Turbine case fatigue reliability optimization design method, device, equipment and medium

Abstract

The invention discloses a turbine case fatigue reliability optimization design method, device and equipment and a medium, and the method comprises the steps: S1, building a turbine case axial symmetry parametric model based on a parametric design idea; s2, aiming at the service environment of the turbine case, applying a temperature field and a cyclic load, carrying out thermosetting coupling analysis, and calculating and obtaining stress-strain data of a dangerous point; s3, carrying out fatigue reliability analysis by adopting an AK-MCS method, and replacing the implicit performance function by the self-adaptive agent model in the fatigue reliability analysis process; and S4, taking the maximum fatigue life mean value under the reliability constraint as a target, carrying out optimization by adopting a nested SQP method based on a self-adaptive agent model, and outputting optimal design parameters of the turbine case. According to the method, the efficiency of the turbine case reliability analysis method is improved, the turbine case reliability analysis precision is greatly improved, and efficient and high-precision reliability optimization design of the turbine case is achieved.

Description

Technical Field

[0001] This application relates to the field of aero-engine structural design and reliability engineering technology, and in particular to turbine casing fatigue reliability optimization design methods, devices, equipment and media. Background Technology

[0002] The turbine casing is a core load-bearing component of an aero-engine, responsible for supporting the rotor, forming an airflow channel, and containing high-temperature combustion gases. Its service environment exposes it to high temperatures (600-650℃), complex cyclic loads (internal and external pressure differences, thermal stress), and multiple uncertainties (material fluctuations, dimensional errors, load fluctuations). More than 70% of aero-engine casing accidents are caused by fatigue failure. For example, the failure rate of weld cracks in the rear casing mounting plate of a certain type of engine reached 3%, with the longest crack reaching 100mm, seriously threatening flight safety.

[0003] Current reliability calculation methods include traditional design, Monte Carlo simulation, simplified methods such as FORM / AFOSM, and the AK-MCS method. However, existing turbine casing fatigue reliability design methods have the following drawbacks: 1) Traditional design adopts deterministic methods based on allowable stress and empirical safety factors, without considering parameter uncertainties, which can easily lead to "over-design" or "under-design". 2) Among existing reliability analysis methods, direct Monte Carlo simulation (MCS) requires massive finite element calculations (a single thermo-mechanical coupling analysis of the casing takes more than 10 hours), which is extremely inefficient; 3) Simplification methods such as FORM / AFOSM have poor adaptability to implicit function types; 4) The AK-MCS method lacks application and adaptation optimization algorithms for turbine casings, making it difficult to meet the requirements of "high reliability + long life". Summary of the Invention

[0004] This application provides a fatigue reliability optimization design method for turbine casings, which solves the technical problems of low efficiency, low accuracy, poor adaptability and lack of specificity in the existing technology.

[0005] This application is achieved through the following solution: The turbine casing fatigue reliability optimization design method includes the following steps: S1. Based on the parametric design concept, construct an axisymmetric parametric model of the turbine casing; S2. Apply temperature field and cyclic load to the service environment of turbine casing, conduct thermo-mechanical coupling analysis, calculate and obtain stress and strain data at critical points; S3. The AK-MCS method is used to conduct fatigue reliability analysis. During the fatigue reliability analysis, an adaptive surrogate model is used to replace the implicit function. S4. To maximize the average fatigue life under reliability constraints, an optimization method based on an adaptive surrogate model and a nested SQP method is used to output the optimal design parameters of the turbine casing.

[0006] Furthermore, in step S1, when constructing the axisymmetric parameterized model of the turbine casing, the model retains all the core features of the original high-pressure turbine casing, including reinforcing ribs, mounting edges, hook rings and mounting holes, and ignores non-critical structures including supports to simplify calculations.

[0007] Furthermore, step S3 specifically includes the following steps: S31. Functional Function Definition: Based on Miner's linear cumulative damage theory, let the fatigue failure critical damage D be... CR =1, cumulative damage D=∑(n i / N fi Define the function: ; Where, n i N represents the number of cycles under multiple loads. fi To address fatigue life under various load conditions, n Indicates the number of cycles under a single load. N f This indicates the fatigue life under multiple loads. S32. Determination of random input variables: Select the model size variable that has a significant impact on fatigue life as a random input variable, and determine its distribution form, which includes normal distribution, uniform distribution and exponential distribution; S33. The AK-MCS method is used to conduct fatigue reliability analysis. The fatigue reliability analysis is based on the Kriging surrogate model and adaptive iterative update to calculate the failure probability and coefficient of variation.

