Method and System for Collecting and Analyzing 3D Printing Tooling Fixture Data Based on Big Data

Abstract

The present invention provides a method and system for collecting and analyzing 3D printing tooling fixture data based on big data, which relates to the fields of 3D printing and intelligent manufacturing technology. It includes obtaining fixture historical design schemes, processing parameters, and sensor data, and performing feature extraction and data cleaning to form structured data. Then, based on the BERT-CNN model, the design intent is parsed, and a fixture applicability classification database is constructed in combination with a classification model. Subsequently, a lightweight fixture structure is generated using a generative adversarial network and topology optimization, and its topology structure is optimized through an optimization model. Next, simulation analysis is carried out to evaluate the force, deformation, and thermal stress, and the error is optimized based on the simulation parameters to obtain the optimized fixture design data. Finally, the fixture is manufactured and the force data is collected in real time, and the processing process parameters are optimized through dynamic analysis. The present invention optimizes the fixture design and manufacturing through data driving, improving the design efficiency and fixture performance.

Description

Technical Field

[0001] The present invention relates to the technical fields of 3D printing and intelligent manufacturing. Specifically, it relates to a method and system for collecting and analyzing 3D printing tooling fixture data based on big data. Background Art

[0002] With the continuous pursuit of precision and efficiency in the industrial manufacturing field, the application of 3D printing technology in the manufacturing of tooling fixtures has gradually become a trend. Traditional methods for designing and manufacturing tooling fixtures rely on manual experience and fixed design rules. This method often fails to provide flexible and efficient solutions when faced with complex processing technologies and changing production requirements. Existing fixture design methods usually cannot fully utilize big data and advanced algorithm technologies for optimization, thus easily leading to data redundancy, low design efficiency, and insufficient fixture applicability during the design process.

[0003] Therefore, there is an urgent need for a method and system for collecting and analyzing 3D printing tooling fixture data based on big data to solve the problems existing in the prior art, such as incomplete data processing, lack of dynamic optimization, and insufficient applicability classification. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for collecting and analyzing 3D printing tooling fixture data based on big data to improve the above problems. To achieve the above purpose, the technical solutions adopted by the present invention are as follows:

[0005] In a first aspect, the present application provides a method for collecting and analyzing 3D printing tooling fixture data based on big data, including: obtaining first information, where the first information includes fixture historical design plans, processing process parameters, and sensor data;

[0006] Performing feature extraction and data cleaning on the first information to obtain second information, where the second information includes the first information after data cleaning;

[0007] Analyzing the design intent of the second information based on a BERT-CNN dual-channel and classifying the fixture historical design plans in combination with a preset classification model to obtain a fixture applicability classification database;

[0008] According to the fixture applicability classification database, generating a lightweight fixture structure through a generative adversarial network and topology optimization, and optimizing the lightweight fixture structure through a preset optimization model to obtain an optimized fixture topology structure;

[0009] Performing simulation on the optimized fixture topology structure and optimizing errors based on the simulation parameters to obtain error-optimized fixture design data;

[0010] Manufacture the fixture according to the fixture design data optimized based on the error, and collect the fixture force data for dynamic optimization to obtain the optimized fixture processing process parameters.

[0011] In a second aspect, the present application also provides a big data-based acquisition and analysis system for 3D printing tooling fixture materials, including:

[0012] An acquisition unit for acquiring first information, where the first information includes fixture historical design schemes, processing process parameters, and sensor data;

[0013] A cleaning unit for performing feature extraction and data cleaning on the first information to obtain second information, where the second information includes the first information after data cleaning;

[0014] A classification unit for analyzing the design intention of the second information based on BERT-CNN dual channels and classifying the fixture historical design schemes in combination with a preset classification model to obtain a fixture applicability classification database;

[0015] An optimization unit for generating a lightweight fixture structure through a generative adversarial network and topology optimization according to the fixture applicability classification database, and optimizing the lightweight fixture structure through a preset optimization model to obtain an optimized fixture topology structure; A simulation unit for performing simulation on the optimized fixture topology structure and performing error optimization based on the simulation parameters to obtain error-optimized fixture design data;

[0016] An analysis unit for manufacturing the fixture according to the error-optimized fixture design data and collecting the fixture force data for dynamic optimization to obtain the optimized fixture processing process parameters.

[0017] The beneficial effects of the present invention are:

[0018] Through innovative algorithm combinations and data processing methods, the present invention comprehensively improves the efficiency and adaptability of fixture design and manufacturing. The method obtains first information including fixture historical design schemes, machining process parameters, and sensor data, and optimizes this information using various data processing techniques (such as feature extraction, data cleaning, dimensionality reduction, clustering, etc.). Subsequently, based on BERT-CNN dual-channel analysis of the design intent, and combined with a preset classification model, the fixture applicability is classified. By combining a generative adversarial network with topology optimization, a lightweight fixture structure is generated, and its topology is optimized and simulated to further optimize the error, ultimately achieving the dynamic optimization and efficient manufacturing of the fixture. In addition, based on the dynamically collected force data, time series analysis and multi-objective optimization methods are used to achieve real-time optimization of the fixture machining process parameters. The present invention provides a complete closed-loop system from design to manufacturing and then to optimization, significantly improving the accuracy and production efficiency of fixture design, overcoming the limitations of traditional methods, and possessing high innovation and practicality.

