A method for predicting the porosity of coatings prepared by cold spraying

Abstract

The present invention proposes a method for predicting the porosity of a coating prepared by cold spraying, by establishing a three-dimensional randomly distributed multi-particle collision model, calculating the collision velocity of different particle sizes under a certain cold spraying pressure, assigning particles of corresponding particle sizes as their initial velocity, applying initial temperature to the particles and the matrix, and dividing the grid of appropriate size based on the model size. After calculation, a three-dimensional coating is obtained, and the output can be selected as EVF (Void / Material volume fraction in element) in the post-processing interface to observe the porosity of the coating, and the cross-sectional morphology of the coating is exported, and the porosity of the coating can be measured using software such as Image pro.

Description

Technical Field

[0001] The present invention belongs to the field of cold spray solid additive manufacturing, and particularly relates to a method for predicting the porosity of a coating prepared by cold spray. Background Art

[0002] Cold spray is a new type of solid additive manufacturing technology. During the cold spray process, powder particles are accelerated to an ultra-high speed of 300 - 1200 m / s under supersonic conditions and are combined with the substrate through their own strong plastic deformation at a temperature below the melting point of the material. It has the advantages of high spraying rate, good bonding strength, and little thermal influence on the substrate, and is widely used in fields such as part repair, surface protection, and corrosion resistance. Porosity is an important performance index of the coating. The lower the porosity of the coating, the better the coating quality. The factors affecting the coating porosity include cold spray working pressure, gas preheating temperature, spraying distance, particle size, etc.

[0003] In order to improve the coating quality, researchers use a combination of cold spray experiments and numerical simulations to study the bonding mechanism between the coating and the substrate, as well as the influence of process parameters on the coating quality, so as to optimize the cold spray process and expand its application range. Experiments can directly observe the influence of process parameters on the coating quality, but coatings under each parameter cannot be prepared. When conducting cold spray experiments on pure aluminum powder, powder clogging often occurs due to too high gas preheating temperature or relatively small particle size (about 15 μm), resulting in poor coating quality or inability to prepare the coating. Therefore, numerical simulation methods can be used to analyze the influence of particle size or preheating temperature on the coating. The numerical simulation method was initially used to establish a single-particle collision model to observe the changes in deformation, stress, strain, temperature field, and energy of the particle and the substrate during the process of the particle colliding with the substrate, and to determine the critical velocity of particle deposition by observing the sudden change in stress or the velocity of the particle jet.

[0004] Different simulation methods have their own advantages and disadvantages, and researchers choose appropriate simulation methods according to the needs of the research content. With the progress of the research, particle collision models have also emerged, which are often used to study the influence of parameters such as powder materials, powder particle size, substrate materials, and cold spray pressure in cold spray on the performance of the coating and the substrate. Most of them are two-dimensional models, and the particle size and position are evenly distributed, which has a certain gap with the actual cold spray process. Summary of the Invention

[0005] The technical problem solved by the present invention is as follows: In the existing cold spraying technology, due to the small density and low hardness of pure aluminum powder, powder clogging will occur during the cold spraying process when the particle size is small or the gas preheating temperature is high, resulting in the inability to study the influence of particle size and gas temperature on the coating performance. The present invention proposes a method for predicting the porosity of a coating prepared by cold spraying. By establishing a three-dimensional multi-particle collision model, assigning particle collision velocities and initial temperatures, calculating the cross-sectional morphology of the coating, and using software such as Image pro to calculate the coating porosity, the prediction of the coating porosity is achieved.

