An intelligent light-cured 3D printing model optimization method

Abstract

The application provides a kind of intelligent light curing 3D printing model optimization method, it is related to light curing 3D printing technical field, the method obtains the 3D printing model file uploaded by user;Analysis geometric information in model file, extract triangle mesh structure, normal vector information and surface area;Identify the key feature area of surface, corner and edge in model, collect printing parameter data, and generate target model summary file;Define optimization printing target, generate optimization model based on generative adversarial network, evaluate model quality through discriminator network, analyze the result of multi-objective optimization, find the optimal balance point between different targets;Test the optimized model and printing parameter using virtual reality, make adaptive correction according to simulation test result, adjust to the best printing parameter.The application automatically analyzes the geometric characteristics and material properties of 3D model, identifies potential weaknesses in the model, and intelligently optimizes the 3D printing model.

Description

Technical Field

[0001] The present invention relates to the technical field of photocuring 3D printing, and in particular to an intelligent photocuring 3D printing model optimization method. Background Art

[0002] Stereolithography (SLA) is a high-precision additive manufacturing technology that creates three-dimensional objects by solidifying photosensitive resin layer by layer. This technology offers high resolution and excellent surface finish, making it widely used in medical devices, precision industrial components, jewelry, and other fields. Despite its significant advantages, SLA still faces several challenges in practical application.

[0003] Traditional stereolithography 3D printing relies on manual experience to set printing parameters (such as infill ratio, layer thickness, and print speed), which have a direct impact on print quality. However, due to model complexity and variations in the printing environment, a single parameter setting cannot meet the needs of all models, resulting in unstable print quality, surface defects, and insufficient structural strength. Due to the lack of intelligent optimization methods, the traditional printing process often results in material waste. The support structure and infill method in the model are often manually set by the user, which can lead to inappropriate material usage, increasing production costs and time.

[0004] With the development of technology, the market demand for efficient and precise light-curing 3D printing technology is growing. In order to improve printing efficiency, reduce material waste, and ensure print quality, an intelligent method is urgently needed to optimize models and adjust printing parameters to meet different printing requirements. Summary of the Invention

[0005] In order to solve the above problems, the present invention proposes an intelligent light-curing 3D printing model optimization method, which automatically analyzes the geometric features and material properties of the 3D model, identifies potential weaknesses in the model, and intelligently optimizes the 3D printing model.

[0006] The present invention is achieved by adopting the following technical solutions:

[0007] In a first aspect, the present invention provides an intelligent light-curing 3D printing model optimization method, the method comprising the following steps:

[0008] Get the 3D printing model file uploaded by the user;

[0009] Analyze the geometric information in the model file and extract the triangle mesh structure, normal vector information and surface area;

[0010] Identify key feature areas of surfaces, corners, and edges in the model, collect printing parameter data, and generate target model summary files;

[0011] Define the optimized printing target according to the target model summary file, generate the optimization model based on the generative adversarial network, evaluate the model quality through the discriminator network, analyze the results of multi-objective optimization, and find the optimal balance between different objectives;

[0012] Use virtual reality to test the optimized model and printing parameters, and make adaptive corrections based on the simulation test results to adjust to the optimal printing parameters.

[0013] As a further solution of the present invention, when parsing geometric information in a model file, a 3D printing model file uploaded by a user is received through a user interface and the file is stored on a server; 3D model data is read according to the file format; wherein, when parsing geometric information, the process includes:

[0014] Extract the vertex, edge and face information of the model to obtain the triangular mesh structure;

[0015] Calculate the normal vector of each face and triangle to obtain the normal vector information;

[0016] The total area of ​​the model surface and the area of ​​the triangles are calculated to obtain the surface area of ​​the 3D printed model.

[0017] As a further solution of the present invention, the format of the 3D printing model file includes but is not limited to STL and OBJ formats. When calculating the normal vector information, the normal vector n of the triangle △ABC is calculated by vector intersection:

[0018] n=(AB×AC)

[0019] Among them, AB and AC are vectors from point A to B and C respectively; that is: AB=BA; AC=CA; among them, A, B, C are the coordinates of the three vertices of the triangle, A( , , ), B ( , , ), C ( , , ); × represents the vector cross product operation, and the result is a normal vector perpendicular to the plane △ABC.

