Dynamic biomechanical field construction method for 3D printing protector customization

Abstract

The invention discloses a dynamic biomechanical field construction method for 3D printing protector customization, and the method comprises the steps: comprehensively collecting the dynamic pressure distribution data of a user in movement through a multi-angle impact experiment and maximum pooling operation, and generating an original impact thermodynamic diagram of the user; the optimal matching between the public model and the user impact thermodynamic diagram is quickly determined by using a multi-scale feature matching technology, and anisotropic scaling is performed on the public model to ensure that the model size adapts to the stature of the user and the impact force distribution feature. Through mechanical compensation parameter calculation, the crystal lattice rod diameter is adjusted to meet the mechanical property requirement of a user, the crystal lattice structure is adjusted in real time in the 3D printing process, and the accuracy and consistency of the protector are ensured. According to the scheme, the period from data acquisition to finished product delivery is remarkably shortened, the 3D customization cost is reduced, the production efficiency is improved, and wide application prospects are achieved.

Description

Technical Field

[0001] The present invention relates to a method for constructing a dynamic biomechanical field for customization of 3D printed protective gear. Background Art

[0002] There are two major bottlenecks in the current 3D printing customized protective gear technology: 1. Static Data Limitations: Existing technologies rely on CT or optical scanning to obtain static anatomical data, which is costly and cannot capture dynamic pressure distribution during movement, resulting in incomplete data. Due to muscle deformation, the contact surface between the protective gear and the human body changes rapidly, making traditional protective gear prone to displacement during movement.

[0003] 2. Offline design delays: Conventional processes require importing measurement data into CAD software for manual modeling, followed by 3D printing. Model design is time-consuming, requiring editing and uploading. This results in an average time delay of over several days from data collection to finished product delivery, resulting in high costs and a difficulty in scaling up. Summary of the Invention

[0004] The present invention aims to solve the problems existing in the above-mentioned prior art and provide a method for constructing a dynamic biomechanical field for customization of 3D printed protective gear.

[0005] The technical solutions adopted in the present invention are: A method for constructing a dynamic biomechanical field for 3D printed protective gear customization includes the following steps: 1) The user wears a force acquisition device with multiple thin-film force sensors and performs multiple impact tests. During each impact test, the maximum force value of each sensor is measured. Based on the fixed position relationship of the thin-film sensors in the force acquisition device, the maximum force value of each sensor is mapped to a two-dimensional pressure distribution matrix. 2) Perform a maximum pooling operation on the two-dimensional pressure distribution matrix generated by multiple impact experiments to synthesize a global feature map, namely the user's original impact heat map; 3) Based on the user's original impact heat map, the pre-designed standardized reference model is scaled and adjusted to suit the user's body shape and impact force distribution characteristics; 4) Perform multi-scale feature matching to determine the scaling coefficients kx and ky of the public model so that the impact heat map of the public model is optimally matched with the user's original impact heat map; 5) Anisotropically scale the public model according to the scaling coefficients kx and ky, and calculate the changes in the morphological parameters of the scaled model; 6) Calculate the mechanical compensation parameters to customize the rod diameter. According to the user's impact force requirements, adjust the lattice rod diameter to ensure that the scaled model meets the user's mechanical performance requirements; 7) In the real-time additive manufacturing process, the rod diameter of the lattice structure is adjusted in real time during the printing process based on the calculated customized rod diameter, enabling rapid customization of the protective gear.

[0006] Furthermore, suppose the user completes p shock experiments, each shock t∈[1, p], generating: ; in: represents the two-dimensional pressure distribution matrix generated by the t-th impact experiment; Represents an m×n real matrix, where m is the number of rows and n is the number of columns; It represents the force value of the thin film force sensor at position (i, j) in the t-th impact test. t represents the number of impact tests and its value range is t=1,2,…,p, where p is the total number of impact tests. The global feature map is then synthesized by maximum pooling, that is, each position (i, j) retains the maximum value within p collisions: .

[0007] Furthermore, the impact heat map G of the public version model s The original impact heat map of the user G u When matching, keep the thickness z of the protective gear unchanged, and scale the public version model by kx and ky times along the x and y directions.