[0019] Other features and advantages of the present invention will be described in the subsequent specification, and, in part, will be obvious from the specification, or can be understood by implementing the embodiments of the present invention. The objectives and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the written specification, claims, and drawings. Brief Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0021] Figure 1 It is a schematic flow chart of the method for collecting and analyzing 3D printing tooling fixture data based on big data described in the embodiments of the present invention;

[0022] Figure 2 It is a schematic structural diagram of the system for collecting and analyzing 3D printing tooling fixture data based on big data described in the embodiments of the present invention.

[0023] In the figure: 701, acquisition unit; 702, cleaning unit; 703, classification unit; 704, optimization unit; 705, simulation unit; 706, analysis unit. Detailed Embodiments

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. The components of the embodiments of the present invention usually described and illustrated in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0025] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive distinction and cannot be understood as indicating or implying relative importance.

[0026] Embodiment 1:

[0027] This embodiment provides a method for collecting and analyzing data of 3D printing tooling jigs based on big data.

[0028] See Figure 1 , which shows that this method includes step S1, step S2, step S3, step S4, step S5, and step S6.

[0029] Step S1: Obtain first information, where the first information includes fixture historical design schemes, processing process parameters, and sensor data;

[0030] It can be understood that in this step, the fixture historical design schemes include previous design examples, manufacturing processes, and applicable engineering parameters. These data can provide valuable references for the current design. The processing process parameters cover various parameters involved in the manufacturing process of the fixture, such as cutting speed, tool selection, coolant usage, etc. These are key factors affecting the quality and accuracy of the fixture. The sensor data comes from various sensors installed on the fixture during use, collecting real-time data such as the force, deformation, vibration, and temperature of the fixture. These data can reflect the dynamic behavior of the fixture during operation and provide true and objective information for subsequent performance analysis and optimization.

[0031] Step S2: Perform feature extraction and data cleaning on the first information to obtain second information, where the second information includes the first information after data cleaning;

[0032] It can be understood that in this step, while cleaning the data, not only the noise is effectively eliminated, but also the effective information in the data is enhanced, and more accurate and representative features are extracted. The processed data is more usable and can help the subsequent algorithms provide higher accuracy and robustness in optimizing the fixture design and manufacturing process. In this step, step S2 includes step S21, step S22, step S23, and step S24.

[0033] Step S21: According to the first information, perform high-dimensional feature dimensionality reduction and abnormal data detection processing. Specifically, use a variational autoencoder to perform high-dimensional feature embedding on the first information, and combine an isolation forest to identify abnormal data in the first information. At the same time, use the DBSCAN clustering algorithm to analyze the distribution pattern of the abnormal data, remove the isolated points and data in the low-density area, and obtain a denoised feature data set.

[0034] It can be understood that in this step, first, the original features are extracted from the first information. However, the dimensions of these features are usually very high and contain redundant information. Directly using this data will lead to too high computational complexity and may interfere with subsequent analysis and optimization. To reduce this complexity, a variational autoencoder is used for high-dimensional feature embedding. A variational autoencoder is a deep learning model that maps data to a latent space through a neural network and reduces the dimensions through a reconstruction process. The variational autoencoder will compress and embed the input high-dimensional data into the latent space while preserving the important features of the data for subsequent processing.

[0035] For the embedded feature data, this step performs abnormal data detection. Among them, there are some outliers in the high-dimensional data. If these outliers are not removed, they will affect the accuracy of subsequent analysis. Therefore, an isolation forest algorithm is used for processing, which is specifically used to handle abnormal data. The working principle of the isolation forest is to randomly divide the data space and recursively divide out the isolated points. Abnormal data is easy to be identified because it is relatively isolated. It can then efficiently process a large number of features and quickly identify the isolated points without being affected by the data dimensions.

[0036] However, there will still be some noise data or low-density areas after abnormal detection. These data may be wrongly considered as abnormal data and removed. In fact, they are just a normal distribution pattern in the data. Therefore, this step also uses the DBSCAN clustering algorithm to analyze the data, identify the low-density areas and isolated points in the data, and remove them. The DBSCAN clustering algorithm can judge which data belongs to the "group" and which data is distributed on the "edge" or is isolated according to the density characteristics of the data.

[0037] After high-dimensional feature dimensionality reduction, abnormal data detection, and clustering analysis, the denoised feature dataset can more accurately reflect the key factors in the fixture design and manufacturing process, and the data quality is significantly improved.

[0038] Step S22: According to the denoised feature dataset, perform multi-scale feature extraction and alignment processing. Specifically, extract features at different time scales in the fixture design and manufacturing process through wavelet transform to obtain the dataset after feature extraction.

[0039] It can be understood that the input dataset in this step is the denoised feature dataset, which contains various types of feature data at different time points in the fixture design and manufacturing process. These data include information such as the mechanical parameters of the fixture, temperature changes, and machining accuracy, which usually appear as time series signals. Then, select the Daubechies wavelet to perform multi-level decomposition on the original signal to achieve multi-scale analysis, decomposing the signal into a series of low-frequency (approximation) and high-frequency (detail) parts. Specifically, use the Daubechies wavelet to convolve the signal and calculate the projection of the signal at different time scales. Then, perform downsampling operations (in this step, the data length is reduced by half each time of decomposition) so that the low-frequency part obtained at each layer represents the larger-scale changes of the signal, while the high-frequency part retains the smaller-scale changes.

[0040] During the decomposition process, each layer of the signal represents features at different time scales. For example, the low-frequency part obtained from the first-layer decomposition corresponds to the large-scale changes of the signal (such as the gradual wear of the fixture over a long period), while the high-frequency part corresponds to the sudden changes in the signal (such as the instantaneous failure of the fixture). By continuing the decomposition, more levels of signal details are obtained. The decomposition effect of each layer is demonstrated by different features of the signal in the time scale.

