Method, device, equipment, medium and product for optimizing probability strength design of turbine blade

Abstract

The application discloses a turbine blade probability strength design optimization method, device, equipment, medium and product, relates to the blade strength design technical field, and the method comprises the following steps: obtaining a preliminary blade geometric structure and initially selecting material parameters according to strength requirements and operating load conditions; taking the probability distribution of load conditions, geometric structures, material parameters and model parameters as input variables of finite elements to obtain stress and strain responses; designing a structure similar to a structure, carrying out multilevel verification of the accuracy of a life prediction model, calculating the total damage of the blade based on the life prediction model, obtaining the probability distribution of the total damage, and then obtaining the reliability under the design life that meets the strength requirements; judging whether the reliability meets the design requirements, and if not, optimizing the design with the goal of improving the reliability value until the reliability meets the design requirements. The application can consider the objective uncertainty in the design process, improve the component life prediction accuracy, and achieve more accurate strength design.

Description

Technical Field

[0001] This application relates to the field of blade strength design technology, and in particular to a method, apparatus, equipment, medium and product for probabilistic strength design optimization of turbine blades. Background Technology

[0002] Turbine blades, as typical hot-end components of advanced gas turbines, embody high reliability and long service life in their design. However, they are also a weak link in gas turbine lifespan design, becoming a bottleneck restricting gas turbine development. Gas turbine hot-end components operate in high-temperature, high-speed, and complex environments, exhibiting significant characteristics such as severe and complex stress, multiple failure modes, multi-mode compound failures, a large number of parts, and the requirement for long service life. Therefore, reliable blade strength design is crucial.

[0003] Traditional turbine blade strength design is based on deterministic methods. Starting from the design strength requirements, finite element simulations are performed. A conservative blade life is calculated based on a simple life prediction model with a large safety factor. If the life requirement is not met, the geometry and materials are adjusted, and the above calculation process is repeated until the life requirement is met. This traditional strength design method has redundancy in the turbine blade life assessment and cannot fully utilize the entire life of the turbine blade, leading to premature replacement and increased operating costs. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, equipment, medium, and product for probabilistic strength design optimization of turbine blades, which can take into account objective uncertainties in the design process, improve the accuracy of component life prediction, and achieve more accurate strength design.

[0005] To achieve the above objectives, this application provides the following solution:

[0006] Firstly, this application provides a method for probabilistic strength design optimization of turbine blades, the method comprising:

[0007] Based on strength requirements and operating load conditions, a preliminary blade geometry was obtained and material parameters were initially selected.

[0008] The probability distributions of load conditions, geometry, material parameters, and model parameters are used as input variables for finite element analysis to obtain stress-strain response.

[0009] Based on the structural characteristics of the dangerous parts of the blade, structural components are designed. The accuracy of the life prediction model is verified at the material level and the structural component level. Based on the life prediction model verified at the multi-level level, the total damage of the blade is calculated and the probability distribution of the total damage is obtained.

[0010] The reliability is calculated based on the probability distribution of total damage, and the reliability under the design life that meets the strength requirements is obtained.

[0011] Determine whether the reliability meets the design requirements. If not, optimize the design of the geometry and material parameters with the goal of improving the reliability value until the reliability meets the design requirements.

[0012] Optionally, the step of using the probability distributions of load conditions, geometry, material parameters, and model parameters as inputs to the finite element method to perform finite element calculations and obtain the stress-strain response specifically includes:

[0013] The load conditions are obtained by using historical data of blade operation, the probability distribution of geometric structure is obtained by multiple measurements, and the probability distribution of material parameters and model parameters is obtained by multiple parallel tests under the same load conditions.

[0014] Random sampling is performed based on the probability distribution of the input variables. Each set of sampling results is input into the finite element calculation to obtain multiple sets of stress-strain responses.

[0015] Optionally, the multi-level verification of the accuracy of the lifetime prediction model at the material and structural component levels specifically includes:

[0016] Fatigue damage is obtained through a fatigue life calculation model;

[0017] Creep damage is obtained through a creep life calculation model;

[0018] The fatigue damage and the creep damage are added together to obtain the total damage;

[0019] The predicted blade life is obtained by taking the reciprocal of the total damage.

[0020] The blade's test life is obtained through mechanical property testing at the material and structural component levels, and the predicted blade life is verified by the blade test life.

