Rapid design method and system for turbine blade, electronic equipment and storage medium

Abstract

The invention discloses a rapid design method and system for a turbine blade, electronic equipment and a storage medium, and the rapid design method comprises the steps: firstly, obtaining boundary conditions of a gas side and a cold air side of the turbine blade in an actual working environment based on flow heat exchange test data, and then constructing a heat exchange boundary model based on the boundary conditions; according to the method, temperature field calculation becomes a pure solid heat conduction problem, and only depends on known boundary conditions and material attributes, so that a heat exchange boundary model can be directly applied to a blade three-dimensional model for transient temperature field calculation, and CFD simulation calculation needing iterative solution is replaced; by means of the method, the most time-consuming flow field solving and flow thermal coupling iteration link in the whole analysis process is thoroughly eliminated, the calculation efficiency is remarkably improved while the calculation precision is guaranteed, a thermal-force coupling chain is cut off through decoupling calculation of the transient temperature field and the transient stress field, and the calculation efficiency is further improved.

Description

Technical Field

[0001] This invention relates to the field of turbine blade technology, and in particular to a rapid design method and system for turbine blades, electronic devices, and computer-readable storage media. Background Technology

[0002] Existing turbine blade design methods are based on flow-thermal coupling turbine blade temperature field calculation methods. The core technical approach can be summarized in four stages: First, a three-dimensional model of the turbine blade, including complex features such as cooling channels and film cooling holes, is constructed, and a high-precision structured mesh is used to discretize the fluid and solid domains. Second, the flow characteristics of the blade surface and internal cooling channels are solved through three-dimensional CFD simulation to obtain the blade's temperature field distribution. Third, the blade temperature field calculation results are transferred to the finite element analysis (FEM) module to calculate the stress distribution of the turbine blade. Finally, the blade strength is checked based on the stress analysis results, and feedback is given to the design end to adjust cooling and structural parameters. However, due to the enormous time consumption of 3D CFD modeling and solving, a single simulation requires more than ten standard steps (including geometric simplification, multi-block mesh generation, turbulence model selection, etc.). Especially when dealing with local fine structures such as trailing edge turbulence pillars and leading edge film air holes, the mesh size often reaches tens of millions to ensure the accuracy of flow separation and vortex capture. The calculation time for a single working condition exceeds 12 hours. This contradiction between high-fidelity simulation and inefficient iterative optimization makes it difficult for traditional design methods to meet the rapid iterative design requirements of modern aero-engines. Especially in scenarios with high thermodynamic cycle parameters (such as inlet temperature > 1800K), small adjustments to the cooling structure may cause nonlinear effects such as flow separation and hot spot shift, making it difficult for the turbine blade optimization design process to simultaneously meet the requirements of high computational accuracy and short design cycle. In addition, the fluid-thermal coupled iterative process is redundant. Each round of design improvement requires repeated execution of the complete CFD→FEM data transfer and convergence determination. Furthermore, due to the difference in mesh topology between the fluid domain and the solid domain, the data mapping process needs to be implemented through interpolation algorithms, introducing additional computational errors. Summary of the Invention

[0003] This invention provides a rapid design method and system for turbine blades, electronic devices, and a computer-readable storage medium, which can significantly improve computational efficiency while ensuring computational accuracy.

[0004] According to one aspect of the present invention, a rapid design method for turbine blades is provided, comprising the following: The boundary conditions of the turbine blades on the gas and cold air sides under actual working conditions were obtained based on the flow heat transfer test data. A heat transfer boundary model is constructed based on the boundary conditions of the gas side and the cold gas side; Input a 3D model of the turbine blade, and perform decoupled calculations of the transient temperature field and transient stress field based on the 3D model and the heat transfer boundary model; Based on the decoupling calculation results, local design iterative optimization is performed, and the optimized 3D model is output.

[0005] Furthermore, the process of obtaining the boundary conditions of the turbine blades on the cold gas side under actual operating conditions includes the following: Flow tests were conducted on the cooling channel of the turbine blades. The velocity distribution of the cooling airflow was measured using PIV technology. The convective heat transfer coefficient of the inner wall of the cooling channel was calculated based on the velocity distribution measurement results and Reynolds analogy.

[0006] Furthermore, for the rim plate region, the convective heat transfer coefficient of the inner wall of the cooling channel is corrected based on the following formula: ; ; in, This represents the corrected convective heat transfer coefficient. This represents the correction factor. This represents the convective heat transfer coefficient before correction. Represents the proportionality coefficient. This represents the reference vorticity intensity in the region without secondary flow. This indicates the vortex intensity in the rim plate region.

[0007] Furthermore, when constructing the heat transfer boundary model, a piecewise modeling strategy is adopted to construct the Nusselt number correlation model of the leading edge region, which is expressed as: ; ; in, Nusselt number representing the stationed area and The constant to be fitted is obtained through inversion calibration based on experimentally measured stagnation heat transfer data. Indicated based on leading edge diameter The Reynolds number, Representing Prandtl numbers, The angle between the point of residence and the leading edge is . The Nusselt number at the position, Indicates the attenuation coefficient. The two decay parameters, representing the decay index, are obtained by fitting and optimizing the leading-edge axial temperature / heat flux distribution data obtained experimentally.

