Intelligent monitoring system and method applied to composite material forming process

Abstract

The invention discloses an intelligent monitoring system and method applied to a composite material forming process, and relates to the technical field of composite material forming. The system comprises a three-dimensional model construction module, an irregular region screening module, a conventional region temperature acquisition module and an irregular region temperature sensor arrangement module. The three-dimensional model construction module is used for performing three-dimensional scanning to obtain electric heating mold point cloud data to generate a three-dimensional model; the irregular region screening module is used for identifying curvature salient points by combining curvature calculation through a curvature-based edge detection algorithm, and calculating a region area, a convex hull area and a convexity shape factor to judge an irregular region; the conventional area temperature acquisition module is used for arranging sensors in a conventional area according to unit space to acquire temperature to obtain a threshold value; and the irregular region temperature sensor arrangement module is used for adjusting irregular region sensor arrangement and establishing a database. According to the invention, intelligent monitoring of the temperature of the electric heating mold in the composite material forming process can be realized, and the composite material forming quality and the process stability are improved.

Description

Technical Field

[0001] This invention relates to the field of composite material molding technology, specifically to an intelligent monitoring system and method applied to composite material molding processes. Background Technology

[0002] In the composite material molding process, the temperature uniformity of the electrically heated mold directly affects the quality of the composite product. Traditional temperature monitoring methods typically use sensors arranged at fixed intervals, which are difficult to adapt to the complex geometry of the mold surface, especially in irregular areas with significant curvature changes. This can easily lead to blind spots or redundancy in temperature monitoring, resulting in localized overheating or underheating, affecting the degree of curing and mechanical properties of the composite. Existing technologies lack intelligent sensor placement schemes based on mold geometry characteristics, making it impossible to accurately match heat conduction characteristics. Therefore, a monitoring method that can dynamically optimize sensor distribution is needed. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent monitoring system and method for composite molding processes to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an intelligent monitoring system and method applied to composite molding processes, the method comprising: Step S1: Use laser 3D scanning technology to scan the electric heating mold from all angles to obtain the 3D point cloud data of the mold. Import the point cloud data into 3D modeling software for processing and model reconstruction to generate a 3D solid model of the electric heating mold. Step S2: Select a curvature-based edge detection algorithm to calculate the curvature value of each point on the surface of the 3D model, identify the significant curvature points as the edges of irregular regions, extract the edge contours using 3D modeling software, calculate the area, convex hull area, and convexity shape factor geometric parameters of the region enclosed by the edge contours, and determine the irregular regions by calculating the convexity shape factor. Step S3: Based on the geometric features and thermal conductivity of the conventional area, formulate a sensor layout scheme, arrange temperature sensors in a unit space layout manner, collect temperature data, and obtain the temperature threshold of the conventional area; Step S4: By comparing the temperature of the irregular area with the temperature threshold of the regular area, adjust the arrangement of the temperature sensors in the irregular area to obtain the distribution pattern of the temperature sensors in the irregular area and establish a database of temperature sensor distribution patterns in the irregular area.