[0021] Optionally, the life prediction model based on multi-level verification calculates the total blade damage to obtain the probability distribution of the total damage, specifically including:

[0022] Fatigue damage and creep damage are calculated based on the stress-strain response, and the input dataset and output dataset are accumulated after batch calculations.

[0023] Train a neural network agent model using the input and output datasets;

[0024] Sampling is performed on the input variables, and the damage response is obtained through the trained neural network surrogate model. Multiple sets of corresponding damage responses are obtained through multiple samplings, and finally the probability distributions of fatigue damage, creep damage, and total damage are obtained.

[0025] Optionally, the reliability calculation based on the probability distribution of total damage to obtain the reliability under the design life that meets the strength requirements specifically includes:

[0026] The probability distribution of the damage threshold is determined based on the leaf test data. The formula for calculating the damage threshold is:

[0027] ;

[0028] In the formula, For the lifespan of a certain test, The mean lifetime is the result of multiple tests under the same working conditions. The probability distribution of the damage threshold can be obtained by conducting multiple tests under the same working conditions.

[0029] Based on the probability distribution of total damage under the current operating condition of the blade, the reliability under the strength design requirements is calculated.

[0030] The reliability is calculated using the cumulative damage-damage threshold interference criterion, and the specific formula is as follows:

[0031] ;

[0032] In the formula, P f It is the probability of failure. R It's about reliability. D crit It is a damage threshold related to material properties. D(N) The total damage to the blade under its current operating condition accumulates over time. Both are random variables, and reliability is the area of ​​intersection between them. Let Pr(·) be the limit state equation, and let Pr(·) denote the probability calculation.

[0033] Optionally, the optimization design of the geometric structure and material parameters with the goal of improving the reliability value is specifically formulated as follows:

[0034] ;

[0035] In the formula, Design variables for blade geometry. For material properties, Let be the objective function. This represents the probability of failure. For the remaining variables that are not optimized, For the first The maximum probability of failure allowed by each constraint.

[0036] Secondly, this application provides a turbine blade probabilistic strength design optimization device, comprising:

[0037] The parameter selection unit is used to obtain the preliminary blade geometry and preliminarily select material parameters based on strength requirements and operating load conditions.

[0038] The first calculation unit is used to take the probability distribution of load conditions, geometric structure, and material parameters as input variables of the finite element method, perform finite element calculations, and obtain stress-strain response.

[0039] The second calculation unit is used to design structural components based on the structural characteristics of the dangerous parts of the blade, to conduct multi-level verification of the accuracy of the life prediction model at the material level and the structural component level, and to calculate the total damage of the blade based on the multi-level verified life prediction model to obtain the probability distribution of the total damage.

[0040] The third calculation unit is used to calculate the reliability based on the probability distribution of total damage, and obtain the reliability under the design life that meets the strength requirements;

[0041] The optimization design unit is used to determine whether the reliability meets the design requirements. If not, the geometric structure and material parameters are optimized with the goal of improving the reliability value until the reliability meets the design requirements.

[0042] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the turbine blade probabilistic strength design optimization method described in any one of the above.

[0043] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the turbine blade probabilistic strength design optimization method described above.

[0044] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the turbine blade probabilistic strength design optimization method described above.

[0045] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0046] This application provides a probabilistic strength design optimization method, apparatus, equipment, medium, and product for turbine blades, which can effectively overcome the conservatism and redundancy of traditional deterministic strength design methods. By introducing multi-level verification of material-level and structural component-level life prediction models, the accuracy and reliability of life prediction for critical critical parts of the blade can be improved. By considering the uncertainties of load conditions, geometry, material parameters, and model parameters in the design, a shift from deterministic design to a more scientific and reasonable probabilistic design is achieved. Finally, using the calculated reliability as a key indicator to guide structural optimization and material selection, the optimal balance between economy and safety in turbine blade design can be achieved while meeting strength requirements and design life reliability, effectively solving the problem of excessive conservatism caused by the reliance on large safety factors in traditional methods. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A flowchart illustrating a probabilistic strength design optimization method for turbine blades provided in an embodiment of this application;

[0049] Figure 2 A finite element model of a turbine blade and a wheel disk is provided in one embodiment of this application;

[0050] Figure 3 An irregular notch-type structural component provided in one embodiment of this application;

[0051] Figure 4a A radial comparison diagram of strain and strain gradient at the tenon of a structural component and a turbine blade provided in an embodiment of this application;

[0052] Figure 4b A circumferential comparison diagram of strain and strain gradient at the tenon of a structural component and a turbine blade provided in an embodiment of this application;

[0053] Figure 5 A schematic diagram illustrating the verification results of the lifetime prediction method provided in an embodiment of this application using material and structural component test data.

