Combustion chamber stress and strain calculation method, device, equipment and medium

Abstract

The invention discloses a combustion chamber stress and strain calculation method, device and equipment and a medium, and relates to the field of rapid stress and strain prediction.The method comprises the steps that the range of input load parameters is determined according to the actual service working condition of a combustion chamber, a plurality of working condition points are generated, finite element simulation is conducted on each working condition point, and a simulation result is obtained; obtaining full-order model stress-strain field data of the combustion chamber; performing intrinsic orthogonal decomposition on the stress-strain field data of the full-order model, and constructing a reduced-order model of a stress field and a strain field; establishing a regression model for reflecting a regression mapping relation from the input load parameters to the coefficient matrix of the reduced-order model; fusing the reduced-order model and the regression model to form a stress field reduced-order agent model and a strain field reduced-order agent model; input load parameters monitored in real time are input into the stress field reduced-order agent model and the strain field reduced-order agent model respectively, the stress field and the strain field of the combustion chamber are obtained through real-time calculation, and real-time and rapid prediction of stress and strain is achieved.

Description

Technical Field

[0001] This application relates to the field of stress and strain prediction, and in particular to a method, apparatus, equipment and medium for calculating stress and strain in a combustion chamber. Background Technology

[0002] The combustor, a critical hot-end component of an aero-engine, is located between the compressor and turbine. Its function is to mix and burn high-pressure air with fuel, converting the fuel's chemical energy into thermal energy to form high-temperature, high-pressure combustion gas. During service, the combustor is subjected to the thermal shock of high-temperature airflow and the effects of overall engine vibration, resulting in a relatively high probability of failure under the interaction of thermal stress, creep stress, and vibration stress. Furthermore, the complex configuration of the combustor and the presence of geometric discontinuities cause stress concentration. As a key component of the hot-end components of a gas turbine, the combustor plays a crucial role in the performance stability and safe operation throughout the gas turbine's life cycle. Therefore, achieving stress and strain prediction of the combustor is of great significance for improving its safe and reliable operation during service, enhancing the economy of operation and maintenance management, and reducing the failure rate.

[0003] Current research on combustion chambers has proposed a method for predicting the ultimate pressure resistance of the combustion chamber casing. This method establishes a finite element model of the combustion chamber casing, sets material parameters, applies loads and constraints for large deformation finite element analysis, uses the arc-length method to calculate the casing's deformation response, extracts the curve of the internal pressure of the combustion chamber casing as a function of radial deformation, and determines the ultimate pressure resistance value of the combustion chamber casing. However, this technique focuses on predicting the pressure resistance of the combustion chamber casing and does not address the stress and strain prediction of the combustion chamber itself. Related technologies have also proposed a method, apparatus, and equipment for predicting the life of the combustion chamber flame tube. This method applies different calculation weights to ablation life prediction models, creep life prediction models, low-cycle fatigue life prediction models, and thermal shock life prediction models to establish a flame tube life prediction model. The flame tube life is then predicted using this model. However, this technique focuses on the assessment and prediction of the combustion chamber life; stress and strain, as important conditions for the driving life model, are not mentioned. Related technologies have proposed a method for predicting the life of aero-engine combustors. However, none of the above solutions mention a method for predicting combustor stress and strain, especially a method for rapid prediction of stress and strain. Using finite element simulation of combustor stress and strain data requires large computational resources and is time-consuming. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, equipment, and medium for calculating stress and strain in a combustion chamber, which can realize real-time and rapid prediction of stress and strain.

[0005] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a method for calculating combustion chamber stress and strain, including: The range of input load parameters is determined based on the actual service conditions of the combustion chamber. Multiple operating points are generated using experimental design methods, and finite element simulation is performed on each operating point to obtain the full-order model stress field data and full-order model strain field data of the combustion chamber. The stress field data and strain field data of the full-order model are subjected to intrinsic orthogonal decomposition to construct a reduced-order model of the stress field and strain field; the reduced-order model represents the full-order model data as a combination of low-dimensional eigenvectors and coefficient matrices. Establish a regression model that reflects the regression mapping relationship between the input load parameters and the coefficient matrix of the reduced-order model; By integrating the reduced-order model and the regression model, a stress field reduced-order proxy model and a strain field reduced-order proxy model are formed that can output predicted values ​​of stress field and strain field respectively based on input load parameters. The real-time monitored input load parameters are input into the stress field reduced-order surrogate model and the strain field reduced-order surrogate model respectively, and the stress field and strain field of the combustion chamber are calculated in real time.

[0006] Optionally, the experimental design method is a central composite design method, a Latin hypercube sampling method, or an orthogonal experimental design method.

[0007] Optionally, finite element simulation is performed at each operating point to obtain full-order model stress field data and full-order model strain field data of the combustion chamber, specifically including: Combustion flow field calculations were performed at each operating point to obtain the temperature field data of the fluid in the combustion chamber; The fluid temperature field data is mapped to a finite element solid domain mesh to obtain fluid-structure interaction data. Steady-state thermal analysis was performed on the fluid-structure interaction data at each operating point to obtain the steady-state thermal analysis results. The steady-state thermal analysis results for each operating point were added to the solid domain for stress-strain finite element calculations to obtain the full-order model stress field data and full-order model strain field data of the combustion chamber.

