Novel ship gas turbine compressor blade vibration stress optimization method

Abstract

The invention relates to the field of gas turbines, in particular to a novel ship gas turbine compressor blade vibration stress optimization method, which comprises the following steps: carrying out vibration stress simulation analysis on a constructed finite element simulation model to obtain definition data of a vibration stress concentration position, and then completing the correlation between the stress concentration position and blade profile geometric characteristics. And obtaining blade profile modeling coefficient data of each sensitive section, positioning a sensitive section range, outputting a standard aerodynamic performance evaluation report and a strength vibration analysis report, and carrying out integration optimization to form a final blade structure. According to the method, the fatigue performance of the blade can be easily and efficiently improved, the stress concentration phenomenon of the blade is optimized, compared with methods such as redesign optimization and blade profile installation angle adjustment, the excellent aerodynamic performance of an original advanced blade profile is reserved, the requirement for blade vibration stress distribution optimization can be met, and the method is suitable for large-scale popularization and application. And the anti-fatigue performance of the blade is further improved.

Description

Technical Field

[0001] This invention relates to the field of gas turbines, and more specifically to a novel method for optimizing the vibration stress of compressor blades in marine gas turbines. Background Technology

[0002] Current designs for new gas turbine compressor blades employ high-performance compressor airfoils, but this often results in severe stress concentration due to blade vibration. During fatigue testing, this leads to blades failing to meet design performance expectations and fracturing prematurely. To address this challenge, compressor blade design often employs optimization methods such as directly redesigning the airfoil, simply increasing the blade transition radius, or adjusting the airfoil's installation angle.

[0003] However, obtaining new and satisfactory blade profiles through these methods typically involves multiple iterations, constantly balancing performance and vibration stress, resulting in extremely low design efficiency and unsatisfactory final results. A key technology that urgently needs to be addressed is how to quickly optimize vibration stress distribution while maintaining the functional performance profile of the blade, thereby improving its fatigue performance. Furthermore, it is essential to consider both blade vibration reliability and aerodynamic performance requirements. Summary of the Invention

[0004] This invention addresses the technical problems existing in the prior art by providing a novel method for optimizing the vibration stress of marine gas turbine compressor blades.

[0005] The technical solution of this invention to solve the above-mentioned technical problems is as follows: A novel method for optimizing the vibration stress of marine gas turbine compressor blades, the method comprising: S101. Based on the basic design data of the compressor blades of the gas turbine of the target ship, a finite element simulation model is constructed, and the operating parameters of the gas turbine are introduced. After applying the equivalent load, vibration stress simulation analysis is carried out on the constructed finite element simulation model to obtain the definition data of the vibration stress concentration location. S102. Based on the obtained definition data of vibration stress concentration location, complete the association between stress concentration location and airfoil geometric features, divide the benchmark and range of the modeling coefficient, and obtain the airfoil modeling coefficient data of each sensitive section. S103. Based on the definition data of the vibration stress concentration location and the air profile modeling coefficient data of each sensitive section, locate the range of the sensitive section, implement the modeling thickening of the sensitive section and calculate the key dimensions of the thickened air profile to obtain locally optimized thickened air profile data and thickened air profile key dimension parameters. S104. Based on the locally optimized thickened airfoil data and key dimensional parameters of the thickened airfoil, conduct aerodynamic performance evaluation and strength vibration analysis. If any indicator fails to meet the standard, return to S102, adjust the modeling coefficient, and repeat the operation until all indicators meet the standard. Finally, output a qualified aerodynamic performance evaluation report and strength vibration analysis report. S105, combining the compliant aerodynamic performance evaluation report, strength and vibration analysis report, and locally optimized thickened airfoil data and key dimensional parameters of the thickened airfoil, integrates and optimizes to form the final blade structure.