[0054] Figure 6 Fatigue damage, creep damage, and total damage cloud diagrams of a turbine blade obtained from finite element calculations in an embodiment of this application;

[0055] Figure 7a A turbine blade design life-failure probability curve provided in an embodiment of this application;

[0056] Figure 7b A pie chart illustrating the variable failure sensitivity analysis of a turbine blade provided in an embodiment of this application;

[0057] Figure 8 A schematic diagram of the functional modules of a turbine blade probabilistic strength design optimization device provided in an embodiment of this application;

[0058] Figure 9 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0059] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0060] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0061] In one exemplary embodiment, such as Figure 1 As shown, a probabilistic strength design optimization method for turbine blades is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, it includes the following steps 101 to 105. Wherein:

[0062] Step 101: Based on the strength requirements and operating load conditions, obtain the preliminary blade geometry and preliminarily select the material parameters;

[0063] In this embodiment, a gas turbine blade (hereinafter referred to as the blade for ease of description) is used as an example to illustrate the application process of the probabilistic strength design and optimization method. The simplified finite element model of the blade and its assembly with the turbine disk is as follows: Figure 2 As shown. The centrifugal force resulting from the overall loading at a rotational speed of 314 rad / s, and the aerodynamic surface force of 1 MPa applied to the inner side of the blade, the waveform is loading-holding-unloading, the holding time is 3600 s, and the maximum temperature is 900 ℃. The blade tenon and the wheel disk tenon are tightly fitted and set as friction contact.

[0064] In this embodiment of the application, based on the strength requirement of fatigue life exceeding 300 cycles, the material selected for finite element modeling is MarM247.

[0065] Step 102: Using the probability distributions of load conditions, geometric structure, material parameters, and model parameters as input variables for the finite element method, perform finite element calculations to obtain the stress-strain response.

[0066] As an optional implementation, the step of using the probability distributions of load conditions, geometric structures, material parameters, and model parameters as input variables for finite element calculations specifically includes:

[0067] The load conditions are obtained by using historical data of blade operation, the probability distribution of geometric structure is obtained by multiple measurements, and the probability distribution of material parameters and model parameters is obtained by multiple parallel tests under the same load conditions.

[0068] Random sampling is performed based on the probability distribution of the input variables. Each set of sampling results is input into the finite element calculation to obtain multiple sets of stress-strain responses.

[0069] In this embodiment, the probability distribution of the load condition, geometric structure, material parameters, and model parameters is determined by the KS test. The probability distribution to be tested includes, but is not limited to, the normal distribution, the log-normal distribution, and the Weibull distribution.

[0070] In this embodiment of the application, based on the actual multi-source uncertainty factors, specific uncertain variables in material parameters, load conditions, geometric structures, and model parameters are determined, wherein: uncertain variables in material parameters include density, coefficient of thermal expansion, and elastic modulus; uncertain variables in load conditions include maximum rotational speed and holding time; uncertain variables in geometric structures include the critical dimension of the blade root tenon chamfer; and uncertain variables in model parameters include several model parameters of the life model to be selected.

[0071] In this embodiment of the application, the probability distribution, mean, and coefficient of variation of the input variables of the blade, namely load conditions, geometric structure, material parameters, and model parameters, are statistically calculated to obtain the multi-source uncertainty input data of the blade, as shown in Table 1.

[0072] Table 1. Input data for multi-source uncertainties of the blade.

[0073]

[0074] In Table 1, For fatigue strength factor, It is the fatigue ductility factor. For creep material parameters, It is the critical strain energy density.

[0075] This implementation method uses the probability distributions of load conditions, geometry, material parameters, and model parameters as inputs to perform finite element analysis (FEM) calculations. This allows for the determination of the blade's stress-strain response, facilitating the calculation of fatigue damage, creep damage, and lifespan. Furthermore, after multiple sets of sampling calculations, the corresponding probability distributions—namely, the probability distributions of fatigue damage, creep damage, and lifespan—can be obtained.

