Powder material 3d intelligent printing system

Abstract

This invention discloses an intelligent 3D printing system for powder materials. After data preprocessing, feature engineering, model selection, model training, and model verification, a molten pool porosity prediction model is generated. The system receives molten pool state data, atmosphere distribution data, and laser state data, and analyzes and processes them to generate molten pool quality monitoring index, atmosphere distribution index, and laser state index, respectively. These indices are then transmitted to the molten pool porosity prediction model, which generates a molten pool porosity prediction index. Based on this index, printing parameters or atmosphere control are adjusted in real time. Correlation analysis is performed on the molten pool state data, laser state data, and atmosphere distribution data to establish the molten pool porosity prediction model. The system uses the porosity prediction results to adjust printing parameters or control the atmosphere in real time to reduce the porosity incidence rate.

Description

Technical Field

[0001] This invention belongs to the field of 3D printing technology, specifically relating to an intelligent 3D printing system for powder materials. Background Technology

[0002] Additive manufacturing, commonly known as 3D printing, is a process based on laser powder bed melting (LPBF). Laser powder bed sintering is a widely used 3D printing process for materials such as metals and plastics, also known as selective laser melting or laser sintering. The principle of this process involves the steps of powder layering – laser scanning – melting and sintering – stacking and solidification – component cooling – part extraction. Laser printing parameters include laser power, scanning speed, and preheating temperature. The gas environment used in the laser printing process is a gas (such as nitrogen or a nitrogen-nitrogen mixture) or a mixture of gases. These gases constitute the atmosphere in the LPBF process. The paper "Metal Vaporization and Its Influence in Laser Powder Bed Melting" published by Liu Jingge and Peng Wen describes the influence of metal vaporization on plume, spatter, erosion, and atmosphere in the LPBF process. The link to the paper is as follows:

[0003] The following is an excerpt from the article: https: / / www.sciencedirect.com / science / article / pii / S0264127522001265? via %3Dihub:

[0004] In the LPBF process, the area on the powder layer that is melted by the laser is called the keyhole. After the metal powder in the keyhole melts and sinters, a molten pool is formed. If the power of the laser or electron beam is very high, the metal powder may rapidly cool to its boiling point while melting, resulting in vaporization. In this step, some of the metal powder transforms into metal vapor or gaseous metal. The metal vapor is a mixture of metal vapor, plasma, and small condensed particles above the molten pool. When vaporization occurs, the generated metal vapor expands rapidly and produces a strong gas flow within the laser printing area. As the laser energy input increases, due to the vapor... As the laser beam intensifies, powder particles near the laser trajectory are removed. This removal of powder leads to uneven thickness and unstable melting. When the metal vapor is ejected, powder particles around the keyhole are blown up from the powder layer. In addition, due to the Bernoulli effect, a low-pressure zone is generated, and particles slightly away from the keyhole begin to move towards the metal vapor. At the same time, liquid particles ejected from the molten pool splash. The plume of the molten pool flow refers to the flow and morphology of the liquid metal inside the molten pool after the metal powder melts due to laser irradiation and energy transfer. This includes parameters such as the flow rate, temperature distribution, turbulence level, and surface tension of the liquid metal.

[0005] As the vapor jet moves with the laser scanning, the particles in the powder layer are always entrained into the metal vapor, thus forming an erosion zone near the scanning track. The particles entrained into the jet hole are blown up from the powder layer and form solid spatter. Some particles that accumulate on the track may partially melt and then adhere to the track surface, thereby deteriorating the surface roughness of the formed workpiece.

[0006] However, the above-mentioned solutions have the following shortcomings: In the LPBF process, when small particles are present in the generated metal vapor, they scatter the laser beam and cause the focus to shift. Both of these factors weaken the energy absorbed by the powder bed. At the same time, the generation of metal vapor will cause changes in the composition and quality of the molten pool, which in turn leads to the formation of pores and defects in the workpiece under the influence of spatter and erosion, reducing the strength and density of the workpiece. However, monitoring these pores requires specialized equipment such as X-rays, CT scans, or thermal imaging, which is costly. The current direction for improvement is to conduct correlation data monitoring and analysis of molten pool state data, laser state data, and atmosphere distribution data in the LPBF process, and to establish a molten pool pore prediction model based on this data. The goal is to adjust printing parameters or control the atmosphere in real time based on the pore prediction results to reduce the pore occurrence rate. Summary of the Invention

[0007] The purpose of this invention is to provide an intelligent 3D printing system for powder materials to solve the problems existing in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a powder material 3D intelligent printing system, comprising:

[0009] Molten pool porosity prediction unit: acquires molten pool state data, atmosphere distribution data and laser state data when different types of porosity are generated in the training database, and generates a molten pool porosity prediction model after data preprocessing, feature engineering, model selection, model training and model validation.

[0010] Data acquisition module: used to collect molten pool state data and atmosphere distribution data in the molten pool area of ​​the laser-printed component, as well as laser state data of the laser head;

[0011] Data processing module: used to receive molten pool state data, atmosphere distribution data and laser state data, and perform analysis and processing to generate molten pool quality monitoring index YCZl, atmosphere distribution index QFFb and laser state index JGZt respectively;

[0012] Output unit: Acquires the molten pool quality monitoring index YCZl, atmosphere distribution index QFFb, and laser state index JGZt and transmits them to the molten pool porosity prediction model. The molten pool porosity prediction model generates the molten pool porosity prediction index KXYc, and adjusts the printing parameters or controls the atmosphere in real time based on the molten pool porosity prediction index KXYc.

[0013] Preferably, the molten pool porosity prediction unit acquires molten pool state data, atmosphere distribution data, and laser state data by collecting a large number of pore experimental data datasets corresponding to different conditions and types. The datasets include metal spatter and metal plume in the molten pool region, as well as control parameters such as laser power, scanning speed, metal vapor parameters, atmosphere flow rate, and composition. A model is established based on a neural network approach. The collected data should cover multiple process conditions and material combinations to ensure the robustness of the model.

[0014] Preferably, the molten pool state data includes molten pool trajectory splash parameters GJFj, molten pool trajectory plume parameters GJYl, and molten pool trajectory erosion parameters GJBs;

[0015] The molten pool trajectory erosion parameter GJBs consists of depth Sd, width Kd, velocity Vd, and temperature Wd. The real-time depth Sd, width Kd, velocity Vd, and temperature Wd of the molten pool are normalized and mapped to the following value range:

[0016] 0≤Sd≤1, 0≤Kd≤1, 0≤Vd≤1, 0≤Wd≤1, the normalization calculation formula is as follows:

[0017]

[0018] Where GY represents the numerical normalization target, Ys represents the original value, max represents the maximum value, min represents the minimum value, and X can be any one of the following: depth Sd, width Kd, velocity Vd, and temperature Wd.

