Model parameter optimization method and device, equipment, storage medium and program product

Abstract

The invention relates to a model parameter optimization method and device, equipment, a storage medium and a program product, and the method comprises the steps: obtaining working condition information and a plurality of initial parameters corresponding to different air film cooling working conditions, inputting the working condition information and the plurality of initial parameters into a turbulence model, and carrying out the prediction. And finally, according to the error information of the parameters under the cooling working conditions of the air films, optimizing the initial parameters to obtain optimized model parameters. According to the method, the error information under each air film cooling working condition can be simulated and calculated according to a plurality of initial parameters, the influence of each closing coefficient of the turbulence model on the air film cooling calculation error under different blowing ratios can be systematically analyzed, and the turbulence model can be optimized and analyzed according to each error information; thus, the optimized turbulence model is obtained, and the prediction accuracy of the turbulence model can be improved.

Description

Technical Field

[0001] The present application relates to the field of air film cooling technology, and in particular to a method, device, equipment, storage medium and program product for optimizing model parameters. Background Art

[0002] Film cooling is an external cooling technology that uses the heat and mass transfer principle of a solid wall coolant thin film to achieve cooling and thermal protection effects. It can greatly reduce the surface temperature of solid metals, and is therefore widely used in gas turbines, aircraft engine combustion chambers, turbine blades, and other thermal protection-related engineering and technical fields. As the design conditions of gas turbines and aircraft engines are increasingly approaching the high temperature and high pressure limits of materials, the importance of film cooling technology is increasing, and the demand for improving its cooling efficiency is also increasing.

[0003] At present, the numerical modeling problem of film cooling is closely related to turbulence-related modeling. The Reynolds-Averaged Navier-Stokes turbulence model (RANS model for short) is usually used to predict the cooling efficiency of film cooling technology in order to guide the design of film cooling engineering.

[0004] However, in the related art, the RANS model is easily affected by various factors and has the problem of inaccurate prediction. Summary of the invention

[0005] Based on this, it is necessary to provide a method, device, equipment, storage medium and program product for optimizing model parameters that can improve prediction accuracy in response to the above technical problems.

[0006] In a first aspect, the present application provides a method for optimizing model parameters, the method comprising:

[0007] Obtaining operating condition information and multiple initial parameters corresponding to different film cooling operating conditions;

[0008] The operating condition information and multiple initial parameters are input into the turbulence model for prediction, and the error information of the parameters under each film cooling condition is obtained;

[0009] The initial parameters are optimized according to the error information of the parameters under various film cooling conditions to obtain the optimized model parameters.

[0010] In one embodiment, each initial parameter is optimized according to the error information of the parameter under each film cooling condition to obtain the optimized model parameters, including:

[0011] According to the error information of parameters under each film cooling condition and each initial parameter, the active subspace under each film cooling condition is determined;

[0012] The initial parameters of the active subspace under various film cooling conditions are optimized to obtain the optimized model parameters.

[0013] In one embodiment, the initial parameters of the active subspace under each film cooling condition are optimized to obtain optimized model parameters, including:

[0014] According to the active subspace under each film cooling condition, the parameters to be optimized under each film cooling condition are determined;

[0015] According to the active subspace and initial parameters under each film cooling condition, the parameter with the minimum error under each film cooling condition is determined;

[0016] According to the parameters to be optimized and the parameters with the smallest error under various film cooling conditions, the initial parameters are optimized to obtain the optimized model parameters.

[0017] In one embodiment, each initial parameter is optimized according to the parameters to be optimized and the parameters with the smallest error under each film cooling condition to obtain the optimized model parameters, including:

[0018] Determine the constraint relationship based on the parameters with the smallest error under each film cooling condition;

[0019] According to the value range and constraint relationship of the parameters to be optimized under various film cooling conditions, a constraint surface is constructed;

[0020] By solving the constraint surface, each initial parameter is optimized to obtain the optimized model parameters.

[0021] In one embodiment, each initial parameter is optimized by solving the constraint surface to obtain optimized model parameters, including:

[0022] Solve the constraint surface and obtain the solution result;

[0023] If the solution result indicates that there is a feasible domain, then the initial parameters are optimized according to the feasible domain of the constraint surface to obtain the optimized model parameters;

[0024] If the solution result indicates that there is no feasible domain, the multiple initial parameters are updated, and based on the updated initial parameters and the operating condition information corresponding to different film cooling conditions, the step of returning to execute the operating condition information and the multiple initial parameters is input into the turbulence model for prediction to obtain the error information of the parameters under each film cooling condition.

[0025] In one embodiment, each initial parameter is optimized according to the feasible domain of the constraint surface to obtain the optimized model parameters, including:

[0026] Determine whether the feasible domain of the constraint surface meets the preset error range;

[0027] If satisfied, the parameters included in the feasible domain are determined as optimized model parameters;

[0028] If it is not satisfied, multiple initial parameters are updated according to the parameters with the smallest error under each film cooling condition, and based on the updated initial parameters and the operating condition information corresponding to different film cooling conditions, return to execute the step of inputting the operating condition information and multiple initial parameters into the turbulence model for prediction to obtain the error information of the parameters under each film cooling condition.

[0029] In one embodiment, the method further comprises:

[0030] Determine whether there are at least two common parameters of the parameters to be optimized under different film cooling conditions;

[0031] If it exists, when constructing the constraint surface, executing the step of constructing the constraint surface according to the value range and constraint relationship of the parameter to be optimized including at least two common parameters;

[0032] If it does not exist, when constructing the constraint surface, the step of constructing the constraint surface is performed according to the value range and constraint relationship of the parameter to be optimized under each film cooling condition.