[0006] In a preferred embodiment, the basic design data of the gas turbine compressor blade of the target ship in S101 specifically includes: a three-dimensional geometric model of the blade shape, including complete coordinates of the leading edge, trailing edge, suction surface, and pressure surface of each section along the blade height; mechanical parameters of the blade material, such as elastic modulus, Poisson's ratio, density, and fatigue strength; blade installation boundary conditions; and connection method, constraint position, and constraint type with the turbine disk. The core operating parameters of a gas turbine are: design speed, inlet air temperature under typical operating conditions, inlet air pressure, and working fluid flow rate. The model is meshed based on the three-dimensional geometric model of the blade. Fine meshes are used in areas prone to stress concentration, such as the blade root and transition fillets, while transition meshes are used in other areas of the blade body. Based on the actual installation state of the blade, constraints are defined in the model, such as setting fixed constraints on the connection surface between the tenon and the wheel disk to simulate the force transmission after actual assembly. Combined with the operating parameters of the gas turbine, equivalent loads including centrifugal loads, aerodynamic loads, and vibration excitation loads are applied. The centrifugal load is calculated based on the inertial force of each mass point on the blade according to the design speed. The aerodynamic load is converted into a distributed load on the blade surface according to the aerodynamic characteristics. The vibration excitation load is a simulation of vibration loads caused by airflow excitation, rotor imbalance, etc.

[0007] In a preferred embodiment, after applying the equivalent load, step S101 sets the simulation analysis type to transient dynamic analysis, focuses on a preset vibration mode of interest, and solves the vibration response of the blade under the corresponding mode. The preset vibration mode of interest is based on common failure modes of gas turbine compressor blades, and identifies the vibration modes that need to be analyzed, such as bending vibration, torsional vibration, and combined bending-torsional vibration. The vibration stress distribution data of the entire blade is extracted, and stress distribution cloud map, stress variation curve along blade height, and stress distribution curve along cross section are generated. By comparing the stress values ​​of each region of the blade, the parts with stress values ​​higher than the surrounding areas and abrupt changes in stress gradient are identified, and the specific location information of the parts is determined, including the height range along blade height and the circumferential position along cross section, such as the inlet side, exhaust side, leading edge, and trailing edge. The definition data of the vibration stress concentration location, including the height coordinate range along blade height of the stress concentration area, the specific orientation along cross section circumferentially, the stress peak value under the corresponding vibration mode of interest, and the initial value of the stress concentration factor, is obtained, and the vibration stress concentration location is determined.

[0008] In a preferred embodiment, step S102 determines the vibration stress concentration location based on the definition data of the vibration stress concentration location, locates the corresponding blade sensitive section, extracts the leading and trailing edge radius and curvature distribution data of the sensitive section and adjacent sections, and clarifies the correspondence between the leading and trailing edge radius, curvature change rate and stress concentration coefficient based on the leading and trailing edge radius and curvature distribution data. For example, the smaller the leading and trailing edge radius and the larger the curvature, the more significant the stress concentration phenomenon, thus completing the association between the stress concentration location and the blade geometry.

[0009] In a preferred embodiment, step S102 uses the key geometric parameters of the leading and trailing edges of the sensitive section airfoil as the reference dimensions for modeling and amplification, such as the minimum radius of curvature of the elliptical airfoil on the inlet side and the radius of the trailing edge arc. Combined with the initial value of the stress concentration factor and aerodynamic performance constraints, the modeling coefficient range of each sensitive section is defined to ensure that the leading and trailing edges of the airfoil still maintain an aerodynamically friendly geometry after amplification. This completes the division of the modeling coefficient reference and range, and obtains the airfoil modeling coefficient data for each sensitive section, including the modeling amplification reference dimension of each sensitive section, the finally determined modeling coefficient, the modeling amplification direction, and the estimated value of the leading and trailing edge radius after amplification.

[0010] In a preferred embodiment, after S103 locates the range of sensitive sections, a set of sensitive sections with thickened templates is obtained. Based on the original blade shape coordinate data and the determined template coefficient and magnification direction, the blade shape of each sensitive section in the set of sensitive sections is uniformly magnified and the profile is adjusted. Specifically, for the leading edge elliptical segment and the trailing edge arc segment, the radius of curvature is magnified synchronously according to the template coefficient. For the straight or curved segments of the blade body, the corresponding distance is translated along the normal direction of the profile to ensure the geometric continuity of the profile after magnification. Extract the maximum thickness, leading edge radius, and trailing edge radius of each thickened sensitive section and compare them with the original airfoil. Record the dimensional change data, where the maximum thickness is the maximum distance along the chord of the airfoil, the leading edge radius is the radius corresponding to the minimum curvature of the enlarged elliptical airfoil, and the trailing edge radius is the enlarged arc radius. Obtain locally optimized thickened airfoil data including the complete airfoil coordinates after thickening each sensitive section, as well as key dimensional parameters of the thickened airfoil including the maximum thickness, leading edge radius, and trailing edge radius values ​​of each sensitive section, and the dimensional changes compared to the original airfoil.

