A sintering inversion-based lunar soil-like DLP printing size compensation method

Abstract

This invention discloses a dimensional compensation method for lunar regolith-like DLP printing based on sintering inversion, belonging to the field of additive manufacturing technology. The method includes: classifying and refining irregularly shaped lunar regolith-like powder; constructing a high-solids-content photosensitive slurry system; printing standard samples using DLP technology and performing gradient debinding and sintering; measuring the anisotropic shrinkage rate of the samples; constructing a finite element model based on the Skorhod-Olevsky viscous sintering (SOVS) theory; coupling microscopic grain growth kinetics with transverse anisotropy correction; identifying the high-temperature rheological parameters and anisotropy factor of the material through parameter inversion; and using the calibrated model to predict the sintering deformation field of the target component, generating a reverse pre-deformation compensation model. This invention effectively solves the problem of non-uniform shrinkage caused by gravity and microstructure orientation during liquid-phase sintering of lunar regolith-like material photopolymerized printed parts, achieving near-net-shape forming of aerospace precision spare parts.

Description

Technical Field

[0001] This invention belongs to the field of additive manufacturing technology, specifically relating to a lunar soil-like DLP printing size compensation method based on sintering inversion, which is applicable to near-net-shape manufacturing of aerospace precision components in the context of in-situ resource utilization in deep space exploration. Background Technology

[0002] In-situ resource utilization (ISRU) is a core supporting technology for future deep space exploration. Its core logic is to achieve self-sufficiency in construction materials through the on-site conversion of lunar surface resources, thus overcoming the cost and capacity bottlenecks of the Earth-Moon transportation system. As a key branch of the ISRU technology system, lunar regolith additive manufacturing technology directly serves the structural construction and equipment maintenance of lunar research stations. The extreme conditions of microgravity, ultra-high vacuum, and intense thermal cycling on the lunar surface place stringent requirements on the forming mechanism and forming precision of lunar regolith materials.

[0003] Current lunar soil 3D printing technologies mostly employ extrusion molding processes combined with alkali activators to achieve low-temperature curing. This approach is only suitable for constructing large lunar structures, suffers from low printing precision, and cannot meet the high-resolution molding requirements of precision aerospace components such as gears and differentials. Furthermore, the curing process relies on chemical reactions, making it difficult to mold complex internal structures. While traditional Digital Light Processing (DLP) technology offers micron-level molding precision, its application to lunar soil-like materials faces core challenges: coarse powder particle size, poor slurry suspension stability, and large and uneven sintering shrinkage. This easily leads to insufficient strength, deformation, and cracking of the molded parts. In particular, lunar soil-like simulated powders contain multiphase minerals such as SiO2, Al2O3, Fe2O3, and TiO2, which are fundamentally different from the uniform composition of single-phase ceramic materials. During sintering, the mismatch in thermal expansion coefficients of each phase, combined with the microstructural directionality introduced by the DLP layering manufacturing process, results in strong shrinkage anisotropy and stress concentration, directly affecting the assembly and fitting accuracy of the parts.

[0004] Existing shrinkage control schemes for DLP-printed ceramic materials mostly employ global linear dimensional compensation based on measured average shrinkage rates. This neglects the multiphase heterogeneity and shrinkage anisotropy of lunar regolith-like materials, failing to accurately predict localized uneven shrinkage in complex structures. This ultimately leads to excessive dimensional deviations after sintering, part assembly failures, and high trial-and-error costs, as well as long process iteration cycles. Furthermore, existing sintering numerical models are primarily designed for homogeneous ceramic materials, neglecting the high-temperature rheological properties of lunar regolith-like multiphase systems, failing to correct for the transverse anisotropy caused by the DLP process, and lacking precise calibration of model parameters through experimental inversion. Consequently, they cannot achieve high-precision prediction of sintering shrinkage in lunar regolith-like materials, making it difficult to directly guide part dimensional compensation. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a lunar soil-like DLP printing size compensation method based on sintering inversion. This method solves the core problems of sintering shrinkage anisotropy caused by the multiphase heterogeneity of lunar soil, low dimensional accuracy, and high trial-and-error costs of traditional compensation methods. By constructing a high-precision sintering simulation model specifically for lunar soil, and completing the rheological parameter inversion calibration with a small number of standard parts, the geometric compensation of complex parts can be directly guided by simulation prediction, realizing near-net-shape forming of precision transmission parts in the scenario of in-situ resource utilization on the moon.

[0006] To achieve the above-mentioned objectives, the technical solution of this invention is as follows: A method for size compensation in lunar soil-like DLP printing based on sintering inversion, comprising the following steps:

[0007] S1: Select silicate-based lunar soil simulant powder and mechanically ball-mill it to obtain fine powder that meets the requirements of photocuring layup.