[0020] As a further solution of the present invention, when the normal vector information is calculated, the calculation formula for the area S of the triangle △ABC is:

[0021]

[0022] in, is the area of ​​triangle △ABC; It represents the modulus of the vector cross product AB×AC, that is, the size of the cross product of the two side vectors of the triangle, and represents the area of ​​the parallelogram formed by the two side vectors on the plane of the triangle.

[0023] As a further solution of the present invention, key feature areas of surfaces, corners and edges in the model are identified. When identifying the surface in the model, the normal vector n of each point P on the surface is calculated using the curvature formula:

[0024]

[0025] Where, Curvature is used to describe the degree of curvature of a point on the surface. The greater the curvature, the more severe the surface curvature around the point. is the normal vector of the point on the actual surface; Represents the estimated value of the normal vector of the point, which is obtained by averaging the normal vectors of neighboring points; is the modulus of the difference between the two normal vectors, indicating the difference between the normal vector of the point and its estimated value; Represents the distance between adjacent vertices on the model, calculated using the Euclidean distance in three-dimensional space.

[0026] As a further solution of the present invention, the key feature area of ​​the corner in the model is identified. When identifying the corner, the normal vector of the point is used to calculate the angle, where:

[0027]

[0028] Where, It is the angle between two adjacent faces and is used to identify corner features in the model; The inverse cosine function is used to find the angle from the dot product of the normal vector. The result is the angle. and are the normal vectors of two adjacent triangular faces, representing the directions of the two faces in three-dimensional space; is the dot product of two normal vectors, which is used to calculate the angle between the two normal vectors. The calculation formula for the dot product value is:

[0029]

[0030] Where, is the angle between the two vectors.

[0031] As a further solution of the present invention, the optimization printing target defined according to the target model summary file includes the strength of the model, material consumption and printing time; wherein the objective function of the optimization target setting is:

[0032]

[0033] Where, 、 and is the target weight, 、 、 They are the strength of the model, material consumption and printing time.

[0034] As a further solution of the present invention, when generating an optimization model based on a generative adversarial network, a group of high-quality model samples, including models before and after optimization, are collected as training data; a generator network and a discriminator network are established, wherein the input of the generator is the optimization target parameter, and the output is the generated 3D model; the input of the discriminator is the comparison between the generated model and the real model, and the output is a score of the model quality; the generator loss and the discriminator loss are calculated respectively through the loss function; adversarial training is performed with the real model and the generated model, and the trained discriminator is used to evaluate the generated model, and a quality score is output, and a comprehensive evaluation is performed using the quality score of the model and the result that meets the objective function.

[0035] As a further solution of the present invention, when analyzing the results of multi-objective optimization and finding the optimal balance point between different objectives, Pareto optimization is used to find the optimal balance point based on multiple optimization objectives of the determined model strength, material consumption and printing time; the trade-off relationship between different objectives is determined by drawing the Pareto front of the objective function, and the Pareto front is analyzed to reach the optimal balance point according to the objective weights.

[0036] As a further solution of the present invention, when using virtual reality to test the optimized model and printing parameters, Unity is selected as the VR platform, the optimized 3D model is imported into the VR environment, and the printing parameters such as printing speed, material type, and layer thickness are set for virtual simulation; the printing parameters are adjusted according to the test results, and the adjusted parameters are retested in the virtual environment. The testing and adjustment process is repeated until the optimal printing parameters are found.

[0037] The present invention also includes a computer device comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the intelligent light-curing 3D printing model optimization method.

[0038] The present invention also includes a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the intelligent light-curing 3D printing model optimization method.

[0039] Compared with the existing technology, the intelligent light-curing 3D printing model optimization method provided by the present invention has the following beneficial effects:

[0040] 1. Improved printed model quality, accuracy, and stability. This invention precisely analyzes the model's geometric information (such as triangular mesh structure, normal vectors, and surface area) to ensure that every detail in the printing process is accurately captured and processed, reducing printing defects. By identifying and optimizing key feature areas in the model (such as curved surfaces, corners, and edges), it effectively improves the strength and stability of the printed model and prevents structural problems during the printing process.

