Finite element analysis method and system for flow regulating valve geometry design

Abstract

The application provides a flow regulating valve geometry finite element analysis method and system, and relates to the technical field of data processing.The method comprises the following steps: step 1, spatial geometry feature analysis is performed on each data point in a discrete profile coordinate data set, the principal curvature value of each data point on the metal diaphragm sealing surface is solved, the point with the maximum average curvature value on the metal diaphragm sealing surface is determined as a core topography reference point; step 2, a local coordinate system is established with the core topography reference point as the origin, four sealing state points are screened from the discrete profile coordinate data set, the corresponding spatial plane equation is solved, and the spatial intersection line between the spatial plane and the metal diaphragm sealing surface is calculated. Through accurate positioning of the core reference point, definition of the analysis subdomain, mesh division, mechanical solving, leakage rate calculation and parameter optimization, the application effectively improves the dynamic particulate matter problem, improves the sealing performance and micro flow regulating precision.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a finite element analysis method and system for the geometric design of flow control valves. Background Technology

[0002] Currently, when designing flow control valves for use in high-cleanliness environments (such as metal diaphragm valves for semiconductor equipment), the geometric design of the sealing surface between the valve body and the metal diaphragm is crucial, as it directly affects whether the valve can achieve extremely low static leakage rates (e.g., less than 1×). (mbar·L / s); Currently, the design and analysis of this type of valve sealing pair usually relies on general computer-aided design (CAD) software and finite element analysis (FEA) software for overall modeling and simulation.

[0003] Taking the design process of a metal diaphragm valve for semiconductor process gas delivery as an example, after completing the 3D model of the valve body and diaphragm assembly, engineers typically import it into finite element analysis software for overall structural mechanics analysis, including contact pair setting, material property assignment, and boundary condition application (such as fixing constraints and bolt preload). In this process, the geometric model relied upon by the analysis is usually an idealized CAD surface, which fails to effectively integrate the microscopic contour measurement data of the actual machined surface of the parts. Therefore, when establishing the contact area, dividing the mesh, and especially performing local refinement, it is mostly necessary to manually select based on macroscopic geometric features (such as chamfers and fillets) or engineering experience. It is difficult to automatically and accurately identify the local areas that play a decisive role in sealing behavior (such as the location with the largest average curvature and the most likely to occur initial contact or separation) based on microscopic geometric features such as the curvature distribution on the sealing surface. This leads to possible deviations in the simulation of potential microscopic leakage channels in subsequent nonlinear contact analysis, making design optimization and reliable prediction of leakage performance still partially dependent on the later physical prototype processing and repeatability testing, affecting R&D efficiency and cost control. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a finite element analysis method and system for the geometric design of flow control valves, improve the efficiency of finite element analysis, and at the same time ensure that the analysis focuses on the core sealing part, thereby improving the effectiveness of the analysis results.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] Firstly, a finite element analysis method for the geometric design of flow control valves, the method comprising:

[0007] Step 1: Perform spatial geometric feature analysis on each data point in the discrete contour coordinate dataset, solve for the principal curvature value of each data point on the metal diaphragm sealing surface, and determine the point with the largest average curvature value on the metal diaphragm sealing surface as the core morphology reference point.

[0008] Step 2: Establish a local coordinate system with the core morphology reference point as the origin, select four sealing state points from the discrete contour coordinate dataset, solve the corresponding spatial plane equation, and calculate the spatial intersection line between the spatial plane and the sealing surface of the metal diaphragm.

[0009] Step 3: Based on the spatial intersection line, define a closed analysis subdomain on the metal diaphragm sealing surface of the three-dimensional parametric geometric model of the valve body, and mesh the three-dimensional parametric geometric model of the valve body containing the closed analysis subdomain to generate a finite element calculation model.

[0010] Step 4: Apply boundary conditions and load conditions to the finite element calculation model, and perform nonlinear contact mechanics calculation on the finite element calculation model with applied boundary conditions and load conditions to obtain simulation data of microscopic leakage channels in the closed analysis subdomain.

[0011] Step 5: Based on the simulation data of the microscopic leakage channel, calculate the predicted static leakage rate of the contact sealing pair, adjust the design parameters of the three-dimensional parametric geometric model of the valve body, and repeat the process from establishing the three-dimensional parametric geometric model of the valve body to calculating the predicted static leakage rate. Analyze the influence of different geometric design parameters on the predicted static leakage rate and mechanical response, and complete the optimization verification of the geometric design of the valve body and diaphragm seal.

[0012] Secondly, the finite element analysis system for the geometric design of flow control valves includes:

[0013] The geometric feature analysis module is used to perform spatial geometric feature analysis on each data point in the discrete contour coordinate dataset, solve the principal curvature value of each data point on the metal diaphragm sealing surface, and determine the point with the largest average curvature value on the metal diaphragm sealing surface as the core morphology reference point.

[0014] The spatial intersection calculation module is used to establish a local coordinate system with the core topography reference point as the origin, select four sealing state points from the discrete contour coordinate dataset, solve the corresponding spatial plane equation, and calculate the spatial intersection between the spatial plane and the sealing surface of the metal diaphragm.

[0015] The finite element model generation module is used to define a closed analysis subdomain on the metal diaphragm sealing surface of the three-dimensional parametric geometric model of the valve body based on spatial intersection lines, and to mesh the three-dimensional parametric geometric model of the valve body containing the closed analysis subdomain to generate a finite element calculation model.

[0016] The nonlinear contact solution module is used to apply boundary conditions and load conditions to the finite element calculation model, and to perform nonlinear contact mechanics solution calculations on the finite element calculation model with applied boundary conditions and load conditions to obtain simulation data of microscopic leakage channels in the closed analysis subdomain.

[0017] The optimization and verification module is used to calculate the predicted static leakage rate of the contact sealing pair based on the simulation data of the micro leakage channel. This allows for the adjustment of the design parameters of the three-dimensional parametric geometric model of the valve body and the repeated execution of the process from establishing the three-dimensional parametric geometric model of the valve body to calculating the predicted static leakage rate. The module analyzes the influence of different geometric design parameters on the predicted static leakage rate and mechanical response, and completes the optimization and verification of the geometric design of the valve body and diaphragm seal.

[0018] The above-described solution of the present invention has at least the following beneficial effects:

[0019] By analyzing the spatial geometric features of each data point in the discrete contour coordinate dataset, solving for the principal curvature value, and locking the core morphological reference point with the largest average curvature, the key morphological feature area of ​​the metal diaphragm sealing surface can be captured. This core reference point avoids the blindness of the analysis scope. A local coordinate system is established with the core morphological reference point as the origin. Sealing state points are selected and the intersection of the spatial plane and the metal diaphragm sealing surface is solved, thereby defining a closed analysis subdomain. This process can delineate the key areas for sealing performance analysis, avoid indiscriminate analysis of the entire valve body model, reduce unnecessary calculations, improve the efficiency of finite element analysis, and ensure that the analysis focuses on the core sealing part, thus improving the effectiveness of the analysis results. The three-dimensional parametric model of the valve body containing the closed analysis subdomain is meshed and solved by nonlinear contact mechanics, which can realistically simulate the sealing contact state between the metal diaphragm and the valve body. The nonlinear contact characteristics of the sealing pair are fully considered, effectively restoring the real situation of the microscopic leakage channel. The obtained simulation data is more in line with the actual working scenario, ensuring the reliability of the analysis results.

[0020] By calculating and predicting the static leakage rate based on simulation data of microscopic leakage channels, the sealing performance of flow control valves can be predicted in advance, avoiding the lag of relying on physical test verification. The prediction results can be used to adjust the design parameters of the valve body's three-dimensional parametric model, quickly locate key geometric design factors affecting sealing performance, and effectively reduce the risk of internal leakage. By repeatedly executing the process from model establishment to leakage rate calculation, the system analyzes the influence of different geometric design parameters on the predicted static leakage rate and mechanical response, providing a quantitative optimization basis for the geometric design of flow control valves. Based on this law, the geometric parameters of key components such as valve body, metal diaphragm, and threads can be precisely adjusted. This improves flow regulation accuracy, enhances dynamic particulate matter control, strengthens structural mechanical stability, reduces diaphragm fatigue damage, and achieves synergistic optimization of sealing performance, regulation accuracy, and structural reliability, shortening the product design iteration cycle. Attached Figure Description

[0021] Figure 1 This is a schematic flowchart of the finite element analysis method for the geometric design of a flow regulating valve provided in an embodiment of the present invention.

[0022] Figure 2 This is a schematic diagram of a finite element analysis system for the geometric design of a flow regulating valve provided in an embodiment of the present invention. Detailed Implementation

[0023] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0024] like Figure 1 As shown, embodiments of the present invention propose a finite element analysis method for the geometric design of flow control valves, the method comprising the following steps:

[0025] Step 1: Perform spatial geometric feature analysis on each data point in the discrete contour coordinate dataset, solve for the principal curvature value of each data point on the metal diaphragm sealing surface, and determine the point with the largest average curvature value on the metal diaphragm sealing surface as the core morphology reference point.

[0026] Step 2: Establish a local coordinate system with the core morphology reference point as the origin, select four sealing state points from the discrete contour coordinate dataset, solve the corresponding spatial plane equation, and calculate the spatial intersection line between the spatial plane and the sealing surface of the metal diaphragm.

[0027] Step 3: Based on the spatial intersection line, define a closed analysis subdomain on the metal diaphragm sealing surface of the three-dimensional parametric geometric model of the valve body, and mesh the three-dimensional parametric geometric model of the valve body containing the closed analysis subdomain to generate a finite element calculation model.

[0028] Step 4: Apply boundary conditions and load conditions to the finite element calculation model, and perform nonlinear contact mechanics calculation on the finite element calculation model with applied boundary conditions and load conditions to obtain simulation data of microscopic leakage channels in the closed analysis subdomain.

[0029] Step 5: Based on the simulation data of the microscopic leakage channel, calculate the predicted static leakage rate of the contact sealing pair, adjust the design parameters of the three-dimensional parametric geometric model of the valve body, and repeat the process from establishing the three-dimensional parametric geometric model of the valve body to calculating the predicted static leakage rate. Analyze the influence of different geometric design parameters on the predicted static leakage rate and mechanical response, and complete the optimization verification of the geometric design of the valve body and diaphragm seal.

[0030] In this embodiment of the invention, by analyzing the spatial geometric features of each data point in the discrete contour coordinate dataset, solving for the principal curvature value, and locking the core morphological reference point with the largest average curvature, the key morphological feature area of ​​the metal diaphragm sealing surface can be captured. This core reference point avoids the blindness of the analysis range. A local coordinate system is established with the core morphological reference point as the origin, sealing state points are screened, and the intersection of the spatial plane and the metal diaphragm sealing surface is solved, thereby defining a closed analysis subdomain. This process can delineate the key area for sealing performance analysis, avoid indiscriminate analysis of the entire valve body model, reduce invalid calculations, improve the efficiency of finite element analysis, and ensure that the analysis focuses on the core sealing part, thus improving the effectiveness of the analysis results. The three-dimensional parametric model of the valve body containing the closed analysis subdomain is meshed, and combined with nonlinear contact mechanics solution, the sealing contact state between the metal diaphragm and the valve body can be realistically simulated. The nonlinear contact characteristics of the sealing pair are fully considered, effectively restoring the real situation of the microscopic leakage channel. The obtained simulation data is more in line with the actual working scenario, ensuring the reliability of the analysis results.

[0031] By calculating and predicting the static leakage rate based on simulation data of microscopic leakage channels, the sealing performance of flow control valves can be predicted in advance, avoiding the lag of relying on physical test verification. The prediction results can be used to adjust the design parameters of the valve body's three-dimensional parametric model, quickly locate key geometric design factors affecting sealing performance, and effectively reduce the risk of internal leakage. By repeatedly executing the process from model establishment to leakage rate calculation, the system analyzes the influence of different geometric design parameters on the predicted static leakage rate and mechanical response, providing a quantitative optimization basis for the geometric design of flow control valves. Based on this law, the geometric parameters of key components such as valve body, metal diaphragm, and threads can be precisely adjusted. This improves flow regulation accuracy, enhances dynamic particulate matter control, strengthens structural mechanical stability, reduces diaphragm fatigue damage, and achieves synergistic optimization of sealing performance, regulation accuracy, and structural reliability, shortening the product design iteration cycle.

