High-ventilation-quantity seat surface cover processing method and system

Abstract

The invention belongs to the technical field of production big data application, and particularly relates to a high-ventilation seat cover processing method and system.The high-ventilation seat cover processing method comprises the steps that in a fabric cutting piece design drawing, a high-pressure area range is recognized according to pressure distribution data and a unified standard; the contour range capable of covering the high-pressure area range is determined, the shape and pressure distribution data of the high-pressure area range in the contour range and the material attribute data of the fabric cut-parts are input into a neural network model, the neural network model outputs a silica gel structure scheme, the silica gel structure scheme is converted into a three-dimensional form and adjusted, and the silica gel structure scheme is obtained. And forming a silica gel structure according to the adjusting scheme, and hot-pressing the corresponding fabric cutting pieces on the back surface of the outline range. According to the method, the neural network model is used for analyzing and optimizing the high-pressure area, the silica gel structure scheme is output, it is ensured that the high-pressure area is effectively supported, meanwhile, good ventilation performance is kept through the reasonably-distributed air holes, and the conflict between the ventilation requirement and the structural strength requirement is pointedly solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of production big data application, and in particular relates to a high-air-flow seat cover processing method and system. Background Art

[0002] The high-ventilation seat cover is a special structure applied to the seat surface, designed to provide better ventilation, especially when the seat is used for a long time, it can improve the user's comfort. By enhancing the circulation of air, the high-ventilation seat cover can effectively reduce the accumulation of heat and moisture, thereby alleviating sitting discomfort and reducing sweating. It is especially suitable for environments such as car seats and office chairs where there is long-term contact with the human body.

[0003] At present, with respect to the design and manufacture of high-ventilation seat covers, there is often a certain conflict between the ventilation requirements and the structural strength requirements. When the ventilation is good, the structural strength is often reduced due to the increase in ventilation area. Moreover, it is usually difficult to achieve a targeted balance between structural strength and ventilation performance for different positions of the seat cover. Summary of the invention

[0004] The present invention provides a high-ventilation seat cover processing method and system, which can effectively solve the problems in the background technology.

[0005] To achieve the above object, the technical solution of the present invention is as follows: A high ventilation volume seat cover processing method, comprising: Obtaining pressure distribution data of the seat during use; In the design drawings of each fabric piece of the seat cover, the high pressure area is identified according to the pressure distribution data and the unified standard; For each of the fabric pieces, determine a contour range that can cover the high-pressure area range, the contour range is greater than or equal to the high-pressure area range, input the shape and pressure distribution data of the high-pressure area range within the contour range and the material property data of the fabric piece into a neural network model, and the neural network model outputs a silicone structure solution, and the silicone structure is used to support the contour range; The silicone structure scheme is a data matrix, each element in the data matrix corresponds to a different area within the contour range, and the elements include position information of the center of the area on the silicone structure, average thickness information within the area, and judgment information on whether to set air holes; The silicone structure scheme is converted into a three-dimensional form and adjusted according to processing requirements, the silicone structure is formed according to the adjustment scheme, and the corresponding fabric pieces are hot-pressed on the back of the contour range to complete the processing of the required fabric pieces.

[0006] Furthermore, the neural network model includes: An input layer for inputting the shape and pressure distribution data and material property data; A convolutional neural network, which captures the shape and force characteristics in the shape and pressure distribution data through convolution kernels, and gradually captures the characteristics of different scales in the force distribution through multiple layers of convolution; A fully connected network, adding a plurality of hidden layers, wherein the hidden layers have a plurality of neurons, and capturing different combinations of material properties and the effects of the combinations on the silicone structure scheme through the neurons; A fusion layer, which combines the output features of the convolutional neural network and the fully connected network through a connection operation; The output layer outputs the data matrix of the silica gel structure scheme according to the merging results.

[0007] Furthermore, through the connection operation, the fusion layer obtains a characteristic graph including position information of the center of the region on the silicone structure, average thickness information in the region, and judgment information on whether to set air holes; The neural network model also includes an interpolation smoothing layer, which performs an interpolation operation on the feature map to smooth the thickness values ​​of adjacent areas; The output layer outputs a data matrix of the silica gel structure scheme according to the smoothing result.

