods focus on thermal and temperature analysis, but fail to adequately consider the physical aspects of high-frequency vibration damage. Due to the significant differences between various types of turbine blades, conventional methods lack a scientific screening method to identify high-risk points for targeted lifespan analysis.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions, which specifically include:

[0006] Step S1: Laser scan the blade to be analyzed and obtain a finite element model through point cloud modeling;

[0007] Step S2: obtaining an equivalent stress distribution diagram and an equivalent strain distribution diagram based on finite element model simulation, dividing the blade area and performing a primary screening process to obtain target points;

[0008] Step S3: Measure the target points to obtain target vibration data, calculate the correlation coefficients of adjacent points, and perform secondary screening to obtain a set of high-risk points;

[0009] Step S4: Calculate the frequency response ratio function between adjacent high-risk points, and obtain the interaction damage factor based on the frequency response ratio function;

[0010] Step S5: Calculate the nonlinear coupling total loss value based on the finite element model, the equivalent stress distribution diagram, the equivalent strain distribution diagram and the interactive damage factor, and calculate the remaining life of the blade based on the nonlinear coupling total loss value.

[0011] Preferably, the blade to be analyzed is scanned by 3D laser to obtain point cloud data, the point cloud data is denoised by Gaussian filtering, and an initial geometric model is established based on the point cloud data and NURBS surface reconstruction algorithm;

[0012] The initial geometric model is meshed using eight-node hexahedral elements. The unit distortion of the model mesh is checked using the Jacobian matrix determinant. When the unit distortion is greater than or equal to a first threshold, the meshing algorithm is triggered. The point cloud is aligned iteratively using the ICP algorithm to obtain a mesh model.

[0013] The elastic modulus tensor, Poisson's ratio matrix, and thermal expansion coefficient tensor corresponding to the blade material to be analyzed are input into the mesh model through the ANSYS APDL script, and the boundary conditions are imported for solidification to obtain the blade finite element model;

[0014] The boundary conditions include turbine inlet temperature, aerodynamic load, outlet static pressure and rotational speed.

[0015] Preferably, simulation is performed based on a finite element model to obtain an equivalent stress distribution diagram and an equivalent strain distribution diagram of the blade to be tested;

[0016] The processing logic for obtaining the equivalent stress distribution diagram and the equivalent strain distribution diagram of the blade to be tested includes:

[0017] The aerodynamic load distribution is obtained by calculating the Reynolds-averaged Navier-Stokes equations through CFX, and the pressure flow field is obtained. The pressure flow field and temperature field are mapped to the structural grid of the blade finite element model through a high-order interpolation algorithm. The nonlinear equilibrium equation is calculated using the Newton-Raphson algorithm and the preset displacement increment limit to obtain the equivalent stress distribution diagram and the equivalent strain distribution diagram.

[0018] Preferably, the blade to be analyzed is divided into four sub-regions, wherein the four sub-regions include a blade root, a blade leading edge, a blade trailing edge, and a blade cooling hole portion;

[0019] A target point coordinate set is obtained by performing a screening based on the equivalent stress distribution map and the equivalent strain distribution map. The processing logic of the screening includes:

[0020] Based on the equivalent stress distribution diagram, the equivalent stresses of the four sub-areas are sorted respectively, and A1 high stress load points are selected based on the preset first screening ratio set;

[0021] Based on the equivalent strain distribution diagram, the equivalent strains of the four sub-regions are sorted respectively, and A2 strain high load points are selected based on the preset second screening ratio;

[0022] Matching is performed based on the historical damage database, and the maintenance data of the blade of the same model as the blade to be analyzed is retrieved to obtain the historical damage points. Based on the historical damage points, A1 stress high load points and A2 strain high load points, the target point coordinate set is obtained.

[0023] Preferably, a vibration sensor is set based on the target point coordinate set of the blade to be analyzed, and vibration time domain data is acquired based on the vibration sensor, and the sampling frequency of the vibration sensor is set to twice the highest-order modal frequency of the blade to be analyzed;

[0024] The first vibration frequency domain data is obtained by calculating the vibration time domain data through the fast Fourier transform algorithm. The frequency domain power spectrum density data is calculated based on the first vibration frequency domain data corresponding to the adjacent target points. The adjacent point mutual correlation coefficient is calculated based on the frequency domain power spectrum density data. The calculation expression includes:

[0025] ;

[0026] Among them, CR represents the neighbor correlation coefficient, represents the frequency, p represents the target point number, q represents the target point number adjacent to p, Represents the cross-power spectrum density between target point p and target point q, represents the autopower spectrum density of the target point p, Represents the autopower spectral density of the target point q.

