Continuous fiber composite 3D printing process parameter optimization method and system

Abstract

The invention belongs to the related technical field of composite material additive manufacturing, and discloses a continuous fiber composite material 3D printing process parameter optimization method and system.The method comprises the steps that a first sample set is constructed, and samples in the first sample set are combinations of printing parameters and corresponding printing quality indexes and printing efficiency indexes; constructing a prediction model: training the prediction model by using a first sample set to obtain the trained prediction model; and parameter optimization: aiming at comprehensive optimization of the printing quality index and the printing efficiency index, optimizing the combination of the printing parameters by using a multi-objective optimization algorithm based on the trained prediction model to obtain the optimized printing parameter combination. According to the method, parameter optimization can be dynamically carried out in real time in the printing process, the printing parameter combination with the optimal comprehensive performance can be obtained in the printing process, and the optimization efficiency and the optimization effect of a traditional optimization method are effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field related to composite material additive manufacturing, and more specifically, relates to a method and system for optimizing process parameters of continuous fiber composite material 3D printing. Background Art

[0002] Continuous fiber composite material (CFRP) is a composite material with continuous fibers as reinforcement and thermoplastic materials such as polymer resin materials as the matrix. Compared with traditional polymer materials, its specific strength, specific stiffness and fatigue resistance have been greatly improved. Therefore, it has been widely used in military and civilian industrial fields such as aerospace, automobiles and large ships, which have extremely high requirements for performance and weight. However, in the early stages of development, due to the limitations of traditional manufacturing processes, it was impossible to produce according to the specific complex structures designed, which greatly limited the application of continuous fiber composite materials. As a disruptive manufacturing technology, the rapid development of 3D printing (also known as additive manufacturing) has greatly solved the problem of difficult forming of complex structures, which has greatly developed the application of continuous fiber composite materials.

[0003] At the same time, there are many printing process parameters (such as layer angle, nozzle temperature, layer thickness, and fiber filling density) in the 3D printing process, and it is difficult to find a printing parameter combination with the best overall performance. In the absence of a large number of experiments and expert experience guidance, it is difficult to find a printing parameter combination that takes into account both printing efficiency and printing quality. This factor greatly restricts the application of continuous fiber composite 3D printed components in important fields such as aerospace. Therefore, it is very meaningful to predict and optimize the printing parameters in the 3D printing process of continuous fiber composites.

[0004] The technology for optimizing the process parameters of 3D printing of continuous fiber composites has developed relatively slowly. The mainstream method uses the performance test results obtained from a large number of experiments as feedback, and then adjusts the parameters one by one. However, the current 3D printing process of continuous fiber composites involves a wide range of parameters, and there are coupling effects. The adjustment of a single parameter is difficult to meet the optimization needs. In addition, traditional optimization methods are limited by sample performance testing, and the verification cycle of the optimization parameter combination is long. Feedback can only be obtained after printing and mechanical performance testing are completed. If the optimization effect is not up to standard, parameter optimization and sample printing need to be re-performed, and real-time dynamic optimization cannot be achieved. Summary of the invention

[0005] In view of the above defects or improvement needs of the prior art, the present invention provides a method and system for optimizing the 3D printing process parameters of continuous fiber composite materials, which is used to solve the problem that the traditional 3D printing process parameter optimization method adjusts the printing parameters to conduct printing experiments, and then performs performance tests on the samples to select a better printing parameter combination, but it is difficult to achieve multi-parameter coupling optimization and the optimization parameter combination verification cycle is long.

[0006] To achieve the above object, according to one aspect of the present invention, a method for optimizing process parameters of 3D printing of continuous fiber composite materials is provided, comprising:

[0007] Constructing a first sample set: the samples in the first sample set are combinations of printing parameters and corresponding printing quality indicators and printing efficiency indicators;

[0008] Constructing a prediction model: the prediction model is used to predict the printing quality index and the printing efficiency index according to the combination of printing parameters, and the prediction model is trained by using the first sample set to obtain the trained prediction model;

[0009] Parameter optimization: With the goal of achieving the best comprehensive results for printing quality and printing efficiency indicators, the combination of printing parameters is optimized using a multi-objective optimization algorithm based on the trained prediction model to obtain the optimized printing parameter combination, and printing is performed according to the optimized printing parameter combination.

[0010] The method for optimizing process parameters of 3D printing of continuous fiber composite materials provided by the present invention also includes:

[0011] Constructing a second sample set: the samples of the second sample set are monitoring information during the printing process and corresponding defect information;

[0012] Constructing a printing defect monitoring model: the printing defect monitoring model is used to judge defect information according to monitoring information in the printing process, and the printing defect monitoring model is trained using the second sample set to obtain the trained printing defect monitoring model;

[0013] Online defect monitoring: obtaining online monitoring information during the printing process, and obtaining online defect information using the online monitoring information and the trained printing defect monitoring model;

[0014] If the online defect information indicates that a defect exists, a parameter optimization operation is performed.