[0008] Furthermore, a global constraint optimization function is introduced into the neural network model to ensure that the interpolation algorithm automatically adjusts the thickness value during the smoothing process so that the average thickness of each of the regions remains unchanged.

[0009] Furthermore, in the process of adjusting the three-dimensional model, the total air permeability area of ​​the silicone structure and the air permeability area of ​​a single air permeability hole are determined, wherein the total air permeability area is proportional to the size of the high-pressure area, and the air permeability area of ​​a single air permeability hole is proportional to the sum of the average thicknesses of the regions where the air permeability hole is set; Calculating the required number of air holes according to the total air permeability area and the air permeability area of ​​a single air hole; All of the ventilation holes are distributed at least in a partial area within the contour range where ventilation holes need to be set.

[0010] Furthermore, when the contour range is larger than the high-pressure area range, during the training of the neural network model, the element distribution density in the high-pressure area range is dynamically adjusted, and the element distribution density in other areas remains unchanged.

[0011] High air flow seat cover processing system, including: A data acquisition module, for acquiring pressure distribution data of the seat during use; A design drawing processing module, in the design drawings of each fabric piece of the seat cover, identifies the high-pressure area range according to the pressure distribution data and the unified standard; and, for each fabric piece, determines a contour range that can cover the high-pressure area range, wherein the contour range is greater than or equal to the high-pressure area range; A neural network module is integrated with a neural network model, wherein the shape and pressure distribution data of the high-pressure area within the contour range and the material property data of the fabric piece are used as inputs of the neural network model, and the silicone structure scheme is used as output of the neural network model, and the silicone structure is used to support the contour range; The silicone structure scheme is a data matrix, each element in the data matrix corresponds to a different area within the contour range, and the elements include position information of the center of the area on the silicone structure, average thickness information within the area, and judgment information on whether to set air holes; A three-dimensional model conversion module converts the silicone structure scheme into a three-dimensional form and adjusts it according to processing requirements; The production and processing module shapes the silicone structure according to the adjustment plan, and hot presses the corresponding fabric pieces on the back of the contour range to complete the processing of the required fabric pieces.

[0012] Furthermore, the neural network model includes: An input layer for inputting the shape and pressure distribution data and material property data; A convolutional neural network, which captures the shape and force characteristics in the shape and pressure distribution data through convolution kernels, and gradually captures the characteristics of different scales in the force distribution through multiple layers of convolution; A fully connected network, adding a plurality of hidden layers, wherein the hidden layers have a plurality of neurons, and capturing different combinations of material properties and the effects of the combinations on the silicone structure scheme through the neurons; A fusion layer, which combines the output features of the convolutional neural network and the fully connected network through a connection operation; The output layer outputs the data matrix of the silica gel structure scheme according to the merging results.

[0013] Furthermore, the fusion layer obtains a characteristic graph including position information of the center of the region on the silicone structure, average thickness information in the region, and judgment information on whether to set air holes through the connection operation; The neural network model also includes an interpolation smoothing layer, which performs an interpolation operation on the feature map to smooth the thickness values ​​of adjacent areas; The output layer outputs a data matrix of the silica gel structure scheme according to the smoothing result.

[0014] Further, the three-dimensional model conversion module includes a vent design unit; In the process of adjusting the three-dimensional model, the vent design unit determines the total vent area of ​​the silicone structure and the vent area of ​​a single vent, and calculates the required number of vents according to the total vent area and the vent area of ​​a single vent, wherein the total vent area is proportional to the size of the high-pressure area, and the vent area of ​​a single vent is proportional to the sum of the average thicknesses of the regions where the vent is set; And, all the ventilation holes are distributed at least in a partial area within the contour range where ventilation holes need to be set.

[0015] The technical solution of the present invention can achieve the following technical effects: The present invention collects pressure distribution data during the use of the seat, accurately identifies the high-pressure area ranges in different areas, and uses a neural network model to analyze and optimize these areas, outputs a silicone structure solution, and can design a support structure with appropriate thickness and strength for different pressure areas to ensure that the high-pressure area is effectively supported, while maintaining good ventilation performance through reasonably distributed air holes. This data-driven optimization design can accurately control the use of silicone materials, while reducing material waste and seat weight, and improving user comfort. Through the use of silicone structures, the face cover can be made of materials with better ventilation. While the overall ventilation performance is improved, the overall service life of the product is guaranteed through effective support of high-pressure positions, and the conflict between ventilation needs and structural strength needs is specifically resolved.