[0008] Furthermore, the G s With the G u The matching process is: (a) Center pre-alignment First calculate the G s and the G u Heat map centroid coordinates: C s and C u : ; ; Among them: G s (i, j) is the value of the public model heat map at position (i, j); G u (i, j) is the value of the user's heat map at position (i, j); ∑G s is the sum of all values of the public version model heat map; ∑G u is the sum of all values in the user heat map; Calculate the initial translation based on the center of mass to pre-align the two heatmap centroids: ; Where: Δx0 is the initial translation in the x-axis direction; Δy0 is the initial translation in the y-axis direction; C s (x) 、C s (y) These are the x- and y-coordinates of the centroid of the heat map of the public model; C u (x) 、C u (y) The x-coordinate and y-coordinate of the centroid of the user's heat map respectively; (b) Translational search Define the translation search space so that the impact heat map Gs of the public version model and the user's original impact heat map Gu are The best similarity is achieved within the range: ; Where: Δx is the translation in the x-axis direction, Δy is the translation in the x-axis direction; δ1 is the range parameter of translation search, which is used to define the value range of translation amount; Keep the valid part of the translated image and calculate the similarity index ρT between the translated public version model heat map and the original impact heat map: ; in Molecules: Calculation With G u The covariance of With G u The sum of the products of the deviations at the corresponding positions after translation; Denominator: Calculation With G u The product of the standard deviations of is used to normalize the covariance so that the similarity value is in the range of [-1,1]; The closer ρT is to 1, the With G u The closer to -1, the more dissimilar; is the mean of the effective part of the translated public model heat map; is the mean of the effective part of the user's original impact heat map; Ω is With G u The overlapping area of the valid part that remains after translation; (c) Rotation search After obtaining the displacement Δx and Δy corresponding to the maximum similarity value, perform a rotation search and define the angle search range: ; Where: θ is the rotation angle of the public version model heat map; θ0 is the initial rotation angle; δ2 is the range parameter of the rotation search, which is used to define the value range of the rotation angle; The translated image is rotated by θ, and then the original impact heat map G is calculated. u The similarity of , get the rotation angle with the highest matching degree; (d) Zoom Search Based on translation and rotation, perform x,y anisotropic scaling optimization, and scale in the x and y dimensions to obtain the optimal scaling coefficients kx and ky: ; δ3 is the range parameter of the zoom search, which is used to define the value range of the zoom factor.

[0009] Furthermore, anisotropic scaling is performed on the public version model: ; After scaling, while keeping the z thickness unchanged, the lattice rod diameter r s , angle θ and lattice rod length L s will change; It is known that the lattice rod diameter in the original public version model becomes: ; The length of the lattice members in the original public model is: ; After scaling, the lattice member lengths are: ; The angles of the lattice members and the xy plane in the original reference model are scaled to: ; Where: M s For the original public version model; M' s To represent the scaled public version model; k x x is the scaled coordinate in the x direction; k y y is the coordinate after scaling in the y direction; z is the coordinate in the z direction, which remains unchanged because the z direction is not scaled; lx, ly, and lz correspond to the lengths of the lattice cubic unit in the original public model along the x, y, and z axes, respectively.

[0010] Furthermore, when calculating the mechanical compensation parameters, the average bearing capacity F of the original public version model s , and the average carrying capacity F for the custom modelu It is calculated in the following way: (1) Impact heat map G of the public version model s In the equation, find all regions where the force values are greater than the set threshold r, and then calculate the average force values of these regions, that is, the average bearing capacity F s ; (2) In the original shock heat map G u In the equation, find all regions where the force values are greater than the set threshold r, and then calculate the average force values of these regions, that is, the average bearing capacity F u ; The F s With F u The calculation formula corresponds to: ; .

[0011] Furthermore, the calculation process of the customized rod diameter is: Assuming that the lattice members are inclined members, the material obeys Hooke's law, and is within the elastic deformation range, the axial stiffness of the lattice members is: ; Where, E is the elastic modulus, A is the cross-sectional area of the lattice rod, and L is the s is the length of the lattice rod in the original public model; When a lattice rod is subjected to a force F, its component in the z direction is: ; The effective stiffness of the member in the z direction is: ; According to Hooke's law, the relationship between the force and deformation in the z direction is: ; Combining the axial stiffness and effective stiffness, we get: ; Among them: F s is the support force of the original public model along the z direction; r s is the lattice rod diameter of the original public model; θs is the angle between the lattice rod and the xy plane in the original public model; δz is the deformation in the z direction; Substitute the scaled lattice rod diameter r s ′, angle θ s ′ and length L s ′, the formula of the support force Fs′ of the scaled model is: ; According to the scaling relationship: ; ; ; Substituting the scaling relationship into the support force formula of the scaled model, we obtain: ; User-customized load capacity F u The formula is: ; Comparing the customized capacity with the scaled model capacity, we get: ; Further deduction to obtain the customized rod diameter r u formula: .