[0008] Optionally, intrinsic orthogonal decomposition is performed on the full-order model stress field data and the full-order model strain field data to construct reduced-order models of the stress field and strain field, specifically including: The full-order model stress field data and the full-order model strain field data are formed Matrix A; This indicates the number of nodes in the finite element model of the combustion chamber. Indicates the number of operating points; The singular value decomposition algorithm is used to decompose matrix A into a U matrix, an S matrix, and a V matrix; the S matrix is ​​the singular value matrix, which is composed of the singular values ​​of matrix A. The diagonal matrix formed by the matrix A; matrix U is a left singular matrix, and matrix V is a right singular matrix, with matrices U and V each formed by the left singular vectors of matrix A. Right singular vector composition; The first k singular values ​​and their corresponding left and right singular vectors are selected to approximate the description matrix, and a reduced-order model of the stress field and strain field is constructed; k is a positive integer.

[0009] Alternatively, the price reduction model can be represented as follows: ; ; in, Let A be a matrix; It is an m-row, m-column left singular matrix (eigenvector matrix). It is an m x m singular value matrix; A right singular matrix with m rows and n columns The transpose of the matrix; The eigenvector matrix consists of m rows and k columns; This is the singular value matrix corresponding to the first k order data; The right singular matrix corresponding to the first k order data The transpose of the matrix; This is a coefficient matrix, which represents the weight of different modes under the corresponding operating conditions; The reconstruction matrix using the first k order data; It is an m-row, n-column eigenvector matrix; It is an n x n singular value matrix; A right singular matrix with n rows and n columns The transpose of .

[0010] Optionally, the reduced-order model and the regression model are integrated to form a stress field reduced-order surrogate model and a strain field reduced-order surrogate model that can output predicted values ​​of stress field and strain field respectively based on input load parameters, specifically including: Multiplying the coefficient matrix of the regression model for different operating conditions with the eigenvector matrix yields a stress field reduction surrogate model and a strain field reduction surrogate model that can output the predicted values ​​of stress field and strain field respectively based on the input load parameters.

[0011] Alternatively, the stress field reduced-order surrogate model can be represented as follows: ; The reduced-order surrogate model for the strain field is represented as follows: ; in, T , F These are temperature and force, respectively, in the input load parameters; and These are the stress field and strain field corresponding to the input parameters of the surrogate model, respectively; and These represent the mapping relationships between input load parameters and the stress and strain fields of the combustion chamber, respectively.

[0012] Secondly, this application provides a combustion chamber stress and strain calculation device, comprising: The finite element simulation module is used to determine the range of input load parameters based on the actual service conditions of the combustion chamber, generate multiple operating points using experimental design methods, and perform finite element simulation on each operating point to obtain full-order model stress field data and full-order model strain field data of the combustion chamber. The order reduction model construction module is used to perform intrinsic orthogonal decomposition on the stress field data and strain field data of the full-order model to construct an order reduction model of the stress field and strain field; the order reduction model represents the full-order model data as a combination of low-dimensional eigenvectors and coefficient matrices; The regression model building module is used to build a regression model that reflects the regression mapping relationship between the input load parameters and the coefficient matrix of the reduced-order model; The proxy model construction module is used to integrate the reduced-order model and the regression model to form a stress field reduced-order proxy model and a strain field reduced-order proxy model that can output the predicted values ​​of stress field and strain field respectively according to the input load parameters. The real-time stress and strain field calculation module is used to input the real-time monitored input load parameters into the stress field reduced-order surrogate model and the strain field reduced-order surrogate model respectively, and calculate the stress and strain fields of the combustion chamber in real time.

[0013] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the above-described method for calculating combustion chamber stress and strain.

[0014] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for calculating combustion chamber stress and strain.

[0015] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a method, apparatus, equipment, and medium for calculating stress and strain in a combustion chamber. The method determines the range of input load parameters based on the actual service conditions of the combustion chamber, generates multiple operating points, and performs finite element simulation on each operating point to obtain full-order model stress field data and full-order model strain field data for the combustion chamber. Intrinsic orthogonal decomposition is performed on the full-order model stress field data and full-order model strain field data to construct reduced-order models of the stress and strain fields. A regression model is established to reflect the regression mapping relationship between the input load parameters and the coefficient matrix of the reduced-order model. The reduced-order model and the regression model are fused to form a reduced-order surrogate model for the stress field and a reduced-order surrogate model for the strain field. The real-time monitored input load parameters are input into the reduced-order surrogate models for the stress and strain fields respectively, and the stress and strain fields of the combustion chamber are calculated in real time, achieving real-time and rapid prediction of stress and strain. By employing the order reduction method of intrinsic orthogonal decomposition, a reduced-order model of the stress-strain field is established. While retaining its key information and main influences, this method can significantly reduce computation time and storage requirements. This application proposes a modeling method for a reduced-order model of combustion chamber stress-strain prediction. A proxy model of the combustion chamber stress-strain field is established through singular value decomposition and regression techniques. Using operating load monitoring data, the stress and strain of the combustion chamber can be calculated in real time. This solves the problem that finite element calculation of combustion chamber stress and strain requires large computational resources and is time-consuming, which cannot meet the real-time calculation requirements of digital twin models. Attached Figure Description

[0016] 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.

