Evaluation Methods for the Flow Field Solving Capabilities of Integrated CAD and CAE Software

Abstract

This invention discloses an integrated method for evaluating the flow field solving capability of CAD and CAE software, comprising: S1: importing the model and summarizing and normalizing the quantitative parameters of the solving process; S2: quantifying the qualitative parameters in the mesh generation capability to comprehensively evaluate the mesh generation capability; S3: quantifying the qualitative parameters in the solver capability analysis to comprehensively evaluate the solver's physical model; S4: calculating the full parameter evaluation weights based on the quantification results of S1-S3, and generating a final analysis report on the software's solving capability. This invention has the advantages of good versatility, high quantification degree, and strong comprehensiveness of evaluation parameters, and is also easy to automate the entire process on a computer using a programming language.

Description

Technical Field

[0001] This invention relates to the field of computer-aided design and computation technology, and in particular to an evaluation method for the flow field solving capability of integrated CAD and CAE software. Background Technology

[0002] Currently, with technological advancements, many technological fields have entered the era of big data. Research in various disciplines is no longer limited to the calculation of lumped parameters, but rather pursues higher computational accuracy and spatiotemporal resolution in theoretical and technological applications. Taking flow field calculation as an example, compared to the initial qualitative judgment of flow fields based on directly measurable parameters at the macroscopic scale, current flow field simulation technology tends to perform multi-parameter coupled state reconstruction. In sub-fields such as heat transfer, multiphase flow, and combustion, the problems faced by flow field simulation are becoming increasingly complex, leading to the emergence of computer-aided computation and design technologies, further enhancing the research capabilities of different fields in addressing complex problems. However, various CAD and CAE software for computer-aided design and computation exist, each with its own advantages. Therefore, comprehensively evaluating the flow field solving capabilities of software for different physical processes and selecting the optimal solution based on comprehensive analysis results will greatly benefit the improvement of research and design efficiency.

[0003] Current evaluation methods for CAD and CAE software primarily rely on single-point analysis of scattered indicators, focusing mainly on aspects such as mesh generation quality, computational speed, and accuracy. A single parameter can only reflect one aspect of the computer-aided software's performance. For a specific simulation requirement, this evaluation system cannot provide a complete and comprehensive analysis report to determine the optimal solution software selection. Furthermore, this method ignores many parameters that are not directly quantifiable but directly reflect the software's computational capabilities, resulting in a lack of comprehensiveness in the final analysis. Additionally, while some detailed quantitative methods exist in specific subfields of flow field simulation, evaluating simulation methods from a multi-parameter perspective, these methods are specific to particular problems within a specific domain. When faced with a wide variety of CAD and CAE software, their lack of universality means that the evaluation effectiveness needs improvement. Summary of the Invention

[0004] The purpose of this invention is to provide an integrated method for evaluating the flow field solving capability of CAD and CAE software, in order to solve the problems of poor universality and single evaluation results of existing flow field solving capability evaluation methods.

[0005] The above-mentioned objectives of the present invention can be achieved by the following technical solutions:

[0006] This invention provides an integrated method for evaluating the flow field solving capability of CAD and CAE software, including:

[0007] S1: Import the model and summarize and normalize the quantitative parameters of the solution process;

[0008] S2: Quantification of qualitative parameters in mesh generation capability to enable comprehensive evaluation of mesh generation capability;

[0009] S3: Quantify the qualitative parameters in the solver capability analysis so that they can be used to complete a comprehensive evaluation of the solver's physical model;

[0010] S4: Based on the quantization results of S1-S3, calculate the full parameter evaluation weights and generate the final analysis report on the software's solving capability.

[0011] Preferably, S1 includes:

[0012] S11: Perform preliminary clustering of the parameters of the imported model, and denote the parameters representing geometric processing capabilities as α1, ..., α2. n The parameters characterizing the mesh properties are denoted as β1, ..., β2. n The parameters characterizing the solver properties are denoted as γ1, ..., γ2. n The parameters characterizing the post-processing properties are denoted as δ1, ..., δ n ,:

[0013] S12: Select the parameters that can be directly quantified, and separate the positive incentive parameter α from them. + ,β + γ + δ + and the negative positivity parameter α - ,β - γ - δ - ;

[0014] S13: Find the standard values ​​of all directly quantifiable parameters;

[0015] S14: Normalize the parameters according to their characteristics to unify the final evaluation criteria.

