Method for predicting pore defects of additive manufacturing formed part by considering melt channel deviation

Abstract

The invention is applicable to the field of metal additive manufacturing, and provides an additive manufacturing formed part pore defect prediction method considering melt channel deviation, which can comprehensively consider the influence of laser power, scanning speed, scanning spacing, powder laying thickness and interlayer rotation angle on pore defects. Porous defects possibly appearing in the sample under given process parameters can be rapidly, visually and accurately predicted, and a forming process scheme for preparing the high-density sample can be rapidly predicted. The key technology of the method is that the influence of the melt channel deviation on the pore defects in the forming process is considered, the generation of the pore defects caused by insufficient melt channel overlapping can be effectively predicted, the accuracy of density prediction is improved, and the repeatability and stability of an additive manufacturing sample are enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of metal additive manufacturing, and in particular relates to a method for predicting pore defects of additively manufactured parts taking into account melt path deviation. Background Art

[0002] As a common metal additive manufacturing process, laser powder bed fusion (LPBF) technology is capable of manufacturing products with complex geometries; however, the complex layer-by-layer manufacturing process has many uncertainties, resulting in poor quality stability of LPBF-manufactured parts, which poses a challenge to the repeated manufacture of high-quality products in large-scale production and limits the widespread application of this technology in industry; among them, pore defects are the main reason for the poor repeatability and stability of specimens.

[0003] At present, researchers mainly use the molten pool size to judge the possible defects in the material. For example, the ratio of the molten pool depth and the powder thickness is used to judge the unfused defects caused by spheroidization and discontinuity, and the ratio of the molten pool depth and the molten pool width is used to judge the porosity defects caused by the keyhole effect.

[0004] The above method predicts possible porosity defects based on the size of the molten pool, which has certain practical value. However, during the LPBF forming process, the sample relies on the molten pool to form a dense part, which is not only related to the molten pool size, but also closely related to the molten pool position. In fact, during the LPBF forming process, the molten path will shift left and right along the laser scanning direction, such as Figure 1 As shown in Figure 3, this will result in the failure to form a good metallurgical bond between the melt channels, which will inevitably introduce additional porosity defects in the formed specimen.

[0005] It should be noted that the melt path refers to the area formed after laser scanning, and the cross-section of the melt path is called the molten pool. This type of melt path deviation is affected by many factors, such as laser power, scanning speed and powder distribution. By increasing the heat input, the melt path deviation can be alleviated to a certain extent, but due to the random distribution of powder particles, the melt path deviation cannot be completely eliminated, which will inevitably affect the accuracy of the existing pore prediction model. How to establish a fast and accurate pore defect prediction model taking into account the melt path deviation is a key issue that needs to be urgently solved to improve the stability of LPBF prepared samples. Summary of the invention

[0006] In view of the above-mentioned deficiencies in the prior art, an object of the present invention is to provide a method for predicting porosity defects of additively manufactured parts taking into account melt path deviation. The prediction model established by this method takes into account the influence of melt path deviation on porosity defects, thereby improving the prediction accuracy.

[0007] To achieve the above purpose, the present invention adopts the following technical route:

[0008] Step 1: Calculation of molten pool size; analytical calculation is used to quickly predict the molten pool width and depth under different laser powers and scanning speeds, where the laser value range is 100-400W, the scanning speed range is 200-1200mm / s, and the laser line energy density is between 0.2-2J / mm; under this process condition, it is ensured that there is no spheroidization and discontinuity in the melt channel, as well as keyhole porosity defects. At this time, the molten pool is a typical semi-elliptical morphology. The following formula is used to calculate the molten pool width and depth:

[0009]

[0010] Where: a is the molten pool width, α is the laser absorptivity; P is the laser power, ρ s is the material density, c s and c l is the specific heat capacity of the material in solid and liquid states, T0 is the ambient temperature, T m is the melting temperature of the metal, v is the scanning speed, u is the laser diameter, κ is the thermal conductivity, K is the temperature gradient on the solid-liquid interface, h is the convection heat transfer coefficient, ε is the material emissivity, σ b is the Boltzmann constant, L is the latent heat of fusion;