[0006] The technical solution of the present invention is: A method for predicting the porosity of a coating prepared by cold spraying, comprising the following steps:

[0007] Step 1: Calculate the collision velocities of particles with different particle sizes under a certain cold spraying pressure using an empirical formula;

[0008] The calculation formula is:

[0009]

[0010] Step 2: Establish a three-dimensional randomly distributed multi-particle collision model, and assign the collision velocities at different particle sizes in Step 1 to the particles corresponding to the particle sizes, including the following sub-steps:

[0011] Step 2.1: Establish particle, Euler domain, and substrate models in the part module of ABAQUS software. In the assembly module, assemble the particles with the same particle size into one part, and then assemble the particles with different particle sizes with the Euler domain and the substrate;

[0012] Step 2.2: Set the material properties in the property module and assign the material properties to the corresponding part;

[0013] Step 2.3: Perform mesh division on the Euler domain and the substrate in the mesh module, including the settings of the mesh size and element type. The element type of the outermost mesh unit of the substrate is different from that of the internal mesh, which is convenient for subsequent infinite element settings;

[0014] Step 2.4: Set the boundary conditions and initial conditions for the model in the load module. The initial conditions include the initial velocity of the particles and the initial temperature of the model. The particle velocity adopts the calculation result in Step 1 and is set based on the Euler volume fraction method with the particles as the entity reference.

[0015] Step 2.5: Establish an analysis step in the step module, and set the analysis duration according to the particle velocity magnitude in the load module to ensure that the particles are completely deposited on the substrate;

[0016] Step 2.6: Create a job in the job module, export the inp file under the current job, modify the mesh type of the outermost elements of the substrate in the exported inp file, save the inp file after completing the setting of the infinite element type in Step 2.2, recreate the job and submit the modified inp file for simulation calculation.

[0017] Step 3: Obtain a three-dimensional coating through the simulation calculation in Step 2. In the post-processing of ABAQUS, select the output as EVF (Void / Material volume fraction in element) to observe the porosity of the coating, and export the porosity of different cross-sections of the coating as pictures.

[0018] Step 4: Use Image pro software to calculate the porosity of the pictures exported in Step 3, and the average value obtained is the predicted porosity value of the coating under this parameter.

[0019] Further, in the above Step 1, in the particle velocity calculation formula, V P is the collision velocity of the particle; M is the nozzle Mach number; is N 2 is the molecular weight; γ is the specific heat capacity of the gas, 1.6 for monatomic and 1.4 for diatomic; R is the gas constant; T is the gas temperature; D is the powder particle size; x is the distance from the nozzle throat to the substrate; ρ P is the powder density; P 0 is the cold spray gas pressure.

[0020] Further, in the above Step 2.1, the particle size is normally distributed, the particle size is between 15 and 65 μm, and the number of particles is between 100 and 200.

[0021] Further, in the above Step 2.1, the particles are randomly assembled during assembly to ensure that the particles do not intersect with each other.

[0022] Further, in the above Step 2.2, the material property model is the Johnson-Cook constitutive model. The parameters included are: A, B, n, m, T m 、T r , where A and B are strain hardening parameters, n is the strain hardening power exponent, m is the thermal softening power exponent, T m is the melting point of the material, T r is the material transition temperature, hardening parameter (select the Johnson-Cook hardening model), density, thermal conductivity and specific heat.

[0023] Further, in step 2.2, the mesh type of the matrix is C3D8RT, the element type of the outermost layer of the mesh is C3D8T, and the mesh type of the Euler domain is EC3D8RT.

[0024] Further, in step 2.4, the analysis step is a temperature-displacement coupled dynamic explicit analysis step, and the analysis duration is 600 - 1500 ns.

[0025] Further, in step 2.5, modify the outermost single type of the substrate in the inp file, changing from C3D8T to an infinite element of CIN3D8 to avoid the influence of stress waves and reduce the model size.

[0026] Advantages of the Invention

[0027] The technical effects of the present invention are as follows:

[0028] 1. The present invention makes the entire model closer to reality in terms of three aspects: particle size, particle quantity, and particle distribution. In step 2, when establishing the three-dimensional model, the particle size shows a normal distribution, the number of particles can reach more than one hundred, and the particle distribution is random to ensure no contact between two particles. Compared with the previous two-dimensional, regularly distributed, and equal-particle-size models, the model of the present invention is more comprehensive and systematic, achieving a cold spray deposition process closer to reality.