[0032] In another preferred embodiment of the present invention, the process of obtaining the discrete contour coordinate dataset includes:

[0033] Step 001: Based on the structural design parameters of the flow control valve, establish a three-dimensional parametric geometric model of the valve body, including the contact sealing pair between the metal diaphragm and the valve body. Specifically, this includes: First, clarifying the core structural design parameters of the flow control valve to ensure that the parameter values ​​meet the actual needs of semiconductor industry applications. Assume the overall dimensions of the valve body are set to a length of 80 mm, a width of 50 mm, and a height of 60 mm, with an internal flow channel diameter of 10 mm and a flow channel length of 40 mm; the metal diaphragm is made of nickel-cobalt alloy with an elastic modulus of 2. The valve body has a pressure of 0.00 GPa, a Poisson's ratio of 0.3, a diaphragm thickness of 0.5 mm, and an initial shape of a spherical arc surface with a radius of 30 mm. The contact area between the valve body and the metal diaphragm sealing pair is annular, with an inner diameter of 8 mm and an outer diameter of 12 mm. The valve stem diameter is 8 mm, and the basic pitch of the double thread is 1.5 mm, with a pitch difference of 0.02 mm. The steel wire diameter of the spring above the diaphragm is 1 mm, the number of spring coils is 6, and the free length is 20 mm. Based on the above-determined structural design parameters, a 3D modeling software (…) was used. Modeling is performed using tools such as SolidWorks and UG. First, the main structure of the valve body is constructed. According to the designed external dimensions and internal flow parameters, the outline and internal cavity of the valve body are drawn, and the annular sealing surface structure that contacts the metal diaphragm on the valve body is defined. Then, based on the thickness of the metal diaphragm, the initial spherical arc shape, and material-related parameters, a nickel-cobalt metal diaphragm model is constructed to ensure that the size of the annular contact area of ​​the diaphragm is completely matched with the size of the valve body sealing surface. Next, the valve stem structure is built. Based on the basic pitch of the double thread and the pitch difference parameter of 0.02 mm, the double thread structure on the valve stem is drawn to ensure that the tooth profile and tooth height of the two threads are consistent and the pitch difference is accurate. Finally, the spring above the diaphragm is assembled. The spring model is constructed according to the spring wire diameter, spring coil number, and free length parameters. The spring, metal diaphragm, valve stem, and valve body are precisely assembled so that the spring axis coincides with the diaphragm center axis, and the metal diaphragm and valve body sealing surface are tightly fitted to form a contact sealing pair. Finally, the three-dimensional parametric geometric model of the valve body, including the contact sealing pair between the metal diaphragm and the valve body, is established.

[0034] Step 002: Based on the three-dimensional parametric geometric model of the valve body, continuous geometric contour curve data is collected on the sealing surface of the metal diaphragm along the normal cross-sectional path of the sealing contour using a geometric engine or measuring device. Specifically, this includes: based on the three-dimensional parametric geometric model of the valve body established in Step 001, firstly, accurately locating the sealing surface of the metal diaphragm in the model, defining the sealing surface as an annular region with an inner diameter of 8 mm and an outer diameter of 12 mm, and determining the sealing contour as the inner and outer edge curves of the annular shape. Next, determining the normal cross-sectional path along the sealing contour, i.e., with the center of the annular sealing surface as the center, uniformly setting 5 normal cross-sectional paths along the radial direction of the annular shape, each path perpendicular to the... The plane containing the sealing profile must cover the entire sealing area from the inner edge to the outer edge of the ring. This ensures that the collected data can fully reflect the geometric features of different radial positions on the sealing surface. Then, using the geometry engine built into 3D modeling software (such as the curve extraction function of SolidWorks) or a high-precision laser measuring device with an accuracy of 0.0001 mm, the metal diaphragm sealing surface is continuously scanned along five defined normal cross-sectional paths. The geometric profile curve data on each path is recorded in real time. These data must fully present the spherical arc surface undulations and geometric shape of the sealing surface. After integration, a continuous geometric profile curve dataset covering the entire sealing area is formed.

[0035] Step 003: For continuous geometric contour curve data, data points are sampled using a fixed step size to ensure that the spacing between adjacent sampling points on the curve path is equal. This transforms the continuous contour curve into a series of discrete spatial coordinate points, generating a discrete contour coordinate dataset of the metal diaphragm surface to be analyzed. Specifically, this includes: firstly, preprocessing the continuous geometric contour curve data collected in step 002 by using a moving average filter to remove abnormal data points that deviate from the normal spherical arc surface contour due to scanning errors, ensuring the accuracy and continuity of the data; then, determining the value of the fixed sampling step size based on the geometric complexity of the metal diaphragm sealing surface and the accuracy requirements of subsequent analysis. Since the metal diaphragm sealing surface is a regular spherical arc surface with moderate geometric complexity, and moderate-precision geometric feature analysis and finite element simulation are required subsequently, the fixed sampling step size is set to 0.005 mm. The method for calculating this fixed sampling step size is as follows: firstly, the total length of the five continuous geometric contour curves is obtained through the measurement function of the 3D modeling software. Assuming the total length of a certain curve is... Based on a fixed sampling step size of 0.005 mm, calculate the required number of sampling points. The calculation formula is: = ÷0.005 mm, if the calculated value is... If the value is not an integer, round up to determine the number of sampling points. Recalculate the spacing between adjacent sampling points to ensure that the spacing does not exceed 0.005 mm and is closest to this value. Perform the above calculation process on each of the five curves to determine the number of sampling points corresponding to each curve. According to the final determined fixed sampling step size and number of sampling points, starting from the starting point of each continuous geometric contour curve, select sampling points sequentially. After selecting each sampling point, verify the spacing between the point and the previous sampling point using 3D modeling software to ensure that the spacing between adjacent sampling points is strictly controlled within the range of 0.005 mm ± 0.0001 mm. Transform each continuous contour curve into a series of ordered discrete spatial coordinate points. Finally, organize all discrete spatial coordinate points and record the 3D spatial coordinate values ​​(x-axis coordinate, y-axis coordinate, z-axis coordinate) of each point in the sampling order of each curve. After integration, form a discrete contour coordinate dataset covering the entire sealing surface of the metal diaphragm.

[0036] This embodiment establishes a precise three-dimensional parametric geometric model of the valve body, providing a reliable model foundation for subsequent geometric feature analysis and finite element simulation, ensuring the accuracy of subsequent analysis results. The standardized acquisition of continuous geometric contour curve data can comprehensively capture the true geometric morphology of the metal diaphragm sealing surface. By using a fixed step size to sample data and generate a discrete contour coordinate dataset, the discretized coordinate points can uniformly and accurately reflect the geometric features of the continuous contour, helping to improve the sealing performance and flow regulation accuracy of the flow control valve, while ensuring the standardization and repeatability of the entire analysis process.

[0037] In a preferred embodiment of the present invention, step 1 includes:

[0038] Step 100: Based on the discrete contour coordinate dataset, for each data point in the dataset, find all data points whose Euclidean distance to that data point is less than or equal to a preset radius threshold, to form a set of neighboring data points for that data point. Specifically, this includes: First, determining the value of the preset radius threshold. Considering the discrete sampling density and geometric feature scale of the metal diaphragm sealing surface, and taking into account that the subsequent quadratic surface fitting needs to ensure a sufficient number of neighboring points to improve fitting accuracy, while avoiding too many irrelevant points interfering with the calculation, the preset radius threshold is set to 0.01 mm. Then, the discrete contour coordinate dataset of the metal diaphragm sealing surface generated in step 003 is retrieved. Each data point in the dataset is selected as a target data point. For the currently selected target data point, its three-dimensional spatial coordinates (x-axis coordinate, y-axis coordinate, z-axis coordinate) are extracted. Then, the three-dimensional spatial coordinates of all other data points in the dataset are extracted in turn. Then, the Euclidean distance between the target data point and each other data point is calculated. All data points whose Euclidean distance from the target data point is less than or equal to 0.01 mm are selected and together with the target data point form the set of neighboring data points of the target data point. The above operation is repeated until the corresponding set of neighboring data points is constructed for each data point in the discrete contour coordinate dataset.

[0039] Step 101: Using the spatial coordinates of the data point and all data points in its neighboring data point set, fit a quadratic surface equation using the least squares method, minimizing the sum of squared distances from the surface represented by the quadratic surface equation to the data point and all data points in its neighboring data point set. Based on the quadratic surface equation, calculate the two principal curvature values ​​of the data point on the surface. Specifically, for each target data point and its neighboring data point set constructed in step 100, first summarize the three-dimensional spatial coordinate values ​​of all data points in the set to form a coordinate dataset for fitting calculation. Then, use the least squares method to fit the quadratic surface equation. The specific formula for the quadratic surface equation is... Here, a, b, c, d, e, f, g, h, i, and j are all equation coefficients. The core objective during the fitting process is to adjust these coefficients so that the sum of squared distances from the surface represented by the equation to each data point in the set reaches its minimum. Specifically, the calculation method is as follows: first, calculate the distance from each data point to the surface equation corresponding to the assumed coefficients; then, square each distance and sum them. Iteratively adjust the values ​​of each coefficient, recalculating the sum of squared distances after each adjustment, until the sum reaches its minimum value. The equation formed by these coefficients is the fitted quadratic surface equation. After obtaining the quadratic surface equation, calculate the two principal curvature values ​​of the target data point on the surface. The calculation process is as follows: first, for the quadratic surface... The equations are used to calculate the first and second partial derivatives along the x and y axes, respectively. The first-order fundamental quantities of the surface at the target data point are then calculated using the first-order partial derivatives (including E, F, and G, where E is the sum of squares of the partial derivatives along the x-axis, F is the dot product of the partial derivatives along the x and y axes, and G is the sum of squares of the partial derivatives along the y-axis). The second-order fundamental quantities of the surface at the target data point are then calculated using the second-order partial derivatives (including L, M, and N, where L is the dot product of the second-order partial derivative along the x-axis and the normal vector, M is the dot product of the mixed second-order partial derivative along the x and y axes and the normal vector, and N is the dot product of the second-order partial derivative along the y-axis and the normal vector). Finally, a curvature matrix is ​​constructed based on the first and second-order fundamental quantities. The expression for the curvature matrix is ​​as follows: By solving for the eigenvalues ​​of the curvature matrix, two eigenvalues ​​are obtained. These two eigenvalues ​​are the two principal curvature values ​​of the target data point on the surface. The specific process is as follows: First, construct the characteristic equation det(λI-K)=0, where λ is the eigenvalue, I is the identity matrix of the same order as the curvature matrix, K is the curvature matrix constructed above, and det represents calculating the determinant of the matrix. After substituting the curvature matrix into the characteristic equation, expand the determinant to obtain a quadratic algebraic equation in λ. ,in , , The constant term obtained by calculating through determinant expansion ( =1, =-(Sum of the main diagonal elements of the curvature matrix) = determinant of curvature matrix); then use the quadratic equation root-finding formula λ=[- ± Substitute , , The specific values ​​are calculated to obtain two roots, which are the two eigenvalues ​​of the curvature matrix, that is, the two principal curvature values ​​of the target data point on the surface.

[0040] Step 102: Traverse each data point in the discrete contour coordinate dataset, repeatedly performing the steps of finding the nearest neighbor set, fitting the quadratic surface equation, and calculating the principal curvature values ​​to obtain two principal curvature values ​​for each data point on the metal diaphragm sealing surface. Specifically, this includes: following the method for constructing the nearest neighbor set in step 100 and the method for fitting the quadratic surface equation and calculating the principal curvature values ​​in step 101, traversing each data point in the discrete contour coordinate dataset. During the traversal, for each data point, sequentially performing the steps of finding the nearest neighbor set, summarizing coordinate data, and fitting the quadratic surface equation based on the least squares method. The complete process involves formulating the quadratic surface equation, calculating the first and second fundamental quantities, constructing the curvature matrix, and solving for the principal curvature values ​​using the characteristic equation and the quadratic equation root-finding formula. This ensures that each data point can obtain the corresponding two principal curvature values. The process continues until all data points in the dataset have been processed, ultimately obtaining the two principal curvature values ​​for each data point on the metal diaphragm sealing surface.

