Production process optimization control method of mineral fiber composite hanging plate

Abstract

The invention discloses a production process optimization control method of a mineral fiber composite hanging plate. The method comprises the following steps: acquiring working condition data of the mineral fiber composite hanging plate in a typical use scene; constructing a three-dimensional structure model according to the working condition data, and obtaining a three-dimensional stress vector field map; extracting a corresponding fiber stress optimal arrangement direction in the stress vector field graph, and generating a target fiber arrangement vector graph; acquiring real-time distribution image data of the mineral fibers in the slurry; extracting an actual fiber distribution vector diagram in the current slurry state; according to the target fiber arrangement vector diagram and the actual fiber distribution vector diagram in each unit area, calculating an included angle difference value and a direction deviation value in each unit area to form a fiber arrangement error matrix; and generating a corresponding decentralized control parameter adjustment instruction based on the fiber arrangement error matrix. According to the control method, the mechanical strength, bending rigidity and durability of the composite hanging plate can be improved, and the service life of the composite hanging plate is effectively prolonged.

Description

Technical Field

[0001] This invention relates to the field of composite panel production technology, and in particular to a method for optimizing and controlling the production process of mineral fiber composite panels. Background Technology

[0002] Mineral fiber composite sidings are widely used as decorative and protective layers for building facades, tunnel interiors, and municipal facilities due to their advantages such as light weight, high strength, corrosion resistance, and good decorative properties. However, the structural performance of the sidings largely depends on the arrangement of the internal mineral fibers. Currently, the fiber arrangement in the production process of composite sidings mostly relies on empirical settings or natural settlement and random dispersion methods, lacking a system design and control mechanism that matches the response to actual working conditions. This leads to problems such as insufficient local strength and low bending stiffness during use, affecting service life and safety. Summary of the Invention

[0003] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0004] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for optimizing and controlling the production process of mineral fiber composite siding, comprising the following steps: Obtain operating condition data for mineral fiber composite panels under typical usage scenarios; Based on the aforementioned working condition data, a three-dimensional structural model of the mineral fiber composite panel was constructed, and a three-dimensional stress vector field map was obtained. Extract the optimal fiber stress distribution direction from the stress vector field map and generate the target fiber arrangement vector map; During the slurry forming process, real-time image data of the distribution of mineral fibers in the slurry is acquired; The image data is preprocessed to extract the actual fiber distribution vector map under the current slurry state; Based on the target fiber arrangement vector diagram and the actual fiber distribution vector diagram within each unit area, calculate the angle difference and direction deviation within each unit area to form a fiber arrangement error matrix. Based on the fiber arrangement error matrix, corresponding distributed control parameter adjustment instructions are generated.

[0005] As a preferred embodiment of the production process optimization and control method for the mineral fiber composite siding of the present invention, the working condition data includes temperature load, wind load, and siding constraint boundary conditions.

[0006] As a preferred embodiment of the production process optimization and control method for the mineral fiber composite siding described in this invention, the method includes the following steps: constructing a three-dimensional structural model of the mineral fiber composite siding based on the operating data and obtaining a three-dimensional stress vector field map. Establish a loading dataset that includes temperature load, wind load, and boundary constraint data; Based on the loaded dataset, a three-dimensional structural model of the mineral fiber composite panel is constructed using structural modeling tools. The temperature load, wind load, and boundary conditions are applied to the three-dimensional structural model to construct a multi-physics coupled finite element simulation model. Finite element analysis is performed on the finite element simulation model with multi-physics coupling to obtain the distribution results of stress direction and stress intensity covering the entire hanging plate structure, forming a three-dimensional stress vector field map.

[0007] As a preferred embodiment of the production process optimization and control method for the mineral fiber composite siding of the present invention, the method includes the following steps: extracting the optimal fiber stress distribution direction corresponding to the stress vector field spectrum and generating a target fiber arrangement vector map. In the three-dimensional stress vector field map, the hanging plate structure is divided into several unit regions with a preset division granularity; Within each unit region, the maximum principal stress direction vector of all finite element elements in that region is extracted to form a set of principal stress direction vectors. The target fiber arrangement principal direction is obtained by weighted averaging of multiple maximum principal stress direction vectors in each unit region. The main direction of the target fiber arrangement in each unit area is represented as a two-dimensional direction vector, generating a target fiber arrangement vector map, which is then image-encoded and stored.

[0008] As a preferred embodiment of the production process optimization and control method for the mineral fiber composite siding of the present invention, the image data includes fiber position and orientation information.

[0009] As a preferred embodiment of the production process optimization and control method for the mineral fiber composite siding of the present invention, the method includes the following steps: preprocessing the image data to extract the actual fiber distribution vector map under the current slurry state: Based on the preprocessed image data, the orientation gradient algorithm is used to identify the local orientation information of fibers at each pixel in the image and construct a pixel-level orientation gradient map. The constructed pixel-level gradient orientation map is divided into vectors aligned with the target fiber. Figure 1 The unit area grid is defined, and the pixel orientation distribution is extracted within each unit area. The main orientation is extracted and statistically aggregated. Based on the extraction of the main direction, and combining the image grayscale distribution and pixel gradient magnitude, the fiber density value of each unit region is estimated. ; Integrate the main direction and fiber density values ​​of all unit regions Generate the actual fiber distribution vector map of the slurry at the current moment and represent it in a structured matrix form.