[0041] 2. Reduced material consumption and shortened printing time. This invention uses a generative adversarial network (GAN) optimization model for multi-objective optimization, which can reduce material consumption while meeting strength requirements. For example, the system can identify and reduce unnecessary support structures, thereby saving material. The calculation of material consumption in the optimization objective function can accurately estimate and optimize material usage, reducing waste. Furthermore, by optimizing the model and printing parameters, the optimal printing speed, layer thickness, and other parameters can be selected to reduce printing time. Using a virtual reality platform to simulate and test the optimized model, printing time and results can be verified in a virtual environment, thus avoiding time waste in actual printing.

[0042] 3. Improved automation and intelligence of the optimization process. This invention can automatically parse user-uploaded 3D model files and extract geometric information, reducing manual operations and improving work efficiency. By leveraging a generative adversarial network (GAN) and the Pareto optimization algorithm, it can intelligently find the optimal balance point for model optimization and automatically adjust the model to meet multiple optimization objectives (such as strength, material consumption, and printing time).

[0043] 4. Optimize design flexibility to achieve multi-objective balance and personalized adjustment. The Pareto optimization method allows users to make trade-offs between strength, material consumption, and printing time to ensure the optimal balance between different objectives. For example, the weights of the optimization objectives can be adjusted according to actual needs to adapt to different application scenarios. According to the user's specific needs and printing tasks, the system can make adaptive adjustments to ensure that the final printing effect meets expectations. By testing printing parameters through virtual reality (VR) simulation, the system can make optimization adjustments in a virtual environment, reducing the number and cost of trials in actual printing. The discriminator network in the generative adversarial network can automatically evaluate the quality of the optimization model, reducing dependence on manual evaluation.

[0044] 5. Improved the printing adaptability and printing efficiency of the model. By adaptively correcting the model, the present invention enables the system to adjust the printing parameters to adapt to different printing conditions and requirements, thereby improving the adaptability of the model in different printing environments. Virtual simulation and adjustment using the VR platform can determine the optimal printing settings before actual printing to ensure that the printing effect of the model meets the requirements. Through automated geometric information analysis and optimized model generation, the model design and printing preparation process is accelerated, and overall work efficiency is improved. By accurately calculating normal vectors, curvatures, and corners, errors in model design and printing are reduced, and printing accuracy and consistency are improved.

[0045] In summary, the intelligent photo-curing 3D printing model optimization method of the present invention significantly improves the quality, efficiency and material utilization of printed models through automated geometric information analysis, curvature and corner calculation, generative adversarial network optimization, Pareto optimization and virtual reality testing. It not only improves the accuracy and strength of the model, but also reduces material waste and printing time, while enhancing the intelligence and automation level of the system, ultimately bringing a more efficient and economical 3D printing solution.

[0046] These and other aspects of the present invention will become more readily apparent in the following description of the embodiments. It should be understood that the above general description and the following detailed description are merely exemplary and explanatory and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or related technologies, the following briefly introduces the drawings required for the exemplary embodiments or related technical descriptions. The drawings are used to provide a further understanding of the present invention and constitute part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the drawings:

[0048] Figure 1 Flowchart of the intelligent light-curing 3D printing model optimization method according to an embodiment of the present invention.

[0049] Figure 2 This is a flow chart of parsing geometric information in the intelligent light-curing 3D printing model optimization method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0051] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., and do not represent the order of precedence, nor do they limit "first" and "second" to be different types.

[0052] The following will clearly and completely describe the technical solutions in the exemplary embodiments of the present invention in conjunction with the accompanying drawings of the exemplary embodiments of the present invention. Obviously, the exemplary embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0053] Due to the complexity of the model and changes in the printing environment, a single parameter setting cannot meet the needs of all models, resulting in unstable printing quality, and often problems such as model surface defects and insufficient structural strength. Due to the lack of intelligent optimization methods, the traditional printing process often results in material waste. The support structure and filling method in the model are usually set manually by the user. These settings may lead to unreasonable use of materials, thereby increasing production costs and time. The present invention provides an intelligent light-curing 3D printing model optimization method, which automatically analyzes the geometric features and material properties of the 3D model, identifies potential weaknesses in the model, and intelligently optimizes the 3D printing model.