[0016] This method converts the silicone solution output by the neural network into a three-dimensional form, and through careful adjustment and hot pressing processes, ensures the precise combination of the silicone structure and the fabric pieces, further enhancing the support and ventilation performance of the seat. Through precise processing and sewing processes, the final high-ventilation seat cover not only has a good support effect, but also has the characteristics of uniform ventilation, comfort and durability, meeting the needs of long-term use. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A flow chart of a method for processing a high ventilation volume seat cover; Figure 2A schematic diagram of the position of the silicone structure relative to the back of the fabric piece; Figure 3 This is the framework diagram of the neural network model; Figure 4 The design flow chart for the vent hole; Figure numerals: 1. fabric piece; 2. silicone structure. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0021] Embodiment 1 High ventilation seat cover processing method, such as Figure 1 and 2 As shown, including: S1: Obtaining pressure distribution data of the seat during use; the pressure distribution data here can be collected during actual use, or can be obtained through targeted tests. Specifically, the pressure changes of different parts of the seat during use can be recorded by sensors, including the pressure data exerted on different parts of the seat by the user's actions such as sitting down, moving, and adjusting posture; S2: In the design drawings of each fabric piece of the seat cover, the high-pressure area is identified according to the pressure distribution data and the unified standard; the unified standard here means that the identification standard for the high-pressure area is consistent on each different fabric piece, and the high-pressure area can be identified by setting a threshold; S3: for each fabric piece, determine the contour range that can cover the high-pressure area range, the contour range is greater than or equal to the high-pressure area range, the shape and pressure distribution data of the high-pressure area range within the contour range, and the material property data of the fabric piece are input into the neural network model, and the neural network model outputs a silicone structure solution, and the silicone structure is used to support the contour range; whether the contour range is greater than the high-pressure area range in this step depends on whether the high-pressure area range is relatively regular, because the contour range determines the outer edge contour of the processed silicone structure, so when the high-pressure area range is irregular, covering it with a larger contour range can ensure the processability of the silicone structure, and if the outer edge contour of the high-pressure area range itself is relatively regular, it can be directly used as the contour range for processing the silicone structure; The silicone structure scheme is a data matrix, and each element in the data matrix corresponds to a different area within the contour range. The elements include the position information of the center of the area on the silicone structure, the average thickness information within the area, and the judgment information of whether to set the air hole; In this step, the shape and pressure distribution of the high-pressure area are analyzed through the neural network model, and a silicone structure that can adapt to complex force distribution can be designed, thereby ensuring that the support structure has different thicknesses and strengths in different areas, and can specifically support the high-pressure area to reduce the user's discomfort caused by long-term pressure. Through the automatic optimization of the neural network, the thickness and layout of the silicone structure are reasonably designed, and the thickness is only increased in areas that require additional support, while the use of materials is reduced in areas with lower pressure within the area, while also avoiding affecting the ventilation performance of the raw materials. This precise position and thickness control greatly reduces material waste and reduces the overall weight of the seat cover, making the seat lighter and suitable for cover materials with stronger ventilation but optimized structural strength; through the judgment information of the air holes, the location of the air holes can be reasonably planned; S4: Convert the silicone structure scheme into a three-dimensional form and adjust it according to the processing requirements, mold the silicone structure according to the adjustment scheme, and heat-press the corresponding fabric pieces on the back of the contour range to complete the processing of the required fabric pieces; in this step, the three-dimensional form will be carefully adjusted according to the thickness, air hole distribution, etc. to ensure that the final product meets the ergonomic requirements; in order to adapt to the specific needs of the processing technology, the three-dimensional silicone structure is fine-tuned, such as adjusting the smooth transition of the thickness, the size and distribution of the air holes in the air hole setting area, etc., and finally the corresponding mold and equipment can be used to mold a silicone structure that meets the design requirements; After the above steps are completed, the fabric pieces are sewn to obtain a high-ventilation seat cover. In the present invention, by collecting the pressure distribution data during the use of the seat, the high-pressure area range of different areas is accurately identified, and the neural network model is used to analyze and optimize these areas, and the silicone structure scheme is output. A support structure with appropriate thickness and strength can be designed for different pressure areas to ensure that the high-pressure area is effectively supported, while maintaining good ventilation performance through reasonably distributed air holes. This data-driven optimization design can accurately control the use of silicone materials, while reducing material waste and seat weight, and improving the comfort of the user. Through the use of silicone structures, the cover can be made of materials with better ventilation. While the overall ventilation performance is improved, the overall service life of the product is guaranteed through effective support of high-pressure positions, which specifically solves the conflict between ventilation needs and structural strength needs.

