Design parameter extraction method and system based on geometric features

Abstract

The invention belongs to the technical field of data processing, and provides a design parameter extraction method and system based on geometric features, a geometric model is disassembled in combination with the geometric features, all components in the model and the mutual relation of the components can be more accurately understood and analyzed, and the accuracy of data processing is improved; the geometric design parameters are set according to the geometric feature types, so that the accuracy, reliability and scientificity of the design parameters are ensured, and meanwhile, the efficiency of setting the design parameters can be remarkably improved; the integrity of the geometric model is verified, so that the reasonability of the constructed design parameters can be ensured, and geometric loss cannot be caused; by optimizing the design parameters, the total quantity of the design parameters of the complete geometric model can be effectively reduced, the complexity and the data volume of the model can be remarkably reduced, the maintainability and the expandability of the relevant performance corresponding to the design parameters can be improved, and the cost required for performance evaluation in the later period can be further reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method and system for extracting design parameters based on geometric features. Background Art

[0002] Design optimization plays a crucial role in modern engineering and product development. Its core goal is to improve the design scheme to meet various requirements such as specific performance, cost, and time. In the process of realizing design optimization, the selection of design parameters is an important link that cannot be ignored. Especially in geometric shape design, the selection of these design parameters will directly affect the characteristic shape of the geometry, thus affecting its related performances, such as structural strength, flow characteristics, heat transfer, and electromagnetism. Therefore, design parameters not only affect the performance and quality of the final product but also help improve the efficiency of the design process. Therefore, in-depth exploration of the selection of design parameters is crucial for achieving efficient design optimization.

[0003] Currently, there are two main sources for parameters in geometric design optimization: First, engineers can obtain inspiration from design cases of similar products and quickly determine design parameters. Although this method is efficient, it may lack a scientific basis and is commonly used in conventional designs. Another type is to compare their related performances through experimental data or numerical means and then adjust the design parameters accordingly. This method is widely used in most engineering fields, especially in high-tech fields such as aerospace, automotive, and shipbuilding. By comparing, the related performances of each optimization scheme are analyzed, and then the values of each design parameter are optimized and adjusted. The cycle is iterated until the design requirements are met. Even with the combination of current various optimization algorithms to obtain the final design parameters, the final design parameters often limit the boundary of the optimized related performances. Summary of the Invention

[0004] In order to solve the problems in the prior art that the determination of design parameters overly relies on manual judgment and lacks a scientific and rigorous basis; and the design parameters obtained by adjusting through experimental data or numerical analysis methods often limit the further improvement and optimization of related performances, a method and system for extracting design parameters based on geometric features are proposed, ensuring the accuracy, reliability, and scientific nature of design parameters; in addition, effectively reducing the complexity and data volume of geometric models and significantly improving the maintainability and scalability of the performances corresponding to design parameters.

[0005] The solution is as follows:

[0006] A method for extracting design parameters based on geometric features, the specific steps include:

[0007] S1: Decompose the geometric model: Decompose the original geometric model based on geometric features to obtain corresponding geometric features; the geometric features include points, lines, and planes.

[0008] S2: Set design parameters: Set design parameters for the geometric features extracted in S1; for the points, set design parameters in the form of coordinates; for the lines, first identify the type of the line, and then set design parameters according to the type of the line; for the surfaces, first identify the type of the surface, and then set design parameters according to the type of the surface.

[0009] S3: Combine design parameters and construct a geometric model: Combine all the design parameters set in S2, and reconstruct a geometric model according to the combined design parameters.

[0010] S4: Determine whether the geometric model is complete; Determine whether the geometric model constructed in S3 is a complete geometric model based on the surface coincidence degree detection method; if it is a complete geometric model, mark the design parameters of the currently constructed geometric model as valid, record the current design parameters and the corresponding parameter quantity, and enter step S5; if not, mark the current design parameters as invalid, and return to step S3 to reconstruct the geometric model.