[0010] As a preferred embodiment of the production process optimization and control method for the mineral fiber composite siding of the present invention, the formula for calculating the fiber density value is as follows:

[0011] in: : Represents the fiber density value per unit area; : The range of two-dimensional spatial coordinates for the current unit region; : for pixels The normalized grayscale value reflects the brightness information of the pixel, directly from the input image. Obtain the original grayscale value; : Indicates the main direction of the current pixel; : Represents an image At point The gradient vector represents the rate of change of gray level of the image at that point; : The gradient magnitude of a pixel, extracted from the edge information of the image; : A weighting function for the directional consistency of pixels in the main direction, used to enhance the contribution of pixels that are consistent with the overall direction of the region; The product of these three factors reflects the weighted contribution of each pixel to the fiber structure density. molecular : Represents the weighted sum of fiber structure values ​​for all pixels within a unit area, which is the total weighted contribution value; denominator : Indicates a region The area is used for normalization; Meaning of the range: This indicates the degree of fiber density per unit area; when This indicates that the region has no obvious fibrous structure or that the image has low grayscale and inconsistent orientation. when This indicates that the fibers in this area are clear, have high grayscale, and are aligned in the same direction.

[0012] As a preferred embodiment of the production process optimization and control method for the mineral fiber composite siding of the present invention, the method includes the following steps: calculating the angle difference and direction deviation values ​​within each unit area to construct a fiber arrangement error matrix: Obtain the principal direction vectors of corresponding unit regions in the target fiber arrangement vector map and the actual fiber distribution vector map, and denote them as follows: and ,in Indicates the index number of the unit area; Based on the vector dot product formula, the angle difference between the target fiber principal direction vector and the actual fiber principal direction vector is calculated. ; Calculate the directional deviation of the main direction of fiber distribution within a unit area. ; The angle difference value of each unit area Deviation from direction Combined into deviation pairs And form an error matrix. ; The constructed fiber arrangement error matrix is ​​stored in a structured manner.

[0013] As a preferred embodiment of the production process optimization and control method for the mineral fiber composite siding of the present invention, the method includes the following steps: Based on the fiber arrangement error matrix, generating corresponding distributed control parameter adjustment instructions. The fiber arrangement error matrix is ​​analyzed to extract the angle difference values ​​of each unit region. and direction deviation value Then, based on a preset error threshold, the unit area is divided into two priority areas: when Exceeding the first threshold, or When the threshold is exceeded, it is marked as high priority and will be corrected first. when Less than or equal to the first threshold, or When the value is less than or equal to the second threshold, it is marked as a progressive adjustment level, and progressive adjustment is performed according to the magnitude of the error; Establish the angle difference value and direction deviation value The dynamic mapping relationship with distributed control parameters; The mapped control parameters are sorted according to region priority to generate a time-seriesd control instruction set. The control instructions include, but are not limited to, segmented speed and direction instructions of the stirrer, and frequency and amplitude gradual change instructions of the vibration device. The actual fiber distribution vector map is continuously monitored by a visual acquisition device. The error matrix is ​​updated every Δt interval, and the control commands are dynamically corrected until the error values ​​in all areas are reduced to within the tolerance range.

[0014] As a preferred embodiment of the production process optimization and control method for the mineral fiber composite siding of the present invention, wherein: the included angle difference value and direction deviation value The dynamic mapping relationship with the distributed control parameters is as follows: For high-priority regions: if If the deviation is positive, increase the stir bar speed by 100 rpm per adjustment cycle until it reaches the upper limit of 1200 rpm; if the deviation is negative, decrease the speed by 100 rpm per adjustment cycle, with a minimum of 600 rpm. when When the display direction deviates clockwise, the stir bar switches to counterclockwise rotation; otherwise, it switches to clockwise rotation. Based on fiber density value Deviation from direction Product, through a nonlinear function Adjust the frequency of the vibration device and amplitude .

[0015] The beneficial effects of this invention are: 1. This invention constructs a three-dimensional structural model of a mineral fiber composite panel and performs finite element simulation based on actual working conditions to generate a three-dimensional stress vector field map. Based on this, the optimal fiber arrangement principal directions for each region are extracted to form a target fiber arrangement vector map, achieving precise synergistic matching between structural function and fiber arrangement. Compared to traditional methods based on empirical distribution or random arrangement, this invention significantly improves the mechanical strength, bending stiffness, and durability of the composite panel, effectively extending its service life.