[0015] According to the method for optimizing process parameters of continuous fiber composite material 3D printing provided by the present invention, constructing the first sample set and the second sample set includes:

[0016] Pre-printing is performed with different printing parameter combinations, wherein a layer of sample is printed under each printing parameter combination; and monitoring information is obtained once when each layer of sample is printed;

[0017] Obtaining corresponding defect information according to the monitoring information, selecting a printing parameter combination corresponding to a defect-free state, and obtaining a printing quality index and a printing efficiency index of a sample printed under the printing parameter combination to construct a first sample set;

[0018] Corresponding defect information is obtained according to the monitoring information, and a second sample set is constructed.

[0019] The method for optimizing process parameters of 3D printing of continuous fiber composite materials provided by the present invention further includes: determining the parameter value space of each printing parameter;

[0020] Printing parameters include multiple of layering angle, nozzle temperature, fiber filling density, wire feed speed, layer thickness, and fiber printing speed;

[0021] The print quality index includes at least one of tensile strength, flexural strength, and porosity;

[0022] Printing efficiency indicators include printing time.

[0023] According to the continuous fiber composite material 3D printing process parameter optimization method provided by the present invention, the parameter optimization adopts the NSGA-Ⅱ algorithm to optimize the combination of printing parameters, and the NSGA-Ⅱ algorithm inputs the initial population into the trained prediction model to predict the printing quality index and printing efficiency index corresponding to each individual in the initial population; then, based on the printing quality index and printing efficiency index predicted by each individual, the individuals in the initial population are evaluated by non-dominated sorting and crowding distance; the next generation of population is generated based on the evaluation results; it is iterated in sequence, and finally the Pareto optimal solution set is output.

[0024] According to the continuous fiber composite material 3D printing process parameter optimization method provided by the present invention, based on the optimal objective function w min Determine the distance between each point in the Pareto optimal solution set and the Utopia point, so as to calculate the optimal solution and the optimal objective function w min The calculation expression is:

[0025]

[0026] Among them, n is the number of optimization objectives, f cτ (x) is the τth target value in the cth Pareto solution, f τ (x) is the value of the Utopia point.

[0027] According to the continuous fiber composite material 3D printing process parameter optimization method provided by the present invention, the non-dominated sorting is to divide individuals into different levels according to the dominance relationship of individuals on multiple targets. Given two individuals A and B, the dominance of A on B is recorded as The mathematical expression is:

[0028]

[0029] Among them, f m (A) is the value of individual A on target m, M is the total number of targets;

[0030] The crowding distance is used to measure the distance between an individual and its neighboring individuals in the target space; for target m, the calculation formula of the crowding distance l of individual i is:

[0031]

[0032] Among them, f m (i+1) and f m (i-1) are the target values ​​of the previous and next individuals of individual i after the target m is sorted; and The maximum and minimum values ​​of the target m.

[0033] According to the method for optimizing process parameters of continuous fiber composite 3D printing provided by the present invention, the monitoring information during the printing process includes multiple of temperature information, surface image information, contact pressure information between the printing nozzle and the deposited layer, acoustic signal information generated by the friction between the printing nozzle and the continuous fiber, and point cloud information of the printing layer;

[0034] Parameter optimization is specifically as follows: parameter optimization is performed at the beginning of printing;

[0035] And conduct online defect monitoring for each printed layer, and selectively optimize parameters based on the online defect information.

[0036] According to another aspect of the present invention, a continuous fiber composite material 3D printing process parameter optimization system is provided, comprising a 3D printing platform; the 3D printing platform comprises a six-axis industrial robot, a print nozzle, a printing forming platform and a host computer; the print nozzle is installed at the end of the six-axis industrial robot, the host computer is connected to the six-axis industrial robot, the print nozzle and the printing forming platform respectively, and the six-axis industrial robot drives the print nozzle to perform printing on the printing forming platform;

[0037] The host computer includes a memory and a processor, the memory stores a computer program, and the processor executes any one of the above-mentioned continuous fiber composite material 3D printing process parameter optimization methods when executing the computer program.

[0038] According to the continuous fiber composite material 3D printing process parameter optimization system provided by the present invention, the 3D printing platform is also integrated with a sensor component for obtaining monitoring information during the printing process, and the sensor component includes at least one of an infrared camera, a visible light camera, a pressure sensor, an acoustic emission sensor and a laser scanner; the infrared camera and the visible light camera are used to collect temperature information and surface state information during the printing process; the pressure sensor is located below the printing forming platform, and is used to collect contact pressure information between the printing nozzle and the deposition layer during the printing process; the acoustic emission sensor is located at the upper feeding position of the printing nozzle, and is used to collect acoustic signal information generated by the friction between the printing nozzle and the continuous fiber; the laser scanner is located below the printing nozzle, and is used to collect point cloud information of the printing layer; the host computer is connected to the sensor component.

[0039] In general, compared with the prior art, the above technical solutions conceived by the present invention provide a method and system for optimizing process parameters of continuous fiber composite material 3D printing:

[0040] 1. A training prediction model is proposed. Based on the trained prediction model, a multi-objective optimization algorithm can be used to quickly optimize the printing parameter combination. The optimization process does not require printing experiments and performance tests of printed samples. It can realize real-time dynamic parameter optimization during the printing process. Real-time dynamic optimization can quickly adjust the printing parameters to significantly improve the optimization efficiency, reduce waste samples and significantly reduce printing costs. It can also achieve coupled optimization of multiple printing parameters, which is conducive to obtaining the printing parameter combination with the best comprehensive performance during the printing process. Multi-objective optimization can achieve simultaneous optimization of printing quality and printing efficiency, solving the problem that the printing quality and printing efficiency are mutually restricted and difficult to optimize during the 3D printing of continuous fiber composite materials, effectively improving the optimization efficiency and optimization effect of traditional optimization methods, and filling the gap in the field of parameter optimization of 3D printing of continuous fiber composite materials.