[0008] Furthermore, the fatigue life N under the corresponding multi-load conditions f Calculated using the Moro model (considering mean stress correction): ; In the formula, The average strain amplitude at the critical point. The average stress at the critical point. E The elastic modulus of the material. (fatigue strength coefficient) The fatigue ductility coefficient is , b For fatigue strength index, c The fatigue ductility index is the parameter value mentioned above, which can be obtained from the fatigue test data of GH4169 material through linear heteroscedasticity regression.

[0009] Furthermore, step S33 specifically includes the following steps: S331. Sample pool generation: Based on the joint distribution of random variables, a Monte Carlo sample pool S is generated using Latin hypercube sampling; S332. Initial training set construction: Randomly select n0 samples from S, calculate fatigue life through thermo-mechanical coupling analysis and Moro model, obtain the function value g(x), and form the initial training set T; S333, Kriging surrogate model construction: A zeroth-order regression polynomial and a Gaussian correlation function are selected. The correlation parameter θ is determined through maximum likelihood estimation, and a Kriging surrogate model g is constructed. K (x), outputting the predicted mean μ g (x) and the prediction standard deviation σ g (x); S334, Adaptive Iterative Update: Calculate the U-learning function value (U=|μ) for all samples in the sample pool S. g (x) / σ g (x)|), select the sample x with the smallest U. new (The most likely misclassified sample), calculate its true function value g(x) new Add the model to the training set T and update the Kriging proxy model; repeat this process until minU(x)≥2; S335. Failure Probability Calculation: The failure state of each sample in the sample pool S is determined using a convergent Kriging surrogate model, where g K (x)≤0 indicates failure; calculate the failure probability. and coefficient of variation , This represents the number of failed samples.

[0010] Furthermore, step S4 specifically includes the following steps: S41. Optimization model construction, including design variables, objective function, and constraints. The design variables are: the mean of the desired optimization variable is selected as the design variable, and its value range and coefficient of variation are set. The objective function is to maximize the average fatigue life E(N). f The optimization model is calculated using the Kriging surrogate model; the constraints define reliability constraints and design variable boundary constraints, and the mathematical expression of the optimization model is: ; in, , representing the design variables for reliability optimization design. This represents the random input size variables of the model, which are related to the design variables. related, Represents the random input environment variables of the model. Indicates design variables The generated random input variables, This represents the fatigue life value of the turbine casing model obtained based on the Moro model, and the objective function is... This represents the average fatigue life, which is a variable that varies with the design life. A changing function, The function representing turbine casing fatigue failure determines whether a sample point of the input variable has failed by setting the fatigue life threshold to a target cycle number A. ; This represents the fatigue failure probability of the casing model, which is related to the design variables. Relevant variables, The threshold representing the fatigue reliability constraint of the casing model. and These represent design variables. Upper and lower bound constraints; S42, Nested SQP Optimization Method: This method uses a surrogate model, SQP subproblem solving, and a nested SQP optimization method based on an adaptive surrogate model to solve and output the optimal design parameters.

[0011] Further, step S42 specifically includes the following steps: S421, Agent Model Construction: Expanding the Design Variable Space (Based on Reliability Indicators) β 0 =3 determines the confidence interval [q] j - , q j + Within the scope of ]), fatigue life surrogate models N are constructed respectively. f K(x) (H learning function update) and failure boundary surrogate model g K (x) (U learning function update); Solving S422 and SQP subproblems: At the current design variable iteration point μX k The optimization problem can be approximated as a quadratic programming (QP) subproblem: ; in, H k It is a Hessian matrix. C k The gradient vector is given by N. f K(x) is obtained through finite difference. A k To constrain the gradient matrix, B k The constraint value is obtained from the failure probability calculated by gK(x) and the AK-MCS method; S423, Adaptive Update Agent Model: At each iteration point μX k A new sample pool is generated, and gK(x) and N are updated using the U / H learning function. f K(x) ensures the accuracy of the surrogate model; S424. Step Size Search and Convergence Judgment: Along the direction of the optimal solution of the QP subproblem S k Perform a one-dimensional search to determine the step size. α k Update design variable μX (k+1) =μX k +α k S k When ||μX (k+1) -μX k ||<10⁻ 4 When converged, the optimal design parameters of the turbine casing are output.