[0033] In one embodiment, according to the error information of the parameters under each film cooling condition and each initial parameter, the active subspace under each film cooling condition is determined, including:

[0034] According to the error information of parameters under various film cooling conditions and initial parameters, a linear response surface and covariance matrix are constructed;

[0035] The linear response surface is solved based on the covariance matrix to obtain the active subspace.

[0036] In a second aspect, the present application also provides a device for optimizing model parameters, the device comprising:

[0037] An acquisition module, used to acquire operating condition information and multiple initial parameters corresponding to different film cooling operating conditions;

[0038] The prediction module is used to input the working condition information and multiple initial parameters into the turbulence model for prediction, and obtain the error information of the parameters under each film cooling working condition;

[0039] The optimization module is used to optimize each initial parameter according to the error information of the parameters under each film cooling condition to obtain the optimized model parameters.

[0040] In a third aspect, the present application further provides a computer device, the computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program:

[0041] Obtaining operating condition information and multiple initial parameters corresponding to different film cooling operating conditions;

[0042] The operating condition information and multiple initial parameters are input into the turbulence model for prediction, and the error information of the parameters under each film cooling condition is obtained;

[0043] The initial parameters are optimized according to the error information of the parameters under various film cooling conditions to obtain the optimized model parameters.

[0044] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0045] Obtaining operating condition information and multiple initial parameters corresponding to different film cooling operating conditions;

[0046] The operating condition information and multiple initial parameters are input into the turbulence model for prediction, and the error information of the parameters under each film cooling condition is obtained;

[0047] The initial parameters are optimized according to the error information of the parameters under various film cooling conditions to obtain the optimized model parameters.

[0048] In a fifth aspect, the present application further provides a computer program product, the computer program product comprising a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0049] Obtaining operating condition information and multiple initial parameters corresponding to different film cooling operating conditions;

[0050] The operating condition information and multiple initial parameters are input into the turbulence model for prediction, and the error information of the parameters under each film cooling condition is obtained;

[0051] The initial parameters are optimized according to the error information of the parameters under various film cooling conditions to obtain the optimized model parameters.

[0052] The optimization method, device, equipment, storage medium and program product of the above-mentioned model parameters, the method obtains the working condition information and multiple initial parameters corresponding to different film cooling working conditions, and then inputs the working condition information and multiple initial parameters into the turbulence model for prediction, obtains the error information of the parameters under each film cooling working condition, and finally optimizes each initial parameter according to the error information of the parameters under each film cooling working condition to obtain the optimized model parameters. The above-mentioned method can simulate and calculate the error information under each film cooling working condition according to multiple initial parameters, can systematically analyze the influence of each closure coefficient of the turbulence model on the calculation error of film cooling under different blowing ratios, and can optimize and analyze the turbulence model according to each error information, thereby obtaining the optimized turbulence model, which can improve the prediction accuracy of the turbulence model. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is an internal structure diagram of a computer device in one embodiment;

[0054] Figure 2 A schematic diagram of a flow chart of a method for optimizing model parameters in one embodiment;

[0055] Figure 3 is a schematic flow chart of a method for optimizing model parameters in another embodiment;

[0056] Figure 4 is a schematic flow chart of a method for optimizing model parameters in another embodiment;

[0057] Figure 5 is a schematic flow chart of a method for optimizing model parameters in another embodiment;

[0058] Figure 6 is a schematic flow chart of a method for optimizing model parameters in another embodiment;

[0059] Figure 7 is a schematic flow chart of a method for optimizing model parameters in another embodiment;

[0060] Figure 8 is a schematic flow chart of a method for optimizing model parameters in another embodiment;

[0061] Fig. 9 is a schematic flow chart of a method for optimizing model parameters in another embodiment;

[0062] Fig.10 is a schematic flow chart of a method for optimizing model parameters in another embodiment;

[0063] Fig.11 is a schematic diagram of the structure of a turbulence model in an embodiment;

[0064] Fig.12A schematic diagram of one-dimensional active subspace coefficients in an embodiment;

[0065] Fig.13 A schematic diagram of a reconstructed response surface in one embodiment;

[0066] Fig.14 A schematic diagram of a solution domain for reconstructing a response surface in one embodiment;

[0067] Fig.15 A schematic diagram of verification results in one embodiment;

[0068] Fig.16 It is a structural block diagram of a device for optimizing model parameters in one embodiment. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0070] Film cooling is an external cooling technology that uses the heat and mass transfer principle of solid wall coolant thin film to achieve cooling and thermal protection effects. It can greatly reduce the surface temperature of solid metals, and is therefore widely used in gas turbines, combustion chambers of aircraft engines, turbine blades, and other engineering and technical fields related to thermal protection. As the design conditions of gas turbines and aircraft engines are increasingly approaching the high temperature and high pressure limits of materials, the importance of film cooling technology is becoming increasingly important, and the demand for improving its cooling efficiency is also increasing. At present, the numerical modeling problem of film cooling is closely related to turbulence-related modeling. The Reynolds-Averaged Navier-Stokes turbulence model (abbreviated as RANS model) is usually used to predict the cooling efficiency of film cooling technology in order to guide the design of film cooling engineering. However, in the related art, the RANS model is easily affected by various factors and has the problem of inaccurate prediction. The present application provides a method for optimizing model parameters, aiming to solve the above-mentioned technical problems. The following embodiments will specifically illustrate the optimization method of model parameters described in the present application.