[0022] This method converts the silicone solution output by the neural network into a three-dimensional form, and through careful adjustment and hot pressing processes, ensures the precise combination of the silicone structure and the fabric pieces, further enhancing the support and ventilation performance of the seat. Through precise processing and sewing processes, the final high-ventilation seat cover not only has a good support effect, but also has the characteristics of uniform ventilation, comfort and durability, meeting the needs of long-term use.

[0023] As a preferred embodiment of the above, Figure 3 As shown, the neural network model includes: The input layer is used for inputting shape and pressure distribution data and material property data; the convolutional neural network captures the shape and force characteristics in the shape and pressure distribution data through the convolution kernel, and gradually captures the characteristics of different scales in the force distribution through multi-layer convolution; the fully connected network adds several hidden layers, and the hidden layers have several neurons. The neurons capture different combinations of material properties and the impact of the combination on the silicone structure scheme; the fusion layer merges the output features of the convolutional neural network and the fully connected network through a connection operation; the output layer outputs the data matrix of the silicone structure scheme according to the merged result.

[0024] In this preferred embodiment, a convolutional neural network is used to capture local features in shape and pressure distribution data using convolution kernels, which can automatically identify important patterns and features in the data and is particularly suitable for processing image-like pressure distribution data. Through multi-layer convolution operations, the network can gradually capture features of different scales in the force distribution. Specifically, shallow convolution can capture local subtle features, such as small high-pressure points or pressure concentration areas of specific shapes, while deep convolution can capture broader, global pressure distribution patterns. This multi-scale feature extraction capability enables the model to have high flexibility and accuracy in processing complex force shapes.

[0025] By adjusting the neuron connection weights during the training process in the fully connected network, the network can optimize the combination of material properties, such as elasticity, breathability, strength, etc., thereby generating silicone support structure designs suitable for different materials. The flexibility of this design enables the model to effectively adapt to a variety of material properties.

[0026] Finally, the fusion operation helps the model to better optimize the silicone structure solution, so that it can provide the best support effect and ventilation performance under the interaction of material properties and stress characteristics. The model not only pays attention to the shape of the pressure concentration area, but also combines the physical properties of the material to generate a design that is more suitable for the actual use environment. The neural network model has powerful feature capture, fusion, and optimization capabilities. Combining the advantages of convolutional neural networks and fully connected networks, it can effectively process shape and pressure distribution data and material property data, and output high-precision silicone structure solutions.

[0027] In the above optimization scheme, the design scheme of the silicone structure is a two-dimensional data matrix output by the neural network, including position information, thickness information and information on whether to set air holes. However, in the actual manufacturing process, when converting the two-dimensional data into a three-dimensional silicone structure, it is necessary to adjust it according to the specific processing technology to ensure that the design scheme can be smoothly implemented in actual production. Different processing technologies, such as mold forming, hot pressing, injection molding, etc., have specific requirements for the shape, size and pore distribution of silicone. For example, the mold may have restrictions on the range of variation of silicone thickness, and the hot pressing process may have specific requirements for the layout of the holes. Therefore, the three-dimensional model must be adjusted according to the actual processing parameters, such as accuracy, tolerance, etc., to ensure that the manufactured silicone structure meets the expected function and ensure that the molded structure has sufficient stability and support in terms of physical properties.