[0017] Figure 1 This is an application environment diagram of a combustion chamber stress and strain calculation method according to an embodiment of this application.

[0018] Figure 2 This is a flowchart illustrating a method for calculating the stress and strain of a combustion chamber, provided as an embodiment of this application.

[0019] Figure 3 This is a schematic diagram of a combustion chamber structure model provided in an embodiment of this application.

[0020] Figure 4 A schematic diagram of the modeling process for a rapid calculation model of combustion chamber stress and strain provided in an embodiment of this application.

[0021] Figure 5 A schematic diagram of the design process of a central composite experimental space provided in an embodiment of this application.

[0022] Figure 6 The combustion chamber outer wall temperature field under a certain design condition is provided in an embodiment of this application.

[0023] Figure 7 The steady-state thermal analysis results of the combustion chamber outer wall temperature field under a certain design condition are provided in an embodiment of this application.

[0024] Figure 8 The stress-strain field analysis results of the combustion chamber under a certain design condition provided in an embodiment of this application, wherein Figure 8 (a) in the figure represents the stress field analysis results; Figure 8 (b) in the figure represents the results of the strain field analysis.

[0025] Figure 9 The stress-strain field of a combustion chamber under a certain monitoring load is provided in an embodiment of this application, wherein... Figure 9 (a) in the figure represents the stress field result; Figure 9 (b) in the figure represents the strain field result.

[0026] Figure 10 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0027] 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.

[0028] 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.

[0029] Explanation of relevant terms: 1. Stress: When an object deforms due to external factors (force, humidity, temperature field changes, etc.), the internal forces generated between the different parts of the object are called stress. 2. Strain: The local relative deformation of an object under the action of external forces and non-uniform temperature fields. 3. Reduced-order model: A reduced-order model refers to a lower-order model obtained by using model aggregation to reduce the order of a state-space model expressed by a state-space method, or by ignoring higher-order terms in a model established by differential equations, difference equations, or time series analysis.

[0030] The combustion chamber stress and strain calculation method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on other servers. Terminal 102 can send requests to be processed to server 104. After receiving the request, server 104 determines the range of input load parameters based on the actual service conditions of the combustion chamber, generates multiple operating points, and performs finite element simulation on each operating point to obtain the full-order model stress-strain field data of the combustion chamber. It then performs intrinsic orthogonal decomposition on the full-order model stress-strain field data to construct reduced-order models of the stress and strain fields. A regression model is established reflecting the regression mapping relationship between the input load parameters and the coefficient matrix of the reduced-order model. The reduced-order model and the regression model are fused to form a reduced-order surrogate model for the stress field and a reduced-order surrogate model for the strain field. The real-time monitored input load parameters are input into the reduced-order surrogate models for the stress and strain fields respectively, and the stress and strain fields of the combustion chamber are calculated in real time. Server 104 can feed back the obtained video tags for the video to terminal 102. In addition, in some embodiments, the combustion chamber stress and strain calculation method can also be implemented by server 104 or terminal 102 alone. For example, terminal 102 can directly perform stress and strain calculation for the request to be processed, or server 104 can obtain the request to be processed from the data storage system and perform stress and strain calculation for the request to be processed.

[0031] The terminal 102 can be, but is not limited to, various desktop computers and laptops. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers, or it can be a cloud server.

[0032] In one exemplary embodiment, such as Figure 2 As shown, a method for calculating combustion chamber stress and strain 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, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 205.

[0033] Step 201: Determine the range of input load parameters based on the actual service conditions of the combustion chamber, generate multiple operating points using experimental design methods, and perform finite element simulation on each operating point to obtain the full-order model stress field data and full-order model strain field data of the combustion chamber.

[0034] Step 202: Perform eigenorthogonal decomposition on the full-order model stress field data and the full-order model strain field data to construct a reduced-order model of the stress field and strain field; the reduced-order model represents the full-order model data as a combination of low-dimensional eigenvectors and coefficient matrices.

[0035] Step 203: Establish a regression model that reflects the regression mapping relationship between the input load parameters and the coefficient matrix of the reduced-order model.

[0036] Step 204: Integrate the reduced-order model and the regression model to form a stress field reduced-order proxy model and a strain field reduced-order proxy model that can output predicted values ​​of stress field and strain field respectively based on the input load parameters.

[0037] Step 205: Input the real-time monitored input load parameters into the stress field reduced-order surrogate model and the strain field reduced-order surrogate model respectively, and calculate the stress field and strain field of the combustion chamber in real time.