[0008] Furthermore, when constructing the heat transfer boundary model, the Nusselt number correlation model of the trailing edge region is defined as follows: ; ; in, This represents the corrected Nusselt number for the trailing edge region. Represents the fundamental value of the Nusselt number. f Indicates the geometric correction factor. A Represents the geometric correction factor. This indicates the longitudinal spacing of the spoiler columns in the trailing edge region. This indicates the lateral spacing of the spoiler columns in the trailing edge region. D This indicates the diameter of the turbulence column.

[0009] Furthermore, when constructing the heat transfer boundary model, the recovery temperature model is defined as: ; in, This represents the recovery temperature of the radial position x at time t. This indicates the time-varying air conditioning reference temperature. Indicates the radial time-varying gas temperature distribution. Indicates Ye Gao, The dimensionless radial distribution function is calibrated using high-temperature gas wind tunnel test data at the steady-state design point.

[0010] Furthermore, the process of decoupling the transient temperature field and transient stress field calculation based on the three-dimensional model and the heat transfer boundary model includes the following: The three-dimensional model is meshed, and the heat transfer boundary model is applied to the three-dimensional model to calculate the transient temperature field. Then, a stress field finite element model is constructed, and the mesh node distribution of the stress field finite element model is completely consistent with the temperature field. The transient temperature field is applied to the stress field finite element model through a one-to-one mapping of mesh nodes to calculate the transient stress field.

[0011] In addition, the present invention also provides a rapid design system for turbine blades, comprising: The boundary condition acquisition module is used to acquire the boundary conditions of the turbine blades on the gas side and the cold gas side under actual working conditions based on the flow heat transfer test data. The heat transfer boundary model construction module is used to construct a heat transfer boundary model based on the boundary conditions of the gas side and the cold gas side. The transient temperature-stress decoupling calculation module is used to input a three-dimensional model of the turbine blade and perform decoupling calculations of the transient temperature field and transient stress field based on the three-dimensional model and the heat transfer boundary model. The local design optimization module is used to perform iterative optimization of local design based on the decoupling calculation results and output the optimized 3D model.

[0012] In addition, the present invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the method described above by calling the computer program stored in the memory.

[0013] In addition, the present invention provides a computer-readable storage medium for storing a computer program for rapid design of turbine blades, wherein the computer program performs the steps of the method described above when run on a computer.

[0014] The present invention has the following beneficial effects: The rapid design method for turbine blades of this invention first obtains the boundary conditions of the turbine blade on the gas and cooling sides under actual working conditions based on flow heat transfer test data. Then, a heat transfer boundary model is constructed based on this, making the temperature field calculation a pure solid heat conduction problem that only depends on known boundary conditions and material properties. Thus, the heat transfer boundary model can be directly applied to the three-dimensional model of the blade for transient temperature field calculation, replacing the CFD simulation calculation that requires iterative solution. This completely eliminates the most time-consuming flow field solution and flow-heat coupling iteration links in the entire analysis process, significantly improving computational efficiency while ensuring computational accuracy. Furthermore, by decoupling the calculation of transient temperature field and transient stress field, the thermo-mechanical coupling chain is cut off, further improving computational efficiency.

[0015] In addition, the rapid design system for turbine blades of the present invention also has the above-mentioned advantages.

[0016] In addition to the objectives, features, and advantages described above, the present invention has other objectives, features, and advantages. The invention will now be described in further detail with reference to the figures. Attached Figure Description

[0017] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a flowchart illustrating a rapid design method for turbine blades according to a preferred embodiment of this application. Figure 2 This is a schematic diagram of the module structure of a rapid design system for turbine blades according to another embodiment of this application. Detailed Implementation

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] Reference Figure 1A preferred embodiment of this application provides a rapid design method for turbine blades, comprising the following: Step S1: Based on the flow heat transfer test data, obtain the boundary conditions of the turbine blades on the gas side and the cold gas side under actual working conditions; Step S2: Construct a heat transfer boundary model based on the boundary conditions of the gas side and the cold gas side; Step S3: Input the three-dimensional model of the turbine blade, and perform decoupled calculations of the transient temperature field and transient stress field based on the three-dimensional model and the heat transfer boundary model; Step S4: Perform local design iterative optimization based on the decoupling calculation results, and output the optimized 3D model.

[0020] It is understood that the rapid design method for turbine blades in this embodiment first obtains the boundary conditions of the turbine blade on the gas side and the cooling side under actual working conditions based on flow heat transfer test data. Then, it constructs a heat transfer boundary model based on this, making the temperature field calculation a pure solid heat conduction problem that only depends on known boundary conditions and material properties. Thus, the heat transfer boundary model can be directly applied to the three-dimensional model of the blade for transient temperature field calculation, replacing the CFD simulation calculation that requires iterative solution. This completely eliminates the most time-consuming flow field solution and flow-heat coupling iteration links in the entire analysis process, significantly improving calculation efficiency while ensuring calculation accuracy. Furthermore, by decoupling the calculation of transient temperature field and transient stress field, the thermo-mechanical coupling chain is cut off, further improving calculation efficiency.

