A method for constructing a particle size distribution model for 3D printing powder bed particles

By constructing a particle size distribution model of powder bed in 3D printing, using the Box-Muller algorithm to generate a normal distributed particle radius and optimize the particle distribution, the problem of unstable particle distribution of powder materials is solved, and the high accuracy of 3D printing is achieved and stable performance.

CN117521171BActive Publication Date: 2024-11-22CHANGZHOU WEIREN DIGITAL TECH CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202311274870.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-28
Publication Date
2024-11-22
Estimated Expiration
2043-09-28

AI Technical Summary

Technical Problem

In the existing 3D printing technology, the distribution of powder material particles is affected by a variety of factors, making it difficult to achieve ideal nested distribution, resulting in unstable printing performance and difficult to be suitable for precision component manufacturing.

Method used

Using the construction method of the 3D printed powder bed particle size distribution model, the normal distributed particle radius is generated through the Box-Muller algorithm, and the particles are randomly placed in the powder generation space to perform overlap judgment and reverse overlap treatment to ensure particle distribution and bonding In actual situation, meet the optimal volume fraction requirements.

Benefits of technology

The powder particle distribution similarity between the digital model and the actual printed components is improved, and the performance difference fluctuation is significantly reduced. It is suitable for 3D printing of precision components, ensuring the stability and efficiency of component performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure SMS_7
    Figure SMS_7
  • Figure SMS_15
    Figure SMS_15
  • Figure SMS_22
    Figure SMS_22
Patent Text Reader

Abstract

In order to overcome the problems existing in the prior art, the present invention provides a method for constructing a particle size distribution model of a 3D printing powder bed, comprising: S1. Obtaining model target parameters. S2. Generate a random number Kx. S3. Convert the random number Kx into a normally distributed particle radius by using the Box‑Muller algorithm. S4. Randomly place Kx particles of normally distributed particle radius generated in step S3 within a specified powder generation space. S5. Perform overlap judgment and perform anti-overlap processing on overlapping particles. S6. Perform back-bound processing on out-of-bounds particles. S7. Store the particle radius and particle spatial position data of the particles placed this time. S8. Repeat the above steps S2‑S7 so that the volume percentage of particles of each particle size meets the relevant objectives. The present invention takes into account the distribution of particle size particles in the component when constructing the model, so that the performance difference fluctuation between the digital model and the actual printed component is significantly reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of additive manufacturing, and in particular to a method for constructing a particle size distribution model of 3D printing powder bed particles. Background Art

[0002] Additive manufacturing technology is a manufacturing technology that uses adhesive materials to construct objects by stacking them layer by layer based on digital model files. The most common additive manufacturing technology is 3D printing technology. When designing 3D printed components, it is often necessary to first design the component model and analyze the performance parameters that can be obtained from the required materials to determine whether the 3D printed component meets the design requirements and form the corresponding digital model file.

[0003] In the existing 3D printing technology, powder materials are usually used as raw materials for 3D printing. The distribution of coarse and fine particles in the powder is one of the key factors affecting the performance of the component. The existing technology generally uses the ideal coarse and fine particles to nest in sequence to build the 3D printing model. However, since the flow of particles in the 3D printing process is affected by many factors such as printing temperature, material fluidity, gravity, and printing environment factors, the particle distribution in the print is currently unable to achieve the distribution method of coarse and fine particles nested in sequence in the theoretical model during printing. Therefore, the performance of the obtained printed component is always unstable compared with the theoretical performance of the theoretical model. There is a downward difference, making 3D printing difficult to apply to the manufacture of precision components. Summary of the invention

[0004] In order to solve the problems raised in the above background technology, the present invention provides a method for constructing a particle size distribution model of 3D printing powder bed particles, comprising the following steps:

[0005] S1. Obtain particles Xn of different and non-overlapping radius ranges in the material to be printed, and sort them according to the radius range from large to small, the target printing model, and the theoretically optimal volume percentage Ym corresponding to the particles Xn.

[0006] S2. Generate a random number Kx through a random function.