[0012] The present invention has the following beneficial effects: 1) Through multi-angle impact experiments and maximum pooling operations, this invention can comprehensively collect dynamic pressure distribution data of users during exercise, avoiding the limitations of traditional static data collection methods and effectively solving the problem of contact surface changes caused by muscle deformation.

[0013] 2) Use thin film force sensors to accurately record the maximum force value at each position and generate the user's original impact thermal map, providing an accurate data basis for subsequent model customization.

[0014] 3) Through multi-scale feature matching, including center pre-alignment, translation search, rotation search, and scaling search, the optimal match between the public version model and the user impact heat map can be quickly determined, greatly shortening the model adjustment time.

[0015] 4) The public version model is anisotropically scaled according to the user's impact heat map to ensure that the model size is adapted to the user's body shape and impact force distribution characteristics, achieving true personalized customization.

[0016] 5) Through mechanical compensation parameter calculation, the lattice rod diameter is adjusted to meet the user's mechanical performance requirements, ensuring the stability and support of the protective gear during exercise, and effectively reducing the risk of protective gear displacement.

[0017] 6) The lattice structure is adjusted according to the calculated custom rod diameter during the printing process, ensuring the accuracy and consistency of the protective gear.

[0018] 7) The average time required for the entire process, from data collection to finished product delivery, is significantly shortened. Compared to the traditional method's delivery cycle of several days, this technology enables rapid customization and improves production efficiency. This reduces the workload of manual modeling and support editing, lowers the cost of 3D customization, and makes the technology more valuable for promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1a Detect frontal impact map for the arm.

[0020] Figure 1b Detect the lower elbow impact graph for the arm.

[0021] Figure 1c Detect medial impaction maps for the arm.

[0022] Figure 1d Detect lateral impact maps for the arm.

[0023] Figure 2a This is the impact heat map of the public version model.

[0024] Figure 2b A heat map of the user's original impact.

[0025] Figure 3 This is a scaled version of the public model. DETAILED DESCRIPTION

[0026] The present invention will be further described below with reference to the accompanying drawings.

[0027] The present invention provides a method for constructing a dynamic biomechanical field for 3D printing protective gear customization, comprising the following steps: 1) The user wears a force acquisition device with multiple thin-film force sensors and performs multiple impact tests. During each impact test, the maximum force value of each sensor is measured. Based on the fixed position relationship of the thin-film sensors in the force acquisition device, the maximum force value of each sensor is mapped to a two-dimensional pressure distribution matrix. 2) Perform a maximum pooling operation on the two-dimensional pressure distribution matrix generated by multiple impact experiments to synthesize a global feature map, namely the user's original impact heat map; 3) Based on the user's original impact heat map, the pre-designed standardized reference model is scaled and adjusted to suit the user's body shape and impact force distribution characteristics; 4) Perform multi-scale feature matching, including center pre-alignment, translation search, rotation search, and scaling search, to determine the scaling factor of the public version model kx and ky , so that the impact heat map of the public model is best matched with the user's original impact heat map; 5) According to the scaling factor kx and ky, perform anisotropic scaling on the public version model and calculate the changes in the morphological parameters of the scaled model; 6) Calculate the mechanical compensation parameters to customize the rod diameter. According to the user's impact force requirements, adjust the lattice rod diameter to ensure that the scaled model meets the user's mechanical performance requirements; 7) In the real-time additive manufacturing process, the rod diameter of the lattice structure is adjusted in real time during the printing process based on the calculated customized rod diameter, enabling rapid customization of the protective gear.

[0028] The implementation process of this technical solution is further described below.