[0016] 2. This invention acquires real-time fiber distribution images in the slurry state through a vision acquisition device installed on the molding cavity, extracts the actual fiber arrangement vector map, and constructs a fiber arrangement error matrix composed of angle differences and directional deviations. This error matrix enables quantitative comparative analysis between the fiber arrangement and the design target, and can output error maps at the unit area level, providing real-time, visual, and structured feedback information for subsequent dynamic control. This mechanism effectively overcomes the bottlenecks of existing technologies where fiber distribution states cannot be quantified and cannot be fed back.

[0017] 3. This invention establishes a mapping relationship between error values ​​and stirring speed, stirring direction, vibration frequency, and amplitude based on a fiber arrangement error matrix. It further generates multi-region, multi-parameter collaborative control commands to achieve targeted intervention in the fiber flow behavior of the slurry. Simultaneously, through a closed-loop feedback mechanism, image data is periodically collected and updated, and control parameters are dynamically corrected to ensure that the error continuously converges within a preset tolerance range. This control method offers rapid response and precise regulation, significantly improving the consistency and uniformity of fiber distribution in the slurry and effectively avoiding the generation of weak mechanical properties. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart illustrating a production process optimization and control method for a mineral fiber composite siding according to the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0020] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0022] Secondly, the present invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In addition, actual fabrication should include three-dimensional spatial dimensions of length, width, and depth.

[0023] Example 1 Reference Figure 1As an embodiment of the present invention, a method for optimizing and controlling the production process of mineral fiber composite siding is provided, mainly including: S1: Obtain working condition data of mineral fiber composite panels under typical usage scenarios.

[0024] Specifically, the operating data includes temperature load, wind load, and the constraint boundary conditions of the hanging plate.

[0025] It should be noted that temperature load refers to the stress caused by thermal expansion and contraction of mineral fiber composite panels due to changes in ambient temperature under typical usage conditions. Temperature load includes, but is not limited to, typical temperature change curves, maximum temperature difference, temperature duration intervals, and thermal conductivity boundary conditions in the usage environment.

[0026] Wind load is one of the main external mechanical loads that sidings suffer from in outdoor environments, whether over a long period or suddenly. It has a certain directionality and dynamic variation characteristics. Wind load includes, but is not limited to, prevailing wind direction, maximum wind pressure intensity, and wind field change frequency.

[0027] The constraint boundary conditions of a mounting plate refer to the physical constraints that the edge area of ​​a mounting plate is typically subjected to by the mounting structure or connectors during actual installation and use, resulting in different boundary conditions such as fixed, sliding, or elastic support. These constraint boundary conditions include, but are not limited to, the location of fixed points, the direction of restricted degrees of freedom, and the connection stiffness value.

[0028] In this invention, by collecting the above-mentioned working condition data, the stress vector field simulation analysis of the hanging panel can be performed based on the actual use environment, thereby providing a basis for the subsequent generation of target fiber arrangement vector diagram and correction of fiber arrangement error, and significantly improving the mechanical properties and durability stability of the finished hanging panel in complex environments.

[0029] S2: Construct a three-dimensional structural model of the mineral fiber composite panel based on the working condition data, and obtain a three-dimensional stress vector field map.

[0030] Specifically, a three-dimensional structural model of the mineral fiber composite panel is constructed based on the working condition data, and a finite element stress field simulation is performed on the three-dimensional structural model to obtain a three-dimensional stress vector field map containing stress direction and stress intensity distribution. This includes the following steps: S21: Create a loading dataset that includes temperature load, wind load, and boundary constraint data.

[0031] In practical applications, mineral fiber composite siding is often installed on building facades or structural cladding. Its long-term stability is closely affected by environmental temperature variations, wind load impacts, and the boundary constraints of the supporting structure. Therefore, this step first extracts the main load condition data triplets from "typical usage scenarios": temperature load + wind load + boundary constraints, establishing a complete and quantifiable loading dataset. This allows for the maximum possible reproduction of the stress loading path of the siding during actual service, thus avoiding the problem of simulations deviating from actual working conditions caused by modeling a single load.

[0032] S22: Based on the loaded dataset, call the structural modeling tool to construct a three-dimensional structural model of the mineral fiber composite panel.

[0033] Constructing a three-dimensional structural model is fundamental to finite element simulation. However, unlike traditional reinforced concrete or metal components, composite panels are multi-layered heterogeneous structures, and the anisotropy of materials and layered structure have a decisive impact on the simulation results. Therefore, advanced modeling tools that support heterogeneous material input (such as ANSYS and Abaqus) are needed to achieve layered modeling and separate definition of material parameters. Considering layering and anisotropy allows for a more realistic reflection of the surface layer's wind stress resistance, the load transferred by the core material, and the strength-enhancing effect of fiber reinforcement direction, thereby improving the targeting and scientific rigor of fiber arrangement direction control.

[0034] It should be noted that the three-dimensional structural model takes into account the geometric dimensions of the panel, the layered structure of the materials (such as the surface layer, core layer and fiber reinforcement layer), the anisotropic mechanical parameters of the materials, and the boundary constraint methods to ensure the engineering authenticity and analytical accuracy of the simulation results.