[0008] S2: Using photosensitive resin as the matrix, add the powder obtained from S1 and additives to prepare a high solids content lunar soil-like photosensitive slurry;

[0009] S3: Use a DLP printer to form a reference sample with standard geometric features, and perform staged gradient debinding and high-temperature sintering on the green blank of the reference sample.

[0010] S4: Measure the anisotropic linear shrinkage rate of the reference sample after sintering. Construct a finite element simulation model based on the Skorhod-Olevsky viscous sintering theory. This model treats the green body as a nonlinear viscoelastic body and assumes that the deformation during the sintering process is driven by both external load and internal pore surface tension. Identify the viscosity parameters and anisotropy factors of the material through parameter scanning inversion.

[0011] S5: Apply the viscosity parameters and anisotropy factors obtained from the inversion identification in S4 to the finite element simulation model of the target component, apply gravity and bottom friction boundary conditions, and calculate the displacement field during the sintering process.

[0012] S6: Based on the displacement field obtained in S5, extract the anisotropic shrinkage rate of the target component, perform reverse geometric compensation on the CAD model of the target component, and export the corrected model for final printing and sintering.

[0013] According to a preferred embodiment of the present invention, in S1, the chemical composition of the silicate-based lunar soil simulant powder includes multiphase minerals such as SiO2, Al2O3, Fe2O3, and TiO2, which matches the chemical composition of real lunar soil. The contents of several major chemical components, by mass percentage, are as follows: SiO2 45-50 wt.%, Al2O3 14-16 wt.%, Fe2O3 5-10 wt.%, and TiO2 1-3 wt.%.

[0014] According to a preferred embodiment of the present invention, in S1, the ball milling refining process employs a staged grinding strategy to reduce the median particle size d of the powder. 50 The powder is refined to 4-6 μm. Typically, but not exclusively, a planetary ball mill can be used for staged mechanical ball milling. The first stage uses a higher speed to initially break up powder agglomerates, and the second stage uses a lower speed to precisely refine the particle size, obtaining fine powder that meets the requirements for photocuring layup, with a median particle size d. 50 The size should be controlled within 4-6 μm.

[0015] According to a preferred embodiment of the present invention, in S2, the photosensitive resin matrix is ​​a ternary system, comprising polyurethane acrylate (PUA), 1,6-hexanediol diacrylate (HDDA), and trimethylolpropane triacrylate (TMPTA) as the matrix, mixed in a mass ratio of (40-55): (25-35): (15-25); pretreated lunar soil-like powder and additives are added, and the content of lunar soil-like simulated powder in the photosensitive slurry is controlled to be 55-65 wt.%.

[0016] According to a preferred embodiment of the present invention, the additives include a dispersant, a silane coupling agent, a photoinitiator, and a cosolvent; wherein the dispersant is BYK-9077, added at 1-4 wt.%, the silane coupling agent is KH-570, added at 5-10 wt.%, the photoinitiator is diphenyl(2,4,6-trimethylbenzoyl)phosphine oxide, added at 3-6 wt.%, and the cosolvent is PEG-300, added at 30-45 wt.%; the added amounts of the additives are all calculated based on 100% of the total mass of the ternary photosensitive resin matrix composed of PUA, HDDA, and TMPTA; the slurry is degassed by an ultrasonic degassing process to remove air bubbles.

[0017] According to a preferred embodiment of the present invention, in S3, the wavelength of the light source for DLP printing is 405nm, the single-layer exposure time is 2-4s, the layer thickness is set to 0.05-0.1mm, and the light intensity is controlled at 0-20mW / cm²; the reference sample is a ring or cylinder structure, with an outer diameter of 10-15mm, an inner diameter of 3-5mm, and a thickness of 1-3mm for the ring, and a diameter of 8-12mm and a height of 5-10mm for the cylinder.

[0018] According to a preferred embodiment of the present invention, in S3, the staged gradient degreasing and high-temperature sintering includes: heating from room temperature to 200°C at a heating rate of 1-2°C / min; continuing to heat up, performing multi-platform heat preservation degreasing in the 200-600°C range, i.e., controlling the heating rate within the range to be less than 1°C / min, and holding at multiple set heat preservation temperatures for 1-2 hours respectively; heating at 600-800°C at a heating rate of 1-1.5°C / min, and continuing to heat up from 800°C at a heating rate of 1-2°C / min until the target maximum sintering temperature, wherein the maximum sintering temperature is 1100-1200°C, and performing high-temperature densification sintering at the maximum sintering temperature for 4-8 hours; and then naturally cooling with the furnace.