[0054] The technical solution of the present invention is further described below with reference to specific embodiments:

[0055] See Figure 1 As shown, Figure 1 A flow chart of an intelligent light-curing 3D printing model optimization method provided by the present invention. An intelligent light-curing 3D printing model optimization method provided in one embodiment of the present invention includes the following steps:

[0056] Step S10: Obtain the 3D printing model file uploaded by the user.

[0057] Step S20: parse the geometric information in the model file to extract the triangle mesh structure, normal vector information and surface area.

[0058] In this step, when parsing the geometric information in the model file, the 3D printing model file uploaded by the user is received through the user interface and the file is stored on the server; the 3D model data is read according to the file format; wherein, see Figure 2 As shown, when parsing geometric information, it includes:

[0059] Step S201: extracting vertex, edge, and face information of the model to obtain a triangular mesh structure;

[0060] Step S202: Calculate the normal vector of each face and triangle to obtain normal vector information;

[0061] Step S203: Calculate the total area of ​​the model surface and the area of ​​the triangles to obtain the surface area of ​​the 3D printed model.

[0062] In this embodiment, the format of the 3D printing model file includes but is not limited to STL and OBJ formats. When calculating the normal vector information, the normal vector n of the triangle △ABC is calculated by vector intersection:

[0063] n=(AB×AC)

[0064] Among them, AB and AC are vectors from point A to B and C respectively; that is: AB=BA; AC=CA; among them, A, B, C are the coordinates of the three vertices of the triangle, A( , , ), B ( , , ), C ( , , ); × represents the vector cross product operation, and the result is a normal vector perpendicular to the plane △ABC.

[0065] Among them, when the normal vector information is calculated, the calculation formula for the area S of the triangle △ABC is:

[0066]

[0067] in, is the area of ​​triangle △ABC; It represents the modulus of the vector cross product AB×AC, that is, the size of the cross product of the two side vectors of the triangle, and represents the triangle formed by the two side vectors on the plane.

[0068] Among them, the key feature areas of the surface, corners and edges in the model are identified. When identifying the surface in the model, the normal vector n of each point P on the surface is calculated using the curvature formula:

[0069]

[0070] Where, Curvature is used to describe the degree of curvature of a point on the surface. The greater the curvature, the more severe the surface curvature around the point. is the normal vector of the point on the actual surface; Represents the estimated value of the normal vector of the point, which is obtained by averaging the normal vectors of neighboring points; is the modulus of the difference between the two normal vectors, indicating the difference between the normal vector of the point and its estimated value; Represents the distance between adjacent vertices on the model, calculated using the Euclidean distance in three-dimensional space.

[0071] Step S30: Identify key feature areas of surfaces, corners, and edges in the model, collect printing parameter data, and generate a target model summary file.

[0072] In this step, the key feature areas of the corners in the model are identified. When identifying corners, the angle is calculated using the normal vector of the point, where:

[0073]

[0074] Where, It is the angle between two adjacent faces and is used to identify corner features in the model; The inverse cosine function is used to find the angle from the dot product of the normal vector. The result is the angle. and are the normal vectors of two adjacent triangular faces, representing the directions of the two faces in three-dimensional space; is the dot product of two normal vectors, which is used to calculate the angle between the two normal vectors. The calculation formula for the dot product value is:

[0075]

[0076] Where, is the angle between the two vectors.

[0077] The optimization printing targets defined in the target model summary file include the strength, material consumption, and printing time of the model; the objective function of the optimization target setting is:

[0078]

[0079] Where, 、 and is the target weight, 、 、 They are the strength of the model, material consumption and printing time.

[0080] For example, when identifying key feature areas of surfaces, corners, and edges in a model, if a sharp edge in the model is identified as an area requiring special treatment, the printing parameter record is set to a light curing time of 10 seconds, a light source intensity of 80%, and a speed of 20 mm / s. The summary file content includes model data, feature areas, and parameter recommendations, forming a JSON file for further processing, which can ensure the logical correctness and completeness of the 3D printing model optimization method.