[0076] Step 103: Design structural components based on the structural characteristics of the dangerous parts of the blade, and conduct multi-level verification of the accuracy of the life prediction model at the material level and structural component level. Calculate the total blade damage based on the verified life prediction model and obtain the probability distribution of the total damage.

[0077] In this embodiment, the structural component is designed based on the structural characteristics of the dangerous (vulnerable) parts of the blade, requiring geometric similarity and similar stress states. For example, in this embodiment, the blade tenon is a dangerous part, requiring the strain gradient of the structural component to be similar to the finite element calculation results at the blade tenon. The designed structural component specimen is as follows: Figure 3 As shown, this is an irregular notch. The strain and strain gradient curves at the structural component and blade tenon are as follows: Figure 4a and Figure 4b As shown, where Figure 4a For radial contrast, Figure 4b For circumferential comparison.

[0078] As an optional implementation, the multi-level verification of the accuracy of the lifetime prediction model at the material and structural component levels specifically includes:

[0079] Fatigue damage is obtained through a fatigue life calculation model;

[0080] Creep damage is obtained through a creep life calculation model;

[0081] The fatigue damage and the creep damage are added together to obtain the total damage;

[0082] The predicted blade life is obtained by taking the reciprocal of the total damage.

[0083] The blade's test life is obtained through mechanical property testing at the material and structural component levels, and the predicted blade life is verified by the blade test life.

[0084] In this embodiment, multiaxial fatigue damage is calculated using the critical distance theory. The critical distance method can account for the influence of strain gradients, avoiding singular values ​​that may exist in single-point calculations within the finite element method. The equivalent strain is obtained by integrating the strain sequence within a certain distance and then substituted into the conventional fatigue life calculation model, Manson-Coffin (MC). The critical distance and fatigue life have a power-law relationship, meaning they are coupled. The critical distance, fatigue life, and damage under different working conditions can be obtained through iterative methods. The expression for the fatigue life calculation model is as follows:

[0085] ;

[0086] In the formula, L This is the critical distance. N f The fatigue life is the factorial factor. Taking the reciprocal of the fatigue life yields the fatigue damage rate per week. A and B These are material parameters.

[0087] Equivalent stress and equivalent strain are derived from the stress and strain integrals within the critical distance and substituted into the creep life calculation model using the strain energy density dissipation method (SEDE). Since the critical distance is related to the holding time, the creep life and holding time are also affected. t Given the existing relationship, the expression for the creep life calculation model is as follows:

[0088] ;

[0089] In the formula, L This is the critical distance. N c Given the creep life, taking the reciprocal of the creep life yields the creep damage per week. t It is the load holding time. A , B , C and D All of these are material parameters.

[0090] In this embodiment, fatigue damage and creep damage are linearly accumulated to obtain total damage. When the total damage accumulates to 1 cycle, all cycles are added together (or the reciprocal of the total damage is taken to obtain the lifespan) to obtain the predicted lifespan of the blade under the creep-fatigue interaction mode.

[0091] In this embodiment of the application, the predicted life of the blade is verified through mechanical performance test data at the material level and the structural component level. Specifically, this means that test data, i.e., experimental life, is obtained by conducting mechanical performance tests at the material level and the structural component level. By comparing the experimental life with the predicted life, a low error is required, thus verifying the accuracy of the blade life prediction model. Figure 5As shown, both material-level and structural component-level test points fall within the 2x error band, validating the accuracy of the life prediction model.

[0092] In this embodiment of the application, if the error between the test life and the predicted life is too large, for example, if the test points at the material level and the structural component level exceed the 2x error band, then the accuracy of the blade life prediction model is considered to be too low. The model can be corrected by modifying the form of the life prediction model or adding material parameters to the life prediction model, thereby improving the accuracy of the life prediction model. The life prediction model is then verified again until the life prediction model meets the accuracy requirements.

[0093] As an optional implementation, the life prediction model based on multi-level verification calculates the total damage to the blade and obtains the probability distribution of the total damage, specifically including:

[0094] Based on the stress-strain response obtained in step 102, fatigue damage and creep damage of the blade are calculated, and the input dataset and output dataset are accumulated after batch calculation.

[0095] Train a neural network agent model using the input and output datasets;

[0096] Sampling is performed on the probability distribution of the input variables, and the damage response is obtained through the trained neural network surrogate model. Multiple sets of corresponding damage responses are obtained through multiple samplings, and finally the probability distributions of fatigue damage, creep damage, total damage, and lifespan are obtained.