[0041] 2. Innovatively combines the online defect monitoring process with the parameter optimization process, realizing a closed-loop dynamic optimization process of multi-sensor information collection - deep learning prediction model learning and training - prediction and optimization of the best printing parameter combination - defect monitoring layer by layer during the printing process - secondary optimization of printing parameters after defects are found;

[0042] 3. The proposed closed-loop dynamic optimization system can achieve the best printing parameter combination with comprehensive performance in one printing process, without the need for repeated parameter adjustment optimization and sample printing, thus improving the stability of the continuous fiber composite material 3D printing parameter optimization method, realizing the cyclic and dynamic optimization of printing parameters, and improving the optimization efficiency and effect while reducing the optimization cost;

[0043] 4. By integrating sensors such as infrared cameras, visible light cameras, pressure sensors, acoustic emission sensors and laser scanners on the print head, a 3D printing platform with multi-sensor in-situ online monitoring function is formed, which realizes the in-situ data collection of the continuous fiber composite material 3D printing process and timely feedback to the host computer, providing a reference basis for subsequent related research on printing parameter optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic diagram of the composition of a continuous fiber composite material 3D printing platform with multi-sensor in-situ online monitoring function provided by the present invention;

[0045] Figure 2 It is a schematic diagram of a closed-loop dynamic optimization method for continuous fiber reinforced composite material 3D printing process parameters based on deep learning provided by the present invention;

[0046] Figure 3 It is a schematic diagram of the network architecture of a CNN neural network in a specific embodiment provided by the present invention;

[0047] Figure 4 It is a flow chart of the NSGA-Ⅱ algorithm in a specific embodiment provided by the present invention;

[0048] Figure 5 It is a Pareto optimal result diagram of a single optimization in a specific embodiment provided by the present invention;

[0049] Figure 6 It is a comparison diagram used to verify the optimization results in the specific embodiment provided by the present invention;

[0050] Throughout the drawings, the same reference numerals are used to denote the same elements or structures, wherein:

[0051] 1-Visible light camera; 2-Infrared camera; 3-Acoustic emission sensor; 4-Pressure sensor; 5-Laser scanner; 6-Print nozzle; 7-Host computer; 8-Six-axis industrial robot; 9-Printing platform. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with 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. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0053] See also Figure 1 and Figure 2This embodiment provides a method for optimizing process parameters of 3D printing of continuous fiber composite materials, and the method for optimizing process parameters of 3D printing of continuous fiber composite materials includes:

[0054] Constructing a first sample set: the samples in the first sample set are a combination of printing parameters and corresponding printing quality indicators and printing efficiency indicators; the combination of printing parameters is a combination of specific values ​​of different printing parameters, and the printing quality indicator and printing efficiency indicator are the quality indicator value and efficiency indicator value of the sample printed under the printing parameter combination;

[0055] Constructing a prediction model: the prediction model is used to predict the printing quality index and the printing efficiency index according to the combination of printing parameters, and the prediction model is trained using the first sample set to obtain the trained prediction model; the trained prediction model can output the predicted values ​​of the printing quality index and the printing efficiency index according to the combination of printing parameters;

[0056] Parameter optimization: With the purpose of achieving the best comprehensive print quality index and print efficiency index, the combination of print parameters is optimized using a multi-objective optimization algorithm based on the trained prediction model, the optimized print parameter combination is obtained, and printing is performed according to the optimized print parameter combination. In the optimization process, the object of optimization is the combination of print parameters, and the purpose of optimization is to achieve the best comprehensive print quality index and print efficiency index corresponding to the print parameter combination, wherein the print quality index and print efficiency index corresponding to the print parameter combination are predicted and obtained through the trained prediction model. By adopting a multi-objective optimization algorithm, it is possible to achieve the best print parameter combination in multiple optimization goals.

[0057] The continuous fiber reinforced composite material in this embodiment includes, but is not limited to, continuous fiber reinforced resin-based composite materials such as carbon fiber, glass fiber, aramid fiber, plant fiber, and basalt fiber.

[0058] The prediction model can be a forward prediction deep learning model, which is used to forward predict the mapping relationship between printing process parameters and mechanical properties and manufacturing efficiency, including but not limited to CNN neural network, Bp neural network, LSTM neural network, RNN neural network, N-BEATS and other deep learning models. Multi-objective optimization algorithm, based on the mapping relationship between printing process parameters and mechanical properties and manufacturing efficiency, obtains the optimal printing parameter combination, including but not limited to NSGA-Ⅱ, MOGWO, MOPSO, SPEA2, NSGA-Ⅲ and other multi-objective optimization algorithms.