[0028] The output of the convolutional neural network is usually a multidimensional feature map, and the output of the fully connected network is a vector. Through the fusion operation, the convolutional network provides spatial information in the region, and the fully connected network provides optimization on material properties. The feature map obtained after merging contains all information related to shape, material properties and thickness design. As a preferred embodiment of the above, Through the connection operation, the fusion layer obtains a feature map including the position information of the center of the region on the silicone structure, the average thickness information in the region and the judgment information of whether to set air holes; the neural network model also includes an interpolation smoothing layer, which performs an interpolation operation on the feature map to smooth the thickness values ​​of adjacent regions; the output layer outputs the data matrix of the silicone structure scheme according to the smoothing results.

[0029] In this preferred solution, the interpolation smoothing layer smoothes the thickness values ​​of adjacent areas to ensure that the regional transition is more natural and smooth; the smoothed data matrix is ​​converted into a specific silicone structure design, which is ultimately helpful in optimizing the processing feasibility in production when it is formed in a three-dimensional form. The smooth transition of the thickness value makes the flow and forming of the silicone material smoother during the manufacturing process, especially through mold forming or hot pressing processes, and reduces manufacturing defects and poor forming phenomena caused by sudden changes in thickness. In addition, in actual applications, too abrupt thickness changes may affect the comfort of the seat, especially during long-term use. The thickness transition of the body contact area should be as smooth as possible. Through the interpolation smoothing layer, it can be ensured that the thickness changes in different pressure areas are more natural, thereby enhancing the support effect of the silicone structure on the user's body, and making the pressure distribution of the seat cover in different areas more uniform.

[0030] Compared with the three-dimensional model of the silicone structure, the feature map has a lower dimension and a smaller amount of data. Therefore, interpolation and smoothing operations on the feature map can significantly reduce computational complexity and resource usage, which not only saves computational time but also reduces the demand for computing power. Smoothing on the feature map can optimize and train the model more quickly, because the convolutional neural network has compressed the complex three-dimensional shape information into a two-dimensional feature map when extracting features. By interpolating the feature map, direct processing of complex three-dimensional data structures is avoided. Since the interpolation and smoothing operations are performed on the feature map of the neural network, the entire process can be optimized through gradient back propagation, allowing the network to automatically learn the optimal smoothing parameters. Smoothing operations in three-dimensional models are difficult to integrate into the network optimization process and require additional post-processing steps.

[0031] As a preferred embodiment of the above embodiment, a global constraint optimization function is introduced into the neural network model to ensure that the interpolation algorithm automatically adjusts the thickness value during the smoothing process so that the average thickness of each area remains unchanged.

[0032] This optimization solution can maintain the overall mechanical properties and design requirements of each area during thickness smoothing. Keeping the average thickness unchanged can ensure that while the transition is smooth, the support performance and material properties will not be changed by local adjustments, thereby avoiding imbalance in regional thickness distribution. This not only improves the structural stability and service life of the product, but also ensures that the functional requirements of the initial design are met. Keeping the key performance indicators in the design unchanged can improve the adaptability of the product to different body shapes and postures, and reduce the discomfort caused by excessive local thickness changes.

[0033] By introducing the global constrained optimization function during the implementation process, the interpolation algorithm can not only smooth the thickness transition of adjacent areas, but also strictly abide by the thickness limit of each area when performing the smoothing operation. This gives the interpolation algorithm greater flexibility and control, allowing the smooth transition to be carried out under the given constraints rather than blindly performing numerical smoothing. This control can ensure that the thickness change is not only smooth but also remains within a reasonable range, enhancing the stability of the interpolation algorithm.

[0034] In the above embodiment, information on whether ventilation holes are set in the set area of ​​the silicone structure can be obtained, that is, the setting requirements of the ventilation holes are clarified. In the actual application of the silicone structure, the reasons why ventilation holes are not set in individual areas may include: In high-pressure areas or areas where the body's weight is concentrated, the silicone structure needs to provide stronger support. Setting air holes in these areas may weaken the overall strength of the structure, resulting in reduced support effect, or even deformation or failure. At the edges or joints of the silicone structure, structural strength and stability are crucial. These areas are often subject to large tension or shear forces. If vents are opened, the strength of these areas may be weakened, resulting in the risk of tearing or breaking. In some high-friction or high-stress concentration areas, such as areas of prolonged contact with the skin or other surfaces and frequent friction, the setting of vents may accelerate the wear or fatigue of these areas, leading to premature failure of the silicone structure.