[0019] The following calculation formula for the molten pool trajectory erosion parameter GJBs is further obtained:

[0020] GIBs = Sd GY ×a1+Kd GY ×a2+Vd GY ×a3+Wd GY ×a4+C1

[0021] Where a1, a2, a3, and a4 are weighting coefficients, and 0 < a1 < a2 < a3 < a4, C1 is a correction factor, and the upper and lower thresholds of GJBs are set to GJBs. ma x 、 GJBs min The following molten pool trajectory erosion degree (GJBs) was obtained. cd The calculation formula is as follows:

[0022]

[0023] Among them, the degree of erosion of the molten pool trajectory GJBscd The range of continuous values ​​is limited to 0 to 1, where 0 represents no erosion, 1 represents complete erosion, and the values ​​in between represent different degrees of erosion.

[0024] Preferably, the spatter parameters GJFj of the molten pool trajectory are composed of the number of spatter particles FJSl, the velocity of spatter particles FJSd, and the size of spatter particles FJCc. The number of spatter particles FJSd, the velocity of spatter particles FJSd, and the size of spatter particles FJCc of the molten pool are normalized and mapped to the following value range:

[0025] 0≤FJSl≤1, 0≤FJSd≤1, 0≤FJCc≤1, the normalization calculation formula is as follows:

[0026]

[0027] Where GY represents the numerical normalization target, Ys represents the original value, max represents the maximum value, min represents the minimum value, and X can be any one of the splash parameters GJFj, which can be the number of splashed particles FJSl, the velocity of splashed particles FJSd, or the size of splashed particles FJCc.

[0028] The following formula is obtained for calculating the spatter parameter GJFj of the molten pool trajectory:

[0029] GJFj=FJSl GY ×b1+SJSd GY ×b2+FJCc GY ×b3+C2

[0030] Where b1, b2, and b3 are weighting coefficients, and 0 < b1 < b2 < b3, C2 is a correction factor, and the upper and lower thresholds of GJFj are set as GJFj. max GJFj min The following molten pool trajectory splashing degree GJFj was obtained. cd The calculation formula is as follows:

[0031]

[0032] The degree of splashing in the molten pool trajectory GJFj cd The range of continuous values ​​is limited to 0 to 1, where 0 represents no splashing, 1 represents complete splashing, and the values ​​in between represent different degrees of splashing.

[0033] Preferably, the plume parameter GJYl of the molten pool trajectory is composed of fluidity LDx, temperature gradient WDTd, turbulence intensity CLCs, and surface tension BMZl. The fluidity LDx, temperature gradient WDTd, turbulence intensity CLCs, and surface tension BMZl of the molten pool are normalized and mapped to the following value range:

[0034] 0≤LDx≤1, 0≤WDTd≤1, 0≤CLCs≤1, 0≤BMZl≤1, the normalization calculation formula is as follows:

[0035]

[0036] Where GY represents the numerical normalization target, Ys represents the original value, max represents the maximum value, min represents the minimum value, and X can be any one of the following: fluidity LDx, temperature gradient WDTd, turbulence degree CLCs, and surface tension BMZl.

[0037] The following calculation formula for the plume parameter GJYl of the molten pool trajectory is further obtained:

[0038] GJYl=LDx GY ×d1+WDTd GY ×d2+CLCs GY ×d3+BMZl GY ×d4+C3

[0039] Where d1, d2, d3, and d4 are weighting coefficients, and 0 < d1 < d2 < d3 < d4, C3 is a correction factor, and the upper and lower thresholds of GJYl are set to GJYl respectively. max 、GJYl min The following melt pool trajectory plume degree GJYl was obtained. cd The calculation formula is as follows:

[0040]

[0041] Among them, the degree of plume trajectory of the molten pool is GJYl cd The range of continuous values ​​is limited to 0 to 1, where 0 represents no flow, 1 represents complete flow, and the values ​​in between represent different degrees of flow.

[0042] Preferably, the metal vapor parameter JSZq includes the vapor temperature value ZQWd, the vapor concentration value ZQNd, and the vapor distribution orientation value ZQFb, which are then formalized to obtain the following formula:

[0043] JSZqw=ZQWd×e1+ZSNd×e2+ZQFb×e3+C4

[0044] Where e1, e2, e3, and e4 are weighting coefficients, and 0 < e1 < e2 < e3 < e4, C4 is a correction factor, and the upper and lower thresholds of JSZq are set to JSZq. max 、JsZq min And obtained the following metal vapor degree JSZq cd The calculation formula is as follows:

[0045]

[0046] The degree of metal vapor JSZq cd The continuous value range is limited to 0 to 1, where 0 represents the minimum metal vapor quantity m, 1 represents the maximum vapor quantity M, and intermediate values ​​represent different levels of metal vapor quantity. The values ​​of m and M are adaptively adjusted according to the type and fineness of the metal powder to be processed. The molten pool quality monitoring index YCZl is determined by the degree of splashing along the molten pool trajectory GJFj. cd , Molten pool trajectory plume degree GJYl cd Degree of erosion of molten pool trajectory (GJBs) cd and the degree of metal vapor JSZq cd Obtained through formula processing:

[0047]

[0048] Furthermore, the range of the molten pool quality monitoring index YCZl is -1≤YCZl≤1;

[0049] When -1 ≤ YCZl < 0, this indicates the degree of plume GJYl on the molten pool trajectory. cd Metal vapor level JSZq cd Degree of erosion of molten pool trajectory (GJBs) cd Any one or more combined values ​​in the value range are greater than the spatter intensity of the molten pool trajectory GJFj cd At this point, the molten pool mass depends on the plume intensity of the molten pool trajectory (GJYl). cd Metal vapor level JSZq cd Degree of erosion of molten pool trajectory (GJBs) cd To determine the stability;

[0050] When 0 ≤ YCZl < 0.5, this indicates ① the degree of plume on the molten pool trajectory GJYl. cd Metal vapor level JSZq cd Or the degree of erosion of the molten pool trajectory GJBs cd Any one of them is related to the spatter intensity of the molten pool trajectory GJFj cd The values ​​are the same, and other values ​​are 0; ② Molten pool trajectory plume degree GJYl cd Metal vapor level JSZq cd and the degree of erosion of the molten pool trajectory GJBs cd The numerical combination and the degree of splashing of the molten pool trajectory GJFj cd The values ​​are the same; ③ The degree of splashing on the molten pool trajectory GJFj cd The value is greater than the plume degree of the molten pool trajectory GJYl cd Metal vapor level JSZq cdand the degree of erosion of the molten pool trajectory GJBs cd The sum of the values;

[0051] When 0.5≤YCZl≤1, ④ the degree of splashing on the molten pool trajectory GJFj cd The value is greater than the plume degree of the molten pool trajectory GJYl cd Metal vapor level JSZq cd and the degree of erosion of the molten pool trajectory GJBs cd The values ​​of ④ are equal, and the value of ④ is greater than that of ③.