[0038] Finite element method (FEM) simulation of combustor stress and strain data requires significant computational resources and is time-consuming. Therefore, a reduced-order model for stress and strain calculation is needed to quickly obtain combustor stress and strain field data. By implementing steps 201 to 205 above, a reduced-order model of the stress and strain field is established using the intrinsic orthogonal decomposition method. This significantly reduces computation time and storage requirements while retaining key information and major influences. This application proposes a modeling method for reducing the order of combustor stress and strain prediction. A proxy model of the combustor stress and strain field is established using singular value decomposition and regression techniques. Based on load monitoring data, the stress and strain of the combustor can be calculated in real time, solving the problem that finite element method calculations of combustor stress and strain require large computational resources and are time-consuming, failing to meet the real-time calculation requirements of digital twin models. Real-time monitoring of combustor load conditions enables real-time prediction of stress and strain, which will play a crucial role in the design and operation of combustors.

[0039] Step 1. Through experimental design, select an appropriate sampling method to determine the operating points and clarify the input load of each operating point, so as to achieve the equivalent results of a large number of comprehensive experiments with the fewest number of experiments. The purpose is to obtain the most representative training data.

[0040] Step 2. Based on the loads at each working point determined in Step 1, perform finite element simulation on the combustion chamber to obtain stress and strain results under different working conditions.

[0041] Step 3. Based on the stress-strain field calculated in Step 2, establish a reduced-order model for stress-strain calculation. While retaining the key information and main influences of the model, this model can significantly reduce computation time and storage requirements to meet the real-time requirements of subsequent digital twin model establishment.

[0042] Step 4. Based on the stress-strain data under different load conditions in Step 2, establish a regression model to obtain the mapping relationship between the load and the stress-strain field. Based on this mapping relationship, stress-strain data under any load can be obtained, meeting the requirements of speed and real-time performance.

[0043] Step 5. Combine the reduced-order model from Step 3 and the regression model from Step 4 to establish a proxy model for the stress and strain fields of the combustion chamber.

[0044] Step 6. Calculate the stress-strain field data under the load value based on the real-time monitored load and the proxy model in Step 5.

[0045] Based on data reduction and data fitting regression techniques, this application establishes a proxy model for the stress and strain fields of the combustion chamber, enabling rapid calculation of combustion chamber stress and strain under different operating conditions.

[0046] This embodiment uses a simulated bottom component of the combustion chamber as an example. The specific implementation steps are as follows: Based on the actual service conditions of the combustion chamber, the fuel inlet velocity range is considered to be 32-48 m / s; the main air inlet velocity range is 8-12 m / s; and the auxiliary air inlet velocity range is 4.8-7.2 m / s. The structural model of the combustion chamber in this case is as follows: Figure 3 As shown. The modeling process for the rapid calculation model of combustion chamber stress and strain is as follows. Figure 4 As shown.

[0047] First, the operating points are obtained through orthogonal experimental design. Experimental design methods can include central composite design (CCD), Latin hypercube sampling, or orthogonal experimental design, etc.

[0048] Within the actual operating conditions of the combustion chamber, conduct DOE experimental design (not limited to orthogonal, Latin square, etc.). The experimental conditions are not limited to the temperature and load (mechanical, aerodynamic, pressure load, etc.) of the combustion chamber.

[0049] In this embodiment, the experimental space design adopts a central composite design sampling method, setting upper and lower limits for input parameters to solve for experimental design points within the design space. Different combinations of each design parameter constitute orthogonal experimental design schemes, as shown in Table 1. A schematic diagram of the experimental space for the central composite design is shown below. Figure 5 As shown.

[0050] In this embodiment, based on the actual operating conditions of the combustion chamber during service, an orthogonal experimental design was conducted with the main fuel inlet, main air inlet, and secondary air inlet as input variables, and stress and strain as output variables. Experimental design points were automatically generated within the design space, resulting in 15 experimental schemes, as shown in Table 2.

[0051] Table 1 Experimental Design Scheme

[0052] Table 2 Experimental Design Table

[0053] Training data is obtained through simulation experiments. Using the orthogonal experimental design scheme obtained in the above steps, fluid-structure-thermal coupled finite element simulation analysis is performed on the combustion chamber to obtain the stress and strain fields under various operating conditions, forming a simulation database. This includes, but is not limited to, using finite element software such as ANSYS, Abaqus, and Comsol to perform the aforementioned stress and strain calculations.

[0054] In step 201 above, finite element simulation is performed for each operating point to obtain the full-order model stress field data and full-order model strain field data of the combustion chamber. Specifically, this includes: calculating the combustion flow field for each operating point to obtain the temperature field data of the fluid in the combustion chamber; mapping the fluid temperature field data to a finite element solid domain mesh to obtain fluid-structure interaction data; performing steady-state thermal analysis calculations on the fluid-structure interaction data for each operating point to obtain steady-state thermal analysis results; and adding the steady-state thermal analysis results for each operating point to the solid domain for stress-strain finite element calculations to obtain the full-order model stress field data and full-order model strain field data of the combustion chamber.