[0011] S5: Optimize the total quantity of design parameters and output the final design parameters: Calculate the total quantity of the design parameters obtained in S4, reduce the quantity of the current design parameters, and repeat steps S3 - S4 to perform optimization iteration on the total quantity of design parameters. When the number of optimization iterations reaches the threshold or the total quantity of the design parameters cannot be reduced compared with the previous output, output the design parameters with the lowest total quantity.

[0012] Preferably, the types of the lines in S2 include: regular lines and fitted lines; the regular lines refer to regular lines that can be described by expressions; the fitted lines refer to curves obtained through fitting techniques; the types of the surfaces include: elementary analytic surfaces and free-form surfaces; the elementary analytic surfaces are surfaces that can be constructed through preliminary analysis, including: planes, spherical surfaces, and cylindrical surfaces; the free-form surfaces are surfaces composed of surfaces that vary freely in a complex manner.

[0013] Preferably, the specific method for setting design parameters for the lines in S2 is:

[0014] A1: When the type of the line is a regular line, construct design parameters through the design expression of the regular line, or set the design parameters of the regular line by combining the design parameters of the points.

[0015] A2: When the type of the line is a fitted line; first construct a regular line, and optimize the regular line through a mathematical model and an optimization method; when the optimized regular line is close to the original fitted line, use the design parameters of the regular line at this time as the design parameters of the fitted line.

[0016] Preferably, the specific method for setting design parameters for the fitted lines is:

[0017] B1: Construct regular lines; select models, and determine the parametric form of the regular lines through the selected models; the models include: polynomial models, exponential models, Fourier series models, spline curve models, and neural network models;

[0018] B2: Set an objective function and quantify the fitting error for the parametric form obtained in A1; optimize the set objective function by the least squares method, absolute error method, or regularization method; when the value of the objective function reaches the set quantified fitting error, use the current parametric form as the design parameters of the fitting line.

[0019] Preferably, the specific method for setting design parameters for the surface in S2 is as follows:

[0020] 1) For a plane, split the plane into lines, and construct the design parameters of the lines according to the type of the lines; form the design parameters of the plane by combining the design parameters of all the lines;

[0021] 2) For a complete spherical surface, set the design parameters through the center coordinates and radius of the sphere; for an incomplete spherical surface, construct the design parameters of the incomplete spherical surface by combining the characteristics of the spherical boundary curve and according to the processing methods of regular lines and fitting lines;

[0022] 3) The cylindrical surface is a surface formed by stretching or sweeping an A curve along one or more B curves; for the cylindrical surface, decompose the cylindrical surface into an A curve and a B curve, and set the design parameters for the A curve and the B curve;

[0023] 4) For a free-form surface, construct the design parameters for the surface boundary of the free-form surface by combining regular lines and fitting lines; at the same time, construct transverse and longitudinal grid curves on the surface to further analyze the surface, and the design parameters of each line need to be constructed by combining regular lines and fitting lines; combine the design parameters of all the constructed lines to form the design parameters of the free-form surface.

[0024] Preferably, the method for judging the integrity of the geometric model in S4 is judged based on the surface coincidence degree detection method; the surface coincidence degree detection method is to perform dissection on the original geometric model and the newly built geometric model based on three-dimensional profiles and three directions of the three-dimensional coordinate system to obtain the dissection lines of the original geometric model and the newly built geometric model at the three-dimensional profile positions, and compare the coincidence degree of the dissection lines of the original geometric model and the newly built geometric model at the three-dimensional profile positions.

[0025] Preferably, the specific method for parameter optimization in S5 is as follows:

[0026] C1: Reduce the total amount of current design parameters; call steps S3 and S4 to determine whether the design parameters after the reduction in quantity can form a complete geometric model; when the optimization iteration count reaches the threshold or the total amount of the design parameters cannot reduce the total number of design parameters compared to the last output, proceed to the next step, otherwise return to C1 to perform the operation of reducing the total amount of parameters; the situation where the total number of design parameters cannot be reduced further means that after reducing the total number of design parameters, the design parameters cannot form a complete geometric model.

[0027] C2: Compare the total amounts of design parameters of all geometric models marked as valid, and output the group with the least total amount of design parameters as the final design parameters.