[0011] Step 2: Calculation of the cross-sectional dimensions of the melt channel; During LPBF forming, the laser inputs energy into the molten pool, causing the powder particles to melt and solidify to form a new forming zone; at the same time, this process will also cause the remelting of the solidified material in the previous layer. During the solidification stage, the newly deposited material will form a "small hat" shape on the previously deposited material under the action of surface tension. Based on the observation of the morphological characteristics of the melt channel, it is assumed that the remelting zone and the forming zone in the forming process are both semi-elliptical in shape, where the horizontal major axis length of the remelting zone ellipse is the melt channel width a, and the vertical major axis length d is the melt pool depth. The horizontal major axis length of the forming zone ellipse is the same as that of the remelting zone, which is a, and the vertical major axis length is the residual height of the melt channel d. t ; Considering the influence of melt path overlap on melt path residual height during multi-pass forming, it is assumed that within a single scanning interval, the powder is only accumulated in this area after melting; based on this assumption and ignoring the evaporation loss of the material, the melt path residual height d is obtained by the condition of constant volume before and after forming. t , described by the following formula:

[0012]

[0013] Among them, ρ p is the loose packing density of the powder, ρ s is the material density, t is the powder thickness, h is the scanning interval, and a is the molten pool width;

[0014] Step 3: Prediction model of pore defects: A computational model for simulating the melt stacking in the LPBF forming process, namely the pore defect prediction model, was developed based on Python language; by inputting the melt width a, height d and residual height d t , as well as the scanning spacing h, powder thickness t and inter-layer rotation angle parameters, the model can realize the stacking of the melt path layer by layer along the construction direction, and can predict the pore defects that may occur during the forming process; when simulating and predicting, the stacking area size is set to 1000×800μm, and the stacking process is carried out from left to right and from bottom to top; in order to consider the influence of melt path offset on pore defects, each molten pool is randomly offset in the width direction based on the theoretical position; the offset is randomly determined according to the Gaussian distribution, where the concentrated position of the x value is randomly generated in the range of 0-3, and the concentration is randomly determined in the range of 4-6;

[0015] Step 4: Density prediction; The porosity δ of the sample can be obtained by the ratio of the number of white pixels to the total pixels in the model prediction results; Considering the influence of slag inclusions and impurities that may exist in the material, it is assumed that the density of the sample without pore defects is 99%, and in the sample with pore defects, the density of the pore area filled by powder is the ratio of the powder bulk density to the theoretical density of the material; Based on these assumptions, the model predicts the density of the sample ρ 预测 Calculated by the following formula:

[0016]

[0017] Where: p is the loose packing density of the powder, ρ s is the material density and δ is the porosity.

[0018] The technical effects of the present invention relative to the prior art are:

[0019] The prediction model established by the present invention takes into account the influence of melt path deviation on pore defects, thereby improving the prediction accuracy; at the same time, the model can also comprehensively consider the influence of laser power, scanning speed, scanning spacing, powder thickness and interlayer rotation angle on pore defects; therefore, the present invention can effectively predict the generation of unpredictable pore defects caused by melt path deviation under high-quality melt path forming conditions, improve the prediction accuracy of sample density, improve the repeatability and stability of additively manufactured formed samples, and promote the popularization and application of additive manufacturing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is the melt path deviation diagram;

[0021] Figure 2 It is a schematic diagram of the cross-sectional dimensions of the melt channel;

[0022] Figure 3is the prediction result of the model for the pore defects in Example 10;

[0023] Figure 3 (a) does not consider the melt path deviation, (b) the melt path is randomly deviated, and (c) the actual test result;

[0024] Figure 4 It is the prediction result of the model for the density of each embodiment. DETAILED DESCRIPTION

[0025] In order to make the contents of the present invention easier to understand, the technical solution of the present invention is further described below in conjunction with specific implementation methods, but the scope of the present invention is not limited.

[0026] A solution of the present invention:

[0027] A method for predicting porosity defects of additively manufactured parts taking into account melt path deviation comprises the following steps:

[0028] Step 1: Calculation of the cross-sectional dimensions of the melt channel; the melt channel width a and depth d are calculated using an analytical method, and the melt channel residual height dt is calculated based on the condition that the volume remains unchanged before and after forming, and the cross-sectional dimension information of the formed melt channel under different laser powers and scanning speeds is obtained;

[0029] Step 2: Based on the principle of melt channel stacking and considering the law of melt channel deviation, a pore defect prediction model is established; by inputting laser power, scanning speed, scanning spacing, powder thickness and interlayer rotation angle into the prediction model, the model is used to predict the pore defects that may appear in the sample and calculate its density;

[0030] In step 1, by matching the laser power and the scanning speed, the laser line energy density is between 0.2 J / mm and 2 J / mm, ensuring that the formed melt channel does not have spheroidization, discontinuity, or keyhole porosity defects;

[0031] In step 1, the melt channel width a and depth d are calculated as follows:

[0032]