[0012] This application also provides a turbine casing fatigue reliability optimization design device, including: The model building module is used to construct an axisymmetric parametric model of the turbine casing based on the parametric design concept. The stress-strain data calculation module is used to apply temperature fields and cyclic loads to the service environment of the turbine casing, conduct thermo-mechanical coupling analysis, calculate and obtain stress-strain data at critical points; The reliability analysis module is used to conduct fatigue reliability analysis using the AK-MCS method. During the fatigue reliability analysis process, an adaptive surrogate model is used to replace the implicit function. The design parameter optimization module aims to maximize the average fatigue life under reliability constraints. It employs a nested SQP method based on an adaptive surrogate model to perform optimization and output the optimal design parameters for the turbine casing.

[0013] This application also provides an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the turbine casing fatigue reliability optimization design method when executing the computer program.

[0014] This application also provides a computer-readable storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the turbine casing fatigue reliability optimization design method.

[0015] Compared with the prior art, this application has the following advantages: This application proposes a fatigue reliability optimization design method for turbine casings that balances efficiency and accuracy. This method establishes an integrated "modeling-analysis-optimization" process for turbine casings, integrating parametric modeling, thermo-mechanical coupling analysis, AK-MCS quantitative reliability assessment, and nested SQP optimization. Combined with adaptive surrogate model optimization, it achieves efficient and high-precision reliability optimization design of turbine casings, improves the efficiency of turbine casing reliability analysis methods, significantly improves the accuracy of turbine casing reliability analysis, and meets the requirement of maximizing fatigue life under preset failure probability constraints. It is suitable for the design of key components of aero-engines with high reliability requirements.

[0016] In addition to the purposes, features, and advantages described above, this application has other purposes, features, and advantages. A further detailed description of this application will be provided below with reference to the figures. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein: Figure 1 This is a flowchart illustrating the turbine casing fatigue reliability optimization design method according to a preferred embodiment of this application; Figure 2 This is a schematic diagram of the axisymmetric parametric model of the turbine casing; Figure 3 This is a schematic diagram of the temperature boundary of the parameterized model of the turbine casing; Figure 4 This is a schematic diagram of the main cycle load; Figure 5 This is a schematic diagram of the stress distribution results at critical points obtained from thermo-mechanical coupling analysis; Figure 6 This is a schematic diagram of the fatigue reliability analysis process using the AK-MCS method; Figure 7 This is a schematic diagram of the module of the turbine casing fatigue reliability optimization design device according to a preferred embodiment of this application; Figure 8 This is a schematic block diagram of an electronic device according to a preferred embodiment of this application; Figure 9 This is a schematic diagram of the internal structure of a computer device according to a preferred embodiment of this application. Detailed Implementation

[0019] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0020] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or a turbine casing fatigue reliability optimization design device capable of performing the above functions. The following description uses a turbine casing fatigue reliability optimization design device as the executing entity to illustrate this embodiment and the subsequent embodiments.

[0021] like Figure 1 As shown, a preferred embodiment of this application provides a turbine casing fatigue reliability optimization design method, including the following steps: S1. Based on the parametric design concept, an axisymmetric parametric model of the turbine casing is constructed using UG software (see...). Figure 2 The model retains all the core features of the original high-pressure turbine casing (such as reinforcing ribs, mounting edges, hook rings, and 50 mounting holes) and ignores non-critical structures (such as supports) to simplify calculations. S2. For the turbine casing's service environment, a temperature field and cyclic load are applied to conduct thermo-mechanical coupling analysis, calculating and obtaining stress and strain data at critical points. Specifically, finite element analysis is used to obtain the overall stress and strain distribution of the component, and then the stress and strain values ​​at critical points are extracted. For the axisymmetric parametric model of the turbine casing established in step one, thermal analysis is performed, and a temperature field is applied in ANSYS. Based on the characteristics of the cooling flow path and temperature distribution of the high-pressure turbine casing, the temperature boundary of the casing can be divided into two categories: high-temperature region and low-temperature region. The low-temperature region can be further divided into flow cooling region and impact cooling region. Specific temperature boundary conditions are detailed in [link to relevant documentation]. Figure 3 Applying a pressure to the axisymmetric model of the turbine casing Figure 4 The cyclic load shown has a maximum internal and external pressure difference of ΔP = 1.2 MPa. The constraint condition is that the rear mounting edge of the turbine casing is fixed. The stress-strain data at the critical point from the finite element analysis are as follows: Figure 5 As shown, the maximum strain value of the casing model occurs at the connection between the tail section casing and the mounting edge; S3. The AK-MCS method is used to conduct fatigue reliability analysis. During the fatigue reliability analysis, an adaptive surrogate model is used to replace the implicit function. S4. To maximize the average fatigue life under reliability constraints, an optimization method based on an adaptive surrogate model and a nested SQP method is used to output the optimal design parameters of the turbine casing.