[0059] Furthermore, the method for optimizing process parameters of 3D printing of continuous fiber composite materials provided in this embodiment further includes:

[0060] Constructing a second sample set: the samples of the second sample set are monitoring information during the printing process and corresponding defect information;

[0061] Constructing a printing defect monitoring model: the printing defect monitoring model is used to judge defect information according to monitoring information in the printing process, and the printing defect monitoring model is trained using the second sample set to obtain the trained printing defect monitoring model;

[0062] Online defect monitoring: obtaining online monitoring information during the printing process, and obtaining online defect information using the online monitoring information and the trained printing defect monitoring model;

[0063] If the online defect information indicates that a defect exists, a parameter optimization operation is performed.

[0064] In some specific embodiments, constructing the first sample set and the second sample set includes:

[0065] Pre-printing is performed with different printing parameter combinations, wherein a layer of sample is printed under each printing parameter combination; and monitoring information is obtained once when each layer of sample is printed;

[0066] According to the monitoring information, the corresponding defect information is obtained, the corresponding printing parameter combination when there is no defect is selected, and the printing quality index and printing efficiency index of the sample printed under the printing parameter combination are obtained to construct the first sample set; when constructing the first sample set, the printing parameter combination when there is no defect or no obvious defect and the corresponding index efficiency index are needed. At this time, the monitoring information can be an image, and then the image can be used to quickly determine whether there is a defect, so that images without defects or no obvious defects can be quickly screened out, and the corresponding printing parameter combination can be used to construct the first sample set. The index information in the first sample set can be obtained by measuring the printed sample.

[0067] The corresponding defect information is obtained according to the monitoring information to construct the second sample set. When constructing the second sample set, because the printing defect information needs to be judged according to the monitoring information, such as whether there is a defect, etc., a variety of monitoring information can be used to improve the comprehensiveness and accuracy of the judgment.

[0068] Furthermore, the method for optimizing process parameters of 3D printing of continuous fiber composite materials provided in this embodiment further includes: determining a parameter value space of each printing parameter;

[0069] Printing parameters include multiple of layering angle, nozzle temperature, fiber filling density, wire feed speed, layer thickness, and fiber printing speed;

[0070] The printing quality index includes at least one of tensile strength, bending strength and porosity; the printing quality index can be a relevant index that can reflect the quality of the printed sample, such as a strength index or a defect index, etc. Multiple indicators can be used, and the specific type and quantity are not limited.

[0071] Printing efficiency indicators include printing time.

[0072] The monitoring information during the printing process includes multiple of temperature information, surface image information, contact pressure information between the printing nozzle and the deposited layer, acoustic signal information generated by the friction between the printing nozzle and the continuous fiber, and point cloud information of the printed layer;

[0073] Parameter optimization is specifically as follows: parameter optimization is performed at the beginning of printing;

[0074] And conduct online defect monitoring for each printed layer, and selectively optimize parameters according to the online defect information. The preset time for pausing printing for each printed layer can be controlled to collect the online monitoring information of the printed layer and conduct online defect monitoring. When the online defect information shows that there are defects, parameter optimization is performed to adjust the printing parameters so that the printing process is printed with the optimized parameter combination.

[0075] Furthermore, the present embodiment provides a continuous fiber composite material 3D printing process parameter optimization system, which includes a 3D printing platform; the 3D printing platform includes a six-axis industrial robot 8, a print nozzle 6, a printing forming platform 9 and a host computer 7; the print nozzle 6 is installed at the end of the six-axis industrial robot 8, the host computer 7 is connected to the six-axis industrial robot 8, the print nozzle 6 and the printing forming platform 9 respectively, and the six-axis industrial robot 8 drives the print nozzle 6 to print on the printing forming platform 9;

[0076] The host computer 7 includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, any one of the above-mentioned continuous fiber composite material 3D printing process parameter optimization methods is executed.

[0077] Furthermore, the 3D printing platform is also integrated with a sensor component for obtaining monitoring information during the printing process, and the sensor component includes at least one of an infrared camera 2, a visible light camera 1, a pressure sensor 4, an acoustic emission sensor 3 and a laser scanner 5; the infrared camera 2 and the visible light camera 1 are used to collect temperature information and surface state information during the printing process; the pressure sensor 4 is located below the printing forming platform 9, and is used to collect contact pressure information between the print head 6 and the deposition layer during the printing process; the acoustic emission sensor 3 is located at the feeding position at the upper end of the print head 6, and is used to collect acoustic signal information generated by the friction between the print head 6 and the continuous fiber; the laser scanner 5 is located below the print head 6, and is used to collect point cloud information of the printing layer; the host computer 7 is connected to the sensor component.

[0078] In some specific embodiments, the present specific embodiments provide a closed-loop dynamic optimization system and method for continuous fiber reinforced composite 3D printing process parameters based on deep learning, including two parts: a multi-sensor in-situ online monitoring system and a dynamic cycle optimization method. The two parts are interconnected to form a closed-loop dynamic optimization system integrating prediction, optimization and monitoring, which makes up for the shortcomings of traditional continuous fiber composite 3D printing parameter optimization methods, effectively reduces the defects generated during the printing process, and reduces the time cost and material cost of optimization.