[0071] The optimization method of model parameters provided in the embodiment of the present application can be applied to Figure 1 In the computer device shown in the figure, the computer device can be a terminal or a server, and its internal structure diagram can be as shown in FIG. Figure 1As shown, the computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for optimizing model parameters is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0072] Those skilled in the art will understand that Figure 1 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0073] In one embodiment, Figure 2 As shown in the figure, a method for optimizing model parameters is provided. Figure 1 The computer device in the example is used to illustrate, including the following steps:

[0074] S201, obtaining operating condition information and a plurality of initial parameters corresponding to different film cooling operating conditions.

[0075] Among them, the film cooling working conditions include working conditions corresponding to different blowing ratios, which can specifically include working conditions corresponding to high blowing ratios and working conditions corresponding to low blowing ratios. Working condition information includes mainstream Mach number , boundary layer thickness , mainstream turbulence intensity , turbulence integral characteristic length , mainstream and cold flow density ratio , Blowing ratio , low blowing ratio , high blowing ratio The multiple initial parameters are the closure coefficients of the turbulence model, which are sample input sets generated based on the object parameters of the turbulence model. , the turbulence model is the RANS model.

[0076] In the embodiment of the present application, the computer device can obtain the working condition information corresponding to different film cooling working conditions through the information acquisition device, and can generate a sample input set as multiple initial parameters based on the object parameters of the turbulence model. Specifically, N object parameters can be determined first, and then their corresponding parameter ranges can be set to , and finally randomly generate Sample input set , and obtain multiple initial parameters.

[0077] S202, inputting the operating condition information and a plurality of initial parameters into a turbulence model for prediction, and obtaining error information of the parameters under various film cooling operating conditions.

[0078] Among them, the turbulence model is the RANS model.

[0079] In the embodiment of the present application, after the computer device obtains the working condition information and multiple initial parameters corresponding to different film cooling working conditions based on the above steps, the working condition information and multiple initial parameters can be input into the turbulence model, and the corresponding solution of the parameters under each film cooling working condition can be obtained through the calculation of the turbulence model, and the solution is output as error information. Specifically, by using the sample input set Input into the turbulence model, and then calculate the turbulence model to obtain the corresponding solution (i.e. error information) for each sample input at different blowing ratios: and ,in, Indicates the error information at low blowing ratio, Indicates error information at high blowing ratio.

[0080] S203, optimizing each initial parameter according to the error information of the parameter under each film cooling working condition to obtain optimized model parameters.

[0081] Among them, the optimized model parameters are the optimized initial parameters.

[0082] In the embodiment of the present application, the computer device can set a preset error range according to the requirements in advance. (Error limit), for example, the error limit is set to 5%. After the computer device obtains the error information of the parameters under each film cooling condition based on the above steps, each initial parameter can be optimized according to the error information of the parameters under each film cooling condition and the preset error range to obtain the optimized model parameters. Specifically, each error information can be compared with the preset error range, and then each initial parameter can be optimized according to the comparison result to obtain the optimized model parameters.

[0083] The optimization method of model parameters provided in the embodiment of the present application obtains the working condition information and multiple initial parameters corresponding to different film cooling working conditions, and then inputs the working condition information and multiple initial parameters into the turbulence model for prediction, obtains the error information of the parameters under each film cooling working condition, and finally optimizes each initial parameter according to the error information of the parameters under each film cooling working condition to obtain the optimized model parameters. The above method can simulate and calculate the error information under each film cooling working condition according to multiple initial parameters, can systematically analyze the influence of each closure coefficient of the turbulence model on the calculation error of film cooling under different blowing ratios, and can optimize and analyze the turbulence model according to each error information, thereby obtaining the optimized turbulence model, which can improve the prediction accuracy of the turbulence model.

[0084] In one embodiment, a specific implementation method for optimizing each initial parameter according to the error information of the parameter under each film cooling condition is also provided, such as Figure 3 As shown, the above step S203 of "optimizing each initial parameter according to the error information of the parameter under each film cooling condition to obtain the optimized model parameter" includes:

[0085] S301, determining the active subspace under each film cooling working condition according to the error information of the parameters under each film cooling working condition and each initial parameter.

[0086] Among them, the active subspace is used to reduce the high-dimensional data to one-dimensional data.

[0087] In the embodiment of the present application, after the computer device obtains the error information and initial parameters of the parameters under each film cooling condition, the error information and initial parameters of the parameters under each film cooling condition can be input into a preset algorithm for algorithm solution to obtain the active subspace under each film cooling condition. Optionally, the error information and initial parameters of the parameters under each film cooling condition can also be input into a preset model for prediction, and the active subspace under each film cooling condition can be output.

[0088] Optional, such as Figure 4 As shown, the above step S301 of "determining the active subspace under each film cooling condition according to the error information of the parameters under each film cooling condition and each initial parameter" includes:

[0089] S401, constructing a linear response surface and a covariance matrix according to the error information of the parameters under each film cooling condition and each initial parameter.

[0090] S402, solving the linear response surface based on the covariance matrix to obtain an active subspace.