[0035] On the basis of the above situation, since the forming of the silicone structure is also affected by other factors, the specific setting scheme of the vent holes needs to be further optimized and selected, and is not directly output through the neural network model. As a preferred embodiment of the above embodiment, in the process of adjusting the three-dimensional model, Figure 4 As shown, the total air permeability area of ​​the silicone structure and the air permeability area of ​​a single air hole are determined, wherein the total air permeability area is proportional to the size of the high-pressure area, and the air permeability area of ​​a single air hole is proportional to the sum of the average thicknesses of the regions where the air holes are set; the required number of air holes is calculated based on the total air permeability area and the air permeability area of ​​a single air hole; all the air holes are distributed at least in the partial region where the air holes are required to be set within the contour range.

[0036] In the above embodiment, three steps are specifically adopted for the design of the vent hole: The first step is to determine the setting position on the silicone structure, which is completed by the neural network model; Step 2: Determine the ventilation area of ​​the ventilation hole, which is completed based on the average thickness information in the area in the obtained data matrix; Step 3: Determine the number of ventilation holes, which is achieved under the premise that the total ventilation area is determined.

[0037] The above logic effectively combines intelligent tools with precise thickness, area, and quantity control, ensuring that the silicone structure maintains its support and durability while providing efficient breathability, greatly improving the design accuracy, manufacturing feasibility, and user comfort.

[0038] The specific distribution scheme here can be determined based on the requirements of aesthetics and processability. During implementation, when the high-pressure area is larger, the overall pressure distribution area is larger, so more ventilation is needed to dissipate heat and sweat. By making the total ventilation area of ​​all air holes proportional to the size of the high-pressure area, sufficient ventilation capacity is ensured in areas with higher pressure, thereby effectively regulating the airflow and improving the comfort of the user. The design in which the area of ​​a single air hole is proportional to the sum of the average thickness of the areas where the air hole is set helps to appropriately increase the size of a single air hole in a thicker silicone structure, thereby increasing air permeability, and appropriately reducing the size of a single air hole in a thinner silicone structure to avoid insufficient support due to excessively large air holes, thereby achieving a balance between structural support and air permeability.

[0039] At the same time, the use of a ventilation hole design with a uniform ventilation area simplifies the manufacturing process, improves production efficiency and design flexibility, and enables the silicone structure to meet ventilation needs while also having good structural stability and durability.

[0040] When the contour range is larger than the high-pressure area, during the training of the neural network model, the element distribution density in the high-pressure area is dynamically adjusted, while the element distribution density in other areas remains unchanged. Specifically, during the training process, the model will gradually increase its element distribution density and reduce the error through error feedback to the high-pressure area. This refined dynamic adjustment can better optimize the design accuracy of the high-pressure area and ensure support performance, comfort and durability. Dynamically adjusting the element density in the high-pressure area avoids the computational burden of increasing the number of elements globally, optimizes the computational efficiency, and the model invests more computing resources in the high-pressure area while maintaining a simplified computational process in the low-pressure area, thereby achieving effective allocation of computing resources, improving training speed, saving overall training time, and achieving faster convergence. By limiting the density of the low-pressure area to remain unchanged, the model focuses on the optimization task in the high-pressure area, reduces the error fluctuation in the low-pressure area, and improves the generalization ability of the overall model. Specifically, a maximum element distribution density upper limit can be set in the high-pressure area to ensure that the model will not be difficult to train or cause unnecessary computational burden due to too many elements.

[0041] Embodiment 2 High air flow seat cover processing system, including: A data acquisition module, for acquiring pressure distribution data of the seat during use; A design drawing processing module, in the design drawings of each fabric piece of the seat cover, identifies the high-pressure area range according to the pressure distribution data and the unified standard; and, for each fabric piece, determines the contour range that can cover the high-pressure area range, and the contour range is greater than or equal to the high-pressure area range; The neural network module integrates a neural network model, takes the shape and pressure distribution data of the high-pressure area within the contour range and the material property data of the fabric piece as the input of the neural network model, and takes the silicone structure scheme as the output of the neural network model, and the silicone structure is used to support the contour range; The silicone structure scheme is a data matrix, and each element in the data matrix corresponds to a different area within the contour range. The elements include the position information of the center of the area on the silicone structure, the average thickness information within the area, and the judgment information of whether to set the air hole; 3D model conversion module, which converts the silicone structure scheme into 3D form and adjusts it according to processing requirements; The production and processing module forms the silicone structure according to the adjustment plan, and hot presses the corresponding fabric pieces on the back of the contour range to complete the processing of the required fabric pieces.