[0028] A design parameter extraction system based on geometric features, comprising:

[0029] Geometric import module: used to import the original geometric model that needs to optimize the design parameters;

[0030] Geometric decomposition module: used to decompose the original geometric model imported by the geometric import module according to geometric features;

[0031] Point design parameter setting module: used to set design parameters for the points obtained by the geometric decomposition module;

[0032] Line design parameter setting module: a type recognition unit used to identify the type of the lines obtained by the geometric decomposition module; a design parameter setting unit used to set design parameters for the lines according to the line types obtained by the type recognition unit;

[0033] Face design parameter setting module, comprising: a type recognition unit used to identify the type of the faces obtained by the geometric decomposition module; a design parameter setting unit used to set design parameters for the faces according to the line types obtained by the type recognition unit;

[0034] Design parameter combination module: used to combine the design parameters output by the point design parameter setting module, the line design parameter setting module, and the face design parameter setting module, recombine the design parameters for which the geometric integrity verification module's judgment result is no, or combine the design parameters output by the design parameter optimization unit to obtain a geometric model;

[0035] Geometric integrity verification module: used to verify the integrity of the geometric model output by the design parameter combination module;

[0036] Design parameter optimization module: comprising: a judgment unit used to judge whether design parameter optimization is required for the geometric model for which the geometric integrity verification module's judgment result is yes; a design parameter optimization unit used to optimize the design parameters of the geometric model when the judgment result of the judgment unit is yes; an output unit used to output the final design parameters when the judgment result of the judgment unit is no.

[0037] Beneficial effects:

[0038] The present invention provides a method and system for extracting design parameters based on geometric features. By disassembling the geometric model in combination with geometric features, it is possible to more accurately understand and analyze each component in the model and their interrelationships, while helping to reduce errors in data processing and improve the accuracy of data processing. By setting geometric design parameters according to geometric feature types, the accuracy, reliability, and scientific nature of the design parameters are ensured, and at the same time, the efficiency of setting design parameters can be significantly improved. By verifying the integrity of the geometric model, errors and defects in the design can be detected and corrected in a timely manner, such as missing faces, edges, or points, as well as unreasonable geometric relationships, etc., which helps to improve the accuracy and reliability of the design and reduce potential problems caused by design mistakes. By automatically optimizing the design parameters, the total amount of design parameters of the complete geometric model can be effectively reduced, and at the same time, the complexity and data volume of the model can be significantly reduced, and it helps to improve the maintainability and scalability of the relevant performance corresponding to the design parameters, and further reduce the costs required for performance evaluation in the later stage. Description of the drawings

[0039] Figure 1 It is a flowchart of a method for extracting design parameters based on geometric features.

[0040] Figure 2 It is an architecture diagram of a system for extracting design parameters based on geometric features. Detailed implementation manners

[0041] The present invention will be further described below with reference to the drawings and embodiments.

[0042] As Figure 1 shown, the flow of a method for extracting design parameters based on geometric features is as follows:

[0043] S1: Decompose the geometric model: Decompose the original geometric model based on geometric features to obtain corresponding geometric features; the geometric features include points, lines, and faces.

[0044] S2: Set design parameters: Set design parameters for the geometric features extracted in S1; for the points, set the design parameters in the form of coordinates; for the lines, first identify the type of the line, and then set the design parameters according to the type of the line; for the faces, first identify the type of the face, and then set the design parameters according to the type of the face.

[0045] S3: Combine the design parameters and construct the geometric model: Combine all the design parameters set in S2, and reconstruct the geometric model according to the combined design parameters.

[0046] S4: Determine whether the geometric model is complete; based on the surface coincidence detection method, determine whether the geometric model constructed in S3 is a complete geometric model; if it is a complete geometric model, mark the design parameters of the currently constructed geometric model as valid, record the current design parameters and the corresponding number of parameters, and enter step S5; if not, mark the current design parameters as invalid and return to step S3 to reconstruct the geometric model;

[0047] S5: Optimize the total amount of design parameters and output the final design parameters: Calculate the total amount of design parameters obtained in S4, reduce the number of current design parameters, and repeat steps S3 - S4 to optimize and iterate the total amount of design parameters. When the number of optimization iterations reaches the threshold or the total amount of the design parameters cannot be further reduced compared to the previous output, output the design parameters with the lowest total amount.