[0091] In the embodiment of the present application, for different blowing ratios, the computer device can fit the linear response surface according to the error information and the initial parameters. , , for each blowing ratio, the covariance matrix composed of gradients can be constructed ,in, is a semi-positive definite matrix. After eigendecomposition, we can get ,in represents the eigenvalue, and by solving it we can get represents a one-dimensional active subspace, which is a dimensional vector (the dimension is the same as the input vector), use represents its components, then , we can see that the one-dimensional active subspaces corresponding to low and high blowing ratios are: and .

[0092] S302, optimizing initial parameters of the active subspace under various film cooling conditions to obtain optimized model parameters.

[0093] In the embodiment of the present application, after obtaining the active subspace, the computer device can optimize each initial parameter according to the active subspace to obtain optimized model parameters.

[0094] Specifically, Figure 5 As shown, the above step S302 of "optimizing each initial parameter of the active subspace under each film cooling condition to obtain optimized model parameters" includes:

[0095] S501, determining parameters to be optimized under each film cooling working condition according to the active subspace under each film cooling working condition.

[0096] In the embodiment of the present application, after the computer device obtains the active subspace under each film cooling condition, it can analyze the initial parameters in the active subspace under each film cooling condition and determine the parameters to be optimized from various initial parameters. Specifically, the parameters to be optimized can be part of the initial parameters or all of the initial parameters. For example, the active subspace is and , sort by the size of the coefficients in the active subspace, and get the top three input parameters with the greatest impact under different blowing ratios , and determine the first three input parameters as the parameters to be optimized (i.e., the main control parameters).

[0097] S502, determining a parameter with the minimum error under each film cooling working condition according to the active subspace and each initial parameter under each film cooling working condition.

[0098] In the embodiment of the present application, after the computer device obtains the active subspace and initial parameters under each film cooling condition, the active subspace under each film cooling condition can be used to calculate the initial parameters by formula and , reduce the N initial parameters to one dimension and reconstruct the response surface and Then, the reconstructed response surface is solved according to the initial parameters after dimensionality reduction to obtain the one-dimensional input value with the lowest error, and the one-dimensional input value with the lowest error is used as the parameter with the lowest error. For example, the input value corresponding to zero error is obtained: and .

[0099] S503, optimizing each initial parameter according to the parameters to be optimized and the parameters with the smallest error under each film cooling condition to obtain optimized model parameters.

[0100] In an embodiment of the present application, after the computer device obtains the parameters to be optimized and the parameters with the smallest error, it can optimize the initial parameters according to the parameters to be optimized and the parameters with the smallest error in accordance with a preset optimization method to obtain the optimized model parameters.

[0101] In one embodiment, a specific implementation method for optimizing each initial parameter according to the parameters to be optimized and the parameters with the smallest error under each film cooling condition is also provided, such as Figure 6 As shown, the above step S503 of "optimizing each initial parameter according to the parameters to be optimized and the parameters with the smallest error under each film cooling condition to obtain the optimized model parameters" includes:

[0102] S601, determining a constraint relationship according to a parameter with the smallest error under each film cooling working condition.

[0103] In the embodiment of the present application, the computer equipment obtains the parameter with the minimum error under each film cooling condition. and Afterwards, the constraint relationships can be determined based on the relationships in the reconstructed response surface:

[0104]

[0105]

[0106] Among them, the left side of the equation is known, and the active subspace parameters on the right side of the equation are known. Therefore, this constraint relationship can be used to perform subsequent optimization work by changing specific parameters (parameters to be optimized).

[0107] S602: construct a constraint surface according to the value range and constraint relationship of the parameter to be optimized under each film cooling condition.

[0108] In the embodiment of the present application, the computer device can construct a constraint surface according to the value range and constraint relationship of the parameter to be optimized under each film cooling condition.

[0109] Optionally, before constructing the constraint surface, e.g. Figure 7 As shown, a method for optimizing model parameters is also provided, the method comprising:

[0110] S701, determining whether there are at least two common parameters among the parameters to be optimized under different film cooling conditions.

[0111] In the embodiment of the present application, the computer device obtains the parameters to be optimized under different film cooling conditions based on the above steps, and then can further determine whether there are at least two common parameters in the parameters to be optimized under different film cooling conditions. For example, the parameters to be optimized under the high blowing ratio condition are ABC, and the parameters to be optimized under the low blowing ratio condition are BCD, which means that there are at least two common parameters BC in the parameters to be optimized under different film cooling conditions. For another example, the parameters to be optimized under the high blowing ratio condition are ABC, and the parameters to be optimized under the low blowing ratio condition are BEF, which means that there are not at least two common parameters in the parameters to be optimized under different film cooling conditions.

[0112] S702, if it exists, when constructing the constraint surface, execute the step of constructing the constraint surface according to the value range and constraint relationship of the parameters to be optimized including at least two common parameters.

[0113] In an embodiment of the present application, if there are at least two common parameters in the parameters to be optimized under different film cooling conditions, it means that the parameters to be optimized in the conditions corresponding to the high blowing ratio and the conditions corresponding to the low blowing ratio can be uniformly optimized. Then, when constructing a constraint surface based on the value range and constraint relationship of the parameters to be optimized under each film cooling condition, the constraint surface formed by the two common parameters can be determined according to the value range and constraint relationship of the parameters to be optimized.

[0114] S703, if it does not exist, when constructing the constraint surface, execute the step of constructing the constraint surface according to the value range and constraint relationship of the parameter to be optimized under each film cooling condition.