[0041] Step S23: According to the dataset after feature extraction, perform local consistency enhancement and structure preservation processing. Specifically, use Laplacian eigenmaps to preserve the local structure of the fixture geometric features, and combine t-SNE to optimize the data distribution in the low-dimensional space to obtain the feature data with optimized structure.

[0042] It can be understood that this step takes the dataset after wavelet transform and feature extraction as the input data, which contains key feature information in the fixture design and manufacturing process (such as multi-scale features of time series data such as temperature, pressure, and displacement). Then, construct an adjacency matrix based on the input data, where each data point is connected to its most similar neighbor points in the feature space. The elements of the adjacency matrix represent the similarity between data points, which is represented by a weighted graph in this step, and the weights are calculated by the Gaussian kernel function.

[0043] Using the adjacency matrix, calculate the Laplacian matrix of the graph. The Laplacian matrix is used to describe the local structural information of each node in the graph and can help reveal the local geometric structure in the dataset. By performing eigenvalue decomposition on this matrix, eigenvectors are obtained, and these eigenvectors capture the geometric relationships in the local neighborhoods of the data. And through eigenvalue decomposition, the eigen data of low-dimensional embedding is obtained. Laplacian eigenmaps can more effectively reflect the local structural changes in the data by preserving local similarity rather than global variance.

[0044] The dataset processed by Laplacian eigenmaps will be used as input and fed into the t-SNE algorithm for processing. By calculating the similarity between high-dimensional data points, the probability distribution is used to describe the degree of similarity. Among them, for each pair of data points, t-SNE first calculates their similarity in the high-dimensional space, then calculates their similarity in the low-dimensional space, and tries to make the similarity in the low-dimensional space as consistent as possible with that in the high-dimensional space. The optimization process of t-SNE is achieved by iteratively minimizing the Kullback-Leibler divergence (KL divergence), that is, making the distribution of data points in the low-dimensional space reflect the relative relationships of data points in the high-dimensional space as much as possible. Through this optimization, the relative distances between data points can be retained, and the group structure and cluster structure in the original data will be effectively presented in the low-dimensional space. And finally, the eigen data with optimized structure is obtained.

[0045] Step S24: According to the eigen data with optimized structure, perform noise suppression and data smoothing processing to obtain the second information.

[0046] It can be understood that there will be some abnormal fluctuations or high-frequency noises in the eigen data with optimized structure in this step. This step uses the physical model of the fixture processing process (such as the dynamic change law of fixture displacement). Kalman filtering can predict the state of the fixture at a certain moment and update the prediction value by fusing the measurement data of the sensor, thereby effectively suppressing the noise in the sensor data. For example, if the data of the temperature sensor fluctuates instantaneously due to external environmental changes, Kalman filtering can smooth these instantaneous fluctuations and eliminate the high-frequency noises in the data. After Kalman filtering suppresses most of the noises, this step performs data smoothing processing on the data through locally weighted regression to obtain a more smooth and accurate eigen dataset.

[0047] Step S3: Analyze the design intention of the second information based on the BERT-CNN dual-channel and classify the historical design schemes of the fixture in combination with the preset classification model to obtain the fixture applicability classification database;

[0048] It is understandable that in this step, a dual-channel analysis method of BERT and CNN is used. The BERT-CNN dual-channel analysis means that BERT is first responsible for extracting rich semantic representations from the input text, and then CNN further extracts features and optimizes the spatial structure of these representations, so as to achieve a more efficient feature learning and classification task model. This step can comprehensively consider the text semantics and structural features in the fixture design scheme, and comprehensively and accurately judge the applicability of each fixture design. This method not only improves the accuracy of classification, but also can efficiently handle the complexity of the fixture design scheme, ensuring that the most suitable design scheme can be quickly found during the fixture selection and optimization process, improving production efficiency and reducing resource waste. In this step, step S3 includes step S31, step S32 and step S33.

[0049] Step S31: According to the second information after data cleaning, perform text and semantic understanding processing. Among them, the fixture historical design scheme in the second information is encoded by the BERT model to extract the context information and potential semantic features of the fixture design, and obtain the design features after semantic encoding.

[0050] It is understandable that in this step, the input data is the second information after feature extraction, denoising and data smoothing processing. This information includes the text description of the fixture design scheme, the technical parameters of the processing process, the processing records of the historical design scheme, etc. In this process, the input data is mainly text information (such as the description of the fixture design scheme), and these text information contains important semantic information such as design intent, usage environment, and technical requirements.

[0051] In order to enable the BERT model to efficiently process the input text, the text data is first preprocessed, including removing stop words, word segmentation, and normalization processing, etc. These processing steps can reduce the computational burden of the model and improve the accuracy of semantic understanding at the same time.

[0052] The BERT model receives this text data and captures the context information in the text through its bidirectional encoding ability. For example, BERT can identify the relationship between "precision milling" and "high-precision clamping", and then understand the core requirements of the fixture design scheme. Through the processing of the BERT model, each word and sentence in the fixture design scheme is transformed into a high-dimensional semantic vector, and BERT can extract potential semantic features such as fixture function, applicable scenarios, and design limitations from it.

[0053] After the encoding process of the BERT model, the obtained design features are the deep semantic representations of the fixture design scheme, and these features contain information such as the functional requirements, potential constraint conditions, and applicable scenarios of the fixture design.