[0097] In this embodiment of the application, the neural network proxy model refers to the BP artificial neural network model.

[0098] In this embodiment, the fatigue life calculation model and creep fatigue life calculation model described above are applied, and the mean values ​​of each variable in Table 1 are substituted into the finite element method for calculating the blade to obtain the fatigue damage, creep damage, and total damage under conventional deterministic calculation. Figure 6 As shown, the creep-fatigue life is the reciprocal of the total damage, calculated to be 306 cycles, which can be used to compare with the results of probability intensity calculation.

[0099] In this embodiment, the input variables of the multi-source uncertain input data in Table 1 are sampled and substituted into the finite element method to obtain the stress-strain response and calculate fatigue damage and creep damage. After batch calculation, the input and output datasets are accumulated. A neural network surrogate model is trained using the input and output datasets. Then, Monte Carlo sampling is performed on the probability distribution of the input variables in Table 1, and the damage response is obtained through the trained neural network surrogate model. Multiple sets of corresponding damage responses are obtained through multiple samplings, and finally, the probability distribution of fatigue damage is obtained. The probability distribution of creep damage is as follows: The probability distribution of total damage is as follows: And the probability distribution of lifespan is Meanwhile, the average lifespan was found to be 364 cycles.

[0100] Step 104: Calculate the reliability based on the probability distribution of total damage to obtain the reliability under the design life that meets the strength requirements.

[0101] As an optional implementation, the reliability calculation based on the probability distribution of total damage to obtain the reliability under the design life that meets the strength requirements specifically includes:

[0102] The probability distribution of the damage threshold is determined based on the leaf test data. The formula for calculating the damage threshold is:

[0103] ;

[0104] in, For the lifespan of a certain test, The mean lifetime is the result of multiple tests under the same working conditions. The probability distribution of the damage threshold can be obtained by conducting multiple tests under the same working conditions.

[0105] Based on the probability distribution of total damage under the current operating condition of the blade, the reliability under the strength design requirements is calculated.

[0106] The reliability is calculated using the cumulative damage-damage threshold interference criterion, and the specific formula is as follows:

[0107] ;

[0108] in, P f It is the probability of failure. R It's about reliability. D crit It is a damage threshold related to material properties. D(N) The total damage to the blade under its current operating condition accumulates over time. Both are random variables, and reliability is the area of ​​intersection between them. Let Pr(·) be the limit state equation, and let Pr(·) denote the probability calculation.

[0109] In this embodiment of the application, the design life-failure probability curve and variable sensitivity analysis results obtained based on the probability distribution of total damage and the failure probability are as follows: Figure 7a and Figure 7bAs shown, the design life at 99% reliability is 336 cycles, which meets the strength design specifications. Compared to the aforementioned deterministic calculation result of 306 cycles, engineering practice often requires applying a large empirical coefficient to the deterministic result from a conservative perspective. However, the 336 cycles obtained using the probabilistic strength design method of this application corresponds to a reliability of 99%, which is considered quite reliable. It can be seen that the redundancy in lifespan provided by the method in this application is significantly less than that of the deterministic calculation method. According to... Figure 7b It can be seen that the most important sensitive variable is the chamfer radius, accounting for 30% of the sensitivity, followed by the rotation speed, accounting for 27.71%. This indicates that the fluctuation of these two variables has the greatest impact on the life and reliability results. The chamfer radius is affected by the machining accuracy. Controlling the machining accuracy and reducing the machining error can reduce the dispersion of the results.

[0110] Step 105: Determine whether the reliability meets the design requirements. If not, optimize the design of the geometric structure and material parameters with the goal of improving the reliability value until the reliability meets the design requirements.

[0111] As an optional implementation, the optimization design of the geometric structure and material parameters aims to improve the reliability value. The specific optimization design formula is as follows:

[0112] ;

[0113] In the formula, Design variables for blade geometry. For material properties, Let be the objective function. This represents the probability of failure. To enable, represents a constraint. For the remaining variables that are not optimized, For the first The maximum probability of failure allowed by each constraint.