[0035] S23: Load temperature load, wind load and boundary conditions onto the three-dimensional structural model to construct a multi-physics coupled finite element simulation model.

[0036] It should be noted that composite panels face multi-physics coupling problems in actual service (thermal stress caused by temperature changes + wind-driven stress + stress concentration induced by local boundary rigidity differences). Therefore, it is necessary to construct a multi-physics coupling simulation model, which combines thermal-structural coupling, dynamic wind pressure simulation, and boundary conditions onto a three-dimensional model. This joint simulation can accurately obtain the stress distribution trends in different time periods and spatial regions, thereby identifying the areas in the structure most prone to failure or requiring reinforcement, providing a theoretical basis for fiber orientation control.

[0037] S24: Perform finite element solution on the multi-physics coupled finite element simulation model to obtain the distribution results of stress direction and stress intensity covering the entire hanging plate structure, and form a three-dimensional stress vector field map.

[0038] It should be noted that traditional simulations only focus on the magnitude of stress values ​​or the total deformation, while this step focuses on the distribution of principal stress directions in each region, using it as a "guide vector" for subsequent control of fiber arrangement. Therefore, the simulation results need to extract the principal stress direction vectors of each finite element to form a three-dimensional vector field map. Compared to simply outputting a color heatmap, this step provides directional information, which can be converted into a subsequent target arrangement vector map, serving as a crucial bridge guiding manufacturing control.

[0039] S3: Extract the optimal fiber stress distribution direction from the stress vector field map and generate the target fiber arrangement vector map; The target fiber arrangement vector diagram is used to represent the expected fiber arrangement direction in different regions.

[0040] This step involves converting the three-dimensional stress vector field map obtained in step S2 into a target fiber arrangement vector map required for manufacturing control, which is used to guide the orientation control of the fibers in the actual molding process.

[0041] Specifically, the optimal fiber stress distribution direction is extracted from the stress vector field spectrum, and a target fiber arrangement vector map is generated, including the following steps: S31: In the three-dimensional stress vector field map, the hanging plate structure is divided into several unit regions with a preset division granularity; in each unit region, the maximum principal stress direction vector of all finite element elements in the region is extracted to form a principal stress direction vector set.

[0042] It should be noted that a unit region is composed of multiple finite element elements, and the maximum principal stress direction vector is the extraction result of each element. Each finite element element typically has three principal stress values. , , and its corresponding direction vector , , To achieve regionalized control, we divided the entire panel structure into m×n two-dimensional regional elements (e.g., 10×20). Within each element, we calculated the direction of the maximum principal stress of all elements in that region (i.e., Corresponding direction ).

[0043] In this step, by dividing the entire panel structure into several unit regions (e.g., 10×20), differentiated control of fiber orientation in local areas can be achieved, rather than a one-size-fits-all overall orientation. This allows for more precise handling of stress characteristics in different regions. Furthermore, the direction of the maximum principal stress in each finite element element is the direction in which the material is most likely to produce a stress response, and it is also the optimal direction for the reinforcing fibers. Extracting and compiling these directions to construct a regional "orientation database" lays the foundation for subsequent orientation optimization.

[0044] S32: The target fiber arrangement principal direction is obtained by weighted averaging of multiple maximum principal stress direction vectors in each unit area.

[0045] It should be noted that the advantages of weighted averaging of multiple principal stress direction vectors within each unit region are as follows: Although the principal stress directions of different elements are different, their contributions to the local mechanical response vary. Weighted averaging can make the target direction more consistent with the principle of "most effective enhancement" by assigning higher weights to high-stress areas.

[0046] Directly arranging fibers based on stress direction units may result in severe directional jumps, while weighted averaging can yield a target fiber arrangement main direction with better continuity and higher manufacturing controllability.

[0047] S33: Generate the target fiber arrangement vector map and encode and store it in an image format.

[0048] Specifically, generating the target fiber arrangement vector map includes the following steps: The "target fiber arrangement main direction" of each unit region obtained in step S32 is represented as a two-dimensional direction vector (e.g., unit vector). , The entire panel is divided into m×n unit regions, and directional arrows are drawn in each region to form a complete target fiber arrangement vector diagram, which can be displayed in the form of a vector field.

[0049] It should be noted that, to achieve rapid transmission, retrieval, and visual analysis, the target fiber arrangement vector map undergoes image encoding. This image-encoded vector map serves as a "baseline template" for subsequent comparison with actual fiber orientation images. During the slurry forming process, the real-time images acquired by the visual acquisition device will extract the fiber distribution vector map, which will be compared one-to-one with the target map generated in this step to analyze the angle difference. This forms a fiber arrangement error matrix, driving the intelligent control unit to adjust stirring and vibration parameters, thus completing closed-loop control optimization.

[0050] S4: During the slurry molding process, a vision acquisition device installed on the molding cavity is used to acquire real-time image data of the distribution of mineral fibers in the slurry. The image data includes fiber position and orientation information.