[0019] Compared with the prior art, the present invention has the following significant advantages:

[0020] This invention constructs a dedicated simulation model based on the complete SOVS theory. For the first time, it combines the Skorhod-Olevsky viscous sintering theory with the multiphase heterogeneity of lunar soil and the anisotropic characteristics of DLP process. Through constitutive equation correction, grain growth coupling, and parameter inversion calibration, a complete modeling system of "theory-correction-verification" is formed, which controls the sintering size prediction error to within 1%, and improves the accuracy compared with existing empirical models.

[0021] This invention pioneers a dual coupling mechanism of transverse anisotropy and grain growth, precisely adapting to the sintering characteristics of lunar regolith-like materials. It addresses the interlayer / intralayer shrinkage differences in the DLP process through viscosity tensor correction, and corrects the densification stagnation phenomenon in the later stages of sintering through grain growth kinetics equations. This dual coupling ensures the model can reproduce the non-uniform shrinkage behavior of "high shrinkage rate in the XY plane and low shrinkage rate in the Z-axis," solving the technical pain point that single-phase models cannot adapt to lunar regolith-like materials.

[0022] This invention optimizes the slurry system and process parameters to ensure molding stability. Specifically, the 55-65 wt.% solids content photosensitive slurry achieves a balance between low viscosity and high suspension stability through the synergistic effect of additives; the staged gradient degreasing and high-temperature sintering process effectively avoids crack defects caused by rapid decomposition of the organic phase, resulting in a sintered part with a compressive strength ≥60 MPa.

[0023] The method of this invention has significant dimensional compensation effect and strong engineering applicability. Through the reverse geometric compensation strategy, the dimensional deviation rate of the lunar soil-like printed components is ≤1.5%, which is lower than that of the no-compensation scheme (deviation rate >10%) and the traditional linear compensation scheme (deviation rate >3%), and can meet the assembly requirements of aerospace precision transmission parts; at the same time, it eliminates the need for a large number of sintering experiments for each new structure, which significantly reduces the trial and error cost and process iteration cycle.

[0024] The method of this invention uses lunar soil-like simulated material as the core raw material, and the process flow is adapted to the extreme lunar environment. It can serve the equipment operation and maintenance of lunar research stations and realize the on-site manufacturing of precision spare parts such as differential gears and drive shafts. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating the entire process of the size compensation method described in this invention.

[0026] Figure 2 A comparison of particle size distribution before and after refining lunar soil-like powder;

[0027] Figure 3 Flowchart for configuring lunar soil-like DLP printing slurry;

[0028] Figure 4 This is a process curve diagram of phased gradient debinding and sintering;

[0029] Figure 5 Comparison of the appearance of the standard circular ring calibration sample before and after sintering;

[0030] Figure 6 This is a graph showing the results of viscosity parameter inversion.

[0031] Figure 7 This is a graph showing the results of the anisotropy factor scan.

[0032] Figure 8 The result shows the error between the simulated predicted size of the annulus and the experimental value.

[0033] Figure 9 The stress-strain curves of the sintered product under compression testing are shown. Detailed Implementation

[0034] The present invention will be further described in detail below with reference to specific embodiments and comparative examples, but the scope of protection of the present invention is not limited to the following embodiments.

[0035] like Figure 1 The diagram shows a flowchart of the lunar soil-like DLP printing size compensation method based on sintering inversion proposed in this invention. In some specific embodiments of this invention, the method proposed in this invention is mainly implemented according to the following process:

[0036] S1: Select silicate-based lunar soil simulant powder and mechanically ball-mill it to obtain fine powder that meets the requirements of photocuring layup.

[0037] In an optional embodiment of the present invention, the silicate-based lunar soil simulant powder uses SC-071A simulated lunar soil (commercially available), whose main chemical components are: SiO2 47.71wt.%, TiO2 1.59wt.%, Al2O3 15.02wt.%, Fe2O3 10.79wt.%, MnO 0.19wt.%, MgO 9.39wt.%, CaO 9.9wt.%, Na2O 2.7wt.%, K2O 0.82wt.%, P2O5 0.66wt.%. It is subjected to staged ball milling with a ball-to-material ratio of 3-10:1. The first stage ball milling speed is 500-600 rpm and the milling time is 1-3 h; the second stage ball milling speed is 200-300 rpm and the milling time is 6-10 h. The average particle size d of the refined powder is... 50 Controlled within 4-6μm, Figure 2 This is a comparison diagram of the particle size distribution of lunar soil powder before and after refining.