[0052] Preferably, the atmosphere distribution data includes the atmosphere flow rate parameter QFLs in the molten pool region and the atmosphere composition parameter QFCf in the molten pool region.

[0053] The atmosphere flow rate parameter QFLs is obtained by measuring with a gas velocity sensor or a laser Doppler velocimeter, and the atmosphere composition parameter QFCf in the molten pool region is obtained by measuring with a gas mass spectrometer, an infrared spectrometer, or a gas sensor. After normalization analysis, the following calculation formula for the atmosphere distribution index QFFb is obtained:

[0054] QFFb=QFLs×g1+QFCf×g2

[0055] Where g1 and g2 are weighting coefficients, and 0 < g1 < g2; the upper and lower thresholds of the atmosphere distribution index QFFb are set to QFFb and QFFb respectively. max QFFb min And obtain the following atmosphere distribution degree QFFb cd The calculation formula is as follows:

[0056]

[0057] The degree of atmosphere distribution QFFb cd The range of continuous values ​​is limited to 0 to 1, where 0 indicates normal distribution, 1 indicates uneven distribution, and the values ​​in between represent different degrees of atmospheric distribution.

[0058] Preferably, the laser state data includes laser power parameter JGGl, laser scanning speed parameter JGSm, and laser fluctuation parameter JGBd.

[0059] The laser power parameter JGGl and laser scanning speed parameter JGSm are obtained through the specifications of the corresponding laser head;

[0060] The laser fluctuation parameter JGBd represents the degree of vibration offset of the laser head caused by external vibration or other interference. The value range of the laser fluctuation parameter JGBd is set to be limited to 1 to 10. The larger the value, the greater the offset of the laser head. The degree of offset is related to the offset, deviation, vibration amplitude, vibration frequency and vibration period.

[0061] Normalization analysis and processing of laser power parameter JGGl, laser scanning speed parameter JGSm, and laser fluctuation parameter JGBd yielded the following calculation formula for the laser state index JGZt:

[0062]

[0063] Where θ is the factor, 0≤θ≤1, and the upper and lower thresholds of JGZt are set as JGZt respectively. max 、JGZt min The following laser state level JGZt was obtained. cd The calculation formula is as follows:

[0064]

[0065] Among them, the laser state level JGZt cd The continuous values ​​are limited to the range of 0 to 1, where 0 indicates that the laser is in normal condition, 1 indicates that the laser offset is large, and the values ​​in between indicate different degrees of laser head offset.

[0066] Preferably, upper thresholds (max and min) are set for acquiring the molten pool quality monitoring index YCZl, atmosphere distribution index QFFb, and laser state index JGZt, and correlation analysis is performed to obtain the following formula for calculating the correlation coefficient ρ: JGZt max 、JGZt min

[0067]

[0068] The correlation coefficient ρ is then transmitted to the molten pool porosity prediction model for analysis and processing, and the molten pool porosity prediction index KXYc is obtained. The calculation formula is as follows:

[0069] KXYc=ρ×δ

[0070] Where δ is the revision factor, which is less than or equal to 0 ≤ δ ≤ 1;

[0071] When 0≤KXYc<3, it indicates that the generated pores are micron-sized pores (less than 100 microns), which have little impact on the structural strength of the parts, and there is no need to adjust the printing parameters and control atmosphere;

[0072] When 3≤KXYc<7, it indicates that the generated pores are sub-millimeter level pores (100 micrometers to 1 millimeter), which will have a certain impact on the structural strength of the parts, and fine-tuning of printing parameters and control atmosphere is required;

[0073] When 7≤KXYc≤10, it indicates that the generated pores are millimeter-level pores (greater than 1 mm), which will significantly reduce the structural strength of the parts, and the printing parameters and control atmosphere need to be checked and adjusted.

[0074] Compared with existing technologies, the beneficial effects of this invention are as follows: After data preprocessing, feature engineering, model selection, model training, and model verification, a molten pool porosity prediction model is generated. It receives molten pool state data, atmosphere distribution data, and laser state data, and analyzes and processes them to generate a molten pool quality monitoring index YCZl, an atmosphere distribution index QFFb, and a laser state index JGZt, respectively. These indices are then transmitted to the molten pool porosity prediction model. The molten pool porosity prediction model generates a molten pool porosity prediction index KXYc. Based on this index, printing parameters or atmosphere control are adjusted in real time. Correlation data monitoring and analysis are performed on the molten pool state data, laser state data, and atmosphere distribution data, and this data is used to establish the molten pool porosity prediction model. The printing parameters or atmosphere control are adjusted in real time based on the porosity prediction results to reduce the porosity incidence rate. Attached Figure Description

[0075] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0076] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0077] Please see Figure 1 The present invention provides a technical solution:

[0078] Example 1:

[0079] The material 3D intelligent printing system includes:

[0080] Molten pool porosity prediction unit: acquires molten pool state data, atmosphere distribution data and laser state data when different types of porosity are generated in the training database, and generates a molten pool porosity prediction model after data preprocessing, feature engineering, model selection, model training and model validation.

[0081] Data acquisition module: used to collect molten pool state data and atmosphere distribution data in the molten pool area of ​​the laser-printed component, as well as laser state data of the laser head;

[0082] Data processing module: used to receive molten pool state data, atmosphere distribution data and laser state data, and perform analysis and processing to generate molten pool quality monitoring index YCZl, atmosphere distribution index QFFb and laser state index JGZt respectively;

[0083] Output unit: Acquires the molten pool quality monitoring index YCZl, atmosphere distribution index QFFb, and laser state index JGZt and transmits them to the molten pool porosity prediction model. The molten pool porosity prediction model generates the molten pool porosity prediction index KXYc, and adjusts the printing parameters or controls the atmosphere in real time based on the molten pool porosity prediction index KXYc.