[0051] S5: Preprocess the image data to extract the actual fiber distribution vector map under the current slurry state.

[0052] It should be noted that since the original image may have problems such as noise interference and insufficient contrast, image preprocessing operations are required, including grayscale conversion, bilateral filtering, contrast enhancement and edge sharpening, to improve the accuracy and robustness of subsequent fiber recognition.

[0053] Specifically, extracting the actual fiber distribution vector map under the current slurry state includes the following steps: S51: Based on the preprocessed image data, the orientation gradient algorithm is used to identify the local orientation information of fibers at each pixel in the image and construct a pixel-level orientation gradient map.

[0054] In this step, image processing algorithms such as the Sobel operator, Canny edge detection, or structural tensor analysis are applied to obtain the gradient direction of each pixel in the image, thereby inferring the local arrangement trend of fibers in the image. Combining structural tensor analysis further enhances the detection capability of long, strip-shaped fiber textures, thus establishing a pixel-level fiber orientation field across the entire image.

[0055] S52: Divide the constructed pixel-level gradient orientation map into vectors aligned with the target fiber arrangement. Figure 1 The system generates a uniform unit area grid. Within each unit area, it extracts the pixel orientation distribution and performs main orientation extraction and statistical aggregation.

[0056] In this step, to achieve one-to-one control with the target fiber arrangement vector map, the image is divided into m×n unit regions, the same as in S31. In each unit region, the distribution of pixel orientation information is statistically analyzed based on the orientation gradient map, a principal orientation probability model for that region is constructed using the orientation histogram, and the principal fiber orientation (i.e., the fiber principal orientation vector) is calculated using weighted average or principal component analysis (PCA).

[0057] For example, the entire pixel-level gradient orientation map is divided into 10×20 regions, and the statistical orientation distribution of a certain region is as follows: There are 50 pixels oriented at 30°-35°; and 10 pixels oriented at 90°. The principal direction of the region is determined to be 32° by using a weighted histogram or PCA, and this is taken as the principal direction vector of the unit region.

[0058] S53: Based on the extraction of the main direction, the fiber density value of each unit region is estimated by combining the image grayscale distribution and pixel gradient magnitude.

[0059] It should be noted that direction alone is insufficient to fully represent the fiber distribution; distribution density also needs to be evaluated. Therefore, the fiber distribution concentration within a region is determined by combining pixel grayscale values ​​with gradient magnitude.

[0060] The formula for calculating fiber density is as follows:

[0061] in: : Represents the fiber density value per unit area; : The range of two-dimensional spatial coordinates for the current unit region (i.e. (This can be set manually, or it can be a sliding window, image segmentation results, or candidate regions.) : for pixels The normalized grayscale value is usually a linear scaling of the image grayscale values ​​to 10 ... The range reflects the brightness information of a pixel, directly from the input image. Obtain the original grayscale value; : Indicates the main direction of the current pixel (calculated from the structure tensor or gradient, in radians); : Represents an image At point The gradient vector represents the rate of change of gray level of the image at that point, and is usually approximated by the Sobel operator, Scharr operator or structure tensor. : The gradient magnitude of a pixel, extracted from the edge information of the image; : A directional consistency weighting function for pixels, used to enhance the contribution of pixels consistent with the overall direction of the region. It is often defined as: ,in The main direction angle of the region, the range of this function is The more consistent the pixel orientation is with the main direction of the region, the higher the weight; among which... It can be obtained through the mean direction of all pixels in the region or through principal component analysis (PCA); The product of these three factors reflects the weighted contribution of each pixel to the fiber structure density. molecular : Represents the weighted sum of fiber structure values ​​for all pixels within a unit area, which is the total weighted contribution value; denominator : Indicates a region The area (i.e., the integral representation of the total number of pixels) is used for normalization; Meaning of the range: This indicates the degree of fiber density per unit area; when This indicates that the region has no obvious fibrous structure or that the image has low grayscale and inconsistent orientation. when This indicates that the fibers in this area are clear, have high grayscale, and are aligned in the same direction.

[0062] In the specific implementation process, the input image is first subjected to grayscale normalization processing to obtain each pixel point. grayscale value Subsequently, the gradient components in the horizontal and vertical directions of the image are extracted using the Sobel operator, and the gradient magnitude at that point is calculated. and direction angle The overall main direction of the region It is obtained by weighted averaging of all gradient directions within the region, and the directional consistency function is further calculated. Finally, multiply the above terms in the region. After integrating the inner components, divide by the area of ​​the region to obtain an estimated fiber density per unit area. .

[0063] S54: Integrate the main direction and fiber density values ​​of all unit regions to generate the actual fiber distribution vector map of the slurry state at the current moment, and represent it in a structured matrix form.

[0064] This step involves combining the principal direction vector of each unit region with the fiber density value to form a "region vector," which is then uniformly arranged to form the entire fiber distribution vector map (an m×n dimensional matrix, where each unit is a vector with direction and density).