[0005] Furthermore, step S2 includes: Step S2-1: Extract surface geometric information from the 3D mold CAD model to generate 3D point cloud data. Through sampling processing, form a uniformly distributed set of sampling points. ,in This represents the first discrete sampling point on the corresponding mold surface. This represents the second discrete sampling point on the corresponding mold surface. This represents the third discrete sampling point on the corresponding mold surface, ... This represents the b-th discrete sampling point on the surface of the corresponding mold; then, the point cloud data is denoised and filtered to eliminate discrete errors in the model building process; Step S2-2: For each sampled vertex Select its a nearest neighbors to form a local neighborhood M( Within the local neighborhood, the least squares method is used to evaluate the local neighborhood M( Fit a quadratic surface Calculate the fitted surface midpoint Normal vector at point ; Step S2-3: Calculate the fitted surface At point The coefficients of the first and second fundamental forms are used to solve the characteristic equation and obtain the point. Principal curvature at and And satisfy ≥ Calculate Gaussian curvature and mean curvature principal curvature and These represent the curvature values ​​of the surface in the directions of maximum and minimum curvature, respectively; Step S2-4: For sampling vertices The formula for calculating its curvature gradient is: Gaussian curvature gradient: Mean curvature gradient: Where w(h) is the weight of the local neighborhood vertex h, Let g be the unit vector pointing from the sampled vertex g to the local neighborhood vertex h. Step S2-5: Combine gradient magnitudes: , where α and β are weighting coefficients, and a gradient threshold T is set. The sampling vertex where G>T is the curvature salient point; Step S2-6: Arrange all significant curvature points in clockwise order to obtain an ordered point set. ,in This represents the first point in the point set. This represents the second point in the point set. Let represent the 3rd point in the point set, ..., This represents the nth point in the point set; n represents the number of points in the point set; taking a point of significant curvature as the starting point, the X-axis extends horizontally to the right as the positive X-axis, and the Y-axis extends vertically downwards as the positive Y-axis, represented by the coordinates as follows: =( ); Step S2-7: Calculate the area of ​​the polygon using the shoelace formula: ,in Let x be the x-coordinate of the i-th vertex in the two-dimensional coordinate system. Let be the ordinate of the i-th vertex in the two-dimensional coordinate system. Let x be the x-coordinate of the (i+1)th vertex in the two-dimensional coordinate system. Let be the ordinate of the (i+1)th vertex in the two-dimensional coordinate system. Point and When the points coincide, a closed polygon is formed; Step S2-8: After sorting the point set by coordinates from smallest to largest, construct the lower convex hull by pushing it onto a stack, then traverse and sort it in reverse order to construct the upper convex hull, and finally merge the two to obtain the complete convex hull. Step S2-9: Arrange the vertices of the convex hull in clockwise order to obtain an ordered set of points. ,in This represents the first point in the point set. This represents the second point in the point set. Let represent the 3rd point in the point set, ..., This represents the nth point in the point set; n represents the number of points in the point set; taking a point of significant curvature as the starting point, the X-axis extends horizontally to the right as the positive X-axis, and the Y-axis extends vertically downwards as the positive Y-axis, represented by the coordinates as follows: =( ); Step S2-10: Calculate the area of ​​the convex hull using the shoelace formula. The formula is as follows: , Let x be the x-coordinate of the j-th vertex in the two-dimensional coordinate system. Let be the ordinate of the j-th vertex in the two-dimensional coordinate system. Let x be the x-coordinate of the (j+1)th vertex in the two-dimensional coordinate system. Let be the ordinate of the (j+1)th vertex in the two-dimensional coordinate system. Point and When the points coincide, a closed polygon is formed; Step S2-11: The formula for calculating the convexity shape factor is: , where C is the convexity shape factor; if C=1, it is convex; if C<1, it is concave.

[0006] Furthermore, step S3 includes: Step S3-1: Divide the regular area into equal intervals along the X, Y, and Z axes according to the 3D solid model to form several cubic unit space areas, and arrange Q sensors in each unit space area; Step S3-2: Based on the temperature data collected by the sensor in a unit space area, L times are recorded, and the average value is taken to obtain the temperature threshold of the regular area, which is denoted as . .

[0007] Furthermore, step S4 includes: step S4-1: set up spatial unit areas and arrange sensors in the irregular area according to step S3-1, the number of which is denoted as Q; Step S4-2: Obtain the temperature collected by the temperature sensor in the unit area of ​​space, collect data L times, and take the average value. ; Step S4-3: When the weighted average is not within the threshold range of 0.95 < <1.05 At this time, a temperature sensor is added to the corresponding unit space area, and the number is recorded as Q+1; Step S4-4: Distribute the Q+1 temperature sensors evenly in a unit space area to obtain the distribution pattern. ; Step S4-5: Repeat steps S6-2 to S6-4. When the weighted average value of the temperature sensor first falls within the threshold range, output the distribution pattern. Distribution pattern As a corresponding distribution pattern of a certain three-dimensional feature; Step S4-6: Establish a database based on the distribution pattern of temperature sensors in irregular areas.

[0008] Compared with the prior art, the beneficial effects of the present invention are: 1. By accurately identifying irregular areas of the mold through a curvature-based edge detection algorithm, the arrangement of temperature sensors is adaptively adjusted accordingly. This changes the traditional method of arranging sensors at fixed intervals, reduces temperature monitoring errors in irregular areas, and significantly improves monitoring accuracy.

[0009] 2. It enables more uniform monitoring of the temperature field of the electric heating mold, effectively improves temperature uniformity, thereby increasing the yield of composite molding, significantly reducing curing defects caused by uneven temperature, and improving the quality of composite molding.