[0029] 2. Combining with the empirical formula for predicting particle velocity in step 1, the present invention can predict the porosity of cold spray coatings obtained under different cold spray pressures, gas preheating temperatures, and powder particle sizes. Numerical simulation calculations for parameters that cannot effectively prepare coatings can better analyze the influence of process parameters on coating quality, facilitating subsequent optimization of equipment or processes. In addition, compared with preparing coatings through cold spray experiments and calculating the porosity of the coating by grinding metallographic specimens, the present invention can quickly predict the porosity through calculations, providing certain reference for cold spray experimental research.

[0030] 3. When establishing the matrix model, the present invention introduces an infinite element layer, which can change the outermost layer of the mesh to an infinite element, avoiding the influence of stress wave rebound and reducing the model size and calculation time. Description of the Drawings

[0031] Figure 1 Multi-particle modeling in Example 1

[0032] Figure 2 Euler domain and matrix modeling in Example 1

[0033] Figure 3 Multi-particle and matrix model diagram in Example 1

[0034] Figure 4 Coating porosity result diagram in Example 1

[0035] Figure 5 Optical micrograph of the cold-sprayed test coating in Example 1

[0036] Figure 6 Multi-particle modeling in Example 2

[0037] Figure 7 Eulerian domain and substrate modeling in Example 2

[0038] Figure 8 Multi-particle and substrate model diagram in Example 2

[0039] Figure 9 Coating porosity result diagram in Example 2

[0040] Figure 10 Optical micrograph of the cold-sprayed test coating in Example 2

[0041] Figure 11 Coating porosity results obtained from simulation and experiment in Example 1 and Example 2 Detailed implementation manners

[0042] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention.

[0043] Refer to Figures 1-11 , the present invention establishes a three-dimensional randomly distributed multi-particle collision model, calculates the collision velocities of different particle sizes under a certain cold spraying pressure, assigns them as the initial velocities of the corresponding particle sizes, applies an initial temperature (generally 25°C) to the particles and the substrate, and divides a grid of appropriate size based on the model size. After software calculation, a three-dimensional coating is obtained. In the post-processing interface, it can be selected to output as EVF (Void / Material volume fraction in element) to observe the porosity of the coating. The cross-sectional morphology of the coating is exported, and the porosity of the coating can be measured using software such as Image pro.

[0044] A prediction method for the porosity of a coating prepared by cold spraying, characterized in that the simulation method is carried out in the finite element simulation software ABAQUS according to the following steps:

[0045] Step 1: Calculate the collision velocity of particles with different particle sizes under a certain cold spraying pressure using an empirical formula;

[0046] The empirical formula is as follows:

[0047]

[0048] where V P is the collision velocity of the particles; M is the nozzle Mach number; is N 2 molecular weight; γ is the specific heat capacity of the gas, 1.6 for monatomic and 1.4 for diatomic; R is the gas constant; T is the gas temperature; D is the powder particle size; x is the distance from the nozzle throat to the substrate; ρ P is the powder density; P 0 is the cold spraying gas pressure.

[0049] Step 2: Establish a three-dimensional random distribution multi-particle collision model and assign the collision velocities at different particle sizes in Step 1 to the particles corresponding to the particle sizes. Specifically, it includes the following sub-steps:

[0050] Step 2.1: Establish models of particles, Eulerian domain, and substrate in the part module of ABAQUS software. In the assembly module, assemble the particles with a particle size difference of no more than 10% into one part, and then assemble the particles with different particle sizes with the Eulerian domain and the substrate;

[0051] In Step 2.1, the particle sizes are normally distributed, the particle sizes are between 15 and 65 μm, and the number of particles is between 100 and 200. During assembly, random assembly is performed to ensure that there is no intersection between particles.

[0052] Step 2.2: Set the material properties in the property module and assign the material properties to the corresponding part;

[0053] The material model is the Johnson-Cook constitutive model. The parameters included are: A, B, n, m, T m 、T r , where A and B are strain hardening parameters, n is the strain hardening power exponent, m is the thermal softening power exponent, T m is the melting point of the material, T r is the material transformation temperature, hardening parameter (select the Johnson-Cook hardening model), density, thermal conductivity, and specific heat.