[0079] like Figure 1 As shown, the present embodiment provides a 3D printing platform with a multi-sensor in-situ online monitoring function, which is composed of: a six-axis industrial robot 8, a print nozzle 6, a printing forming platform 9, a host computer 7, an infrared camera 2, a visible light camera 1, a pressure sensor 4, an acoustic emission sensor 3 and a laser scanner 5; the six-axis industrial robot 8 is the basis of the entire closed-loop dynamic optimization system; the print nozzle 6 is installed at the end of the six-axis industrial robot 8 and moves under its drive; the infrared camera 2 and the visible light camera 1 surround the print nozzle 6 and are aligned with the position of the nozzle at the bottom of the print nozzle 6 to collect temperature information and surface state information during printing; the pressure sensor 4 is located on the printing forming platform 9 is located below the printing nozzle 6, and is used to collect the contact pressure information between the nozzle and the deposited layer during the printing process; the acoustic emission sensor 3 is located at the feeding position at the upper end of the printing nozzle 6, and is used to collect the acoustic signal information generated by the friction between the nozzle and the continuous fiber; the laser scanner 5 is located below the printing nozzle 6, and is aimed at the side of the printing layer to collect the point cloud information of the printing layer; the host computer 7 is connected to the six-axis industrial robot 8, the printing nozzle 6 and the printing forming platform 9, and is connected to various sensors, and the various printing information collected by the above sensors is feature extracted and fused, based on which the deep learning model is used to predict and optimize the printing parameters, and the optimized printing process is monitored online, and closed-loop dynamic optimization is performed in real time.

[0080] This embodiment also provides a method for dynamic cycle optimization of printing parameters associated with the above-mentioned multi-sensor in-situ online monitoring system, the method comprising the following steps:

[0081] (1) Pre-printing is performed with different printing parameters, and a Python script is used to control the print head 6 to pause for 3 to 5 seconds after each layer is printed, so that multiple sensors can collect information;

[0082] (2) Preliminarily selecting 30 sets of printing parameter combinations without obvious printing defects based on the visible light image information collected in step (1);

[0083] (3) dividing the several groups of printing parameters in step (2) into two groups: a training set and a test set, and using them to train the CNN neural network prediction model so that it can predict the mapping relationship between the printing parameters and the printing quality and printing efficiency;

[0084] (4) The mapping relationship between the printing parameters obtained in step (3) and the printing quality and printing efficiency is input into the NSGA-II multi-objective optimization algorithm as the objective function, and the optimal printing parameter combination is predicted and optimized by the optimization algorithm;

[0085] (5) Preprocessing the information collected in step (1) to extract and fuse the characteristic values ​​of the information, and establish a connection with the printing defects, and then input it into the YOLOv8 model, i.e., the printing defect monitoring model, for learning, so that it can judge whether a defect occurs based on the information fed back by the multi-sensor;

[0086] (6) Based on the result of the judgment in step (5), if no defects occur, printing is continued; if defects occur, printing is stopped, and the above steps (4) to (5) are repeated, thereby realizing closed-loop dynamic optimization of continuous fiber reinforced composite material 3D printing process parameters.

[0087] Firstly, based on the monitoring results of multiple sensors, several groups of printing parameter combinations without obvious printing defects are screened out and divided into training sets and test sets to train the forward prediction model. Then, the mapping relationship between the printing process parameters and the mechanical properties and manufacturing efficiency obtained by the forward prediction model is used as the objective function of the multi-objective optimization algorithm to obtain the optimal solution. Finally, the collected data is preprocessed to extract features and input into the defect detection deep learning model for feature fusion. A corresponding relationship is established between the fusion results and the printing defects to realize defect identification, which is used to identify defects layer by layer in the printing process of verifying the optimal solution, thereby realizing real-time iterative optimization.

[0088] Among them, the preprocessing method in step (5) includes but is not limited to image denoising, median filtering, 3D point cloud reconstruction and other methods. The defect detection deep learning model fuses the preprocessed data at the feature level and establishes a connection with the printing defects, so that it can judge whether a defect has occurred according to the input sensor information during the printing process, including but not limited to YOLOv8, GAN network, GRU gated recurrent unit and other deep learning models. Feature-level fusion can first extract the features of each monitoring information, and then perform feature fusion, and perform defect monitoring and judgment based on the fused features. Feature fusion can be a feature splicing method, etc., and the specific fusion method is not limited.

[0089] The optimization method provided by the present invention can realize a closed-loop dynamic optimization process through the above steps: multi-sensor information collection - CNN neural network learning and training - NSGA-Ⅱ multi-objective optimization algorithm prediction and optimization of the best printing parameters - YOLOv8 model layer-by-layer monitoring of defects in the printing process - secondary optimization of printing parameters after defects are found. Figure 2 As shown, the generation of defects in the printing process can be effectively reduced and the mechanical properties of the formed samples can be significantly improved.

[0090] Furthermore, this specific embodiment selects five printing parameters, namely, layer angle α (°), nozzle temperature T (°C), fiber filling density D (%), layer thickness L (mm) and fiber printing speed V (mm / s), as optimization targets.

[0091] Furthermore, the parameter value space of the optimization algorithm given in this specific embodiment is as follows. The parameter value space can be set according to the achievable range of the parameters of the printing device, or a better range can be selected according to empirical values. The specific determination method is not limited:

[0092]

[0093] Furthermore, 30 groups of printing parameter combinations without obvious printing defects screened out in this specific embodiment are shown in Table 1.