[0010] 3. It is compatible with electric heating molds of different geometries. The sensor arrangement scheme can automatically match the geometric characteristics of the mold, eliminating the need for extensive manual debugging, reducing manual debugging costs, and enhancing adaptability to different molds. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the structure of an intelligent monitoring system applied to composite molding process according to the present invention; Figure 2 This is a flowchart illustrating an intelligent monitoring method for composite molding processes according to the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] Example: Figures 1-2 As shown, the present invention provides a technical solution: an intelligent monitoring system and method applied to composite molding processes, the method comprising: Step S1: Use laser 3D scanning technology to scan the electric heating mold from all angles to obtain the 3D point cloud data of the mold. Import the point cloud data into 3D modeling software for processing and model reconstruction to generate a 3D solid model of the electric heating mold. Step S1 includes: using a FARO Edge laser 3D scanner with a scanning accuracy of 0.02mm to perform an all-around scan of the electric heating mold to acquire point cloud data with a point cloud density of 10 points / ... Import the point cloud data into SolidWorks 3D modeling software, perform noise reduction, Gaussian filtering with a threshold of 0.05mm, and surface reconstruction through the "Point Cloud Processing" module to generate a 3D solid model of the mold.

[0014] Step S2: Select a curvature-based edge detection algorithm to calculate the curvature value of each point on the surface of the 3D model, identify the significant curvature points as the edges of irregular regions, extract the edge contours using 3D modeling software, calculate the area, convex hull area, and convexity shape factor geometric parameters of the region enclosed by the edge contours, and determine the irregular regions by calculating the convexity shape factor. Step S2 includes: Step S2-1: Extract surface geometric information from the CAD model and generate a set A consisting of 10,000 uniformly distributed sampling points. Use CloudCompare software to perform statistical filtering to remove outliers. Step S2-2: For each sampling point Selecting a=20 nearest neighbor points, the quadratic surface is fitted using MATLAB's "fitgeosurface" function, and the normal vector is calculated. ; Step S2-3: Solve for the first and second fundamental form coefficients of the quadratic surface to obtain the principal curvatures. and satisfy ≥ Calculate Gaussian curvature mean curvature ; Step S2-4: Set weights Calculate the Gaussian curvature gradient ∇K and the mean curvature gradient ∇G, where It is a unit direction vector; Step S2-5: Take =0.6、 =0.4, calculate the combined gradient magnitude. Set a threshold T=0.03 and select sampling vertices with G>T as significant curvature points; Steps S2-6 to S2-7: Sort the significant points of curvature clockwise and calculate the area of ​​the region using the shoelace formula. =0.23 ; Step S2-8: Initialize an empty stack. Iterate through the sorted points in sequence. If there are fewer than two elements in the stack, push them onto the stack directly. Otherwise, check if the top two points of the stack form a right turn with the current point. If they do, pop the top element from the stack. Continue this process until a left turn condition is met, then push the current point onto the stack. After the traversal is complete, the elements in the stack form the lower convex hull. After reversing the sorting, repeat the construction process of the lower convex hull to obtain the upper convex hull. Finally, merge the lower and upper convex hulls to obtain the complete convex hull. Steps S2-9 to S2-11: Construct the convex hull and then calculate its area. =0.25 Convexity shape factor C= =0.92<1, therefore the region is determined to be an irregular region.

[0015] Step S3: Based on the geometric features and thermal conductivity of the conventional area, formulate a sensor layout scheme, arrange temperature sensors in a unit space layout manner, collect temperature data, and obtain the temperature threshold of the conventional area; Step S3 includes: Step S3-1: For the regular area excluding irregular areas, divide the space into cubic units at 10cm intervals along the X, Y, and Z axes, for example, a total of 150 unit areas, and place Q=1 PT100 temperature sensor in each area; Step S3-2: Assuming the molding process requires a temperature range of 115-125℃, collect temperature data L=10 times and take the average value to obtain the temperature threshold for the normal area. =120℃; Step S4: By comparing the temperature of the irregular area with the temperature threshold of the regular area, adjust the arrangement of the temperature sensors in the irregular area to obtain the distribution pattern of the temperature sensors in the irregular area and establish a database of temperature sensor distribution patterns in the irregular area. Steps S4-1 to S4-2: Divide the irregular area into unit spaces at 10cm intervals, arrange Q=1 sensors, and collect L=10 temperature data points, averaging the results. =112℃, exceeding 0.95 =Lower limit of 114℃; Steps S4-3 to S4-5: Add one sensor to the corresponding unit area, i.e., Q=2, and re-collect the temperature after uniform distribution. At 116℃, 0.95 -1.05 Within the range, record the distribution pattern Record that at this point, the two sensors are diagonally distributed; Step S4-6: ... The data is associated with the curvature characteristics of the region, such as the average curvature G=0.02, and stored in a MySQL database to form a "three-dimensional feature - sensor distribution" mapping table.

[0016] The system includes: a 3D model building module, an irregular area filtering module, a regular area temperature acquisition module, and an irregular area temperature sensor layout module; The 3D model building module is used to generate a 3D solid model of the electric heating mold; The irregular area filtering module is used to filter out irregular areas of the electrically heated mold; The normal area temperature acquisition module is used to acquire the temperature threshold of the normal area; The irregular area temperature sensor layout module is used to obtain the layout of temperature sensors in irregular areas.