[0054] Step 2.3: Perform mesh division on the Eulerian domain and the substrate in the mesh module, including the settings of mesh size and element type. The element type of the outermost grid of the matrix is different from that of the internal grid, which is convenient for subsequent infinite element settings.

[0055] In step 2.2, the mesh type of the substrate is C3D8RT, the element type of the outermost layer of the mesh is C3D8T, and the mesh type of the Euler domain is EC3D8RT.

[0056] Step 2.4: Set the boundary conditions and initial conditions for the model in the load module. The initial conditions include the initial velocity of the particles and the initial temperature of the model. The particle velocity adopts the calculation result in step 1 and is set based on the Euler volume fraction method with the particles as the solid reference.

[0057] Step 2.5: Establish an analysis step in the step module. Set the analysis duration according to the particle velocity magnitude in the load module to ensure that the particles are completely deposited on the substrate.

[0058] In step 2.5, the analysis step is a temperature-displacement coupled dynamic explicit analysis step, and the analysis duration is 600 - 1500 ns.

[0059] Step 2.6: Create a job in the job module, export the inp file under the current job, modify the mesh type of the outermost cells of the substrate in the exported inp file, save the inp file after completing the setting of the infinite element type in step 2.2, recreate the job and submit the modified inp file for simulation calculation.

[0060] In step 2.6, modify the outermost cell type of the substrate in the inp file from C3D8T to CIN3D8.

[0061] Step 3: Obtain a three-dimensional coating through the simulation calculation in step 2. In the post-processing of ABAQUS, select the output as EVF (Void / Material volume fraction in element) to observe the porosity of the coating, and export the porosity of different cross-sections of the coating as pictures.

[0062] Step 4: Use Image pro software to calculate the porosity of the pictures exported in step 3, and the average value is the predicted porosity value of the coating under this parameter.

[0063] The present invention will be further described below in conjunction with specific examples.

[0064] Example 1

[0065] Taking the deposition of pure aluminum powder with a particle size distribution of 15 - 45 μm on a copper substrate as an example to predict the porosity of the pure aluminum coating, the specific method of this example includes the following steps:

[0066] Step 1: Use an empirical formula to calculate the particle collision velocity at different particle sizes under a certain cold spraying pressure.

[0067] The formula is as follows:

[0068]

[0069] where V P is the collision velocity of the particles; M is the nozzle Mach number; is N 2 molecular weight; γ is the specific heat capacity of the gas, 1.6 for monatomic and 1.4 for diatomic; R is the gas constant; T is the gas temperature; D is the powder particle size; x is the distance from the nozzle throat to the substrate; ρ P is the powder density; P 0 is the cold spray gas pressure.

[0070] The cold spray pressure is 3.0 MPa, the spraying distance is 30 mm, the loading gas is diatomic nitrogen, and the gas preheating temperature is 300 °C. The velocities of particles with different particle sizes are shown in Table 3.

[0071] Table 3 Collision velocities of particles with different particle sizes

[0072]

[0073] Step 2: Establish a three-dimensional randomly distributed multi-particle collision model, and assign the collision velocities at different particle sizes in Step 1 to the particles corresponding to the particle sizes;

[0074] In the part module of ABAQUS software, establish particles with particle sizes of 15 μm, 25 μm, 35 μm, and 45 μm. The number of particles is shown in Table 1. Establish a cuboid Euler domain with dimensions of 320 * 320 * 570 μm, which completely covers the multi-particle region, and a cuboid substrate with dimensions of 340 * 340 * 800 μm to complete the establishment of the Euler domain and the substrate;

[0075] In the Assembly module, assemble the particles with different corresponding particle sizes together as closely as possible without contacting each other. The particles with different particle sizes are evenly distributed to avoid concentration. The particle distribution range is within a cylindrical region with a diameter of 180 μm and a height of 470 μm;

[0076] In the Assembly module, delete the previous assembly, and assemble the new multi-particle model with different particle sizes, the Euler domain, and the substrate into a whole. The Euler domain coincides with the top of the substrate by 100 μm to facilitate the deformation after particle deposition, as Figure 2 shown. After the assembly is completed, hide the Euler domain and display the multi-particle and substrate parts, as Figure 3 shown.