[0016] Among them, the larger the value of the normalized parameter, the better the performance of the corresponding parameter; the larger the positive positivity parameter, the better the performance of the software; the smaller the negative positivity parameter, the better the performance of the software.

[0017] Preferably, the standard values ​​of all directly quantifiable parameters are obtained in the following way:

[0018] Acquire standard models within multiple target application domains;

[0019] The parameters generated by simulating or directly manipulating standard physical processes based on various standard models are the standard values.

[0020] Preferably, the normalization method is as follows:

[0021]

[0022]

[0023] Preferably, S2 includes:

[0024] S21: After importing the mesh model, establish the coordinate origin and record the coordinate values ​​of all meshes in the current model;

[0025] S22: Assuming that a mesh model composed entirely of regular hexahedral meshes is a standard high-quality mesh model, in the mesh subdivision capability evaluation algorithm, multiple hexahedrals are combined into a geometric aggregate so that the imported model boundary is exactly and completely contained within the geometric aggregate.

[0026] S23: Continue to perform dichotomy fractal on the geometric aggregate until the number of cubes contained in the mesh structure is the same as the number of meshes in the mesh structure;

[0027] S24: Calculate the Hausdorff distance h between the point set of the geometric aggregate and the mesh structure. ρ ;

[0028] S25: Assuming the highest quality mesh perfectly coincides with the geometry aggregate in S24, the quality of the normalized mesh in S24 is characterized by the following formula:

[0029]

[0030] Preferably, S22 includes:

[0031] S221: Use multiple cubes with side length a to completely cover the mesh structure and record the coordinates of each vertex;

[0032] S222: Calculate the Hausdorff distance between the cubic aggregate and the mesh structure in S221 according to the following formula:

[0033] h ρ (E, A) = sup m∈E d(m, A)∨sup n∈A d(n, E)

[0034]

[0035] Among them, hρ Let A represent the Hausdorff distance, and let E represent the point set of the aggregate in S221;

[0036] S223: If h calculated in S222 ρ If (E, A) is less than the preset threshold σ, then the grid structure is considered to be exactly and completely contained.

[0037] S224: If h calculated in S222 ρ If (E, A) is greater than the threshold σ, then each cube with side length a in the geometric aggregate in S221 is subjected to dichotomous fractal analysis to remove cubes that fall completely outside the grid structure.

[0038] S225: Repeat steps S222 to S224 until h ρ (E, A) is less than the threshold σ.

[0039] Preferably, S3 includes:

[0040] S31: Import basic cases and experimental data from different target flow field simulation fields into the evaluation model;

[0041] S32: Load the standard test conditions into the model to be evaluated and use them to solve the flow field characteristics in the corresponding flow field simulation field;

[0042] S33: Compare the spatial points and calculation results at different time points output by the solution model with the corresponding experimental results in the experimental data to obtain the difference Δz. i，j , where i and j represent different spatial and temporal positions, respectively;

[0043] S34: Calculate the degree of twinning between the solution model and the actual physical process using the following formula:

[0044]

[0045] Where, |Δz max | indicates the maximum difference between experimental values ​​on the time and spatial scales.

[0046] Preferably, the experimental data includes experimental data at spatial points and all time data.

[0047] Preferably, S4 includes:

[0048] S41: Distinguish between parameters that have a direct impact on the calculation result output and evaluation parameters that have only an indirect impact on the calculation result or are unrelated to the calculation result, where the weight coefficient of the latter is assumed to be 1;

[0049] S42: For evaluation parameters whose output results are directly affected by the calculation results, calculate the weight coefficient of the corresponding parameter based on the sensitivity of the calculation results to their changes.

[0050] S43: Combine the weighting coefficients to draw a polygonal diagram to evaluate the software's solving capability;

[0051] S44: Combine all the quantitative indicators calculated in steps S1 to S3 with the weighting coefficients in S41, mark the indicators on each coordinate dimension of the polygon graph, and connect them to form a closed surface.

[0052] S45: Calculate the total area of ​​the closed surface;

[0053] The total area of ​​the closed surface represents the overall performance of the software's solving capability; the larger the total area, the stronger the software's solving capability.