[0084] Example 2:

[0085] Based on Example 1, the molten pool porosity prediction unit acquires molten pool state data, atmosphere distribution data, and laser state data by collecting a large number of pore experimental data corresponding to different conditions and types. The dataset includes metal spatter and metal plume in the molten pool region, as well as control parameters such as laser power, scanning speed, metal vapor parameters, atmosphere flow rate and composition, and establishes a model based on a neural network.

[0086] The data should cover multiple process conditions and material combinations to ensure the robustness of the model. The processing steps are as follows:

[0087] Data preprocessing:

[0088] Data is cleaned, normalized, and outliers are removed to ensure data quality and consistency;

[0089] Feature engineering:

[0090] Features related to the molten pool were extracted from the data, including molten pool temperature gradient, molten pool trajectory plume, molten pool trajectory spatter, molten pool trajectory erosion, laser power variation, laser scanning speed variation, laser fluctuation variation, atmosphere composition distribution, and atmosphere flow rate distribution characteristics, in order to capture the relationship with molten pool region parameters.

[0091] Model selection:

[0092] Choose either a machine learning model or a mathematical model to build a steam molten pool prediction model. The model includes neural networks, regression analysis, physical modeling, etc. In this embodiment, a neural network method is chosen to build the model.

[0093] Data splitting: The dataset is divided into training set, validation set and test set, with 70%-80% of the data used as training set, 10%-15% of the data used as validation set and 10%-15% of the data used as test set;

[0094] Choosing a neural network architecture: Select an appropriate neural network architecture based on the complexity of the problem and the characteristics of the data. For the problem with multiple inputs and multiple outputs in this embodiment, a multiple input multiple output (MIMO) neural network is used.

[0095] Constructing a neural network model: Designing a neural network structure, including an input layer, hidden layers, and an output layer. For multi-input multi-output problems, a multi-branch neural network is used, with each branch processing one input feature, such as the branch features in this embodiment concerning molten pool state data, laser state data, and atmosphere distribution data.

[0096] Loss function and optimizer: Choose an appropriate loss function, such as the mean squared error (MSE) in the regression problem, and choose a suitable optimizer, such as stochastic gradient descent (SGD) or an adaptive optimizer (such as Adam).

[0097] Model training: The neural network model is trained using a training set. In each training iteration, the model weights are updated using the backpropagation algorithm to reduce the value of the loss function.

[0098] Model validation and tuning: Use a validation set to monitor the model's performance to prevent overfitting. Try different hyperparameter settings and model structures to improve performance.

[0099] Model evaluation: The performance of the final model is evaluated using a test set, using various evaluation metrics such as mean squared error, mean absolute error, and coefficient of determination.

[0100] Deployment and use: Once the model is trained and evaluated, it can be deployed to a real laser printing system for real-time prediction of the properties of the molten pool porosity.

[0101] Example 3:

[0102] Based on Example 1, the molten pool state data includes molten pool trajectory splash parameters GJFj, molten pool trajectory plume parameters GJYl, and molten pool trajectory erosion parameters GJBs.

[0103] The molten pool trajectory erosion parameter GJBs consists of depth Sd, width Kd, velocity Vd, and temperature Wd. The real-time depth Sd, width Kd, velocity Vd, and temperature Wd of the molten pool are normalized and mapped to the following value range:

[0104] 0≤Sd≤1, 0≤Kd≤1, 0≤Vd≤1, 0≤Wd≤1, the normalization calculation formula is as follows:

[0105]

[0106] Where GY represents the numerical normalization target, Ys represents the original value, max represents the maximum value, min represents the minimum value, and x can be any one of the depth value Sd, width value Kd, velocity value Vd, and temperature value Wd;

[0107] The following calculation formula for the molten pool trajectory erosion parameter GJBs is further obtained:

[0108] GJBs = Sd GY ×a1+Kd GY ×a2+Vd GY ×a3+Wd GY ×a4+C1

[0109] Where a1, a2, a3, and a4 are weighting coefficients, and 0 < a1 < a2 < a3 < a4, C1 is a correction factor, and the upper and lower thresholds of GJBs are set to GJBs. max GJBs min The following molten pool trajectory erosion degree (GJBs) was obtained. cd The calculation formula is as follows:

[0110]

[0111] Among them, the degree of erosion of the molten pool trajectory GJBs cd The range of continuous values ​​is limited to 0 to 1, where 0 represents no erosion, 1 represents complete erosion, and the values ​​in between represent different degrees of erosion.

[0112] The data collection steps are as follows:

[0113] Experimental observation: The 3D laser printing process of metal powder materials is observed through experiments, and the formation process of the molten pool and the erosion of the molten pool surface are recorded. This is achieved by using a high-speed telescope, laser scanner or other related equipment.

[0114] Image analysis: Image processing technology is used to analyze images or videos during the laser printing process to extract molten pool trajectory erosion parameters, and computer vision technology is used to identify information such as the shape, size, and location of the molten pool;

[0115] Simulation and modeling: Numerical simulation or computational fluid dynamics (CFD) models are used to simulate the 3D laser printing process of metal powder materials, obtain the molten pool trajectory erosion parameters, and obtain the molten pool depth Sd, width Kd, velocity Vd, and temperature Wd parameters;

[0116] Data processing and analysis: The collected parameters of molten pool depth Sd, width Kd, velocity Vd, and temperature Wd are processed and analyzed using statistical analysis, image processing techniques, and pattern recognition algorithms.

[0117] Experimental design and optimization: By continuously adjusting laser printing parameters and material properties, new experimental schemes are designed to obtain more accurate and comprehensive molten pool trajectory erosion parameters. This requires multiple experiments and optimization processes.

[0118] Example 4:

[0119] Based on Example 3, the spatter parameter GJFj of the molten pool trajectory is composed of the number of spatter particles FJSl, the velocity of spatter particles FJSd, and the size of spatter particles FJCc. The number of spatter particles FJSd, the velocity of spatter particles FJSd, and the size of spatter particles FJCc of the molten pool are normalized and mapped to the following value range:

[0120] 0≤FJSl≤1, 0≤FJSd≤1, 0≤FJCc≤1, the normalization calculation formula is as follows:

[0121]

[0122] Where GY represents the numerical normalization target, Ys represents the original value, max represents the maximum value, min represents the minimum value, and X can be any one of the splash parameters GJFj, which can be the number of splashed particles FJSl, the velocity of splashed particles FJSd, or the size of splashed particles FJCc.