[0022] This embodiment proposes a fatigue reliability optimization design method for turbine casings that balances efficiency and accuracy. This method establishes an integrated "modeling-analysis-optimization" process for turbine casings, integrating parametric modeling, thermo-mechanical coupling analysis, AK-MCS quantitative reliability assessment, and nested SQP optimization. Combined with adaptive surrogate model optimization, it achieves efficient and high-precision reliability optimization design of turbine casings, improves the efficiency of turbine casing reliability analysis methods, significantly improves the accuracy of turbine casing reliability analysis, and meets the requirement of maximizing fatigue life under preset failure probability constraints. It is suitable for the design of key components of aero-engines with high reliability requirements.

[0023] Preferably, step S3 specifically includes the following steps: S31. Functional Function Definition: Based on Miner's linear cumulative damage theory, let the fatigue failure critical damage D be... CR =1, cumulative damage D=∑(n i / N fi Define the function: ; Where, n i N represents the number of cycles under multiple loads. fi To address fatigue life under various load conditions, n Indicates the number of cycles under a single load. N f This represents the fatigue life under multiple loads, where N represents the fatigue life under multiple loads. f Calculated using the Moro model (considering mean stress correction): ; In the formula, The average strain amplitude at the critical point. The average stress at the critical point. E The elastic modulus of the material. (fatigue strength coefficient) The fatigue ductility coefficient is , b For fatigue strength index, c The fatigue ductility index is the parameter value mentioned above, which can be obtained from the fatigue test data of GH4169 material through linear heteroscedasticity regression; the temperature value of the critical point is... T =300 0 C. Combining the fatigue performance parameters of GH4169 material at typical temperatures in the material data handbook, the fatigue performance of GH4169 material at 300°C was obtained through interpolation. 0 Fatigue strength coefficient under C fatigue ductility coefficient Fatigue strength index b =-0.07, fatigue ductility index c=-0.875, elastic modulus E =188 GPa Substituting the above fatigue performance parameters and the finite element analysis results of the casing model into the Moro model, the fatigue life of the casing model under the main cyclic load can be obtained as follows: ; S32. Determination of Random Input Variables: Model size variables that significantly affect fatigue life are selected as random input variables, and their distribution is determined. The distribution includes normal, uniform, and exponential distributions. Specifically, when determining the size variables for fatigue reliability analysis of the turbine casing axisymmetric parametric model, model size variables that have a significant impact on the maximum strain value of the casing model are primarily selected as random input variables. Internal and external pressure difference is selected. High temperature zone Temperature of the flow cooling zone Impact cooling zone temperature The tail section casing right side rounding radius ra6, mounting edge thickness b3, and auxiliary variables Seven influencing factors were selected as random input variables. Assuming that each random input variable is independent and follows a normal distribution, the statistical characteristics of each input variable are shown in Table 1 below.