[0065] S6: Based on the target fiber arrangement vector diagram and the actual fiber distribution vector diagram within each unit area, calculate the angle difference and direction deviation values ​​within each unit area to form a fiber arrangement error matrix.

[0066] Specifically, the angle difference and directional deviation values ​​within each unit area are calculated to form a fiber arrangement error matrix, including the following steps: S61: Obtain the principal direction vectors of corresponding unit regions in the target fiber arrangement vector map and the actual fiber distribution vector map, denoted as follows: and ,in Indicates the index number of the unit area; S62: Based on the vector dot product formula, calculate the angle difference between the target fiber principal direction vector and the actual fiber principal direction vector. The formula is:

[0067] in: Represents the dot product of vectors; Indicates the magnitude of the vector; S63: Calculate the directional deviation of the main direction of fiber distribution within a unit area. ; In a two-dimensional plane coordinate system, and Convert to polar coordinates angle and And calculate the direction deviation value:

[0068] in: Indicates unit area In the middle, the principal direction vector of the target fiber and the actual fiber principal direction vector The angular offset between the two directions, in other words, is the angle difference between the two directions. It is a function that represents a two-dimensional vector. The direction angle, that is, its relationship with The angle between axes, in radians, typically ranges from... This function uses mathematical... It is an improved arctangent function used to correctly determine which quadrant the angle is in.

[0069] For example: in a certain unit area middle: The target fiber direction vector is , indicating completely along Axial direction; The actual fiber direction vector is , indicates a 45° direction (in the first quadrant); Substitute into the formula to calculate the directional deviation value : radian

[0070] radian Therefore, the directional offset between these two vectors is 45° (or radian).

[0071] S64: Calculate the angle difference value for each unit region. Deviation from direction Combined into deviation pairs And form an error matrix. The details are as follows: The angle difference value of each unit area Deviation from direction Combined into deviation pairs And arrange them according to the spatial number of the unit area to construct the fiber arrangement error matrix. .

[0072] S65: The constructed fiber arrangement error matrix is ​​stored in a structured manner, that is, the fiber arrangement error matrix is ​​stored in a preset database in matrix form or image encoding form for real-time feedback input in the subsequent process parameter control process.

[0073] S7: Based on the fiber arrangement error matrix, generate corresponding dispersion control parameter adjustment instructions. The instructions are used to adjust the stirring speed and direction of the dispersion control unit, as well as the frequency and amplitude of the vibration device, in order to correct the fiber orientation in the slurry in real time.

[0074] Specifically, based on the fiber arrangement error matrix, corresponding distributed control parameter adjustment instructions are generated, including the following steps: S71: Analyze the fiber layout error matrix and extract the angle difference values ​​of each unit region. and direction deviation value Then, based on a preset error threshold, the unit area is divided into two types of areas: when Exceeding the first threshold (e.g., 15°), or If the value exceeds the second threshold (e.g., 0.3), it is marked as high priority and will be corrected first. when Less than or equal to the first threshold, or When the value is less than or equal to the second threshold, it is marked as a progressive adjustment level, and progressive adjustments are made according to the magnitude of the error.

[0075] This step effectively reduces the overall system response time and improves the efficiency and accuracy of fiber orientation correction by prioritizing the adjustment of areas with significant errors, while avoiding ineffective resource consumption and excessive system load.

[0076] S72: A step-by-step adjustment and limiting mechanism is adopted to establish a dynamic mapping relationship between the error value and the distributed control parameters. The mapping relationship is as follows: For high-priority areas, if For a positive deviation (actual direction lags behind target direction), increase the stir bar speed by 100 rpm per adjustment cycle until it reaches the upper limit of 1200 rpm; for a negative deviation, decrease the speed by 100 rpm per adjustment cycle, with a minimum of 600 rpm. when When the display direction deviates clockwise, the stir bar switches to counter-clockwise rotation; conversely, it switches to clockwise rotation. The rotation duration... Calculate using the following formula:

[0077] in This is a time adjustment factor, such as 2 seconds per unit of deviation density product.

[0078] Based on fiber density value Deviation from direction Product, through a nonlinear function Adjust the frequency of the vibration device and amplitude Vibration parameters are dynamically adjusted within the equipment's safe range, with adjustment steps not exceeding 5Hz and 0.5mm.

[0079] Based on the above mapping relationship, a multi-parameter collaborative control model is generated to ensure the coordination between stirring speed, direction of rotation and vibration parameters.

[0080] This step utilizes a refined, step-by-step parameter adjustment mechanism to achieve differentiated responses of stirring and vibration actions to different types of deviations, thereby improving the adaptability and robustness of control and effectively preventing oscillations or hysteresis caused by over- or under-control.

[0081] S73: Sort the mapped control parameters according to region priority to generate a time-seriesd control instruction set.

[0082] Control commands include, but are not limited to: segmented speed and direction commands for the stirrer; and gradual frequency and amplitude commands for the vibration device.