[0055] (1) Combustion and thermal model calculation: In this embodiment, a vortex dissipation model is used, but not limited to, to simulate the combustion process of methane and oxygen in a canister burner. Other boundary conditions are as follows: the temperature of the fuel and air inlet is 300K; the mass fraction of methane is 1.0; the mass fraction of oxygen is 0.232; and the relative pressure at the outlet is 0MPa. By changing the three input variables in the experimental space design, the experimental design conditions are simulated respectively, and the wall temperatures of the inner wall, outer wall, and auxiliary air inlet are obtained respectively. Figure 6 The figure shows the temperature field on the outer wall of the combustion chamber.

[0056] (2) Data mapping: The fluid-to-solid mapping in this embodiment is completed through the Ansys mechanical module, including but not limited to the direct assignment method, triangulation method, distance-based averaging method, etc. The temperature field of the fluid calculated by step 1 is mapped to the finite element solid domain mesh through data mapping.

[0057] (3) Steady-state thermal calculation: The temperature fields generated by fluid calculations under various operating conditions in the experimental space are imported into the solid domain for steady-state thermal analysis. In this embodiment, the steady-state thermal calculation is performed using the Ansys Steady-State Thermal module. The steady-state thermal analysis results are as follows: Figure 7 As shown. The governing equation for steady-state heat conduction can be expressed as: (1); in, , , It refers to the thermal conductivity of a material along different directions. It is the intensity of the internal heat source. It is the heat generation rate per unit volume. This is a material density, steady-state temperature field problem that is independent of time. Using the Galerkin method with the equivalent integral form of the equations, an equivalent variational principle can be established. After discretization using the finite element method with interpolation functions, the stationary values ​​of the functionals can be directly obtained to solve the finite element equations. Finally, the system of linear algebraic equations that all nodal temperatures must satisfy is obtained: (2); in, It is the global heat conduction matrix. It is an unknown node temperature vector. The temperature value at the discrete node is obtained by solving the system of equations shown in equation (2).

[0058] The temperature field after steady-state thermal analysis under various working conditions in the experimental space was added to the solid domain for stress-strain finite element calculation.

[0059] Finite element method (FEM) discretizes continuous structures, using nodal displacements as the fundamental unknowns. By solving a large-scale system of linear equations, strain and stress are ultimately calculated based on the displacement results. The structural discretization divides the continuum into a finite number of elements connected by nodes. Taking the displacement method as an example, based on the principle of minimum potential energy or virtual work, the overall stiffness equation is established: (3); in, Let be the stiffness matrix of the nodes. Given the force vectors at the nodes, solve for the nodal displacement vectors using boundary conditions. .

[0060] Nodal displacements via strain matrix Calculate element strain: (4); in, For element strain; For nodal displacement.

[0061] According to the generalized Hooke's law, stress is calculated from strain. : (5); in This is the elasticity matrix. In this embodiment, stress-strain calculations are performed using the Ansys Static-Structural module. The stress-strain analysis results are as follows: Figure 8 As shown.

[0062] Based on the designed 15 orthogonal test schemes, the stress and strain fields under various working conditions were obtained through the above numerical simulation process, forming a simulation database. Table 3 shows the stress field data for all working conditions, and the strain field data is in a similar format.

[0063] Table 3 Stress field data (MPa)

[0064] This application employs Singular Value Decomposition (SVD) to achieve Proper Orthogonal Decomposition (POD), reducing the order of the full-order model stress field and strain field data from step 201. Proper Orthogonal Decomposition is a mathematical method for extracting feature information from discrete data. It approximates multidimensional stochastic processes and extracts the essential characteristics of complex systems using low-dimensional approximations. The principle of SVD is to transform a high-dimensional vector linearly to find the main components in the data, projecting a set of eigenvectors into a low-order vector space, thereby reducing the dimensionality of the data. This eigenvector retains the main features of the original high-order vector, and can be used to reconstruct the corresponding original high-order vector.

[0065] In step 202 above, the stress field data and strain field data of the full-order model are subjected to intrinsic orthogonal decomposition to construct a reduced-order model of the stress field and strain field, specifically including the following steps 301 to 303.

[0066] Step 301: Form the full-order model stress field data and the full-order model strain field data. Matrix A; This indicates the number of nodes in the finite element model of the combustion chamber. This indicates the number of operating points.

[0067] Step 302: Decompose matrix A into matrix U, matrix S, and matrix V using the singular value decomposition algorithm; matrix S is the singular value matrix, which is composed of the singular values ​​of matrix A. The diagonal matrix formed by the matrix A; matrix U is a left singular matrix, and matrix V is a right singular matrix, with matrices U and V each formed by the left singular vectors of matrix A. Right singular vector composition.

[0068] Step 303: Select the first k singular values ​​and their corresponding left and right singular vectors to approximate the description matrix, and construct a reduced-order model of the stress field and strain field; k is a positive integer.