[0094] Table 1

[0095]

[0096] Furthermore, this specific embodiment selects tensile strength (σ / MPa) and printing time (η / min) as the prediction results of the CNN neural network, which are respectively used as representative values ​​of the printing quality and printing efficiency described in the present invention.

[0097] Furthermore, CNN convolutional neural network is a feedforward neural network with deep structure and convolutional calculation. It realizes end-to-end learning through a series of convolution, pooling, full connection and other operations. It has strong generalization and adaptive learning capabilities. The network architecture diagram is as follows: Figure 3 This specific embodiment performs forward prediction based on the CNN convolutional neural network, converts the printing parameter data into a feature matrix, and then extracts the input data features through operations such as the constructed convolution layer, batch normalization layer, and activation layer. The network learns and the prediction results are output through the fully connected layer. The input function is:

[0098] V = conv2(W,X,"valid") + b;

[0099] Where conv2() is the function used for convolution operation in Matlab R2024a, W is the convolution kernel matrix, X is the input matrix, valid is the type of convolution operation, and b is the bias.

[0100] The output function of the CNN convolutional neural network is:

[0101]

[0102] in is the activation function, V is the above input function. For the last fully connected layer, set it as the Lth layer, the output result is y in vector form L , the expected output, i.e. the true value, is d, then the total error formula is:

[0103]

[0104] It should be noted that the above input and output functions are for each convolutional layer. Each convolutional layer has a different weight matrix W, and W, X and Y are all in matrix form.

[0105] CNN convolutional neural network is trained by gradient descent and back propagation algorithm. The gradient formula of its convolution layer and pooling layer is:

[0106]

[0107] Among them, E is the total error, W is the convolution kernel matrix, and V is the input function.

[0108] Further, 24 groups of the 30 groups of data in Table 1 are randomly selected as training samples to complete the training of the CNN neural network, and the remaining 6 groups are used as validation sets to verify the model prediction effect. The root mean square error (RMSE) and mean absolute error (MAE) are used to quantitatively characterize the performance of the CNN neural network in this specific embodiment. The calculation expressions of RMSE and MAE are as follows:

[0109]

[0110] Among them, y pred Represents the predicted value obtained by the trained model, y true represents the true value in the validation set, sum is the summation function, n is the number of elements, For n y true The average value of .

[0111] As shown in Table 2, the RMSE value and MAE value of the CNN neural network in this specific embodiment on the tensile strength verification set are 0.889 and 1.168 respectively; the RMSE value and MAE value on the printing time verification set are 2.733 and 2.569 respectively.

[0112] Table 2

[0113]

[0114] Furthermore, the NSGA-Ⅱ algorithm has become an excellent multi-objective optimization algorithm due to its fast non-dominated sorting, elite strategy and congestion calculation features, which can ensure that a uniformly distributed solution close to the Pareto front can be obtained in the multi-objective optimization problem. This specific embodiment optimizes the printing parameters based on the NSGA-Ⅱ algorithm, and its flow chart is as follows: Figure 4 shown.

[0115] The purpose of non-dominated sorting is to divide individuals into different levels according to their dominance relationship on multiple objectives. Given two individuals A and B, if A is not worse than B in all objectives and is better than B in at least one objective, then A is said to dominate B (denoted as ), its mathematical expression is:

[0116]

[0117] Among them, f m (A) is the value of individual A on target m, and M is the dimension of the target.

[0118] The crowding distance is used to measure the distance between an individual and its neighboring individuals in the target space, and is used to maintain the diversity of the population. For each individual, its crowding degree on the target is calculated. The crowding distance is the difference between the distances of two adjacent individuals on the target. For target m, the calculation formula for the crowding distance l of individual i is:

[0119]

[0120] Among them, f m (i+1) and f m(i-1) are the target values ​​of the previous and next individuals of individual i after the target m is sorted; and The maximum and minimum values ​​of the target m.

[0121] Furthermore, the specific optimization steps of CNN neural network and NSGA-Ⅱ algorithm are as follows:

[0122] First, tensile strength (σ / MPa) and printing time (η / min) are selected as representative values ​​of the printing quality and printing efficiency described in the present invention. The objective function and constraint conditions of tensile strength and printing time are defined as:

[0123]

[0124] Among them, minmize_F(α,T,D,L,V) is defined as f σ and f η The combination of five printing process parameters that achieves the minimum value, f σ and f η They are the mapping relationships between the printing process parameters (laying angle α (°), nozzle temperature T (°C), fiber filling density D (%), layer thickness L (mm) and fiber printing speed V (mm / s)) and the tensile strength and printing time obtained by the CNN neural network, and the printing process parameters are selected within the given parameter space.