[0021] In step S1, regarding the boundary conditions on the gas side, the actual working environment of the turbine blade is simulated in a high-temperature gas wind tunnel test rig. For example, test parameters such as the total gas inlet temperature (1600-2000K), total pressure (2.5-3.5MPa), and flow velocity (80-150m / s) are set. The temperature distribution and heat flux density on the blade surface (especially key areas such as the leading and trailing edges) are measured in real time using a high-precision infrared thermal imager and a heat flux sensor array. For high-temperature gradient regions such as the leading and trailing edges, the measured wall temperature and heat flux density are substituted into the heat conduction equation to invert the local convective heat transfer coefficient and recovery temperature. The accuracy of the test data is then verified by comparing the results with CFD simulations under the same operating conditions. For areas where the error exceeds a preset value (e.g., 5%), recalibration or supplementary testing is required. The inversion process is as follows: According to Fourier's law of heat conduction, under steady-state conditions, the heat flux density of the blade wall... q With convective heat transfer coefficient h and recovery temperature T aw Satisfy the equation: , This indicates the wall temperature measured in the experiment, which was simultaneously measured during the experiment. q and Local values ​​can be obtained through inversion. h and T aw However, since there are two unknowns in the equation ( h and T aw Additional conditions are needed, and the mainstream gas temperature is known. T g Under conditions of flow rate and recovery factor r Calculate recovery temperature and recovery factor. r Related to the flow state, such as laminar flow r≈Pr 1 / 2 Turbulent flow r≈Pr 1 / 3 Pr represents the Prandtl number, which is calculated using the following formula: , Indicates the gas flow rate. The specific heat capacity is expressed as the equation, which can then be substituted into the equation to find the solution. However, considering that the leading and trailing edges are regions with high temperature gradients, experimental measurement data may be distorted. Therefore, this invention uses high-precision CFD simulation results to calibrate the inversion results. For example, the convective heat transfer coefficient obtained from the inversion based on experimental data is used. h and recovery temperature T aw The results obtained from CFD simulation under the same working conditions h and T aw To compare, if the two convective heat transfer coefficients h The error between and the two recovery temperatures T aw If the error between all values ​​is less than 5%, the test measurement data is considered accurate; otherwise, the test measurement data is considered distorted. For distorted areas, supplementary tests or recalibration are required.

[0022] It is understood that traditional inverse heat conduction methods are usually based on the inversion of boundary conditions from measurement data. However, this invention introduces a verification closed loop and error control with CFD simulation to ensure that the accuracy error in the high temperature gradient region is ≤5%, thereby improving the accuracy and reliability of the experimental measurement data.

[0023] Furthermore, for the main body region of the blade (regions with non-complex geometric features), the average convective heat transfer coefficient can be quickly estimated using empirical formulas and one-dimensional network models (referred to as the one-dimensional network method). The one-dimensional network model can calculate the average heat transfer coefficient based on gas flow rate, blade geometric parameters (such as chord length and thickness), and cooling channel layout through empirical correlations. The one-dimensional network method is an existing technology, and its core formula is based on Reynolds analogy; for example, the Dittus-Boelter formula can be used to calculate turbulent heat transfer. ,in, Nu Represents the Nusel number, Nu=h⋅D h / k , D h Indicates the hydraulic diameter. k Indicates thermal conductivity. Re Represents the Reynolds number. Re=ρUD h / μ , ρ For density, U For flow rate, μ For viscosity, Pr represents Prandtl number, and then based on blade geometry parameters (such as chord length)... L ,thickness t (etc.) and cooling channel layout, calculate the regional average convective heat transfer coefficient. h It is understandable that the one-dimensional network method has the advantage of fast calculation speed, and the temperature gradient in the main body of the blade is small, so the convective heat transfer coefficient of the main body of the blade can be calculated quickly and accurately.

[0024] Optionally, this invention considers that the curvature change is large near the leading edge of the blade, resulting in a significant convective impact effect on the combustion gas side. The middle region of the blade resembles flat plate flow, but also exhibits some curvature change, albeit slowly. Furthermore, the one-dimensional network model neglects secondary flow and three-dimensional effects, leading to insufficient computational accuracy. Therefore, this invention introduces a three-dimensional correction factor to further correct the convective heat transfer coefficient of the main blade region. The expression is as follows: , h 3D This represents the three-dimensional corrected convective heat transfer coefficient. h This represents the convective heat transfer coefficient calculated based on the one-dimensional network method. β This represents a three-dimensional correction factor, the value of which is related to the radial position and local curvature of the blade; the closer to the leading edge region, the better. β The larger the value, for example, the closer to the leading edge region. β The value is 1.5, and in the middle region of the leaf. β The value is 1.2.