[0029] 1. Construction of dynamic biomechanical fields (1) Multi-angle impact information collection Assume the user completes p Shock experiment, each shock t∈[1, p ],generate: ; in: represents the two-dimensional pressure distribution matrix generated by the t-th impact experiment; Represents an m×n real matrix, where m is the number of rows and n is the number of columns; It represents the force value of the thin film force sensor at position (i, j) in the t-th impact test. t represents the number of impact tests and its value range is t=1,2,…,p, where p is the total number of impact tests. The global feature map is then synthesized by maximum pooling, that is, each position (i, j) retains the maximum value within p collisions: .

[0030] (2) Multi-scale feature matching There is an assumption that the impact heat maps and shapes of different users are basically similar, but because the user's body parts may be different from the size of the public version of the protective gear, the public version model needs to be changed. G s Heatmap of original impact with users G u When matching, keep the thickness z of the protective gear unchanged and match the public version model along the x and y directions kx and ky times the zoom.

[0031] Impact heat map of the public version model G s Heatmap of original impact with users G uWhen matching, four sub-processes are required, namely center pre-alignment, translation transformation, rotation transformation and scaling transformation. The corresponding ones are: (a) Center pre-alignment First calculate the G s and stated G u The centroid coordinates of the heat map: C s and C u : ; ; in: G s ( i , j ) is the public version model heat map at position ( i , j ) value; G u (i,j) Heatmap for users at location ( i , j ) value;∑ G s is the sum of all values of the public version model heat map; ∑ G u is the sum of all values in the user heat map; Calculate the initial translation based on the center of mass to pre-align the two heatmap centroids: ; Where: Δ x 0 is the initial translation in the x-axis direction; Δ y 0 is the initial translation in the y-axis direction; C s (x) 、 C s (y) These are the x- and y-coordinates of the centroid of the heat map of the public model; C u (x) 、 C u (y) The x-coordinate and y-coordinate of the centroid of the user's heat map respectively; (b) Translational search Define the translation search space so that the impact heat map Gs of the public version model and the user's original impact heat map Gu are The best similarity is achieved within the range: ; Where: Δ x is the translation in the x-axis direction, Δ y is the translation amount in the x-axis direction; δ 1 The range parameter for translation search is used to define the value range of the translation amount; Keep the valid part of the image after translation and calculate the heat map of the public version model after translation Heat map with original shock Gu Similarity index between ρT : ; in Molecules: Calculation With G u The covariance of With G u The sum of the products of the deviations at the corresponding positions after translation; Denominator: Calculation With G u The product of the standard deviations of is used to normalize the covariance so that the similarity value is in the range of [-1,1]; The closer ρT is to 1, the With G u The closer to -1, the more dissimilar; is the mean of the effective part of the translated public model heat map; is the mean of the effective part of the user's original impact heat map; Ω is With G u The overlapping area of the valid part that remains after translation; (c) Rotation search Get the displacement Δ corresponding to the maximum similarity value x , Δ y Then, perform a rotation search and define the angle search range: ; Where: θ is the rotation angle of the public version model heat map; θ0 is the initial rotation angle; δ2 is the range parameter of the rotation search, which is used to define the value range of the rotation angle; The translated image is rotated by θ, and then the original impact heat map G is calculated. u The similarity of , get the rotation angle with the highest matching degree; (d) Zoom Search Based on translation and rotation, perform x,y anisotropic scaling optimization, and scale in x and y dimensions to obtain the optimal scaling factor. kx and ky : ; δ 3 The range parameter for zoom search is used to define the value range of the zoom factor.

[0032] 2. Public model deformation and mechanical compensation algorithm (1) Scaling of the public version model Perform anisotropic scaling on the original lattice model (because we only want the area of the armor to change, and we don't want it to become thicker or thinner, so keep z unchanged): .

[0033] (2) Calculation of morphological parameter changes After scaling, while keeping the z thickness unchanged, the lattice rod diameter r s ,angle θ and lattice rod length L s will change; It is known that the lattice rod diameter in the original public version model becomes: ; The length of the lattice members in the original public model is: ; After scaling, the lattice member lengths are: ; The angles of the lattice members and the xy plane in the original reference model are scaled to: ; in: M s It is the original public version model; M ' s To represent the scaled public version model; k x x is the scaled coordinate in the x direction; k y y is the scaled coordinate in the y direction; z is the coordinate in the z direction, which remains unchanged; lx , ly , lzThey correspond to the lengths of the lattice cubic unit in the original public model on the x, y and z axes respectively.