[0048] The specific implementation process is as follows:

[0049] First step: Decompose the three-dimensional geometry. The three-dimensional geometry can be curves, surfaces, or three-dimensional geometric bodies, and extract them based on points, lines, and surfaces respectively. Points are the endpoints of the three-dimensional geometry or the intersection points of lines. Lines are three-dimensional curves, including straight lines, arcs, curves, etc. Surfaces can be divided into planes, spherical surfaces, cylindrical surfaces, and free-form surfaces, etc.;

[0050] Second step, set design parameters for the extracted points, lines, and surfaces;

[0051] For points, use coordinates to construct design parameters, with a total of N1;

[0052] For lines, there are mainly two processing methods: regular lines and fitted lines;

[0053] Regular lines usually refer to lines with regular shapes. For example, linear curves can be described by specific expressions, such as straight lines, arcs, hyperbolas, parabolas, etc. A total of N2 design parameters are constructed;

[0054] Specifically, for a straight line, it can be determined by its two end points. At this time, the construction of the design parameters of the straight line is transformed into the construction of point design parameters; for a parabola, the construction of the design parameters of this line can be constructed through the parabola function, and at the same time, the start and end positions of this section of the parabola are limited, and the start and end positions can be transformed into the construction of point design parameters.

[0055] Fitted lines mainly refer to curves obtained through various fitting techniques, which usually cannot be described by specific expressions, such as B-Spline curves, NURBS curves, etc. A total of N3 design parameters are constructed;

[0056] If the curve is composed of multiple spliced and integrated surfaces, it can be split first, and then the corresponding curve type can be selected for processing;

[0057] For surfaces, there are mainly four processing methods: The first category is elementary analytic surfaces, such as planes, spheres, and cylinders; the second category is free-form surfaces;

[0058] The first category - elementary analytic surfaces, which are surfaces that can be constructed through preliminary analysis, such as planes, spheres, and cylinders;

[0059] A plane is usually constructed by plane curves, and its boundary is usually composed of regular curves or fitted curves. The design parameter processing method for lines can be combined. If the plane is a regular geometry (such as an equilateral triangle, isosceles triangle, square, rectangle, regular N-sided polygon, circle, sector, etc.), the design parameters can be optimized, and a total of N4 design parameters can be constructed;

[0060] Specifically, since a plane is composed of one or more curves, the design parameters of the lines in the plane can be constructed by borrowing the design parameters of constructing regular lines and fitted lines, and then the design parameters of the plane can be formed;

[0061] A sphere is usually a spherical surface. A complete sphere can be expressed by the center coordinates and radius of the sphere, but an incomplete sphere can construct the design parameters by combining the characteristics of the curves on its boundary and the processing method for lines, with a total of N5;

[0062] A cylinder is usually a surface formed by stretching or sweeping an A curve along another or multiple B curves (straight lines, curves, or three-dimensional curves). Its design parameters can be decomposed into the design parameters of the A curve and the design parameters of the B curve, and can be disassembled into the processing method for lines, with a total of N6 design parameters constructed;

[0063] The second category - free-form surfaces;

[0064] Free-form surfaces cannot be simply analyzed like elementary analytic surfaces, but are surfaces composed of surfaces that freely change in complex ways, such as the shapes of airplanes, cars, and ships. The curves distributed horizontally and vertically are all fitted lines, and the construction of their design parameters is the most complex. Each horizontal or vertical curve is a spliced or fitted curve, and more design parameters need to be constructed to design free-form surfaces, with N7 design parameters constructed;

[0065] If a surface is composed of multiple surfaces spliced and integrated, it can be disassembled first, and then the corresponding surface type can be selected for processing;

[0066] The third step is to combine all the design parameters and reconstruct the geometric model;

[0067] The fourth step is to determine whether the constructed geometric model can completely construct all the geometries by combining the surface coincidence detection method in the patent with the patent number 202311086498.3. If it can, proceed to the next step; if not, return to the previous step to re-iterate and combine the design parameters and reconstruct the geometric model;