[0123] The following formula is obtained for calculating the spatter parameter GJFj of the molten pool trajectory:

[0124] GJFj=FJSl GY ×b1+SJSd GY ×b2+FJCc GY ×b3+C2

[0125] Where b1, b2, and b3 are weighting coefficients, and 0 < b1 < b2 < b3, C2 is a correction factor, and the upper and lower thresholds of GJFj are set as GJFj. max GJFj min The following molten pool trajectory splashing degree GJFj was obtained. cd The calculation formula is as follows:

[0126]

[0127] The degree of splashing in the molten pool trajectory GJFj cd The range of continuous values ​​is limited to 0 to 1, where 0 represents no splashing, 1 represents complete splashing, and the values ​​in between represent different degrees of splashing.

[0128] The specific steps are as follows:

[0129] Data collection and processing: Obtain parameters such as the number, velocity, and size of splashing particles through sensors, cameras, or other monitoring devices.

[0130] Data representation: Each parameter value is represented as a data point, and these data points constitute a dataset, where each data point corresponds to a sampling or time step in the printing process.

[0131] Statistical analysis: Using statistical analysis methods, such as correlation analysis, to study the relationship between different parameters. Correlation is usually measured using correlation coefficients (such as Pearson correlation coefficient) to determine the degree of correlation between parameters.

[0132] Calculation formula: The correlation calculation formula is used to quantify this relationship, and quantitative results about the relationship between parameters are obtained. These results can be used to optimize the laser printing process, improve printing parameter settings, or predict the possibility of spatter.

[0133] Example 5:

[0134] Based on Example 3, the plume parameter GJYl of the molten pool trajectory is composed of fluidity LDx, temperature gradient WDTd, turbulence intensity CLCs, and surface tension BMZl. The fluidity LDx, temperature gradient WDTd, turbulence intensity CLCs, and surface tension BMZl of the molten pool are normalized and mapped to the following value range:

[0135] 0≤LDx≤1, 0≤WDTd≤1, 0≤CLCs≤1, 0≤BMZl≤1, the normalization calculation formula is as follows:

[0136]

[0137] Where GY represents the numerical normalization target, Ys represents the original value, max represents the maximum value, min represents the minimum value, and X can be any one of the following: fluidity LDx, temperature gradient WDTd, turbulence degree CLCs, and surface tension BMZl.

[0138] The following calculation formula for the plume parameter GJYl of the molten pool trajectory is further obtained:

[0139] GJYl=LDx GY ×d1+WDTd GY ×d2+CLCs GY ×d3+BMZl GY ×d4+C3

[0140] Where d1, d2, d3, and d4 are weighting coefficients, and 0 < d1 < d2 < d3 < d4, C3 is a correction factor, and the upper and lower thresholds of GJYl are set to GJYl respectively. max、GJYl min The following melt pool trajectory plume degree GJYl was obtained. cd The calculation formula is as follows:

[0141]

[0142] Among them, the degree of plume trajectory of the molten pool is GJYl cd The range of continuous values ​​is limited to 0 to 1, where 0 represents no flow, 1 represents complete flow, and the values ​​in between represent different degrees of flow.

[0143] Specifically as follows:

[0144] Flow velocity and direction (fluidity): refers to the speed and direction of flow of molten metal within a molten pool. Laser irradiation and energy transfer cause metal powder to melt and form molten metal in a molten pool. Its flow velocity and direction depend on various factors such as laser irradiation and material properties.

[0145] Temperature gradient: The temperature gradient change in different regions of the molten pool. Due to the high temperature irradiation of the laser, the metal powder melts to form a molten pool, and the temperature gradient in the molten pool affects the liquid flow characteristics of the metal.

[0146] Turbulence intensity: The intensity or degree of turbulence in the molten metal pool. This refers to the turbulent characteristics that may exist in the flow of liquid metal, and has a certain impact on printing quality and physical properties.

[0147] Surface tension: The surface tension of molten metal in a molten pool refers to the tension generated on the surface of the liquid metal due to intermolecular forces, which also has a certain impact on the flow characteristics inside the molten pool.

[0148] Example 6:

[0149] Based on Example 5, the metal vapor parameter JSZq includes the vapor temperature value ZQWd, the vapor concentration value ZQNd, and the vapor distribution orientation value ZQFb, which are then formalized to obtain the following formula:

[0150] JSZqw=ZQWd×e1+ZSNd×e2+ZQFb×e3+C4

[0151] Where e1, e2, e3, and e4 are weighting coefficients, and 0 < e1 < e2 < e3 < e4, C4 is a correction factor, and the upper and lower thresholds of JSZq are set to JSZq. max 、JsZq min And obtained the following metal vapor degree JSZq cd The calculation formula is as follows:

[0152]

[0153] The degree of metal vapor JsZq cd The continuous values ​​are limited to the range of 0 to 1, where 0 represents the minimum metal vapor quantity m, 1 represents the maximum vapor quantity M, and the intermediate values ​​represent different levels of metal vapor quantity. The values ​​of m and M are adaptively adjusted according to the type and fineness of the metal powder to be processed. The specific influencing factors are as follows:

[0154] Metal powder type: Different metal powders produce different amounts of vapor when melted. Therefore, some types of metal powders will directly affect the upper and lower limits of metal vapor. Metals may begin to evaporate at lower temperatures, while others require higher temperatures.

[0155] Process parameters: Parameters during the printing process, such as laser power, scanning speed, and layer height, directly affect the amount of metal vapor generated. Therefore, it is necessary to determine appropriate upper and lower thresholds based on the process parameters used.

[0156] Atmosphere: Using an inert atmosphere (usually a suppressing gas) will affect the behavior of metal vapors. The type and concentration of gases in the atmosphere will affect the release of metal vapors.

[0157] The molten pool quality monitoring index YCZl is determined by the degree of splashing along the molten pool trajectory, GJFj. cd , Molten pool trajectory plume degree GJYl cd Degree of erosion of molten pool trajectory (GJBs) cd and the degree of metal vapor JSZq cd Obtained through formula processing:

[0158]

[0159] Furthermore, the range of the molten pool quality monitoring index YCZl is -1≤YCZl≤1;

[0160] When -1 ≤ YCZl < 0, this indicates the degree of plume GJYl on the molten pool trajectory. cd Metal vapor level JSZq cd Degree of erosion of molten pool trajectory (GJBs) cd Any one or more combined values ​​in the value range are greater than the spatter intensity of the molten pool trajectory GJFj cd At this point, the molten pool mass depends on the plume intensity of the molten pool trajectory (GJYl). cd Metal vapor level JSZq cd Degree of erosion of molten pool trajectory (GJBs) cd To determine the stability;