[0077] Table 1 Multi-particle size distribution in Example 1

[0078]

[0079] Set the material properties in the property module and assign the material properties to the corresponding part;

[0080] In this embodiment, the material model is the Johnson-Cook constitutive model. The material parameters are shown in Table 2. Create two materials in the Property module, input the material parameters of aluminum and copper, establish the corresponding cross-sectional properties, assign the properties of material aluminum to the Euler domain, and the properties of material copper to the matrix;

[0081] Table 2 Material parameters of aluminum and copper

[0082]

[0083] Perform mesh division in the mesh module, including the settings of mesh size, division method, and element type;

[0084] The mesh type of the Euler domain is EC3D8RT, and the mesh size is 2.5μm. Perform volume division on the matrix, and divide it at 100μm from the top and 20μm close to the outermost side; the mesh type in the middle part is C3D8RT, the mesh type of the outermost side is set to C3D8T, the mesh size is 2.5μm, perform layout and seeding on the part below 100μm from the top, with the number being 10; perform local seeding on the outermost side, with the number being 1, and the mesh division method is sweeping;

[0085] Set the boundary conditions and initial conditions in the load module;

[0086] The initial conditions are the particle velocity and the overall temperature of the model. Assign the velocities calculated in step 1 to particles with different particle sizes. The initial temperatures of the particles and the substrate are 25°C. The boundary condition is to fully constrain the bottom of the substrate, and set displacement constraints on the outer surface of the Euler domain to prevent the material from flowing out of the Euler domain;

[0087] Establish an analysis step in the step module and set the analysis duration;

[0088] In this embodiment, the analysis step is a temperature-displacement coupled dynamic explicit analysis step. Establish a temperature-displacement coupled dynamic explicit analysis step in the Step module, set the time to 8.0e-7s, and select temperature, stress, strain, void / material volume fraction in elements, and displacement as output variables in the history output;

[0089] Create a job in the job module, export the inp file under the current job, modify the mesh type of the outermost side in the inp file from C3D8T to CIN3D8, and save. Resubmit the modified inp file in the job module for simulation calculation;

[0090] Step 3: The three-dimensional coating is obtained through the simulation calculation in Step 2. In the post-processing of ABAQUS, select the output as EVF (Void / Material volume fraction in element) to observe the porosity of the coating. As shown in Figure 4 , select the porosity of the cross-sections at three different positions of the coating and export it as a picture;

[0091] Step 4: Use Image pro software to measure the porosity and calculate its average value as the porosity of the coating under this parameter. According to the EVF value less than 0.5 and below being voids, select the cross-sectional views of the coating at three different positions, use Image pro software to measure the porosity and calculate its average value as the porosity of the coating under this parameter, which is 4.52%.

[0092] Conduct cold spraying experiments with the same cold spraying parameters. The particle size of the powder used is 15 - 45 μm. After obtaining the coating specimens, cut, grind, and polish them to prepare metallographic specimens. Observe the cross-sectional morphology of the coating using an optical microscope. As shown in Figure 5 , use Image pro software to measure the porosity of the coating, which is 3.8%.

[0093] The error between the simulation data and the actual data is within 20%, and the porosity of the cold-sprayed coating can be predicted to a certain extent. Due to the small particle size and a certain gap compared with the actual cold spraying process in terms of the number of particles, only a part can be selected. When the number of particles reaches a certain level, more accurate prediction data closer to the actual situation than this example will be obtained.