[0007] S3. The random number Kx is converted into a normally distributed particle radius by the Box-Muller algorithm.

[0008] S4. Randomly place Kx particles with a normally distributed particle radius generated in step S3 within the specified powder generation space.

[0009] S5. Perform overlap judgment to confirm whether the particle placed in step S4 overlaps with the adjacent particle in the particle group P already placed in the specified powder generation space. When the particle placed in step S4 overlaps with the adjacent particle in the particle group P already placed in the specified powder generation space, anti-overlap processing is performed on the placed particle.

[0010] S6. If the boundary of the particle placed in step S4 exceeds the powder generation space, it is considered that the particle exceeds the boundary and needs to be returned to the boundary.

[0011] S7. After all particles are placed in step S4, the particle radius and particle spatial position data of the currently placed particles are stored, and the previously placed particle groups P are combined into a new placed particle group P, and the volume number T of all current particles is calculated.

[0012] S8. Repeat the above steps S2-S7 so that the volume percentage of particles Xn satisfies Ym in descending order.

[0013] Furthermore, the method of converting the random number Kx into a normally distributed particle radius by using the Box-Muller algorithm in step S3 includes:

[0014] Step 1. Design two independent standard normal distributions X ~ N (0,1) and Y ~ N (0,1). Since they are independent of each other, the joint probability density function is:

[0015] Formula 1.

[0016] Step 2. Transform equation 1 into polar coordinates to obtain:

[0017] Formula 2.

[0018] Step 3. Convert equation 2 to a standard uniform distribution, then:

[0019] Formula 3.

[0020] Another density function is:

[0021] Formula 4.

[0022] Step 4. Combine equation 3 and equation 4 into the cumulative distribution function CDF:

[0023] Formula 5.

[0024] Step 5. Reverse equation 5 to get:

[0025] Formula 6.

[0026] Step 6. According to the inverse transform sampling principle, if we have the PDF of P ( R ), then the sample distribution obtained by uniform sampling of the inverse function of the aligned CDF will conform to the distribution of P ( R ). If u is uniformly distributed, then U = 1 -u is also uniformly distributed, so replace 1 - u with U, and finally we can get:

[0027]

[0028] The two uniformly distributed random numbers U and V in equation 7 and equation 8 are obtained by a random function.

[0029] Step 7. Substitute the random numbers U and V into the Box-Buller algorithm to obtain the normally distributed particle radius corresponding to the random number Kx and randomly set the position of the particle.

[0030] Furthermore, the powder generation space specified in step S5 is: in the three-dimensional coordinate system, the XY plane at Z=0 is used as the bottom surface of the coordinate system, and the bottom surface of the target printing model is proportionally placed on the bottom surface of the coordinate system. The boundary of the model is used as the virtual space surface to construct a hollow powder generation space.

[0031] Optionally, the method for performing overlap determination in step S5 includes:

[0032] First, the center position of particle 1 in the model space is (x1, y1, z1), and the radius is R1. The center position of particle 2 is (x2, y2, z2), and the radius is R2. ... The center position of particle V is (x v ,y v ,z v ), with a radius of R v The particle V is centered on particle 1 and is numbered in natural number order of the particles adjacent to particle 1.

[0033] Secondly, taking particle 1 as the target, the distances from particle 1 to the surrounding particles 2 and particle V are calculated as and get

[0034] Then, taking particle 1 as the target, calculate the The sum of the radii of the corresponding particles C m =R1+R m . Where m is The corresponding particle number.

[0035] Finally, make a judgment: If <C m , they are considered overlapping.

[0036] Furthermore, the anti-overlapping processing method in step S5 includes:

[0037] First, the center position of particle 1 (x1, y1, z1) and the center position of particle m (x m ,y m ,z m ) and make a line O between them.

[0038] Next, calculate D m =(C m - ).

[0039] Finally, the center position of particle 1 is displaced D away from particle m along the extension line of line O. m , completing the anti-overlapping process.