[0114] In this embodiment of the application, the chamfer radius of the sensitive random variable is optimized. r This ensures that the system meets a certain level of reliability. R Under the premise of maximizing the design life N The specific formula is expressed as follows:

[0115] ;

[0116] The final chamfer was optimized from 2 mm to 2.15 mm, and the reliability at a design life of 300 cycles was improved from 99.17% to 99.89%.

[0117] Implementing steps 101 to 105 effectively overcomes the conservatism and redundancy of traditional deterministic strength design methods. By introducing multi-level verification of material-level and structural component-level life prediction models, the accuracy and reliability of life prediction for key blade components are improved. By considering the uncertainties of loads, geometry, and material parameters throughout the entire design process, a shift from deterministic design to a more scientific and rational probabilistic design is achieved. Finally, using the calculated reliability as a key indicator to guide structural optimization and material selection, the optimal balance between economy and safety in turbine blade design can be achieved while meeting strength requirements and design life reliability, effectively solving the problem of excessive conservatism caused by traditional methods relying on large safety factors.

[0118] Based on the same inventive concept, this application also provides a turbine blade probabilistic strength design optimization device for implementing the aforementioned turbine blade probabilistic strength design optimization method. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the turbine blade probabilistic strength design optimization device provided below can be found in the limitations of the turbine blade probabilistic strength design optimization method described above, and will not be repeated here.

[0119] In one exemplary embodiment, such as Figure 8 As shown, a turbine blade probabilistic strength design optimization device is provided, comprising:

[0120] The turbine blade probabilistic strength design optimization device includes:

[0121] The parameter selection unit 201 is used to obtain the preliminary blade geometry and preliminarily select material parameters based on strength requirements and operating load conditions.

[0122] The first calculation unit 202 is used to take the probability distribution of load conditions, geometric structure, and material parameters as input variables of the finite element method, perform finite element calculations, and obtain stress-strain response.

[0123] The second calculation unit 203 is used to design structural components based on the structural characteristics of the dangerous parts of the blade, to conduct multi-level verification of the accuracy of the life prediction model at the material level and the structural component level, to calculate the total damage of the blade based on the multi-level verified life prediction model, and to obtain the probability distribution of the total damage.

[0124] The third calculation unit 204 is used to calculate the reliability based on the probability distribution of total damage, and obtain the reliability under the design life that meets the strength requirements.

[0125] The optimization design unit 205 is used to determine whether the reliability meets the design requirements. If it does not, the geometric structure and material parameters are optimized with the goal of improving the reliability value until the reliability meets the design requirements.

[0126] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 9 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores video tag processing data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When executed by the processor, the computer program implements a probabilistic strength design optimization method for turbine blades.

[0127] 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 present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0128] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0129] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0130] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0131] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0132] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0133] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0134] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0135] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for probabilistic strength design optimization of turbine blades, characterized in that, The probabilistic strength design optimization method for turbine blades includes: Based on strength requirements and operating load conditions, a preliminary blade geometry was obtained and material parameters were initially selected. The probability distributions of load conditions, geometry, material parameters, and model parameters are used as input variables for finite element analysis to obtain stress-strain response. Based on the structural characteristics of the dangerous parts of the blade, structural components are designed. The accuracy of the life prediction model is verified at the material level and the structural component level. Based on the life prediction model verified at the multi-level level, the total damage of the blade is calculated and the probability distribution of the total damage is obtained. The reliability is calculated based on the probability distribution of total damage, and the reliability under the design life that meets the strength requirements is obtained. Determine whether the reliability meets the design requirements. If not, optimize the design of the geometric structure and material parameters with the goal of improving the reliability value until the reliability meets the design requirements. The multi-level verification of the accuracy of the lifetime prediction model at the material and structural component levels specifically includes: Fatigue damage is obtained through a fatigue life calculation model; Creep damage is obtained through a creep life calculation model; The fatigue damage and the creep damage are added together to obtain the total damage; The predicted blade life is obtained by taking the reciprocal of the total damage. The blade's test life is obtained through mechanical property testing at the material level and structural component level, and the predicted blade life is verified by the blade test life. The multi-level validation-based lifetime prediction model calculates the total blade damage and obtains the probability distribution of the total damage, specifically including: Fatigue damage and creep damage are calculated based on the stress-strain response, and the input dataset and output dataset are accumulated after batch calculations. Train a neural network agent model using the input and output datasets; Sampling is performed on the input variables, and the damage response is obtained through the trained neural network surrogate model. Multiple sets of corresponding damage responses are obtained through multiple samplings, and finally the probability distributions of fatigue damage, creep damage, and total damage are obtained.