[0115] In the embodiment of the present application, if there are not at least two common parameters in the parameters to be optimized under different film cooling conditions, it means that the parameters to be optimized in the conditions corresponding to the high blowing ratio and the conditions corresponding to the low blowing ratio need to be optimized respectively. Then, when constructing the constraint surface based on the value range and constraint relationship of the parameters to be optimized under each film cooling condition, the constraint surface can be constructed according to the value range and constraint relationship of the parameters to be optimized under the conditions corresponding to the high blowing ratio, and the constraint surface can be constructed according to the value range and constraint relationship of the parameters to be optimized under the conditions corresponding to the low blowing ratio.

[0116] S603, optimizing each initial parameter by solving the constraint surface to obtain optimized model parameters.

[0117] In an embodiment of the present application, after the computer device constructs the constraint surface based on the above steps, it can solve the constraint surface, optimize each initial parameter according to the result of the settlement, and obtain the optimized model parameters.

[0118] Specifically, Figure 8 As shown, the above step S603 of "optimizing each initial parameter by solving the constraint surface to obtain the optimized model parameters" includes:

[0119] S801, solving the constraint surface and obtaining a solution result.

[0120] In the embodiment of the present application, the computer device can solve the constraint surface by mathematical methods to obtain a solution result, and can further determine whether the solution result has a feasible domain, that is, determine whether the constraint surface has a solution.

[0121] S802: If the solution result indicates that a feasible domain exists, each initial parameter is optimized according to the feasible domain of the constraint surface to obtain optimized model parameters.

[0122] In the embodiment of the present application, if the solution result indicates that there is a feasible domain, it means that the constraint surface has a solution, and then the initial parameters can be optimized according to the feasible domain of the constraint surface to obtain the optimized model parameters.

[0123] When the computer equipment determines the feasible domain of the constraint surface, it is necessary to further determine whether the feasible domain meets the preset error range, such as Fig. 9 As shown, the above step S802 of "optimizing each initial parameter according to the feasible region of the constraint surface to obtain optimized model parameters" includes:

[0124] S901, determining whether the feasible domain of the constraint surface satisfies a preset error range.

[0125] The preset error range is a preset error limit, for example, the error limit is 5%.

[0126] In the embodiment of the present application, when the computer device determines the feasible domain where the constraint surface exists, it is necessary to further determine whether the feasible domain where the constraint surface exists satisfies a preset error range.

[0127] S902: If satisfied, the parameters included in the feasible domain are determined as optimized model parameters.

[0128] In the embodiment of the present application, if the feasible domain of the constraint surface satisfies the preset error range, the parameters to be optimized included in the feasible domain are output as the optimized model parameters.

[0129] S903, if not satisfied, then multiple initial parameters are updated according to the parameters with the smallest error under each film cooling condition, and based on the updated initial parameters and the operating condition information corresponding to different film cooling conditions, return to the step of inputting the operating condition information and multiple initial parameters into the turbulence model for prediction to obtain the error information of the parameters under each film cooling condition.

[0130] In the embodiment of the present application, if the feasible domain of the constraint surface does not satisfy the preset error range, multiple initial parameters are updated according to the parameters with the smallest error under each film cooling condition. Specifically, the parameter value range can be narrowed to the parameter with the smallest error. Then re-update the value range of the parameter to be optimized , other parameters are default values, and the process returns to the step of inputting the operating condition information and multiple initial parameters into the turbulence model for prediction to obtain the error information of the parameters under each film cooling operating condition, that is, returns to the step of executing S202.

[0131] S803, if the solution result indicates that there is no feasible domain, multiple initial parameters are updated, and based on the updated initial parameters and the operating condition information corresponding to different film cooling conditions, the step of returning to execute the step of inputting the operating condition information and multiple initial parameters into the turbulence model for prediction to obtain the error information of the parameters under each film cooling condition is performed.

[0132] In the embodiment of the present application, if the solution result indicates that there is no feasible domain, it means that the constraint surface has no solution, which means that the multiple initial parameters, parameter ranges, and preset error ranges set this time do not meet the optimization requirements.

[0133] The parameter range and the preset error range are then updated to obtain multiple updated initial parameters. Then, based on the updated initial parameters and the operating condition information corresponding to different film cooling conditions, the step of inputting the operating condition information and multiple initial parameters into the turbulence model for prediction is returned to obtain the error information of the parameters under each film cooling condition, that is, returning to the step of executing S202.

[0134] The method described in the embodiment of the present application uses a given turbulence model closure coefficient, takes the gap between the simulated calculated spanwise average cooling efficiency and the experimental value as the optimization target, and uses random sampling, response surface construction and active subspace analysis to optimize the calculation results under a wide blowing ratio to reduce the calculation error. This method can systematically analyze the impact of the turbulence model closure coefficient on the calculation error of film cooling under different blowing ratios, find out the main control parameters, and give an optimized coefficient set as the optimized model parameters. According to the average gap between the spanwise average cooling efficiency predicted by the target numerical value and the experimental value, the turbulence closure coefficient set that can minimize the target error is determined by performing active subspace analysis and linear programming analysis within the specified parameter optimization range.

[0135] Based on all the above embodiments, a method for optimizing model parameters is also provided, and the method includes:

[0136] S1001, obtaining operating condition information and multiple initial parameters corresponding to different film cooling operating conditions.