[0040] Furthermore, after each anti-overlapping process is completed, the overlap judgment and anti-overlapping process are repeated with the new center position of particle 1 until ≥C m .

[0041] Optionally, the method for performing overlap determination in step S5 includes:

[0042] First, the center position of particle 1 in the model space is (x1, y1, z1), and the radius is R1. The center position of particle 2 is (x2, y2, z2), and the radius is R2. ... The center position of particle V is (x v ,y v ,z v ), with a radius of R v The particle V is centered on particle 1 and is numbered in natural number order of the particles adjacent to particle 1.

[0043] Secondly, taking particle 1 as the target, the distances from particle 1 to the surrounding particles 2 and particle V are calculated as

[0044] Then, taking particle 1 as the target, the sum of the radius C of the particles between particle 1 and particle V is calculated. v =R1+R v .

[0045] Finally, count all d v <C v to obtain particle group D. If there are particles in particle group D, it is determined that there are overlapping particles.

[0046] Furthermore, the anti-overlapping processing method in step S5 includes:

[0047] First, a vector group E is formed from particle 1 pointing to the centers of all particles in particle group D.

[0048] Afterwards, the sum vector F of the vector group E is calculated, and the inverse vector G of the sum vector F is obtained by calculation.

[0049] Finally, the center of particle 1 is displaced according to the reverse vector G to complete the anti-overlapping process.

[0050] Furthermore, after each anti-overlapping process is completed, the overlap judgment and anti-overlapping process are repeated with the new center position of particle 1 until no particles exist in particle group D.

[0051] Furthermore, the boundary return process in step S6 includes:

[0052] First, the ratio S between the volume Hc of the particle J exceeding the powder generation space and the total volume H0 of the particle is determined.

[0053] Afterwards, determine: if S≥0.5, delete the particle. If S<0.5, proceed to the following steps.

[0054] (1) Determine whether there is a particle above particle J. If so, delete the particle. If not, proceed to step (2).

[0055] (2) Make a vector group I between the center of particle J and the centers of the surrounding particles, and make the sum vector I0 of the vector group I.

[0056] (3) Make the sum vector I0 point to the perpendicular vector IC of the powder generation space, and the inverse vector ID of the perpendicular vector IC.

[0057] (4) Taking the center of particle J as the starting point, draw a line W connecting the center of the circle and the particle boundary surface along the direction of the reverse vector ID.

[0058] (5) With the powder generation space as the boundary, the line W is divided into a portion W1 inside the powder generation space and a portion W2 outside the powder generation space.

[0059] (6) The particle J is displaced by W2 along the vertical vector IC to complete the boundary return process.

[0060] (7) After completing the boundary return process, the overlap judgment described in step S5 is performed again. If there are overlapping particles, anti-overlapping processing is performed.

[0061] Compared with the prior art, the present invention has at least one of the following beneficial effects:

[0062] 1. The present invention introduces a normal distribution design of the particles of the powder material when designing the 3D printing model, so that it fits the particle distribution morphology in the actual printed component to the greatest extent.

[0063] 2. The 3D printing model of the present invention takes into account the particle volume fraction required for optimal performance, so that the designed model can meet the design and use requirements. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships that are relative orientations or positional relationships and are 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 therefore should not be understood as limiting the present invention.

[0066] Example 1

[0067] A method for constructing a particle size distribution model of 3D printing powder bed particles comprises the following steps:

[0068] S1. Obtain particles Xn of different and non-overlapping radius ranges in the material to be printed, and sort them according to the radius range from large to small, the target printing model, and the theoretically optimal volume percentage Ym corresponding to the particles Xn.

[0069] S2. Generate a random number Kx through a random function.

[0070] S3. The random number Kx is converted into a normally distributed particle radius by the Box-Muller algorithm.

[0071] S4. Randomly place Kx particles with a normally distributed particle radius generated in step S3 within the specified powder generation space.