[0038] S2: Using photosensitive resin as the matrix, add the powder obtained from S1 and additives to prepare a high solids content lunar soil-like photosensitive slurry;

[0039] In this invention, the photosensitive resin matrix refers to the photocurable film-forming body composed of photoactive oligomers and active diluent monomers, excluding functional additives such as dispersants, coupling agents, photoinitiators, and cosolvents; the inorganic solid content refers to the percentage of the mass of the lunar soil-like simulated powder to the total mass of the slurry components; the component ratios and addition amounts are all based on mass.

[0040] The photosensitive resin matrix is ​​a ternary resin system with the following mass ratio: polyurethane acrylate (PUA) 50 wt.%, 1,6-hexanediol diacrylate (HDDA) 30 wt.%, and trimethylolpropane triacrylate (TMPTA) 20 wt.%. The amount of additives added is based on the total mass of the ternary photosensitive resin matrix, specifically: co-solvent PEG-300 38 wt.%, dispersant BYK-9077 3 wt.%, silane coupling agent KH-570 6-7 wt.%, and photoinitiator TPO 5-6 wt.%. The inorganic solids content of the high-solids-content lunar soil photosensitive slurry is 55-65 wt.%, with an optimal inorganic solids content of 60 wt.%. Figure 3 This is a flowchart illustrating the preparation process of the lunar soil-like DLP printing slurry (high solids content lunar soil-like photosensitive slurry) used in this embodiment.

[0041] S3: Use a DLP printer to form a reference sample with standard geometric features, and perform staged gradient debinding and high-temperature sintering on the green blank of the reference sample.

[0042] In step S3, the specific method for testing the curing depth is as follows: The prepared slurry is added to the DLP printer feed tank, a fixed light intensity of 3.44 mW / cm² is set, and the single-layer exposure time is set with gradients. After exposure, the cured layer is removed, and the uncured slurry on the surface is cleaned with anhydrous ethanol. The curing depth is measured using a micrometer. After repeating multiple sets of experiments, a Jacobs working curve is fitted to determine the critical exposure time and penetration depth. The resin conversion rate is calculated using Fourier transform infrared spectroscopy (FTIR), and the calculation formula is: Resin conversion rate = ;in, The absorption intensity of the characteristic peak of the C=C double bond in the slurry before curing. The absorption intensity of the characteristic peak of the C=C double bond on the lower surface of the cured layer after curing is given. The optimized printing parameters are: single-layer exposure time 3s, curing depth 0.07mm, and light source wavelength 405nm.

[0043] Figure 4 The diagram illustrates the phased gradient degreasing and sintering process curves used in the embodiment. In step S3, the optimal process for multi-stage gradient sintering is as follows: maximum sintering temperature 1145℃, sintering atmosphere is air; low-temperature degreasing stage: room temperature → 200℃, heating rate 1℃ / min, holding at 200℃ for 1 hour; 200℃ → 335℃, heating rate 0.5℃ / min, holding at 335℃ for 2 hours; 335℃ → 430℃, heating rate 0.5℃ / min, holding at 430℃ for 2 hours; 430℃ → 490℃, heating rate 0.5℃ / min. The process involves: holding at 490℃ for 2 hours, then heating from 490℃ to 600℃ at a rate of 0.5℃ / min, holding at 600℃ for 2 hours to ensure complete removal of the organic phase; the transition phase involves heating from 600℃ to 800℃ at a rate of 1℃ / min, holding at 800℃ for 2 hours to promote the recombination of the silicate matrix; the high-temperature densification phase involves heating from 800℃ to 1145℃ at a rate of 1.5℃ / min, holding at 1145℃ for 6 hours; and the cooling phase involves cooling from 1145℃ to room temperature, allowing the furnace to cool naturally.

[0044] S4: Measure the anisotropic linear shrinkage rate of the reference sample after sintering. Construct a finite element simulation model based on the Skorhod-Olevsky viscous sintering theory. This model treats the green body as a nonlinear viscoelastic body and assumes that the deformation during the sintering process is driven by both external load and internal pore surface tension. Identify the viscosity parameters and anisotropy factors of the material through parameter scanning inversion.

[0045] In some embodiments of the present invention, S4 includes:

[0046] S41: Measure the radial and axial dimensions of the sintered reference sample and calculate the linear shrinkage rate in the XY plane and Z direction as target data for finite element simulation model calibration.

[0047] S42: Establish a three-dimensional model according to the original size of the reference sample, set the material constitutive relationship based on the Skorhod-Olevsky viscous sintering theory, and define core parameters including initial porosity and sintering stress.