[0024] Table 1 Statistical characteristics of each random variable S33. Fatigue reliability analysis is conducted using the AK-MCS method. The fatigue reliability analysis is based on the Kriging surrogate model and adaptive iterative updates to calculate the failure probability and coefficient of variation. Specific steps include (see...). Figure 6 ): S331. Sample Pool Generation: Based on the joint distribution of random variables, Latin hypercube sampling is used to generate a sample pool of size 1×10n. 5 Monte Carlo sample pool S; S332. Initial training set construction: Randomly select n0=30 samples from S, calculate fatigue life through thermo-mechanical coupling analysis and Moro model, obtain the function value g(x), and form the initial training set T={(x1,g(x1)),…,(x30,g(x30))}; S333, Kriging surrogate model construction: A zero-order regression polynomial and a Gaussian correlation function (R(xi,xj;θ)=exp(-∑θk|xi,k-xj,k|²)) are selected. The correlation parameter θ is determined through maximum likelihood estimation, and the Kriging surrogate model g is constructed. K (x), outputting the predicted mean μ g (x) and the prediction standard deviation σ g (x); S334, Adaptive Iterative Update: Calculate the U-learning function value (U=|μ) for all samples in the sample pool S. g (x) / σ g (x)|), select the sample x with the smallest U. new (The most likely misclassified sample), calculate its true function value g(x) new Add the model to the training set T and update the Kriging surrogate model; repeat this process until minU(x)≥2 (sample classification accuracy≥97.7%). S335. Failure Probability Calculation: The failure state of each sample in the sample pool S is determined using a convergent Kriging surrogate model, where g K (x)≤0 indicates failure; calculate the failure probability. and coefficient of variation , For the number of failed samples, when the coefficient of variation The results are valid when the failure rate is less than 5%. Table 2 shows the estimation results of the fatigue failure probability of the turbine casing model using the AK-MCS method. Table 2. Estimation results of fatigue failure probability of turbine casing model using AK-MCS method Preferably, step S4 specifically includes the following steps: S41. Optimization Model Construction: This includes design variables, objective function, and constraints. The design variables are the mean values ​​of the variables to be optimized, with their ranges and coefficients of variation set. The objective function is to maximize the mean fatigue life E(N). f The constraints are calculated by the Kriging surrogate model; the constraints define reliability constraints and design variable boundary constraints. In this embodiment: Design variables: Choose the mean μ of ra6 and b3. ra6 μ b3 As a design variable, its value range is μ. ra6 ∈[1.3,1.5]mm, μ b3 The range is ∈[1.8,2.2] mm, and the coefficient of variation is 0.01 for all values. The design variables are shown in Table 3. Table 3 Design variables for turbine casing fatigue reliability optimization Objective function: Maximize the mean fatigue life E(N) f The result is calculated by the Kriging surrogate model (updated by the H learning function, with convergence condition maxH(x)≤1); Constraints: ① Reliability constraints: P f ≤ Pf *=0.005; ② Design variable boundary constraints: μ ra6 l ≤μ ra6 ≤μ ra6 u μ b3 l ≤μ b3 ≤μ b3 u ; Therefore, the mathematical expression for the optimization model is: ; in, , representing the design variables for reliability optimization design. This represents the random input size variables of the model, which are related to the design variables. related, Represents the random input environment variables of the model. Indicates design variables The generated random input variables, This represents the fatigue life value of the turbine casing model obtained based on the Moro model, and the objective function is... This represents the average fatigue life, which is a variable that varies with the design life. A changing function, The function representing turbine casing fatigue failure determines whether a sample point of the input variable has failed by setting the fatigue life threshold to a target cycle number A. ; This represents the fatigue failure probability of the casing model, which is related to the design variables. Relevant variables, The threshold representing the fatigue reliability constraint of the casing model. and These represent design variables. Upper and lower bound constraints; Specifically, in this embodiment... Design variables representing reliability optimization design, This represents the random input size variables of the model, which are related to the design variables. related, Represents the random input environment variables of the model; The function representing turbine casing fatigue failure determines whether a sample point of the input variable has failed by setting a fatigue life threshold of 400 loops. ; This represents the fatigue failure probability of the casing model, which is also related to design variables. Regarding the relevant variables, the threshold for fatigue reliability constraints is set to... , and These represent design variables. Upper and lower bound constraints; S42. Nested SQP Optimization Method: This method uses a surrogate model, SQP subproblem solving, and a nested SQP optimization method based on an adaptive surrogate model to solve and output the optimal design parameters. Specific steps include: S421, Agent Model Construction: Expanding the Design Variable Space (Based on Reliability Indicators) β 0 =3 determines the confidence interval [q] j - , q j + Within the scope of ]), fatigue life surrogate models N are constructed respectively. f K(x) (H learning function update) and failure boundary surrogate model g K (x) (U learning function update); Solving S422 and SQP subproblems: At the current design variable iteration point μX k The optimization problem can be approximated as a quadratic programming (QP) subproblem: ; in, H k It is a Hessian matrix. C k The gradient vector is given by N. f K(x) is obtained through finite difference. A k To constrain the gradient matrix, B k The constraint value is obtained from the failure probability calculated by gK(x) and the AK-MCS method; S423, Adaptive Update Agent Model: At each iteration point μX k A new sample pool is generated, and gK(x) and N are updated using the U / H learning function. f K(x) ensures the accuracy of the surrogate model; S424. Step Size Search and Convergence Judgment: Along the direction of the optimal solution of the QP subproblem S k Perform a one-dimensional search to determine the step size. α k Update design variable μX (k+1) =μX k +α k S k When ||μX (k+1) -μX k ||<10⁻ 4 Upon convergence, the optimal design parameters for the output turbine casing are: μ ra6=1.5mm, μ b3 =2.0mm.