[0072] S5. Perform overlap judgment to confirm whether the particle placed in step S4 overlaps with the adjacent particle in the particle group P already placed in the specified powder generation space. When the particle placed in step S4 overlaps with the adjacent particle in the particle group P already placed in the specified powder generation space, anti-overlap processing is performed on the placed particle.

[0073] S6. If the boundary of the particle placed in step S4 exceeds the powder generation space, it is considered that the particle exceeds the boundary and needs to be returned to the boundary.

[0074] S7. After all particles are placed in step S4, the particle radius and particle spatial position data of the currently placed particles are stored, and the previously placed particle groups P are combined into a new placed particle group P, and the volume number T of all current particles is calculated.

[0075] S8. Repeat the above steps S2-S7 so that the volume percentage of particles Xn satisfies Ym in descending order.

[0076] Compared with the ideal distribution model of powder particles used in the existing 3D model design, the present invention gives priority to the distribution of particles of different particle sizes in the component when constructing the model, and constructs the 3D printed digital model in a random normal distribution method that best fits the actual printed component, thereby significantly reducing the floating variation of the performance difference between the theoretical performance of the digital model and the performance of the actual printed component, and is suitable for the construction of digital models of 3D printed precision components.

[0077] In addition, the present invention needs to meet the optimal volume percentage of particles of various particle sizes when constructing the digital model, so that the theoretical performance of the final digital model can reach or exceed the design requirements. Combined with the small difference between the performance of the digital model of the present invention and the actual component, the actual performance of the 3D printed component can be made to be excellent and stable.

[0078] Example 2

[0079] Based on the method for constructing a particle size distribution model of 3D printing powder bed particles in Example 1, the method of converting the random number Kx into a normally distributed particle radius by using the Box-Muller algorithm in step S3 includes:

[0080] Step 1. Design two independent standard normal distributions X ~ N (0,1) and Y ~ N (0,1). Since they are independent of each other, the joint probability density function is:

[0081]

[0082] Step 2. Transform equation 1 into polar coordinates to obtain:

[0083]

[0084] The above result can be regarded as the product of the density functions of two probability distributions, one of which can be regarded as a uniform distribution on [0,2π]. Converting Equation 2 to a standard uniform distribution gives:

[0085]

[0086] Another density function is:

[0087]

[0088] Step 4. Combine equation 3 and equation 4 into the cumulative distribution function CDF:

[0089]

[0090] Step 5. Reverse equation 5 to get:

[0091]

[0092] Step 6. According to the inverse transform sampling principle, if we have the PDF of P ( R ), then the sample distribution obtained by uniform sampling of the inverse function of the aligned CDF will conform to the distribution of P ( R ). If u is uniformly distributed, then U = 1 -u is also uniformly distributed, so replace 1 - u with U, and finally we can get:

[0093]

[0094] The two uniformly distributed random numbers U and V in equation 7 and equation 8 are obtained by a random function.

[0095] Step 7. Substitute the random numbers U and V into the Box-Buller algorithm to obtain the normally distributed particle radius corresponding to the random number Kx and randomly set the position of the particle.

[0096] By constructing the random particle distribution model, the particle distribution in the powder layer model can be made highly similar to the actual distribution, thereby improving the similarity of the powder particle distribution morphology between the digital model and the actual printed component, making the theoretical performance of the digital model more consistent with the performance of the printed component, which is conducive to obtaining high-precision 3D printed components.

[0097] Example 3

[0098] Based on the method for constructing a 3D printing powder bed particle size distribution model of Example 1, the powder generation space specified in step S5 is: in a three-dimensional coordinate system, the XY plane at Z=0 is used as the bottom surface of the coordinate system, and the bottom surface of the target printing model is placed on the bottom surface of the coordinate system in equal proportion. The boundary of the model is used as the virtual space surface to construct a hollow powder generation space.

[0099] The method for performing overlap determination in step S5 includes:

[0100] First, the center position of particle 1 in the model space is (x1, y1, z1), and the radius is R1. The center position of particle 2 is (x2, y2, z2), and the radius is R2. ... The center position of particle V is (x v ,y v ,z v ), with a radius of R v The particle V is centered on particle 1 and is numbered in natural number order of the particles adjacent to particle 1.