[0027] Preferably, the neighbor point mutual correlation coefficients are secondary screened using a preset mutual correlation coefficient threshold to obtain a high-risk point set. The processing logic for the secondary screening includes:

[0028] When the mutual correlation coefficient of neighboring points is less than the preset mutual correlation coefficient threshold, the corresponding target point will be deleted;

[0029] When the mutual correlation coefficient of adjacent points is greater than or equal to the preset mutual correlation coefficient threshold, the corresponding two target points are retained;

[0030] Summarize all the retained target points to obtain a set of high-risk points.

[0031] Preferably, data corresponding to the high-risk point set is extracted from the first vibration frequency domain data to obtain second vibration frequency domain data, and the second vibration frequency domain data is calculated by an H1 estimator to obtain a frequency response function corresponding to each high-risk point;

[0032] Based on the frequency response function of each high-risk point, the frequency response ratio function between adjacent high-risk points is calculated. Based on the frequency response ratio function, the interaction damage factor IDF is calculated. The calculation expression includes:

[0033] ;

[0034] ;

[0035] in, represents the frequency, m represents the target point number, n represents the target point number adjacent to m, Represents the frequency response ratio function between target point m and target point n, Represents the reference frequency response ratio function between target point m and target point n, Represents the system response function matrix at the target point m, represents the system response function matrix at the target point n, IDF represents the interaction damage factor, Indicates the upper limit of the frequency range, Indicates the lower limit of the frequency range.

[0036] Preferably, the nonlinear coupling total loss value is calculated based on the finite element model, the equivalent stress distribution diagram, the equivalent strain distribution diagram and the interactive damage factor, and the blade remaining life is calculated based on the nonlinear coupling total loss value, and its calculation expression includes:

[0037] ;

[0038] in, Indicates the remaining life of the blade. Indicates the blade design life, Represents the total nonlinear coupling loss value.

[0039] Preferably, the stress amplitude and stress value data corresponding to the high-risk point are obtained based on the equivalent stress distribution diagram, the stress value data are averaged to calculate the average stress, the number of failure cycles is calculated by combining the average stress, stress amplitude and Morrow elastic stress correction model, and the fatigue damage accumulation value is calculated based on the number of failure cycles and Miner's rule;

[0040] Based on the equivalent strain distribution diagram, the steady-state creep stress and temperature field data corresponding to the high-risk points are obtained. The creep rupture time is calculated in combination with the Larson-Miller equation. The creep damage accumulation value is calculated based on the creep rupture time and the Miner rule. The calculation expression includes:

[0041] ;

[0042] ;

[0043] in, represents the cumulative value of fatigue damage, i represents the stress number, I represents the total stress, represents the actual number of cycles corresponding to the i-th stress, represents the number of cycles of material fatigue failure corresponding to the i-th stress, represents the cumulative value of creep damage, j represents the stress-temperature combination number, and J represents the total number of stress-temperature combinations. represents the actual action time corresponding to the jth stress-temperature combination, represents the creep rupture time corresponding to the jth stress-temperature combination.

[0044] Preferably, the interactive damage factor is introduced to improve the Chaboche model, and the total nonlinear coupling loss value is calculated based on the fatigue damage accumulation value and the creep damage accumulation value. The calculation expression includes:

[0045] ;

[0046] in, represents the total loss value of nonlinear coupling, IDF represents the interaction damage factor, represents the segmentation threshold of the interaction damage factor, Indicates the material damage accumulation coefficient. Represents the second-stage coefficient of material damage accumulation.

[0047] The beneficial effects of the present invention are as follows: a screening is performed in combination with the finite element model, which narrows the selection range of key points, is conducive to the rational arrangement of sensors to collect vibration data at vulnerable locations, reduces the amount of calculation for subsequent analysis, and at the same time, combined with the records of damage locations in the historical maintenance database, increases the accuracy of identifying high-risk points. Compared with slowly developing creep damage and easily observed thermal corrosion damage, the hidden damage and cracks on heavy-duty gas turbine blades have the characteristics of rapid development and huge hidden dangers. From the perspective of vibration data, secondary screening is performed to accurately locate vulnerable areas with potential high risks. An interactive damage factor IDF is established, and the transmission law of IDF damage between adjacent points is established, thereby enhancing the accuracy of the calculation of blade fatigue-creep interactive damage. The introduction of the interactive damage factor IDF to modify the life calculation coupling model increases the applicability to blades in different conditions, which is conducive to obtaining accurate blade remaining life. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of the basic flow of a turbine blade life performance analysis model provided by one embodiment of the present invention;