[0125] Furthermore, the NSGA-Ⅱ algorithm is used to find the minimum value of the objective function. The objective function f is constructed using the trained CNN neural network prediction model. σ and f η In the reverse optimization process of process parameters, the NSGA-Ⅱ algorithm will autonomously generate many combinations of process parameters for initial filling, where the process parameters include layer angle, nozzle temperature, fiber filling density, layer thickness and fiber printing speed. The initial population is input into the objective function f σ and f η , used to predict tensile strength and printing time; then, based on the objective function f σ and f η , the initial population is evaluated by non-dominated sorting and crowding distance calculation. The excellent part of the initial population is selected to enter the next generation population according to the non-dominated level and crowding degree; the rest generates a new population through crossover and mutation and enters the next generation population; when the size of the next generation population reaches a predetermined size, the reverse model will enter the next iteration. After the iterative process is completed, the Pareto optimal solution set is output. The optimization process established in this specific embodiment is completed on MatlabR2024a, and the NSGA-Ⅱ algorithm population is initialized to search within the given parameter space.

[0126] That is, parameter optimization uses the NSGA-Ⅱ algorithm to optimize the combination of printing parameters. The NSGA-Ⅱ algorithm inputs the initial population into the trained prediction model to predict the printing quality index and printing efficiency index corresponding to each individual in the initial population; then, based on the printing quality index and printing efficiency index predicted by each individual, the individuals in the initial population are evaluated through non-dominated sorting and crowding distance; the next generation of population is generated based on the evaluation results; it is iterated in sequence, and finally the Pareto optimal solution set is output.

[0127] Furthermore, in this specific embodiment, the parameters are determined by repeated debugging, the initial population size is set to 50, the crossover probability is 0.85, the variance probability is 0.01, the mutation probability is 0.2, the maximum number of iterations is 120, and the mapping relationship between the printing parameters and the printing quality and printing efficiency obtained by the CNN neural network is used as the objective function of the NSGA-Ⅱ algorithm. The Pareto optimal boundary is finally obtained as follows: Figure 5 This specific embodiment is based on the optimal objective function w min Determine the distance between each point in the Pareto optimal solution set and the Utopia point, so as to calculate the optimal solution and the optimal objective function w min The calculation expression is:

[0128]

[0129] Where n is the number of targets, f cτ (x) is the τth target value among the c Pareto solutions, f τ (x) is the value of the Utopia point, that is, the ideal optimal point; the point in the Pareto optimal solution set that makes the optimal objective function reach the minimum value can be taken as the optimal solution.

[0130] Furthermore, the YOLOv8 model is one of the excellent deep learning models in the field of feature fusion and defect detection. This specific embodiment selects the YOLOv8 model as a deep learning model for feature fusion and establishing the relationship between feature fusion results and printing defects, and is used for real-time monitoring of the entire optimization process. In the verification printing process of the above-mentioned Pareto optimal solution, whether secondary optimization is needed is determined layer by layer based on the established relationship between the feature fusion results and printing defects.

[0131] Furthermore, in order to prove that each optimization in this specific embodiment is effective, the optimized process parameters were subjected to actual experiments and corresponding mechanical property tests, and the test results were compared with No. 21 with the best comprehensive performance in Table 1. The experimental results show that with 0° ply angle, 260°C nozzle temperature, 80% filling rate, 0.6mm layer thickness and 10mm / s printing speed as the optimized parameters, the maximum tensile strength of the sample is 144.7MPa, and the printing time is only 77.8min. Compared with No. 21 with the best comprehensive performance in Table 1, the manufacturing efficiency has increased by about 27% year-on-year while the mechanical properties have been improved by about 53%. Figure 6 shown.

[0132] The present invention provides a closed-loop dynamic optimization system and method for 3D printing process parameters of continuous fiber reinforced composite materials based on deep learning. The system is based on a 3D printing platform that integrates multiple sensors such as an infrared camera 2, a visible light camera 1, a pressure sensor 4, an acoustic emission sensor 3, and a laser scanner 5, and incorporates deep learning models such as a forward prediction model, a multi-objective optimization algorithm, and a target detection model. The present invention innovatively combines the online defect monitoring process with the parameter optimization process to form a closed-loop dynamic optimization system that integrates prediction, optimization, and monitoring. While solving the problem that the printing quality and printing efficiency are mutually restricted and difficult to optimize during the 3D printing of continuous fiber composite materials, the system cooperates with a multi-sensor in-situ monitoring device to detect defects layer by layer, and performs real-time and dynamic cyclic iterative optimization of printing parameters. The optimal parameter combination can be verified in real time and layer by layer, and feedback is given through real-time monitoring, and the printing parameters are cyclically iteratively optimized. Finally, the printing parameter combination with the best comprehensive performance can be obtained in one printing process, without the need for repeated parameter optimization and sample printing.

[0133] It will be easily understood by those skilled in the art that 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 protection scope of the present invention.

Claims

1. A method for optimizing process parameters of 3D printing of continuous fiber composite materials, characterized in that: include: Constructing a first sample set: the samples in the first sample set are combinations of printing parameters and corresponding printing quality indicators and printing efficiency indicators; Constructing a prediction model: the prediction model is used to predict the printing quality index and the printing efficiency index according to the combination of printing parameters, and the prediction model is trained by using the first sample set to obtain the trained prediction model; Parameter optimization: With the goal of achieving the best comprehensive results for printing quality and printing efficiency indicators, the combination of printing parameters is optimized using a multi-objective optimization algorithm based on the trained prediction model to obtain the optimized printing parameter combination, and printing is performed according to the optimized printing parameter combination.