2. The turbine blade probabilistic strength design optimization method according to claim 1, characterized in that, The process of using the probability distributions of load conditions, geometric structures, material parameters, and model parameters as inputs for finite element analysis to obtain stress-strain response specifically includes: The load conditions are obtained by using historical data of blade operation, the probability distribution of geometric structure is obtained by multiple measurements, and the probability distribution of material parameters and model parameters is obtained by multiple parallel tests under the same load conditions. Random sampling is performed based on the probability distribution of the input variables. Each set of sampling results is input into the finite element calculation to obtain multiple sets of stress-strain responses.

3. The turbine blade probabilistic strength design optimization method according to claim 1, characterized in that, The reliability calculation based on the probability distribution of total damage, yielding the reliability under the design life that meets the strength requirements, specifically includes: The probability distribution of the damage threshold is determined based on the leaf test data. The formula for calculating the damage threshold is: ； In the formula, For the lifespan of a certain test, The mean lifetime is the result of multiple tests under the same working conditions. The probability distribution of the damage threshold can be obtained by conducting multiple tests under the same working conditions. Based on the probability distribution of total damage under the current operating condition of the blade, the reliability under the strength design requirements is calculated. The reliability is calculated using the cumulative damage-damage threshold interference criterion, and the specific formula is as follows: ； In the formula, P f It is the probability of failure. R It's about reliability. D crit It is a damage threshold related to material properties. D(N) The total damage to the blade under its current operating condition accumulates over time. Both are random variables, and reliability is the area of ​​intersection between them. Let Pr(·) be the limit state equation, and let Pr(·) denote the probability calculation.

4. The turbine blade probabilistic strength design optimization method according to claim 1, characterized in that, The optimization design of geometric structure and material parameters with the goal of improving reliability is specifically formulated as follows: ； In the formula, Design variables for blade geometry. For material properties, Let be the objective function. This represents the probability of failure. For the remaining variables that are not optimized, For the first The maximum probability of failure allowed by each constraint.

5. A turbine blade probabilistic strength design optimization device, characterized in that, The turbine blade probabilistic strength design optimization device includes: The parameter selection unit is used to obtain the preliminary blade geometry and preliminarily select material parameters based on strength requirements and operating load conditions. The first calculation unit is used to take the probability distribution of load conditions, geometric structure, and material parameters as input variables for finite element calculation, and to obtain stress-strain response. The second calculation unit is used to design structural components based on the structural characteristics of the dangerous parts of the blade, to conduct multi-level verification of the accuracy of the life prediction model at the material level and the structural component level, and to calculate the total damage of the blade based on the multi-level verified life prediction model to obtain the probability distribution of the total damage. The multi-level verification of the accuracy of the lifetime prediction model at the material and structural component levels specifically includes: Fatigue damage is obtained through a fatigue life calculation model; Creep damage is obtained through a creep life calculation model; The fatigue damage and the creep damage are added together to obtain the total damage; The predicted blade life is obtained by taking the reciprocal of the total damage. The blade's test life is obtained through mechanical property testing at the material level and structural component level, and the predicted blade life is verified by the blade test life. The multi-level validation-based lifetime prediction model calculates the total blade damage and obtains the probability distribution of the total damage, specifically including: Fatigue damage and creep damage are calculated based on the stress-strain response, and the input dataset and output dataset are accumulated after batch calculations. Train a neural network agent model using the input and output datasets; Sampling is performed on the input variables, and the damage response is obtained through the trained neural network surrogate model. Multiple sets of corresponding damage responses are obtained through multiple sets of sampling, and finally the probability distributions of fatigue damage, creep damage, and total damage are obtained. The third calculation unit is used to calculate the reliability based on the probability distribution of total damage, and obtain the reliability under the design life that meets the strength requirements; The optimization design unit is used to determine whether the reliability meets the design requirements. If not, the geometric structure and material parameters are optimized with the goal of improving the reliability value until the reliability meets the design requirements.

6. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the turbine blade probabilistic strength design optimization method according to any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the turbine blade probabilistic strength design optimization method as described in any one of claims 1-4.

8. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the turbine blade probabilistic strength design optimization method as described in any one of claims 1-4.

Images