[0069] First, organize the full-order model stress field data and full-order model strain field data obtained in step 201 from the combustion chamber model simulation database into... A matrix A is defined such that each column of matrix A corresponds to the calculated stress field (strain field) under different loading conditions, and the number of rows m represents the number of nodes in the combustion chamber finite element model. The number of nodes is determined by the type of solid domain mesh.

[0070] Singular value decomposition decomposes matrix A into matrices U, S, and V, where S is the singular value matrix, which is composed of the singular values ​​of A. The diagonal matrix formed by A, where U is the left singular matrix and V is the right singular matrix, are composed of the left and right singular vectors of A, respectively. , It is composed of a unitary matrix.

[0071] In many cases, the sum of the first 10% or even 1% of singular values ​​accounts for more than 99% of the total sum of singular values. The matrix can be approximated by the first k singular values ​​and their corresponding left and right singular vectors. The SVD method only needs to extract the first few eigenvectors as the reduced-order modes to represent the original data features to the greatest extent. Matrix A is approximately decomposed into an eigenvector matrix U and a coefficient matrix C. The price reduction model obtained by SVD decomposition is shown in formulas (6) and (7).

[0072] (6); (7); in, Let A be a matrix; It is an m-row, m-column left singular matrix (eigenvector matrix). It is an m x m singular value matrix; A right singular matrix with m rows and n columns The transpose of the matrix; The eigenvector matrix consists of m rows and k columns; This is the singular value matrix corresponding to the first k order data; The right singular matrix corresponding to the first k order data The transpose of the matrix; This is a coefficient matrix, which represents the weight of different modes under the corresponding operating conditions; The reconstruction matrix using the first k order data; It is an m-row, n-column eigenvector matrix; It is an n x n singular value matrix; A right singular matrix with n rows and n columns The transpose of .

[0073] In this embodiment, the full-order model stress field data from step 201 is decomposed using SVD to obtain three matrices: U, S, and V. S is the singular value matrix, and U and V are the left and right singular vectors, respectively. The first 9 singular values ​​(other values ​​can be chosen, with no limitation; higher values ​​result in higher accuracy) and their corresponding left and right singular vectors are selected to approximate the matrices. This means that the first 9 orders of U, S, and V are extracted and named the first 9 order eigenvector matrix U_, the first 9 order singular value matrix S_, and the first 9 order right singular matrix, respectively. V_ and reconstruct the stress matrix. When K=9, the reduced-order model is as follows: (8); This approximates the decomposition of the original full-order model stress field data into an eigenvector matrix. and a coefficient matrix C=S_ V_ T Meanwhile, the original large 15×328960 matrix is ​​simplified into a small 15×9 matrix multiplied by a fixed coefficient matrix C, greatly reducing storage requirements. Table 4 shows the reconstructed matrix of the stress field. The reduction of the strain field matrix is ​​similar to that of the stress field.

[0074] Table 4. Stress field data (MPa) of the first 9 order reduced-order models.

[0075] A proxy model of the stress and strain fields is constructed to enable real-time calculation of stress and strain under different input conditions.

[0076] Step 204 above integrates the reduced-order model and the regression model to form a stress field reduced-order proxy model and a strain field reduced-order proxy model that can output predicted values ​​of stress field and strain field respectively based on the input load parameters. Specifically, it includes multiplying the coefficient matrix of different working conditions predicted by the regression model with the eigenvector matrix to obtain the stress field reduced-order proxy model and the strain field reduced-order proxy model that can output predicted values ​​of stress field and strain field respectively based on the input load parameters.

[0077] A regression algorithm is used to fit the input load parameters (temperature, force) to a coefficient matrix, obtaining predicted values ​​for the coefficient matrix under different load parameters. Then, data reconstruction is used to predict the stress and strain fields under those load parameters. Multiplying the coefficient matrix fitted by the regression model with the eigenvector matrix yields the stress and strain fields under that operating condition. This results in a reduced-order surrogate model for the combustion chamber stress and strain fields. The stress field reduced-order surrogate model is represented as follows: (9); The reduced-order surrogate model for the strain field is represented as follows: (10); in, T , F These are temperature and force, respectively, in the input load parameters; and These are the stress field and strain field corresponding to the input parameters of the surrogate model, respectively; and These are the mapping relationships from input load parameters to the stress and strain fields of the combustion chamber, based on SVD decomposition and regression models, respectively.

[0078] In this embodiment, the function is trained using regression tools (not limited to the Matlab Regression Toolbox, etc.): (11); For new working conditions Make predictions. The number of columns represents the number of operating conditions, allowing for the prediction of one or more sets of operating conditions. The output data is the coefficient matrix predicted by the regression model.

[0079] Multiplying the coefficient matrix predicted by the regression model with the eigenvector matrix obtained from the singular value decomposition in step 202, a reduced-order surrogate model of the stress field is constructed, thus obtaining the predicted values ​​of the stress field under different working conditions. Similarly, a reduced-order surrogate model of the stress field can be obtained, thereby obtaining the predicted values ​​of the strain field.