[0041] Step 103: Based on the two principal curvature values ​​of each data point on the metal diaphragm sealing surface, calculate the average curvature value of each data point. Compare the average curvature values ​​of all data points on the metal diaphragm sealing surface, and determine the data point with the largest average curvature value as the core morphological reference point of the metal diaphragm sealing surface. Specifically, for each data point on the metal diaphragm sealing surface, calculate the average curvature value of the data point based on the two principal curvature values ​​obtained in step 102. The calculation method is to add the two principal curvature values, divide the sum by 2, and the quotient is the average curvature value of the data point. After the average curvature values ​​of all data points are calculated, compare all average curvature values, sort them according to their numerical values, and select the data point with the largest average curvature value, and determine this data point as the core morphological reference point of the metal diaphragm sealing surface.

[0042] In this embodiment, by setting a preset radius threshold and accurately calculating the Euclidean distance, the constructed set of neighboring data points can accurately reflect the local geometric environment of each data point. The least squares method is used to fit the quadratic surface equation and calculate the principal curvature value, ensuring the accuracy of the surface fitting and the calculation of the principal curvature value, thus truly reflecting the local geometric features of the metal diaphragm sealing surface. By traversing all data points to obtain complete principal curvature and average curvature values, a comprehensive data foundation is provided for determining the core morphology reference points, making the selection of core morphology reference points more scientific and reasonable. This allows for precise positioning of the key geometric positions on the metal diaphragm sealing surface, providing accurate reference for subsequent steps such as establishing a spatial coordinate system and screening sealing state points, thereby helping to improve the sealing performance of the flow control valve and the effectiveness of geometric design optimization.

[0043] In a preferred embodiment of the present invention, step 2 includes:

[0044] Step 200: Establish a local Cartesian coordinate system with the core morphology reference point as the origin and the directions of maximum and minimum principal curvature at the core morphology reference point as the coordinate axes. Specifically, this includes: First, retrieving the core morphology reference point of the metal diaphragm sealing surface determined in step 103. Using the coordinate query function of the 3D modeling software, accurately extract the global 3D spatial coordinates (X0, Y0, Z0) of this point and directly set this point as the origin of the local Cartesian coordinate system. Next, retrieve the two principal curvature values ​​corresponding to the core morphology reference point recorded in step 102, and record them as follows: and The direction of curvature is determined by comparing the numerical values. > ,but The corresponding tangent direction is the direction of maximum principal curvature. The corresponding tangent direction is the direction of minimum principal curvature; if > Conversely, if the direction is not directly proportional to the curvature, then the direction is directly proportional to the curvature. To ensure accurate direction determination, it is necessary to combine the spherical arc surface characteristics of the metal diaphragm sealing surface to verify the consistency between the curvature direction and the tangent direction of the arc surface. Then, taking the core morphology reference point as the origin, the determined maximum principal curvature direction is extended along the tangent of the arc surface and set as the positive X-axis direction of the local rectangular coordinate system. The minimum principal curvature direction is extended along another tangent of the arc surface perpendicular to the maximum principal curvature direction and set as the positive Y-axis direction. Then, according to the right-hand screw rule, the direction perpendicular to the XY plane, with the thumb pointing to the positive X-axis direction, the index finger pointing to the positive Y-axis direction, and the middle finger naturally pointing, is set as the positive Z-axis direction. This completes the establishment of the local rectangular coordinate system. This coordinate system is precisely matched with the geometric features of the metal diaphragm sealing surface, providing a dedicated coordinate reference for the subsequent geometric analysis of the local area.

[0045] Step 201: In the local Cartesian coordinate system, from the discrete contour coordinate dataset, select the point with the largest curvature value along the positive direction of the maximum principal curvature as the first sealed state point; select the point with the smallest curvature value along the positive direction of the minimum principal curvature as the second sealed state point; select the point with the largest absolute value of the rate of change of the maximum principal curvature as the third sealed state point; and select the point closest to the third sealed state point and with the opposite sign of the rate of change of the maximum principal curvature as the fourth sealed state point. Specifically, this includes: In the established local Cartesian coordinate system, firstly, transform the global three-dimensional coordinate values ​​of all data points in the discrete contour coordinate dataset to obtain the three-dimensional coordinate values ​​of each data point in the local Cartesian coordinate system. The transformation process is as follows: Local X-axis coordinate value = global X-axis coordinate value of data point - global X-axis coordinate value of origin (X0); Local Y-axis coordinate value = global Y-axis coordinate value of data point - global Y-axis coordinate value of origin (Y0); Local Z-axis coordinate value = global Z-axis coordinate value of data point - global Z-axis coordinate value of origin (Z0), completing the coordinate transformation of all data points. Standardization processing is performed; subsequently, four sealing state points are selected according to the following specific process: traversing the local coordinate values ​​of all data points, selecting data points with positive local X-axis coordinate values ​​(i.e., along the positive direction of the maximum principal curvature), and counting the total number of such data points to ensure sufficient sample size; extracting the principal curvature value corresponding to each data point that meets the condition, comparing the magnitude of these principal curvature values ​​one by one, recording the specific data point corresponding to the principal curvature value with the largest value, determining it as the first sealing state point, and saving the local coordinate value of this point; focusing on data points with positive local Y-axis coordinate values ​​(i.e., along the positive direction of the minimum principal curvature), similarly counting the total number of such data points; extracting the principal curvature value corresponding to each data point, selecting the specific data point corresponding to the principal curvature value with the smallest value through comparison, determining it as the second sealing state point, and simultaneously saving its local coordinate value; for each data point in the discrete contour coordinate dataset, determining its previous and next adjacent data points according to the sampling order, extracting the maximum principal curvature values ​​corresponding to these three data points, and recording them as follows. , , ; Calculate the first difference = - The second difference = - Add the absolute values ​​of the two differences and divide by 2 to obtain the maximum principal curvature change rate of the data point; iterate through all data points, calculate and record the maximum principal curvature change rate of each data point, compare the absolute values ​​of all change rates, and select the data point with the largest absolute value as the third sealing state point; use the Euclidean distance calculation method in step 100 to calculate the Euclidean distance between all data points in the discrete contour coordinate dataset and the third sealing state point; sort by distance from smallest to largest, and select the 5 closest data points as candidate data points (ensure sufficient filtering space); extract the sign (positive or negative) of the maximum principal curvature change rate of the third sealing state point, and then calculate the maximum principal curvature change rate of the 5 candidate data points respectively, compare their signs with the sign of the third sealing state point, and select the candidate data point with the opposite sign; if there are multiple candidate data points with opposite signs, select the one with the smallest Euclidean distance to the third sealing state point and determine it as the fourth sealing state point.

[0046] Step 202: Based on the spatial coordinates of the first, second, third, and fourth sealing state points, the spatial plane equation passing through the four sealing state points is obtained using a least-squares plane fitting algorithm. Specifically, this includes: first, extracting the three-dimensional coordinates of the first, second, third, and fourth sealing state points in the local Cartesian coordinate system, denoted as (x1, y1, z1), (x2, y2, z2), (x3, y3, z3), and (x4, y4, z4), respectively; and then using the least-squares plane fitting algorithm to solve for the spatial plane equation. The spatial plane equation has the following form: x+ y+ z+ =0, where , , , The coefficients of the equations are shown below. The fitting process is as follows: Substitute the coordinates of the four sealed points into the plane equations to obtain four equations, forming a system of equations. Define the error function as the sum of the squares of the left-hand side values ​​after substituting all data points into the equations, i.e. (i=1, 2, 3, 4), the core objective is to adjust , , , The value of is such that the error function reaches its minimum; the error function is respectively about _ . , , , Find the partial derivatives and set them all equal to 0 to obtain a system of linear equations; solve this system of linear equations to calculate... , , , Substituting the specific values ​​into the plane equation, we obtain the spatial plane equation passing through the four sealed state points.

[0047] Step 203 involves calculating the intersection of the spatial plane equation with the metal diaphragm sealing surface represented by the discrete contour coordinate dataset using a surface reconstruction method. This yields the spatial intersection line between the spatial plane and the metal diaphragm sealing surface. Specifically, this includes: First, based on the discrete contour coordinate dataset, using cubic spline interpolation, a complete surface model of the metal diaphragm sealing surface is constructed. The reconstruction process involves determining the number of segments based on the total number of points in the discrete contour coordinate dataset. If the dataset contains… If there are 10 data points, they will be divided into 10 data points according to the sampling order. - Two segments are defined, each containing three consecutive adjacent data points, ensuring that every two adjacent segments share an intermediate data point, thus guaranteeing the continuity and integrity of the segments. For each segment with three adjacent data points, let the coordinates of the three points be (x0, z0), (x1, z1), and (x2, z2) respectively. A univariate cubic spline interpolation function is constructed with the local X-axis as the independent variable and the Z-axis as the dependent variable. The function formula is S(x) = x³+ x²+ x+ ( Indicates the current segment number. =1, 2, ..., -2); Next, a system of equations is established using boundary conditions and continuity conditions, as follows: boundary conditions are set, first segment ( The starting point of the interpolation function (x0, z0) is the first data point in the entire dataset, where the first derivative is equal to 0. The first derivative of the interpolation function in the first segment is then calculated to equal 0; the last segment (x0, z0) is the first data point in the entire dataset. = -2) The endpoint is the last data point in the entire dataset. , The first derivative of the function is equal to 0. The first derivative of the interpolation function in the last segment is also equal to 0. The continuity condition is set for any two adjacent segments (the first segment, the second segment, the third segment, the fourth segment, the fifth segment, the sixth segment, the seventh ... Section and the +1 segment, =1, 2, ..., -3), the shared point is ( , The interpolation function must satisfy three conditions: first, the function values ​​must be equal; second, the first derivatives must be equal; and third, the second derivatives must be equal. The data point constraint within each segment is that the interpolation function must pass through both endpoints of that segment, i.e., the first... Section ( , )and( , The above boundary conditions, continuity conditions, and segmented constraints are integrated to form a system containing 4 ( A system of equations consisting of -2 equations is solved using matrix solving or elimination methods to obtain the coefficients of the interpolation function for each segment. , , , Then, the function calculates the coordinates of the midpoint between two adjacent data points within the current segment. The distance between the midpoints remains consistent with the fixed sampling step size (0.005 mm) in step 003. This process is repeated for all data points in sequence. Interpolation calculations for -2 segments supplement the geometric information between adjacent data points, ultimately forming a continuous, smooth, and unbroken sealing surface model, ensuring that the model can completely reproduce the spherical arc surface geometry of the metal diaphragm sealing surface.

[0048] Then, the spatial plane equation obtained in step 202 is intersected with the reconstructed metal diaphragm sealing surface model. The specific calculation process is as follows: the sealing surface model is divided into a grid with an interval of 0.001 mm to obtain several uniformly distributed discrete points, and the three-dimensional coordinate values ​​of each discrete point are extracted; the coordinate values ​​of each discrete point are substituted into the spatial plane equation, and the result value on the left side of the equation is calculated; a small error range is set to ±0.0001. If the calculated result value is within this error range, the discrete point is determined to be the intersection point of the spatial plane and the sealing surface; all discrete points after grid division are traversed, all intersection points that meet the conditions are selected, and the intersection points are sorted according to the geometric contour of the sealing surface. Adjacent intersection points are connected by straight lines to form a continuous and closed curve. This curve is the spatial intersection line between the spatial plane and the metal diaphragm sealing surface.