[0137] S1002, inputting the operating condition information and a plurality of initial parameters into the turbulence model for prediction, and obtaining the error information of the parameters under each film cooling operating condition.

[0138] S1003: construct a linear response surface and a covariance matrix according to the error information of the parameters under each film cooling condition and each initial parameter.

[0139] S1004, solving the linear response surface based on the covariance matrix to obtain the active subspace.

[0140] S1005, determining parameters to be optimized under each film cooling working condition according to the active subspace under each film cooling working condition.

[0141] S1006, determining a parameter with the minimum error under each film cooling working condition according to the active subspace and each initial parameter under each film cooling working condition.

[0142] S1007, determining a constraint relationship according to the parameter with the smallest error under each film cooling working condition.

[0143] S1008: Determine whether there are at least two common parameters among the parameters to be optimized under different film cooling conditions.

[0144] S1009: If it exists, when constructing a constraint surface, the constraint surface is constructed according to the value range and constraint relationship of the parameters to be optimized that include at least two common parameters.

[0145] S1010, if it does not exist, when constructing the constraint surface, the constraint surface is constructed according to the value range and constraint relationship of the parameter to be optimized under each film cooling condition.

[0146] S1012, solving the constraint surface to obtain a solution result.

[0147] S1013: If the solution result indicates that a feasible domain exists, determine whether the feasible domain of the constraint surface satisfies a preset error range.

[0148] S1014: If satisfied, the parameters included in the feasible domain are determined as optimized model parameters.

[0149] S1015, if not satisfied, then multiple initial parameters are updated according to the parameters with the smallest error under each film cooling working condition, and based on the updated initial parameters and the working condition information corresponding to different film cooling working conditions, return to the step of executing S1002.

[0150] S1016, if the solution result indicates that there is no feasible domain, multiple initial parameters are updated, and based on the updated initial parameters and the operating condition information corresponding to different film cooling operating conditions, the process returns to step S1002.

[0151] In the present application embodiment, Fig.10 As shown in Figure 2, the optimization process of the turbulence model can be specifically:

[0152] 1) Obtain film cooling operating information, including mainstream Mach number , boundary layer thickness , mainstream turbulence intensity , turbulence integral characteristic length And so on, there is also the mainstream and cold flow density ratio , Blowing ratio Cover wide blowing ratio, pay attention to select two groups of blowing ratios, namely low and high blowing ratios: low blowing ratio , high blowing ratio . Determine the optimized turbulence model and object parameters N, and set their corresponding ranges and error limits .

[0153] 2) For different blowing ratios, randomly and uniformly generate Sample input set: ; Through calculation, the corresponding solution of each sample input under different blowing ratios (i.e. the average error of the spanwise average cooling efficiency) is obtained: and .

[0154] 3) Construct a linear response surface based on the sampled data , , obtain the gradient covariance matrix and calculate the one-dimensional active subspace , According to the size of the one-dimensional active subspace coefficient, the main control mechanism of the top three influences is obtained, and the one-dimensional parameter is obtained and The one-dimensional solution with the lowest error limit and .

[0155] 4) For different situations, choose unified optimization or separate optimization according to whether there are more than two common main control parameters for different blowing ratios. During the optimization process, optimize the top three main control parameters, and keep the default values ​​for other parameters. , and one-dimensional active subspace, obtain multiple constraint surfaces spanned by the main control parameters, and examine the feasibility of the solution domain. If feasible, go to step 5) for output; if not feasible, return to step 1) for iteration by adjusting the parameter range, error limit, etc.

[0156] 5) Obtain the optimized parameter solution and output it.

[0157] Among them, 3) includes 31) to 33):

[0158] 31) Calculate one-dimensional active subspace

[0159] For different blowing ratios, a linear response surface is fitted based on the sample data. , Therefore, for each blowing ratio, the covariance matrix composed of gradients can be constructed: ,in, is a semi-positive definite matrix. After eigendecomposition, we can get ,in represents the eigenvalue, and by solving it we can get Represents a one-dimensional active subspace. From the above, we can get the one-dimensional active subspace corresponding to low and high blowing ratios , . According to the size of the coefficients in the subspace, the parameter sensitivity analysis is performed, and the top three input parameters (i.e., main control parameters) with the greatest impact under different blowing ratios are obtained. .

[0160] 32) Parameter Dimensionality Reduction

[0161] For different blowing ratios, the parameters are reduced to one dimension using the one-dimensional active subspace: and , and reconstruct the response surface and .

[0162] 33) Calculate the one-dimensional solution with the lowest error limit

[0163] For different blowing ratios, the one-dimensional input value with the lowest error limit (usually 0) is obtained by using the reduced dimension parameters and response surface: and .

[0164] Among them, 4) includes 41) to 46):

[0165] 41) Determine whether different blowing ratios contain ≥ 2 common parameters. If yes, select step 42); if no, select step 43)

[0166] 42) Unified Optimization

[0167] 421) According to the constraint relationship given by the active subspace and the range of values ​​of each parameter, the constraint surface spanned by the two common parameters is determined.

[0168] 422) Determine whether there is a feasible solution domain. If yes, select step 44); if not, select step 43)

[0169] 43) Optimize separately

[0170] 431) According to the constraint relationship given by each active subspace and the range of each parameter, the constraint surface spanned by the top three influencing parameters is determined.