[0049] Figure 2 An equivalent stress distribution diagram provided for one embodiment of the present invention;

[0050] Figure 3 This is an equivalent strain distribution diagram provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0052] Reference Figure 1-Figure 3 , as one embodiment of the present invention, provides a turbine blade life performance analysis model, including:

[0053] Step S1: Laser scan the blade to be analyzed and obtain a finite element model through point cloud modeling;

[0054] Step S2: obtaining an equivalent stress distribution diagram and an equivalent strain distribution diagram based on finite element model simulation, dividing the blade area and performing a primary screening process to obtain target points;

[0055] Step S3: Measure the target points to obtain target vibration data, calculate the correlation coefficients of adjacent points, and perform secondary screening to obtain a set of high-risk points;

[0056] Step S4: Calculate the frequency response ratio function between adjacent high-risk points, and obtain the interaction damage factor based on the frequency response ratio function;

[0057] Step S5: Calculate the nonlinear coupling total loss value based on the finite element model, the equivalent stress distribution diagram, the equivalent strain distribution diagram and the interactive damage factor, and calculate the remaining life of the blade based on the nonlinear coupling total loss value.

[0058] In this embodiment, 3D laser scanning is performed on the blade to be analyzed to obtain point cloud data, the point cloud data is denoised using Gaussian filtering, and an initial geometric model is established based on the point cloud data and the NURBS surface reconstruction algorithm;

[0059] The initial geometric model is meshed using eight-node hexahedral elements. The unit distortion of the model mesh is checked using the Jacobian matrix determinant. When the unit distortion is greater than or equal to a first threshold, the meshing algorithm is triggered. The point cloud is aligned iteratively using the ICP algorithm to obtain a mesh model.

[0060] The elastic modulus tensor, Poisson's ratio matrix, and thermal expansion coefficient tensor corresponding to the blade material to be analyzed are input into the mesh model through the ANSYS APDL script, and the boundary conditions are imported for solidification to obtain the blade finite element model;

[0061] The boundary conditions include turbine inlet temperature, aerodynamic load, outlet static pressure and speed.

[0062] In this embodiment, simulation is performed based on a finite element model to obtain an equivalent stress distribution diagram and an equivalent strain distribution diagram of the blade to be tested;

[0063] The processing logic for obtaining the equivalent stress distribution diagram and the equivalent strain distribution diagram of the blade to be tested includes:

[0064] The aerodynamic load distribution is obtained by calculating the Reynolds-averaged Navier-Stokes equations through CFX, and the pressure flow field is obtained. The pressure flow field and temperature field are mapped to the structural grid of the blade finite element model through a high-order interpolation algorithm. The nonlinear equilibrium equation is calculated using the Newton-Raphson algorithm and the preset displacement increment limit to obtain the equivalent stress distribution diagram and the equivalent strain distribution diagram.

[0065] In this embodiment, the blade to be analyzed is divided into four sub-regions, including the blade root, the blade leading edge, the blade trailing edge and the blade cooling hole portion;

[0066] A screening is performed based on the equivalent stress distribution diagram and the equivalent strain distribution diagram to obtain a target point coordinate set. The processing logic of the screening includes:

[0067] Based on the equivalent stress distribution diagram, the equivalent stresses of the four sub-areas are sorted respectively, and A1 high stress load points are selected based on the preset first screening ratio set;

[0068] Based on the equivalent strain distribution diagram, the equivalent strains of the four sub-regions are sorted respectively, and A2 strain high load points are selected based on the preset second screening ratio;

[0069] Matching is performed based on the historical damage database, and the maintenance data of the blade of the same model as the blade to be analyzed is retrieved to obtain the historical damage points. Based on the historical damage points, A1 stress high load points and A2 strain high load points, the target point coordinate set is obtained.

[0070] Among them, the scope of key points is narrowed through one screening, which is conducive to the rational arrangement of sensors to collect vibration data at easily damaged locations, reducing the amount of calculation for subsequent analysis. At the same time, combined with the records of damage locations in the historical maintenance database, the identification accuracy of high-risk points is increased.