[0161] When 0 ≤ YCZl < 0.5, this indicates ① the degree of plume on the molten pool trajectory GJYl. cd Metal vapor level JSZqcd Or the degree of erosion of the molten pool trajectory GJBs cd Any one of them is related to the spatter intensity of the molten pool trajectory GJFj cd The values ​​are the same, and other values ​​are 0; ② Molten pool trajectory plume degree GJYl cd Metal vapor level JSZq cd and the degree of erosion of the molten pool trajectory GJBs cd The numerical combination and the degree of splashing of the molten pool trajectory GJFj cd The values ​​are the same; ③ The degree of splashing on the molten pool trajectory GJFj cd The value is greater than the plume degree of the molten pool trajectory GJYl cd Metal vapor level JSZq cd CJBs degree of erosion along the molten pool trajectory cd The sum of the values;

[0162] When 0.5≤YCZl≤1, ④ the degree of splashing on the molten pool trajectory GJFj cd The value is greater than the plume degree of the molten pool trajectory GJYl cd Metal vapor level JSZq cd and the degree of erosion of the molten pool trajectory GJBs cd The values ​​of ④ are equal, and the value of ④ is greater than that of ③.

[0163] Example 7:

[0164] The atmosphere distribution data includes the atmosphere flow rate parameter QFLs and the atmosphere composition parameter QFCf in the molten pool region.

[0165] The atmosphere flow rate parameter QFLs is obtained by measuring with a gas velocity sensor or a laser Doppler velocimeter, and the atmosphere composition parameter QFCf in the molten pool region is obtained by measuring with a gas mass spectrometer, an infrared spectrometer, or a gas sensor. After normalization analysis, the following calculation formula for the atmosphere distribution index QFFb is obtained:

[0166] QFFb=QFLs×g1+QFCf×g2

[0167] Where g1 and g2 are weighting coefficients, and 0 < g1 < g2; the upper and lower thresholds of the atmosphere distribution index QFFb are set to QFFb and QFFb respectively. max QFFb min And obtain the following atmosphere distribution degree QFFb cd The calculation formula is as follows:

[0168]

[0169] The degree of atmosphere distribution QFFb cdThe range of continuous values ​​is limited to 0 to 1, where 0 indicates normal distribution, 1 indicates uneven distribution, and the values ​​in between indicate different degrees of atmosphere distribution.

[0170] Gas flow rate measurement: Measuring the gas flow rate in the molten pool area can typically be accomplished using a gas flow meter, turbine flow meter, or gas velocity sensor. These sensors monitor the speed and direction of gas flow within the printing chamber. This data can be used to understand the gas flow during the printing process.

[0171] Atmosphere composition analysis: Understanding the atmosphere composition in the molten pool region typically requires the use of gas analysis instruments, such as gas mass spectrometers, infrared spectrometers, or gas sensors. These instruments can measure the concentration of various components in the gas, including inert gases (such as nitrogen or argon) and any other gases that may be present.

[0172] Turbine flow meter: A turbine flow meter is a sensor used to measure the flow rate of gas. It is typically used to monitor the flow rate of gas in pipes or channels. In 3D printing, they can be used to monitor the flow rate of atmospheric gases.

[0173] Optical methods: Optical methods can be used to monitor particles in a gas or the flow of particles. For example, a laser Doppler velocimeter can be used to measure the velocity of particles in an atmosphere.

[0174] Example 8:

[0175] The laser status data includes laser power parameter JGGl, laser scanning speed parameter JGSm, and laser fluctuation parameter JGBd.

[0176] The laser power parameter JGGl and laser scanning speed parameter JGSm are obtained through the specifications of the corresponding laser head;

[0177] Laser power parameters: Laser power is the energy output of a laser beam, usually measured in watts (W). Laser power is typically measured using power sensors or power meters. These devices measure the energy passing through the laser beam, thus determining the laser's power. This measurement is usually performed at the laser's output or at a specific location in the laser's optical path.

[0178] Laser scanning speed parameter: Laser scanning speed refers to the speed at which the laser spot moves across the working surface, usually measured in millimeters per second (mm / s). Laser scanning speed is typically measured using encoders or position sensors within a motion control system. These sensors measure the motion of the laser scanning system and provide speed information in the form of digital signals.

[0179] The laser fluctuation parameter JGBd represents the degree of vibration offset of the laser head caused by external vibration or other interference. The value range of the laser fluctuation parameter JGBd is set to be limited to 1 to 10. The larger the value, the greater the offset of the laser head. The degree of offset is related to the offset, deviation, vibration amplitude, vibration frequency and vibration period.

[0180] Offset: Offset usually represents the deviation of the laser head position from its original position. This can be represented by displacement values ​​in a coordinate system, such as X, Y and Z offsets, which are usually measured in millimeters or micrometers.

[0181] Deviation: Deviation refers to the distance between the actual position of the laser head and its ideal position. It is usually used to describe the magnitude of the deviation and can be expressed in distance units such as millimeters or micrometers.

[0182] Vibration amplitude: The offset caused by the vibration of the laser head can be represented by the vibration amplitude, which refers to the peak offset of the vibration, usually in millimeters or micrometers.

[0183] Vibration frequency: If the laser head is affected by periodic vibration, the vibration frequency represents the frequency of the vibration, usually in Hertz (Hz).

[0184] Vibration period: The vibration period is the reciprocal of the vibration frequency, representing the time required for the vibration to complete one cycle, usually measured in seconds.

[0185] Normalization analysis and processing of laser power parameter JGGl, laser scanning speed parameter JGSm, and laser fluctuation parameter JGBd yielded the following calculation formula for the laser state index JGZt:

[0186]

[0187] Where θ is the factor, 0≤θ≤1, and the upper and lower thresholds of JGZt are set as JGZt respectively. max 、JGZt min The following laser state level JGZt was obtained. cd The calculation formula is as follows:

[0188]

[0189] Among them, the laser state level JGZt cd The continuous values ​​are limited to the range of 0 to 1, where 0 indicates that the laser is in normal condition, 1 indicates that the laser offset is large, and the values ​​in between indicate different degrees of laser head offset.