[0054] Step S32: Perform image-based feature extraction processing based on the semantically encoded design features. Specifically, perform multi-level convolution and pooling operations on the semantically encoded design features through a CNN (Convolutional Neural Network). The multi-level convolution includes extracting spatial correlation features through a sliding window mechanism and fusing these features through a fully connected layer to obtain the visual features of the fixture's historical design solutions.

[0055] It can be understood that this step takes the semantically encoded design features as input, that is, the high-dimensional vector representations from the BERT model. These vectors encode the text information of the fixture design solutions and contain key features such as the functional description of the fixture, the processing environment, and design parameters.

[0056] In this step, the semantic features are arranged into a two-dimensional matrix according to a preset rule. The preset rule is to map different categories of semantic features to fixed rows / columns. Through the sliding window mechanism, perform convolution operations on the input semantic feature matrix to extract local spatial correlation features. For example: The first layer of convolution extracts low-level features (local associations between semantic units in the present invention). The second layer of convolution extracts intermediate-level features (syntactic structures and common design patterns in the present invention). The third layer of convolution extracts high-level features (complex semantic relationships in the present invention, such as the "matching of clamping force and stiffness"). The convolution kernels of each layer automatically learn the patterns of the data. For example: Small-sized convolution kernels (such as 3×3) can extract local semantic relationships, and large-sized convolution kernels (such as 5×5) can identify larger semantic structures, such as the design intent expressed by the entire sentence. After each layer of convolution, use a pooling layer to reduce the data dimension, improve computational efficiency, and retain key features. After multiple layers of convolution and pooling, flatten the extracted features and input them into a fully connected layer. The role of the fully connected layer is to fuse features at different levels to generate the final visual feature representation. For example, through the fully connected layer, the model can associate features such as "clamping force", "fixture material", and "processing accuracy" to form an overall fixture design feature representation. After processing by the CNN convolutional neural network, the data of the fixture design solutions can be represented in an image-based manner, making the matching of similar solutions more intuitive.

[0057] Step S33: Concatenate the semantically encoded design features with the visual features of the fixture's historical design solutions to form a multi-modal feature vector, and classify the fixture's historical design solutions in combination with a support vector machine to obtain the applicability classification result of the fixture's historical design solutions.

[0058] It can be understood that the semantically encoded design features in this step contain semantic information of the fixture design solutions, such as features like fixture type, design purpose, and usage environment. The visual features of the fixture's historical design solutions contain the spatial correlation features of the fixture solutions extracted through the convolutional neural network, such as clamping methods and processing error control.

[0059] In this step, different weights are assigned to the semantic features and visual features in a weighted splicing manner, and then the weighted features are spliced to obtain the spliced features;

[0060] Then, a support vector machine is used to train a dataset: containing a large number of labeled fixture design schemes, and annotating their applicability categories (such as "suitable for aluminum alloy processing", "suitable for precision machining", etc.). The optimal hyperplane is trained to maximize the classification interval of different categories of fixture schemes. Furthermore, a trained support vector machine model is obtained. The trained support vector machine model can classify newly input fixture design schemes and output their applicability categories. Using a support vector machine for classification enables the system to accurately determine the applicability of new fixture schemes based on historical fixture data. It greatly improves the automation ability of fixture design and provides strong support for efficient fixture scheme recommendation and optimization.

[0061] Step S4: According to the fixture applicability classification database, a lightweight fixture structure is generated through a generative adversarial network and topology optimization, and the lightweight fixture structure is optimized through a preset optimization model to obtain the optimized fixture topology structure; It can be understood that this step uses a deep learning generative adversarial network and topology optimization, combined with a preset optimization model, to generate a lightweight fixture structure that meets specific manufacturing requirements, improving the mechanical properties of the fixture while reducing material consumption. In this step, step S4 includes step S41, step S42, and step S43.

[0062] Step S41: According to the fixture applicability classification database, perform lightweight fixture structure generation processing. Among them, a candidate fixture structure is generated through a generative adversarial network. The generator network generates different fixture structure schemes according to the design parameters, material properties, and process requirements in the applicability classification database, while the discriminator network judges the authenticity of these generated fixture structures to obtain a fixture structure that meets the preset lightweight design requirements;

[0063] It can be understood that in this step, the fixture applicability classification database includes design parameters (fixture size, fixing method, load capacity), material properties (density, strength, toughness, corrosion resistance), and process requirements (whether it is suitable for 3D printing, CNC machining, etc.). After taking them as inputs, the design parameters, material properties, and process requirements are initially feature fused through a fully connected layer, and the three-dimensional shape of the fixture is constructed layer by layer using a convolutional neural network. The structural rationality is improved through skip connections to avoid information loss, and thus a forged fixture design scheme is obtained;

[0064] The authenticity of the forged fixture design scheme is evaluated through the discriminator, and then it is predicted whether the fixture structure meets the lightweight design requirements. Among them, adversarial loss is used for optimization, as shown below:

[0065]

[0066] Among them, L GAN represents the adversarial loss function, x represents the real fixture design scheme, and P data (x) represents the data sample of the real fixture design scheme, D(x) represents the discriminant output of the real data, z represents the random noise vector, and P z (z) represents the probability distribution of the noise variable, G(z) represents the forged fixture design scheme, D(G(z)) represents the discriminant result of the forged fixture design scheme, and E is the expectation operator, representing the calculation of the mathematical expectation of the data distribution.