[0033] Where: a is the molten pool width, α is the laser absorptivity; P is the laser power, ρ s is the material density, c s and c l is the specific heat capacity of the material in solid and liquid states, T0 is the ambient temperature, T m is the melting temperature of the metal, v is the scanning speed, u is the laser diameter, κ is the thermal conductivity, K is the temperature gradient on the solid-liquid interface, h is the convection heat transfer coefficient, ε is the material emissivity, σ b is the Boltzmann constant, L is the latent heat of fusion;

[0034] The residual height of the melt channel dt is calculated as follows:

[0035]

[0036] Among them, ρ p is the loose packing density of the powder, ρ s is the material density, t is the powder thickness, h is the scanning interval, and a is the molten pool width;

[0037] During stacking, the melt path is randomly offset in the width direction at its theoretical position, and the offset is randomly determined according to Gaussian distribution, where the concentration position of the value is randomly generated in the range of 0-3, and the concentration degree is randomly determined in the range of 4-6;

[0038] When calculating the density, it is assumed that the density of the sample without pore defects is 99%. In the sample with pore defects, the density of the pore area filled with powder is the ratio of the powder bulk density to the theoretical density of the material. The model predicts the density of the sample ρ 预测 Calculated by the following formula:

[0039]

[0040] Where: p is the loose packing density of the powder, ρ s is the material density and δ is the porosity.

[0041] A more specific solution of the present invention:

[0042] A prediction method for porosity defects in additively manufactured parts considering melt path deviation.

[0043] Step 1: Calculation of molten pool size; analytical calculation is used to quickly predict the molten pool width and depth under different laser powers and scanning speeds, where the laser value range is 100-400W, the scanning speed range is 200-1200mm / s, and the laser line energy density is between 0.2-2J / mm; under this process condition, it is ensured that there is no spheroidization and discontinuity in the melt channel, as well as keyhole porosity defects. At this time, the molten pool is a typical semi-elliptical morphology. The following formula is used to calculate the molten pool width and depth:

[0044]

[0045] Where: a is the molten pool width, α is the laser absorptivity; P is the laser power, ρ s is the material density, c s and c l is the specific heat capacity of the material in solid and liquid states, T0 is the ambient temperature, T mis the melting temperature of the metal, v is the scanning speed, u is the laser diameter, κ is the thermal conductivity, K is the temperature gradient on the solid-liquid interface, h is the convection heat transfer coefficient, ε is the material emissivity, σ b is the Boltzmann constant, L is the latent heat of fusion;

[0046] Step 2: Calculation of the cross-sectional dimensions of the melt channel; During LPBF forming, the laser inputs energy into the molten pool, causing the powder particles to melt and solidify to form a new forming zone; at the same time, this process will also cause the remelting of the solidified material in the previous layer. During the solidification stage, the newly deposited material will form a "small hat" shape on the previously deposited material under the action of surface tension. Based on the observation of the morphological characteristics of the melt channel, it is assumed that the remelting zone and the forming zone in the forming process are both semi-elliptical in shape, where the horizontal major axis length of the remelting zone ellipse is the melt channel width a, and the vertical major axis length d is the melt pool depth. The horizontal major axis length of the forming zone ellipse is the same as that of the remelting zone, which is a, and the vertical major axis length is the residual height of the melt channel d. t ; Considering the influence of melt path overlap on melt path residual height during multi-pass forming, it is assumed that within a single scanning interval, the powder is only accumulated in this area after melting; based on this assumption and ignoring the evaporation loss of the material, the melt path residual height d is obtained by the condition of constant volume before and after forming. t , described by the following formula:

[0047]

[0048] Among them, ρ p is the loose packing density of the powder, ρ s is the material density, t is the powder thickness, h is the scanning interval, and a is the molten pool width;

[0049] Step 3: Prediction model of pore defects: A computational model for simulating the melt stacking in the LPBF (laser powder bed fusion) forming process, namely the pore defect prediction model, was developed based on Python language; by inputting the melt width a, height d and residual height d t , as well as the scanning spacing h, powder thickness t and inter-layer rotation angle parameters, the model can realize the stacking of the melt path layer by layer along the construction direction, and can predict the pore defects that may occur during the forming process; when simulating and predicting, the stacking area size is set to 1000×800μm, and the stacking process is carried out from left to right and from bottom to top; in order to consider the influence of melt path offset on pore defects, each molten pool is randomly offset in the width direction based on the theoretical position; the offset is randomly determined according to the Gaussian distribution, where the concentrated position of the x value is randomly generated in the range of 0-3, and the concentration is randomly determined in the range of 4-6;

[0050] Step 4: Density prediction; The porosity δ of the sample can be obtained by the ratio of the number of white pixels to the total pixels in the model prediction results; Considering the influence of slag inclusions and impurities that may exist in the material, it is assumed that the density of the sample without pore defects is 99%, and in the sample with pore defects, the density of the pore area filled by powder is the ratio of the powder bulk density to the theoretical density of the material; Based on these assumptions, the model predicts the density of the sample ρ 预测 Calculated by the following formula:

[0051]

[0052] Where: p is the loose packing density of the powder, ρ s is the material density and δ is the porosity.