[0011] In a preferred embodiment, step S104 constructs a compressor flow channel computational domain including the thickened airfoil, covering the inlet, blade passage, and exhaust. After meshing the computational domain, a structured mesh is used on the blade surface to ensure the accuracy of the boundary layer simulation, while an unstructured mesh is used in the flow channel region. Aerodynamic simulation boundary conditions are set, and a turbulence model suitable for turbomachinery is selected to solve the flow field distribution under the design conditions. The aerodynamic simulation boundary conditions include a total temperature and total pressure boundary on the inlet side, a static pressure boundary on the exhaust side, and an adiabatic no-slip boundary on the blade surface. The actual efficiency, flow rate, pressure ratio, and surge margin of the thickened airfoil are obtained by solving and compared with the original design parameters to complete the aerodynamic performance evaluation.

[0012] In a preferred embodiment, after aerodynamic performance evaluation, the finite element simulation model is updated based on the thickened airfoil data. The constraint conditions and equivalent loads of S101 are applied, and strength vibration simulation is carried out. Specifically, the static stress distribution and fatigue life of the blade under steady-state conditions are calculated, and the magnitude of static stress and material yield strength, and the magnitude of fatigue life and design requirements are verified. The static stress must be less than or equal to the material yield strength, and the fatigue life must be greater than or equal to the design requirements. The strength analysis is then completed. Solve for the natural frequency and vibration stress distribution of the thickened airfoil, verify the magnitude of the vibration stress and the maximum allowable vibration stress, and ensure that the vibration stress is less than or equal to the maximum allowable vibration stress to complete the vibration analysis; If any of the indicators in the strength analysis, vibration analysis, or aerodynamic performance evaluation fails to meet the standard, the modeling coefficient should be adjusted and the operation repeated. When all indicators of strength analysis, vibration analysis, and aerodynamic performance evaluation meet the standards, the output includes flow field distribution data, actual aerodynamic performance parameters, and a strength and vibration analysis report including static stress distribution, fatigue life calculation results, natural frequency data, vibration stress distribution, and improved stress concentration factor. The flow field distribution data includes velocity field, pressure field, and temperature field, while the actual aerodynamic performance parameters include efficiency, flow rate, pressure ratio, and surge margin.

[0013] In a preferred embodiment, step S105 integrates the locally optimized thickened airfoil data and the key dimension parameters of the thickened airfoil, and the coordinates of the thickened airfoil at each sensitive section with the original airfoil coordinates at the non-sensitive section according to the original stacking rule, so that the airfoil has continuity and smoothness along the airfoil height direction. The original stacking rule refers to keeping the stacking axis and stacking angle unchanged. Global geometric optimization is performed on the integrated airfoil. For the transition area between the sensitive section and the non-sensitive section, the profile is optimized by curve interpolation method to avoid geometric abrupt changes and prevent the generation of new stress concentrations, thereby obtaining optimized airfoil data. Based on the optimized blade profile data, a complete three-dimensional solid model of the blade is generated, including all structural parts such as the blade body, tenon, and transition fillet. The complete design data of the final blade is extracted, the blade vibration stress is optimized, and the final blade structure is output.

[0014] The beneficial effects of this invention are: it can simply and efficiently improve the fatigue performance of blades. The newly designed blade profile selects the following positions along the blade height: inside the tenon, inside the original transition fillet, and inside the position close to the transition fillet. The scale line is enlarged proportionally along the normal direction of the profile, and a local thickening design is made along the blade height to optimize the stress concentration phenomenon of the blade. Compared with the methods of redesigning and optimizing, and adjusting the blade profile installation angle, it not only retains the excellent aerodynamic performance of the original advanced blade profile, but also meets the needs of optimizing the blade vibration stress distribution, thereby improving the fatigue resistance of the blade. Attached Figure Description