[0071] In this embodiment, a vibration sensor is set based on the target point coordinate set of the blade to be analyzed, and vibration time domain data is collected based on the vibration sensor. The sampling frequency of the vibration sensor is set to twice the highest modal frequency of the blade to be analyzed;

[0072] The first vibration frequency domain data is obtained by calculating the vibration time domain data through the fast Fourier transform algorithm. The frequency domain power spectrum density data is calculated based on the first vibration frequency domain data corresponding to the adjacent target points. The adjacent point mutual correlation coefficient is calculated based on the frequency domain power spectrum density data. The calculation expression includes:

[0073] ;

[0074] Among them, CR represents the neighbor correlation coefficient, represents the frequency, p represents the target point number, q represents the target point number adjacent to p, Represents the cross-power spectrum density between target point p and target point q, represents the autopower spectrum density of the target point p, Represents the autopower spectral density of the target point q.

[0075] In this embodiment, the neighboring point mutual correlation coefficients are secondary screened using a preset mutual correlation coefficient threshold to obtain a high-risk point set. The processing logic for the secondary screening includes:

[0076] When the mutual correlation coefficient of neighboring points is less than the preset mutual correlation coefficient threshold, the corresponding target point will be deleted;

[0077] When the mutual correlation coefficient of adjacent points is greater than or equal to the preset mutual correlation coefficient threshold, the corresponding two target points are retained;

[0078] Summarize all the retained target points to obtain a set of high-risk points.

[0079] In this embodiment, data corresponding to the high-risk point set is extracted from the first vibration frequency domain data to obtain second vibration frequency domain data, and the second vibration frequency domain data is calculated by the H1 estimator to obtain the frequency response function corresponding to each high-risk point;

[0080] Based on the frequency response function of each high-risk point, the frequency response ratio function between adjacent high-risk points is calculated. Based on the frequency response ratio function, the interaction damage factor IDF is calculated. The calculation expression includes:

[0081] ;

[0082] ;

[0083] in, represents the frequency, m represents the target point number, n represents the target point number adjacent to m, Represents the frequency response ratio function between target point m and target point n, Represents the reference frequency response ratio function between target point m and target point n, Represents the system response function matrix at the target point m, represents the system response function matrix at the target point n, IDF represents the interaction damage factor, Indicates the upper limit of the frequency range, Indicates the lower limit of the frequency range.

[0084] Compared to slowly developing creep damage and easily observable thermal corrosion damage, hidden damage and cracks on heavy-duty gas turbine blades develop rapidly and pose significant risks. By analyzing vibration data and performing secondary screening, we pinpointed potentially high-risk, vulnerable areas. We established an interactive damage factor (IDF) and characterized the transmission of IDF damage between adjacent points, thereby enhancing the accuracy of blade fatigue-creep interaction damage calculations.

[0085] In this embodiment, the nonlinear coupling total loss value is calculated based on the finite element model, the equivalent stress distribution diagram, the equivalent strain distribution diagram, and the interaction damage factor. The remaining life of the blade is calculated based on the nonlinear coupling total loss value. The calculation expression includes:

[0086] ;

[0087] in, Indicates the remaining life of the blade. Indicates the blade design life, Represents the total nonlinear coupling loss value.

[0088] In this embodiment, the stress amplitude and stress value data corresponding to the high-risk point are obtained based on the equivalent stress distribution diagram, the stress value data are averaged to calculate the average stress, and the number of failure cycles is calculated by combining the average stress, stress amplitude and Morrow elastic stress correction model. The fatigue damage accumulation value is calculated based on the number of failure cycles and Miner's rule;

[0089] Based on the equivalent strain distribution diagram, the steady-state creep stress and temperature field data corresponding to the high-risk points are obtained. The creep rupture time is calculated in combination with the Larson-Miller equation. The creep damage accumulation value is calculated based on the creep rupture time and the Miner rule. The calculation expression includes:

[0090] ;

[0091] ;

[0092] in, represents the cumulative value of fatigue damage, i represents the stress number, I represents the total stress, represents the actual number of cycles corresponding to the i-th stress, represents the number of cycles of material fatigue failure corresponding to the i-th stress, represents the cumulative value of creep damage, j represents the stress-temperature combination number, and J represents the total number of stress-temperature combinations. represents the actual action time corresponding to the jth stress-temperature combination, represents the creep rupture time corresponding to the jth stress-temperature combination.