[0025] The optimization results are verified below: The performance of the optimized turbine casing model was verified. Table 4 shows a comparison between the optimal solution of the obtained turbine casing fatigue reliability optimization model and the initial turbine casing solution. Table 4 Comparison of Turbine Casing Fatigue Reliability Optimization Results The optimization results of the average fatigue life show that the average fatigue life of the turbine casing was 1475 before optimization, and increased to 1920 after optimization, representing a 30.2% improvement. Furthermore, the coefficient of variation of the fatigue life after reliability optimization was lower than that before optimization. Table 4 shows that the optimized model dimensions improved the average fatigue life of the turbine casing model to some extent and met certain fatigue reliability requirements. The average fatigue life is the largest under the condition that the present application has been proven to have high practical value through simulation.

[0026] The key protection points of the above embodiments include: 1) Perform thermo-mechanical coupling analysis on the parameterized model of the turbine casing to obtain stress and strain data at critical points; 2) Conduct fatigue reliability analysis of the turbine casing based on the AK-MCS method and estimate the probability of fatigue failure; 3) The fatigue reliability optimization model is established by using the nested SQP method based on the adaptive surrogate model to solve for the optimal design parameters.

[0027] like Figure 7 As shown, another preferred embodiment of this application also provides a turbine casing fatigue reliability optimization design device, including: The model building module is used to construct an axisymmetric parametric model of the turbine casing based on the parametric design concept. The stress-strain data calculation module is used to apply temperature fields and cyclic loads to the service environment of the turbine casing, conduct thermo-mechanical coupling analysis, calculate and obtain stress-strain data at critical points; The reliability analysis module is used to conduct fatigue reliability analysis using the AK-MCS method. During the fatigue reliability analysis process, an adaptive surrogate model is used to replace the implicit function. The design parameter optimization module aims to maximize the average fatigue life under reliability constraints. It employs a nested SQP method based on an adaptive surrogate model to perform optimization and output the optimal design parameters for the turbine casing.

[0028] The turbine casing fatigue reliability optimization design device provided in this embodiment adopts the turbine casing fatigue reliability optimization design method in the above embodiment, which solves the technical problems of low efficiency, low accuracy, poor adaptability and lack of specificity in the prior art. Compared with the prior art, the beneficial effects of the turbine casing fatigue reliability optimization design device provided in this embodiment are the same as the beneficial effects of the turbine casing fatigue reliability optimization design method provided in the above embodiment. Moreover, other technical features in the turbine casing fatigue reliability optimization design device are the same as the features disclosed in the method of the above embodiment, and will not be repeated here.

[0029] like Figure 8 As shown, a preferred embodiment of this embodiment also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the turbine casing fatigue reliability optimization design method in the above embodiment.

[0030] This embodiment provides an electronic device that employs the turbine casing fatigue reliability optimization design method described in the above embodiments. This method addresses the technical problems of low efficiency, low precision, poor adaptability, and lack of specificity in the prior art. Compared with the prior art, the beneficial effects of the electronic device provided in this embodiment are the same as those of the turbine casing fatigue reliability optimization design method described in the above embodiments. Furthermore, the other technical features of the electronic device are the same as those disclosed in the methods of the above embodiments, and will not be elaborated upon here.

[0031] like Figure 9 As shown in the preferred embodiment, this embodiment also provides a computer device, which may be a terminal or a liveness detection server, and its internal structure diagram may be as follows. Figure 9 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with other external computer devices via a network connection. When the computer program is executed by the processor, it implements the steps of the aforementioned turbine casing fatigue reliability optimization design method.

[0032] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the solution of this embodiment, and does not constitute a limitation on the computer device to which the solution of this embodiment is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0033] The computer equipment provided in this application adopts the turbine casing fatigue reliability optimization design method in the above embodiments, which solves the technical problems of low efficiency, low accuracy, poor adaptability and lack of specificity in the prior art. Compared with the prior art, the beneficial effects of the computer equipment provided in this embodiment are the same as the beneficial effects of the turbine casing fatigue reliability optimization design method provided in the above embodiments. In addition, other technical features in the electronic equipment are the same as the features disclosed in the method of the above embodiments, and will not be repeated here.