[0171] 432) Determine whether there is a feasible solution domain. If yes, select step 44); if not, select step 45)

[0172] 44) Obtain the solution set within the solution domain and verify whether the solution set meets the set error range. If so, select step 5); if not, select step 46)

[0173] 45) There is no optimal solution within the current parameter range and error limit. Redefine the parameter value range. With error limit , return to step 1) and iterate again.

[0174] 46) Narrow the parameter value range to Nearby, redefine the value: , other parameters are default values, and return to step 1) to iterate again.

[0175] The following is an example of optimizing the closure coefficient of the SST model in this embodiment:

[0176] This embodiment constructs a turbulence model based on experimental conditions (Saumweber C et al, Effect of Geometry Variation on the Cooling Performance of Fan-Shaped Cooling Holes, Journal of Turbomachinery, 2012, 134: 061008), such as Fig.11 According to the model definition, 9 parameters in the SST model are selected for research and optimization. The specific parameters and their value ranges are shown in the table below. It is 4%.

[0177]

[0178] After proceeding to step 31), the one-dimensional active subspace coefficients at low and high blowing ratios are shown in Fig.12 In the figure, blue is low blowing ratio and brown is high blowing ratio. The main control parameters can be obtained: After proceeding to step 33), the response surface at low and high blowing ratios reconstructed by one-dimensional input is as follows: Fig.13 As shown, the solution , .

[0179] Go to step 4), there are at least two common master control parameters, go to step 42) unified optimization, solve the domain analysis such as Fig.14 As shown. Determine whether there is a feasible solution domain and select the solution domain parameter set: =[0.68,0.06,0.78,0.62], go to step 44) for verification. The verification result is as follows Fig.15 As shown, the average errors of low and high blowing ratios are 0.2% and 0.9% respectively, which are lower than the set error limit of 4%. The solution set is feasible and the current solution is output.

[0180] Compared with the values ​​predicted by the default parameters: the average errors of low and high blowing ratios are 50.6% and 13.2% respectively, and the average errors after optimization are greatly reduced. It can be seen that the optimized coefficient set greatly improves the prediction ability of the RANS model under a wide blowing ratio, which illustrates the effectiveness of the method described in this embodiment in the main control mechanism analysis and coefficient optimization.

[0181] The methods described in the above steps are all described in the above embodiments. Please refer to the above description for details and will not be repeated here.

[0182] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0183] Based on the same inventive concept, the embodiment of the present application also provides a device for optimizing model parameters for implementing the above-mentioned method for optimizing model parameters. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific definition in the embodiment of the device for optimizing one or more model parameters provided below can refer to the definition of the method for optimizing model parameters above, and will not be repeated here.

[0184] In one embodiment, Fig.16 As shown, a device for optimizing model parameters is provided, comprising:

[0185] The acquisition module 10 is used to acquire operating condition information and a plurality of initial parameters corresponding to different film cooling operating conditions.

[0186] The prediction module 11 is used to input the operating condition information and a plurality of initial parameters into the turbulence model for prediction, and obtain the error information of the parameters under various film cooling operating conditions.

[0187] The optimization module 12 is used to optimize each initial parameter according to the error information of the parameter under each film cooling working condition to obtain the optimized model parameters.

[0188] In one embodiment, the optimization module 12 includes:

[0189] The first determination unit is used to determine the active subspace under each film cooling working condition according to the error information of the parameters under each film cooling working condition and each initial parameter.

[0190] The optimization unit is used to optimize the initial parameters of the active subspace under various film cooling conditions to obtain optimized model parameters.

[0191] In one embodiment, the above-mentioned optimization unit is specifically used to determine the parameters to be optimized under each film cooling condition according to the active subspace under each film cooling condition; determine the parameters with the smallest error under each film cooling condition according to the active subspace under each film cooling condition and each initial parameter; optimize each initial parameter according to the parameters to be optimized and the parameters with the smallest error under each film cooling condition to obtain the optimized model parameters.

[0192] In one embodiment, the above-mentioned optimization unit is specifically used to determine the constraint relationship according to the parameter with the smallest error under each air film cooling condition; construct a constraint surface according to the value range and constraint relationship of the parameter to be optimized under each air film cooling condition; and optimize each initial parameter by solving the constraint surface to obtain the optimized model parameters.

[0193] In one embodiment, the above-mentioned optimization unit is specifically used to solve the constraint surface and obtain a solution result; if the solution result indicates that there is a feasible domain, then the initial parameters are optimized according to the feasible domain of the constraint surface to obtain optimized model parameters; if the solution result indicates that there is no feasible domain, then the multiple initial parameters are updated, and based on the updated initial parameters and the operating condition information corresponding to different film cooling conditions, the operating condition information and the multiple initial parameters are input into the turbulence model for prediction to obtain the error information of the parameters under each film cooling condition.

[0194] In one embodiment, the above-mentioned optimization unit is specifically used to determine whether the feasible domain of the constraint surface satisfies a preset error range; if so, the parameters included in the feasible domain are determined as optimized model parameters; if not, multiple initial parameters are updated according to the parameters with the smallest error under each film cooling condition, and based on the updated initial parameters and the operating condition information corresponding to different film cooling conditions, return to execute the step of inputting the operating condition information and multiple initial parameters into the turbulence model for prediction to obtain the error information of the parameters under each film cooling condition.

[0195] In one embodiment, the optimization module 12 further includes:

[0196] The second determination unit is used to determine whether there are at least two common parameters among the parameters to be optimized under different film cooling conditions.