[0101] Secondly, taking particle 1 as the target, the distances from particle 1 to the surrounding particles 2 and particle V are calculated as and get

[0102] Then, taking particle 1 as the target, calculate the The sum of the radii of the corresponding particles C m =R1+R m . Where m is The corresponding particle number.

[0103] Finally, make a judgment: If <C m , they are considered overlapping.

[0104] The anti-overlapping processing method in step S5 includes:

[0105] First, the center position of particle 1 (x1, y1, z1) and the center position of particle m (x m ,y m ,z m ) and make a line O between them.

[0106] Next, calculate D m =(C m - ).

[0107] Finally, the center position of particle 1 is displaced D away from particle m along the extension line of line O. m , completing the anti-overlapping process.

[0108] After each anti-overlapping process is completed, the overlap judgment and anti-overlapping process are repeated with the new center position of particle 1 until ≥C m .

[0109] Example 4

[0110] Based on the method for constructing a 3D printing powder bed particle size distribution model of Example 1, the powder generation space specified in step S5 is: in a three-dimensional coordinate system, the XY plane at Z=0 is used as the bottom surface of the coordinate system, and the bottom surface of the target printing model is placed on the bottom surface of the coordinate system in equal proportion. The boundary of the model is used as the virtual space surface to construct a hollow powder generation space.

[0111] The method for performing overlap determination in step S5 includes:

[0112] First, the center position of particle 1 in the model space is (x1, y1, z1), and the radius is R1. The center position of particle 2 is (x2, y2, z2), and the radius is R2. ... The center position of particle V is (x v ,y v ,z v ), with a radius of R v The particle V is centered on particle 1 and is numbered in natural number order of the particles adjacent to particle 1.

[0113] Secondly, taking particle 1 as the target, the distances from particle 1 to the surrounding particles 2 and particle V are calculated as

[0114] Then, taking particle 1 as the target, the sum of the radius C of the particles between particle 1 and particle V is calculated. v =R1+R v .

[0115] Finally, count all d v <C v to obtain particle group D. If there are particles in particle group D, it is determined that there are overlapping particles.

[0116] The anti-overlapping processing method in step S5 includes:

[0117] First, a vector group E is formed from particle 1 pointing to the centers of all particles in particle group D.

[0118] Afterwards, the sum vector F of the vector group E is calculated, and the inverse vector G of the sum vector F is obtained by calculation.

[0119] Finally, the center of particle 1 is displaced according to the reverse vector G to complete the anti-overlapping process.

[0120] After the anti-overlapping process is completed each time, the overlap judgment and anti-overlapping process are repeated with the new center position of particle 1 until there is no particle in particle group D.

[0121] Embodiment 3 and Embodiment 4 provide two methods for particle overlap judgment and anti-overlap processing, wherein Embodiment 3 has a better processing effect when the particle size distribution range is relatively different, and is generally applicable to mixed printing of particles of multiple materials. Embodiment 4 has a better processing effect when the particle size distribution range is relatively different, and is generally applicable to printing of a single material or alloy material.

[0122] Example 5

[0123] Based on the method for constructing a particle size distribution model of a 3D printing powder bed particle in Example 1, the return boundary processing in step S6 includes:

[0124] First, the ratio S between the volume Hc of the particle J exceeding the powder generation space and the total volume H0 of the particle is determined.

[0125] Afterwards, determine: if S≥0.5, delete the particle. If S<0.5, proceed to the following steps.

[0126] (1) Determine whether there is a particle above particle J. If so, delete the particle. If not, proceed to step (2).

[0127] (2) Make a vector group I between the center of particle J and the centers of the surrounding particles, and make the sum vector I0 of vector group I.

[0128] (3) Make a perpendicular vector IC pointing to the powder generation space with the sum vector I0, and an inverse vector ID of the perpendicular vector IC.

[0129] (4) Taking the center of particle J as the starting point, draw a line W connecting the center of the circle and the particle boundary surface along the direction of the reverse vector ID.