[0083] The purpose of this step is to transform the mapped control parameters into time-sequential and regionalized specific control commands, and to achieve coordination and synergy among multiple control parameters (such as speed, steering, frequency, and amplitude).

[0084] S74: Continuously monitor the actual fiber distribution vector map using a visual acquisition device, update the error matrix every Δt (e.g., 5 seconds), and dynamically correct control commands until the error values ​​in all areas are reduced to within the tolerance range. , It enters steady-state maintenance mode. If the detected regional error value is still higher than the tolerance threshold (e.g., or If the control command is not improved after two consecutive adjustments, it is marked as a "disturbed non-response area" and the image abnormal resampling mechanism is triggered.

[0085] The purpose of this step is to use vision devices for real-time error feedback and trend prediction to achieve a closed-loop control mechanism, enabling the control system to have dynamic adjustment and adaptive capabilities.

[0086] In summary, this invention constructs a three-dimensional structural model of a mineral fiber composite panel and performs finite element simulation based on actual working conditions (temperature load, wind load, and boundary constraints) to generate a three-dimensional stress vector field map. Based on this, the optimal fiber arrangement principal direction for each region is extracted to form a target fiber arrangement vector map, achieving precise synergistic matching between structural function and fiber arrangement. Compared to traditional empirical distribution or random arrangement methods, this invention significantly improves the mechanical strength, bending stiffness, and durability of the composite panel, effectively extending its service life. This invention uses a visual acquisition device installed on the molding cavity to acquire real-time fiber distribution images in the slurry state, extracts the actual fiber arrangement vector map, and constructs a fiber arrangement error matrix composed of angle differences and directional deviations. This error matrix enables quantitative comparative analysis between fiber arrangement and design targets, outputting error maps at the unit area level, providing real-time, visual, and structured feedback information for subsequent dynamic control. This mechanism effectively overcomes the bottleneck of existing technologies where fiber distribution states are not quantifiable or can not be fed back. This invention establishes a mapping relationship between error values ​​and stirring speed, stirring direction, vibration frequency, and amplitude based on a fiber arrangement error matrix. It further generates multi-region, multi-parameter collaborative control commands to achieve targeted intervention in the fiber flow behavior of the slurry. Simultaneously, through a closed-loop feedback mechanism, image data is periodically collected and updated, and control parameters are dynamically corrected to ensure that the error continuously converges within a preset tolerance range. This control method offers rapid response and precise adjustment, significantly improving the consistency and uniformity of fiber distribution in the slurry and effectively avoiding the formation of weak mechanical properties.

[0087] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for optimizing and controlling the production process of a mineral fiber composite siding, characterized in that, Includes the following steps: Obtain operating condition data for mineral fiber composite panels under typical usage scenarios; Based on the aforementioned working condition data, a three-dimensional structural model of the mineral fiber composite panel was constructed, and a three-dimensional stress vector field map was obtained. Extract the optimal fiber stress distribution direction from the stress vector field map and generate the target fiber arrangement vector map; During the slurry forming process, real-time image data of the distribution of mineral fibers in the slurry is acquired; The image data is preprocessed to extract the actual fiber distribution vector map under the current slurry state; Based on the target fiber arrangement vector diagram and the actual fiber distribution vector diagram within each unit area, calculate the angle difference and direction deviation within each unit area to form a fiber arrangement error matrix. Based on the fiber arrangement error matrix, corresponding distributed control parameter adjustment instructions are generated.

2. The method for optimizing and controlling the production process of mineral fiber composite siding as described in claim 1, characterized in that: The operating data includes temperature load, wind load, and hanging plate constraint boundary conditions.

3. The method for optimizing and controlling the production process of mineral fiber composite siding as described in claim 2, characterized in that: Based on the aforementioned working condition data, a three-dimensional structural model of the mineral fiber composite panel is constructed to obtain a three-dimensional stress vector field map, including the following steps: Establish a loading dataset that includes temperature load, wind load, and boundary constraint data; Based on the loaded dataset, a three-dimensional structural model of the mineral fiber composite panel is constructed using structural modeling tools. The temperature load, wind load, and boundary conditions are applied to the three-dimensional structural model to construct a multi-physics coupled finite element simulation model. Finite element analysis is performed on the multi-physics coupled finite element simulation model to obtain the distribution results of stress direction and stress intensity covering the entire hanging plate structure, forming a three-dimensional stress vector field map.

4. The method for optimizing and controlling the production process of mineral fiber composite siding as described in claim 3, characterized in that: Extracting the optimal fiber stress distribution direction from the stress vector field spectrum and generating a target fiber arrangement vector map includes the following steps: In the three-dimensional stress vector field map, the hanging plate structure is divided into several unit regions with a preset division granularity; Within each unit region, the maximum principal stress direction vector of all finite element elements in that region is extracted to form a set of principal stress direction vectors. The target fiber arrangement principal direction is obtained by weighted averaging of multiple maximum principal stress direction vectors in each unit region. The main direction of the target fiber arrangement in each unit area is represented as a two-dimensional direction vector, generating a target fiber arrangement vector map, which is then image-encoded and stored.