[0034] (3) Calculation of mechanical compensation parameters When calculating the mechanical compensation parameters, the average bearing capacity of the original public model F s , and the average carrying capacity for custom models F u It is calculated in the following way: (1) Impact heat map of the public version model G s Find all the force values greater than the set threshold r Then calculate the average value of the force values in these areas, that is, the average bearing capacity F s ; (2) In the original shock thermogram G u Find all the force values greater than the set threshold r Then calculate the average value of the force values in these areas, that is, the average bearing capacity F u ; described F s and F u The calculation formula corresponds to: ; .

[0035] The calculation process for custom rod diameter is: Assuming that the lattice members are inclined members, the material obeys Hooke's law, and is within the elastic deformation range, the axial stiffness of the lattice members is: ; in, E is the elastic modulus, A is the cross-sectional area of the lattice rod, L s is the length of the lattice rod in the original public model; When the lattice rod is subjected to F When , its component in the z direction is: ; The effective stiffness of the member in the z direction is: ; According to Hooke's law, the relationship between the force and deformation in the z direction is: ; Combining the axial stiffness and effective stiffness, we get: ; in: F s is the support force of the original public model along the z direction; r s is the lattice rod diameter of the original public model; θs is the angle between the lattice rod and the xy plane in the original reference model; δz is the deformation in the z direction; Substitute the scaled lattice rod diameter r s ',angle θ s ′ and length L s ′, the support force of the scaled model Fs The formula for ′ is: ; According to the scaling relationship: ; ; ; Substituting the scaling relationship into the support force formula of the scaled model, we obtain: ; User-customized load capacity F u The formula is: ;

[0036] ; Further deduction to obtain customized rod diameter r u formula: .

[0037] In the 3D printing process, in order to make the lattice rod diameter change from the scaled lattice rod diameter r s 'Change to custom rod diameter ru , it is necessary to adjust the outline of the printed image in real time. The specific steps are as follows: (1) Relationship between pixel size and rod diameter: It is known that the pixel size of the light-curing printer is l p (the actual physical size of a single pixel), when projecting, if you need to make the lattice rod diameterr s ′ changes to r u , we need to calculate the pixel change Δ p . Pixel change calculation: ; (2) Explanation of pixel change: If Δ p >0, indicating that the rod diameter becomes thicker and the image contour needs to be expanded.

[0038] If Δ p <0, indicating that the rod diameter becomes thinner and the image contour needs to be eroded.

[0039] Δ p May be a fractional number, representing small changes in pixels.

[0040] (3) Image contour adjustment: Dilate or erode the image contour by Δ p Pixels are used to adjust the lattice rod diameter; an interpolation algorithm is used between two pixels to ensure a smooth transition of the image.

[0041] Case: Dynamic biomechanical field construction and protective gear customization 1) Construction of dynamic biomechanical fields 1. Multi-angle impact information collection: The user wears a mechanical acquisition device with multiple thin film force sensors and performs multiple impact experiments (such as Figure 1a 、 1b , 1c, and 1d).

[0042] In each impact test, the maximum force value of each sensor is detected and mapped to a two-dimensional pressure distribution matrix. Through the maximum pooling operation, the user's original impact heat map is synthesized. G u (like Figure 2a and Figure 2b ).

[0043] 2. Multi-scale feature matching: The impact heat map of the public version model G s Heatmap of original impact with users G u to match.

[0044] Determine the scaling factor of the public model through center pre-alignment, translation search, rotation search, and scaling search kx =1.15 and ky =1.085.

[0045] 2) Public model deformation and mechanical compensation 1. Public version model scaling (such as Figure 3 shown): According to the zoom factor k x and k y , anisotropic scaling of the public version model.

[0046] Lattice rod diameter of the original model r s =1.2mm, lattice rod diameter after scaling r s ′=1.311mm.

[0047] Length of lattice members of the original model L s =15.1mm, length after scaling L s ′=15.899mm.

[0048] 2. Calculation of mechanical compensation parameters: Average bearing capacity of the original model F s =10N / cm 2 .

[0049] User-customized load capacity F u =12 N / cm 2 .

[0050] Calculate the support force of the scaled model F s ′=12.22 N / cm 2 .

