A method and system for measuring pressure distribution during filling valve machining

Abstract

The application belongs to the technical field of pressure measurement, and specifically relates to a pressure distribution measurement method and system for filling valve processing, which comprises the following steps: collecting three-dimensional topographic data of a filling valve conical surface and a flow channel and digitally reconstructing to generate a digital model, constructing micron-level annular pressure guiding groove features on a mandrel, registering the mandrel channel and the filling valve flow channel, building a leak-free pressure transmission path, transmitting analog pressure data and obtaining discrete pressure digital signals, pre-processing the signals to obtain standardized pressure data, constructing a pressure spatial distribution equation, extracting a gradient vector from the equation, generating a pressure depth map, locating abnormal candidate positions and calculating pressure grayscale values, combining pressure curvature and binary grayscale values to identify abnormalities, identifying gradient mutations through line curvature, fusing multiple results, and outputting filling valve processing pressure distribution measurement results. In the application, accurate pressure measurement is realized through three-dimensional reconstruction and fluid simulation, and the detection accuracy and reliability are improved.

Description

Technical Field

[0001] This invention belongs to the field of pressure measurement and analysis technology, specifically a method and system for measuring pressure distribution during the processing of a filling valve. Background Technology

[0002] Existing methods often employ contact-based single-point or limited-point measurements, making it difficult to obtain the full-domain pressure distribution of the conical surface and flow channel, resulting in narrow measurement range and low resolution. Relying on direct measurement using physical sensors is susceptible to installation interference and environmental disturbances, limiting measurement accuracy and real-time performance, and failing to fully reflect the changing patterns of the pressure field. Traditional simulation analyses often simplify geometric models, resulting in insufficient matching with actual machining morphology and pressure-applying structures, and coarse modeling of pressure transmission paths, leading to significant deviations between simulation results and actual operating conditions. Anomaly detection relies on a single pressure index, lacking multi-dimensional feature fusion and multi-level verification mechanisms such as gradient, curvature, and grayscale, resulting in low anomaly localization accuracy, prone to missed detections and misjudgments, and unable to support accurate defect identification and structural optimization. Summary of the Invention

[0003] To overcome the shortcomings of existing technologies, this invention proposes a method for measuring pressure distribution during the processing of filling valves. This invention primarily addresses the problems of low measurement accuracy, difficulty in measuring the entire pressure range, and inaccurate anomaly detection in existing methods.

[0004] The present invention provides a method for measuring the pressure distribution during the processing of a filling valve, comprising: S1: collecting the original three-dimensional topographic data of the cone surface and flow channel after the filling valve is processed and performing digital reconstruction to generate a digital model of the filling valve cone surface-flow channel, and constructing a micron-level annular pressure groove digital feature on the mandrel.

[0005] S2: Based on the digital characteristics of the pressure tapping groove, the mandrel center channel and the internal flow channel of the filling valve are registered to build a leak-free pressure transmission path. The simulated pressure distribution data of the cone surface is transmitted to the virtual static pressure acquisition along the pressure transmission path to obtain the discrete pressure digital signal corresponding to the cone surface position.

[0006] S3: Preprocess the discrete pressure digital signal to obtain standardized pressure data adapted to the processing characteristics of the filling valve, extract the effective pressure measurement point cloud and construct the spatial distribution equation.

[0007] S4: Extract the pressure distribution gradient vector from the spatial distribution equation, use the gradient vector to generate a pressure depth map of the injection valve's machined surface and locate candidate locations of pressure anomalies, and calculate the pressure grayscale value of the standardized pressure data.

[0008] S5: Calculate the pressure curvature value within the detection area according to the pressure depth map calculation rules, perform binarization processing on the pressure grayscale value to obtain the binarized pressure grayscale value, combine the pressure curvature value and the binarized pressure grayscale value to determine the candidate location of pressure anomaly, and obtain the pressure anomaly determination result.

[0009] S6: Extract pressure distribution lines using binarized pressure grayscale values ​​and perform segmented cutting. Calculate the curvature of the cut lines to identify pressure gradient abrupt changes. Integrate the pressure anomaly candidate locations, pressure anomaly identification results, and pressure gradient identification results to output the measurement results of the pressure distribution during valve processing.

[0010] According to the pressure distribution measurement method for filling valve processing provided by the present invention, the specific steps for constructing the digital features of the micron-level annular pressure tapping groove in step S1 are as follows: S11: Collect the original three-dimensional topographic data of the conical surface and flow channel after the filling valve is processed, use a point cloud denoising algorithm to filter out measurement noise, and register and stitch the denoised point cloud data to obtain a three-dimensional point cloud dataset of the conical surface and flow channel.

[0011] S12: Mesh the 3D point cloud dataset of the conical surface and flow channel, and reconstruct a continuous and smooth digital structure through a surface fitting algorithm to generate the digital structure of the filling valve conical surface-flow channel.

[0012] S13: Extract key geometric parameters of the conical surface based on the digital structure of the filling valve conical surface-flow channel, plan the layout of the pressure-applying groove at the corresponding position of the mandrel based on the geometric parameters, and construct the digital features of the micron-level annular pressure-applying groove that are adapted to the conical surface using a micron-level precision modeling algorithm.

[0013] According to the pressure distribution measurement method for the filling valve processing provided by the present invention, the specific steps for constructing the digital features of the micron-level annular pressure tapping groove in step S13 are as follows: Based on the digital structure of the filling valve's cone-channel, extract the cone angle, sealing strip width, and channel inner diameter geometric parameters, and output the geometric parameter dataset.

[0014] Based on the geometric parameter dataset, the corresponding area that matches the conical surface is located on the digital model of the mandrel, and the circumferential position, groove width, and groove depth parameters of the annular pressure groove are planned to form the pressure groove layout planning data.

[0015] Based on the layout planning data of the pressure groove, a digital model of the corresponding area of ​​the mandrel is performed to generate a digital feature of the annular pressure groove that matches the geometric parameters of the conical surface.

[0016] According to the pressure distribution measurement method for the filling valve processing provided by the present invention, the specific steps for obtaining the discrete pressure digital signal corresponding to the cone surface position in step S2 are as follows: S21: Extract the geometric coordinates and dimensional parameters of the digital features of the pressure groove, and use a coordinate alignment algorithm to accurately match the digital coordinates of the mandrel center channel with the coordinates of the internal flow channel of the filling valve.

[0017] S22: A gap compensation algorithm is used to digitally seal the connection between the mandrel center channel and the internal flow channel of the filling valve, thus establishing a leak-free pressure transmission path.

[0018] S23: Based on the geometric boundary determined by the digital structure of the filling valve cone-flow channel and the digital characteristics of the pressure tapping groove, the pressure distribution of the cone surface is calculated iteratively using fluid pressure field simulation. Discrete sampling is performed according to the spatial coordinates of the cone surface to generate simulated pressure distribution data of the cone surface.

[0019] S24: Import the simulated pressure distribution data of the cone surface into the leak-free pressure transmission path, and use the pressure transmission simulation algorithm to accurately transmit the simulated pressure along the path to the virtual static pressure acquisition terminal to obtain discrete pressure digital signals that correspond one-to-one with each position of the cone surface.

[0020] According to the pressure distribution measurement method for the filling valve processing provided by the present invention, the specific steps for generating the cone-shaped simulated pressure distribution data in step S23 are as follows: Extract the geometric parameters of the digital structure of the filling valve cone-flow channel and the digital features of the pressure tapping groove, integrate them to determine the geometric boundary, clarify the scope and constraints of the pressure simulation, and output the geometric boundary parameter set.