[0025] It is understood that this invention first uses the existing one-dimensional network method to quickly solve the average convective heat transfer coefficient of the main body region of the blade, and then introduces a three-dimensional correction factor to correct it, thereby reducing the impact of secondary flow and three-dimensional effects on the calculation accuracy and achieving a balance between calculation efficiency and calculation accuracy.

[0026] In addition, the process of obtaining the boundary conditions of the turbine blades on the cold gas side under actual operating conditions includes the following: Flow tests were conducted on the cooling channel of the turbine blades. The velocity distribution of the cooling airflow was measured using PIV technology. The convective heat transfer coefficient of the inner wall of the cooling channel was calculated based on the velocity distribution measurement results and Reynolds analogy.

[0027] Specifically, the flow characteristics of cooling airflow (80-200 m / s) within the blade cavity are simulated in a low-speed wind tunnel to conduct cooling channel flow tests and measure the convective heat transfer coefficient of the inner wall of the cooling channel. The traditional method involves using a miniature heat flux sensor (such as a thin-film heat flux meter) to measure the wall heat flux density during the test. q At the same time, the wall temperature was measured using thermocouples. T w and mainstream cooling airflow temperature T c Then, the convective heat transfer coefficient formula is used for conversion. h , This invention employs PIV (Pivot-Induced Velocity) technology to measure the velocity distribution U(y) of the cooling airflow during the experiment, where y is the distance from the wall, thereby obtaining the velocity gradient within the boundary layer. Local shear stress is then calculated based on the PIV data. Then, by relating the convective heat transfer coefficient through the Reynolds analogy (for turbulence), it can be expressed as: , St Represents Stanton numbers, C f Indicates the coefficient of friction. , ρ For density, U For flow rate, μ Viscosity, C p Specific heat capacity.

[0028] It is understandable that PIV technology can provide flow details (such as vortex structures) to improve the convective heat transfer coefficient. h The calculation is closer to the actual flow, and compared with the traditional method, it can improve the conversion accuracy of the convective heat transfer coefficient.

[0029] Furthermore, to address the complex secondary flows in the upper and lower edge plate regions, this invention also constructs a heat transfer coefficient correction model through vorticity intensity analysis. Specifically, the vorticity is first calculated from PIV data. , U =( u , v , w ), representing the three-dimensional velocity vector at any point in the flow field. ω The vorticity vector represents the curl of the velocity field, characterizing the rotational intensity of a fluid element. ∇ × denotes the curl operator and defines the vorticity intensity. By performing regression analysis on the flow test data of the cooling channel, a heat transfer coefficient enhancement factor (i.e., a correction factor) was established. Relationship with vorticity intensity: , This represents the proportionality coefficient, with a value range of 0.1 to 0.3. This represents the reference vorticity intensity in the region without secondary flow. The vortex intensity in the rim region is represented by the following formula, which is then used to correct the convective heat transfer coefficient of the inner wall of the cooling channel: ; in, This represents the corrected convective heat transfer coefficient. This represents the convective heat transfer coefficient before correction, obtained through the Reynolds analogy above.

[0030] It is understood that this invention constructs a heat transfer coefficient enhancement factor correction model through vorticity intensity analysis, quantifies the enhancement effect of secondary flow on heat transfer in the rim plate region, improves the conversion accuracy of convective heat transfer coefficient in the rim plate region, and thus improves the temperature prediction accuracy in the rim plate region.

[0031] It is understood that in step S1, high-precision boundary condition data (including convective heat transfer coefficient and recovery temperature) for the gas and cold air sides have been obtained based on the flow heat transfer test data. However, these high-precision boundary condition data are discrete and distributed in points / regions. In step S2, the goal is to construct a high-precision three-dimensional transient heat transfer boundary model that covers the entire blade surface, is spatiotemporally continuous, and is based on these discrete high-precision boundary condition data.

[0032] In constructing the heat transfer boundary model, a segmented modeling strategy is used to build the Nusselt number correlation model of the leading edge region, which is expressed as: ; ; in, Nusselt number representing the stationed area and The constants to be fitted are obtained through inversion calibration based on experimentally measured stagnation heat transfer data. For example, a certain experimental condition might calibrate the constants to be fitted. C =0.95、 m =0.5, Indicated based on leading edge diameter The Reynolds number, Representing Prandtl numbers, The angle between the point of residence and the leading edge is . The Nusselt number at the position, Indicates the attenuation coefficient. The two decay parameters, representing the decay index, are obtained by fitting and optimizing the experimentally measured leading-edge axial temperature / heat flux distribution data, thereby accurately capturing the decay law of the convective heat transfer coefficient along the leading edge.