[0015] Figure 1 This is a flowchart of the present invention; Figure 2 This is a schematic diagram of the blade design of the present invention; Figure 3 A schematic diagram of the vibration stress distribution of a compressor blade under a high-performance airfoil. Figure 4 A schematic diagram of the location selection for leaf shape optimization; Figure 5 A schematic diagram of the three-dimensional magnification factor of the sensitive section; Figure 6 A schematic diagram of the magnification factor for the sensitive section in two models; Figure 7 A schematic diagram of the magnification factor for the sensitive section 1; Figure 8 This diagram illustrates the improvement in stress concentration factor from 1.5 to 1.21. Figure 9 This invention is used to evaluate the impact of aerodynamic performance. Figure 10 This is the result of the vibration fatigue test of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] As attached Figure 1 As shown in the figure, this embodiment provides: a novel method for optimizing the vibration stress of marine gas turbine compressor blades, the method comprising: S101. Based on the basic design data of the compressor blades of the gas turbine of the target ship, a finite element simulation model is constructed, and the operating parameters of the gas turbine are introduced. After applying the equivalent load, vibration stress simulation analysis is carried out on the constructed finite element simulation model to obtain the definition data of the vibration stress concentration location. The basic design data of the gas turbine compressor blade of the target ship in S101 are as follows: three-dimensional geometric model of the blade, including complete coordinates of the leading edge, trailing edge, suction surface and pressure surface of each section along the blade height, mechanical parameters of the blade material, such as elastic modulus, Poisson's ratio, density, fatigue strength, etc., blade installation boundary conditions, connection method with the turbine disk, constraint position and constraint type; The core operating parameters of a gas turbine are: design speed, inlet air temperature under typical operating conditions, inlet air pressure, and working fluid flow rate. The model is meshed based on the three-dimensional geometric model of the blade. Fine meshes are used in areas prone to stress concentration, such as the blade root and transition fillets, while transition meshes are used in other areas of the blade body. Based on the actual installation state of the blade, constraints are defined in the model, such as setting fixed constraints on the connection surface between the tenon and the wheel disk to simulate the force transmission after actual assembly. Combined with the operating parameters of the gas turbine, equivalent loads including centrifugal loads, aerodynamic loads, and vibration excitation loads are applied. The centrifugal load is calculated based on the inertial force of each mass point on the blade according to the design speed. The aerodynamic load is converted into a distributed load on the blade surface according to the aerodynamic characteristics. The vibration excitation load is a simulation of vibration loads caused by airflow excitation, rotor imbalance, etc.

[0018] After applying the equivalent load, S101 sets the simulation analysis type to transient dynamics analysis, focuses on the preset vibration modes of interest, and solves the vibration response of the blade under the corresponding modes. The preset vibration modes of interest are based on common failure modes of gas turbine compressor blades, and the vibration modes that need to be analyzed are identified, such as bending vibration, torsional vibration, and combined bending and torsional vibration. The vibration stress distribution data of the entire blade is extracted, and stress distribution cloud map, stress variation curve along blade height, and stress distribution curve along cross section are generated. By comparing the stress values ​​of each region of the blade, the parts with stress values ​​higher than the surrounding areas and abrupt changes in stress gradient are identified, and the specific location information of the parts is identified, including the height range along blade height and the circumferential position along cross section, such as the inlet side, exhaust side, leading edge, and trailing edge. The definition data of the vibration stress concentration location, including the height coordinate range along blade height of the stress concentration area, the specific orientation along cross section circumferentially, the stress peak value of the corresponding vibration mode of interest, and the initial value of the stress concentration factor, are obtained, and the vibration stress concentration location is determined.

[0019] S102. Based on the obtained definition data of vibration stress concentration location, complete the association between stress concentration location and airfoil geometric features, divide the benchmark and range of the modeling coefficient, and obtain the airfoil modeling coefficient data of each sensitive section. S102 determines the vibration stress concentration location based on the definition data of the vibration stress concentration location, locates the corresponding blade sensitive section, extracts the leading and trailing edge radius and curvature distribution data of the sensitive section and adjacent sections, and clarifies the correspondence between the leading and trailing edge radius, curvature change rate and stress concentration coefficient based on the leading and trailing edge radius and curvature distribution data. For example, the smaller the leading and trailing edge radius and the larger the curvature, the more significant the stress concentration phenomenon is, thus completing the association between the stress concentration location and the blade geometry.