[0021] Based on the set of geometric boundary parameters, fluid simulation parameters are set, and fluid dynamics simulation algorithms are used to perform multiple rounds of iterative calculations to gradually optimize the pressure distribution data and obtain the initial results of the pressure distribution over the entire conical surface.

[0022] The initial pressure distribution results are discretely sampled according to the spatial coordinates of the cone surface, and uniform sampling nodes are divided to ensure that each node corresponds to a specific position on the cone surface, thereby generating basic data.

[0023] Outlier removal is performed on the basic data, the accuracy of the correspondence between coordinates and pressure values ​​is calibrated, and cone-shaped simulated pressure distribution data with one-to-one correspondence with the pressure is generated.

[0024] According to the pressure distribution measurement method for the filling valve processing provided by the present invention, the specific steps for constructing the spatial distribution equation in step S3 are as follows: S31: The filtering algorithm is used to filter out simulation noise and transmission interference in the discrete pressure digital signal, calibrate the pressure value to a preset range that matches the processing characteristics of the filling valve, and output the pre-processed pressure data.

[0025] S32: Perform validity screening on the preprocessed pressure data, remove abnormal data points that exceed the reasonable pressure range, and form a cloud of valid pressure measurement points.

[0026] S33: Based on the coordinates of the effective pressure measurement point cloud and the pressure data, perform data fitting to establish the functional relationship between the pressure value and the spatial coordinates of the cone surface, and construct the spatial distribution equation.

[0027] According to the pressure distribution measurement method for filling valve processing provided by the present invention, the specific steps for obtaining the pressure grayscale value of standardized pressure data in step S4 are as follows: S41: Take the first-order partial derivatives of the spatial distribution equation of pressure on the conical surface along the three-dimensional coordinates of the conical surface to obtain the pressure change rate in each coordinate direction, and combine them to form the global pressure distribution gradient vector.

[0028] S42: Based on the magnitude of the gradient vector of the global pressure distribution and the spatial coordinates, the cone surface is divided into grids and assigned corresponding gradient magnitudes to generate pressure depth map data that characterizes the strength of pressure changes.

[0029] S43: Based on the pressure depth map data, mark the locations where the gradient magnitude exceeds the set threshold as candidate locations for pressure anomalies.

[0030] S44: Normalize and map the candidate locations of pressure anomalies and the corresponding pressure data across the entire region according to the maximum and minimum values ​​to obtain standardized pressure data with uniform intervals.

[0031] S45: Linearly transform the standardized pressure data according to the grayscale mapping rule to obtain the pressure grayscale value that corresponds one-to-one with the pressure magnitude.

[0032] According to the pressure distribution measurement method for the filling valve processing provided by the present invention, the specific steps for obtaining the pressure anomaly judgment result in step S5 are as follows: S51: Based on the pressure depth map data, the second derivative of the pressure depth value within the regular detection area is calculated to obtain the pressure curvature value corresponding to each grid position, thus quantifying the degree of curvature of the pressure distribution.

[0033] S52: Set a preset grayscale threshold based on the statistical distribution of the grayscale value of the pressure across the entire conical surface and the allowable fluctuation range of the processing.

[0034] S53: Binarize the pressure grayscale value according to the preset grayscale threshold, divide the grayscale value into black and white categories, and output the binarized pressure grayscale value.

[0035] S54: Combines pressure curvature value and binary pressure gray value, sets dual discrimination criteria, verifies each candidate pressure anomaly location, and outputs pressure anomaly discrimination result.

[0036] According to the pressure distribution measurement method for the filling valve processing provided by the present invention, the specific steps in step S6 for outputting the measurement results of the pressure distribution during filling valve processing are as follows: S61: Perform connected component analysis on the binarized pressure grayscale values ​​to extract continuous line data representing the pressure distribution pattern. Discretize the lines according to preset segmentation rules to obtain several ordered line sub-segments.

[0037] S62: Solve for the second derivative of the spatial coordinate sequence of each line segment to calculate the line curvature value of each segment.

[0038] S63: Set a gradient abrupt change discrimination threshold based on the distribution range of line curvature values, compare the line curvature values ​​of each line segment with the gradient abrupt change discrimination threshold one by one, mark the positions of lines whose curvature exceeds the threshold, and form a pressure gradient discrimination result.

[0039] S64: Using the set of spatial coordinates of candidate pressure anomaly locations as a reference, align the anomaly level information in the pressure anomaly discrimination results with the abrupt change location information in the pressure gradient discrimination results in terms of spatial coordinates.

[0040] S65: Based on the spatial alignment and attribute association results, the three types of data after association are fused and matched, and the measurement results of the processing pressure distribution of the filling valve are output.

[0041] The present invention also provides a pressure distribution measurement system for the processing of a filling valve, comprising: The 3D modeling module is used to collect the original 3D morphological data of the conical surface and flow channel after the filling valve is processed and to perform digital reconstruction, generating a digital model of the filling valve conical surface-flow channel, and constructing a micron-level annular pressure groove digital feature on the mandrel.

[0042] The pressure building path module is used to register the mandrel center channel and the internal flow channel of the filling valve according to the digital characteristics of the pressure tapping groove, build a leak-free pressure transmission path, and transmit the simulated pressure distribution data of the cone surface to the virtual static pressure acquisition along the pressure transmission path to obtain the discrete pressure digital signal corresponding to the position of the cone surface.

[0043] The data fitting module is used to preprocess discrete pressure digital signals to obtain standardized pressure data adapted to the processing characteristics of the filling valve, extract the effective pressure measurement point cloud, and construct the spatial distribution equation.

[0044] The gradient analysis module is used to extract the pressure distribution gradient vector from the spatial distribution equation, use the gradient vector to generate a pressure depth map of the machined surface of the filling valve and locate candidate locations of pressure anomalies, and calculate the pressure grayscale value of the standardized pressure data.

[0045] The anomaly detection module is used to calculate the pressure curvature value within the detection area according to the pressure depth map, perform binarization processing on the pressure grayscale value to obtain the binarized pressure grayscale value, and combine the pressure curvature value and the binarized pressure grayscale value to determine the candidate location of pressure anomaly and obtain the pressure anomaly detection result.

[0046] The results output module is used to extract pressure distribution lines using binarized pressure grayscale values ​​and perform segmented cutting, calculate the line curvature of the cut lines to identify pressure gradient abrupt changes, integrate the pressure anomaly candidate locations, pressure anomaly identification results and pressure gradient identification results, and output the measurement results of the pressure distribution of the filling valve.

[0047] This invention provides a method for measuring pressure distribution during the processing of a filling valve. Through three-dimensional reconstruction and fluid simulation, it achieves accurate pressure measurement, improving detection accuracy and reliability. The beneficial effects of this invention are as follows: 1. This invention achieves ultra-high precision matching of geometry and pressure transmission based on three-dimensional digital reconstruction and micron-level pressure-injecting groove modeling. The invention employs non-contact three-dimensional measurement to acquire the original morphology of the conical surface and flow channel. Through point cloud denoising, registration and stitching, meshing, and surface fitting, a digital model highly consistent with the actual workpiece is constructed, eliminating measurement errors and geometric distortions at the source. Simultaneously, based on the key parameters of the conical surface, micron-level annular pressure-injecting groove digital features are planned and constructed on the mandrel, ensuring precise matching of the groove position, width, and depth with the conical surface sealing area. This solves the problems of low matching degree between the pressure-injecting structure and the measured surface and large pressure transmission deviations in traditional methods. A leak-free pressure transmission path is constructed through coordinate alignment and gap compensation algorithms, avoiding virtual leakage and signal interference, ensuring a continuous, stable, and traceable pressure transmission process. This provides a realistic and reliable geometric and flow channel foundation for subsequent pressure calculations, significantly improving overall measurement accuracy and reliability.