[0042] As a preferred embodiment of this embodiment, the neural network model includes: The input layer is used for inputting shape and pressure distribution data and material property data; the convolutional neural network captures the shape and force characteristics in the shape and pressure distribution data through the convolution kernel, and gradually captures the characteristics of different scales in the force distribution through multi-layer convolution; the fully connected network adds several hidden layers, and the hidden layers have several neurons. The neurons capture different combinations of material properties and the impact of the combination on the silicone structure scheme; the fusion layer merges the output features of the convolutional neural network and the fully connected network through a connection operation; the output layer outputs the data matrix of the silicone structure scheme according to the merged result.

[0043] As a preferred embodiment of this embodiment, the fusion layer obtains a feature map including the position information of the center of the region on the silicone structure, the average thickness information in the region and the judgment information of whether to set the air vents through a connection operation; the neural network model also includes an interpolation smoothing layer, which performs an interpolation operation on the feature map to smooth the thickness values ​​of adjacent regions; the output layer outputs the data matrix of the silicone structure scheme according to the smoothing results.

[0044] Among them, the three-dimensional model conversion module includes a vent design unit; in the process of adjusting the three-dimensional model, the vent design unit determines the total vent area of ​​the silicone structure and the vent area of ​​a single vent, and calculates the required number of vents based on the total vent area and the vent area of ​​a single vent, wherein the total vent area is proportional to the size of the high-pressure area, and the vent area of ​​a single vent is proportional to the sum of the average thicknesses of the areas where the vents are set; and all the vents are distributed at least in the partial area where the vents need to be set within the contour range.

[0045] The technical effect achieved by this embodiment is the same as that of the above-mentioned embodiment 1, and will not be repeated here.

[0046] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for processing a seat cover with high ventilation volume, characterized in that: include: Obtaining pressure distribution data of the seat during use; In the design drawings of each fabric piece of the seat cover, the high pressure area is identified according to the pressure distribution data and the unified standard; For each of the fabric pieces, determine a contour range that can cover the high-pressure area range, the contour range is greater than or equal to the high-pressure area range, input the shape and pressure distribution data of the high-pressure area range within the contour range and the material property data of the fabric piece into a neural network model, and the neural network model outputs a silicone structure solution, and the silicone structure is used to support the contour range; The silicone structure scheme is a data matrix, each element in the data matrix corresponds to a different area within the contour range, and the elements include position information of the center of the area on the silicone structure, average thickness information within the area, and judgment information on whether to set air holes; The silicone structure scheme is converted into a three-dimensional form and adjusted according to processing requirements, the silicone structure is formed according to the adjustment scheme, and the corresponding fabric pieces are hot-pressed on the back of the contour range to complete the processing of the required fabric pieces.

2. The high air flow rate seat cover processing method according to claim 1, characterized in that: The neural network model includes: An input layer for inputting the shape and pressure distribution data and material property data; A convolutional neural network, which captures the shape and force characteristics in the shape and pressure distribution data through convolution kernels, and gradually captures the characteristics of different scales in the force distribution through multiple layers of convolution; A fully connected network, adding a plurality of hidden layers, wherein the hidden layers have a plurality of neurons, and capturing different combinations of material properties and the effects of the combinations on the silicone structure scheme through the neurons; A fusion layer, which combines the output features of the convolutional neural network and the fully connected network through a connection operation; The output layer outputs the data matrix of the silica gel structure scheme according to the merging results.

3. The high air flow rate seat cover processing method according to claim 2, characterized in that: Through the connection operation, the fusion layer obtains a characteristic graph including position information of the center of the region on the silicone structure, average thickness information in the region, and judgment information on whether to set air holes; The neural network model also includes an interpolation smoothing layer, which performs an interpolation operation on the feature map to smooth the thickness values ​​of adjacent areas; The output layer outputs a data matrix of the silica gel structure scheme according to the smoothing result.