[0034] A preferred embodiment of this example also provides a storage medium, which includes a stored program that, when the program is executed, controls the device containing the storage medium to perform the steps of the turbine casing fatigue reliability optimization design method described in the above embodiment.

[0035] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0036] If the functions described in this embodiment are implemented as software functional units and sold or used as independent products, they can be stored in one or more computing device-readable storage media. Based on this understanding, the parts of this embodiment that contribute to the prior art or the technical solution can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computing device (which may be a personal computer, server, mobile computing device, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this embodiment. The aforementioned storage media include: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0037] Those skilled in the art will understand that embodiments of this example can be provided as methods, systems, or computer program products. Therefore, this example can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this example 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 this example can be implemented using various computer languages, such as the object-oriented programming language C++ and the embedded programming language C.

[0038] This embodiment is described with reference to flowchart illustrations and / or block diagrams of the method, apparatus (system), and computer program product according to this embodiment. It should 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 illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0039] 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.

[0040] 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.

[0041] This embodiment also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the turbine casing fatigue reliability optimization design method described above.

[0042] The computer program product provided in this embodiment solves the technical problems of low efficiency, low accuracy, poor adaptability, and lack of specificity in the prior art. Compared with the prior art, the beneficial effects of the computer program product provided in this embodiment are the same as those of the turbine casing fatigue reliability optimization design method provided in the above embodiments, and will not be repeated here.

[0043] Obviously, those skilled in the art can make various modifications and variations to this embodiment without departing from the spirit and scope of this embodiment. Therefore, if these modifications and variations of this embodiment fall within the scope of the claims of this embodiment and their equivalents, this embodiment is also intended to include these modifications and variations.

Claims

1. A method for optimizing the fatigue reliability of turbine casings, characterized in that, Including the following steps: S1. Based on the parametric design concept, construct an axisymmetric parametric model of the turbine casing; S2. Apply temperature field and cyclic load to the service environment of turbine casing, conduct thermo-mechanical coupling analysis, calculate and obtain stress and strain data at critical points; S3. The AK-MCS method is used to conduct fatigue reliability analysis. During the fatigue reliability analysis, an adaptive surrogate model is used to replace the implicit function. S4. To maximize the average fatigue life under reliability constraints, an optimization method based on an adaptive surrogate model and a nested SQP method is used to output the optimal design parameters of the turbine casing.

2. The turbine casing fatigue reliability optimization design method according to claim 1, characterized in that, In step S1, when constructing the axisymmetric parametric model of the turbine casing, the model retains all the core features of the original high-pressure turbine casing, including reinforcing ribs, mounting edges, hook rings and mounting holes, and ignores non-critical structures including supports to simplify the calculation.

3. The turbine casing fatigue reliability optimization design method according to claim 1, characterized in that, Step S3 specifically includes the following steps: S31. Functional Function Definition: Based on Miner's linear cumulative damage theory, let the fatigue failure critical damage D be... CR =1, cumulative damage D=∑(n i / N fi Define the function: ； Where, n i N represents the number of cycles under multiple loads. fi To address fatigue life under various load conditions, n Indicates the number of cycles under a single load. N f This indicates the fatigue life under multiple loads. S32. Determination of random input variables: Select the model size variable that has a significant impact on fatigue life as a random input variable, and determine its distribution form, which includes normal distribution, uniform distribution and exponential distribution; S33. The AK-MCS method is used to conduct fatigue reliability analysis. The fatigue reliability analysis is based on the Kriging surrogate model and adaptive iterative update to calculate the failure probability and coefficient of variation.

4. The turbine casing fatigue reliability optimization design method according to claim 3, characterized in that, The fatigue life N under the corresponding multi-load conditions f Calculated using the Moro model: ； In the formula, The average strain amplitude at the critical point. The average stress at the critical point. E The elastic modulus of the material. (fatigue strength coefficient) The fatigue ductility coefficient is , b For fatigue strength index, c The fatigue ductility index is the parameter value mentioned above, which can be obtained from the fatigue test data of GH4169 material through linear heteroscedasticity regression.