2. The method for optimizing process parameters of continuous fiber composite material 3D printing according to claim 1, characterized in that: Also includes: Constructing a second sample set: the samples of the second sample set are monitoring information during the printing process and corresponding defect information; Constructing a printing defect monitoring model: the printing defect monitoring model is used to judge defect information according to monitoring information in the printing process, and the printing defect monitoring model is trained using the second sample set to obtain the trained printing defect monitoring model; Online defect monitoring: obtaining online monitoring information during the printing process, and obtaining online defect information using the online monitoring information and the trained printing defect monitoring model; If the online defect information indicates that a defect exists, a parameter optimization operation is performed.

3. The method for optimizing process parameters of 3D printing of continuous fiber composite materials according to claim 2, characterized in that: Constructing the first sample set and the second sample set includes: Pre-printing is performed with different printing parameter combinations, wherein a layer of sample is printed under each printing parameter combination; and monitoring information is obtained once when each layer of sample is printed; Obtaining corresponding defect information according to the monitoring information, selecting a printing parameter combination corresponding to a defect-free state, and obtaining a printing quality index and a printing efficiency index of a sample printed under the printing parameter combination to construct a first sample set; Corresponding defect information is obtained according to the monitoring information, and a second sample set is constructed.

4. The method for optimizing process parameters of continuous fiber composite material 3D printing according to claim 1, characterized in that: Also includes: Determine the parameter value space of each printing parameter; Printing parameters include multiple of layer angle, nozzle temperature, fiber filling density, wire feed speed, layer thickness, and fiber printing speed; The print quality index includes at least one of tensile strength, flexural strength, and porosity; Printing efficiency indicators include printing time.

5. The method for optimizing process parameters of 3D printing of continuous fiber composite materials according to claim 1, characterized in that: The parameter optimization uses the NSGA-Ⅱ algorithm to optimize the combination of printing parameters. The NSGA-Ⅱ algorithm inputs the initial population into the trained prediction model to predict the printing quality index and printing efficiency index corresponding to each individual in the initial population; then, based on the printing quality index and printing efficiency index predicted by each individual, the individuals in the initial population are evaluated by non-dominated sorting and crowding distance; the next generation of population is generated based on the evaluation results; it is iterated in sequence and finally the Pareto optimal solution set is output.

6. The method for optimizing process parameters of continuous fiber composite material 3D printing according to claim 5, characterized in that: Based on the optimal objective function w min Determine the distance between each point in the Pareto optimal solution set and the Utopia point, so as to calculate the optimal solution and the optimal objective function w min The calculation expression is: Among them, n is the number of optimization objectives, f cτ (x) is the τth target value in the cth Pareto solution, f τ (x) is the value of the Utopia point.

7. The method for optimizing process parameters of continuous fiber composite material 3D printing according to claim 5, characterized in that: Non-dominated sorting is to divide individuals into different levels according to their dominance relationship on multiple targets. Given two individuals A and B, the mathematical expression of A dominating B, denoted as A<B, is: Among them, f m (A) is the value of individual A on target m, M is the total number of targets; The crowding distance is used to measure the distance between an individual and its neighboring individuals in the target space; for target m, the calculation formula of the crowding distance l of individual i is: Among them, f m (i+1) and f m (i-1) are the target values ​​of the previous and next individuals of individual i after the target m is sorted; and The maximum and minimum values ​​of the target m.

8. The method for optimizing process parameters of continuous fiber composite material 3D printing according to claim 2, characterized in that: The monitoring information during the printing process includes multiple of temperature information, surface image information, contact pressure information between the printing nozzle and the deposited layer, acoustic signal information generated by the friction between the printing nozzle and the continuous fiber, and point cloud information of the printed layer; Parameter optimization is specifically as follows: parameter optimization is performed at the beginning of printing; And conduct online defect monitoring for each printed layer, and selectively optimize parameters based on the online defect information.

9. A continuous fiber composite material 3D printing process parameter optimization system, characterized in that: It includes a 3D printing platform; the 3D printing platform includes a six-axis industrial robot, a printing nozzle, a printing forming platform and a host computer; the printing nozzle is installed at the end of the six-axis industrial robot, the host computer is connected to the six-axis industrial robot, the printing nozzle and the printing forming platform respectively, and the six-axis industrial robot drives the printing nozzle to print on the printing forming platform; The host computer includes a memory and a processor, the memory stores a computer program, and the processor executes the continuous fiber composite material 3D printing process parameter optimization method described in any one of claims 1 to 8 when executing the computer program.

10. The continuous fiber composite material 3D printing process parameter optimization system according to claim 9, characterized in that: The 3D printing platform is also integrated with a sensor component for obtaining monitoring information during the printing process, and the sensor component includes at least one of an infrared camera, a visible light camera, a pressure sensor, an acoustic emission sensor and a laser scanner; the infrared camera and the visible light camera are used to collect temperature information and surface state information during the printing process; the pressure sensor is located below the printing forming platform and is used to collect contact pressure information between the printing nozzle and the deposition layer during the printing process; The acoustic emission sensor is located at the feeding position at the upper end of the print nozzle, and is used to collect the acoustic signal information generated by the friction between the print nozzle and the continuous fiber; the laser scanner is located below the print nozzle, and is used to collect the point cloud information of the printing layer; the host computer is connected to the sensor component.

Images