[0051] Custom rod diameter r u =1.42mm.

[0052] 3) Real-time additive manufacturing During the 3D printing process, the custom rod diameter is calculated r u , real-time adjustment of the rod diameter of the lattice structure r u - r s ′=1.42 - 1.311 = 0.109mm.

[0053] If the size of a single pixel of the printer l p The image edge is expanded outward by Δ by using the online image contour compensation technology.p =2.18 pixels; if the size of a single pixel on the printer l p The printed image edge should be extended outwards by Δ p =1.09 pixels; and so on, to ensure precise control of the lattice rod diameter.

[0054] This case demonstrates the complete process from dynamic biomechanical data collection to protective gear customization, ensuring that the protective gear's dimensions and mechanical properties meet the user's specific needs. Real-time adjustment and compensation enable rapid protective gear customization. The resulting protective gear model closely matches the user's requirements in both dimensions and mechanical properties.

[0055] The above description is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements can be made without departing from the principles of the present invention. These improvements should also be regarded as the scope of protection of the present invention.

Claims

1. A method for constructing a dynamic biomechanical field for customized 3D printed protective gear, characterized by: The steps include: 1) The user wears a force acquisition device with multiple thin-film force sensors and performs multiple impact tests. During each impact test, the maximum force value of each sensor is measured. Based on the fixed position relationship of the thin-film sensors in the force acquisition device, the maximum force value of each sensor is mapped to a two-dimensional pressure distribution matrix. 2) Perform a maximum pooling operation on the two-dimensional pressure distribution matrix generated by multiple impact experiments to synthesize a global feature map, namely the user's original impact heat map; 3) Based on the user's original impact heat map, the pre-designed standardized reference model is scaled and adjusted to suit the user's body shape and impact force distribution characteristics; 4) Perform multi-scale feature matching to determine the scaling coefficients kx and ky of the public model so that the impact heat map of the public model is optimally matched with the user's original impact heat map; 5) Anisotropically scale the public model according to the scaling coefficients kx and ky, and calculate the changes in the morphological parameters of the scaled model; 6) Calculate the mechanical compensation parameters to customize the rod diameter. According to the user's impact force requirements, adjust the lattice rod diameter to ensure that the scaled model meets the user's mechanical performance requirements; 7) In the real-time additive manufacturing process, the rod diameter of the lattice structure is adjusted in real time during the printing process based on the calculated customized rod diameter, enabling rapid customization of the protective gear.

2. The method for constructing a dynamic biomechanical field for customized 3D printed protective gear according to claim 1, wherein: Assume that the user completes p shock experiments, each shock t∈[1, p], generating: ； in: represents the two-dimensional pressure distribution matrix generated by the t-th impact experiment; Represents an m×n real matrix, where m is the number of rows and n is the number of columns; It represents the force value of the thin film force sensor at position (i, j) in the t-th impact test. t represents the number of impact tests and its value range is t=1,2,…,p, where p is the total number of impact tests. The global feature map is then synthesized by maximum pooling, that is, each position (i, j) retains the maximum value within p collisions: 。 3. The method for constructing a dynamic biomechanical field for customized 3D printed protective gear according to claim 1, wherein: In the impact heat map G of the public version model s The original impact heat map of the user G u When matching, keep the thickness z of the protective gear unchanged, and scale the public version model by kx and ky times along the x and y directions.