[0048] S43: The material viscosity is divided into the XY plane and the Z-axis direction, and an anisotropy factor is introduced to characterize the difference in flow resistance between the two directions;

[0049] S44: Add a grain growth equation to simulate the evolution of grain size during sintering and feed it back into the sintering stress calculation to correct the densification rate in the later stage.

[0050] S45: Apply the temperature curves of the phased gradient degreasing and high-temperature sintering in S3, apply gravity and bottom friction constraints, and perform transient solution;

[0051] S46: Scan the reference viscosity and anisotropy factor within the set range, compare the simulated shrinkage rate with the measured value, find the parameter combination with the smallest error, obtain the viscosity parameters and anisotropy factor of the material, and complete the material parameter inversion and model calibration.

[0052] Step S4 specifically involves using a 3D scanner to measure the key radial (XY plane) and axial (Z axis) dimensions of the reference sample before and after sintering. Each sample is measured three times, and the average value is taken to calculate the anisotropic linear shrinkage rate. A finite element simulation model is then constructed based on the Skorhod-Olevsky viscous sintering (SOVS) theory, as detailed below:

[0053] 1. Core assumptions of SOVS theory: The porous green body resembling lunar soil is regarded as a macroscopically homogeneous continuous medium (equivalent continuous medium assumption), the true density of the solid skeleton is constant (skeleton incompressible assumption), the deformation mechanism at high temperature is mainly diffusion-controlled viscous flow (viscous flow dominant assumption), and short-time elastic deformation and non-dominant plastic deformation are ignored.

[0054] 2. Constitutive Equation Construction: SOVS theory treats the material as a nonlinear viscoelastic body, where deformation is driven by both external loads (such as gravity) and internal pore surface tension. The macroscopic constitutive relation is:

[0055]

[0056] The parameters in the formula are defined as follows:

[0057] Stress tensor (Pa) describes the stress state inside a material;

[0058] The shear viscosity (Pa·s) of a fully dense matrix material varies with temperature according to the Arrhenius law. Where A is the pre-exponential factor, Q is the viscosity activation energy, R is the gas constant, and T is the absolute temperature;

[0059] Porosity (dimensionless) characterizes the degree of material densification. ,in Relative density;

[0060] and are the shape factors (dimensionless) of the bulk viscosity modulus and the shear viscosity modulus, respectively, used to describe the weakening effect of porosity on the macroscopic flowability of the material, and their expressions are: , ;

[0061] : Strain rate tensor (1 / s), which describes the deformation rate of a material;

[0062] The volumetric strain rate is 1 / s. This characterizes the rate of volume shrinkage;

[0063] The symbol for Kronecker. When it is 1, The time is 0;

[0064] Effective sintering stress, i.e., the thermodynamic driving force that drives pore closure and volume contraction, always points inward towards the material, and its expression is: (γ is the surface energy, G(T) is the instantaneous grain diameter).

[0065] 3. Anisotropy Correction: To address the microstructural orientation caused by the layer-by-layer stacking and blade-based material application in the DLP process, a transverse anisotropy assumption is introduced. The scalar viscosity is corrected to a viscosity tensor. The corrected inelastic strain rate is:

[0066]

[0067]

[0068] in, For effective sintering stress (Pa). The anisotropy factor (dimensionless) characterizes the difference in viscous drag along the Z-axis relative to the XY plane. The macroscopic volumetric viscosity (Pa·s) is defined as:

[0069]

[0070] 4. Grain growth kinetic coupling: The evolution of grain size with sintering temperature and time is solved using COMSOL's domain ordinary differential equation module. The core equation is:

[0071]

[0072] in, The grain growth rate constant, which follows the Arrhenius law, is used to correct the decrease in densification rate caused by grain coarsening in the later stage of sintering, and to avoid the model from misjudging due to excessive shrinkage.

[0073] 5. Parameter Inversion: Set Reference Viscosity (10) 17 -10 10 Pa·s) and anisotropy factor The parameter scanning range is (0.2-1.0). The error between the simulated shrinkage rate and the experimental value is ≤1% by using the least squares method, thus completing the parameter inversion and model calibration.

[0074] S5: Apply the viscosity parameters and anisotropy factors obtained from the inversion identification in S4 to the finite element simulation model of the target component, apply gravity and bottom friction boundary conditions, and calculate the displacement field during the sintering process.