[0081] During calculation, if you have the following triangle vertex coordinates:

[0082] A(1,0,0), B(0,1,0), C(0,0,1);

[0083] Normal vectors: AB=(-1,1,0) and AC=(-1,0,1); n=AB×AC=(1,1,1).

[0084] The area of ​​the triangle is: .

[0085] When identifying corners, if the normal vectors of two adjacent faces are (1,0,0) and (0,1,0), the angle between them is:

[0086]

[0087] Step S40: Define the optimized printing target according to the target model summary file, generate the optimization model based on the generative adversarial network, evaluate the model quality through the discriminator network, analyze the results of the multi-objective optimization, and find the optimal balance point between different targets.

[0088] In this step, when generating an optimized model based on a generative adversarial network, a set of high-quality model samples, including models before and after optimization, are collected as training data; a generator network and a discriminator network are established, wherein the input of the generator is the optimization target parameter, and the output is the generated 3D model; the input of the discriminator is the comparison between the generated model and the real model, and the output is a score of the model quality; the generator loss and the discriminator loss are calculated respectively through the loss function; adversarial training is performed with the real model and the generated model, and the trained discriminator is used to evaluate the generated model, and a quality score is output. The quality score of the model and the result that meets the objective function are used for comprehensive evaluation.

[0089] In this embodiment, when analyzing the results of multi-objective optimization and finding the optimal balance point between different objectives, Pareto optimization is used to find the optimal balance point based on multiple optimization objectives of the determined model strength, material consumption and printing time; the trade-off relationship between different objectives is determined by drawing the Pareto front of the objective function, and the Pareto front is analyzed to reach the optimal balance point according to the objective weights.

[0090] Step S50: Use virtual reality to test the optimized model and printing parameters, and make adaptive corrections based on the simulation test results to adjust to the optimal printing parameters.

[0091] In this step, when using virtual reality to test the optimized model and printing parameters, select Unity as the VR platform, import the optimized 3D model into the VR environment, set the printing parameters such as printing speed, material type, and layer thickness for virtual simulation; adjust the printing parameters based on the test results, retest the adjusted parameters in the virtual environment, and repeat the testing and adjustment process until the optimal printing parameters are found.

[0092] Suppose you want to optimize a model of a printed part and test its performance in a VR environment:

[0093] (1) Model import: Import the optimized part model (such as the bracket) into Unity.

[0094] (2) Set printing parameters: Configure the virtual printer to print at a speed of 50 mm / s, the material type to PLA, and the layer thickness to 0.2 mm.

[0095] (3) Simulate printing: Simulate the printing process in Unity to observe whether there is any inter-layer misalignment or support structure failure.

[0096] (4) Performance testing: Using virtual mechanics tools to evaluate the strength of the bracket, it was found that some parts may not be strong enough.

[0097] (5) Parameter adjustment: Reduce the printing speed to 40 mm / s, increase the density of the support structure, and perform virtual testing again.

[0098] (6) Optimization loop: Repeat the above steps until the optimal printing parameters are obtained.

[0099] Through the above steps, the printing model and parameters can be efficiently tested and optimized in a virtual environment, reducing the trial and error costs in actual printing.

[0100] In this invention, the intelligent stereolithography 3D printing model optimization method precisely analyzes the model's geometric information (such as triangular mesh structure, normal vectors, and surface area) to ensure that every detail in the printing process is accurately captured and processed, thereby reducing printing defects. By identifying and optimizing key feature areas in the model (such as curved surfaces, corners, and edges), the strength and stability of the printed model can be effectively improved, preventing structural problems during the printing process. Multi-objective optimization using a generative adversarial network (GAN) optimization model can reduce material consumption while meeting strength requirements. For example, the system can identify and reduce unnecessary support structures, thereby saving material. The calculation of material consumption in the optimization objective function accurately estimates and optimizes material usage, reducing waste. Furthermore, by optimizing the model and printing parameters, the optimal printing speed, layer thickness, and other parameters can be selected to reduce printing time. Simulation testing of the optimized model using a virtual reality platform allows for verification of printing time and results in a virtual environment, thus avoiding time wasted in actual printing.