[0054] Preferably, among the evaluation parameters whose calculation results have a direct impact, the weighting coefficients of the corresponding parameters are calculated based on the sensitivity of the calculation results to their changes, as follows:

[0055] If a 10% change in the evaluation parameter would cause an average fluctuation of a% in the calculation result, then its weighting coefficient would be 1+a.

[0056] This invention has at least the following features and advantages:

[0057] This invention has the advantages of good versatility, high degree of quantification, and strong comprehensiveness of evaluation parameters. At the same time, it is easy to implement the entire process of automatic analysis on a computer through programming languages. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart of the integrated CAD and CAE software flow field solving capability evaluation method of the present invention;

[0060] Figure 2 This is a schematic diagram of the thermal management flow field of a cylindrical lithium battery according to an embodiment of the present invention;

[0061] Figure 3 This refers to the quantification process of qualitative parameters in the mesh generation capability in step S2 of this embodiment of the invention;

[0062] Figure 4This refers to the quantification of qualitative parameters in the analysis of solver capability in step S3 of this embodiment of the invention.

[0063] Figure 5 This is the polygonal diagram given in step S4 of an embodiment of the present invention. Detailed Implementation

[0064] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] This invention provides an integrated method for evaluating the flow field solving capabilities of CAD and CAE software. Please refer to [link / reference]. Figures 1 to 5 ,include:

[0066] S1: Import the model and summarize and normalize the quantitative parameters of the solution process;

[0067] S2: Quantification of qualitative parameters in mesh generation capability to enable comprehensive evaluation of mesh generation capability;

[0068] S3: Quantify the qualitative parameters in the solver capability analysis so that they can be used to complete a comprehensive evaluation of the solver's physical model;

[0069] S4: Based on the quantization results of S1-S3, calculate the full parameter evaluation weights and generate the final analysis report on the software's solving capability.

[0070] In some embodiments, S1 includes:

[0071] S11: Perform preliminary clustering of the parameters of the imported model, and denote the parameters representing geometric processing capabilities as α1, ..., α2. n The parameters characterizing the mesh properties are denoted as β1, ..., β2. n The parameters characterizing the solver properties are denoted as γ1, ..., γ2. n The parameters characterizing the post-processing properties are denoted as δ1, ..., δ n ,:

[0072] S12: Select the parameters that can be directly quantified, and separate the positive incentive parameter α from them. + ,β + γ + δ + and the negative positivity parameter α - ,β - γ - δ - ;

[0073] S13: Find the standard values ​​of all directly quantifiable parameters;

[0074] S14: Normalize the parameters according to their characteristics to unify the final evaluation criteria.

[0075] Among them, the larger the value of the normalized parameter, the better the performance of the corresponding parameter; the larger the positive positivity parameter, the better the performance of the software; the smaller the negative positivity parameter, the better the performance of the software.

[0076] Furthermore, the standard values ​​for all directly quantifiable parameters are obtained as follows:

[0077] Acquire standard models within multiple target application domains;

[0078] The parameters generated by simulating or directly manipulating standard physical processes based on various standard models are the standard values.

[0079] Furthermore, the normalization method is as follows:

[0080]

[0081]

[0082] In some embodiments, S2 includes:

[0083] S21: After importing the mesh model, establish the coordinate origin and record the coordinate values ​​of all meshes in the current model;

[0084] S22: Assuming that a mesh model composed entirely of regular hexahedral meshes is a standard high-quality mesh model, in the mesh subdivision capability evaluation algorithm, multiple hexahedrals are combined into a geometric aggregate so that the imported model boundary is exactly and completely contained within the geometric aggregate.

[0085] S23: Continue to perform dichotomy fractal on the geometric aggregate until the number of cubes contained in the mesh structure is the same as the number of meshes in the mesh structure;

[0086] S24: Calculate the Hausdorff distance h between the point set of the geometric aggregate and the mesh structure. ρ ;

[0087] S25: Assuming the highest quality mesh perfectly coincides with the geometry aggregate in S24, the quality of the normalized mesh in S24 is characterized by the following formula:

[0088]

[0089] Furthermore, S22 includes:

[0090] S221: Use multiple cubes with side length a to completely cover the mesh structure and record the coordinates of each vertex;

[0091] S222: Calculate the Hausdorff distance between the cubic aggregate and the mesh structure in S221 according to the following formula:

[0092] h ρ (E, A) = sup m∈E d(m, A)∨sup n∈A d(n, E)

[0093]

[0094] Among them, h ρ Let A represent the Hausdorff distance, and let E represent the point set of the aggregate in S221;

[0095] S223: If h calculated in S222 ρ If (E, A) is less than the preset threshold σ, then the grid structure is considered to be exactly and completely contained.