[0033] It is understandable that the leading edge of a turbine blade is a stagnation region with highly complex flow and heat transfer characteristics. Existing technologies typically use classical empirical formulas (such as the Dittus-Boelter formula) to uniformly model the leading edge region. However, in the stagnation region (near θ=0°), this area is a flow stagnation point with the strongest heat transfer. Downstream of the stagnation region (|θ|>0°), i.e., from the leading edge stagnation point outwards to both sides (pressure and suction surfaces), a boundary layer begins to form. The stagnation region and the downstream region of the stagnation region have different heat transfer intensities. Therefore, this invention uses a stagnation point heat transfer formula based on boundary layer theory to construct a Nusselt number correlation model for the stagnation region, and a wedge flow model suitable for curved surfaces to construct a Nusselt number correlation model for the downstream region. This achieves segmented modeling of the leading edge region. The segmented model better reflects the physical nature of the leading edge flow, i.e., the stagnation developing towards the boundary layer. Compared with the single modeling using existing classical empirical formulas, this improves the accuracy of the heat transfer boundary model in the leading edge region. Furthermore, the constants to be fitted in the formula... , and attenuation parameters , All models were calibrated using experimental data, effectively overcoming the errors caused by the simplification of complex sets and flows in purely theoretical models, and further improving the model accuracy. Furthermore, after constructing the Nusselt number correlation model for the leading edge region, the heat transfer boundary model for the leading edge region can be obtained based on the theoretical conversion relationship between the Nusselt number and the convective heat transfer coefficient. The specific process is existing technology and will not be elaborated here.

[0034] Furthermore, this invention defines the Nusselt number correlation model for the trailing edge region as follows: ; ; in, This represents the corrected Nusselt number for the trailing edge region. The fundamental value of the Nusselt number is calculated based on the Dittus-Boelter formula. f Indicates the geometric correction factor. A Represents the geometric correction factor. This indicates the longitudinal spacing of the spoiler columns in the trailing edge region. This indicates the lateral spacing of the spoiler columns in the trailing edge region. D This indicates the diameter of the turbulence column.

[0035] It is understandable that the trailing edge region is typically filled with turbulent columns, resulting in significant flow separation and vortex shedding. These local geometric features also affect the convective heat transfer process. Therefore, this invention introduces a geometric correction factor based on classical empirical formulas, directly embedding the specific geometric features of the trailing edge into the heat transfer boundary model. This quantifies the enhancing effect of turbulent column density and arrangement on heat transfer, reflecting the intensifying effect of complex flows such as turbulent column wake vortices on heat transfer, thus improving the accuracy of the heat transfer boundary model in the trailing edge region. Furthermore, when trailing edge design parameters (such as turbulent column spacing) change, the model can be quickly updated simply by updating the geometric parameters, without needing to perform complex CFD calculations again, greatly improving update efficiency. Similarly, after constructing the Nusselt number correlation model of the leading edge region, the heat transfer boundary model of the leading edge region can be obtained based on the theoretical conversion relationship between the Nusselt number and the convective heat transfer coefficient. The specific process is existing technology and will not be elaborated here.

[0036] Furthermore, for the main body region of the blade, since the heat transfer changes in this region are relatively smooth, the radial basis function (RBF) interpolation algorithm can be used to interpolate the convective heat transfer coefficient data corrected by the three-dimensional correction factor in step S1, generating a continuous and smooth three-dimensional convective heat transfer coefficient distribution field, which can be expressed as: , Indicates the coordinates of the point to be interpolated. This represents the on-line heat transfer coefficient at the point to be interpolated. This represents known data points (derived from the calculation points of the one-dimensional network method in step S1). Represents radial basis functions. This represents the weight of the i-th known data point. N This represents the number of known data points. By solving a system of linear equations, the interpolation surface is made to accurately pass through all known data points. RBF interpolation ensures a smooth transition in the heat transfer coefficient distribution across the entire blade surface, avoiding discontinuous jumps at region junctions. It naturally and seamlessly integrates the high-precision leading / tailing edge local model with the main blade region model into a complete three-dimensional boundary condition.

[0037] Furthermore, when constructing the heat transfer boundary model, this invention also constructs a temperature difference function based on the instantaneous change law of the total gas temperature and the cooling gas flow temperature, and defines the recovery temperature model as follows: ; in, This represents the recovery temperature at time t at the radial position x (i.e., the distance from the leaf root). This indicates the time-varying reference temperature of the air conditioner, which changes over time. The radial time-varying combustion gas temperature distribution is represented by the engine transient performance model (e.g., starting, stopping, acceleration / deceleration) and is a function of radial position and time. Indicates Ye Gao, The radial distribution function is a dimensionless function with values ​​between 0 and 1. It is calibrated using high-temperature gas wind tunnel test data at steady-state design points. For example, at the blade root (x / L=0), Φ≈0.2; at the blade tip (x / L=1), Φ≈0.8; intermediate values ​​are determined by spline interpolation.