[0017] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. An intelligent monitoring method applied to composite molding processes, characterized in that: Includes the following steps: Step S1: Use laser 3D scanning technology to scan the electric heating mold from all angles to obtain the 3D point cloud data of the mold. Import the point cloud data into 3D modeling software for processing and model reconstruction to generate a 3D solid model of the electric heating mold. Step S2: Select a curvature-based edge detection algorithm to calculate the curvature value of each point on the surface of the 3D model, identify the significant curvature points as the edges of irregular regions, extract the edge contours using 3D modeling software, calculate the area, convex hull area, and convexity shape factor geometric parameters of the region enclosed by the edge contours, and determine the irregular regions by calculating the convexity shape factor. Step S3: Based on the geometric features and thermal conductivity of the conventional area, formulate a sensor layout scheme, arrange temperature sensors in a unit space layout manner, collect temperature data, and obtain the temperature threshold of the conventional area; Step S4: By comparing the temperature of the irregular area with the temperature threshold of the regular area, adjust the arrangement of the temperature sensors in the irregular area to obtain the distribution pattern of the temperature sensors in the irregular area and establish a database of temperature sensor distribution patterns in the irregular area.

2. The intelligent monitoring method for composite molding process according to claim 1, characterized in that: The curvature-based edge detection algorithm in step S2 uses a combination of Gaussian curvature and average curvature to calculate the curvature values ​​at each point on the surface of the 3D model. The specific steps include: Step S2-1: Extract surface geometric information from the 3D mold CAD model to generate 3D point cloud data. Through sampling processing, form a uniformly distributed set of sampling points. ,in This represents the first discrete sampling point on the corresponding mold surface. This represents the second discrete sampling point on the corresponding mold surface. This represents the third discrete sampling point on the corresponding mold surface, ... This represents the b-th discrete sampling point on the surface of the corresponding mold; then, the point cloud data is denoised and filtered to eliminate discrete errors in the model building process; Step S2-2: For each sampled vertex Select its a nearest neighbors to form a local neighborhood M( Within the local neighborhood, the least squares method is used to evaluate the local neighborhood M( Fit a quadratic surface Calculate the fitted surface midpoint Normal vector at point ; Step S2-3: Calculate the fitted surface At point The coefficients of the first and second fundamental forms are used to solve the characteristic equation and obtain the point. Principal curvature at and And satisfy ≥ Calculate Gaussian curvature and mean curvature principal curvature and These represent the curvature values ​​of the surface in the directions of maximum and minimum curvature, respectively; Step S2-4: For sampling vertices The formula for calculating its curvature gradient is: Gaussian curvature gradient: Mean curvature gradient: Where w(h) is the weight of the local neighborhood vertex h, Let g be the unit vector pointing from the sampled vertex g to the local neighborhood vertex h. Step S2-5: Combine gradient magnitudes: , where α and β are weighting coefficients, and a gradient threshold T is set. The sampling vertex where G>T is the curvature salient point.

3. The intelligent monitoring method for composite molding process according to claim 1, characterized in that: The steps for calculating the convex hull area and convexity shape factor in step S2 include: Step S2-6: Arrange all significant curvature points in clockwise order to obtain an ordered point set. ,in This represents the first point in the point set. This represents the second point in the point set. Let represent the 3rd point in the point set, ..., This represents the nth point in the point set; n represents the number of points in the point set; taking a point of significant curvature as the starting point, the X-axis extends horizontally to the right as the positive X-axis, and the Y-axis extends vertically downwards as the positive Y-axis, represented by the coordinates as follows: =( ); Step S2-7: Calculate the area of ​​the polygon using the shoelace formula: ,in Let x be the x-coordinate of the i-th vertex in the two-dimensional coordinate system. Let be the ordinate of the i-th vertex in the two-dimensional coordinate system. Let x be the x-coordinate of the (i+1)th vertex in the two-dimensional coordinate system. Let be the ordinate of the (i+1)th vertex in the two-dimensional coordinate system. Point and When the points coincide, a closed polygon is formed; Step S2-8: After sorting the point set by coordinates from smallest to largest, construct the lower convex hull by pushing it onto a stack, then traverse and sort it in reverse order to construct the upper convex hull, and finally merge the two to obtain the complete convex hull. Step S2-9: Arrange the vertices of the convex hull in clockwise order to obtain an ordered set of points. ,in This represents the first point in the point set. This represents the second point in the point set. Let represent the 3rd point in the point set, ..., This represents the nth point in the point set; n represents the number of points in the point set; taking a point of significant curvature as the starting point, the X-axis extends horizontally to the right as the positive X-axis, and the Y-axis extends vertically downwards as the positive Y-axis, represented by the coordinates as follows: =( ); Step S2-10: Calculate the area of ​​the convex hull using the shoelace formula. The formula is as follows: , Let x be the x-coordinate of the j-th vertex in the two-dimensional coordinate system. Let be the ordinate of the j-th vertex in the two-dimensional coordinate system. Let x be the x-coordinate of the (j+1)th vertex in the two-dimensional coordinate system. Let be the ordinate of the (j+1)th vertex in the two-dimensional coordinate system. Point and When the points coincide, a closed polygon is formed; Step S2-11: The formula for calculating the convexity shape factor is: , where C is the convexity shape factor; if C=1, it is convex; if C<1, it is concave.