[0067] Step S42: Perform topology optimization on the generated lightweight fixture structure. According to the preset loading conditions and constraints, dynamically adjust the shape and material distribution of the fixture structure to obtain the fixture structure after topology optimization;

[0068] It can be understood that in this step, topology optimization is a method based on mathematical optimization. Its goal is to find the optimal material distribution scheme under the premise of a given design space, load, boundary conditions, and manufacturing constraints, so that the fixture can achieve the maximum degree of lightweight while ensuring stiffness. In this step, the density method is used to optimize the material distribution. Initialize the design area of the fixture, discretize it into finite element meshes, define the material density variable and initialize it to 1 (i.e., the initial model is filled with materials). According to the loading conditions, calculate the stress and strain distributions of the fixture under external forces. Finally, iteratively remove the inefficient material areas to generate the optimized topology structure. The optimized topology structure includes the fixture geometry and material distribution scheme obtained after topology optimization.

[0069] The mathematical expression of topology optimization in this step is as follows:

[0070] Among them, ρ represents the material density distribution vector, C(ρ) represents the objective function, which is the deformation energy of the structure, N represents the total number of elements after the structure is discretized, and ρ e represents the material density of the e-th element, p represents the penalty factor, and c e represents the compliance contribution of the e-th element, V(ρ) represents the material volume of the current topology structure, and v e represents the volume of the e-th element, V max represents the maximum allowable material volume, and ρ min represents the minimum density value.

[0071] Step S43: Optimize the parameters of the fixture structure after topology optimization based on the particle swarm optimization algorithm to obtain the optimized fixture topology structure.

[0072] In this step, the optimization objectives and parameters are defined through the particle swarm optimization algorithm. The optimization objectives include minimizing the fixture mass (reducing material usage and cost), maximizing the fixture stiffness (enhancing structural strength and reducing deformation), optimizing the stress distribution (ensuring stable mechanical properties), and meeting manufacturing process constraints (applicable to 3D printing and CNC machining). The optimization parameters include the position and shape of the key support structures, the hole layout and size (reducing weight), the material distribution density (determining the balance between stiffness and weight), and the rule for changing the fixture structure thickness. Then, the particle swarm is initialized, and the finite element analysis is used to evaluate the deformation and stress distribution of the fixture under external forces, and the fitness function is calculated. The fitness function is as follows:

[0073]

[0074] where F(x) represents the X fitness value of the fixture design scheme, m(x) represents the fixture mass, m min represents the minimum value of the fixture mass, m max represents the maximum value of the fixture mass, k max represents the maximum value of the fixture stiffness, k(x) represents the fixture stiffness, k min represents the minimum value of the fixture stiffness, represents the maximum value of the variance of the fixture stress distribution, σ var (x) represents the variance of the fixture stress distribution, represents the minimum value of the variance of the fixture stress distribution. Among them, w1, w2, and w3 are all weight coefficients.

[0075] Furthermore, the particle velocity and position are iteratively updated through the fitness function to obtain the fixture topology structure with the minimum mass, the maximum stiffness, and the most uniform stress distribution.

[0076] Step S5: Perform simulation based on the optimized fixture topology structure, and optimize the error based on the simulation parameters to obtain the fixture design data with optimized error;

[0077] It can be understood that in this step, the mechanical properties of the fixture are evaluated through finite element simulation, and combined with error modeling and optimization algorithms, the manufacturing and assembly errors are corrected. This step ensures the structural stability, accuracy, and applicability of the fixture, and finally obtains the fixture design data with optimized error, providing a high-precision reference model for subsequent manufacturing and practical applications. In this step, step S5 includes step S51, step S52, and step S53.

[0078] Step S51: Perform simulation processing based on the optimized fixture topology. Among them, through finite element analysis, model and analyze the force conditions, deformations, and thermal stresses of the fixture under different working conditions to obtain simulation results, where the simulation results include the stress distribution, deformation conditions, and fault point positions of the fixture during operation;

[0079] It can be understood that this step adopts an adaptive mesh division strategy to improve the calculation accuracy in high stress concentration areas. Select a suitable mesh element type, which is a hexahedral element in this step, to ensure simulation accuracy and calculation efficiency. Then, set the fixed constraints of the fixture (the contact points with the workpiece in this step). Apply external loads, including clamping force, workpiece weight, and impact force, set thermal boundary conditions, such as heat transfer coefficient and convective heat transfer conditions, and then use the static analysis calculation formula to calculate the force conditions of the fixture, generate a stress distribution contour map, and perform modal analysis to evaluate the vibration characteristics of the fixture, generate a deformation contour map, ensure that the fixture will not affect the machining accuracy due to excessive deformation, perform thermal-mechanical coupling simulation, evaluate the influence of high-temperature environment on the fixture accuracy, identify fault points, find possible failure risks, and improve the service life of the fixture.

[0080] Among them, the static analysis calculation formula is as follows: Ku = F

[0081] Among them, K represents the global stiffness matrix, which is the mechanical properties of the material and the geometric characteristics of the structure, u represents the displacement vector, which is the displacement distribution of the structure under the action of force, and F represents the external force load vector, which is the external load acting on the structure.

[0082] Step S52: Perform error analysis processing based on the simulation results. Among them, evaluate the possible errors of the fixture under different working states by constructing a Gaussian process regression model to obtain potential error sources in the fixture design;

[0083] It can be understood that this step predicts the mean and variance of the error through GPR modeling, and clarifies the error change range of the fixture under different working conditions. Analyze the error sources of the fixture through Gaussian process regression, predict the error range under different working conditions, and finally obtain potential error sources in the fixture design. The error analysis results of this step will be used for error optimization to provide an accurate optimization direction for the fixture design and improve the overall accuracy and stability.