[0053] That is, the prediction method of the pore defect of the additively manufactured part considering the melt path deviation of the present invention is carried out according to the following steps:

[0054] Step 1: Calculate the cross-sectional dimensions of the melt channel; Calculate the width and depth of the melt channel cross-sectional dimensions according to formulas (1) and (2); Calculate the residual height d of the melt channel according to formula (3) t , and then completely determine the cross-sectional dimensions of the melt channel under different laser powers and scanning speeds.

[0055] Step 2: Prediction of pore defects and calculation of density. The established pore defect prediction model is used to predict the morphology of pore defects under different laser powers and scanning speeds, and the density is calculated according to formula (4). To improve the accuracy of model prediction, 10 melt stacking images are randomly generated during prediction to calculate the density and replace the average value.

[0056] Example: Using 2205 duplex stainless steel as the material, the prediction ability of the model for pore defects and sample density under different laser energy densities and scanning speeds is explored. The predicted process parameters are shown in Table 1. During forming, the powder thickness is 30 μm, and the scanning strategy is to rotate 67° layer by layer. As shown in Table 1, the laser energy density covers a wide range of 83.33-312.50 J / mm 3 .

[0057] Table 1 Experimental process parameters

[0058]

[0059] The prediction results of the model for the pore defects in Example 10 are as follows: Figure 3As shown in the figure, the prediction results of regular arrangement of melt paths under the same process parameters and the actual test measurement results are added for comparison; the figure shows that when the melt paths are regularly arranged, no obvious porosity defects are observed in the prediction results; in contrast, under the condition of considering random offset of the melt paths, the prediction results show that additional porosity defects are generated in the sample, and these defects are mainly concentrated in the overlapping area of ​​the molten pool, which is consistent with the distribution position of porosity defects in the actual formed sample, indicating that the model can better predict the unfused defects caused by insufficient overlap of the melt paths.

[0060] The model predicts the density of each embodiment as follows: Figure 4 As shown in the figure, compared with the model prediction results under regular arrangement of melt channels, after introducing melt channel offset, the model prediction results are closer to the experimental measurement results, which further proves that the established pore defect prediction model has high prediction accuracy.

[0061] The above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made according to the scope of the patent application of the present invention should fall within the scope of the present invention.

Claims

1. A method for predicting porosity defects in additively manufactured parts taking into account melt path deviation, characterized in that: The steps include: Step 1: Calculation of the cross-sectional dimensions of the melt channel; the melt channel width a and depth d are calculated using an analytical method, and the melt channel residual height dt is calculated based on the condition that the volume remains unchanged before and after forming, and the cross-sectional dimension information of the formed melt channel under different laser powers and scanning speeds is obtained; Step 2: Based on the principle of melt channel stacking and considering the law of melt channel deviation, a pore defect prediction model is established; by inputting laser power, scanning speed, scanning spacing, powder thickness and interlayer rotation angle into the prediction model, the model predicts the pore defects that may appear in the sample and calculates its density.

2. A method for predicting porosity defects of additively manufactured parts taking into account melt path deviation according to claim 1, characterized in that: In step 1, by matching the laser power and the scanning speed, the laser line energy density is between 0.2 J / mm and 2 J / mm, ensuring that there are no spheroidization, discontinuity, or keyhole defects in the formed melt channel.

3. The method for predicting porosity defects of additively manufactured parts taking into account melt path deviation according to claim 1, characterized in that: In step 1, the melt channel width a and depth d are calculated as follows: Where: a is the molten pool width, α is the laser absorptivity; P is the laser power, ρ s is the material density, c s and c l is the specific heat capacity of the material in solid and liquid states, T0 is the ambient temperature, T m is the melting temperature of the metal, v is the scanning speed, u is the laser diameter, κ is the thermal conductivity, K is the temperature gradient on the solid-liquid interface, h is the convection heat transfer coefficient, ε is the material emissivity, σ b is the Boltzmann constant, L is the latent heat of fusion; The residual height of the melt channel dt is calculated as follows: Among them, ρ p is the bulk density of the powder, ρ s is the material density, t is the powder thickness, h is the scanning interval, and a is the molten pool width.