[0068] Specifically, the surface coincidence degree detection method is to use a three-dimensional section to divide the original geometric model and the newly created geometric model along three directions of the three-dimensional coordinate system, obtain the section lines of the original geometry and the newly created geometry at the position of the three-dimensional section, and judge the integrity of the newly created geometry and the original geometry by comparing the coincidence degree of these two section lines. The higher the coincidence degree, the better the integrity. It should be noted that in this step, the sectioning needs to be carried out along the three directions of the three-dimensional coordinate system with the smallest possible spacing in turn, and then the section lines of the original geometry and the newly created geometry are compared, and the comparison is repeated in turn until all the section lines of the original geometry and the newly created geometry are compared.

[0069] In the fifth step, with the goal of reducing the total number of design parameters, optimize and reduce the total number of design parameters, and nest the third and fourth steps in the process to ensure the integrity of the geometric model until the total number of design parameters can no longer be reduced.

[0070] In the sixth step, output the final design parameters.

[0071] As Figure 2 shown, a design parameter extraction system based on geometric features includes:

[0072] A geometric import module: used to import the original geometric model that needs to optimize the design parameters.

[0073] A geometric decomposition module: used to decompose the original geometric model imported by the geometric import module according to geometric features.

[0074] A point design parameter setting module: used to set design parameters for the points obtained by the geometric decomposition module.

[0075] A line design parameter setting module: including a type recognition unit for type recognition of the lines obtained by the geometric decomposition module; a design parameter setting unit for setting design parameters for the lines according to the line types obtained by the type recognition unit.

[0076] A surface design parameter setting module: including a type recognition unit for type recognition of the surfaces obtained by the geometric decomposition module; a design parameter setting unit for setting design parameters for the surfaces according to the line types obtained by the type recognition unit.

[0077] A design parameter combination module: used to combine the design parameters output by the point design parameter setting module, the line design parameter setting module and the surface design parameter setting module, or to combine the design parameters output by the design parameter optimization unit to obtain a geometric model.

[0078] A geometric integrity verification module: used to verify the integrity of the geometric model output by the design parameter combination module.

[0079] Design parameter optimization module: It includes: a judgment unit for judging whether design parameter optimization is required for a geometric model that is judged to be yes by the geometric integrity verification module; a design parameter optimization unit for optimizing the design parameters of the geometric model when the judgment result of the judgment unit is yes; and an output unit for outputting the final design parameters when the judgment result of the judgment unit is no.

[0080] Specific implementation process:

[0081] First, import the original geometric model with the help of the geometric import module, and then decompose the newly imported geometric model in combination with the geometric decomposition module to obtain the corresponding geometric features; the geometric features include points, lines, and surfaces.

[0082] For points, construct the corresponding design parameters based on the point design parameter setting module, such as setting parameters using three-dimensional coordinates.

[0083] For lines, construct the corresponding design parameters based on the line design parameter setting module. It should be noted here that the construction of the design parameters of the feature lines needs to be constructed separately according to the line features, such as regular lines and fitted lines.

[0084] For surfaces, construct the corresponding design parameters based on the surface design parameter setting module. First, determine the geometric features of the surface according to the type of the surface, such as plane, spherical surface, cylindrical surface, and free-form surface.

[0085] For a plane, the parameters of the lines forming the plane can be set, and the design parameters of the corresponding lines are used to combine and form the design parameters of the plane.

[0086] For a spherical surface, two cases need to be analyzed: for a complete spherical surface, set the design parameters through the center coordinates and radius of the sphere; for an incomplete spherical surface, combine the characteristics of the spherical surface boundary curve and construct the design parameters according to the processing methods for regular lines and fitted lines.

[0087] For a cylindrical surface, it is a surface formed by stretching or sweeping an A curve along one or more B curves; for a cylindrical surface, decompose the cylindrical surface into an A curve and a B curve, and set the design parameters for the A curve and the B curve.