[0084] Step S53: Optimize and adjust the design parameters of the error model based on the potential error sources in the fixture design to obtain the fixture design data after error optimization.

[0085] It can be understood that in this step, a compensation structure is pre-added to the fixture design, including: adding stiffeners to reduce thermal deformation. Optimizing the clamping force distribution to reduce elastic deformation. Adjusting the machining tolerance to match the assembly accuracy. The fixture design data after error optimization makes the fixture more in line with the actual application requirements and improves the machining accuracy. By optimizing the material distribution and error compensation design, the stability of the fixture during long-term use is improved. The fixture after error optimization can adapt to different temperatures, loads and assembly conditions, improving its applicability. The adjusted design parameters obtained through error analysis during the simulation and actual manufacturing processes. These data cover the geometric shape, material properties, stress distribution, deformation conditions and other mechanical performance indicators of the fixture. After error optimization, the fixture design is more in line with the actual working conditions, reducing potential error sources and ensuring higher accuracy and reliability of the fixture during manufacturing and use. In addition, the design data after error optimization also includes machining process parameters based on real-time feedback and dynamic optimization to facilitate more accurate manufacturing and adjustment of the fixture.

[0086] Step S6: Manufacture the fixture according to the fixture design data after error optimization, and collect the fixture force data for dynamic optimization to obtain the optimized fixture machining process parameters.

[0087] It can be understood that the fixture is manufactured according to the fixture design data after error optimization, and the force data is collected for dynamic optimization. By real-time monitoring and optimizing the process parameters, the optimized fixture machining process parameters are obtained. By dynamically adjusting the machining process parameters of the fixture, the machining accuracy and stability of the fixture can be improved, the adaptability and durability of the fixture can be enhanced, and finally the efficiency of the manufacturing process and the product quality can be improved. In this step, step S6 includes step S61, step S62 and step S63.

[0088] Step S61: According to the fixture design data after error optimization, perform fixture manufacturing processing, and based on a preset sensor, collect the force data of the fixture during the machining process in real time. The force data includes the pressure, deformation and vibration data of the fixture during operation.

[0089] It is understandable that in this step, based on the fixture design data optimized according to the error, the actual manufacturing of the fixture is carried out to ensure that the fixture meets the design requirements and has high precision. Based on the preset sensors, the force data of the fixture during the machining process is collected in real time, including the pressure, deformation, and vibration data of the fixture. Through the installed sensors, the force data of the fixture during the machining process is collected in real time. These data include: Pressure data: Record the clamping force applied by the fixture during operation to ensure that the clamping force of the fixture on the workpiece is uniform and stable during the machining process. Deformation data: Monitor the deformation of the fixture through strain gauges or other sensors to ensure that the fixture does not undergo excessive deformation during the machining process, avoiding affecting the machining accuracy. Vibration data: Monitor the vibration of the fixture during machining through vibration sensors, especially during high-speed machining, to ensure that the fixture does not cause unstable machining due to vibration.

[0090] Step S62: According to the force data of the fixture collected in real time, perform dynamic data analysis and processing. Among them, perform time series modeling on the force data through dynamic time series analysis, analyze the force change trend of the fixture under different operation stages, and identify potential anomalies and uneven forces through decision trees to obtain dynamic analysis results;

[0091] It is understandable that in this step, the force data is smoothed, denoised, and normalized to ensure the quality of the data and remove interference factors. Use long short-term memory networks to analyze the time series characteristics of the force data and identify the force change trends of the fixture under different operation stages (such as clamping, machining, loosening, etc.). Based on the time series model, analyze the mechanical changes of the fixture during operation, revealing the changes in the fixture's force over time, including pressure, deformation, and vibration, etc. On the basis of time series analysis, monitor the change trend of the fixture's force data to detect abnormal fluctuations in force in a timely manner. For example: a sudden increase or decrease in the force data may indicate that the fixture has a problem of excessive or insufficient clamping force. A too fast or unstable force change may indicate problems with the deformation or vibration of the fixture. Time series anomalies may reflect the unstable operation of the fixture under different working conditions, or the occurrence of abnormal wear or damage.

[0092] Then, various force data obtained from the timing analysis (such as the pressure, deformation, vibration, etc. of the fixture) are used as input features for the decision tree model. Based on historical data and actual operation experience, the decision tree model is trained to determine the decision rules for various force changes. For example: If the fixture pressure suddenly exceeds the predetermined range, the decision tree determines it as "abnormal". If the deformation amount of the fixture is greater than the preset threshold, it is determined as "uneven force". By obtaining the results of dynamic analysis, through dynamic time series analysis and the decision tree model, the force change trend of the fixture at different stages can be analyzed more accurately, abnormal and uneven force problems can be discovered in a timely manner, and the accuracy of fixture design and operation can be improved. And through real-time dynamic data analysis, immediate feedback on abnormal forces during the fixture operation process can be provided, potential problems can be identified in advance, and production interruptions or product quality problems can be avoided.

[0093] Step S63: According to the results of dynamic analysis, optimize the processing parameters of the fixture to obtain the optimized processing parameters of the fixture.

[0094] It can be understood that in this step, by selecting parameters closely related to the fixture processing process, the fixture processing parameters include fixture clamping force, design of support points, geometric shape of the fixture, material properties, working pressure, etc. The simulated annealing algorithm is used as the optimization algorithm to optimize the selected parameters. The simulated annealing algorithm can efficiently find the optimal solution in the multi-dimensional parameter space. By optimizing the processing parameters of the fixture, it is ensured that the fixture operates stably during the working process, and the processing errors caused by problems such as uneven force and deformation during the processing are reduced, thereby improving the processing accuracy.