[0080] Using the aforementioned rapid stress-strain reduced-order surrogate model for stress and strain fields, the stress and strain fields of the combustion chamber can be calculated in real time based on load monitoring data, and structural hazard points can be monitored in real time. The stress and strain fields of the combustion chamber under a certain input monitoring load are predicted using the stress-strain reduced-order surrogate model, such as... Figure 9 As shown.

[0081] This application also provides an application scenario in which the above-described combustion chamber stress and strain calculation method is applied. Specifically, the combustion chamber stress and strain calculation method provided in this embodiment can be applied in a combustion chamber stress and strain calculation scenario. The combustion chamber stress and strain calculation scenario includes a request issuance stage and a combustion chamber stress and strain calculation stage; the request to be processed enters the combustion chamber stress and strain calculation stage from the request issuance stage, and the stress field and strain field of the combustion chamber are obtained through human-machine collaboration. The combustion chamber stress and strain calculation method provided in this embodiment belongs to the combustion chamber stress and strain calculation stage. Specifically, in the process of calculating the stress and strain of the combustion chamber in response to the request, the range of input load parameters can be determined according to the actual service conditions of the combustion chamber, multiple operating points can be generated, and finite element simulation can be performed on each operating point to obtain the stress and strain field data of the full-order model of the combustion chamber. The stress and strain field data of the full-order model can be decomposed into intrinsic orthogonal models to construct reduced-order models of stress and strain fields. A regression model reflecting the regression mapping relationship of the coefficient matrix from the input load parameters to the reduced-order model can be established. The reduced-order model and the regression model can be fused to form a reduced-order surrogate model of stress field and a reduced-order surrogate model of strain field. The real-time monitored input load parameters can be input into the reduced-order surrogate models of stress field and strain field respectively to calculate the stress and strain fields of the combustion chamber in real time.

[0082] Based on the same inventive concept, this application also provides a combustion chamber stress and strain calculation device for implementing the combustion chamber stress and strain calculation method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more combustion chamber stress and strain calculation device embodiments provided below can be found in the limitations of the combustion chamber stress and strain calculation method described above, and will not be repeated here.

[0083] In one exemplary embodiment, a combustion chamber stress and strain calculation device is provided, comprising: The finite element simulation module is used to determine the range of input load parameters based on the actual service conditions of the combustion chamber, generate multiple operating points using experimental design methods, and perform finite element simulation on each operating point to obtain full-order model stress field data and full-order model strain field data of the combustion chamber. The order reduction model construction module is used to perform intrinsic orthogonal decomposition on the stress field data and strain field data of the full-order model to construct an order reduction model of the stress field and strain field; the order reduction model represents the full-order model data as a combination of low-dimensional eigenvectors and coefficient matrices; The regression model building module is used to build a regression model that reflects the regression mapping relationship between the input load parameters and the coefficient matrix of the reduced-order model; The proxy model construction module is used to integrate the reduced-order model and the regression model to form a stress field reduced-order proxy model and a strain field reduced-order proxy model that can output the predicted values ​​of stress field and strain field respectively according to the input load parameters. The real-time stress and strain field calculation module is used to input the real-time monitored input load parameters into the stress field reduced-order surrogate model and the strain field reduced-order surrogate model respectively, and calculate the stress and strain fields of the combustion chamber in real time.

[0084] 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 10 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 operation of the operating system and computer programs stored in the non-volatile storage media. The database stores stress and strain calculation data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a method for calculating stress and strain in a combustion chamber.

[0085] Those skilled in the art will understand that Figure 10 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.

[0086] In one exemplary embodiment, a computer device is 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.

[0087] 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.

[0088] 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.

[0089] Those skilled in the art will understand that all or part of the processes in the methods of 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, and when executed, it can include the processes of the embodiments of the above methods. 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).

[0090] 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.

[0091] 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.

[0092] 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 combustion chamber stress and strain calculation method characterized by, The combustion chamber stress and strain calculation method comprises: determining the range of input load parameters according to the actual service conditions of the combustion chamber, generating a plurality of working condition points by using an experimental design method, and performing finite element simulation on each working condition point to obtain full-order model stress field data and full-order model strain field data of the combustion chamber; performing intrinsic orthogonal decomposition on the full-order model stress field data and the full-order model strain field data to construct a reduced-order model of the stress field and the strain field; the reduced-order model represents the full-order model data as a combination of low-dimensional eigenvectors and a coefficient matrix; establishing a regression model reflecting the regression mapping relationship from the input load parameters to the coefficient matrix of the reduced-order model; fusing the reduced-order model and the regression model to form a stress field reduced-order surrogate model and a strain field reduced-order surrogate model capable of respectively outputting stress field and strain field prediction values according to input load parameters; inputting real-time monitored input load parameters into the stress field reduced-order surrogate model and the strain field reduced-order surrogate model to obtain the stress field and the strain field of the combustion chamber in real time.

2. The combustion chamber stress and strain calculation method according to claim 1, characterized by, The experimental design method is a central composite design method, a Latin hypercube sampling method or an orthogonal experimental design method.