[0096] S224: If h calculated in S222 ρ If (E, A) is greater than the threshold σ, then perform a dichotomy fractal on each cube with side length a in the geometric aggregate in S221 to remove cubes that fall completely outside the mesh structure.

[0097] S225: Repeat steps S222 to S224 until h ρ (E, A) is less than the threshold σ.

[0098] In some embodiments, S3 includes:

[0099] S31: Import basic cases and experimental data from different target flow field simulation fields into the evaluation model;

[0100] S32: Load the standard test conditions into the model to be evaluated and use them to solve the flow field characteristics in the corresponding flow field simulation field;

[0101] S33: Compare the spatial points and calculation results at different time points output by the solution model with the corresponding experimental results in the experimental data to obtain the difference Δz. i，j , where i and j represent different spatial and temporal positions, respectively;

[0102] S34: Calculate the degree of twinning between the solution model and the actual physical process using the following formula:

[0103]

[0104] Where, |Δz max | indicates the maximum difference between experimental values ​​on the time and spatial scales.

[0105] Furthermore, the experimental data includes experimental data at spatial points as well as all time data.

[0106] In some embodiments, S4 includes:

[0107] S41: Distinguish between parameters that have a direct impact on the calculation result output and evaluation parameters that have only an indirect impact on the calculation result or are unrelated to the calculation result, where the weight coefficient of the latter is assumed to be 1;

[0108] S42: For evaluation parameters whose output results are directly affected by the calculation results, calculate the weight coefficient of the corresponding parameter based on the sensitivity of the calculation results to their changes.

[0109] S43: Combine the weighting coefficients to draw a polygonal diagram to evaluate the software's solving capability;

[0110] S44: Combine all the quantitative indicators calculated in steps S1 to S3 with the weighting coefficients in S41, mark the indicators on each coordinate dimension of the polygon graph, and connect them to form a closed surface.

[0111] S45: Calculate the total area of ​​the closed surface;

[0112] The total area of ​​the closed surface represents the overall performance of the software's solving capability; the larger the total area, the stronger the software's solving capability.

[0113] Furthermore, for the evaluation parameters that directly affect the calculation results, the weighting coefficients of the corresponding parameters are calculated based on the sensitivity of the calculation results to their changes, as follows:

[0114] If a 10% change in the evaluation parameter would cause an average fluctuation of a% in the calculation result, then its weighting coefficient would be 1+a.

[0115] This invention has the advantages of good versatility, high degree of quantification, and strong comprehensiveness of evaluation parameters. At the same time, it is easy to implement the entire process of automatic analysis on a computer through programming languages.

[0116] The present invention will be further described below through several specific embodiments. Please refer to [link / reference]. Figures 1 to 5 In this embodiment, the evaluation of computer-aided software ANSA and Fluent is used as an example for illustration. The specific steps are as follows:

[0117] S1: Importing the model and summarizing and normalizing the quantitative parameters in the solution process. This includes the following steps:

[0118] S11: Based on the imported model, perform preliminary clustering of the model parameters, and denote the parameters representing geometric processing capabilities as α1, ..., α2. n The parameters characterizing the mesh properties are denoted as β1, ..., β2. n The parameters characterizing the solver properties are denoted as γ1, ..., γ2. n The parameters characterizing the post-processing properties are denoted as δ1, ..., δ n ,

[0119] The parameters characterizing geometric processing capability include whether the construction of various complex surfaces is feasible, denoted as α1, ..., α2. i This type of parameter is set to have only two values, 0 and 1, with 1 being the value when it can be completed and 0 being the value when it cannot be completed. Other parameters that characterize geometric processing capabilities include the number of geometric model libraries, the speed and quality of imports, and the ability to clean up and repair geometry.