[0020] S102 uses key geometric parameters of the leading and trailing edges of sensitive section airfoils as the reference dimensions for modeling and amplification, such as the minimum radius of curvature of the elliptical airfoil on the inlet side and the radius of the trailing edge arc. Combining the initial value of the stress concentration factor and aerodynamic performance constraints, the modeling coefficient range of each sensitive section is defined to ensure that the leading and trailing edges of the airfoil still maintain an aerodynamically friendly geometry after amplification. The modeling coefficient reference and range are defined, and the airfoil modeling coefficient data of each sensitive section are obtained, including the modeling amplification reference dimension of each sensitive section, the final determined modeling coefficient, the modeling amplification direction, and the estimated value of the leading and trailing edge radius after amplification.

[0021] S103. Based on the definition data of the vibration stress concentration location and the air profile modeling coefficient data of each sensitive section, locate the range of the sensitive section, implement the modeling thickening of the sensitive section and calculate the key dimensions of the thickened air profile to obtain locally optimized thickened air profile data and thickened air profile key dimension parameters. After locating the sensitive section range in S103, a set of sensitive sections with thickened templates is obtained. Based on the original blade coordinate data and the determined template coefficients and magnification directions, the blade profile of each sensitive section in the set of sensitive sections is uniformly magnified and its profile is adjusted. Specifically, for the leading edge elliptical segment and the trailing edge arc segment, the radius of curvature is magnified synchronously according to the template coefficients. For the straight or curved segments of the blade body, the corresponding distance is translated along the normal direction of the profile to ensure the geometric continuity of the profile after magnification. Extract the maximum thickness, leading edge radius, and trailing edge radius of each thickened sensitive section and compare them with the original airfoil. Record the dimensional change data, where the maximum thickness is the maximum distance along the chord of the airfoil, the leading edge radius is the radius corresponding to the minimum curvature of the enlarged elliptical airfoil, and the trailing edge radius is the enlarged arc radius. Obtain locally optimized thickened airfoil data including the complete airfoil coordinates after thickening each sensitive section, as well as key dimensional parameters of the thickened airfoil including the maximum thickness, leading edge radius, and trailing edge radius values ​​of each sensitive section, and the dimensional changes compared to the original airfoil.

[0022] S104. Based on the locally optimized thickened airfoil data and key dimensional parameters of the thickened airfoil, conduct aerodynamic performance evaluation and strength vibration analysis. If any indicator fails to meet the standard, return to S102, adjust the modeling coefficient, and repeat the operation until all indicators meet the standard. Finally, output a qualified aerodynamic performance evaluation report and strength vibration analysis report. S104 constructs a computational domain for the compressor flow channel, including the thickened airfoil, encompassing the inlet, blade passage, and exhaust. After meshing the computational domain, a structured mesh is used on the blade surface to ensure the accuracy of the boundary layer simulation, while an unstructured mesh is used in the flow channel region. Aerodynamic simulation boundary conditions are set, and a turbulence model suitable for turbomachinery is selected to solve the flow field distribution under the design conditions. The aerodynamic simulation boundary conditions include a total temperature and total pressure boundary on the inlet side, a static pressure boundary on the exhaust side, and an adiabatic no-slip boundary on the blade surface. The actual efficiency, flow rate, pressure ratio, and surge margin of the thickened airfoil are obtained and compared with the original design parameters to complete the aerodynamic performance evaluation.

[0023] After passing the aerodynamic performance evaluation, based on the thickened airfoil data, the finite element simulation model was updated. The constraints and equivalent loads of S101 were applied, and strength vibration simulation was carried out. Specifically, the static stress distribution and fatigue life of the blade under steady-state conditions were calculated, and the magnitude of static stress and material yield strength, and the magnitude of fatigue life and design requirements were verified. The static stress must be less than or equal to the material yield strength, and the fatigue life must be greater than or equal to the design requirements. The strength analysis was then completed. Solve for the natural frequency and vibration stress distribution of the thickened airfoil, verify the magnitude of the vibration stress and the maximum allowable vibration stress, and ensure that the vibration stress is less than or equal to the maximum allowable vibration stress to complete the vibration analysis; If any of the indicators in the strength analysis, vibration analysis, or aerodynamic performance evaluation fails to meet the standard, the modeling coefficient should be adjusted and the operation repeated. When all indicators of strength analysis, vibration analysis, and aerodynamic performance evaluation meet the standards, the output includes flow field distribution data, actual aerodynamic performance parameters, and a strength and vibration analysis report including static stress distribution, fatigue life calculation results, natural frequency data, vibration stress distribution, and improved stress concentration factor. The flow field distribution data includes velocity field, pressure field, and temperature field, while the actual aerodynamic performance parameters include efficiency, flow rate, pressure ratio, and surge margin.