[0038] It is understandable that traditional methods typically simplify the recovery temperature to a steady-state value or only vary with time, ignoring the differences in combustion gas temperature (determined by the combustion chamber outlet temperature field distribution) and cooling airflow temperature distribution at different radial positions of the blades (such as the blade root, blade middle, and blade tip) when engine operating conditions change, leading to distortion in transient temperature field calculations. This invention aims to simulate the recovery temperature... T aw In the spatiotemporal distribution of the transient process, the recovery temperature model is defined as a physics-based bivariate interpolation function, which incorporates the dynamics of time (from...). T g ( t )and T c ( t The model decouples and recouples the description of the temperature field from the spatial non-uniformity (described by the calibrated Φ(x / L)), so that under the transient operating conditions of the engine, the model can not only reflect the changes in the overall temperature level, but also accurately maintain the inherent relative temperature at different heights of the blade. This allows for a more realistic simulation of physical phenomena such as the movement of "hot spots in the blade" during acceleration, significantly improving the physical accuracy of transient temperature field prediction.

[0039] Furthermore, traditional empirical / semi-empirical models, once established, remain fixed and cannot adapt to design changes or new experimental data, lacking self-learning capabilities. This invention, however, designs a confidence-weighted incremental update algorithm to update the heat transfer boundary model. The system periodically (e.g., after each design iteration) updates the wall temperature predicted based on the heat transfer boundary model. T w,pred Wall temperature measured by new high-precision verification tests (or a small number of calibration CFD calculations) T w,meas For comparison, regions where the error exceeds a threshold (e.g., 5%) are denoted as regions. i This triggers the update process. The update strategy is not a simple replacement, but a weighted fusion, which can be represented as: ,in, h i,old This represents the original value of region i. h i,corr Indicates the region i The new corrected values ​​were obtained by inversion calculation based on the new experimental data. ω The weighting coefficients depend on the original values.h i,old The confidence level is related to the data source and the number of times it is used. For example, the confidence level of experimental data is higher than that of one-dimensional network data, and the more times it is used, the higher the confidence level.

[0040] It is understood that, through the above-mentioned incremental update mechanism, the heat transfer boundary model can continuously self-optimize as the design process deepens and more data accumulates, thereby iteratively improving the prediction accuracy and avoiding the static error of the traditional method of "modeling once and always approximating". This enhances the robustness and adaptability of the design method.

[0041] In addition, in step S3, the pre-designed 3D model of the turbine blade and the boundary heat transfer model are input into the ANSYS analysis software, and a dynamic time step control strategy is set. For example, the initial time step is set to 0.1 seconds, which is suitable for cruise conditions with gradual temperature changes. When the temperature change rate is detected to exceed 50K / s (such as during engine emergency stop), the step is automatically shortened to 0.01 seconds to avoid numerical oscillations caused by transient thermal shock. In the Transient Thermal module of ANSYS, the convective heat transfer coefficient and recovery temperature are loaded in batches from the boundary heat transfer model through APDL scripts. Then, the transient temperature field and transient stress field are decoupled and calculated, and a transient stress distribution cloud map is output, marking the location of the maximum stress point (such as the junction of the tenon and the blade). Through time history analysis, the stress peak value at key time points (such as during emergency stop) is extracted. Finally, the dynamic change animation of the temperature-stress field is generated using ANSYS post-processing tools to intuitively display the transient response characteristics of the dangerous area.

[0042] The process of decoupling the transient temperature field and transient stress field based on the three-dimensional model and the heat transfer boundary model includes the following: The three-dimensional model is meshed, and the heat transfer boundary model is applied to the three-dimensional model to calculate the transient temperature field. Then, a stress field finite element model is constructed, and the mesh node distribution of the stress field finite element model is completely consistent with the temperature field. The transient temperature field is applied to the stress field finite element model through a one-to-one mapping of mesh nodes to calculate the transient stress field.

[0043] Specifically, the 3D model is first meshed, and the high-precision transient heat transfer boundary model constructed in step S2 is applied to the 3D model. The corresponding boundary conditions are then assigned to each surface of the 3D model. The Transient Thermal module then solves the following unsteady heat conduction equation: Among them, the material's thermal conductivity k and specific heat capacity C pAll parameters are functions of temperature T, taking into account the nonlinearity of material properties to ensure high fidelity under high thermal loads and accurately capture material behavior. This data comes from a material library. Solving this model yields the transient temperature distribution inside the blade, i.e., the transient temperature field. After solving the transient temperature field, the transient stress field calculation begins. ANSYS's Static Structural module constructs a finite element model of the stress field based on the 3D model, and the mesh node distribution of the stress field finite element model is completely consistent with the temperature field. Within ANSYS, the Transient Thermal module directly transfers the calculated nodal temperature distribution for each time step through memory, accurately mapping it to the corresponding nodes in the Static Structural module using a one-to-one mapping method between mesh nodes, without any interpolation calculations. The Static Structural module solves the thermoelastic equation at each time step based on the mapped transient temperature field to obtain the transient stress distribution results. The thermoelastic equation uses the Green-Naghdi theory, which considers temperature dependence, and the constitutive relation is: Where σ represents the stress tensor and C(T) represents the temperature-dependent elastic matrix. Let α(T) represent the total strain, α(T) represent the temperature-dependent coefficient of thermal expansion, ΔT represent the temperature difference relative to the reference temperature, and I represent the unit tensor.