[0120] Parameters that can be directly quantified to characterize mesh properties include mesh size, mesh warpage, surface mesh, and volume mesh count;

[0121] Parameters that can be directly quantified to characterize the solver include solution speed, solution accuracy, convergence speed, and parallel capability.

[0122] Parameters that can be directly quantified to characterize post-processing features include image processing speed, animation frame rate, image resolution, and the completeness of result reports.

[0123] S12: Select the parameters that can be directly quantified, and separate the positive incentive parameter α from them. + ,β + γ + δ + and the negative positivity parameter α - ,β - γ - δ - The larger the positive positivity parameter, the better the software performance; conversely, the smaller the negative positivity parameter, the better the software performance. For example, solution accuracy and solution speed are positive positivity parameters, while mesh warpage is a negative positivity parameter.

[0124] S13: Find the standard values ​​of all directly quantifiable parameters. This method defines the standard values ​​of parameters as follows: using standard models in different application domains, simulate standard physical processes, or directly manipulate them; the corresponding parameter is the standard value.

[0125] Specifically, the thermal field model can simulate the temperature distribution of lithium-ion batteries under constant current and dynamic operating conditions. For flow field simulation, the turbulence characteristics inside the plug-in cylindrical array can be simulated. For combustion, the combustion process of an internal combustion engine can be selected.

[0126] S14: This example uses a cylindrical lithium battery thermal management system as an example. See Figure 2 Normalization is performed based on the characteristics of the parameters, so that the final evaluation standard for the parameters is unified: the larger the value of the normalized parameter, the better the parameter performance. The specific normalization method is as follows:

[0127]

[0128]

[0129] S2: Quantification of qualitative parameters in mesh generation capability, and comprehensive evaluation of mesh generation capability. Specifically, this includes the following steps:

[0130] S21: After importing the mesh model, establish the coordinate origin and record the coordinate values ​​of all meshes in the current model.

[0131] S22: Assume a mesh model composed entirely of regular hexahedral meshes is a standard "high-quality" mesh model. In the mesh generation capability evaluation algorithm, multiple hexahedrons are combined into a geometric aggregate such that the imported model boundary is exactly and completely contained within this aggregate. The specific method to ensure exactly and completely contained is as follows:

[0132] S221: Taking flow field simulation as an example, such as Figure 3 As shown, multiple cubes with side length 'a' completely cover the mesh structure, and the coordinates of each vertex are recorded.

[0133] S222: Calculate the Hausdorff distance between the cubic aggregate and the mesh structure in S221 using the following formula:

[0134] h ρ (E, A) = sup m∈E d(m, A)∨sup n∈A d(n, E)

[0135]

[0136] Where h ρ Let A represent the Hausdorff distance, A represent the point set of the aggregate in S221, and E represent the point set of the mesh structure.

[0137] S223: If h calculated in S222 ρIf (E, A) is less than a certain threshold σ, set as one-hundredth of the model's maximum scale, then the mesh structure is considered to be exactly and completely contained.

[0138] S224: If h calculated in S222 ρ If (E, A) is greater than a certain threshold σ, set as one-hundredth of the maximum scale of the model, then each cube with side length a in the geometric aggregate of S221 is subjected to bifractal transformation, i.e., it is divided into 8 cubes with side length a / 2. Cubes that fall completely outside the mesh structure are then removed.

[0139] S225: Repeat steps S222 to S224 until h ρ (E, A) is less than a certain threshold σ.

[0140] S23: Continue to perform dichotomy fractal analysis on the geometric aggregate obtained in S22 until the number of cubes contained within the mesh structure is consistent with the number of meshes in the mesh structure.

[0141] S24: Using the Hausdorff distance calculation formula in S221, calculate the Hausdorff distance h between the cube aggregate and the point set of the mesh structure. ρ At the scale of the case study, the Hausdorff distance h of the mesh ultimately obtained in ANSA is... ρ =0.01m

[0142] S25: Assume the "highest quality" mesh perfectly coincides with the geometry aggregate in S24, i.e., h ρ =0, then the normalized mesh quality in S24 is characterized as follows:

[0143]

[0144] S3: Quantification of qualitative parameters in the solver capability analysis, completing the comprehensive evaluation of the solver's physical model. This specifically includes the following steps:

[0145] S31: Import fundamental case studies and detailed experimental data from different flow field simulation fields into the evaluation model, including experimental data at as many spatial points as possible and all time data. In this example, multi-point temperature data and flow velocity data are imported. A comparison of Fluent's calculation results and experimental results can be found... Figure 4 ;

[0146] S32: Load the standard battery constant current test conditions and the constant flow rate conditions of the air inlet into the solution model to be evaluated, and use them to solve the flow field characteristics in the corresponding flow field simulation field.