[0024] S105, combining the compliant aerodynamic performance evaluation report, strength and vibration analysis report, and locally optimized thickened airfoil data and key dimensional parameters of the thickened airfoil, integrates and optimizes to form the final blade structure.

[0025] S105 integrates the locally optimized thickened airfoil data and key dimensional parameters of the thickened airfoil, and combines the thickened airfoil coordinates of each sensitive section with the original airfoil coordinates of the non-sensitive sections according to the original stacking rule. This ensures that the airfoil has continuity and smoothness along the airfoil height direction. The original stacking rule refers to keeping the stacking axis and stacking angle unchanged. Global geometric optimization is performed on the integrated airfoil. For the transition area between sensitive and non-sensitive sections, the profile is optimized using curve interpolation to avoid geometric abrupt changes and prevent new stress concentrations, thus obtaining optimized airfoil data. Based on the optimized blade profile data, a complete three-dimensional solid model of the blade is generated, including all structural parts such as the blade body, tenon, and transition fillet. The complete design data of the final blade is extracted, the blade vibration stress is optimized, and the final blade structure is output.

[0026] In some specific embodiments, this application also provides specific optimization methods: Please refer to the appendix. Figure 2 - Appendix Figure 10 The design process will be explained in detail below.

[0027] 1. Select the sensitive section height of the blade profile based on the original blade profile. To fully improve the vibration stress distribution at the blade root and reduce the stress concentration factor, the optimized blade profile selection is concentrated on three sections containing the vibration stress concentration area: sensitive section 1, sensitive section 2, and sensitive section 3. Among them, sensitive section 3 is located inside the tenon, and sensitive section 2 is close to the transition fillet.

[0028] II. Select the airfoil modeling coefficient for the sensitive section based on the original airfoil profile. Combining the characteristics of advanced airfoil design, the minimum radius of curvature R0 of the elliptical airfoil on the inlet side is selected as the basic dimension for airfoil modeling and enlargement for sensitive section 3, such as... Figure 2The 0.5R0~R0 shown is uniformly scaled up along the normal direction of the original blade profile. Based on the variation law of the blade profile along the blade height direction, the scaled-up size of sensitive section 2 is further selected as 0.3R0~0.5R0, uniformly scaled up along the normal direction of the original blade profile. The scaled-up size of sensitive section 1 is further selected as shown... Figure 7 The 0.1R0~0.3R0 shown are uniformly scaled up along the normal direction of the original blade profile.

[0029] Third, after completing the modeling and enlargement of the local blade shape, other cross-section original blade profile lines are superimposed according to the blade shape superposition rules to form a new optimized blade shape.

[0030] Fourth, conduct aerodynamic assessment and strength vibration analysis to determine whether the root vibration stress distribution of the new blade meets the design requirements. If not, repeat steps two through four; if the aerodynamic performance and vibration stress distribution indicators are met, determine the final blade profile and form a new blade profile coordinate system.

[0031] The following is a specific implementation example of vibration stress optimization analysis of high-pressure compressor blades according to the present invention: I. Refer to Appendix Figure 3 The stress distribution under first-order bending vibration mode was obtained using finite element simulation, and the location of stress concentration was analyzed. In this case, the stress concentration was located at the leading edge of the blade root on the inlet side. The stress concentration factor was approximately 1.5, indicating significant stress concentration. Furthermore, fatigue limit tests showed that the fatigue strength of this stage of blade was significantly lower than that of other stages of blades made of the same material.

[0032] II. Refer to Appendix Figure 4 - Appendix Figure 7 For sensitive section 3, the minimum radius of curvature R0 of the elliptical blade on the intake side is selected as the basic dimension for blade modeling and enlargement, such as... Figure 5 As shown, R0 is uniformly scaled up along the normal direction of the original blade profile. Based on the variation law of the blade profile along the blade height direction, the scaled-up dimensions of sensitive section 2 are further selected as follows: Figure 6 The 0.5R0 shown is uniformly scaled up along the normal direction of the original blade profile. The scaled-up dimensions of sensitive section 1 are then selected as follows: Figure 7 The 0.3R0 shown is uniformly scaled up along the normal direction of the original blade profile.

[0033] Third, based on the blade shape stacking rules in section two, stack other cross-sections of the original blade profiles to form a three-dimensional blade, thus creating a new optimized blade design.