[0197] The first loop unit is used to execute the step of constructing the constraint surface according to the value range and constraint relationship of the parameters to be optimized including at least two common parameters when constructing the constraint surface if it exists.

[0198] The second circulation unit is used to execute the step of constructing the constraint surface according to the value range and constraint relationship of the parameter to be optimized under each film cooling condition when constructing the constraint surface if it does not exist.

[0199] In one embodiment, the first determining unit includes:

[0200] The subunit is constructed to construct a linear response surface and a covariance matrix according to the error information of the parameters under each film cooling condition and each initial parameter.

[0201] The solving subunit is used to solve the linear response surface based on the covariance matrix to obtain the active subspace.

[0202] Each module in the above-mentioned model parameter optimization device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0203] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the method for optimizing model parameters described in any of the above embodiments are implemented.

[0204] In one embodiment, a computer-readable storage medium is provided on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for optimizing the model parameters described in any of the above embodiments are implemented.

[0205] In one embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements the steps of the method for optimizing model parameters described in any of the above embodiments.

[0206] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium 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), magnetoresistive 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. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0207] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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.

[0208] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for optimizing model parameters, characterized in that: The method comprises: Obtaining operating condition information and multiple initial parameters corresponding to different film cooling operating conditions; Inputting the operating condition information and the multiple initial parameters into a turbulence model for prediction to obtain error information of the parameters under each of the film cooling operating conditions; The initial parameters are optimized according to the error information of the parameters under the film cooling conditions to obtain optimized model parameters.

2. The method according to claim 1, characterized in that The step of optimizing each of the initial parameters according to the error information of the parameters under each of the film cooling conditions to obtain optimized model parameters includes: Determining the active subspace under each of the film cooling working conditions according to the error information of the parameters under each of the film cooling working conditions and each of the initial parameters; The initial parameters of the active subspace under the film cooling conditions are optimized to obtain optimized model parameters.

3. The method according to claim 2, characterized in that The optimizing of each initial parameter of the active subspace under each film cooling condition to obtain the optimized model parameters includes: Determining the parameters to be optimized under each of the film cooling working conditions according to the active subspace under each of the film cooling working conditions; Determining parameters with the minimum error under each of the film cooling conditions according to the active subspaces under each of the film cooling conditions and the initial parameters; According to the parameters to be optimized under the film cooling conditions and the parameters with the smallest errors, the initial parameters are optimized to obtain optimized model parameters.

4. The method according to claim 3, characterized in that The step of optimizing the initial parameters according to the parameters to be optimized under the film cooling conditions and the parameters with the smallest error to obtain the optimized model parameters includes: Determining a constraint relationship according to the parameters with the smallest error under each of the film cooling conditions; Constructing a constraint surface according to the value range of the parameter to be optimized under each of the film cooling conditions and the constraint relationship; By solving the constraint surface, each of the initial parameters is optimized to obtain optimized model parameters.

5. The method according to claim 4, characterized in that The step of optimizing each of the initial parameters by solving the constraint surface to obtain optimized model parameters includes: Solving the constraint surface to obtain a solution result; If the solution result indicates that a feasible domain exists, each of the initial parameters is optimized according to the feasible domain of the constraint surface to obtain optimized model parameters; If the solution result indicates that there is no feasible domain, the multiple initial parameters are updated, and based on the updated initial parameters and the operating condition information corresponding to different film cooling conditions, the step of inputting the operating condition information and the multiple initial parameters into the turbulence model for prediction is returned to obtain the error information of the parameters under each film cooling condition.

6. The method according to claim 5, characterized in that The step of optimizing the initial parameters according to the feasible domain of the constraint surface to obtain optimized model parameters includes: Determine whether the feasible domain of the constraint surface satisfies a preset error range; If satisfied, the parameters included in the feasible domain are determined as the optimized model parameters; If it is not satisfied, the multiple initial parameters are updated according to the parameters with the smallest error under each of the film cooling conditions, and based on the updated initial parameters and the operating condition information corresponding to different film cooling conditions, return to execute the step of inputting the operating condition information and the multiple initial parameters into the turbulence model for prediction to obtain the error information of the parameters under each of the film cooling conditions.

7. The method according to claim 4, characterized in that The method further comprises: Determine whether there are at least two common parameters of the parameters to be optimized under different film cooling conditions; If so, when constructing the constraint surface, a step of constructing the constraint surface is performed according to the value range of the parameter to be optimized including the at least two common parameters and the constraint relationship; If it does not exist, when constructing the constraint surface, a step of constructing the constraint surface is performed according to the value range of the parameter to be optimized under each of the film cooling conditions and the constraint relationship.

8. The method according to any one of claims 2 to 7, characterized in that: Determining the active subspace under each of the film cooling working conditions according to the error information of the parameters under each of the film cooling working conditions and each of the initial parameters includes: Constructing a linear response surface and a covariance matrix according to the error information of the parameters under the film cooling conditions and the initial parameters; The linear response surface is solved based on the covariance matrix to obtain the active subspace.

9. A device for optimizing model parameters, characterized in that: The device comprises: An acquisition module, used to acquire operating condition information and multiple initial parameters corresponding to different film cooling operating conditions; A prediction module, used for inputting the working condition information and the multiple initial parameters into a turbulence model for prediction, and obtaining error information of the parameters under each of the film cooling working conditions; The optimization module is used to optimize each of the initial parameters according to the error information of the parameters under each of the film cooling conditions to obtain optimized model parameters.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images