[0147] S33: Compare the spatial points and calculation results at different time points output by the solution model with the corresponding experimental results in the experimental database, and obtain the difference between the two, denoted as Δz. i，j , where i and j represent different spatial and temporal positions, respectively.

[0148] S34: Calculate the twinning degree between the solution model and the actual physical process using the following formula, and then integrate all data points.

[0149]

[0150] S4: Using the quantization results of S1-S3 as input, solve for the full parameter evaluation weights and provide a final analysis report on the software's solution capability. This includes the following steps:

[0151] S41: Distinguish between parameters that have a direct impact on the calculation result output and evaluation parameters that have only an indirect impact on the calculation result or are unrelated to the calculation result, where the weight coefficient of the latter is assumed to be 1;

[0152] S42: For evaluation parameters that directly affect the calculation results, the weight coefficient of the corresponding parameter is calculated based on the sensitivity of the calculation results to their changes. In this method, if a 10% change rate of the evaluation parameter will cause an average fluctuation of a% in the calculation results, then its weight coefficient is 1+a%.

[0153] For example, the typical binary parameters in S11 (parameters with only two values, 0 and 1) are used only as qualitative evaluations of the generated results, with a weight of 1.

[0154] S43: Combining the weighting coefficients, draw a polygonal diagram to evaluate the software's solver capability. Since the weighting coefficients for different parameters vary, the final polygonal diagram is not a regular polygon. In this case, the weight for mesh generation capability is 1.1, the weight for the solver model is 1.25, and the weight for most mesh quality quantification indicators (such as warpage) is 1.05.

[0155] S44: Take all the quantitative indicators calculated in steps S1 to S3, combine them with the weighting coefficients in S41, and plot the indicators on each coordinate dimension of the polygon graph, then connect them to form a closed surface, such as... Figure 5 As shown;

[0156] S45: The total area of ​​a closed surface represents the overall performance of the software's solving capability. The larger the total area, the stronger the software's solving capability.

[0157] This invention utilizes fractal principles and distance calculation methods for point sets to achieve a comprehensive quantification of mesh generation quality, making it a computable variable introduced into the evaluation of computing power.

[0158] This invention provides a quantitative method for characterizing the quality of a solver's physical model, making it a computable variable that can be incorporated into the evaluation of computational power.

[0159] This invention proposes a method for calculating the weight coefficients of all quantifiable indicators, which enables the evaluation process of software solving capabilities to be adaptively adjusted in different models for different physical problems.

[0160] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An integrated CAD and CAE software flow field solver capability assessment method, characterized by, The application relates to a software solver capability analysis method, which comprises the following steps: S1: importing a model and performing normalization processing on quantitative parameters of a solving process; S2: quantifying qualitative parameters in grid division capability so as to comprehensively evaluate the grid division capability; S3: quantifying qualitative parameters in solver capability analysis so as to comprehensively evaluate a physical model of a solver; S4: solving a full parameter evaluation weight based on the quantification results of S1-S3, and generating a final software solver capability analysis report; The S2 comprises: S21: after importing a grid model, establishing a coordinate origin and recording coordinate values of all grids of the current model; S22: assuming that a grid model completely composed of regular hexahedral grids is a standard high-quality grid model, in the evaluation algorithm of grid division capability, a plurality of hexahedral grids are combined into a geometric body aggregate, so that the model boundary is completely contained in the geometric body aggregate; S23: the geometric body aggregate is continuously bifurcated and fractured until the number of cubes contained in the grid structure is consistent with the number of grids in the grid structure; S24: Calculate the Hausdorff distance between the geometric body aggregate and the point set of the grid structure ; S25: assuming that the optimal grid is completely coincident with the geometric body aggregate in S24, the normalized grid quality in S24 is characterized by the following formula: ； The S3 comprises: S31: importing basic cases and experimental data in different target flow field simulation fields into the evaluation model; S32: loading a standard test working condition into the solver to be evaluated, and solving a flow field feature of the corresponding flow field simulation field; S33: comparing the spatial points and the calculation results of different time points output by the solving model with the corresponding experimental results in the experimental data to obtain the difference between the two , wherein i and j represent different spatial positions and time positions, respectively; S34: calculating the twin degree of the solver and the actual physical process by the following formula: wherein represents the maximum difference in the experimental values on the time scale and the space scale.