[0034] IV. Conduct aerodynamic assessment and intensity vibration analysis, such as Figure 8 - Figure 9As shown, the stress concentration factor has improved, decreasing from 1.5 to 1.21. The aerodynamic performance remains essentially unchanged, meeting the overall performance matching requirements.

[0035] V. Compressor blade test pieces were re-machined according to the newly optimized structure designed by the method of this invention, and fatigue tests were conducted simultaneously for comparison. The results show that the fatigue resistance of the blade optimized by the method of this invention is significantly improved. For example... Figure 10 As shown, when the stress level is increased to more than 1.15 times that of the original test state, the fatigue life is the same as that of the low stress level, and some can be increased to 2.5 times the original fatigue life.

Claims

1. A novel method for optimizing the vibration stress of marine gas turbine compressor blades, characterized in that, The method includes: S101. Based on the basic design data of the compressor blades of the gas turbine of the target ship, a finite element simulation model is constructed, and the operating parameters of the gas turbine are introduced. After applying the equivalent load, vibration stress simulation analysis is carried out on the constructed finite element simulation model to obtain the definition data of the vibration stress concentration location. S102. Based on the obtained definition data of vibration stress concentration location, complete the association between stress concentration location and airfoil geometric features, divide the benchmark and range of the modeling coefficient, and obtain the airfoil modeling coefficient data of each sensitive section. S103. Based on the definition data of the vibration stress concentration location and the air profile modeling coefficient data of each sensitive section, locate the range of the sensitive section, implement the modeling thickening of the sensitive section and calculate the key dimensions of the thickened air profile to obtain locally optimized thickened air profile data and thickened air profile key dimension parameters. S104. Based on the locally optimized thickened airfoil data and key dimensional parameters of the thickened airfoil, conduct aerodynamic performance evaluation and strength vibration analysis. If any indicator fails to meet the standard, return to S102, adjust the modeling coefficient, and repeat the operation until all indicators meet the standard. Finally, output a qualified aerodynamic performance evaluation report and strength vibration analysis report. S105, combining the compliant aerodynamic performance evaluation report, strength and vibration analysis report, and locally optimized thickened airfoil data and key dimensional parameters of the thickened airfoil, integrates and optimizes to form the final blade structure.

2. The novel method for optimizing vibration stress of marine gas turbine compressor blades according to claim 1, characterized in that, The basic design data of the gas turbine compressor blade of the target ship in S101 are specifically: three-dimensional geometric model of the blade shape, mechanical parameters of the blade material, and blade installation boundary conditions. The core operating parameters of a gas turbine are: design speed, inlet air temperature under typical operating conditions, inlet air pressure, and working fluid flow rate. The model is meshed based on the three-dimensional geometric model of the blade. A fine mesh is used for areas of stress concentration, while a transition mesh is used for other areas of the blade. Based on the actual installation state of the blade, constraints are defined in the model, and equivalent loads including centrifugal load, aerodynamic load, and vibration excitation load are applied in combination with the operating parameters of the gas turbine.

3. The novel method for optimizing vibration stress of marine gas turbine compressor blades according to claim 1, characterized in that, After applying the equivalent load, S101 sets the simulation analysis type to transient dynamic analysis, focuses on the preset vibration mode of interest, solves the vibration response of the blade under the corresponding mode, and extracts the vibration stress distribution data of the entire blade area, generating stress distribution cloud map, stress variation curve along blade height, and stress distribution curve along cross section circumferentially. By comparing the stress values ​​of each region of the blade, it identifies the parts with stress values ​​higher than the surrounding areas and abrupt changes in stress gradient, and clarifies the specific location information of the parts. It obtains the definition data of the vibration stress concentration location, including the height coordinate interval along blade height of the stress concentration area, the specific orientation along cross section circumferentially, the stress peak value under the corresponding vibration mode of interest, and the initial value of the stress concentration coefficient, and determines the vibration stress concentration location.

4. The novel method for optimizing vibration stress of marine gas turbine compressor blades according to claim 1, characterized in that, S102 determines the vibration stress concentration location based on the definition data of the vibration stress concentration location, locates the corresponding blade sensitive section, extracts the leading and trailing edge radius and curvature distribution data of the sensitive section and adjacent sections, and clarifies the correspondence between the leading and trailing edge radius size, curvature change rate and stress concentration coefficient based on the leading and trailing edge radius and curvature distribution data, thus completing the association between the stress concentration location and the blade geometry.