[0044] It is understandable that since a high-precision transient heat transfer boundary model has been constructed in step S2, the temperature field calculation becomes a pure solid heat conduction problem, relying only on known boundary conditions and material properties. Therefore, the heat transfer boundary model can be directly applied to the three-dimensional model of the blade for transient temperature field calculation, replacing the CFD simulation calculation that requires iterative solution. This completely eliminates the most time-consuming flow field solution and flow-heat coupling iteration in the entire analysis process. Furthermore, the direct node-to-node mapping between the temperature field and stress field eliminates the interpolation error caused by mesh mismatch in traditional coupling, ensuring the accuracy of the data transfer from the temperature field to the stress field.

[0045] In addition, in step S4, the stress peak region is obtained based on the decoupling calculation of transient temperature-stress field. If the stress value exceeds the allowable value of the material, the local stress is reduced by adjusting local parameters such as the local cooling structure or wall thickness gradient. After optimizing the local parameters, the transient temperature-stress field iterative calculation is performed again until the maximum transient stress value of the blade is less than the allowable value of the material. Finally, the optimized three-dimensional model, temperature field results, and stress field results are output. For example, ANSYS's post-processing module automatically identifies the peak stress region and its occurrence time in the transient stress cloud map, and calculates the sensitivity of key design variables (such as local wall thickness, air film orifice diameter, and turbulence column spacing) to the stress peak, specifically through the existing parameter perturbation method for approximate calculation. Then, based on the sensitivity analysis results, the system will prioritize adjusting the parameters most effective in reducing stress (i.e., the parameters with the largest absolute value of sensitivity). For example, if the stress concentration at the junction of the tenon and the blade is the main problem, and the sensitivity to the junction fillet radius R is the highest, the optimization algorithm will automatically increase the value of R. After parameter adjustment, no The flow field analysis needs to be redone. Simply re-execute the decoupling calculation in step S3 based on the updated geometric model, because the heat transfer boundary model can be automatically updated according to the new geometric parameters. For example, when the geometric parameters change, the velocity distribution, Reynolds number, gas temperature distribution, and cold air temperature distribution will all change accordingly, and the leading edge model, trailing edge model, blade mid-section model, and recovery temperature model will be automatically updated. Then, the newly calculated stress peak value is compared with the required stress of the material. If they do not meet the requirements, the parameters are adjusted again based on the new sensitivity analysis, forming an automated closed loop of "analysis-identification-optimization-verification" until convergence.

[0046] It is understood that this invention achieves automatic closed-loop optimization design based on sensitivity analysis, freeing engineers from tedious trial and error, ensuring the effectiveness of design improvements by quantitatively guiding the optimization direction, and significantly reducing the number of iterations, thereby further improving design efficiency.

[0047] In addition, such as Figure 2 As shown, another embodiment of the present invention also provides a rapid design system for turbine blades, preferably employing the rapid design method for turbine blades as described above, comprising: The boundary condition acquisition module is used to acquire the boundary conditions of the turbine blades on the gas side and the cold gas side under actual working conditions based on the flow heat transfer test data. The heat transfer boundary model construction module is used to construct a heat transfer boundary model based on the boundary conditions of the gas side and the cold gas side. The transient temperature-stress decoupling calculation module is used to input a three-dimensional model of the turbine blade and perform decoupling calculations of the transient temperature field and transient stress field based on the three-dimensional model and the heat transfer boundary model. The local design optimization module is used to perform iterative optimization of local design based on the decoupling calculation results and output the optimized 3D model.

[0048] It is understood that the rapid design system for turbine blades in this embodiment first obtains the boundary conditions of the turbine blade on the gas and cooling sides under actual working conditions based on flow heat transfer test data. Then, it constructs a heat transfer boundary model based on this, making the temperature field calculation a pure solid heat conduction problem that only depends on known boundary conditions and material properties. Thus, the heat transfer boundary model can be directly applied to the three-dimensional model of the blade for transient temperature field calculation, replacing the CFD simulation calculation that requires iterative solution. This completely eliminates the most time-consuming flow field solution and flow-heat coupling iteration links in the entire analysis process, significantly improving computational efficiency while ensuring computational accuracy. Furthermore, by decoupling the calculation of transient temperature field and transient stress field, the thermo-mechanical coupling chain is cut off, further improving computational efficiency.

[0049] In addition, another embodiment of the present invention provides an electronic device including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the method described above by calling the computer program stored in the memory.

[0050] In addition, another embodiment of the present invention provides a computer-readable storage medium for storing a computer program for rapid design of turbine blades, the computer program executing the steps of the method described above when run on a computer.

[0051] Common computer-readable storage media include: floppy disks, flexible disks, hard disks, magnetic tapes, any other magnetic media, CD-ROMs, any other optical media, punch cards, paper tape, any other physical media with perforated patterns, random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), flash erasable programmable read-only memory (FLASH-EPROM), any other memory chips or cartridges, or any other media readable by a computer. Instructions may further be transmitted or received by a transmission medium. The term transmission medium can include any tangible or intangible medium used to store, encode, or carry instructions for execution by a machine, and includes digital or analog carrier communication signals or intangible media that facilitate communication of such instructions. Transmission media include coaxial cables, copper wires, and optical fibers, which contain conductors for transmitting a bus of computer data signals.