[0130] (5) Taking the powder generation space as the boundary, the line W is divided into a portion W1 inside the powder generation space and a portion W2 outside the powder generation space.

[0131] (6) Displace particle J by W2 along the vertical vector IC to complete the boundary return process.

[0132] (7) After completing the boundary return process, the overlap judgment described in step S5 is performed again. If there are overlapping particles, anti-overlapping processing is performed.

[0133] Since the positions of particles are randomly distributed during modeling in the present invention, and whether the particles exceed the model boundary is not considered during anti-overlap processing, it is inevitable that some particles will exceed the model boundary. In the actual printing process, it is hoped that particles exceeding the boundary (affecting the surface uniformity of the component) will be avoided as much as possible. This method can make the constructed digital model have no particles exceeding the boundary, so that the total amount of particles exceeding the boundary in the actual component can be reduced during actual 3D printing, making the constructed surface smoother and the component surface performance better.

[0134] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for constructing a particle size distribution model for 3D printing powder bed particles. It is characterized in that The steps include: S1. Obtain particles Xn of different and non-overlapping radius ranges in the material to be printed, and sort them according to the radius range from large to small, the target printing model, and the theoretically optimal volume percentage Ym corresponding to the particles Xn; S2. Generate a random number Kx through a random function; S3. The random number Kx is converted into a normally distributed particle radius by using the Box-Muller algorithm, specifically including: Step 1. Design two independent standard normal distributions X~N(0,1) and Y~N(0,1); since they are independent of each other, the joint probability density function is: Step 2. Transform equation 1 into polar coordinates to obtain: Step 3. Convert equation 2 to a standard uniform distribution, then: θ~Unif(0,2π)=2πV Equation 3; Another density function is: Step 4. Combine equation 3 and equation 4 into the cumulative distribution function CDF: Step 5. Reverse equation 5 to obtain: Step 6. According to the inverse transform sampling principle, if we have the PDF of P(R), then the sample distribution obtained by uniformly sampling the inverse function of the aligned CDF will conform to the distribution of P(R); if u is uniformly distributed, then U=1-u is also uniformly distributed, so replace 1-u with U, and finally we can get: The two uniformly distributed random numbers U and V in equations 7 and 8 are obtained by a random function; Step 7. Substitute the random numbers U and V into the Box-Buller algorithm to obtain the particle radius of the normal distribution corresponding to the random number Kx, and randomly set the position of the particle; S4. randomly placing Kx particles of the normally distributed particle radius generated in step S3 within the specified powder generation space; S5. Performing an overlap judgment to confirm whether the particle placed in step S4 overlaps with the adjacent particle in the particle group P already placed in the specified powder generation space; when the particle placed in step S4 overlaps with the adjacent particle in the particle group P already placed in the specified powder generation space, performing an anti-overlapping process on the placed particle; The prescribed powder generation space is: in a three-dimensional coordinate system, the XY plane at Z=0 is used as the bottom surface of the coordinate system, and the bottom surface of the target printing model is proportionally placed on the bottom surface of the coordinate system; the boundary of the model is used as the virtual space surface to construct a hollow powder generation space; The method for performing overlap judgment comprises: First, the center position of particle 1 in the model space is obtained as (x 1 ,y 1 ,z 1 ), with a radius of R 1 ; The center position of particle 2 is (x 2 ,y 2 ,z 2 ), with a radius of R 2 ; ...the center position of particle V is (x v ,y v ,z v ), with a radius of R v ; V is the natural number sequence number of the particles adjacent to particle 1, with particle 1 as the center; Secondly, taking particle 1 as the target, the distances from particle 1 to the surrounding particles 2 and particle V are calculated as and get min =Min(d 2 ,d 3 ,d 4 …d v ); Then, taking particle 1 as the target, calculate the relationship between particle 1 and d min The sum of the radii of the corresponding particles C m =R 1 +R m ; where m is d min The corresponding particle number; Finally, make a judgment: If d min <C m , they are