5. The method for optimizing and controlling the production process of mineral fiber composite siding as described in claim 1, characterized in that: The image data includes fiber position and orientation information.

6. The method for optimizing and controlling the production process of mineral fiber composite siding as described in claim 5, characterized in that: The image data is preprocessed to extract the actual fiber distribution vector map under the current slurry state, including the following steps: Based on the preprocessed image data, the orientation gradient algorithm is used to identify the local orientation information of fibers at each pixel in the image and construct a pixel-level orientation gradient map. The constructed pixel-level gradient map is divided into a unit region grid consistent with the target fiber arrangement vector map. Within each unit region, the pixel orientation distribution is extracted, and the main orientation is extracted and statistically aggregated. Based on the extraction of the main direction, and combining the image grayscale distribution and pixel gradient magnitude, the fiber density value of each unit region is estimated. ; Integrate the main direction and fiber density values ​​of all unit regions Generate the actual fiber distribution vector map of the slurry at the current moment and represent it in a structured matrix form.

7. The method for optimizing and controlling the production process of mineral fiber composite siding as described in claim 6, characterized in that: The formula for calculating the fiber density value is as follows: in: : Represents the fiber density value per unit area; : The range of two-dimensional spatial coordinates for the current unit region; : for pixels The normalized grayscale value reflects the brightness information of the pixel, directly from the input image. Obtain the original grayscale value; : Indicates the main direction of the current pixel; : Represents an image At point The gradient vector represents the rate of change of gray level of the image at that point; : The gradient magnitude of a pixel, extracted from the edge information of the image; : A weighting function for the directional consistency of pixels in the main direction, used to enhance the contribution of pixels that are consistent with the overall direction of the region; The product of these three factors reflects the weighted contribution of each pixel to the fiber structure density. molecular : Represents the weighted sum of fiber structure values ​​for all pixels within a unit area, which is the total weighted contribution value; denominator : Indicates a region The area is used for normalization; Meaning of the range: This indicates the degree of fiber density per unit area; when This indicates that the region has no obvious fibrous structure or that the image has low grayscale and inconsistent orientation. when This indicates that the fibers in this area are clear, have high grayscale, and are aligned in the same direction.

8. The method for optimizing and controlling the production process of mineral fiber composite siding as described in claim 6, characterized in that: The calculation of the angle difference and direction deviation within each unit area constitutes the fiber arrangement error matrix, including the following steps: Obtain the principal direction vectors of corresponding unit regions in the target fiber arrangement vector map and the actual fiber distribution vector map, and denote them as follows: and ,in Indicates the index number of the unit area; Based on the vector dot product formula, the angle difference between the target fiber principal direction vector and the actual fiber principal direction vector is calculated. ; Calculate the directional deviation of the main direction of fiber distribution within a unit area. ; The angle difference value of each unit area Deviation from direction Combined into deviation pairs And form an error matrix. ; The constructed fiber arrangement error matrix is ​​stored in a structured manner.

9. The method for optimizing and controlling the production process of mineral fiber composite siding as described in claim 8, characterized in that: Based on the fiber arrangement error matrix, corresponding distributed control parameter adjustment instructions are generated, including the following steps: The fiber arrangement error matrix is ​​analyzed to extract the angle difference values ​​of each unit region. and direction deviation value Then, based on a preset error threshold, the unit area is divided into two priority areas: when Exceeding the first threshold, or When the threshold is exceeded, it is marked as high priority and will be corrected first. when Less than or equal to the first threshold, or When the value is less than or equal to the second threshold, it is marked as a progressive adjustment level, and progressive adjustment is performed according to the magnitude of the error; Establish the angle difference value and direction deviation value The dynamic mapping relationship with distributed control parameters; The mapped control parameters are sorted according to region priority to generate a time-seriesd control instruction set. The control instructions include, but are not limited to, segmented speed and direction instructions of the stirrer, and frequency and amplitude gradual change instructions of the vibration device. The actual fiber distribution vector map is continuously monitored by a visual acquisition device. The error matrix is ​​updated every Δt interval, and the control command is dynamically corrected until the error values ​​in all areas are reduced to within the tolerance range, and the system enters a steady-state maintenance mode.

10. The method for optimizing and controlling the production process of mineral fiber composite siding as described in claim 9, characterized in that: The difference in the included angle and direction deviation value The dynamic mapping relationship with the distributed control parameters is as follows: For high-priority regions: if If the deviation is positive, increase the stir bar speed by 100 rpm per adjustment cycle until it reaches the upper limit of 1200 rpm; if the deviation is negative, decrease the speed by 100 rpm per adjustment cycle, with a minimum of 600 rpm. when When the display direction deviates clockwise, the stir bar switches to counterclockwise rotation; otherwise, it switches to clockwise rotation. Based on fiber density value Deviation from direction Product, through a nonlinear function Adjust the frequency of the vibration device and amplitude .

Images