[0048] 2. This invention combines fluid simulation with numerical iteration to achieve continuous characterization and standardized processing of pressure across the entire conical surface. Based on the steady-state fluid control equation and pressure transmission equation, this invention performs multiple rounds of iterative calculations to obtain a continuous pressure distribution across the entire conical surface. Then, through regular sampling, outlier removal, and data fitting, a pressure spatial distribution equation is constructed, transforming discrete measurement points into a continuous equation. Compared to the limitations of traditional single-point measurements that cannot reflect the entire distribution, this method can obtain pressure values ​​at any location, fully depicting the pressure change trend. Simultaneously, standardized pressure data is generated through filtering, calibration, and validity screening. Then, multi-dimensional features such as gradient, curvature, and grayscale are used for visualization and quantitative analysis, transforming the pressure distribution from abstract numerical values ​​into intuitively discernible depth maps and grayscale images. This facilitates both manual observation and supports automated defect identification, balancing engineering practicality and analytical depth.

[0049] 3. This invention achieves precise location and high reliability of pressure anomalies through multi-feature fusion discrimination and multi-level verification. Instead of relying on a single indicator to determine anomalies, this invention integrates multiple features such as pressure gradient, pressure curvature, binarized grayscale, and line curvature to construct a dual or even multi-level discrimination mechanism, effectively eliminating noise, false mutations, and misjudged points. Through connected component analysis, line segmentation, spatial alignment, and data fusion, the candidate locations of pressure anomalies, anomaly levels, and gradient mutation locations are uniformly correlated, and the final measurement results are output after deduplication and correction. This method significantly reduces the false alarm rate and missed detection rate, and can accurately identify pressure anomalies caused by processing defects, structural mutations, and assembly deviations, providing direct and reliable data support for the quality inspection, defect location, and structural optimization of filling valves. Attached Figure Description

[0050] The invention will now be further described with reference to the accompanying drawings.

[0051] Figure 1 This is a flowchart illustrating the steps of a pressure distribution measurement method for the processing of a filling valve, as provided in an embodiment of the present invention. Figure 2 This is a flowchart of a pressure distribution measurement method for the processing of a filling valve provided in an embodiment of the present invention; Figure 3 This is a block diagram of a pressure distribution measurement system for filling valve processing provided in an embodiment of the present invention. Detailed Implementation

[0052] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below according to specific embodiments.

[0053] like Figures 1 to 3 As shown in the figure, an embodiment of the present invention provides a method for measuring the pressure distribution during the processing of a filling valve. The method includes: S1: Collect the original three-dimensional morphological data of the conical surface and flow channel of the filling valve after processing and perform digital reconstruction to generate a digital model of the filling valve conical surface-flow channel, and construct the micron-level annular pressure groove digital feature on the mandrel.

[0054] S11: Collect the original three-dimensional topographic data of the conical surface and flow channel after the filling valve is processed, use a point cloud denoising algorithm to filter out measurement noise, and register and stitch the denoised point cloud data to obtain a three-dimensional point cloud dataset of the conical surface and flow channel.

[0055] Using a non-contact 3D measurement device, the original 3D topographic data of the conical surface and flow channel of the injection valve after processing were collected. This data consists of a large number of discrete spatial coordinate points, i.e., point cloud data, containing the true geometric coordinate information of the workpiece surface, but also mixed with noise interference such as abnormal points and outliers caused by equipment vibration, environmental reflection, and burrs on the workpiece surface. To ensure the accuracy of subsequent processing, a point cloud denoising algorithm was used to process the original point cloud data. By setting a reasonable filtering threshold, abnormal points and outliers that deviate from the normal coordinate range were removed, and valid points that can truly reflect the shape of the workpiece surface were retained. Since a single measurement cannot cover the complete area of ​​the conical surface and flow channel, the denoised multi-view point cloud segments need to be registered and stitched together. A coordinate transformation algorithm was used to unify the point clouds from different perspectives to the same spatial coordinate system. By matching the features of overlapping areas, the positional deviation caused by the difference in perspective was eliminated, forming a 3D point cloud dataset of the conical surface and flow channel that covers the complete area of ​​the conical surface and flow channel, has unified spatial coordinates, and is free of obvious noise.

[0056] S12: Mesh the 3D point cloud dataset of the conical surface and flow channel, and reconstruct a continuous and smooth digital structure through a surface fitting algorithm to generate the digital structure of the filling valve conical surface-flow channel.

[0057] Using the 3D point cloud dataset of the conical surface and flow channel as the sole processing object, the data is first meshed. This involves connecting the discretely distributed point cloud data into triangular facets according to spatial topological relationships, constructing a surface mesh framework corresponding to the point cloud. This gives the originally scattered point cloud a continuous spatial topological relationship, solving the problem that discrete points cannot be directly geometrically analyzed. Based on the mesh framework, a surface fitting algorithm is used to smoothly approximate the transition region of the conical surface and flow channel. By approximating the distribution law of the discrete point cloud through a mathematical surface model, the discreteness and local fluctuations of the point cloud itself are eliminated, generating a geometrically continuous, curvature-smooth, and realistically reflective digital structure of the filling valve's conical surface and flow channel. The digital structure of the filling valve's conical surface and flow channel is a computationally calculable and editable virtual workpiece structure expressed by mathematical surfaces.

[0058] S13: Extract key geometric parameters of the conical surface based on the digital structure of the filling valve conical surface-flow channel, plan the layout of the pressure-applying groove at the corresponding position of the mandrel based on the geometric parameters, and construct the digital features of the micron-level annular pressure-applying groove that are adapted to the conical surface using a micron-level precision modeling algorithm.

[0059] Taking the digital structure of the filling valve's conical surface and flow channel as the core analysis object, a geometric feature analysis algorithm is used to extract the key geometric parameters of the conical surface that determine the layout of the pressure tapping groove. These parameters include the conical angle, sealing strip width, effective contact radius, and the position of the flow channel's central axis. Using this standardized set of geometric parameters, the pressure tapping groove layout is planned on the digital model of the mandrel. Based on the position and size of the conical sealing strip, the circumferential position, groove width, groove depth, and center height of the annular pressure tapping groove are determined, ensuring that the pressure tapping groove completely corresponds to the effective contact area of ​​the conical surface, providing a precise channel for subsequent pressure transmission. Based on this layout planning data, a micron-level precision modeling algorithm is used, i.e., a digital modeling method with micron-level dimensional control. Through precise coordinate positioning and dimensional constraints, the groove outline and spatial dimensions are digitally generated, strictly controlling the dimensional error within the micron range. This constructs a micron-level annular pressure tapping groove digital feature that matches the geometric height of the filling valve's conical surface. This feature is the core digital geometric structure on the mandrel used to form the pressure measurement microcavity.

[0060] S2: Based on the digital characteristics of the pressure tapping groove, the mandrel center channel and the internal flow channel of the filling valve are registered to build a leak-free pressure transmission path. The simulated pressure distribution data of the cone surface is transmitted to the virtual static pressure acquisition along the pressure transmission path to obtain the discrete pressure digital signal corresponding to the cone surface position.