[0075] In some embodiments of the present invention, S5 includes:

[0076] S51: Import the 3D CAD model of the target component into the finite element simulation software to construct the finite element simulation model of the target component;

[0077] S52: Assign the reference viscosity and anisotropy factor obtained from the inversion identification in step S4 to the finite element simulation model of the target component;

[0078] S53: Apply gravity load to simulate the effect of self-weight during sintering, and set friction constraints on the bottom surface of the model to simulate the contact behavior between the component and the sintering plate;

[0079] S54: Load the temperature curves of the phased gradient degreasing and high-temperature sintering in S3, perform transient solution of multi-field coupling of thermo-mechanical-microstructure, and calculate the evolution of displacement field, stress field and density field at each position of the component during the entire sintering process.

[0080] S55: After the solution is completed, extract the displacement of the corresponding part of the target component, calculate the anisotropic shrinkage rate distribution, identify the deformation law and stress concentration area, and obtain the displacement field during the sintering process.

[0081] In a specific embodiment of the present invention, in S5, the calibrated rheological parameters (reference viscosity) are... Anisotropy factor The target component is given a finite element model, a gravity load (9.8 m / s²) and bottom friction boundary conditions (friction coefficient 0.1-0.2) are applied, a sintering process curve consistent with the reference sample is set, and transient thermo-mechanical-microstructure multi-field coupling solution is performed to calculate the displacement field during the sintering process and accurately predict the anisotropic shrinkage behavior and tooth distortion law of the target component.

[0082] Furthermore, in steps S4 and S5, the multiphysics field and boundary conditions are set as follows: the solid heat transfer module loads the actual sintering temperature curve and solves the transient temperature field; the solid mechanics module solves the transverse anisotropic shrinkage through the inelastic strain rate node; the Z-axis negative gravity load and bottom roller support constraint are added to simulate the real sintering conditions; geometric nonlinearity is enabled, and the median time step strategy is used to complete the transient solution.

[0083] S6: Based on the displacement field obtained in S5, extract the anisotropic shrinkage rate of the target component, perform reverse geometric compensation on the CAD model of the target component, and export the corrected model for final printing and sintering.

[0084] In some embodiments of the present invention, step S6 includes: extracting the shrinkage rate of the target component at key positions in the XY plane and Z-axis direction from the displacement field of step S5; calculating the amplification factor in each direction according to the shrinkage rate and the reverse compensation principle, so that the green blank model shrinks to the design size after sintering, and obtaining the compensated three-dimensional CAD model of the green blank; using the printing parameters determined in step S3, performing DLP printing on the compensated model to obtain the green blank component; and processing it using the gradient degreasing and high-temperature sintering conditions determined in step S3 to obtain the final sintered part.

[0085] In a specific embodiment of the present invention, S6 is: based on the displacement field data predicted by simulation, extract the shrinkage rate at key locations of the target component, and then... , , The mapping relationship is used to perform reverse geometric compensation on the original CAD model; the corrected model is exported, and green blank is printed using the optimal printing parameters determined in step S3. After sintering in step S3, the target component with accurate dimensions is obtained; the sintered part is compared with the design standard dimensions to verify the size compensation effect.

[0086] To systematically verify the effectiveness and process tolerance of the dimensional compensation method described in this invention, six sets of test specimens were designed. The model parameters and core applications of each set are shown in the table below:

[0087] Furthermore, in step S6, the specific grouping and design of the size pre-compensation scheme are shown in Table 1 below:

[0088] Table 1 Size Pre-compensation Scheme

[0089]

[0090] Example 1: Implementation and Verification of the Optimal Pre-compensation Scheme (Group B)

[0091] This embodiment corresponds to the test piece in group B of the table above. Using a precision spur gear for lunar exploration as the target spare part, it completes the construction, calibration, and dimensional pre-compensation verification of the simulation model. The specific implementation steps are as follows:

[0092] 1. Pretreatment of lunar soil-like simulated powder

[0093] Commercially available SC-071A simulated lunar soil powder was used. Its main chemical components (determined by XRD) were: SiO2 47.71 wt.%, TiO2 1.59 wt.%, Al2O3 15.02 wt.%, Fe2O3 10.79 wt.%, MnO 0.19 wt.%, MgO 9.39 wt.%, CaO 9.9 wt.%, Na2O 2.7 wt.%, K2O 0.82 wt.%, P2O5 0.66 wt.%; A planetary ball mill was used for staged ball milling with a ball-to-powder ratio of 3:1. Zirconia grinding balls were mixed in a 10:5:1 ratio of 2mm / 5mm / 10mm. The first stage involved milling at 600 rpm for 1 hour, followed by a 0.5-hour interval to prevent overheating and agglomeration. This process was repeated three times. The second stage involved milling at 300 rpm for 3 hours to complete the powder grinding. A laser particle size analyzer was used to detect the particle size distribution of the refined powder, and the average particle size d was measured. 50 =5.386μm, which meets the design requirements of 4-6μm.