[0095] Embodiment 2:

[0096] As Figure 2 shown, this embodiment provides a system for collecting and analyzing 3D printing tooling fixture data based on big data. Refer to Figure 2 The system includes an acquisition unit 701, a cleaning unit 702, a classification unit 703, an optimization unit 704, a simulation unit 705, and an analysis unit 706.

[0097] The acquisition unit 701 is used to acquire the first information, and the first information includes the fixture historical design scheme, processing process parameters, and sensor data;

[0098] The cleaning unit 702 is used to perform feature extraction and data cleaning on the first information to obtain the second information, and the second information includes the first information after data cleaning;

[0099] The classification unit 703 is used to analyze the design intention of the second information based on the BERT-CNN dual-channel and classify the fixture historical design scheme in combination with a preset classification model to obtain a fixture applicability classification database;

[0100] An optimization unit 704, configured to generate a lightweight fixture structure through a generative adversarial network and topology optimization according to a fixture applicability classification database, and optimize the lightweight fixture structure through a preset optimization model to obtain an optimized fixture topology structure; a simulation unit 705, configured to perform simulation according to the optimized fixture topology structure and perform error optimization based on simulation parameters to obtain fixture design data after error optimization;

[0101] An analysis unit 706, configured to manufacture a fixture based on the fixture design data after error optimization, collect fixture force data for dynamic optimization, and obtain optimized fixture processing parameters.

[0102] It should be noted that regarding the system in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0103] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention. The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present invention, and all should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for collecting and analyzing 3D printing tooling fixture data based on big data, characterized in that Including: Obtain first information, where the first information includes fixture historical design schemes, machining process parameters, and sensor data; Perform feature extraction and data cleaning on the first information to obtain second information, where the second information includes the first information after data cleaning; Based on BERT-CNN dual channels, analyze the design intent of the second information, and combine with a preset classification model to classify the fixture historical design schemes, obtaining a fixture applicability classification database; According to the fixture applicability classification database, generate a lightweight fixture structure through a generative adversarial network and topology optimization, and optimize the lightweight fixture structure through a preset optimization model to obtain an optimized fixture topology structure; Perform simulation on the optimized fixture topology structure, and perform error optimization based on the simulation parameters to obtain error-optimized fixture design data; Manufacture a fixture according to the error-optimized fixture design data, and collect fixture force data for dynamic optimization to obtain optimized fixture machining process parameters; Among them, according to the fixture applicability classification database, generate a lightweight fixture structure through a generative adversarial network and topology optimization, and optimize the lightweight fixture structure through a preset optimization model to obtain an optimized fixture topology structure, including: According to the fixture applicability classification database, perform lightweight fixture structure generation processing. Among them, a candidate fixture structure is generated through a generative adversarial network. The generator network generates different fixture structure schemes according to the design parameters, material characteristics, and process requirements in the applicability classification database, while the discriminator network judges the authenticity of these generated fixture structures to obtain a fixture structure that meets the preset lightweight design requirements; According to the generated lightweight fixture structure, perform topology optimization processing. Among them, according to the preset loading conditions and constraints, dynamically adjust the shape and material distribution of the fixture structure to obtain a topologically optimized fixture structure; Optimize the parameters of the topologically optimized fixture structure based on the particle swarm optimization algorithm to obtain an optimized fixture topology structure; Among them, perform simulation on the optimized fixture topology structure, and perform error optimization based on the simulation parameters to obtain error-optimized fixture design data, including: According to the optimized fixture topology structure, perform simulation processing. Among them, through finite element analysis, model and analyze the force, deformation, and thermal stress of the fixture under different working conditions to obtain simulation results. The simulation results include the stress distribution, deformation conditions, and fault point positions of the fixture during operation; According to the simulation results, perform error analysis processing. Among them, evaluate the errors generated by the fixture under different working states by constructing a Gaussian process regression model to obtain potential error sources in the fixture design; Optimize and adjust the design parameters of the error model based on the potential error sources in the fixture design to obtain error-optimized fixture design data.

2. The method for collecting and analyzing the data of the 3D printing tooling fixture based on big data according to claim 1, wherein ,Perform feature extraction and data cleaning on the first information to obtain second information, where the second information includes the first information after data cleaning, including: Perform high-dimensional feature dimensionality reduction and abnormal data detection processing based on the first information. Specifically, use a variational autoencoder to perform high-dimensional feature embedding on the first information, combine an isolation forest to identify abnormal data in the first information, and at the same time use the DBSCAN clustering algorithm to analyze the distribution pattern of the abnormal data, remove the data of isolated points and low-density regions, and obtain a denoised feature data set; Perform multi-scale feature extraction and alignment processing based on the denoised feature data set. Specifically, use wavelet transform to extract features at different time scales in the fixture design and manufacturing process, and obtain a data set after feature extraction; Perform local consistency enhancement and structure preservation processing based on the data set after feature extraction. Specifically, use Laplacian eigenmaps to preserve the local structure of the fixture geometric features, and combine t-SNE to optimize the data distribution in the low-dimensional space, and obtain feature data after structure optimization; Perform noise suppression and data smoothing processing based on the feature data after structure optimization to obtain the second information.