[0061] S21: Extract the geometric coordinates and dimensional parameters of the digital features of the pressure groove, and use a coordinate alignment algorithm to accurately match the digital coordinates of the mandrel center channel with the coordinates of the internal flow channel of the filling valve.

[0062] Digital feature extraction is performed on the pressure tapping groove of the filling valve. Geometric parameters such as the contour coordinates, cross-sectional dimensions, groove depth, groove width, axial direction vector, and spatial coordinates of the start and end points of the pressure tapping groove are obtained through 3D model analysis or high-precision measurement, forming a discrete point set of the entire pressure tapping groove. Simultaneously, the central channel of the mandrel is digitally characterized to obtain a set of spatial coordinate points for the central channel axis, inner wall surface, and inlet and outlet end faces. Furthermore, the internal flow channels of the filling valve are subjected to meshed coordinate extraction to obtain a discrete coordinate set of the flow channel skeleton and the wall surface. A coordinate alignment algorithm is adopted, using the assembly reference plane and the central axis as a common reference. Through three-dimensional coordinate transformation, the mandrel central channel and the internal flow channel of the filling valve are accurately matched, unifying the mandrel central channel and the internal flow channel of the filling valve into the same global coordinate system, eliminating positional misalignment caused by assembly deviation, modeling deviation and coordinate system inconsistency.

[0063] S22: A gap compensation algorithm is used to digitally seal the connection between the mandrel center channel and the internal flow channel of the filling valve, thus establishing a leak-free pressure transmission path.

[0064] After coordinate matching is completed, the interface between the mandrel center channel and the internal flow channel of the filling valve will have radial and axial gaps due to machining, assembly, and structural steps. In the digital model, this manifests as geometric discontinuities such as coordinate discontinuities, abrupt changes in cross-section, and local cavities, which directly cause virtual pressure leakage and signal distortion. A gap compensation algorithm is used to perform digital sealing processing at the registration joint. The gap area is identified by the difference in neighboring coordinates. The gap width is calculated using the following formula: in, To determine the coordinates of the valve flow channel wall, For the axis coordinates of the mandrel center channel, when The region was identified as a gap. The gap was filled by interpolation using cubic B-spline surface fitting; the fitting expression is as follows: In the formula, , As basis functions, To control the vertex coordinates, interpolation reconstruction is used to ensure continuous flow channel cross-section, smooth walls, and closed cavities, achieving digital sealing. After gap compensation is completed, the mandrel central channel, the internal flow channel of the filling valve, and the pressure tapping groove form a continuous, closed, and uninterrupted integrated flow channel structure in digital space, thus constructing a leak-free pressure transmission path.

[0065] S23: Based on the geometric boundary determined by the digital structure of the filling valve cone-flow channel and the digital characteristics of the pressure tapping groove, the pressure distribution of the cone surface is calculated iteratively using fluid pressure field simulation. Discrete sampling is performed according to the spatial coordinates of the cone surface to generate simulated pressure distribution data of the cone surface.

[0066] Extract the geometric parameters of the digital structure of the filling valve cone-flow channel and the digital features of the pressure tapping groove, integrate them to determine the geometric boundary, clarify the scope and constraints of the pressure simulation, and output the geometric boundary parameter set.

[0067] The geometric boundaries of the fluid simulation calculation are determined by the digital structure of the conical flow channel of the filling valve and the digital features of the pressure tapping groove, including the cone half-cone angle α and the inlet diameter. The parameters include the outlet diameter *d*, cone height *H*, no-slip condition on the wall, pressure inlet boundary, and outlet reference pressure. Fluid simulation parameters are set based on the geometric boundary parameter set. A fluid dynamics simulation algorithm is used for multiple iterative calculations to gradually optimize the pressure distribution data, obtaining the initial pressure distribution results across the entire cone surface. Within the geometric boundary, the fluid control equations are used for iterative calculation of the pressure field. The continuity and momentum equations for the steady-state incompressible fluid are: in, For fluid velocity vector, Let ρ be the fluid pressure, ρ be the medium density, and μ be the dynamic viscosity. The continuous pressure distribution over the entire conical surface is obtained by discretizing and iteratively converging using the finite volume method. The initial pressure distribution is discretized using conical spatial coordinates, dividing the data into uniform sampling nodes to ensure each node corresponds to a specific location on the conical surface, thus generating basic data. Discrete sampling is performed using structured spatial coordinates, with sampling points following equal angle and height distribution rules. A one-to-one correspondence exists between the sampling point coordinates and the pressure values. The expression for the sampling pressure is: The system removes outliers from the base data, calibrates the accuracy of the correspondence between coordinates and pressure values, and generates cone-shaped simulated pressure distribution data that corresponds one-to-one with the position coordinates.

[0068] S24: Simulate pressure distribution data from the cone surface A leak-free pressure transmission path is imported, and a pressure transmission simulation algorithm is used to accurately transmit simulated pressure along the path to a virtual static pressure acquisition terminal, obtaining discrete digital pressure signals corresponding one-to-one with each position on the conical surface. The simulated pressure distribution data of the conical surface is used as a pressure source and loaded at the starting end of the constructed leak-free pressure transmission path, i.e., the coupling interface between the conical surface and the flow channel, and the pressure transmission simulation algorithm is started. The pressure transmission within the flow channel follows a one-dimensional pressure transmission equation, which is expressed as: Where c is the pressure wave velocity, s is the length coordinate along the flow path, and f is the friction factor. Where ρ is the equivalent diameter of the flow channel and ρ is the fluid density. For fluid velocity vector, The value represents the scalar magnitude of the flow velocity. The algorithm calculates pressure attenuation, phase change, and transmission delay segment by segment along a leak-free pressure transmission path, ensuring stable transmission of the pressure signal without virtual leakage. When the pressure signal reaches the virtual static pressure acquisition terminal at the end of the path, the acquisition terminal digitizes and quantizes the pressure value corresponding to each sampling point, outputting pressure values ​​that strictly correspond one-to-one with the spatial coordinates of each sampling point on the cone surface. The coordinate information and pressure values ​​are combined to form a standardized data stream, resulting in a discrete digital pressure signal. This discrete digital pressure signal fully preserves the spatial distribution characteristics and numerical magnitude of the pressure across the entire cone surface, achieving a complete conversion from the physical pressure field of the cone surface to a digital pressure signal. This provides reliable input for subsequent pressure distribution identification, anomaly detection, and data inversion.

[0069] S3: Preprocess the discrete pressure digital signal to obtain standardized pressure data adapted to the processing characteristics of the filling valve, extract the effective pressure measurement point cloud and construct the spatial distribution equation.

[0070] S31: The filtering algorithm is used to filter out simulation noise and transmission interference in the discrete pressure digital signal, calibrate the pressure value to a preset range that matches the processing characteristics of the filling valve, and output the pre-processed pressure data.

[0071] After acquiring the discrete pressure digital signals corresponding to each position on the conical surface, the signals contain simulation noise caused by numerical iteration, mesh discretization, and approximation of pressure transmission paths, as well as transmission interference introduced by coordinate matching, sampling interval, and numerical calculation. This transmission interference causes abnormal fluctuations in the pressure values, such as high-frequency jitter, local abrupt changes, and slight shifts, directly affecting the subsequent fitting accuracy and distribution pattern identification. An adaptive filtering algorithm is used to smooth and denoise the discrete pressure digital signals. Taking sliding window mean filtering as an example, the average pressure within the window replaces the instantaneous value at a single point, suppressing high-frequency noise while preserving the overall pressure change trend. After filtering and denoising, a reasonable pressure range is set based on the machining characteristics of the filling valve conical surface, such as machining accuracy, assembly tolerance, and flow channel structure dimensions. The filtered pressure is then calibrated within this range, and the pressure values ​​are constrained to a preset range matching the actual machining conditions through coefficient adjustment. This eliminates the systematic deviation between numerical simulation and physical conditions, resulting in smooth, stable, and range-compliant preprocessed pressure data.