[0190] Example 9:

[0191] The upper limits (max and min) of the thresholds for acquiring the molten pool quality monitoring index YCZl, the atmosphere distribution index QFFb, and the laser state index JGZt are set, and correlation analysis is performed to obtain the following formula for calculating the correlation coefficient ρ: JGZt max 、JGZt min

[0192]

[0193] The correlation coefficient ρ is then transmitted to the molten pool porosity prediction model for analysis and processing, and the molten pool porosity prediction index KXYc is obtained. The calculation formula is as follows:

[0194] KXYc=ρ×δ

[0195] Where δ is the revision factor, which is less than or equal to 0 ≤ δ ≤ 1;

[0196] When 0≤KXYc<3, it indicates that the generated pores are micron-sized pores (less than 100 microns), which have little impact on the structural strength of the parts, and there is no need to adjust the printing parameters and control atmosphere;

[0197] When 3≤KXYc<7, it indicates that the generated pores are sub-millimeter level pores (100 micrometers to 1 millimeter), which will have a certain impact on the structural strength of the parts, and fine-tuning of printing parameters and control atmosphere is required;

[0198] When 7≤KXYc≤10, it indicates that the generated pores are millimeter-level pores (greater than 1 mm), which will significantly reduce the structural strength of the parts, and the printing parameters and control atmosphere need to be checked and adjusted.

[0199] Specifically as follows:

[0200] Micrometer-scale pores (less than 100 micrometers): Micrometer-scale pores generally have little impact on the structural strength of parts, especially in high-strength metallic materials. These small pores may reduce the theoretical ultimate strength of parts, but usually do not cause a significant decrease in mechanical properties during actual use. Pores of this size may affect the fatigue life of materials, but usually do not cause mechanical fracture of parts.

[0201] Submillimeter-scale porosity (100 micrometers to 1 millimeter): Submillimeter-scale porosity can affect the structural strength of parts, especially under high loads or high stresses. These porosities can reduce the bending and compressive strength of parts, particularly when they are unevenly distributed or located in critical stress concentration areas.

[0202] Millimeter-sized pores (greater than 1 mm): Millimeter-sized pores typically significantly reduce the structural strength of parts, especially when they are located in critical load transfer paths or stress concentration areas. Pores of this size can cause a sharp decline in the mechanical properties of parts, and may even lead to mechanical fracture.

[0203] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0204] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0205] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0206] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division of a waterway underwater topography change analysis system and method. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms.

[0207] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0208] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0209] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0210] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A powder material 3D intelligent printing system, characterized in that, include: Molten pool porosity prediction unit: acquires molten pool state data, atmosphere distribution data and laser state data when different types of porosity are generated in the training database, and generates a molten pool porosity prediction model after data preprocessing, feature engineering, model selection, model training and model validation. Data acquisition module: used to collect molten pool state data and atmosphere distribution data in the molten pool area of ​​the laser-printed component, as well as laser state data of the laser head; Data processing module: used to receive molten pool state data, atmosphere distribution data and laser state data, and perform analysis and processing to generate molten pool quality monitoring index YCZl, atmosphere distribution index QFFb and laser state index JGZt respectively; Output unit: acquires the molten pool quality monitoring index YCZl, atmosphere distribution index QFFb, and laser state index JGZt and transmits them to the molten pool porosity prediction model. The molten pool porosity prediction model generates the molten pool porosity prediction index KXYc. Based on the molten pool porosity prediction index KXYc, the printing parameters are adjusted or the atmosphere is controlled in real time. Specifically, the molten pool state data, atmosphere distribution data, and laser state data acquired by the molten pool porosity prediction unit are obtained by collecting a large number of experimental data sets corresponding to different conditions and types of porosity. The datasets include metal spatter and metal plume in the molten pool region, as well as control parameters such as laser power, scanning speed, metal vapor parameters, atmosphere flow rate and composition. A model is established based on a neural network approach. The collected data should cover multiple process conditions and material combinations to ensure the robustness of the model. The molten pool state data includes molten pool trajectory splash parameters GJFj, molten pool trajectory plume parameters GJYl, and molten pool trajectory erosion parameters GJBs; The molten pool trajectory erosion parameter GJBs consists of depth Sd, width Kd, velocity Vd, and temperature Wd. The real-time depth Sd, width Kd, velocity Vd, and temperature Wd of the molten pool are normalized and mapped to the following value range: 0≤Sd≤1, 0≤Kd≤1, 0≤Vd≤1, 0≤Wd≤1, the normalization calculation formula is as follows: Where GY represents the numerical normalization target, Ys represents the original value, max represents the maximum value, min represents the minimum value, and X can be any one of the following: depth Sd, width Kd, velocity Vd, and temperature Wd. The following calculation formula for the molten pool trajectory erosion parameter GJBs is further obtained: Where a1, a2, a3, and a4 are weighting coefficients, and 0 < a1 < a2 < a3 < a4, C1 is a correction factor, and the upper and lower thresholds of GJBs are set as follows: , The following erosion degree of the molten pool trajectory was obtained. The calculation formula is as follows: The degree of erosion of the molten pool trajectory The range of continuous values ​​is limited to 0 to 1, where 0 represents no erosion, 1 represents complete erosion, and the values ​​in between represent different degrees of erosion.

2. The intelligent 3D printing system for powder materials according to claim 1, characterized in that: The spatter parameters GJFj of the molten pool trajectory are composed of the number of spatter particles FJSl, the velocity of spatter particles FJSd, and the size of spatter particles FJCc. The number of spatter particles FJSd, the velocity of spatter particles FJSd, and the size of spatter particles FJCc of the molten pool are normalized and mapped to the following value range: 0≤FJSl≤1, 0≤FJSd≤1, 0≤FJCc≤1, the normalization calculation formula is as follows: Where GY represents the numerical normalization target, Ys represents the original value, max represents the maximum value, min represents the minimum value, and X can be any one of the splash parameters GJFj, which can be the number of splashed particles FJSl, the velocity of splashed particles FJSd, or the size of splashed particles FJCc. The following formula is obtained for calculating the spatter parameter GJFj of the molten pool trajectory: Where b1, b2, and b3 are weighting coefficients, and 0 < b1 < b2 < b3, C2 is a correction factor, and the upper and lower thresholds of GJFj are set as follows: , The following spatter patterns and splash levels in the molten pool were obtained. The calculation formula is as follows: The degree of splashing in the molten pool trajectory The range of continuous values ​​is limited to 0 to 1, where 0 represents no splashing, 1 represents complete splashing, and the values ​​in between represent different degrees of splashing.