[0052] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0053] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0054] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0055] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0056] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0057] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

[0058] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A rapid design method of a turbine blade, characterized by, The method comprises the following steps: The boundary conditions of the gas side and the cold gas side of the turbine blade under the actual working environment are obtained based on the flow heat exchange test data; A heat exchange boundary model is constructed based on the boundary conditions of the gas side and the cold gas side; A three-dimensional model of the turbine blade is input, and decoupling calculation of the transient temperature field and the transient stress field is performed based on the three-dimensional model and the heat exchange boundary model; Local design iteration optimization is performed based on the decoupling calculation results, and an optimized three-dimensional model is output.

2. The rapid design method of a turbine blade according to claim 1, wherein The process of obtaining the boundary conditions of the cold gas side of the turbine blade under the actual working environment comprises the following steps: A cooling channel flow test of the turbine blade is performed, the PIV technology is used to measure the flow velocity distribution of the cooling gas flow, and the convective heat transfer coefficient of the inner wall surface of the cooling channel is obtained based on the flow velocity distribution measurement results and Reynolds analogy calculation.

3. The rapid design method of a turbine blade according to claim 2, wherein For the rim region, the convective heat transfer coefficient of the inner wall surface of the cooling channel is corrected based on the following formula: ； ； wherein, represents the modified convective heat transfer coefficient, represents the correction coefficient, represents the uncorrected convective heat transfer coefficient, represents the proportionality coefficient, represents the reference vorticity intensity of the no secondary flow region, represents the vorticity intensity of the edge plate region.

4. The rapid design method of a turbine blade according to claim 1, wherein In constructing the heat exchange boundary model, the Nusselt number correlation model of the leading edge region is constructed by using a segmented modeling strategy, which is represented as: ； ； in, Nusselt number representing the stationed area and The constant to be fitted is obtained through inversion calibration based on experimentally measured stagnation heat transfer data. Indicated based on leading edge diameter The Reynolds number, Representing Prandtl numbers, The angle between the point of residence and the leading edge is . The Nusselt number at the position, Indicates the attenuation coefficient. The two decay parameters, representing the decay index, are obtained by fitting and optimizing the leading-edge axial temperature / heat flux distribution data obtained experimentally.

5. The rapid design method of a turbine blade according to claim 1, wherein In constructing the heat exchange boundary model, the Nusselt number correlation model of the trailing edge region is defined as: ； ； wherein Nu represents the Nusselt number corrected for the trailing edge region, Nu0 represents a base value for the Nusselt number, f G represents a geometrical correction factor, A G represents a geometrical correction factor, L represents the longitudinal arrangement pitch of the turbulence columns in the trailing edge region, L represents the longitudinal arrangement pitch of the turbulence columns in the trailing edge region, D D represents the diameter of the turbulence columns.

6. The rapid design method of a turbine blade according to claim 1, wherein In constructing the heat exchange boundary model, the recovery temperature model is defined as: ； wherein, represents the recovery temperature at radial position x at time t, represents the time-varying cold gas reference temperature, represents the time-varying radial gas temperature distribution, represents the blade height, represents a dimensionless radial distribution function calibrated from steady-state design point hot gas tunnel test data.

7. The rapid design method of a turbine blade according to claim 1, wherein The process of decoupling calculation of the transient temperature field and the transient stress field based on the three-dimensional model and the heat exchange boundary model comprises the following steps: The three-dimensional model is meshed, and the heat exchange boundary model is applied to the three-dimensional model to calculate the transient temperature field. Then, a stress field finite element model is constructed, and the grid node distribution of the stress field finite element model is completely consistent with the temperature field. The transient temperature field is applied to the stress field finite element model through one-to-one mapping of the grid nodes to calculate the transient stress field.

8. A rapid design system for a turbine blade, characterized by The method comprises the following steps: A boundary condition acquisition module is configured to obtain the boundary conditions of the gas side and the cold gas side of the turbine blade under the actual working environment based on the flow heat exchange test data; A heat exchange boundary model construction module is configured to construct a heat exchange boundary model based on the boundary conditions of the gas side and the cold gas side; A transient temperature-stress decoupling calculation module is configured to input a three-dimensional model of the turbine blade, and perform decoupling calculation of the transient temperature field and the transient stress field based on the three-dimensional model and the heat exchange boundary model; A local design optimization module is configured to perform local design iteration optimization based on the decoupling calculation results, and output an optimized three-dimensional model.

9. An electronic device, comprising: The computer program runs on a computer to perform the steps of the method of any one of claims 1-7.

10. A computer readable storage medium for storing a computer program for performing a rapid design of a turbine blade, characterized in that, The computer program runs on a computer to perform the steps of the method of any one of claims 1-7.

Images