3. The method for collecting and analyzing the 3D printing tooling fixture data based on big data according to claim 2, wherein , Analyze the design intent of the second information based on BERT-CNN dual channels, and combine a preset classification model to classify the fixture historical design schemes, and obtain a fixture applicability classification database, including: Perform text and semantic understanding processing based on the second information after data cleaning. Specifically, use the BERT model to encode the fixture historical design schemes in the second information, extract the context information and potential semantic features of the fixture design, and obtain the design features after semantic encoding; Perform image-based feature extraction processing based on the design features after semantic encoding. Specifically, use a CNN convolutional neural network to perform multi-level convolution and pooling operations on the design features after semantic encoding. The multi-level convolution includes extracting spatial correlation features through a sliding window mechanism, and combining a fully connected layer to fuse these features to obtain the visual features of the fixture historical design scheme; Concatenate the design features after semantic encoding and the visual features of the fixture historical design scheme to form a multi-modal feature vector, and combine a support vector machine to classify the fixture historical design scheme to obtain the applicability classification result of the fixture historical design scheme.

4. A data collection and analysis system for 3D printing tooling fixtures based on big data, characterized in that, Including: An acquisition unit for acquiring the first information, where the first information includes fixture historical design schemes, processing process parameters, and sensor data; A cleaning unit for performing feature extraction and data cleaning on the first information to obtain the second information, where the second information includes the first information after data cleaning; A classification unit for analyzing the design intent of the second information based on BERT-CNN dual channels, and combining a preset classification model to classify the fixture historical design schemes to obtain a fixture applicability classification database; An optimization unit for generating a lightweight fixture structure through a generative adversarial network and topology optimization according to the fixture applicability classification database, and optimizing the lightweight fixture structure through a preset optimization model to obtain an optimized fixture topology structure; A simulation unit for performing simulation according to the optimized fixture topology structure and performing error optimization based on the simulation parameters to obtain fixture design data after error optimization; An analysis unit is used to manufacture a fixture based on the fixture design data after error optimization, and collect fixture force data for dynamic optimization to obtain optimized fixture processing process parameters; Among them, the optimization unit includes: A first optimization subunit is used to perform lightweight fixture structure generation processing according to a fixture applicability classification database. Among them, a candidate fixture structure is generated through a generative adversarial network. The generator network generates different fixture structure schemes according to the design parameters, material characteristics, and process requirements in the applicability classification database, while the discriminator network judges the authenticity of these generated fixture structures to obtain a fixture structure that meets the preset lightweight design requirements; A second optimization subunit is used to perform topology optimization processing according to the generated lightweight fixture structure. Among them, according to the preset loading conditions and constraints, the shape and material distribution of the fixture structure are dynamically adjusted to obtain a fixture structure after topology optimization; A third optimization subunit is used to optimize the parameters of the fixture structure after topology optimization based on a particle swarm optimization algorithm to obtain an optimized fixture topology structure; Among them, the simulation unit includes: A first simulation subunit is used to perform simulation processing according to the optimized fixture topology structure. Among them, through finite element analysis, the force conditions, deformations, and thermal stresses of the fixture under different working conditions are modeled and analyzed to obtain simulation results. The simulation results include the stress distribution, deformation conditions, and fault point positions of the fixture during operation; A second simulation subunit is used to perform error analysis processing according to the simulation results. Among them, by constructing a Gaussian process regression model, the errors generated by the fixture under different working states are evaluated to obtain potential error sources in the fixture design; A third simulation subunit is used to optimize and adjust the design parameters of the error model based on the potential error sources in the fixture design to obtain fixture design data after error optimization.

5. The acquisition and analysis system of 3D printing tooling fixture data based on big data according to claim 4, characterized in that, The cleaning unit includes: A first cleaning subunit is used to perform high-dimensional feature dimensionality reduction and abnormal data detection processing according to the first information. Among them, the first information is embedded with high-dimensional features through a variational autoencoder, and the abnormal data in the first information is identified in combination with an isolation forest. At the same time, the DBSCAN clustering algorithm is used to analyze the distribution pattern of the abnormal data, and the data in the isolated points and low-density regions are removed to obtain a denoised feature data set; A second cleaning subunit is used to perform multi-scale feature extraction and alignment processing according to the denoised feature data set. Among them, the features at different time scales in the fixture design and processing are extracted through wavelet transform to obtain a data set after feature extraction; A third cleaning subunit is used to perform local consistency enhancement and structure preservation processing according to the data set after feature extraction. Among them, the local structure of the fixture geometric features is maintained through Laplacian eigenmaps, and the data distribution is optimized in a low-dimensional space in combination with t-SNE to obtain feature data after structure optimization; A fourth cleaning subunit is used to perform noise suppression and data smoothing processing according to the feature data after structure optimization to obtain the second information.

6. The acquisition and analysis system of 3D printing tooling fixture data based on big data according to claim 5, characterized in that, The classification unit includes: The first classification subunit is used to perform text and semantic understanding processing based on the second information after data cleaning. Among them, the fixture historical design scheme in the second information is encoded by the BERT model to extract the context information and potential semantic features of fixture design, and the design features after semantic encoding are obtained; The second classification subunit is used to perform image feature extraction processing based on the design features after semantic encoding. Among them, multi-level convolution and pooling operations are performed on the design features after semantic encoding through the CNN convolutional neural network. The multi-level convolution includes extracting spatial correlation features through a sliding window mechanism and fusing these features through a fully connected layer to obtain the visual features of the fixture historical design scheme; The third classification subunit is used to splice the design features after semantic encoding with the visual features of the fixture historical design scheme to form a multi-modal feature vector, and classify the fixture historical design scheme in combination with a support vector machine to obtain the applicability classification result of the fixture historical design scheme.

Images