[0049] This embodiment establishes a local rectangular coordinate system based on core morphological reference points, which can accurately match the key geometric positions of the metal diaphragm sealing surface. The standardized screening process of four sealing state points ensures that the selected points can comprehensively reflect the key geometric features of the sealing surface, providing representative data support for fitting the spatial plane equation and improving the accuracy of the equation. The least squares plane fitting algorithm is used to solve the spatial plane equation, and the spatial intersection line is obtained by combining surface reconstruction and intersection calculation. This can capture the key geometric contour of the metal diaphragm sealing surface, helping to improve the accuracy of flow control valve sealing performance analysis and geometric design optimization.

[0050] In a preferred embodiment of the present invention, step 3 includes:

[0051] Step 300: Calculate the vertical projection of the spatial intersection line onto the metal diaphragm sealing surface of the three-dimensional parametric geometric model of the valve body to obtain the projection boundary line. Based on the projection boundary line, define the closed region on the metal diaphragm sealing surface defined by the projection boundary line as a closed analysis subdomain. Specifically, this includes: First, retrieving the spatial intersection line obtained in step 203 and the three-dimensional parametric geometric model of the valve body established in step 001, and accurately locating the metal diaphragm sealing surface using the feature query function of the three-dimensional modeling software. This surface is a spherical arc surface made of nickel-cobalt alloy, with an inner diameter of 8 mm and an outer diameter of 12 mm in the annular contact area. Determine its spatial orientation (parallel and fitted with the valve body sealing surface) and geometric boundaries (inner and outer edges of the annular ring) in the overall model. Next, calculate the vertical projection of the spatial intersection line onto the metal diaphragm sealing surface. That is, given that the metal diaphragm sealing surface is a spherical arc surface, first... The coordinates (Xc, Yc, Zc) of the center of the spherical arc surface are measured using 3D modeling software, and the coordinates (Xr, Yr, Zr) of the core morphology reference point are extracted. The coordinate differences between the center of the sphere and the core morphology reference point are calculated, i.e., ΔX = Xr - Xc, ΔY = Yr - Yc, ΔZ = Zr - Zc. The vector (ΔX, ΔY, ΔZ) formed by these coordinate differences is used as the normal vector of the metal diaphragm sealing surface, ensuring that the normal vector points to one side of the valve body sealing surface and is perpendicular to the tangent plane of the sealing surface. For each discrete point on the spatial intersection line, its 3D coordinates (Xp, Yp, Zp) are extracted sequentially. Starting from this point, the equation of the perpendicular line is constructed along the direction of the normal vector. The parametric equations of the line are X = Xp + t × ΔX, Y = Yp + t × ΔY, Z = Zp + t × ΔZ (t is a parameter). At the same time, the equation of the spherical arc surface of the metal diaphragm sealing surface is retrieved. (R is the radius of the spherical arc surface, i.e., 30 mm); Substitute the linear parametric equation into the spherical arc surface equation to find the unique valid solution for parameter t (ensure that the intersection point is located within the annular contact area of ​​the sealing surface); Substitute the solution for t into the linear parametric equation to calculate the coordinates (Xq, Yq, Zq) of the intersection point between the perpendicular line and the sealing surface. This intersection point is the projection point of the spatial intersection point.

[0052] Following the original point sequence of the spatial intersection line, the projection points of all discrete points are calculated sequentially to ensure a one-to-one correspondence between the projection points and the original intersection line points. The coordinates of all projection points are verified, and abnormal projection points that exceed the annular contact area (inner diameter 8 mm, outer diameter 12 mm) of the sealing surface are removed. The projection points corresponding to the abnormal points are recalculated. The verified projection points are connected end to end in the corresponding order to form a continuous closed curve without any breaks. This curve is the projection boundary line. Finally, a closed analysis subdomain is defined based on the projection boundary line: First, the complete outline of the projection boundary line is locked on the metal diaphragm sealing surface using the region selection function of the 3D modeling software. It is confirmed that the boundary line is a continuous closed curve and is completely located within the annular contact area of ​​the sealing surface. Within an inner diameter of 8 mm and an outer diameter of 12 mm; then, using this projected boundary line as the sole contour boundary, the software's closed region division function is activated to define the entire area enclosed by the boundary line as a closed analysis subdomain; during the division process, the software's ranging tool must be used to verify that the outer edge of the subdomain completely coincides with the projected boundary line, the inner edge covers the core contact area between the metal diaphragm and the valve body sealing surface (the middle area of ​​the annular contact area, with a width of 2 mm), and the entire subdomain does not exceed the annular range of the sealing surface; the final closed analysis subdomain can accurately delineate the key parts of the sealing performance analysis, providing a clear and precise analysis range for subsequent finite element analysis to focus on the core contact area and eliminate interference from irrelevant areas.

[0053] Step 301: Based on the three-dimensional parametric geometric model of the valve body containing the closed analysis subdomain, tetrahedral elements are used for mesh generation. Specifically, the region inside the closed analysis subdomain is meshed using a first preset size; the region of the mesh element at the projected boundary line and the region of the mesh element sharing a common boundary with that mesh element is meshed using a second preset size; and the remaining region in the three-dimensional parametric geometric model of the valve body that does not belong to the region inside the closed analysis subdomain or the region of the projected boundary line is meshed using a third preset size. This generates a finite element calculation model with gradient mesh density. Specifically, this includes: First, clarifying the core parameters and preset sizes for mesh generation. This involves considering the spherical arc surface geometric accuracy requirements of the metal diaphragm sealing surface (requiring accurate reproduction of surface undulations at the 0.005 mm level), the actual range of the closed analysis subdomain (a circular region with a diameter of approximately 4 mm), and the computational efficiency requirements for subsequent nonlinear contact mechanics simulations. After verification with multiple sets of parameters, the first preset size is set to 0.01 mm, and the second… The preset size is 0.005 mm, and the third preset size is 0.02 mm. Tetrahedral elements are selected for mesh generation. These elements have good geometric adaptability and can accurately fit the contours of complex structures such as valve bodies, metal diaphragms, and springs, while avoiding mesh distortion caused by structural abrupt changes, ensuring that the mesh quality meets the simulation calculation requirements. Subsequently, mesh generation is performed in stages according to regions. The specific process is as follows: First, when dividing the internal region of the closed analysis subdomain, for the core contact area of ​​the metal diaphragm sealing surface (the 2 mm wide area in the middle of the annular contact area) and the corresponding internal structure of the diaphragm covered by the closed analysis subdomain, tetrahedral elements with a first preset size of 0.01 mm are used for uniform mesh generation. During the generation, the mesh control function of the 3D modeling software is used to set the element side length deviation to not exceed ±0.001 mm, ensuring that the mesh density of the core analysis area is uniform and high enough to accurately capture the micro-stress distribution and mechanical response such as contact pressure changes at the sealing contact, providing an accurate mesh foundation for subsequent leakage channel simulation.

[0054] When dividing the projection boundary region, the software's mesh query function is first used to identify all mesh cells (approximately 30 to 50 cells) containing the projection boundary line. Then, the adjacent cell search function is used to find adjacent mesh cells (extending 1 to 2 layers of cells on each side) that share an edge or face with these cells, collectively forming the projection boundary region. This region is then densified using tetrahedral cells of a second preset size of 0.005 mm, which is half the size of the first preset size. This densification process ensures a smooth transition of the boundary region's mesh from the high density of the core subdomain to the low density of the outer region, avoiding abrupt changes in mesh size. The stress concentration caused by the change reduces simulation error. When dividing the remaining regions, after removing the internal regions of the closed analysis subdomain and the projected boundary line regions in the three-dimensional parametric geometric model of the valve body, the remaining non-core structural regions such as the valve body body, valve stem, and spring are divided using tetrahedral elements with a third preset size of 0.02 mm. This size is twice that of the first preset size. Under the premise of ensuring that the calculation accuracy of the non-core regions meets the overall simulation requirements (error not exceeding 5%), the total number of meshes is reduced (by about 40% to 50%), improving mesh generation efficiency and subsequent simulation calculation speed, and shortening the overall analysis cycle.

[0055] Finally, after completing the full model mesh generation, mesh quality checking and optimization are carried out. This involves using the mesh quality analysis function of the 3D modeling software to calculate key indicators such as distortion rate, aspect ratio, and warpage of all mesh elements. The specific calculation methods for each indicator are as follows: Distortion rate is based on the ideal shape of the mesh element (tetrahedral shape). First, the volume of the actual element is calculated using the volume measurement function of the 3D modeling software. Then, according to the ideal tetrahedral volume formula, volume = ( ) × the edge length of the ideal tetrahedron³ (where the edge length of the ideal tetrahedron is the same as the average edge length of the corresponding actual mesh element, and the average edge length is the sum of the lengths of the four edges of the actual element divided by 4), calculate the volume of the ideal tetrahedron shape with the same average edge length as the actual element; obtain the distortion rate through the formula (1 - actual volume / ideal volume), and require this value to be less than 0.3 to ensure that the element is not overstretched or compressed; when calculating the aspect ratio, for each mesh element, measure the length of the four edges of the element one by one, extract the length of the longest edge and the length of the shortest edge, and calculate the ratio between the two, requiring this ratio to be less than 5 to ensure that the difference in the length of each edge of the element is moderate and to avoid slender and deformed elements; warpage calculation. For the four triangular surfaces of the mesh element, an ideal plane (i.e., a unique plane determined by three points) is first constructed using the coordinates of the three vertices of each triangular surface. Then, the perpendicular distance from each vertex on the actual triangular surface to the ideal plane is calculated using the distance measurement function of the 3D modeling software. Combining the side lengths between the three vertices of the triangular surface, the deviation angle is calculated using trigonometric functions (with the perpendicular distance as the opposite side and the side length between the vertices as the hypotenuse, arcsin(perpendicular distance / side length) is calculated to obtain the corresponding deviation angle). After calculating the deviation angles of the four triangular surfaces respectively, the maximum value is taken as the warp of the element. This angle is required to be less than 15° to ensure that the element has no obvious twisting deformation.

[0056] The mesh quality standards are defined using the aforementioned calculation methods to comprehensively check for malformed meshes in the model. For any non-compliant mesh elements found during the inspection (such as elements with excessive distortion rates), adjustments are made one by one as follows: If the aspect ratio exceeds the standard, use the element splitting tool in the 3D modeling software to split the slender elements along the line connecting the midpoint of the longest edge to the opposite vertex, splitting them into 2 to 3 short-edge elements at once. After splitting, the difference in the length of the four edges of each sub-element should be controlled within 1.5 times; or, by adjusting the distribution of seed points in the mesh division, the seed point density is increased in the area where the slender elements are located (adding 2 to 3 seed points per millimeter), and the edge lengths are made more uniform when regenerating the mesh, avoiding elements that are excessively extended in one direction; if the warpage exceeds the standard, optimize the structural contour fitting parameters, and adjust the metal membrane... The fitting accuracy parameters for the spherical arc surface of the sealing surface were adjusted from the default value to a higher level (specifically, the fitting tolerance was tightened from 0.005 mm to 0.001 mm, while the number of interpolation points on the surface was increased by 50%), so that the mesh elements better fit the curvature changes of the arc surface during generation. At the same time, the coordinates of the adjacent connecting nodes of the out-of-standard elements were fine-tuned through the node editing function of the software, with each fine-tuning amount not exceeding 0.001 mm, gradually correcting the distortion state of the triangular surface and reducing the deviation angle between the actual surface and the ideal plane. Through the above targeted adjustments, each unqualified mesh element was re-inspected and corrected until the distortion rate of all mesh elements was less than 0.3, the aspect ratio was less than 5, and the warpage was less than 15°, all meeting the quality requirements. Finally, a finite element calculation model with gradient mesh density and qualified mesh quality was generated.

[0057] This embodiment determines the projection boundary line and defines a closed analysis subdomain by perpendicularly projecting the spatial intersection line. This allows for precise focusing on the core area of ​​the metal diaphragm seal, concentrating finite element analysis on key contact areas and avoiding the dispersion of computational resources in irrelevant regions, thus improving the relevance and effectiveness of the analysis. The use of gradient mesh density partitioning sets reasonable mesh sizes for the core sealing area, boundary transition area, and non-critical areas, ensuring both computational accuracy in the core sealing area and overall model efficiency. Furthermore, the use of tetrahedral elements to adapt to complex structures guarantees mesh quality and the reliability of simulation results, facilitating precise optimization of the flow control valve's sealing structure and geometric design.