5. The turbine casing fatigue reliability optimization design method according to claim 3 or 4, characterized in that, Step S33 specifically includes the following steps: S331. Sample pool generation: Based on the joint distribution of random variables, a Monte Carlo sample pool S is generated using Latin hypercube sampling; S332. Initial training set construction: Randomly select n0 samples from S, calculate fatigue life through thermo-mechanical coupling analysis and Moro model, obtain the function value g(x), and form the initial training set T; S333, Kriging surrogate model construction: A zeroth-order regression polynomial and a Gaussian correlation function are selected. The correlation parameter θ is determined through maximum likelihood estimation, and a Kriging surrogate model g is constructed. K (x), outputting the predicted mean μ g (x) and the predicted standard deviation σ g (x); S334, Adaptive Iterative Update: Calculate the U-learning function value (U=|μ) for all samples in the sample pool S. g (x) / σ g (x)|), select the sample x with the smallest U. new Calculate its true function value g(x) new Add the model to the training set T and update the Kriging proxy model; repeat this process until minU(x)≥2; S335. Failure Probability Calculation: The failure state of each sample in the sample pool S is determined using a convergent Kriging surrogate model, where g K (x)≤0 indicates failure; calculate the failure probability. and coefficient of variation , This represents the number of failed samples.

6. The turbine casing fatigue reliability optimization design method according to claim 1, characterized in that, Step S4 specifically includes the following steps: S41. Optimization model construction, including design variables, objective function, and constraints. The design variables are: the mean of the desired optimization variable is selected as the design variable, and its value range and coefficient of variation are set. The objective function is to maximize the average fatigue life E(N). f The optimization model is calculated using the Kriging surrogate model; the constraints define reliability constraints and design variable boundary constraints, and the mathematical expression of the optimization model is: ； in, , representing the design variables for reliability optimization design. This represents the random input size variables of the model, which are related to the design variables. related, Represents the random input environment variables of the model. Indicates design variables The generated random input variables, This represents the fatigue life value of the turbine casing model obtained based on the Moro model, and the objective function is... This represents the average fatigue life, which is a variable that varies with the design life. A changing function, The function representing turbine casing fatigue failure determines whether a sample point of the input variable has failed by setting the fatigue life threshold to a target cycle number A. ; This represents the fatigue failure probability of the casing model, which is related to the design variables. Relevant variables, The threshold representing the fatigue reliability constraint of the casing model. and These represent design variables. Upper and lower bound constraints; S42, Nested SQP Optimization Method: This method uses a surrogate model, SQP subproblem solving, and a nested SQP optimization method based on an adaptive surrogate model to solve and output the optimal design parameters.

7. The turbine casing fatigue reliability optimization design method according to claim 6, characterized in that, Step S42 specifically includes the following steps: S421. Proxy Model Construction: Within the design variable extension space, construct fatigue life proxy models N respectively. f K(x) and the failure boundary surrogate model g K (x); Solving S422 and SQP subproblems: At the current design variable iteration point μX k The optimization problem can be approximated as a quadratic programming subproblem: ； in, H k It is a Hessian matrix. C k The gradient vector is given by N. f K(x) is obtained through finite difference. A k To constrain the gradient matrix, B k The constraint value is obtained from the failure probability calculated by gK(x) and the AK-MCS method; S423, Adaptive Update Agent Model: At each iteration point μX k A new sample pool is generated, and gK(x) and N are updated using the U / H learning function. f K(x) ensures the accuracy of the surrogate model; S424. Step Size Search and Convergence Judgment: Along the direction of the optimal solution of the QP subproblem S k Perform a one-dimensional search to determine the step size. α k Update design variable μX (k+1) =μX k +α k S k When ||μX (k+1) -μX k ||<10⁻ 4 When converged, the optimal design parameters of the turbine casing are output.

8. A turbine casing fatigue reliability optimization design device, characterized in that, include: The model building module is used to construct an axisymmetric parametric model of the turbine casing based on the parametric design concept. The stress-strain data calculation module is used to apply temperature fields and cyclic loads to the service environment of the turbine casing, conduct thermo-mechanical coupling analysis, calculate and obtain stress-strain data at critical points; The reliability analysis module is used to conduct fatigue reliability analysis using the AK-MCS method. During the fatigue reliability analysis process, an adaptive surrogate model is used to replace the implicit function. The design parameter optimization module aims to maximize the average fatigue life under reliability constraints. It employs a nested SQP method based on an adaptive surrogate model to perform optimization and output the optimal design parameters for the turbine casing.

9. An electronic device, the electronic device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the computer program, it implements the turbine casing fatigue reliability optimization design method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the turbine casing fatigue reliability optimization design method as described in any one of claims 1 to 7.

Images