[0094] Example 2

[0095] Taking the deposition of pure aluminum powder with a particle size distribution of 25 - 65 μm on a copper substrate as an example, predict the porosity of the pure aluminum coating. The specific method of this example includes the following steps:

[0096] Step 1: Use the same empirical formula as in Example 1 to calculate the collision velocities of particles with different particle sizes under a certain cold spraying pressure. The velocities of particles with different particle sizes are shown in Table 3;

[0097] Table 3 Collision velocities of particles with different particle sizes

[0098]

[0099] Step 2: Establish a three-dimensional random distribution multi-particle collision model and assign the collision velocities at different particle sizes in Step 1 to the particles with corresponding particle sizes;

[0100] The steps include separately drawing single particles with particle sizes ranging from 25 to 65 μm in the ABAQUS Part module. The particle sizes and quantities are shown in Table 4. In the Assembly module, assemble the particles with corresponding quantities of different sizes together as tightly as possible without touching each other, and ensure that the particles of different sizes are evenly distributed to avoid concentration. The particle distribution range is within a cylindrical region with a diameter of 240 μm and a height of 600 μm. Select particles of the same size and integrate them into one Part, naming it with the particle size, thus completing the establishment of multi-particles, as Figure 6 shown. In the Part module, establish a cuboid Euler domain with dimensions of 380*380*700 μm, completely covering the multi-particle region, and a cuboid matrix with dimensions of 400*400*800 μm, thus completing the establishment of the Euler domain and the matrix.

[0101] In the Assembly module, delete the previous assembly, and assemble the new multi-particle model with different sizes, the Euler domain, and the matrix into a whole. The Euler domain coincides with the top of the matrix by 100 μm, which is convenient for deformation after particle deposition, as Figure 7 shown. After the assembly is completed, hide the Euler domain and display the multi-particle and matrix parts, as Figure 8 shown;

[0102] The multi-particle size distribution in this embodiment is shown in Table 4, which is the same as that in Embodiment 1, and the overall particle size shows a normal distribution.

[0103] Table 4 Multi-particle size distribution in Embodiment 1

[0104]

[0105] Set the material properties in the property module;

[0106] The material properties in this embodiment are the same as those in Embodiment 1. The material parameters in Embodiment 1 are locally saved and directly used in Embodiment 2;

[0107] Perform mesh division in the mesh module, including the settings of mesh size, division method, and element type;

[0108] The mesh division, division method, mesh type, and mesh size are the same as those in Embodiment 1;

[0109] Set the boundary conditions and initial conditions in the load module;

[0110] The settings of the initial conditions and boundary conditions in this embodiment are the same as those in Embodiment 1;

[0111] Establish an analysis step in the step module and set the analysis duration;

[0112] In this embodiment, the analysis steps are the same as those in Embodiment 1, and the analysis step time is set to 1.0e-6 s. Large-sized particles have a lower velocity under the same cold spraying pressure, and the time taken for deformation is also longer than that of small-sized particles;

[0113] Create a job in the job module, generate an inp file, modify the outermost mesh type in the inp file from C3D8T to CIN3D8, and save it. Resubmit the modified inp file in the job module for calculation;

[0114] Step 3: Obtain a three-dimensional coating through the simulation calculation in Step 2. In the post-processing of ABAQUS, select the output as EVF (Void / Material volume fraction in element) to observe the porosity of the coating. As Figure 9 shown, select the porosities of cross-sections at three different positions of the coating and export them as pictures;

[0115] Step 4: Use Image pro software to measure the porosity and calculate its average value as the porosity of the coating under this parameter, which is 2.32%.

[0116] Conduct cold spraying tests with the same cold spraying parameters. The particle size of the powder used is 15 - 45 μm. After obtaining the coating specimens, cut, grind, and polish them to prepare metallographic specimens. Observe the cross-sectional morphology of the coating using an optical microscope. As Figure 10 shown, measure the coating porosity using Imagepro software, which is 1.43%.

[0117] From the perspective of the influence of particle size on the coating porosity, the method of the present invention can achieve qualitative analysis. The simulation and test data of Embodiment 1 and Embodiment 2 are plotted as dash-dotted lines. As Figure 11 shown. It can be seen that the simulation can qualitatively analyze the influence of particle size on the coating porosity. In addition, it is also possible to change the cold spraying pressure, substrate temperature, particle and substrate materials, etc. to qualitatively analyze the influence of these factors on the coating porosity.