3. The intelligent 3D printing system for powder materials according to claim 1, characterized in that: The plume parameter GJYl of the molten pool trajectory is composed of fluidity LDx, temperature gradient WDTd, turbulence intensity CLCs, and surface tension BMZl. The fluidity LDx, temperature gradient WDTd, turbulence intensity CLCs, and surface tension BMZl of the molten pool are normalized and mapped to the following value range: 0≤LDx≤1, 0≤WDTd≤1, 0≤CLCs≤1, 0≤BMZl≤1, the normalization calculation formula is as follows: Where GY represents the numerical normalization target, Ys represents the original value, max represents the maximum value, min represents the minimum value, and X can be any one of the following: fluidity LDx, temperature gradient WDTd, turbulence degree CLCs, and surface tension BMZl. The following calculation formula for the plume parameter GJYl of the molten pool trajectory is further obtained: Where d1, d2, d3, and d4 are weighting coefficients, and 0 < d1 < d2 < d3 < d4, C3 is a correction factor, and the upper and lower thresholds of GJYl are set as follows: , The following melt pool trajectory plume degree was obtained. The calculation formula is as follows: The degree of plume trajectory of the molten pool The range of continuous values ​​is limited to 0 to 1, where 0 represents no flow, 1 represents complete flow, and the values ​​in between represent different degrees of flow.

4. The intelligent 3D printing system for powder materials according to claim 3, characterized in that: The metal vapor parameter JSZq includes the vapor temperature value ZQWd, the vapor concentration value ZQNd, and the vapor distribution orientation value ZQFb, which are then formalized to obtain the following formula: Where e1, e2, e3, and e4 are weighting coefficients, and 0 < e1 < e2 < e3 < e4, C4 is a correction factor, and the upper and lower thresholds of JSZq are set as follows: , And obtain the following metal vapor degree The calculation formula is as follows: The degree of metal vapor The continuous value range is limited to 0 to 1, where 0 represents the minimum metal vapor quantity m, 1 represents the maximum vapor quantity M, and intermediate values ​​represent different levels of metal vapor quantity. The values ​​of m and M are adaptively adjusted according to the type and fineness of the metal powder to be processed. The molten pool quality monitoring index YCZl is determined by the degree of splashing along the molten pool trajectory. , degree of plume trajectory of molten pool Degree of erosion of the molten pool trajectory and the degree of metal vapor Obtained through formula processing: Furthermore, the range of the molten pool quality monitoring index YCZl is -1≤YCZl≤1; When -1 ≤ YCZl < 0, this indicates the degree of plume on the molten pool trajectory. Metal vapor degree Degree of erosion of the molten pool trajectory Any one or more combined values ​​in the range are greater than the spatter intensity of the molten pool trajectory. At this point, the molten pool mass depends on the plume thickness of the molten pool trajectory. Metal vapor degree Degree of erosion of the molten pool trajectory To determine the stability; When 0 ≤ YCZl < 0.5, this indicates ① the degree of plume on the molten pool trajectory. Metal vapor degree Or the degree of erosion of the molten pool trajectory Any one of them is related to the degree of splashing along the trajectory of the molten pool. The values ​​are the same, and all other values ​​are 0; ② Degree of plume trajectory of molten pool Metal vapor degree and the degree of erosion of the molten pool trajectory The numerical combination and the degree of splashing of the molten pool trajectory The values ​​are the same; ③ The degree of splashing in the molten pool trajectory The value is greater than the plume intensity of the molten pool trajectory. Metal vapor degree and the degree of erosion of the molten pool trajectory The sum of the values; When 0.5≤YCZl≤1, ④ the degree of splashing on the molten pool trajectory. The value is greater than the plume intensity of the molten pool trajectory. Metal vapor degree and the degree of erosion of the molten pool trajectory The values ​​of ④ are equal, and the value of ④ is greater than that of ③.

5. The intelligent 3D printing system for powder materials according to claim 1, characterized in that: The atmosphere distribution data includes the atmosphere flow rate parameter QFLs and the atmosphere composition parameter QFCf in the molten pool region. The atmosphere flow rate parameter QFLs is obtained by measuring with a gas velocity sensor or a laser Doppler velocimeter, and the atmosphere composition parameter QFCf in the molten pool region is obtained by measuring with a gas mass spectrometer, an infrared spectrometer, or a gas sensor. After normalization analysis, the following calculation formula for the atmosphere distribution index QFFb is obtained: Where g1 and g2 are weighting coefficients, and 0 < g1 < g2; the upper and lower thresholds of the atmosphere distribution index QFFb are set as follows: , The following atmosphere distribution levels were obtained. The calculation formula is as follows: The degree of atmospheric distribution The range of continuous values ​​is limited to 0 to 1, where 0 indicates normal distribution, 1 indicates uneven distribution, and the values ​​in between represent different degrees of atmospheric distribution.

6. The intelligent 3D printing system for powder materials according to claim 1, characterized in that: The laser status data includes laser power parameter JGGl, laser scanning speed parameter JGSm, and laser fluctuation parameter JGBd. The laser power parameter JGGl and laser scanning speed parameter JGSm are obtained through the specifications of the corresponding laser head; The laser fluctuation parameter JGBd represents the degree of vibration offset of the laser head caused by external vibration or other interference. The value range of the laser fluctuation parameter JGBd is set to be limited to 1 to 10. The larger the value, the greater the offset of the laser head. The degree of offset is related to the offset, deviation, vibration amplitude, vibration frequency and vibration period. Normalization analysis and processing of laser power parameter JGGl, laser scanning speed parameter JGSm, and laser fluctuation parameter JGBd yielded the following calculation formula for the laser state index JGZt: in For factors, Set the upper and lower thresholds of JGZt as follows: , And obtain the following laser state degree The calculation formula is as follows: The degree of laser status The continuous values ​​are limited to the range of 0 to 1, where 0 indicates that the laser is in normal condition, 1 indicates that the laser offset is large, and the values ​​in between indicate different degrees of laser head offset.

7. The intelligent 3D printing system for powder materials according to any one of claims 1-6, characterized in that: The upper limits (max and min) of the thresholds for acquiring the molten pool quality monitoring index YCZl, the atmosphere distribution index QFFb, and the laser state index JGZt were set, and correlation analysis was performed to obtain the following correlation coefficients. The calculation formula is as follows: , and correlation coefficient The data is transmitted to the molten pool porosity prediction model for analysis and processing, and the molten pool porosity prediction index KXYc is obtained. The calculation formula is as follows: in The revision factor is less than or equal to 0. ≤1; when When the value is set to 1, it indicates that the generated pores are micron-sized (less than 100 microns), which has little impact on the structural strength of the part and does not require adjustment of printing parameters and control atmosphere; when When the pore size is sub-millimeter level (100 micrometers to 1 millimeter), it indicates that the generated pores are sub-millimeter level pores (100 micrometers to 1 millimeter), which has a certain impact on the structural strength of the part and requires fine-tuning of the printing parameters and control atmosphere; when When the generated pores are millimeter-level pores (greater than 1 mm), the structural strength of the part will be significantly reduced, and the printing parameters and control atmosphere need to be checked and adjusted.

Images