[0101] The present invention can automatically parse 3D model files uploaded by users, automatically extract geometric information, reduce manual operations, and improve work efficiency. Through the generative adversarial network (GAN) and the Pareto optimization algorithm, it can intelligently find the optimal balance point for model optimization and automatically adjust the model to meet multiple optimization goals (such as strength, material consumption, and printing time). The Pareto optimization method allows users to make trade-offs between strength, material consumption, and printing time to ensure the optimal balance between different goals. For example, the weights of the optimization goals can be adjusted according to actual needs to adapt to different application scenarios. According to the user's specific needs and printing tasks, the system can make adaptive adjustments to ensure that the final printing effect meets expectations. By simulating and testing printing parameters through virtual reality (VR), the system can make optimization adjustments in a virtual environment, reducing the number and cost of trials in actual printing. The discriminator network in the generative adversarial network can automatically evaluate the quality of the optimized model, reducing dependence on manual evaluation.

[0102] By adaptively modifying the model, the system can adjust printing parameters to suit different printing conditions and requirements, thereby improving the model's adaptability in different printing environments. Virtual simulation and adjustment using a VR platform can determine the optimal printing settings before actual printing, ensuring that the model's printing effect meets the requirements. Automated geometric information analysis and optimized model generation accelerate the model design and printing preparation process, improving overall work efficiency. Accurately calculating normal vectors, curvature, and corners reduces errors in model design and printing, improving printing accuracy and consistency.

[0103] In summary, the intelligent photo-curing 3D printing model optimization method of the present invention significantly improves the quality, efficiency and material utilization of printed models through automated geometric information analysis, curvature and corner calculation, generative adversarial network optimization, Pareto optimization and virtual reality testing. It not only improves the accuracy and strength of the model, but also reduces material waste and printing time, while enhancing the intelligence and automation level of the system, ultimately bringing a more efficient and economical 3D printing solution.

[0104] It should be understood that, although the above is described in a certain order, these steps are not necessarily performed in sequence according to the above order. Unless there is clear explanation in this article, the execution of these steps does not have strict order restriction, and these steps can be performed in other orders. Moreover, a part of the steps of the present embodiment may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be performed in turn or alternately with at least a portion of the steps or stages in other steps or other steps.

[0105] In one embodiment, a computer device is further provided in an embodiment of the present invention, comprising at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor performs the steps of the intelligent light-curing 3D printing model optimization method.

[0106] In one embodiment, the present invention further provides a computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to execute the steps of the intelligent light-curing 3D printing model optimization method.

[0107] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program represented by computer instructions. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above-mentioned methods. In addition, any reference to memory, storage, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory.

[0108] Non-volatile memory may include read-only memory, magnetic tape, floppy disk, flash memory, or optical storage. Volatile memory may include random access memory or external cache memory. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory or dynamic random access memory.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent light-curing 3D printing model optimization method, characterized in that: The method comprises the following steps: Get the 3D printing model file uploaded by the user; Analyze the geometric information in the model file and extract the triangle mesh structure, normal vector information and surface area; Identify key feature areas of surfaces, corners, and edges in the model, collect printing parameter data, and generate target model summary files; Define the optimized printing target according to the target model summary file, generate the optimization model based on the generative adversarial network, evaluate the model quality through the discriminator network, analyze the results of multi-objective optimization, and find the optimal balance between different objectives; Use virtual reality to test the optimized model and printing parameters, and make adaptive corrections based on the simulation test results to adjust to the optimal printing parameters; When generating an optimization model based on a generative adversarial network, a set of high-quality model samples, including models before and after optimization, are collected as training data; a generator network and a discriminator network are established, where the input of the generator is the optimization target parameter and the output is the generated 3D model; the input of the discriminator is the comparison between the generated model and the real model, and the output is a score of the model quality; the generator loss and the discriminator loss are calculated respectively through the loss function; adversarial training is performed with the real model and the generated model, and the trained discriminator is used to evaluate the generated model, outputting a quality score, and using the model's quality score and the result that meets the objective function for comprehensive evaluation.