[0093] In this embodiment, the interactive damage factor is introduced to improve the Chaboche model. The total nonlinear coupling loss value is calculated based on the cumulative fatigue damage value and the cumulative creep damage value. The calculation expression includes:

[0094] ;

[0095] in, represents the total loss value of nonlinear coupling, IDF represents the interaction damage factor, represents the segmentation threshold of the interaction damage factor, Indicates the material damage accumulation coefficient. Represents the second-stage coefficient of material damage accumulation.

[0096] Among them, the traditional method that relies on the fatigue Morrow model and creep Larson-Miller model is improved, and the interactive damage factor IDF is introduced to correct the life calculation coupling model, which increases the applicability to blades in different conditions and is conducive to obtaining accurate blade remaining life.

[0097] Those skilled in the art will appreciate that embodiments of the present invention may provide methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage media may be based on any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0098] It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limiting. Although the present invention is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A turbine blade life performance analysis model, characterized in that: include: Step S1: Laser scan the blade to be analyzed and obtain a finite element model through point cloud modeling; Step S2: obtaining an equivalent stress distribution diagram and an equivalent strain distribution diagram based on finite element model simulation, dividing the blade area and performing a primary screening process to obtain target points; Step S3: Measure the target points to obtain target vibration data, calculate the correlation coefficients of adjacent points, and perform secondary screening to obtain a set of high-risk points; Step S4: Calculate the frequency response ratio function between adjacent high-risk points, and obtain the interaction damage factor based on the frequency response ratio function; Step S5: calculating a nonlinear coupling total loss value based on the finite element model, the equivalent stress distribution diagram, the equivalent strain distribution diagram, and the interaction damage factor, and calculating the remaining life of the blade based on the nonlinear coupling total loss value; Wherein, a vibration sensor is set based on the target point coordinate set of the blade to be analyzed, and vibration time domain data is acquired based on the vibration sensor. The sampling frequency of the vibration sensor is set to twice the highest-order modal frequency of the blade to be analyzed; The first vibration frequency domain data is obtained by calculating the vibration time domain data through the fast Fourier transform algorithm. The frequency domain power spectrum density data is calculated based on the first vibration frequency domain data corresponding to the adjacent target points. The adjacent point mutual correlation coefficient is calculated based on the frequency domain power spectrum density data. The calculation expression includes: ； Among them, CR represents the neighbor correlation coefficient, represents the frequency, p represents the target point number, q represents the target point number adjacent to p, Represents the cross-power spectrum density between target point p and target point q, represents the autopower spectrum density of the target point p, Represents the autopower spectral density of the target point q.

2. The turbine blade life performance analysis model according to claim 1, wherein: The blade to be analyzed is scanned by 3D laser to obtain point cloud data, which is then denoised using Gaussian filtering. An initial geometric model is then established based on the point cloud data and the NURBS surface reconstruction algorithm. The initial geometric model is meshed using eight-node hexahedral elements. The unit distortion of the model mesh is checked using the Jacobian matrix determinant. When the unit distortion is greater than or equal to a first threshold, the meshing algorithm is triggered. The point cloud is aligned iteratively using the ICP algorithm to obtain a mesh model. The elastic modulus tensor, Poisson's ratio matrix, and thermal expansion coefficient tensor corresponding to the blade material to be analyzed are input into the mesh model through the ANSYS APDL script, and the boundary conditions are imported for solidification to obtain the blade finite element model; The boundary conditions include turbine inlet temperature, aerodynamic load, outlet static pressure and rotational speed.

3. The turbine blade life performance analysis model according to claim 1, wherein: Perform simulation based on the finite element model to obtain the equivalent stress distribution diagram and equivalent strain distribution diagram of the blade to be tested; The processing logic for obtaining the equivalent stress distribution diagram and the equivalent strain distribution diagram of the blade to be tested includes: The aerodynamic load distribution is obtained by calculating the Reynolds-averaged Navier-Stokes equations through CFX, and the pressure flow field is obtained. The pressure flow field and temperature field are mapped to the structural grid of the blade finite element model through a high-order interpolation algorithm. The nonlinear equilibrium equation is calculated using the Newton-Raphson algorithm and the preset displacement increment limit to obtain the equivalent stress distribution diagram and the equivalent strain distribution diagram.