4. The high air flow rate seat cover processing method according to claim 3, characterized in that: A global constraint optimization function is introduced into the neural network model to ensure that the interpolation algorithm automatically adjusts the thickness value during the smoothing process so that the average thickness of each region remains unchanged.

5. The high air flow rate seat cover processing method according to claim 2, characterized in that: In the process of adjusting the three-dimensional model, the total air permeability area of ​​the silicone structure and the air permeability area of ​​a single air permeability hole are determined, wherein the total air permeability area is proportional to the size of the high-pressure area, and the air permeability area of ​​a single air permeability hole is proportional to the sum of the average thicknesses of the regions where the air permeability hole is set; Calculating the required number of air holes according to the total air permeability area and the air permeability area of ​​a single air hole; All of the ventilation holes are distributed at least in a partial area within the contour range where ventilation holes need to be set.

6. The high air flow rate seat cover processing method according to claim 1, characterized in that: When the contour range is larger than the high-pressure area range, during the training of the neural network model, the element distribution density in the high-pressure area range is dynamically adjusted, and the element distribution density in other areas remains unchanged.

7. High ventilation volume seat cover processing system, characterized in that: include: A data acquisition module, for acquiring pressure distribution data of the seat during use; A design drawing processing module, in the design drawings of each fabric piece of the seat cover, identifies the high-pressure area range according to the pressure distribution data and the unified standard; and, for each fabric piece, determines a contour range that can cover the high-pressure area range, wherein the contour range is greater than or equal to the high-pressure area range; A neural network module is integrated with a neural network model, wherein the shape and pressure distribution data of the high-pressure area within the contour range and the material property data of the fabric piece are used as inputs of the neural network model, and the silicone structure scheme is used as output of the neural network model, and the silicone structure is used to support the contour range; The silicone structure scheme is a data matrix, each element in the data matrix corresponds to a different area within the contour range, and the elements include position information of the center of the area on the silicone structure, average thickness information within the area, and judgment information on whether to set air holes; A three-dimensional model conversion module converts the silicone structure scheme into a three-dimensional form and adjusts it according to processing requirements; The production and processing module shapes the silicone structure according to the adjustment plan, and hot presses the corresponding fabric pieces on the back of the contour range to complete the processing of the required fabric pieces.

8. The high air flow rate seat cover processing system according to claim 7, characterized in that: The neural network model includes: An input layer for inputting the shape and pressure distribution data and material property data; A convolutional neural network, which captures the shape and force characteristics in the shape and pressure distribution data through convolution kernels, and gradually captures the characteristics of different scales in the force distribution through multiple layers of convolution; A fully connected network, adding a plurality of hidden layers, wherein the hidden layers have a plurality of neurons, and capturing different combinations of material properties and the effects of the combinations on the silicone structure scheme through the neurons; A fusion layer, which combines the output features of the convolutional neural network and the fully connected network through a connection operation; The output layer outputs the data matrix of the silica gel structure scheme according to the merging results.

9. The high air flow rate seat cover processing system according to claim 8, characterized in that: The fusion layer obtains a characteristic graph including position information of the center of the region on the silicone structure, average thickness information in the region, and judgment information on whether to set air holes through the connection operation; The neural network model also includes an interpolation smoothing layer, which performs an interpolation operation on the feature map to smooth the thickness values ​​of adjacent areas; The output layer outputs a data matrix of the silica gel structure scheme according to the smoothing result.

10. The high air flow rate seat cover processing system according to claim 7, characterized in that: The three-dimensional model conversion module includes a vent design unit; In the process of adjusting the three-dimensional model, the vent design unit determines the total vent area of ​​the silicone structure and the vent area of ​​a single vent, and calculates the required number of vents according to the total vent area and the vent area of ​​a single vent, wherein the total vent area is proportional to the size of the high-pressure area, and the vent area of ​​a single vent is proportional to the sum of the average thicknesses of the regions where the vent is set; And, all the ventilation holes are distributed at least in a partial area within the contour range where ventilation holes need to be set.

Images