[0088] For a free-form surface, it is a surface composed of a surface that changes freely in a complex manner. The design parameters of its surface boundary can be constructed by combining regular curves and fitted lines; at the same time, transverse and longitudinal grid curves are constructed on the surface to further analyze the surface, and the design parameters of each curve need to be constructed by combining regular curves and fitted lines. The design parameters of this series of curves together constitute the design parameters of the free-form surface.

[0089] All design parameters of the newly imported geometric model can be obtained through the above steps of parsing, and a new geometric model is reconstructed by the design parameter combination module in the order of points, lines, and surfaces. To ensure that the new geometric model is complete and consistent with the imported geometric model, with the help of the geometric integrity verification module, that is, based on the surface coincidence degree detection method to determine whether the currently constructed geometric model is a complete geometric model;

[0090] When it is confirmed that a geometric model with better integrity can be constructed based on the currently constructed design parameters, the current design parameters are marked as valid design parameters. However, differences in the geometric model construction method will result in different corresponding design parameters, which are directly reflected in the different total numbers of design parameters. To minimize the complexity of the design parameters as much as possible, the design parameter optimization module is used to optimize the total number of design parameters, obtaining the minimum total number of design parameters while ensuring the integrity of the newly built geometric model until the total number of design parameters can no longer be reduced, and outputting it as the final design parameters.

[0091] Finally, it should be noted that the above is only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred arrangement, those of ordinary skill in the art should understand that the technical solution of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for extracting design parameters based on geometric features, characterized in that S1: Decompose the geometric model: Decompose the original geometric model based on geometric features to obtain corresponding geometric features; the geometric features include points, lines, and surfaces. S2: Set design parameters: Set design parameters for the geometric features extracted in S1. For the points, set design parameters in the form of coordinates; for the lines, first identify the type of the line, and then set design parameters according to the type of the line; for the surfaces, first identify the type of the surface, and then set design parameters according to the type of the surface. S3: Combine design parameters and construct a geometric model: Combine all the design parameters set in S2, and reconstruct a geometric model according to the combined design parameters. S4: Determine whether the geometric model is complete; Determine whether the geometric model constructed in S3 is a complete geometric model based on the surface coincidence detection method. If it is a complete geometric model, mark the design parameters of the currently constructed geometric model as valid, record the current design parameters and the corresponding number of parameters, and enter step S5. If not, mark the current design parameters as invalid, and return to step S3 to reconstruct the geometric model. S5: Optimize the total amount of design parameters and output the final design parameters: Calculate the total amount of the design parameters obtained in S4, reduce the number of the current design parameters, and repeat steps S3 - S4 to optimize and iterate the total amount of the design parameters. When the number of optimization iterations reaches the threshold or the total amount of the design parameters cannot be reduced compared with the previous output, output the design parameters with the lowest total amount.

2. The design parameter extraction method based on geometric features according to claim 1, characterized in that The types of the lines in S2 include: regular lines and fitting lines; the regular lines refer to regular lines that can be described by expressions; the fitting lines refer to curves obtained by fitting techniques; the types of the surfaces include: elementary analytical surfaces and free-form surfaces; the elementary analytical surfaces are surfaces that can be constructed through preliminary analysis, including: planes, spheres, and cylinders; the free-form surfaces are surfaces composed of surfaces that vary freely in a complex manner.

3. A method for extracting design parameters based on geometric features according to claim 1, characterized in that, The specific method for setting design parameters for the lines in S2 is as follows A1: When the type of the line is a regular line, construct design parameters by designing an expression for the regular line, or set the design parameters of the regular line by combining the design parameters of the points. A2: When the type of the line is a fitting line; first construct a regular line, and optimize the regular line through a mathematical model and an optimization method; when the optimized regular line is close to the original fitting line, use the design parameters of the regular line at this time as the design parameters of the fitting line.

4. A method for proposing design parameters based on geometric features according to claim 3, characterized in that The specific method for setting design parameters for the fitting lines is as follows B1: Construct a regular line; screen the model, and determine the parametric form of the regular line through the screened model; the models include: polynomial models, exponential models, Fourier series models, spline curve models, and neural network models. B2: Set the objective function and the error of quantization fitting for the parameterized form obtained in A1; optimize the set objective function by the least squares method, the absolute error method or the regularization method; when the value of the objective function reaches the set error of quantization fitting, use the current parameterized form as the design parameter of the fitting line.