3. The combustion chamber stress and strain calculation method according to claim 1, characterized by, The finite element simulation on each working condition point to obtain full-order model stress field data and full-order model strain field data of the combustion chamber specifically comprises: performing combustion flow field calculation on each working condition point to obtain temperature field data of the fluid in the combustion chamber; performing data mapping from the temperature field data of the fluid to the finite element solid domain grid to obtain fluid-structure coupling data; performing steady-state thermal analysis calculation on the fluid-structure coupling data of each working condition point to obtain steady-state thermal analysis results; adding the steady-state thermal analysis results of each working condition point to the solid domain for stress and strain finite element calculation to obtain full-order model stress field data and full-order model strain field data of the combustion chamber.

4. The combustion chamber stress and strain calculation method according to claim 1, characterized by, The intrinsic orthogonal decomposition on the full-order model stress field data and the full-order model strain field data to construct a reduced-order model of the stress field and the strain field specifically comprises: The full-order model stress field data and the full-order model strain field data are formed Matrix A; This indicates the number of nodes in the finite element model of the combustion chamber. Indicates the number of operating points; The matrix A is decomposed into a U matrix, an S matrix and a V matrix by using a singular value decomposition algorithm; the S matrix is a singular value matrix composed of singular values of the matrix A ; the U matrix is a left singular matrix, and the V matrix is a right singular matrix; the U matrix and the V matrix are respectively composed of left singular vectors and right singular vectors of the matrix A ; the singular values of the matrix A are obtained by calculating singular values of the matrix A ; and the left singular vectors and the right singular vectors of the matrix A are obtained by calculating left singular vectors and right singular vectors of the matrix A. selecting the first k singular values and the corresponding left and right singular vectors to approximately describe the matrix to construct a reduced-order model of the stress field and the strain field; k is a positive integer.

5. The combustion chamber stress and strain calculation method according to claim 4, characterized by, The reduced-order model is expressed as follows: ； ； wherein, is a matrix A; is a left singular matrix (eigenvector matrix) of m rows and m columns; is a singular value matrix of m rows and m columns; is a right singular matrix of m rows and n columns is a transpose matrix of ; is an eigenvector matrix of m rows and k columns; is a singular value matrix corresponding to the first k orders of data; is a right singular matrix corresponding to the first k orders of data is a transpose matrix of ; is a coefficient matrix, the coefficient matrix representing weights of different modes under corresponding working conditions; is a reconstruction matrix using the first k orders of data; is an eigenvector matrix of m rows and n columns; is a singular value matrix of n rows and n columns; is a right singular matrix of n rows and n columns is a transpose matrix of.

6. The combustion chamber stress and strain calculation method according to claim 5, characterized by, The fusion of the reduced-order model and the regression model to form a stress field reduced-order surrogate model and a strain field reduced-order surrogate model capable of respectively outputting stress field and strain field prediction values according to input load parameters specifically comprises: multiplying the coefficient matrix and the eigenvector matrix of different working condition points predicted by the regression model to obtain a stress field reduced-order surrogate model and a strain field reduced-order surrogate model capable of respectively outputting stress field and strain field prediction values according to input load parameters.

7. The combustion chamber stress and strain calculation method according to claim 4, characterized by, The stress field reduced-order surrogate model is expressed as follows: ； The strain field reduced-order surrogate model is expressed as follows: ； wherein, T , F are the temperature and force in the input load parameters, respectively; and are the stress field and strain field corresponding to the proxy model input parameters, respectively; and are the mapping relationships of the input load parameters to the stress field and strain field of the combustion chamber, respectively.

8. A combustion chamber stress and strain calculation apparatus characterized by comprising: The combustion chamber stress and strain calculation device comprises: a finite element simulation module configured to determine the range of input load parameters according to the actual service conditions of the combustion chamber, generate a plurality of working condition points by using an experimental design method, and perform finite element simulation on each working condition point to obtain full-order model stress field data and full-order model strain field data of the combustion chamber; The reduced-order model construction module is configured to perform intrinsic orthogonal decomposition on the full-order model stress field data and the full-order model strain field data to construct reduced-order models of the stress field and the strain field; and the reduced-order model represents the full-order model data as a combination of a low-dimensional eigenvector and a coefficient matrix; The regression model establishment module is configured to establish a regression model reflecting a regression mapping relationship from the input load parameter to the coefficient matrix of the reduced-order model; The surrogate model construction module is configured to fuse the reduced-order model and the regression model to form a stress field reduced-order surrogate model and a strain field reduced-order surrogate model capable of respectively outputting predicted values of the stress field and the strain field according to the input load parameter; The stress field and strain field real-time calculation module is configured to input the real-time monitored input load parameter into the stress field reduced-order surrogate model and the strain field reduced-order surrogate model respectively to obtain the stress field and the strain field of the combustion chamber in real time.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the combustion chamber stress and strain calculation method of any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the combustion chamber stress and strain calculation method of any one of claims 1-7.

Images