4. The method for constructing a dynamic biomechanical field for customized 3D printed protective gear according to claim 3, wherein: The G s With the G u The matching process is: (a) Center pre-alignment First calculate the G s and the G u Heat map centroid coordinates: C s and C u : ； ； Among them: G s (i, j) is the value of the public model heat map at position (i, j); G u (i, j) is the value of the user's heat map at position (i, j); ∑G s is the sum of all values of the public version model heat map; ∑G u is the sum of all values in the user heat map; Calculate the initial translation based on the center of mass to pre-align the two heatmap centroids: ； Where: Δx0 is the initial translation in the x-axis direction; Δy0 is the initial translation in the y-axis direction; C s (x) 、C s (y) These are the x- and y-coordinates of the centroid of the heat map of the public model; C u (x) 、C u (y) The x-coordinate and y-coordinate of the centroid of the user's heat map respectively; (b) Translational search Define the translation search space so that the impact heat map Gs of the public version model and the user's original impact heat map Gu are The best similarity is achieved within the range: ； Where: Δx is the translation in the x-axis direction, Δy is the translation in the x-axis direction; δ1 is the range parameter of translation search, which is used to define the value range of translation amount; Keep the valid part of the image after translation and calculate the heat map of the public version model after translation The similarity index ρT with the original shock heat map Gu: ； in Molecules: Calculation With G u The covariance of With G u The sum of the products of the deviations at the corresponding positions after translation; Denominator: Calculation With G u The product of the standard deviations of is used to normalize the covariance so that the similarity value is in the range of [-1,1]; The closer ρT is to 1, the With G u The closer to -1, the more dissimilar; is the mean of the effective part of the translated public model heat map; is the mean of the effective part of the user's original impact heat map; Ω is With G u The overlapping area of the valid part that remains after translation; (c) Rotation search After obtaining the displacement Δx and Δy corresponding to the maximum similarity value, perform a rotation search and define the angle search range: ； Where: θ is the rotation angle of the public version model heat map; θ0 is the initial rotation angle; δ2 is the range parameter of the rotation search, which is used to define the value range of the rotation angle; The translated image is rotated by θ, and then the original impact heat map G is calculated. u The similarity of , get the rotation angle with the highest matching degree; (d) Zoom Search Based on translation and rotation, perform x,y anisotropic scaling optimization, and scale in the x and y dimensions to obtain the optimal scaling coefficients kx and ky: ； δ3 is the range parameter of the zoom search, which is used to define the value range of the zoom factor.

5. The method for constructing a dynamic biomechanical field for customized 3D printed protective gear according to claim 4, characterized in that: Anisotropic scaling of the public model: ； After scaling, while keeping the z thickness unchanged, the lattice rod diameter r s , angle θ and lattice rod length L s will change; It is known that the lattice rod diameter in the original public version model becomes: ； The length of the lattice members in the original public model is: ； After scaling, the lattice member lengths are: ； The angles of the lattice members and the xy plane in the original reference model are scaled to: ； Where: M s For the original public version model; M' s To represent the scaled public version model; k x x is the scaled coordinate in the x direction; k y y is the scaled coordinate in the y direction; lx, ly, and lz correspond to the lengths of the lattice cubic unit in the original public model along the x, y, and z axes, respectively.

6. The method for constructing a dynamic biomechanical field for customized 3D printed protective gear according to claim 5, wherein: When calculating the mechanical compensation parameters, the average bearing capacity F of the original public version model s , and the average carrying capacity F for the custom model u It is calculated in the following way: (1) Impact heat map G of the public version model s In the equation, find all regions where the force values are greater than the set threshold r, and then calculate the average force values of these regions, that is, the average bearing capacity F s ; (2) In the original shock heat map G u In the equation, find all regions where the force values are greater than the set threshold r, and then calculate the average force values of these regions, that is, the average bearing capacity F u ; The F s With F u The calculation formula corresponds to: ； 。 7. The method for constructing a dynamic biomechanical field for customized 3D printed protective gear according to claim 6, wherein: The calculation process for custom rod diameter is: Assuming that the lattice members are inclined members, the material obeys Hooke's law, and is within the elastic deformation range, the axial stiffness of the lattice members is: ； Where, E is the elastic modulus, A is the cross-sectional area of the lattice rod, and L is the s is the length of the lattice rod in the original public model; When a lattice rod is subjected to a force F, its component in the z direction is: ； The effective stiffness of the member in the z direction is: ； According to Hooke's law, the relationship between the force and deformation in the z direction is: ； Combining the axial stiffness and effective stiffness, we get: ； Among them: F s is the support force of the original public model along the z direction; r s is the lattice rod diameter of the original public model; θs is the angle between the lattice rod and the xy plane in the original public model; δz is the deformation in the z direction; Substitute the scaled lattice rod diameter r s ′, angle θ s ′ and length L s ′, the formula of the support force Fs′ of the scaled model is: ； According to the scaling relationship: ； ； ； Substituting the scaling relationship into the support force formula of the scaled model, we obtain: ； User-customized load capacity F u The formula is: ； Comparing the customized capacity with the scaled model capacity, we get: ； Further deduction to obtain the customized rod diameter r u formula: 。

Images