5. A novel method for optimizing vibration stress of marine gas turbine compressor blades according to claim 1, wherein step S102 uses the key geometric parameters of the leading and trailing edges of the sensitive section blade as the reference size for modeling and amplification, combines the initial value of the stress concentration factor with aerodynamic performance constraints, delineates the modeling coefficient range of each sensitive section, completes the division of the modeling coefficient reference and range, and obtains the blade modeling coefficient data of each sensitive section, including the modeling amplification reference size of each sensitive section, the finally determined modeling coefficient, the modeling amplification direction, and the estimated value of the leading and trailing edge radius after amplification.

6. The novel method for optimizing vibration stress of marine gas turbine compressor blades according to claim 1, characterized in that, After S103 locates the sensitive section range, a set of sensitive sections with thickened templates is obtained. Based on the original blade shape coordinate data and the determined template coefficient and magnification direction, the blade shape of each sensitive section in the set of sensitive sections is uniformly magnified and adjusted. Specifically, for the leading edge elliptical segment and the trailing edge arc segment, the radius of curvature is magnified synchronously according to the template coefficient. For the straight or curved segments of the blade body, the corresponding distance is translated along the normal direction of the profile. Extract the maximum thickness, leading edge radius, and trailing edge radius of each thickened sensitive section and compare them with the original airfoil. Record the dimensional change data to obtain locally optimized thickened airfoil data, including the complete airfoil coordinates after thickening each sensitive section, as well as the key dimensional parameters of the thickened airfoil, including the maximum thickness, leading edge radius, and trailing edge radius values ​​of each sensitive section, and the dimensional changes compared with the original airfoil.

7. A novel method for optimizing vibration stress of marine gas turbine compressor blades according to claim 1, characterized in that, S104 constructs a computational domain for the compressor flow channel containing thickened airfoils. After meshing the computational domain, aerodynamic simulation boundary conditions are set, and a turbulence model suitable for turbomachinery is selected to solve the flow field distribution under the design conditions. The aerodynamic simulation boundary conditions include setting total temperature and total pressure boundaries on the inlet side, static pressure boundaries on the exhaust side, and adiabatic no-slip boundaries on the blade surface. The actual efficiency, flow rate, pressure ratio, and surge margin of the thickened airfoils are obtained by solving and compared with the original design indicators to complete the aerodynamic performance evaluation.

8. A novel method for optimizing vibration stress of marine gas turbine compressor blades according to claim 7, characterized in that, After the aerodynamic performance evaluation, based on the thickened airfoil data, the finite element simulation model is updated. The constraint conditions and equivalent loads of S101 are applied to carry out strength vibration simulation. Specifically, the static stress distribution and fatigue life of the blade under steady-state conditions are calculated, the magnitude of static stress and material yield strength, and the magnitude of fatigue life and design requirements are verified to complete the strength analysis. Solve for the natural frequency and vibration stress distribution of the thickened airfoil, verify the magnitude of the vibration stress and the maximum allowable vibration stress, and complete the vibration analysis; If any of the indicators in the strength analysis, vibration analysis, or aerodynamic performance evaluation fails to meet the standard, the modeling coefficient should be adjusted and the operation repeated. When all indicators of strength analysis, vibration analysis, and aerodynamic performance evaluation meet the standards, the output includes flow field distribution data, actual aerodynamic performance parameters, and a strength and vibration analysis report including static stress distribution, fatigue life calculation results, natural frequency data, vibration stress distribution, and improved stress concentration factor.

9. A novel method for optimizing vibration stress of marine gas turbine compressor blades according to claim 1, characterized in that, S105 integrates the locally optimized thickened airfoil data and the key dimension parameters of the thickened airfoil, and integrates the thickened airfoil coordinates of each sensitive section with the original airfoil coordinates of the non-sensitive section according to the original superposition rule. Global geometric optimization is performed on the integrated airfoil. For the transition area between the sensitive section and the non-sensitive section, the profile is optimized by curve interpolation method to obtain the optimized airfoil data. Based on the optimized blade profile data, a complete three-dimensional solid model of the blade is generated, and the complete design data of the final blade is extracted to optimize the blade vibration stress and output the final blade structure.

Images