[0072] S32: Perform validity screening on the preprocessed pressure data, remove abnormal data points that exceed the reasonable pressure range, and form a cloud of valid pressure measurement points.

[0073] The preprocessed pressure data contains anomalous data points caused by abrupt changes in boundary conditions, local mesh distortion, and numerical overflow. These anomalous data points exhibit significant deviations from adjacent measurement points, exceed physically achievable ranges, and contradict the overall distribution trend of the conical surface. Directly incorporating these anomalous data into the fitting process would severely distort the spatial distribution pattern. Therefore, a validity screening process is performed on the preprocessed pressure data, setting upper and lower pressure thresholds based on the conical surface structure, fluid medium, and working pressure level. , Based on the neighborhood consistency judgment, the determination rule is as follows: in, Let be the average pressure in the neighborhood of the i-th point. To set an allowable deviation threshold, points are considered valid only if both conditions are met; otherwise, they are considered outliers and removed. After removing all outliers, the remaining valid points retain their original spatial coordinates. With corresponding calibration pressure In three-dimensional space, a dense, continuous set of discrete points with consistent distribution patterns is formed. This set is the effective pressure measurement point cloud. All data in the measurement point cloud meet the requirements of numerical rationality, spatial consistency, and physical feasibility.

[0074] S33: Based on the coordinates of the effective pressure measurement point cloud and the pressure data, perform data fitting to establish the functional relationship between the pressure value and the spatial coordinates of the cone surface, and construct the spatial distribution equation.

[0075] The three-dimensional spatial coordinates of each measuring point in the effective pressure measuring point cloud As the independent variable and the corresponding pressure value p as the dependent variable, a surface fitting algorithm is used to globally fit the measured point cloud, with the objective function being to minimize the fitting residual. Where n is the number of valid measurement points, Let be the pressure spatial distribution function to be constructed. Taking into account the rotational symmetry of the valve cone surface, the effective pressure measurement point cloud is solved using the least squares method to obtain a set of optimal coefficients, making the fitted surface fit all effective measurement points to the greatest extent possible. An explicit functional relationship between pressure values ​​and the spatial coordinates of the cone surface is established, transforming the discrete, disordered measurement point cloud into a continuous, computable, and extrapolable mathematical expression. The expression is... This is the spatial distribution equation of pressure on a conical surface, which can be used for rapid calculation of pressure at any position on a conical surface, solving for pressure gradients, identifying machining defects, structural optimization analysis, and other subsequent applications.

[0076] S4: Extract the pressure distribution gradient vector from the spatial distribution equation, use the gradient vector to generate a pressure depth map of the injection valve's machined surface and locate candidate locations of pressure anomalies, and calculate the pressure grayscale value of the standardized pressure data.

[0077] S41: Take the first-order partial derivatives of the spatial distribution equation of pressure on the conical surface along the three-dimensional coordinates of the conical surface to obtain the pressure change rate in each coordinate direction, and combine them to form the global pressure distribution gradient vector.

[0078] Based on the established cone-shaped pressure spatial distribution equation Based on this, first-order partial derivatives are calculated along the three-dimensional spatial coordinates x, y, and z of the cone surface to obtain the pressure change rate in each coordinate direction. , , By combining the pressure change rates in these three directions as vector components, a global pressure distribution gradient vector can be constructed. The formula is expressed as: The gradient vector accurately reflects the spatial variation trend of pressure at all locations across the entire cone surface. The magnitude of the vector represents the intensity of pressure change, and the direction of the vector points to the direction of the fastest pressure increase, providing core data support for the subsequent generation of pressure depth maps.

[0079] S42: Based on the magnitude of the global pressure distribution gradient vector and spatial coordinates, the conical surface is meshed and assigned corresponding gradient magnitudes to generate pressure depth map data characterizing the strength of pressure changes. (Using the global pressure distribution gradient vector...) Using this as the core input, we first extract the magnitude of the gradient vector corresponding to each spatial coordinate point. Preserve the original spatial coordinates corresponding to each gradient vector. Based on the magnitude and spatial coordinates, the conical surface is divided into a regularized mesh, with each mesh cell corresponding to a unique set of spatial coordinates. A matching gradient magnitude is assigned to each mesh cell; this magnitude is the magnitude of the gradient vector within that cell. By associating the mesh cell coordinates with the gradient magnitudes, pressure depth map data is generated to characterize the intensity of pressure changes. This data comprehensively records the distribution characteristics and severity of pressure changes across the entire conical surface.

[0080] S43: Based on the pressure depth map data, locations where the gradient amplitude exceeds a set threshold are marked as candidate locations for pressure anomalies. A threshold for the severity of pressure changes is preset, using the pressure depth map data as the direct basis. This threshold is set comprehensively based on the actual machining accuracy of the valve cone surface, fluid working characteristics, and simulation calculation error range. All grid cells in the pressure depth map are traversed, and the gradient magnitude of each cell is compared with the set threshold. When the gradient magnitude of a certain grid cell is greater than When the location of the unit is determined to be a candidate location for pressure anomaly, the spatial coordinate range corresponding to the unit is extracted and summarized to form a set of spatial coordinates of the candidate locations for pressure anomaly.

[0081] S44: For candidate pressure anomalies, retrieve the original pressure values ​​corresponding to the entire conical surface. Normalize all pressure data according to the maximum and minimum values, and map the pressure values ​​uniformly to a fixed standard range to eliminate differences in dimensions and numerical ranges, thereby obtaining standardized pressure data that are numerically accurate and comparable.

[0082] S45: The standardized pressure data is linearly converted according to the preset gray level mapping rules, and the standardized pressure values ​​are mapped one by one to the gray level range to generate pressure gray values ​​that strictly match the pressure magnitude at each position of the cone surface and can be directly used for image processing and defect identification.

[0083] S5: Calculate the pressure curvature value within the detection area according to the pressure depth map calculation rules, perform binarization processing on the pressure grayscale value to obtain the binarized pressure grayscale value, combine the pressure curvature value and the binarized pressure grayscale value to determine the candidate location of pressure anomaly, and obtain the pressure anomaly determination result.

[0084] S51: Based on the pressure depth map data, the second derivative of the pressure depth values ​​within the regular detection area is calculated to obtain the pressure curvature value corresponding to each grid position, quantifying the degree of curvature of the pressure distribution. The pressure depth map data is essentially the distribution of pressure gradient amplitudes at each grid node of the cone surface, denoted as... (x, y, z are the 3D coordinates of the mesh nodes). The regular detection area is a pre-defined meshed range covering the key stress area of ​​the cone surface. Invalid meshes outside the detection area must be removed first, and only the depth values ​​of the valid detection meshes are retained for calculation. To quantify the curvature of the pressure distribution, the pressure depth value... Taking the second-order partial derivatives along the three-dimensional coordinates x, y, and z respectively, we obtain the second-order partial derivative matrix. The principal curvatures of each mesh node are then solved using this matrix. , Calculate Gaussian curvature As a pressure curvature value, the obtained pressure curvature value K can quantify the degree of curvature of the pressure distribution. The larger the absolute value of the curvature value, the steeper the pressure distribution and the more obvious the abrupt change at that location.