[0058] In a preferred embodiment of the present invention, step 4 includes:

[0059] Step 400: Based on the finite element calculation model, apply fully constrained boundary conditions to all nodes on the fixed surface of the valve body that mates with the external mounting structure. Apply concentrated force loads perpendicular to the pressure surface of the metal diaphragm to all nodes to simulate elastic preload. Simultaneously apply uniformly distributed pressure loads perpendicular to the pressure surface to simulate the rated working medium pressure. Specifically, this includes: retrieving the finite element calculation model with gradient mesh density generated in step 301; using the boundary condition setting function of the finite element analysis software to accurately locate the fixed surface on the valve body that mates with the external mounting structure. This surface is the valve body flange end face, with a flatness error not exceeding 0.002 mm, and must completely fit the flat contact surface of the mounting base; apply fully constrained boundary conditions to all nodes on this surface, specifically restricting the translational displacement of the nodes in the X, Y, and Z linear directions, and simultaneously restricting the rotational displacement around the X, Y, and Z coordinate axes, ensuring that the valve body maintains a fixed posture throughout the simulation process and will not experience additional displacement or rotational interference due to load, conforming to the actual fixed posture after installation. Next, the pressure-bearing surface of the metal diaphragm is positioned. This surface is the outer side of a spherical arc surface made of nickel-cobalt alloy (i.e., the side of the diaphragm away from the valve body sealing surface). Using the software's load application function, a concentrated force load perpendicular to the pressure-bearing surface is uniformly applied to all nodes on this surface. This is used to simulate the elastic preload required for the metal diaphragm to fit against the valve body sealing surface. The load application direction is strictly directed towards the valve body sealing surface, and the magnitude of the concentrated force at each node is consistent, set to 50N. This ensures that the diaphragm can form a uniform preload contact with the valve body sealing surface in the initial state, avoiding sealing gaps caused by insufficient local contact pressure. Simultaneously, a uniformly distributed pressure load perpendicular to the pressure-bearing surface of the metal diaphragm is applied to the pressure-bearing surface of the metal diaphragm to simulate the pressure generated by the internal medium under the valve's rated operating conditions. The load direction is consistent with the concentrated force load direction, and the range of the uniformly distributed pressure completely covers the entire pressure-bearing surface of the metal diaphragm. The pressure value is set to 1MPa according to the rated operating pressure standard, ensuring that the simulation conditions are completely consistent with the actual working scenario of the valve and accurately reflect the influence of medium pressure on sealing performance.

[0060] Step 401a: Define the lower surface of the metal diaphragm as the contact surface and the sealing surface of the valve body as the target surface to establish a contact pair. Define contact properties for the contact pair, including contact normal behavior that allows separation and contact tangential behavior based on a given friction coefficient. Specifically, this includes: using the contact setting function of the finite element analysis software to clarify the composition of the contact pair; defining the lower surface of the metal diaphragm (i.e., the side surface that directly contacts the valve body sealing surface) as the contact surface, which is a spherical arc surface made of nickel-cobalt alloy with a surface roughness ≤0.02 micrometers, and needs to precisely match the original geometric contour of the diaphragm; defining the sealing surface of the valve body as the target surface, which is precision ground with a flatness error ≤0.001 millimeters to ensure good contact with the diaphragm contact surface. Adaptability is ensured by matching the machining precision of the actual sealing surface. Subsequently, the contact attributes of the established contact pair are defined in detail. In terms of contact normal behavior, a contact mode that allows separation is set. Specifically, when the normal force between the contact surface and the target surface is less than the preset elastic preload threshold (valued at 30N), separation is allowed, fully conforming to the contact state switching that may occur due to load changes during the actual sealing process. In terms of contact tangential behavior, based on the actual friction characteristics of the metal diaphragm (nickel-cobalt alloy) and the valve body sealing surface (valve body metal material), a given friction coefficient of 0.15 is set to accurately simulate the friction effect between the contact surfaces. This avoids discrepancies between the contact sliding behavior and reality due to an unreasonable friction coefficient setting, ensuring that the contact behavior conforms to the laws of physical characteristics.

[0061] Step 401b: The interaction process of the concentrated force load and the uniformly distributed pressure load is divided into multiple load increment steps. For the first load increment step, the geometric state of the finite element calculation model after mesh generation without any boundary conditions or loads is taken as the initial geometric configuration. Based on the initial geometric configuration, an overall stiffness matrix and an initial residual force vector are formed. The initial residual force vector consists of the load values ​​of the concentrated force load and the uniformly distributed pressure load in the first increment step. Solving the linear equation system with the overall stiffness matrix as the coefficient matrix yields the displacement increments of all nodes in the finite element calculation model, specifically including:

[0062] Based on the total magnitude and loading characteristics of the concentrated force load and the uniformly distributed pressure load, multiple load increment steps are defined. The principle for this division is as follows: in the initial stage, the load gradually increases from 0 to the preload threshold (30N), with a small rate of change in load, resulting in denser increment steps, with each increment not exceeding 5% of the total load. Once the load reaches the preload threshold, the subsequent load tends to increase steadily, and the increment steps can be appropriately relaxed, with each increment controlled between 10% and 15% of the total load. This ensures the simulation accuracy of the loading process, effectively avoids calculation convergence difficulties caused by sudden load changes, and balances computational efficiency with result accuracy. For the first load increment step, the unmeshed parts after mesh generation... The initial geometric configuration is the geometric state of the finite element model under any boundary conditions and loads. In this configuration, all node coordinates are the original coordinates after mesh generation, and the mesh elements are free of stress and strain, completely in an initial free state. Based on this initial geometric configuration, the overall stiffness matrix is ​​formed using the stiffness matrix calculation function of the finite element analysis software and the element stiffness matrix assembly method. First, based on parameters such as the type of mesh element (tetrahedral element), the material's elastic modulus (210 GPa for nickel-cobalt alloy), and Poisson's ratio (0.3), the local stiffness matrix of each mesh element is calculated using the tetrahedral element stiffness matrix formula. ,in The element geometry matrix is ​​obtained by first determining the coordinates of the four nodes of the tetrahedral element in the global coordinate system (let the node coordinates be (x5, y5, z5), (x6, y6, z6), (x7, y7, z7), (x8, y8, z8)). The element shape function is then derived using the partial derivatives of the element shape function with respect to the coordinates. The element shape function formula is NI = (aI+bIx+cIy+dIz)(I=5,7,8), where V is the volume of the tetrahedral element, and aI, bI, cI, and dI are shape function coefficients, calculated from the coordinates of the four nodes (e.g., a5=x6(y7z8-y8z7)-y6(x7z8-x8z7)+z6(x7y8-x8y7), the remaining coefficients are derived according to the same coordinate combination rules); element geometric matrix It is a 6x12 matrix, with each row corresponding to a strain component (3 normal strains and 3 shear strains), and each column corresponding to a displacement component of a node (each node has displacements in the X, Y, and Z directions). The matrix elements are calculated using the partial derivatives of the shape function with respect to the coordinates, and are used to establish the relationship between the nodal displacements and strains of the element. Represents the unit geometric matrix The transpose of the matrix, The elasticity matrix of a material (the formula for the elasticity matrix of an isotropic material is as follows) ), The model uses a unit volume micro-element. Based on the global numbering correspondence of each element node in the overall model, the local stiffness matrices of all elements are assembled one by one according to their row and column positions. That is, the elements of each local stiffness matrix are superimposed onto the corresponding positions of the overall stiffness matrix to form the overall stiffness matrix of the entire model. This matrix corresponds one-to-one with the model nodes through row and column indices, and the matrix element values ​​reflect the mechanical correlation strength between corresponding nodes, which can comprehensively reflect the mechanical properties of the entire model. At the same time, an initial residual force vector is constructed. This vector is composed of the load values ​​of concentrated force load and uniformly distributed pressure load in the first incremental step. The calculation method is as follows: first, determine the proportion of the first incremental step to the total load. For example, if the first incremental step accounts for 5% of the total load, then calculate the concentrated force load component (total concentrated force 50N × 5% = 2.5N) and the uniformly distributed pressure load component (total uniformly distributed pressure 1MPa × 5% = 0.05MPa) in this step respectively. Based on the position of each node on the load-bearing surface, the corresponding load components are allocated according to the proportion of the node area (node ​​load component = area of ​​the micro-element on the load-bearing surface where the node is located / total load-bearing surface area × load component). The load components of all nodes are arranged sequentially according to the node number to form an initial residual force vector. The magnitude of each element in the vector corresponds to the magnitude of the force on the corresponding node, and the direction is indicated by the positive and negative signs to ensure that the vector can accurately reflect the load distribution state in the first increment step. Finally, a linear equation system is established with the overall stiffness matrix as the coefficient matrix. The expression of the equation system is: overall stiffness matrix × node displacement increment = residual force vector. The linear equation system is solved by the solver of the finite element analysis software using a direct solution method (such as LU decomposition method) or an iterative solution method (such as conjugate gradient method) to accurately calculate the displacement increments of all nodes in the finite element calculation model in the X, Y, and Z directions in the first load increment step.

[0063] Step 401c: Update the spatial coordinates of all nodes in the finite element calculation model according to the displacement increment to obtain the updated finite element calculation model configuration. Based on the updated finite element calculation model configuration, recalculate the stress inside each mesh element in the finite element calculation model, and update the contact state of the contact pairs. Specifically, based on the node displacement increment obtained in step 401b, update the spatial coordinates of all nodes in the finite element calculation model one by one. The new X coordinate of each node = original X coordinate + X-direction displacement increment, the new Y coordinate = original Y coordinate + Y-direction displacement increment, and the new Z coordinate = original Z coordinate + Z-direction displacement increment. Calculate and update independently in the X, Y, and Z directions to obtain the updated finite element calculation model configuration. This configuration truly reflects the geometric deformation state of the model after the first load increment step, conforming to the stress-deformation law of the actual structure, and providing a geometric basis that conforms to the actual deformation for subsequent stress calculation and contact state update. Based on this updated model configuration, the stress inside each mesh element needs to be recalculated using the elasticity stress calculation method. The specific process is as follows: First, clarify the geometric equation (matrix form) as... ,in The strain vector represents the deformation state of the element. The element nodal displacement vector contains the displacement components of the four nodes of the tetrahedral element in the X, Y, and Z directions, arranged sequentially according to the node order; the specific expression for the strain vector is... , Represents the transpose operation of a vector, normal strain The calculation logic is based on the rate of change of the nodal displacement increment along the X, Y, and Z directions respectively; shear strain. These represent the shear deformations in the XY, YZ, and ZX planes, respectively, and the calculation logic is the superposition of the displacement rates in two perpendicular directions.

[0064] The corresponding strain calculation results are obtained using constitutive equations (matrix form). Calculate the stress, where Let be the stress vector, used to characterize the stress state of the element; the specific expression for the stress vector is: The calculation methods for each stress component are as follows: Normal stress Along the X, Y, and Z directions, respectively, proportional to the corresponding normal strain, the shear stress is solved by multiplying the corresponding element in the elastic matrix [D] with the normal strain; These correspond to three shear strains, and are proportional to the corresponding shear strain. They are calculated as the product of the shear-related elements in the elasticity matrix and the shear strain, for example... (in The shear modulus is calculated using the following formula: ), final elasticity matrix row elements and strain vector The column elements are multiplied sequentially and then summed to obtain the corresponding shear stress component values. Combined with the updated model configuration and the stress data calculated above, the contact state of the contact pair is comprehensively updated. That is, the contact judgment function of the finite element analysis software is used to calculate the straight-line distance between each node of the metal diaphragm contact surface and the nearest node of the valve body target surface. At the same time, combined with the contact normal behavior set in step 401a, the judgment criteria for the contact state are clarified. If the distance between nodes is less than or equal to 0 and the normal force is greater than the preload threshold (30N), it is judged as a contact pressing state. If the distance between nodes is greater than 0 and the normal force is less than the preload threshold (30N), it is judged as a separation state. If there is a relative sliding tendency between nodes and they are not separated, it is judged as a sliding state. The contact state information of each contact node is recorded in detail.