5. A method for extracting design parameters based on geometric features according to claim 1, characterized in that The specific method for setting design parameters for a surface described in S2 is as follows: 1) For a plane, split the plane into lines, and construct the design parameters of the lines according to the type of the lines; form the design parameters of the plane by combining the design parameters of all the lines. 2) For a complete spherical surface, set the design parameters by the center coordinates and radius of the sphere; for an incomplete spherical surface, construct the design parameters of the incomplete spherical surface by combining the characteristics of the spherical boundary curve and according to the processing methods of regular lines and fitting lines. 3) The cylindrical surface is a surface formed by stretching or sweeping an A curve along one or more B curves; for the cylindrical surface, decompose the cylindrical surface into an A curve and a B curve, and set the design parameters for the A curve and the B curve. 4) For a free-form surface, construct the design parameters for the surface boundary of the free-form surface by combining regular lines and fitting lines; at the same time, construct transverse and longitudinal grid curves on the surface to further analyze the surface, and each line needs to combine regular lines and fitting lines to construct the design parameters. Combine all the constructed line design parameters to form the design parameters of the free-form surface.

6. A method for extracting design parameters based on geometric features according to claim 1, characterized in that, The method for judging the integrity of the geometric model described in S4 is judged based on the surface coincidence degree detection method; the surface coincidence degree detection method is to divide the original geometric model and the newly built geometric model based on three directions of a three-dimensional section and a three-dimensional coordinate system, obtain the section lines of the original geometric model and the newly built geometric model at the three-dimensional section position, and compare the coincidence degree of the section lines of the original geometric model and the newly built geometric model at the three-dimensional section position.

7. A method for extracting design parameters based on geometric features according to claim 1, characterized in that, The specific method for parameter optimization described in S5 is as follows: C1: Reduce the total amount of the current design parameters; call steps S3 and S4 to judge whether the design parameters after the reduction in quantity can form a complete geometric model; when the optimization iteration times reach the threshold or the total amount of the design parameters cannot reduce the total number of design parameters compared with the previous output, go to the next step, otherwise return to C1 to perform the operation of reducing the total amount of parameters; the situation that the total number of design parameters cannot be reduced further means that after reducing the total number of design parameters, the design parameters cannot form a complete geometric model. C2: Compare the total amounts of the design parameters of all the geometric models marked as valid, and output the group with the least total amount of the design parameters as the final design parameters.

8. A geometric feature-based design parameter extraction system formed by the method according to any one of claims 1-7, characterized in that A geometric import module: used to import the original geometric model that needs to optimize the design parameters. A geometric decomposition module: used to decompose the original geometric model imported by the geometric import module according to geometric features. A point design parameter setting module: used to set design parameters for the points obtained by the geometric decomposition module. A line design parameter setting module: a type recognition unit used to recognize the type of the lines obtained by the geometric decomposition module. A design parameter setting unit that sets design parameters for a line according to the line type obtained by the type recognition unit; A surface design parameter setting module, including: a type recognition unit that performs type recognition on the surface obtained by the geometric decomposition module; a design parameter setting unit that sets design parameters for the surface according to the line type obtained by the type recognition unit; A design parameter combination module: used to combine the design parameters output by the point design parameter setting module, the line design parameter setting module, and the surface design parameter setting module, recombine the design parameters for which the judgment result of the geometric integrity verification module is no, or combine the design parameters output by the design parameter optimization unit to obtain a geometric model; A geometric integrity verification module: used to perform integrity verification on the geometric model output by the design parameter combination module; A design parameter optimization module: including: a judgment unit that judges whether design parameter optimization is required for the geometric model for which the judgment result of the geometric integrity verification module is yes; a design parameter optimization unit that performs parameter optimization on the design parameters of the geometric model when the judgment result of the judgment unit is yes; an output unit that outputs the final design parameters when the judgment result of the judgment unit is no.

Images