[0085] S52: Set a preset grayscale threshold based on the statistical distribution of the grayscale values ​​of the pressure across the entire conical surface and the allowable fluctuation range during processing. The grayscale values ​​of the pressure across the entire conical surface are denoted as... Statistical analysis was performed on the grayscale values ​​across the entire area, and the key statistical parameters, namely the mean grayscale value, were calculated using the histogram statistical method. Gray standard deviation Maximum grayscale value Minimum grayscale value The statistical formulas are as follows: Where N is the number of effective grid nodes in the entire cone surface. Let be the pressure grayscale value of the i-th node. The allowable pressure fluctuation range for the filling valve cone surface machining is determined by machining accuracy, assembly tolerance, and working pressure rating, and is denoted as . Map the pressure fluctuation range to the corresponding grayscale range. Based on grayscale statistical parameters and the allowable grayscale range, a preset grayscale threshold is determined. The threshold calculation formula is expressed as: Where k is the threshold correction coefficient, ensuring the threshold... Able to accurately distinguish normal pressure areas (grayscale ≤ ) and abnormal pressure areas (grayscale > ), the threshold It is solidified and loaded into the subsequent binarization processing module as a unified criterion for binarization division.

[0086] S53: Binarize the pressure grayscale values ​​according to a preset grayscale threshold, dividing the grayscale values ​​into black and white categories, and outputting the binarized pressure grayscale values. Based on the global pressure grayscale values... With the set preset grayscale threshold A global binarization algorithm is used to determine and assign grayscale values ​​to each grid node point by point. Regions assigned a value of 1 correspond to grayscale values ​​exceeding a threshold and are initially identified as candidate regions for pressure anomalies. Regions assigned a value of 0 correspond to grayscale values ​​within the normal range and are identified as normal regions. To avoid erroneous assignments due to noise interference, simple morphological corrections are performed after binarization to remove isolated single-pixel assignment points, ensuring the continuity of the binarization results and outputting regular and reliable binarized pressure grayscale values, thus enhancing the digital difference between abnormal and normal regions.

[0087] S54: Integrates pressure curvature values ​​and binarized pressure grayscale values, sets dual discrimination criteria, verifies each candidate pressure anomaly location, and outputs the pressure anomaly discrimination result. This is based on the set of candidate pressure anomaly locations and the corresponding pressure curvature values. With binarized pressure gray value To ensure the accuracy of anomaly detection, a dual set of criteria is established. These criteria are: Curvature discrimination criterion: pressure curvature value at candidate location Exceeding the preset curvature threshold , Determined based on the allowable range of pressure fluctuations during processing, and calibrated using experimental data, i.e. .

[0088] Gray-scale discrimination criteria: Binarized pressure gray-scale value of candidate position This means that the location belongs to the anomaly identification area after binarization. Each candidate pressure anomaly location undergoes double verification: if both criteria are met, it is determined to be a true pressure anomaly location. If only one criterion is met, or neither criterion is met, it is determined to be a false positive and eliminated. After verification, the spatial coordinates, corresponding pressure values, and curvature values ​​of all true pressure anomaly locations are compiled to form a structured pressure anomaly discrimination result, clearly defining the anomaly location and degree, providing accurate digital basis for subsequent defect location and structural optimization in the filling valve manufacturing process.

[0089] S6: Extract pressure distribution lines using binarized pressure grayscale values ​​and perform segmented cutting. Calculate the curvature of the cut lines to identify pressure gradient abrupt changes. Integrate the pressure anomaly candidate locations, pressure anomaly identification results, and pressure gradient identification results to output the measurement results of the pressure distribution during valve processing.

[0090] S61: Binarization of pressure grayscale values Connectivity analysis is performed to extract continuous line data representing the pressure distribution pattern. These lines are then discretized according to a predefined segmentation rule, resulting in several ordered line segments. Based on the binarized pressure grayscale values... (Containing only 0 and 1), where regions assigned a value of 1 are candidate regions related to pressure anomalies. An 8-neighborhood connected component analysis algorithm is used to traverse all grid nodes, identify and extract all connected regions with a value of 1, and remove isolated single-pixel noise points to obtain continuous line data. The preset segmentation rules are set based on the trend of line length and curvature changes: the total length of each continuous line is calculated. According to a fixed length threshold Divide the line into uniform segments. If there is a significant abrupt change in curvature at the segmentation point (predicted using a preset temporary curvature threshold), adjust the segmentation point to the abrupt change location to ensure that the curvature change of each line segment is relatively gradual, ultimately resulting in several ordered line segment data. (k=1, 2, ..., m, where m is the number of segments), each segment contains a continuous sequence of spatial coordinates.

[0091] S62: Solve for the second derivative of the spatial coordinate sequence of each line segment to calculate the line curvature value of each segment. Based on the data of each line segment... Extract the spatial coordinate sequence of each segment Let t be the number of nodes in the sub-segment. The line curvature reflects the degree of bending of the line and indirectly characterizes the drastic change in the pressure gradient. The coordinate sequence is smoothed, and the second derivative of the smoothed coordinate sequence is calculated along the line direction. The average curvature and maximum curvature of each line sub-segment are then used as the line curvature value for that sub-segment. This yields the line curvature value corresponding to each line segment, quantifying the degree of curvature of each segment.

[0092] S63: Based on the distribution range of line curvature values, a gradient abrupt change threshold is set. The line curvature value of each line segment is compared one by one with the gradient abrupt change threshold. Lines with curvature exceeding the threshold are marked, forming the pressure gradient discrimination result. This is based on the line curvature values ​​of all line segments. Statistical analysis was performed on all curvature values ​​to calculate the mean and standard deviation of curvature, determine the distribution range of curvature values, and set a gradient change discrimination threshold based on this distribution range and the allowable pressure gradient change range for the filling valve manufacturing process. The line curvature value of each line segment. and Compare them one by one, if If so, mark the spatial location corresponding to the line segment as the location of the pressure gradient abrupt change, and record its coordinate information and curvature value. If If the gradient change is not clear, it is determined to be a gradient-free change. Finally, the coordinates and curvature values ​​of all gradient change locations are sorted out to form a structured pressure gradient discrimination result.

[0093] S64: Using the set of spatial coordinates of candidate pressure anomaly locations as a reference, spatial coordinate alignment is performed between the anomaly level information in the pressure anomaly discrimination results and the abrupt change location information in the pressure gradient discrimination results to complete the attribute association of multi-source information. This is based on the set of spatial coordinates of candidate pressure anomaly locations, the pressure anomaly discrimination results, and the pressure gradient discrimination results. A spatial coordinate nearest neighbor alignment algorithm is used, with the coordinates of the candidate pressure anomaly locations as a reference, to calculate the spatial distance between the pressure gradient abrupt change location and each candidate location. ,when , As a spatial alignment threshold, based on the grid size setting, it is determined that the two correspond to the same abnormal region. The abnormality level information of the region is associated with the gradient mutation information, and assigned the same spatial identifier to complete the spatial alignment and attribute association of all multi-source data.