4. The turbine blade life performance analysis model according to claim 3, wherein: Dividing the blade to be analyzed into four sub-regions, wherein the four sub-regions include the blade root, the blade leading edge, the blade trailing edge and the blade cooling hole portion; A target point coordinate set is obtained by performing a screening based on the equivalent stress distribution map and the equivalent strain distribution map. The processing logic of the screening includes: Based on the equivalent stress distribution diagram, the equivalent stresses of the four sub-areas are sorted respectively, and A1 high stress load points are selected based on the preset first screening ratio set; Based on the equivalent strain distribution diagram, the equivalent strains of the four sub-regions are sorted respectively, and A2 strain high load points are selected based on the preset second screening ratio; Matching is performed based on the historical damage database, and the maintenance data of the blade of the same model as the blade to be analyzed is retrieved to obtain the historical damage points. Based on the historical damage points, A1 stress high load points and A2 strain high load points, the target point coordinate set is obtained.

5. The turbine blade life performance analysis model according to claim 1, wherein: The high-risk point set is obtained by performing secondary screening on the mutual correlation coefficients of adjacent points using a preset mutual correlation coefficient threshold. The processing logic for the secondary screening includes: When the mutual correlation coefficient of neighboring points is less than the preset mutual correlation coefficient threshold, the corresponding target point will be deleted; When the mutual correlation coefficient of adjacent points is greater than or equal to the preset mutual correlation coefficient threshold, the corresponding two target points are retained; Summarize all the retained target points to obtain a set of high-risk points.

6. The turbine blade life performance analysis model according to claim 1, wherein: Extracting data corresponding to a set of high-risk points from the first vibration frequency domain data to obtain second vibration frequency domain data, and calculating the second vibration frequency domain data using an H1 estimator to obtain a frequency response function corresponding to each high-risk point; Based on the frequency response function of each high-risk point, the frequency response ratio function between adjacent high-risk points is calculated. Based on the frequency response ratio function, the interaction damage factor IDF is calculated. The calculation expression includes: ； ； in, represents the frequency, m represents the target point number, n represents the target point number adjacent to m, Represents the frequency response ratio function between target point m and target point n, Represents the reference frequency response ratio function between target point m and target point n, Represents the system response function matrix at the target point m, represents the system response function matrix at the target point n, IDF represents the interaction damage factor, Indicates the upper limit of the frequency range, Indicates the lower limit of the frequency range.

7. The turbine blade life performance analysis model according to claim 1, wherein: The nonlinear coupling total loss value is calculated based on the finite element model, equivalent stress distribution diagram, equivalent strain distribution diagram and interactive damage factor. The remaining life of the blade is calculated based on the nonlinear coupling total loss value. The calculation expression includes: ； in, Indicates the remaining life of the blade. Indicates the blade design life, Represents the total nonlinear coupling loss value.

8. The turbine blade life performance analysis model according to claim 7, characterized in that: Based on the equivalent stress distribution diagram, the stress amplitude and stress value data corresponding to the high-risk points are obtained. The average stress is calculated by averaging the stress value data. The number of failure cycles is calculated by combining the average stress, stress amplitude and Morrow elastic stress correction model. The fatigue damage accumulation value is calculated based on the number of failure cycles and Miner's rule. Based on the equivalent strain distribution diagram, the steady-state creep stress and temperature field data corresponding to the high-risk points are obtained. The creep rupture time is calculated in combination with the Larson-Miller equation. The creep damage accumulation value is calculated based on the creep rupture time and the Miner rule. The calculation expression includes: ； ； in, represents the cumulative value of fatigue damage, i represents the stress number, I represents the total stress, represents the actual number of cycles corresponding to the i-th stress, represents the number of cycles of material fatigue failure corresponding to the i-th stress, represents the cumulative value of creep damage, j represents the stress-temperature combination number, and J represents the total number of stress-temperature combinations. represents the actual action time corresponding to the jth stress-temperature combination, represents the creep rupture time corresponding to the jth stress-temperature combination.

9. The turbine blade life performance analysis model according to claim 8, characterized in that: The Chaboche model is improved by introducing the interactive damage factor. The total nonlinear coupling loss value is calculated based on the cumulative fatigue damage value and the cumulative creep damage value. The calculation expression includes: ； in, represents the total loss value of nonlinear coupling, IDF represents the interaction damage factor, represents the segmentation threshold of the interaction damage factor, Indicates the material damage accumulation coefficient. Represents the second-stage coefficient of material damage accumulation.

Images