[0094] 2. Preparation of high-solids-content photosensitive paste

[0095] Prepare a lunar soil-like photosensitive slurry with a solid content of 60 wt.% by mass fraction, as follows:

[0096] Solid phase composition: 60 wt.% of refined SC-071A type lunar soil powder;

[0097] The liquid phase component is 40 wt.%, of which the ternary resin system consists of approximately 53 wt.% PUA, approximately 30 wt.% HDDA, and approximately 17 wt.% TMPTA.

[0098] Additives (based on the total mass of the ternary photosensitive resin matrix): PEG-300 approximately 38 wt.%, BYK-9077 approximately 3 wt.%, KH-570 approximately 7 wt.%, TPO approximately 5.5 wt.%;

[0099] Preparation process: First, the ternary resin mixture is added to a beaker and stirred at 300 rpm for 10 minutes using a high-speed mixer to obtain a uniform premix. Then, various additives are added and mixed, and stirred again at 400 rpm for 10 minutes. Subsequently, lunar soil-like fine powder is added in batches, and the mixer speed is increased by 50 rpm each time powder is added to prevent powder agglomeration. After all the lunar soil powder has been added and dissolved, the photoinitiator TPO is added, and the stirring speed is maintained for another 10 minutes. Finally, ultrasonic degassing is performed for 30 minutes to remove air bubbles from the slurry, resulting in the finished lunar soil-like DLP printing slurry.

[0100] 3. Printing and sintering of reference samples

[0101] A standard circular reference sample was printed using an industrial-grade DLP photopolymerization 3D printer (light source wavelength 405nm). The design dimensions are: outer diameter 13.2mm, inner diameter 3.0mm, and thickness 2.1mm. The printing parameters were optimized as follows: single-layer exposure time 3s, curing depth 0.07mm, and layer thickness 0.05mm.

[0102] Sintering process: A degreasing sintering furnace is used for sintering in an air atmosphere. The temperature rises from room temperature to 200℃ at a rate of 1℃ / min, and is held at 200℃ for 1 hour; from 200℃ to 335℃ at a rate of 0.5℃ / min, and is held at 335℃ for 2 hours; from 335℃ to 430℃ at a rate of 0.5℃ / min, and is held at 430℃ for 2 hours; from 430℃ to 490℃ at a rate of 0.5℃ / min, and is held at 490℃ for 2 hours; from 490℃ to 600℃ at a rate of 0.5℃ / min, and is held at 600℃ for 2 hours; from 600℃ to 800℃ at a rate of 1℃ / min, and is held at 800℃ for 2 hours; from 800℃ to 1145℃ at a rate of 1.5℃ / min, and is held at 1145℃ for 6 hours; from 1145℃ to room temperature, the furnace is allowed to cool naturally.

[0103] Figure 5 The image shows a comparison of the appearance of the standard circular ring calibration sample before and after sintering. Measurement results: after sintering, the outer diameter of the ring is 9.82 mm, the inner diameter is 2.23 mm, and the thickness is 1.82 mm. The calculated radial shrinkage rate in the XY plane is 25.61%, and the axial shrinkage rate in the Z axis is 13.33%.

[0104] 4. Simulation parameter inversion and model calibration

[0105] A finite element model of the circular toroidal reference sample was constructed using COMSOL Multiphysics 6.3 software.

[0106] Geometric modeling and mesh generation: A 1:1 toroidal model was established, and a hexahedral swept mesh was used, with a maximum mesh cell size of 1.322 mm and a minimum mesh cell size of 0.07692 mm.

[0107] Constitutive equation and parameter settings: Import the modified Skorhod-Olevsky viscous sintering constitutive equation, and set the parameters accordingly. , Define the shape factor; couple the solid heat transfer module to load the above sintering curve, and use the domain ordinary differential equation module to solve the grain growth kinetics;

[0108] Anisotropy settings: Add an inelastic strain rate node to the solid mechanics module, and input the reference shrinkage rate for the XX / YY components. ZZ component input ;

[0109] Parameter Scanning and Inversion: Setting the Reference Viscosity Scan range 10 7 -10 10 Pa·s, anisotropy factor The scanning range was 0.2-1.0; experimental shrinkage data were fitted using the least squares method. =1×10 8 Pa·s、 When the anisotropy coefficient is 0.5, the simulated radial diameter of the annulus is 9.8931 mm and the axial height is 1.8255 mm, with errors of 0.74% and 0.30% respectively compared to the experimental values, both ≤1%. Model calibration is complete. The anisotropy parameter inversion results are as follows: Figures 6 to 8 As shown.