2. The intelligent light-curing 3D printing model optimization method according to claim 1, characterized in that: When parsing the geometric information in the model file, the 3D printing model file uploaded by the user is received through the user interface and the file is stored on the server; the 3D model data is read according to the file format; wherein, when parsing the geometric information, it includes: Extract the vertex, edge and face information of the model to obtain the triangular mesh structure; Calculate the normal vector of each face and triangle to obtain the normal vector information; The total area of ​​the model surface and the area of ​​the triangles are calculated to obtain the surface area of ​​the 3D printed model.

3. The intelligent light-curing 3D printing model optimization method according to claim 2, characterized in that: The formats of the 3D printing model file include STL and OBJ formats. When calculating the normal vector information, the normal vector n of the triangle △ABC is calculated by vector intersection: n=(AB×AC) Among them, AB and AC are vectors from point A to B and C respectively; that is: AB=BA; AC=CA; among them, A, B, C are the coordinates of the three vertices of the triangle, A( , , ), B ( , , ), C ( , , ); × represents the vector cross product operation, and the result is a normal vector perpendicular to the plane △ABC.

4. The intelligent light-curing 3D printing model optimization method according to claim 3, characterized in that: When the normal vector information is calculated, the calculation formula for the area S of triangle △ABC is: in, is the area of ​​triangle △ABC; It represents the modulus of the vector cross product AB×AC, that is, the size of the cross product of the two side vectors of the triangle, and represents the area of ​​the parallelogram formed by the two side vectors on the plane of the triangle.

5. The intelligent light-curing 3D printing model optimization method according to claim 4, characterized in that: Identify key feature areas of surfaces, corners, and edges in the model. When identifying surfaces in the model, the normal vector n of each point P on the surface is calculated using the curvature formula: Where, Curvature is used to describe the degree of curvature of a point on the surface. The greater the curvature, the more severe the surface curvature around the point. is the normal vector of the point on the actual surface; Represents the estimated value of the normal vector of the point, which is obtained by averaging the normal vectors of neighboring points; is the modulus of the difference between the two normal vectors, indicating the difference between the normal vector of the point and its estimated value; Represents the distance between adjacent vertices on the model, calculated using the Euclidean distance in three-dimensional space.

6. The intelligent light-curing 3D printing model optimization method according to claim 5, characterized in that: Identify the key feature areas of the corners in the model. When identifying corners, use the normal vector of the point to calculate the angle, where: Where, It is the angle between two adjacent faces and is used to identify corner features in the model; The inverse cosine function is used to find the angle from the dot product of the normal vector. The result is the angle. and are the normal vectors of two adjacent triangular faces, representing the directions of the two faces in three-dimensional space; is the dot product of two normal vectors, which is used to calculate the angle between the two normal vectors. The calculation formula for the dot product value is: Where, is the angle between the two vectors.

7. The intelligent light-curing 3D printing model optimization method according to claim 6, characterized in that: The optimization printing targets defined in the target model summary file include the model's strength, material consumption, and printing time. The objective function for the optimization target setting is: Where, 、 and is the target weight, 、 、 They are the strength of the model, material consumption and printing time.

8. The intelligent light-curing 3D printing model optimization method according to claim 7, characterized in that: When analyzing the results of multi-objective optimization and finding the optimal balance point between different objectives, Pareto optimization is used to find the optimal balance point based on the multiple optimization objectives of model strength, material consumption, and printing time. The trade-off relationship between different objectives is determined by plotting the Pareto front of the objective function, and the Pareto front is analyzed to reach the optimal balance point based on the objective weights.

9. The intelligent light-curing 3D printing model optimization method according to claim 8, characterized in that: When using virtual reality to test the optimized model and printing parameters, select Unity as the VR platform, import the optimized 3D model into the VR environment, set the printing parameters such as printing speed, material type, and layer thickness for virtual simulation; adjust the printing parameters based on the test results, retest the adjusted parameters in the virtual environment, and repeat the testing and adjustment process until the optimal printing parameters are found.

Images