4. The intelligent monitoring method for composite molding process according to claim 1, characterized in that: In step S3, the regular area is the remaining part after removing the irregular area; The steps for arranging unit space in a conventional area include: Step S3-1: Divide the regular area into equal intervals along the X, Y, and Z axes according to the 3D solid model to form several cubic unit space areas, and arrange Q sensors in each unit space area; Step S3-2: Based on the temperature data collected by the sensor in a unit space area, L times are recorded, and the average value is taken to obtain the temperature threshold of the regular area, which is denoted as . .

5. The intelligent monitoring method for composite molding process according to claim 1, characterized in that: The temperature sensor arrangement steps in step S4 include: Step S4-1: For irregular areas, set up spatial unit areas and arrange sensors according to step S3-1, and record the number as Q; Step S4-2: Obtain the temperature collected by the temperature sensor in the unit area of ​​space, collect data L times, and take the average value. ; Step S4-3: When the weighted average is not within the threshold range of 0.95 < <1.05 At this time, a temperature sensor is added to the corresponding unit space area, and the number is recorded as Q+1; Step S4-4: Distribute the Q+1 temperature sensors evenly in a unit space area to obtain the distribution pattern. ; Step S4-5: Repeat steps S6-2 to S6-4. When the weighted average value of the temperature sensor first falls within the threshold range, output the distribution pattern. Distribution pattern As a corresponding distribution pattern of a certain three-dimensional feature; Step S4-6: Establish a database based on the distribution pattern of temperature sensors in irregular areas.

6. An intelligent monitoring system for composite molding processes, used to execute the intelligent monitoring method for composite molding processes as described in any one of claims 1-5, characterized in that: The system includes: The module includes a 3D model building module, an irregular area filtering module, a regular area temperature acquisition module, and an irregular area temperature sensor layout module. The 3D model building module is used to generate a 3D solid model of the electric heating mold; The irregular area filtering module is used to filter out irregular areas of the electrically heated mold; The normal area temperature acquisition module is used to acquire the temperature threshold of the normal area; The irregular area temperature sensor layout module is used to obtain the layout of temperature sensors in irregular areas.

7. The intelligent monitoring system for composite molding processes according to claim 6, characterized in that: The three-dimensional model construction module includes: using laser three-dimensional scanning technology to perform an all-round scan of the electric heating mold, obtaining three-dimensional point cloud data of the mold, importing the point cloud data into three-dimensional modeling software for processing and model reconstruction, and generating a three-dimensional solid model of the electric heating mold.

8. The intelligent monitoring system for composite molding process according to claim 6, characterized in that: The irregular region filtering module includes: selecting a curvature-based edge detection algorithm to calculate the curvature value of each point on the surface of the 3D model, identifying points with significant curvature changes as the edges of irregular regions, extracting edge contours using 3D modeling software, calculating the area, convex hull area, and convexity shape factor geometric parameters of the region enclosed by the edge contours, and determining the irregular regions by calculating the convexity shape factor.

9. The intelligent monitoring system for composite molding process according to claim 6, characterized in that: The conventional area temperature acquisition module includes: formulating a sensor layout scheme based on the geometric features and thermal conductivity characteristics of the conventional area, arranging temperature sensors in a unit space layout manner, acquiring temperature data, and obtaining the conventional area temperature threshold.

10. The intelligent monitoring system for composite molding process according to claim 6, characterized in that: The irregular area temperature sensor layout module includes: adjusting the layout of temperature sensors in the irregular area by comparing the temperature of the irregular area with the temperature threshold of the regular area, obtaining the distribution pattern of temperature sensors in the irregular area, and establishing a database of temperature sensor distribution patterns in the irregular area.

Images