[0094] S65: Based on the spatial alignment and attribute association results, the three types of associated data are fused and matched, duplicate and conflicting abnormal records are removed, and the pressure distribution measurement results of the filling valve processing are output. Based on the spatial alignment and attribute association results, fusion matching rules are executed on the three types of associated data (coordinates of pressure anomaly candidate locations, anomaly level, and gradient mutation characteristics): under the same spatial identifier, the record with the highest anomaly level is retained, and duplicate anomaly locations are removed. If anomaly level and gradient mutation characteristics conflict, the pressure curvature value and binary grayscale value are combined for re-verification, and valid records are retained after confirming the authenticity of the anomaly. After fusion, the spatial coordinates, anomaly level, gradient mutation location, corresponding curvature value, and grayscale characteristics of all valid anomaly areas are compiled, and the pressure distribution overview of normal areas is supplemented to form a structured pressure distribution measurement result for filling valve processing, providing accurate digital basis for defect location, structural optimization, and quality inspection in filling valve processing.

[0095] like Figure 3 As shown, the present invention also provides a pressure distribution measurement system for the processing of a filling valve, comprising: The 3D modeling module is used to collect the original 3D morphological data of the conical surface and flow channel after the filling valve is processed and to perform digital reconstruction, generating a digital model of the filling valve conical surface-flow channel, and constructing a micron-level annular pressure groove digital feature on the mandrel.

[0096] The pressure building path module is used to register the mandrel center channel and the internal flow channel of the filling valve according to the digital characteristics of the pressure tapping groove, build a leak-free pressure transmission path, and transmit the simulated pressure distribution data of the cone surface to the virtual static pressure acquisition along the pressure transmission path to obtain the discrete pressure digital signal corresponding to the position of the cone surface.

[0097] The data fitting module is used to preprocess discrete pressure digital signals to obtain standardized pressure data adapted to the processing characteristics of the filling valve, extract the effective pressure measurement point cloud, and construct the spatial distribution equation.

[0098] The gradient analysis module is used to extract the pressure distribution gradient vector from the spatial distribution equation, use the gradient vector to generate a pressure depth map of the machined surface of the filling valve and locate candidate locations of pressure anomalies, and calculate the pressure grayscale value of the standardized pressure data.

[0099] The anomaly detection module is used to calculate the pressure curvature value within the detection area according to the pressure depth map, perform binarization processing on the pressure grayscale value to obtain the binarized pressure grayscale value, and combine the pressure curvature value and the binarized pressure grayscale value to determine the candidate location of pressure anomaly and obtain the pressure anomaly detection result.

[0100] The results output module is used to extract pressure distribution lines using binarized pressure grayscale values ​​and perform segmented cutting, calculate the line curvature of the cut lines to identify pressure gradient abrupt changes, integrate the pressure anomaly candidate locations, pressure anomaly identification results and pressure gradient identification results, and output the measurement results of the pressure distribution of the filling valve.

[0101] In summary, this embodiment provides a method and system for measuring pressure distribution during the processing of filling valves. Through multi-feature fusion discrimination and multi-level verification, it achieves accurate location of pressure anomalies and high reliability of results. This invention does not rely on a single indicator to determine anomalies, but rather integrates multiple features such as pressure gradient, pressure curvature, binarized grayscale, and line curvature to construct a dual or even multi-level discrimination mechanism, effectively eliminating noise, false mutations, and misjudged points. Through connected component analysis, line segmentation, spatial alignment, and data fusion, the candidate locations of pressure anomalies, anomaly levels, and gradient mutation locations are uniformly associated, and the final measurement results are output after deduplication and correction. This method significantly reduces the false alarm rate and missed detection rate, and can accurately identify pressure anomalies caused by processing defects, structural mutations, and assembly deviations, providing direct and reliable data support for the quality inspection, defect location, and structural optimization of filling valve processing.

[0102] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for measuring pressure distribution during the processing of a filling valve, characterized in that, include: S1: Collect the original three-dimensional morphological data of the conical surface and flow channel of the filling valve after processing and perform digital reconstruction to generate a digital model of the filling valve conical surface-flow channel, and construct the micron-level annular pressure groove digital feature on the mandrel; S2: Based on the digital characteristics of the pressure groove, register the mandrel center channel and the internal flow channel of the filling valve to build a leak-free pressure transmission path, and transmit the cone-shaped simulated pressure distribution data to the virtual static pressure acquisition along the pressure transmission path to obtain the discrete pressure digital signal corresponding to the cone position; S3: Preprocess the discrete pressure digital signal to obtain standardized pressure data adapted to the processing characteristics of the filling valve, extract the effective pressure measurement point cloud and construct the spatial distribution equation; S4: Extract the pressure distribution gradient vector from the spatial distribution equation, use the gradient vector to generate a pressure depth map of the injection valve processing surface and locate the candidate location of pressure anomaly, and calculate the pressure gray value of the standardized pressure data. S5: Calculate the pressure curvature value within the detection area according to the pressure depth map calculation rules, perform binarization processing on the pressure grayscale value to obtain the binarized pressure grayscale value, and combine the pressure curvature value and the binarized pressure grayscale value to determine the candidate location of pressure anomaly and obtain the pressure anomaly determination result; S6: Extract pressure distribution lines using binarized pressure grayscale values ​​and perform segmented cutting. Calculate the curvature of the cut lines to identify pressure gradient abrupt changes. Integrate the pressure anomaly candidate locations, pressure anomaly identification results, and pressure gradient identification results to output the measurement results of the pressure distribution during valve processing.

2. The method for measuring pressure distribution during the processing of a filling valve according to claim 1, characterized in that: In step S1, the specific steps for constructing the digital features of the micron-scale annular pressure groove are as follows: S11: Collect the original three-dimensional topographic data of the cone surface and flow channel after the filling valve is processed, use the point cloud denoising algorithm to filter out measurement noise, and register and stitch the denoised point cloud data to obtain the three-dimensional point cloud dataset of the cone surface and flow channel. S12: The three-dimensional point cloud dataset of the conical surface and the flow channel is meshed, and a continuous and smooth digital structure is reconstructed by a surface fitting algorithm to generate the digital structure of the filling valve conical surface-flow channel. S13: Extract key geometric parameters of the conical surface based on the digital structure of the filling valve conical surface-flow channel, plan the pressure groove layout at the corresponding position of the mandrel based on the geometric parameters, and construct the micron-level annular pressure groove digital feature that is compatible with the conical surface using a micron-level precision modeling algorithm.

3. The method for measuring pressure distribution during the processing of a filling valve according to claim 2, characterized in that: In step S13, the specific steps for constructing the digital features of the micron-scale annular pressure groove are as follows: Based on the digital structure of the filling valve cone-channel, extract the cone angle, sealing strip width, and channel inner diameter geometric parameters, and output the geometric parameter dataset. Based on the geometric parameter dataset, the corresponding area that matches the conical surface is located on the digital model of the mandrel, and the circumferential position, groove width and groove depth parameters of the annular pressure groove are planned to form the pressure groove layout planning data. Based on the pressure groove layout planning data, the corresponding area of ​​the mandrel is digitally modeled to generate digital features of the annular pressure groove that are adapted to the geometric parameters of the conical surface.