considered to be overlapping; The anti-overlapping processing method comprises: First, the center position of particle 1 (x 1 ,y 1 ,z 1 ) and the center position of particle m (x m ,y m ,z m ) make a line O between them; Next, calculate D m =(C m -d min ); Finally, the center position of particle 1 is displaced D away from particle m along the extension line of line O. m , completing the anti-overlapping process; or, The method for performing overlap judgment comprises: First, the center position of particle 1 in the model space is obtained as (x 1 ,y 1 ,z 1 ), with a radius of R 1 ; The center position of particle 2 is (x 2 ,y 2 ,z 2 ), with a radius of R 2 ; ...the center position of particle V is (x v ,y v ,z v ), with a radius of R v ; V is the natural number sequence number of the particles adjacent to particle 1, with particle 1 as the center; Secondly, taking particle 1 as the target, the distances from particle 1 to the surrounding particles 2 and particle V are calculated as Then, taking particle 1 as the target, the sum of the radius C of the particles between particle 1 and particle V is calculated. v =R 1 +R v ; Finally, count all d v <C v to obtain a particle group D; if there are particles in the particle group D, it is determined that there are overlapping particles; The anti-overlapping processing method comprises: First, a vector group E is formed from particle 1 pointing to the centers of all particles in particle group D; Afterwards, the sum vector F of the vector group E is calculated, and the inverse vector G of the sum vector F is calculated; Finally, the center of particle 1 is displaced according to the reverse vector G to complete the anti-overlapping process; S6. If the boundary of the particle placed in step S4 exceeds the powder generation space, it is considered that the particle exceeds the boundary and needs to be returned to the boundary; The return boundary processing includes: First, determine the ratio S between the volume Hc of the particle J that exceeds the powder generation space and the total volume H0 of the particle; Afterwards, judge: if S≥0.5, delete the particle; if S<0.5, perform the following steps; (1) Determine whether there is a particle above particle J. If so, delete the particle. If not, proceed to step (2); (2) Make a vector group I of the center of particle J and the centers of the surrounding particles, and make the sum vector I0 of the vector group I; (3) Make a perpendicular vector IC pointing to the powder generation space and an inverse vector ID of the perpendicular vector IC; (4) With the center of particle J as the starting point, draw a line W connecting the center of the circle and the particle boundary surface along the direction of the reverse vector ID; (5) Taking the powder generation space as the boundary, the line W is divided into a part W1 inside the powder generation space and a part W2 outside the powder generation space; (6) The particle J is displaced by W2 along the vertical vector IC to complete the boundary return process; After completing the boundary return process, the overlap judgment in step S5 is performed again, and if there are overlapping particles, anti-overlapping processing is performed; S7. After all particles are placed in step S4, the particle radius and particle spatial position data of the particles placed this time are stored, and the particles are combined with the previously placed particle groups P to form a new placed particle group P, and the volume number T of all particles is calculated; S8. Repeat the above steps S2-S7 so that the volume percentage of particles Xn satisfies Ym in order from large to small.

2. The method for constructing a particle size distribution model of a 3D printing powder bed according to claim 1, It is characterized in that After each anti-overlapping process is completed, the overlap judgment and anti-overlapping process are repeated with the new center position of particle 1 until d min ≥C m .

3. The method for constructing a particle size distribution model of a 3D printing powder bed according to claim 1, It is characterized in that After the anti-overlapping process is completed each time, the overlap judgment and anti-overlapping process are repeated with the new center position of particle 1 until there is no particle in particle group D.

Citation Information

Patent Citations

  • Three-dimensional random digital test piece generation method considering aggregate characteristics

    CN109753667A

  • Geometric modeling method for microstructure of planar particle-filled composite material

    CN115034061A

  • Method for establishing mesoscopic geometric model of particle reinforced composite material

    CN116230130A

  • Method for deterministic modeling of powders and systems of particles

    WO2011110755A2

  • Modeling method for high-compactness discrete particle heterogeneous system

    WO2019010859A1