[0065] Step 401d: Based on the updated finite element model configuration, recalculated mesh element stress, and updated contact state, recalculate the residual force vector and determine whether the norm of the residual force vector is less than the preset convergence tolerance. If the norm of the residual force vector is greater than or equal to the convergence tolerance, recalculate the overall stiffness matrix based on the current updated finite element model configuration, mesh element stress, and contact state. Repeat the process of solving the displacement increment, updating the configuration, recalculating the mesh element stress and contact state, recalculating the residual force vector, and determining the norm, performing iterative calculations until the norm of the residual force vector satisfies the convergence criterion. After completing the solution of the current load increment step, use the finite element model configuration, mesh element stress, and contact state that finally satisfied the convergence criterion when the previous load increment step was completed as the initial conditions for the next load increment step, and repeat the calculation process until all load increment steps are completed. Specifically, this includes:

[0066] First, based on the updated finite element model configuration, recalculated mesh element stresses, and updated contact states, the residual force vector is recalculated precisely. The calculation logic is: residual force vector = external load vector - internal force vector corresponding to element stress. Here, the external load vector is the total load vector within the current load increment step, including concentrated force load components and uniformly distributed pressure load components. The internal force vector is obtained by integrating the element stress and element shape function. After calculating the internal force vector of each element, they are assembled according to node numbers to form the overall internal force vector. The residual force vector is obtained by subtracting the corresponding elements of the external load vector and the internal force vector. The absolute value of each element directly quantifies the degree of force imbalance at the corresponding node (the larger the absolute value, the more unbalanced the force at that node). The positive and negative signs clearly reflect the direction of the unbalanced force (positive sign corresponds to the load direction, negative sign corresponds to the opposite direction), ensuring that the residual force vector accurately reflects the balance state between the load and internal forces. Next, the norm of the residual force vector is calculated to comprehensively quantify the magnitude of the residual force in the entire model. The calculation method is as follows: first, all elements in the residual force vector (i.e., the residual force components at each node) are squared separately, then all the squared results are summed, and finally the square root of the sum is taken to obtain the norm value of the residual force vector. This norm value is the core quantitative indicator of the overall force balance of the model. The smaller the value, the closer the overall force balance of the model is to the equilibrium state. Finally, it is determined whether the norm of the residual force vector is less than the preset convergence tolerance (the convergence tolerance is set to 1× according to the simulation accuracy requirements). N, ensuring the result error is within the allowable range), if the norm of the residual force vector is greater than or equal to the convergence tolerance, it indicates that the current model has not reached force equilibrium and iterative calculation is required; based on the updated model configuration, mesh element stress, and contact state, recalculate the overall stiffness matrix, that is, reassemble the local stiffness matrix of each element, while considering the influence of geometric deformation on element stiffness (correcting the stiffness matrix parameters by updating the element geometric matrix); then repeat the steps of solving for displacement increment in step 401b, updating configuration and calculating stress and contact state in step 401c, and recalculating the residual force vector in this step. The complete process of determining the norm continues until the norm of the residual force vector is less than the convergence tolerance. If the norm of the residual force vector is less than the convergence tolerance, the current load increment step is determined to meet the convergence criterion, and the solution for this step is completed. After completing the solution for the current load increment step, the finite element calculation model configuration that finally met the convergence criterion when the previous load increment step was completed, the stress data of all mesh elements, and the contact state of the contact pairs are directly used as the initial conditions for the next load increment step. The complete calculation process from step 401b to this step is repeated to complete the solution for all load increment steps in sequence, ensuring the continuity and accuracy of the entire loading process.

[0067] Step 401f: After all load increment steps are calculated, the final calculation results are output. These results include the final displacements of all nodes in the finite element model, the final stresses of all mesh elements, and the contact pressure distribution data on the contact surface. Specifically, after all load increment steps are calculated, the finite element analysis software outputs the final calculation results to ensure comprehensive and accurate data. This includes three core data types: final node displacements, the final displacements of all nodes in the finite element model (including independent displacement components in the X, Y, and Z directions), and the total displacement of each node (total displacement = ...). The stress distribution (including the sign and angle of the displacement components) fully reflects the overall and local deformation of the model, such as the deformation characteristics of smaller node displacements in the core sealing area and relatively larger displacements in the edge areas. The final stress of each mesh element, including key stress parameters such as normal stress (tensile stress, compressive stress), shear stress, and equivalent stress, is presented as a stress cloud map, dividing the stress distribution into color gradients based on stress values. Different colors correspond to different stress ranges, visually demonstrating the differences in stress distribution within the model. The core sealing area has higher and more uniform stress values ​​due to preload and medium pressure, while non-core areas have relatively lower stress values. The pressure distribution data is relatively low to avoid stress concentration. The contact pressure distribution data includes the magnitude and direction of the contact pressure at each contact node, as well as the distribution pattern of the contact pressure across the entire sealing surface. This distribution pattern shows that the contact pressure gradually decreases towards the edge area, with the central area of ​​the sealing surface as the core. The pressure distribution is uniformly diffused in a ring shape. The contact pressure is highest in the central area (close to the pressure value corresponding to the preload), and lowest in the edge area, but still meets the sealing requirements. This conforms to the contact mechanical characteristics of the spherical arc surface of the metal diaphragm and the valve body sealing surface, ensuring that the output results can comprehensively and accurately reflect the mechanical response and sealing contact state of the model.

[0068] Step 402: Based on the calculation results, extract the normal distance data between corresponding nodes on the contact surface of all mesh elements within the closed analysis subdomain. Identify the set of mesh elements with a normal distance greater than zero that are interconnected through mesh element edges as microscopic leakage channel simulation data. Specifically, based on the calculation results output in step 401f, use the data extraction tool of the finite element analysis software to extract the normal distance data between corresponding nodes on the contact surface of all mesh elements within the closed analysis subdomain. The specific operation process is as follows: Focus on the closed analysis subdomain (covering the core contact area between the metal diaphragm and the valve body), filter each pair of adjacent mesh elements within the domain one by one, and use the software's element association query... The query function identifies the corresponding nodes of two adjacent units on the contact surface between the metal diaphragm and the valve body. These nodes belong to the two adjacent units and their spatial coordinates fall within the contact surface area. Each unit corresponds to 1 to 3 contact surface nodes, ensuring accurate and complete node positioning. Using the local area where the corresponding nodes are located on the contact surface as a reference, the software's surface analysis function constructs the tangent plane of the contact surface. The direction perpendicular to this tangent plane and pointing outwards from the sealing gap is defined as the normal direction of the contact surface. This ensures that the distance calculation direction is strictly perpendicular to the sealing surface, avoiding inaccurate gap measurements due to directional deviations. The software's distance measurement function then calculates the distance between each pair of corresponding nodes along the aforementioned normal direction. The straight-line distance is calculated, with values ​​rounded to four decimal places to ensure data accuracy meets the requirements for microscopic gap analysis and accurately reflects the size of the sealing gap between nodes. Subsequently, all extracted normal distance data undergoes a double-filtering process. The filtering logic is as follows: First, a distance filtering threshold of 0 is set to filter out node pairs with normal distances greater than 0. The corresponding mesh elements of these node pairs are not fully aligned on the contact surface, exhibiting quantifiable microscopic gaps. These gaps provide a potential channel for media leakage and possess the basic conditions for forming a leakage path. Second, the software's element connection relationship query function is used to check whether the mesh elements with gaps after the first filtering stage are interconnected through mesh element edges, i.e., to inspect... Whether two mesh cells with gaps share a complete edge, and whether this edge belongs to the boundary of the contact surface, ensures that these cells can form a continuous and interconnected channel structure, rather than isolated and scattered gaps. Finally, the set of mesh cells that simultaneously meet the two conditions of a normal distance greater than 0 and interconnection through the mesh cell edges are formally identified as micro-leakage channel simulation data. This data contains the core feature information of the channel, namely, the precise location corresponding to the gap area between the metal diaphragm and the valve body sealing surface, the direction extending along the connection path of the mesh cells, and the width quantified based on the normal distance between the corresponding nodes. It can intuitively and accurately reflect the possible micro-leakage paths between the metal diaphragm and the valve body sealing surface.

[0069] This embodiment simulates the actual working conditions of a metal diaphragm seal by applying boundary conditions and loads. Combined with reasonable contact attribute definitions and load increment step division, it ensures that the finite element calculation accurately reflects the mechanical response and contact state of the sealing structure. An iterative calculation method is used until the residual force vector meets the convergence criterion, avoiding errors caused by non-convergence. The output nodal displacement, element stress, and contact pressure distribution data comprehensively reflect the stress and sealing state of the sealing structure, providing rich data support for sealing performance evaluation. By identifying micro-leakage channels in the simulation data, it is possible to locate potential leakage risk areas between the metal diaphragm and the valve body sealing surface, providing a direct basis for optimizing the valve body-metal diaphragm geometry and improving sealing performance. This effectively improves the dynamic particulate matter problem of the flow control valve and enhances the accuracy of micro-flow regulation.

[0070] In a preferred embodiment of the present invention, step 5 includes:

[0071] Step 500: Simplify each connected path in the microscopic leakage channel simulation data into a parallel plate gap model with an equivalent gap height. Based on the equivalent gap height and path length of each parallel plate gap model, calculate the volumetric flow rate through the parallel plate gap model according to the laminar flow equation of viscous fluid. Summarize the volumetric flow rates corresponding to all parallel plate gap models to obtain the predicted static leakage rate of the contact sealing pair under the current load condition. Specifically, this includes: extracting the microscopic leakage channel simulation data identified in step 402; simplifying the parallel plate gap model for each connected path in the data; the simplification principle is to equate the actual cross-sectional profile of the connected path to a rectangle, using the maximum width of the path cross-section as the equivalent gap width, the average normal distance of all nodes on the path as the equivalent gap height, and the actual length of the connected path as the equivalent gap height. The elongation length is used as the path length to form a single parallel plate slot model with equivalent parameters, ensuring that the model can accurately represent the flow characteristics of the original leakage channel. The volumetric flow rate of the single parallel plate slot model is calculated based on the laminar flow equation of viscous fluid. Assuming the fluid flowing through the slot is an incompressible viscous fluid, and the flow state is laminar, ignoring the influence of inertial forces, the laminar flow rate formula for parallel plate slots is used for calculation: Volumetric flow rate = (cubic of equivalent gap height × pressure difference at both ends of the slot × equivalent gap width) ÷ (fixed coefficient 12 × fluid dynamic viscosity × path length). Here, the pressure difference at both ends of the slot is the difference between the uniformly distributed pressure load on the pressure-bearing surface of the metal diaphragm and the external environmental pressure. The fluid dynamic viscosity is determined according to the physical properties of the actual working medium. If the working medium is a commonly used industrial gas (such as nitrogen), it is taken as 1.78 × 12 under normal temperature and pressure conditions. Pa After substituting the values ​​of each parameter, the volumetric flow rate of a single channel is calculated according to the formula. Finally, the volumetric flow rates corresponding to all parallel plate gap models are summarized, and the volumetric flow rates of all single channels are added together to obtain the predicted static leakage rate of the contact sealing pair under the current load condition. This leakage rate directly reflects the static sealing performance of the sealing structure.