4. The method for measuring pressure distribution during the processing of a filling valve according to claim 1, characterized in that: In step S2, the specific steps for obtaining the discrete pressure digital signal corresponding to the position of the cone are as follows: S21: Extract the geometric coordinates and dimensional parameters of the digital features of the pressure groove, and use a coordinate alignment algorithm to accurately match the digital coordinates of the mandrel center channel with the coordinates of the internal flow channel of the filling valve; S22: A gap compensation algorithm is used to digitally seal the connection between the mandrel center channel and the internal flow channel of the filling valve, thus establishing a leak-free pressure transmission path; S23: Based on the geometric boundary determined by the digital structure of the filling valve cone-flow channel and the digital characteristics of the pressure tapping groove, the pressure distribution of the cone surface is calculated iteratively using fluid pressure field simulation. Discrete sampling is performed according to the spatial coordinates of the cone surface to generate simulated pressure distribution data of the cone surface. S24: Import the simulated pressure distribution data of the conical surface into the leak-free pressure transmission path, and accurately transmit the simulated pressure along the path to the virtual static pressure acquisition terminal through the pressure transmission simulation algorithm to obtain discrete pressure digital signals that correspond one-to-one with each position of the conical surface.

5. The method for measuring pressure distribution during the processing of a filling valve according to claim 4, characterized in that: In step S23, the specific steps for generating the cone-shaped simulated pressure distribution data are as follows: Extract the geometric parameters of the digital structure of the filling valve cone-flow channel and the digital features of the pressure tapping groove, integrate them to determine the geometric boundary, clarify the scope and constraints of the pressure simulation, and output the geometric boundary parameter set; Based on the set of geometric boundary parameters, fluid simulation parameters are set, and a fluid dynamics simulation algorithm is used to perform multiple rounds of iterative calculations to gradually optimize the pressure distribution data and obtain the initial results of the pressure distribution over the entire cone surface. The initial pressure distribution results are discretely sampled according to the spatial coordinates of the cone surface, and uniform sampling nodes are divided to ensure that each node corresponds to a specific position on the cone surface, thereby generating basic data. The basic data is processed to remove outliers, and the accuracy of the correspondence between coordinates and pressure values ​​is calibrated to generate cone-shaped simulated pressure distribution data that corresponds one-to-one with the pressure.

6. The method for measuring pressure distribution during the processing of a filling valve according to claim 1, characterized in that: In step S3, the specific steps for constructing the spatial distribution equation are as follows: S31: The filtering algorithm is used to filter out the simulation noise and transmission interference in the discrete pressure digital signal, calibrate the pressure value to a preset range that is compatible with the processing characteristics of the filling valve, and output the pre-processed pressure data. S32: Perform validity screening on the preprocessed pressure data, remove abnormal data points that exceed the reasonable pressure range, and form an effective pressure measurement point cloud; S33: Based on the coordinates of the effective pressure measurement point cloud and the pressure data, perform data fitting to establish the functional relationship between the pressure value and the spatial coordinates of the cone surface, and construct the spatial distribution equation.

7. The method for measuring pressure distribution during the processing of a filling valve according to claim 1, characterized in that: In step S4, the specific steps for obtaining the pressure grayscale values ​​of the standardized pressure data are as follows: S41: Take the first-order partial derivatives of the spatial distribution equation along the three-dimensional coordinates of the cone to obtain the pressure change rate in each coordinate direction, and combine them to form the global pressure distribution gradient vector. S42: Based on the magnitude of the global pressure distribution gradient vector and the spatial coordinates, the cone surface is divided into grids and assigned corresponding gradient magnitudes to generate pressure depth map data that characterizes the strength of pressure changes. S43: Based on the pressure depth map data, the locations where the gradient magnitude exceeds a set threshold are marked as candidate locations for pressure anomalies; S44: Normalize and map the candidate pressure anomaly locations and the corresponding pressure data across the entire region according to the maximum and minimum values ​​to obtain standardized pressure data with uniform intervals. S45: Linearly transform the standardized pressure data according to the grayscale mapping rule to obtain the pressure grayscale value that corresponds one-to-one with the pressure magnitude.

8. The method for measuring pressure distribution during the processing of a filling valve according to claim 1, characterized in that: In step S5, the specific steps to obtain the pressure anomaly detection result are as follows: S51: Based on the pressure depth map data, the second derivative of the pressure depth value within the regular detection area is calculated to obtain the pressure curvature value corresponding to each grid position, thereby quantifying the degree of curvature of the pressure distribution. S52: Set a preset grayscale threshold based on the statistical distribution of the grayscale values ​​of the pressure across the entire conical surface and the allowable fluctuation range of the processing; S53: Binarize the pressure grayscale value according to the preset grayscale threshold, divide the grayscale value into black and white categories, and output the binarized pressure grayscale value. S54: Combine the pressure curvature value and the binary pressure gray value, set dual discrimination criteria, verify each candidate pressure anomaly position, and output the pressure anomaly discrimination result.

9. The method for measuring pressure distribution during the processing of a filling valve according to claim 1, characterized in that: In step S6, the specific steps for outputting the measurement results of the injection valve processing pressure distribution are as follows: S61: Perform connected component analysis on the binarized pressure grayscale value, extract continuous line data representing the pressure distribution pattern, and discretize the lines according to a preset segmentation rule to obtain several ordered line sub-segments. S62: Solve for the second derivative of the spatial coordinate sequence of each line segment to calculate the line curvature value of each segment; S63: Set a gradient abrupt change discrimination threshold according to the distribution range of the line curvature value, compare the line curvature value of each line sub-segment with the gradient abrupt change discrimination threshold one by one, mark the line position where the curvature exceeds the threshold, and form a pressure gradient discrimination result; S64: Using the set of spatial coordinates of candidate pressure anomaly locations as a reference, align the anomaly level information in the pressure anomaly discrimination result with the abrupt change location information in the pressure gradient discrimination result in terms of spatial coordinates. S65: Based on the spatial alignment and attribute association results, the three types of data after association are fused and matched, and the measurement results of the processing pressure distribution of the filling valve are output.

10. A pressure distribution measurement system for filling valve processing, comprising a pressure distribution measurement method for filling valve processing as described in any one of claims 1 to 9, characterized in that, The measurement system includes: The 3D modeling module is used to collect the original 3D morphological data of the conical surface and flow channel after the filling valve is processed and to perform digital reconstruction, generating a digital model of the filling valve conical surface-flow channel, and constructing a micron-level annular pressure groove digital feature on the mandrel; The pressure building path module is used to register the mandrel center channel and the internal flow channel of the filling valve according to the digital characteristics of the pressure tapping groove, build a leak-free pressure transmission path, transmit the cone surface simulated pressure distribution data along the pressure transmission path to the virtual static pressure acquisition, and obtain the discrete pressure digital signal corresponding to the cone surface position. The data fitting module is used to preprocess the discrete pressure digital signal to obtain standardized pressure data adapted to the processing characteristics of the filling valve, extract the effective pressure measurement point cloud and construct the spatial distribution equation. The gradient analysis module is used to extract the pressure distribution gradient vector from the spatial distribution equation, use the gradient vector to generate a pressure depth map of the filling valve's machined surface and locate candidate locations of pressure anomalies, and calculate the pressure grayscale value of the standardized pressure data. The anomaly detection module is used to calculate the pressure curvature value within the detection area according to the pressure depth map, perform binarization processing on the pressure gray value to obtain the binarized pressure gray value, and combine the pressure curvature value and the binarized pressure gray value to determine the candidate location of pressure anomaly and obtain the pressure anomaly detection result. The results output module is used to extract pressure distribution lines using binarized pressure grayscale values ​​and perform segmented cutting, calculate the line curvature of the cut lines to identify pressure gradient abrupt changes, integrate the pressure anomaly candidate locations, pressure anomaly identification results and pressure gradient identification results, and output the measurement results of the pressure distribution of the filling valve.

Images