4. A method for predicting porosity defects of additively manufactured parts taking into account melt path deviation as claimed in claim 1, 2 or 3, characterized in that: During stacking, the melt path is randomly offset in the width direction at its theoretical position, and the offset is randomly determined according to a Gaussian distribution, wherein the concentration position of the values ​​is randomly generated in the range of 0-3, and the concentration degree is randomly determined in the range of 4-6.

5. The method for predicting porosity defects of additively manufactured parts taking into account melt path deviation according to claim 4, characterized in that: When calculating the density, it is assumed that the density of the sample without pore defects is 99%. In the sample with pore defects, the density of the pore area filled with powder is the ratio of the powder bulk density to the theoretical density of the material. The model predicts the density of the sample ρ 预测 Calculated by the following formula: Where: p is the bulk density of the powder, ρ s is the material density and δ is the porosity.

6. A method for predicting porosity defects in additively manufactured parts taking into account melt path deviation, characterized in that: Step 1: Calculation of molten pool size; analytical calculation is used to quickly predict the molten pool width and depth under different laser powers and scanning speeds, where the laser value range is 100-400W, the scanning speed range is 200-1200mm / s, and the laser line energy density is between 0.2-2J / mm; under this process condition, it is ensured that there is no spheroidization and discontinuity in the melt channel, as well as keyhole porosity defects. At this time, the molten pool is a typical semi-elliptical morphology. The following formula is used to calculate the molten pool width and depth: Where: a is the molten pool width, α is the laser absorptivity; P is the laser power, ρ s is the material density, c s and c l is the specific heat capacity of the material in solid and liquid states, T0 is the ambient temperature, T m is the melting temperature of the metal, v is the scanning speed, u is the laser diameter, κ is the thermal conductivity, K is the temperature gradient on the solid-liquid interface, h is the convection heat transfer coefficient, ε is the material emissivity, σ b is the Boltzmann constant, L is the latent heat of fusion; Step 2: Calculation of the cross-sectional dimensions of the melt channel; During LPBF forming, the laser inputs energy into the molten pool, causing the powder particles to melt and solidify to form a new forming zone; at the same time, this process will also cause the remelting of the solidified material in the previous layer. During the solidification stage, the newly deposited material will form a "small hat" shape on the previously deposited material under the action of surface tension. Based on the observation of the morphological characteristics of the melt channel, it is assumed that the remelting zone and the forming zone in the forming process are both semi-elliptical in shape, where the horizontal major axis length of the remelting zone ellipse is the melt channel width a, and the vertical major axis length d is the melt pool depth. The horizontal major axis length of the forming zone ellipse is the same as that of the remelting zone, which is a, and the vertical major axis length is the residual height of the melt channel d. t ; Considering the influence of melt path overlap on melt path residual height during multi-pass forming, it is assumed that within a single scanning interval, the powder is only accumulated in this area after melting; based on this assumption and ignoring the evaporation loss of the material, the melt path residual height d is obtained by the condition of constant volume before and after forming. t , described by the following formula: Among them, ρ p is the bulk density of the powder, ρ s is the material density, t is the powder thickness, h is the scanning interval, and a is the molten pool width; Step 3: Prediction model of pore defects: A computational model for simulating the melt stacking in the LPBF forming process, namely the pore defect prediction model, was developed based on Python language; by inputting the melt width a, height d and residual height d t , as well as the scanning spacing h, powder thickness t and inter-layer rotation angle parameters, the model can realize the stacking of the melt path layer by layer along the construction direction, and can predict the pore defects that may occur during the forming process; when simulating and predicting, the stacking area size is set to 1000×800μm, and the stacking process is carried out from left to right and from bottom to top; in order to consider the influence of melt path offset on pore defects, each molten pool is randomly offset in the width direction based on the theoretical position; the offset is randomly determined according to the Gaussian distribution, where the concentrated position of the x value is randomly generated in the range of 0-3, and the concentration is randomly determined in the range of 4-6; Step 4: Density prediction; The porosity δ of the sample can be obtained by the ratio of the number of white pixels to the total pixels in the model prediction results; Considering the influence of slag inclusions and impurities that may exist in the material, it is assumed that the density of the sample without pore defects is 99%, and in the sample with pore defects, the density of the pore area filled by powder is the ratio of the powder bulk density to the theoretical density of the material; Based on these assumptions, the model predicts the density of the sample ρ 预测 Calculated by the following formula: Where: p is the bulk density of the powder, ρ s is the material density and δ is the porosity.

Images