[0072] Step 501: Based on the predicted static leakage rate and the average contact pressure value extracted from the nonlinear contact mechanics solution within the closed analysis subdomain, adjust the key geometric design parameters of the valve body and metal diaphragm sealing pair in the three-dimensional parametric geometric model of the valve body to obtain the adjusted key geometric design parameters. The key geometric design parameters of the valve body and metal diaphragm sealing pair include the initial radius of curvature of the metal diaphragm, the sealing cone angle of the valve body sealing surface, and the pitch difference of the double threads on the valve stem used for adjustment. Specifically, this includes extracting the contact pressure data of all contact nodes within the closed analysis subdomain from the calculation results output in step 401f and calculating the average contact pressure value. The average contact pressure value is calculated as the sum of the contact pressures of all contact nodes within the closed analysis subdomain divided by the total number of contact nodes. This value ensures that it reflects the overall contact tightness of the sealing pair. Next, combining the predicted static leakage rate obtained in step 500 with the aforementioned average contact pressure value, the rationality of the key geometric design parameters of the current valve body and metal diaphragm sealing pair is analyzed. These key geometric design parameters include the initial radius of curvature of the metal diaphragm (initially 5mm), the sealing cone angle of the valve body sealing surface (initially 15°), and the pitch difference of the double threads on the valve stem used for adjustment (initially 0.02mm). Then, the parameters are adjusted based on the analysis results. If the predicted... A high static leakage rate indicates an excessively large sealing gap. The initial radius of curvature of the metal diaphragm can be increased by 0.5 to 2 mm (adjusted range: 5.5 to 7 mm) to improve the fit after pre-tightening; or the sealing cone angle of the valve body sealing surface can be reduced by 3 to 8° (adjusted range: 7 to 12°) to increase the contact area; or the pitch difference of the double threads can be fine-tuned within the range of 0.01 to 0.03 mm (increase or decrease by 0.005 to 0.01 mm from the initial value) to optimize the diaphragm clamping stroke. If the average contact pressure exceeds the reasonable range (reasonable range: 5 to 15 MPa), excessively high pressure (greater than 15 MPa) may cause diaphragm... If the pressure is too low (less than 5 MPa), the sealing effect will be affected, and the above parameters need to be adjusted in reverse: if the pressure is too high, the initial radius of curvature of the metal diaphragm can be reduced by 0.3 to 1 mm (the adjusted value range is 4 to 4.7 mm), or the sealing cone angle can be increased by 2 to 5° (the adjusted value range is 17 to 20°), or the pitch difference can be reduced by 0.003 to 0.008 mm (the adjusted value range is 0.012 to 0.017 mm); if the pressure is too low, the parameters should be fine-tuned in the direction of the higher leakage rate until the adjusted key geometric design parameters are obtained, ensuring that the parameter adjustment specifically solves the balance problem between sealing performance and structural reliability.

[0073] Step 502: Based on the adjusted key geometric design parameters, re-execute the entire process from establishing the three-dimensional parametric geometric model of the valve body to calculating the predicted static leakage rate. Compare and analyze the predicted static leakage rate and average contact pressure values ​​corresponding to multiple different combinations of key geometric design parameters to determine the influence of changes in key geometric design parameters on the predicted static leakage rate and average contact pressure values. Specifically, this includes: First, based on the key geometric design parameters adjusted in step 501, re-execute the complete simulation and calculation process. This process is completely consistent with the operation flow of steps 301 (establishing the three-dimensional parametric geometric model of the valve body), 302 (mesh generation), 303 (mesh quality check and optimization), 400 (boundary conditions and load application), steps 401a to 401f (nonlinear contact mechanics solution), 402 (microscopic leakage channel identification), and 500 (predicted static leakage rate calculation), ensuring that each set of adjusted parameters corresponds to complete and accurate calculation results. Then, according to the above complete simulation calculation process, the system performs simulation calculations for multiple sets of different combinations of key geometric design parameters. (Specific implementation details follow.) As follows, design no fewer than 15 parameter combinations focusing on three key parameters: the initial radius of curvature of the metal diaphragm, the sealing cone angle of the valve body sealing surface, and the pitch difference of the double threads on the valve stem. Each combination should use either a single-factor variable method or a multi-factor combined variable method to set the parameters. Specifically, in the single-factor variable group, only one parameter's value is changed (e.g., fixing the sealing cone angle and pitch difference, and adjusting only the initial radius of curvature sequentially by 5mm, 5.5mm, 6mm, 6.5mm, and 7mm), while the other two parameters remain unchanged. In the multi-factor combined variable group, two or three parameters are adjusted simultaneously. Parameters (such as combining an initial radius of curvature of 5.5 mm with a sealing cone angle of 12°, or an initial radius of curvature of 6 mm with a pitch difference of 0.025 mm) are used to ensure coverage of key nodes within the parameter value range. Simulation calculations for all parameter combinations are performed under completely consistent conditions. The mesh generation standard follows gradient mesh density requirements (0.01 mm for the core sealing region, 0.02 mm for the boundary transition region, and 0.05 mm for the non-core region). Load parameters are uniformly set to a concentrated force load of 50 N and a uniformly distributed pressure load of 1 MPa, with a convergence tolerance of 1 × 10⁻⁶. N, material parameters (elastic modulus 210 GPa, Poisson's ratio 0.3, friction coefficient 0.15) and contact properties are kept constant to ensure the scientific nature of variable control and avoid interference from irrelevant factors in the analysis of parameter influence patterns. Then, the simulation calculation results of all 15 or more parameter combinations are systematically organized, and a unified data recording table is established. The specific values ​​of key parameters for each group are recorded sequentially according to the parameter combination number (such as initial radius of curvature 5.5 mm, sealing cone angle 10°, pitch difference 0.018 mm), and the corresponding predicted static leakage rate values ​​(accurate to 1×). mbar The data includes L / s and average contact pressure (accurate to 0.1 MPa), ensuring complete and accurate data recording without omissions or errors. Groups with significant differences in key parameter values ​​within the parameter combinations are selected for horizontal comparison. For example, comparing two groups: initial radius of curvature 5mm + sealing cone angle 15° + pitch difference 0.02mm and initial radius of curvature 6mm + sealing cone angle 12° + pitch difference 0.02mm, the analysis shows that when the initial radius of curvature and sealing cone angle change simultaneously, the predicted static leakage rate increases from 3.2 × 10⁻⁶ MPa. mbar L / s decreased to 8.5× mbar The reasons for the difference in L / s and average contact pressure from 8.3 MPa to 10.5 MPa were clarified, and the synergistic effect of multiple parameter combinations on the indicators was identified. Longitudinal comparisons were conducted for single-factor variable groups; for example, focusing on the influence of the initial radius of curvature, the changes in indicators were compared under five values: 5 mm, 5.5 mm, 6 mm, 6.5 mm, and 7 mm. As the initial radius of curvature increased from 5 mm to 7 mm, the predicted static leakage rate increased from 4.1 × 10⁻⁶ MPa. mbar L / s gradually decreased to 6.8× mbar The average contact pressure increased from 7.2 MPa to 11.8 MPa, clearly demonstrating the linear or nonlinear correlation between changes in a single parameter and the overall index. Through the comprehensive comparative analysis above, the influence patterns of various key parameters were clarified: an increase in the initial radius of curvature of the metal diaphragm improves the fit after pre-tightening, reduces the sealing gap, and continuously lowers the predicted static leakage rate. Simultaneously, it increases the contact pressure between the diaphragm and the valve body, leading to a rise in the average contact pressure. Conversely, a decrease in the initial radius of curvature produces the opposite effect, increasing the leakage rate and decreasing the contact pressure. A decrease in the sealing cone angle of the valve body sealing surface expands the contact area between the diaphragm and the valve body, reducing pressure concentration per unit area and improving the predicted static leakage rate. As the static leakage rate decreases, the average contact pressure shows a gentle downward trend; as it increases, the contact area shrinks, the leakage rate rises, and the concentration of contact pressure increases. When the difference in the double thread pitch is between 0.01 and 0.03 mm, an increase in the difference improves the adjustment accuracy of the diaphragm clamping stroke, resulting in more precise control of the sealing gap, a decrease in the predicted static leakage rate, and a slight increase in the average contact pressure. When the difference is too small (less than 0.015 mm), the adjustment sensitivity is insufficient, and the leakage rate is prone to increase. Finally, the above patterns are quantified into a quantitative correspondence between changes in key geometric design parameters and the predicted static leakage rate and average contact pressure, forming a clear parameter influence pattern.

[0074] Step 503: Based on the influence law, select the key geometric design parameter combination that makes the predicted static leakage rate lower than the preset leakage rate threshold and the average contact pressure value within the preset pressure range, and complete the optimization verification of the valve body and diaphragm sealing geometry design. Specifically, this includes: First, clarifying the preset judgment criteria, the preset leakage rate threshold is 1× mbar L / s, meeting the ultra-high purity requirements of semiconductors; the preset pressure range is 5 to 15 MPa, balancing sealing reliability and the fatigue life of the metal diaphragm, ensuring that the standard meets the actual application requirements. Then, based on the parameter influence law obtained in step 502, combinations that meet the judgment criteria are selected from all parameter combinations, that is, combinations with a predicted static leakage rate of less than 1× mbar The key geometric design parameter combination is selected based on L / s and the average contact pressure value is in the range of 5 to 15 MPa. Finally, the selected parameter combination is verified again. The complete simulation calculation process from step 301 to step 500 is strictly re-executed to confirm that the predicted static leakage rate and average contact pressure value continuously meet the preset standards without abnormal fluctuations. The optimization verification of the valve body and diaphragm sealing geometry design is completed, and the final optimized design scheme is formed.

[0075] This embodiment equates the microscopic leakage channel to a parallel plate gap model and calculates the leakage rate. It then adjusts key geometric parameters based on contact pressure data, ensuring precise data support for parameter optimization and enhancing the scientific rigor and relevance of the sealing design. Through comparative analysis of multiple parameter combinations, it clarifies the impact of parameter changes on sealing performance and structural stress, avoiding blind adjustments. By selecting parameter combinations that meet preset standards, it achieves dual optimization of the predicted static leakage rate and average contact pressure value. This ensures that the sealing performance meets high cleanliness requirements while also considering structural reliability, effectively improving dynamic particulate matter issues and enhancing the precision of micro-flow rate regulation.

[0076] like Figure 2 As shown, embodiments of the present invention also provide a finite element analysis system for the geometric design of flow control valves, including:

[0077] The geometric feature analysis module is used to perform spatial geometric feature analysis on each data point in the discrete contour coordinate dataset, solve the principal curvature value of each data point on the metal diaphragm sealing surface, and determine the point with the largest average curvature value on the metal diaphragm sealing surface as the core morphology reference point.

[0078] The spatial intersection calculation module is used to establish a local coordinate system with the core topography reference point as the origin, select four sealing state points from the discrete contour coordinate dataset, solve the corresponding spatial plane equation, and calculate the spatial intersection between the spatial plane and the sealing surface of the metal diaphragm.

[0079] The finite element model generation module is used to define a closed analysis subdomain on the metal diaphragm sealing surface of the three-dimensional parametric geometric model of the valve body based on spatial intersection lines, and to mesh the three-dimensional parametric geometric model of the valve body containing the closed analysis subdomain to generate a finite element calculation model.

[0080] The nonlinear contact solution module is used to apply boundary conditions and load conditions to the finite element calculation model, and to perform nonlinear contact mechanics solution calculations on the finite element calculation model with applied boundary conditions and load conditions to obtain simulation data of microscopic leakage channels in the closed analysis subdomain.

[0081] The optimization and verification module is used to calculate the predicted static leakage rate of the contact sealing pair based on the simulation data of the micro leakage channel. This allows for the adjustment of the design parameters of the three-dimensional parametric geometric model of the valve body and the repeated execution of the process from establishing the three-dimensional parametric geometric model of the valve body to calculating the predicted static leakage rate. The module analyzes the influence of different geometric design parameters on the predicted static leakage rate and mechanical response, and completes the optimization and verification of the geometric design of the valve body and diaphragm seal.