Claims

1. A method for predicting the porosity of a coating prepared by cold spraying, characterized in that: The following steps are involved: Step 1: Use an empirical formula to calculate the collision velocity of particles of different particle sizes under a certain cold spray pressure; The calculation formula is: The particle velocity calculation formula V P is the collision velocity of the particles; M is the nozzle Mach number; is the molecular weight of N2; γ is the specific heat capacity of gas, which is 1.6 for monatomic and 1.4 for diatomic; R is the gas constant; T is the gas temperature; D is the powder particle size; x is the distance between the nozzle throat and the substrate; ρ P is the powder density; P0 is the cold spraying gas pressure; Step 2: Establish a three-dimensional randomly distributed multi-particle collision model, and assign the collision velocities at different particle sizes in step 1 to particles of corresponding particle sizes, including the following sub-steps: Step 2.1: Establish the particle, Euler domain and substrate models in the part module of ABAQUS software, assemble particles with the same particle size into a part in the assembly module, and then assemble particles with different particle sizes with the Euler domain and substrate; Step 2.2: Set the material properties in the property module and assign the material properties to the corresponding part; Step 2.3: Mesh the Euler domain and substrate in the mesh module, including the mesh size and unit type settings. The unit type of the outermost mesh of the substrate is different from that of the inner mesh, which is convenient for the subsequent infinite unit setting. Step 2.4: Set boundary conditions and initial conditions for the model in the load module. The initial conditions include the initial velocity of the particles and the initial temperature of the model. The particle velocity is calculated based on the Euler volume fraction method in step 1 and is set with the particles as the entity reference. Step 2.5: Establish an analysis step in the step module and set the analysis time according to the particle velocity in the load module to ensure that the particles are completely deposited on the substrate; Step 2.6: Create a job in the job module and export the inp file under the current job. Modify the mesh type of the outermost unit of the matrix in the exported inp file. After completing the setting of the infinite unit type in step 2.2, save the inp file, recreate the job and submit the modified inp file for simulation calculation; Step 3: After the simulation calculation in step 2, a three-dimensional coating is obtained. In the post-processing of ABAQUS, the output is selected as EVF to observe the porosity of the coating, and the porosity of the cross-sections at different positions of the coating is exported as a picture; Step 4: Use Image Pro software to calculate the porosity of the image exported in step 3 and find its average value.

2. A method for predicting the porosity of a coating prepared by cold spraying as claimed in claim 1, characterized in that: In the step 2.1, the particle size is normally distributed, the particle size is between 15 and 65 μm, and the number of particles is between 100 and 200.

3. A method for predicting the porosity of a coating prepared by cold spraying as claimed in claim 1, characterized in that: In step 2.1, the particles are randomly assembled during assembly to ensure that the particles do not intersect with each other.

4. A method for predicting the porosity of a coating prepared by cold spraying as claimed in claim 1, characterized in that: The material property model in step 2.2 is the Johnson-Cook constitutive model; the parameters included are: A, B, n, m, T m , T r , where A and B are strain hardening parameters, n is the strain hardening power exponent, m is the thermal softening power exponent, T m is the melting point of the material, T r are material transition temperature, hardening parameters, density, thermal conductivity and specific heat.

5. The method for predicting the porosity of a coating prepared by cold spraying according to claim 1, characterized in that: In step 2.2, the mesh type of the matrix is ​​C3D8RT, the unit type of the outermost mesh is C3D8T, and the mesh type of the Euler domain is EC3D8RT.

6. A method for predicting the porosity of a coating prepared by cold spraying as claimed in claim 1, characterized in that: The analysis step in step 2.4 is a temperature-displacement coupled dynamic display analysis step, and the analysis duration is 600 to 1500 ns.

7. A method for predicting the porosity of a coating prepared by cold spraying as claimed in claim 1, characterized in that: In step 2.5, the outermost single type of the substrate in the inp file is modified from C3D8T to CIN3D8 infinite elements to avoid the influence of stress waves and reduce the model size.

Images