2. The integrated CAD and CAE software flow field solver capability assessment method of claim 1, wherein, The S1 comprises: S11: preliminary clustering division is made to the parameters of the imported model, the parameters representing the geometric processing capability are denoted as α1, …, α n , the parameters representing the grid characteristics are denoted as β1, …, β n , the parameters representing the solver characteristics are denoted as γ1, …, γ n , the parameters representing the post-processing characteristics are denoted as δ1, …, δ n , S12: Selecting the parameters that can be directly quantified, separating the positive parameters a + , b + , g + , d + and the negative parameters a - , b - , g - , d - ; S13: finding standard values of all directly quantifiable parameters; S14: performing normalization processing according to the characteristics of the parameters so as to unify the standards of the final evaluation of the parameters; Wherein, the greater the value of the normalized parameter, the better the performance of the corresponding parameter; the greater the positive parameter, the better the performance of the software; the smaller the negative positive parameter, the better the performance of the software.

3. The integrated CAD and CAE software flow field solver capability assessment method of claim 2, wherein, The standard values of all directly quantifiable parameters are obtained by the following method: Obtaining a plurality of standard models in a plurality of target application fields; The standard values are generated based on the simulation or direct operation of the standard physical processes by respective standard models and the parameters corresponding thereto , , , , , , , .

4. The integrated CAD and CAE software flow field solver capability assessment method of claim 3, wherein, The normalization method is as follows: 。 5. The integrated CAD and CAE software flow field solver capability assessment method of claim 4, wherein, The S22 comprises: S221: covering the grid structure completely by using a plurality of cubes with an edge length of a, and recording the coordinates of each vertex; S222: calculating the Hausdorff distance between the cube aggregate in S221 and the grid structure according to the following formula: wherein denotes the Hausdorff distance, A denotes the point set of the aggregates in S221, E denotes the point set of the grid structure; S223: If the calculated value in S222 is less than a preset threshold then the grid structure is considered to be just and completely contained ; S224: If the calculation in S222 Greater than the threshold Then, for each cube with side length a in the geometric aggregate in S221, a dichotomous fractal is performed to remove cubes that fall completely outside the grid structure. S225: looply performing S222 to S224 until less than the threshold value .

6. The integrated CAD and CAE software flow field solver capability assessment method of claim 5, wherein, The experimental data comprises spatial point experimental data and all time data.

7. The integrated CAD and CAE software flow field solver capability assessment method of claim 6, wherein, The S4 comprises: S41: distinguishing parameters having a direct influence on the output of a calculation result, and evaluation parameters having only an indirect influence on the calculation result or being irrelevant to the calculation result, wherein the weight coefficient of the latter is assumed to be 1; S42: in the evaluation parameters having a direct influence on the output of the calculation result, the weight coefficient of the corresponding parameter is calculated according to the sensitivity of the calculation result to the change; S43: combining the weight coefficient, and drawing a polygon diagram for evaluating the software solving capability; S44: all the quantitative indicators calculated in steps S1 to S3 are combined with the weight coefficients in S41, and the indicators are marked on each coordinate dimension of the polygon graph and connected to form a closed curved surface; S45: the total area of the closed curved surface is calculated; The total area of the closed curved surface represents the comprehensive performance of the software solving ability, and the larger the total area is, the stronger the software solving ability is.

8. The integrated CAD and CAE software flow field solver capability assessment method of claim 7, wherein, Among the evaluation parameters that have a direct impact on the calculation results, the weight coefficients of the corresponding parameters are calculated according to the sensitivity of the calculation results to the changes as follows: If the change rate of the evaluation parameter is 10%, the weight coefficient is taken as 1+a%, and the average a% fluctuation of the calculation results is caused.

Images