[0082] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0083] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A finite element analysis method for the geometric design of flow control valves, characterized in that, The method includes: Step 1: Perform spatial geometric feature analysis on each data point in the discrete contour coordinate dataset, solve for the principal curvature value of each data point on the metal diaphragm sealing surface, and determine the point with the largest average curvature value on the metal diaphragm sealing surface as the core morphology reference point. Step 2: Establish a local coordinate system with the core morphology reference point as the origin, select four sealing state points from the discrete contour coordinate dataset, solve the corresponding spatial plane equation, and calculate the spatial intersection line between the spatial plane and the sealing surface of the metal diaphragm. Step 3: Based on the spatial intersection line, define a closed analysis subdomain on the metal diaphragm sealing surface of the three-dimensional parametric geometric model of the valve body, and mesh the three-dimensional parametric geometric model of the valve body containing the closed analysis subdomain to generate a finite element calculation model. Step 4: Apply boundary conditions and load conditions to the finite element calculation model, and perform nonlinear contact mechanics calculation on the finite element calculation model with applied boundary conditions and load conditions to obtain simulation data of microscopic leakage channels in the closed analysis subdomain. Step 5: Based on the simulation data of the microscopic leakage channel, calculate the predicted static leakage rate of the contact sealing pair, adjust the design parameters of the three-dimensional parametric geometric model of the valve body, and repeat the process from establishing the three-dimensional parametric geometric model of the valve body to calculating the predicted static leakage rate. Analyze the influence of different geometric design parameters on the predicted static leakage rate and mechanical response, and complete the optimization verification of the geometric design of the valve body and diaphragm seal.

2. The finite element analysis method for the geometric design of the flow regulating valve according to claim 1, characterized in that, The process of obtaining the discrete contour coordinate dataset includes: Based on the structural design parameters of the flow control valve, a three-dimensional parametric geometric model of the valve body, including the contact sealing pair between the metal diaphragm and the valve body, is established. Based on the three-dimensional parametric geometric model of the valve body, continuous geometric profile curve data are collected on the sealing surface of the metal diaphragm along the normal section path of the sealing profile of the surface by a geometric engine or measuring device. For continuous geometric contour curve data, a fixed step size is used to sample data points, so that the spacing between adjacent sampling points on the curve path is equal, thus transforming the continuous contour curve into a series of discrete spatial coordinate points, generating a discrete contour coordinate dataset of the metal film to be analyzed surface.

3. The finite element analysis method for the geometric design of the flow regulating valve according to claim 2, characterized in that, Step 1 includes: Based on the discrete contour coordinate dataset, for each data point in the dataset, find all data points whose Euclidean distance to that data point is less than or equal to a preset radius threshold, so as to form a set of neighboring data points of that data point. Using the spatial coordinates of the data point and all data points in its neighboring set, a quadratic surface equation is fitted using the least squares method, such that the sum of squared distances from the surface represented by the quadratic surface equation to the data point and all data points in its neighboring set is minimized; based on the quadratic surface equation, the two principal curvature values ​​of the data point on the surface are calculated. By iterating through each data point in the discrete contour coordinate dataset, repeatedly performing the steps of finding the set of nearest points, fitting the quadratic surface equation and calculating the principal curvature values, two principal curvature values ​​are obtained for each data point on the metal diaphragm sealing surface. Based on the two principal curvature values ​​of each data point on the metal diaphragm sealing surface, the average curvature value of each data point is calculated. The average curvature values ​​of all data points on the metal diaphragm sealing surface are compared, and the data point with the largest average curvature value is determined as the core morphological reference point of the metal diaphragm sealing surface.

4. The finite element analysis method for geometric design of flow regulating valves according to claim 3, characterized in that, Step 2 includes: A local rectangular coordinate system is established with the core morphology reference point as the origin and the directions of the maximum and minimum principal curvature at the core morphology reference point as the coordinate axes. In the local rectangular coordinate system, from the discrete contour coordinate dataset, the point with the largest curvature value along the positive direction of the maximum principal curvature is selected as the first sealing state point, the point with the smallest curvature value along the positive direction of the minimum principal curvature is selected as the second sealing state point, the point with the largest absolute value of the rate of change of the maximum principal curvature is selected as the third sealing state point, and the point closest to the third sealing state point and with the opposite sign of the rate of change of the maximum principal curvature is selected as the fourth sealing state point. Based on the spatial coordinates of the first, second, third and fourth sealing state points, the spatial plane equation passing through the four sealing state points is obtained by using the least squares plane fitting algorithm. The spatial plane equation is intersected with the metal diaphragm sealing surface, which is characterized by a discrete contour coordinate dataset using a surface reconstruction method, to obtain the spatial intersection line between the spatial plane and the metal diaphragm sealing surface.

5. The finite element analysis method for geometric design of flow regulating valves according to claim 4, characterized in that, Step 3 includes: The vertical projection of the spatial intersection line onto the metal diaphragm sealing surface of the three-dimensional parametric geometric model of the valve body is calculated to obtain the projection boundary line. Based on the projection boundary line, the closed region on the metal diaphragm sealing surface defined by the projection boundary line is defined as a closed analysis subdomain. Based on the three-dimensional parametric geometric model of the valve body containing a closed analysis subdomain, tetrahedral elements are used for mesh generation. Specifically, the region inside the closed analysis subdomain is meshed with a first preset size, the region of the mesh element at the projection boundary line and the region of the mesh element that shares a common boundary with the projection boundary line are meshed with a second preset size, and the remaining region in the three-dimensional parametric geometric model of the valve body that does not belong to the region inside the closed analysis subdomain or the region of the projection boundary line is meshed with a third preset size, thereby generating a finite element calculation model with gradient mesh density.

6. The finite element analysis method for geometric design of flow control valves according to claim 5, characterized in that, Step 4 includes: Based on the finite element calculation model, fully constrained boundary conditions are applied to all nodes on the fixed surface of the valve body that mates with the external mounting structure. Concentrated force loads perpendicular to the pressure surface are applied to all nodes on the pressure surface of the metal diaphragm to simulate elastic preload. At the same time, a uniformly distributed pressure load perpendicular to the pressure surface is applied to simulate the rated working medium pressure. Based on the finite element calculation model with fully constrained boundary conditions, concentrated force load and uniformly distributed pressure load, a contact pair is defined between the lower surface of the metal diaphragm and the valve body sealing surface, and nonlinear contact mechanics solution calculation is performed to obtain the calculation results; Based on the calculation results, the normal distance data between the corresponding nodes on the contact surface of all grid cells in the closed analysis subdomain is extracted. The set of grid cells with a normal distance greater than zero and interconnected by grid cell edges is identified as the simulation data of micro-leakage channels.

7. The finite element analysis method for geometric design of flow regulating valves according to claim 6, characterized in that, Based on a finite element model with fully constrained boundary conditions, concentrated force loads, and uniformly distributed pressure loads, a contact pair is defined between the lower surface of the metal diaphragm and the valve body sealing surface. Nonlinear contact mechanics calculations are then performed to obtain the following results: The lower surface of the metal diaphragm is defined as the contact surface, and the sealing surface of the valve body is defined as the target surface to establish a contact pair. Contact properties are defined for the contact pair, including contact normal behavior that allows separation and contact tangential behavior based on a given coefficient of friction. The process of the concentrated force load and the uniformly distributed pressure load is divided into multiple load increment steps. For the first load increment step, the geometric state of the finite element calculation model after meshing without any boundary conditions or loads is taken as the initial geometric configuration. Based on the initial geometric configuration, the global stiffness matrix and the initial residual force vector are formed. The initial residual force vector is composed of the load values ​​of the concentrated force load and the uniformly distributed pressure load in the first increment step. The linear equation system with the global stiffness matrix as the coefficient matrix is ​​solved to obtain the displacement increment of all nodes in the finite element calculation model. The spatial coordinates of all nodes in the finite element calculation model are updated according to the displacement increment to obtain the updated finite element calculation model configuration; based on the updated finite element calculation model configuration, the stress inside each mesh element in the finite element calculation model is recalculated, and the contact state of the contact pair is updated at the same time. Based on the updated finite element model configuration, recalculated mesh element stress, and updated contact state, the residual force vector is recalculated, and it is determined whether the norm of the residual force vector is less than the preset convergence tolerance. If the norm of the residual force vector is greater than or equal to the convergence tolerance, the overall stiffness matrix is ​​recalculated based on the current updated finite element model configuration, mesh element stress, and contact state. The process of solving displacement increments, updating configuration, recalculating mesh element stress and contact state, recalculating residual force vector, and determining norm is repeated for iterative calculation until the norm of the residual force vector satisfies the convergence criterion. After the current load increment step is solved, the finite element model configuration, mesh element stress, and contact state that finally satisfied the convergence criterion when the previous load increment step was solved are used as the initial conditions for the next load increment step. The calculation process is repeated until all load increment steps are completed. After all load increment steps are calculated, the final calculation results are output, which include the final displacement of all nodes in the finite element calculation model, the final stress of all mesh elements, and the contact pressure distribution data on the contact surface.

8. The finite element analysis method for geometric design of flow regulating valves according to claim 7, characterized in that, Step 5 includes: Each connected path in the microscopic leakage channel simulation data is simplified into a parallel plate gap model with an equivalent gap height. Based on the equivalent gap height and path length of each parallel plate gap model, the volumetric flow rate through the parallel plate gap model is calculated according to the laminar flow equation of viscous fluid. The volumetric flow rates corresponding to all parallel plate gap models are summarized to obtain the predicted static leakage rate of the contact sealing pair under the current load condition. Based on the predicted static leakage rate and the average contact pressure value in the closed analysis subdomain extracted from the nonlinear contact mechanics solution, the key geometric design parameters of the valve body and metal diaphragm sealing pair in the three-dimensional parameterized geometric model of the valve body are adjusted to obtain the adjusted key geometric design parameters. Based on the adjusted key geometric design parameters, the entire process from establishing the three-dimensional parametric geometric model of the valve body to calculating the predicted static leakage rate was re-executed. The predicted static leakage rate and average contact pressure value corresponding to multiple different combinations of key geometric design parameters were compared and analyzed to determine the influence of changes in key geometric design parameters on the predicted static leakage rate and average contact pressure value. Based on the influence law, a combination of key geometric design parameters was selected to ensure that the predicted static leakage rate is lower than the preset leakage rate threshold and the average contact pressure value is within the preset pressure range, thereby completing the optimization and verification of the valve body and diaphragm sealing geometry design.

9. The finite element analysis method for geometric design of flow regulating valves according to claim 8, characterized in that, The key geometric design parameters of the valve body and metal diaphragm sealing pair include the initial radius of curvature of the metal diaphragm, the sealing cone angle of the valve body sealing surface, and the pitch difference of the double threads on the valve stem used for adjustment.

10. A finite element analysis system for the geometric design of a flow control valve, the system implementing the method as described in any one of claims 1 to 9, characterized in that, include: The geometric feature analysis module is used to perform spatial geometric feature analysis on each data point in the discrete contour coordinate dataset, solve the principal curvature value of each data point on the metal diaphragm sealing surface, and determine the point with the largest average curvature value on the metal diaphragm sealing surface as the core morphology reference point. The spatial intersection calculation module is used to establish a local coordinate system with the core topography reference point as the origin, select four sealing state points from the discrete contour coordinate dataset, solve the corresponding spatial plane equation, and calculate the spatial intersection between the spatial plane and the sealing surface of the metal diaphragm. The finite element model generation module is used to define a closed analysis subdomain on the metal diaphragm sealing surface of the three-dimensional parametric geometric model of the valve body based on spatial intersection lines, and to mesh the three-dimensional parametric geometric model of the valve body containing the closed analysis subdomain to generate a finite element calculation model. The nonlinear contact solution module is used to apply boundary conditions and load conditions to the finite element calculation model, and to perform nonlinear contact mechanics solution calculations on the finite element calculation model with applied boundary conditions and load conditions to obtain simulation data of microscopic leakage channels in the closed analysis subdomain. The optimization and verification module is used to calculate the predicted static leakage rate of the contact sealing pair based on the simulation data of the micro leakage channel. This allows for the adjustment of the design parameters of the three-dimensional parametric geometric model of the valve body and the repeated execution of the process from establishing the three-dimensional parametric geometric model of the valve body to calculating the predicted static leakage rate. The module analyzes the influence of different geometric design parameters on the predicted static leakage rate and mechanical response, and completes the optimization and verification of the geometric design of the valve body and diaphragm seal.

Images