[0110] 5. Spur Gear Deformation Prediction and Compensation

[0111] Import the 3D model of the optimal pre-compensated gear in Group B of the table above into the calibrated simulation model, apply a gravity load (9.8 m / s²) and a bottom surface friction constraint (friction coefficient 0.15), and set the sintering process consistent with the reference sample; after simulation calculation, the predicted results are output: the average shrinkage rate of the spur gear in the XY plane is 25.2%, and the shrinkage rate in the Z axis is 12.2%.

[0112] Compensation Calculation: Based on the predicted shrinkage rate, calculate the compensation ratio using the formula:

[0113] Magnification in the XY directions =1 / (1-0.252)=1.337;

[0114] Z-direction magnification =1 / (1-0.122)=1.139;

[0115] The original spur gear model was non-uniformly scaled in CAD software to obtain the B-group compensated green blank model (tooth tip circle diameter 21.69mm, thickness 6.83mm).

[0116] 6. Target component printing, sintering, and effect verification

[0117] Print the compensated spur gear blank according to the printing parameters in step 3. After cleaning and drying, process it according to the sintering process in step 3. At the same time, print the standard sample strips of group F in the table above for mechanical property testing.

[0118] Dimensional and performance verification results: After sintering, key dimensions were measured using a 3D scanner. The addendum circle diameter was 16.43 mm, and the thickness was 5.92 mm. The deviation rates from the design dimensions (16.22 mm and 6 mm) were 1.29% and 1.33%, respectively, both ≤1.5%. Meshing tests were conducted between the sintered gear and the theoretically sized gear model. No jamming or abnormal noise was observed during transmission. The compressive strength of the standard specimen was tested using an electronic universal testing machine, and the average compressive strength was measured to be 60.31 MPa (e.g., ...). Figure 9 As shown in the figure, it meets the mechanical requirements of aerospace spare parts.

[0119] Example 2: Verification using a 0.9x conservative pre-compensation scheme (Group C)

[0120] This embodiment corresponds to the test specimens in group C of the table above. The only difference from Embodiment 1 is that the dimensional compensation ratio of the spur gear model is 0.9 times the simulated predicted shrinkage rate, that is, the shrinkage rate in the XY direction is calculated as 22.68% (25.2% × 0.9), and the magnification factor is... =1 / (1-0.2268)=1.293 (corresponding to a tooth tip circle diameter of 20.77mm); Z-axis shrinkage rate is calculated as 10.98% (12.2%×0.9), magnification factor. =1 / (1-0.1098)=1.112 (corresponding to a thickness of 6.67mm); the remaining raw material formulation, printing process, and sintering process are consistent with Example 1. Verification results: the spur gear tip circle diameter after sintering is 16.32mm, and the thickness is 5.81mm; the deviation rates from the design dimensions are 0.62% and 3.17%, respectively. Although the deviations in the XY directions are small, the deviation in the Z direction far exceeds the design requirements, which cannot meet the long-term use requirements of aerospace precision transmission parts. It can only be used as a temporary backup solution for scenarios with stringent radial dimension requirements.

[0121] Example 3 uses a 0.8x conservative pre-compensation scheme (Group D) for verification.

[0122] This embodiment corresponds to the test specimens in group D of the table above. The only difference from Embodiment 1 is that the size compensation ratio of the spur gear model is 0.8 times the simulated predicted shrinkage rate, that is, the shrinkage rate in the XY direction is calculated as 20.16% (25.2% × 0.8), and the magnification factor is... =1 / (1-0.2016)=1.253 (corresponding to a tooth tip circle diameter of 19.79mm); the Z-axis shrinkage rate is calculated as 9.76% (12.2%×0.8), and the magnification factor is... =1 / (1-0.0976)=1.108 (corresponding to a thickness of 6.65mm); the remaining raw material formulations, printing processes, and sintering processes are consistent with those in Example 1. Verification results: the spur gear tip circle diameter after sintering is 15.88mm, and the thickness is 5.75mm; the deviation rates from the design dimensions are 2.10% and 4.17%, respectively. The dimensions are slightly out of tolerance, and only loose clearance assembly can be achieved. It cannot meet the requirements of precision transmission and is only used as a fault-tolerant backup solution under extreme working conditions.

[0123] Comparative Example 1: No Size Pre-compensation Scheme

[0124] This comparative example corresponds to the test pieces in group A of the table above. The only difference from Example 1 is that the spur gear model did not undergo any dimensional compensation, and the scaling ratio in the X / Y / Z directions was 100% (tooth tip circle diameter 16.22mm, thickness 6mm). The remaining raw material formula, printing process, and sintering process were consistent with Example 1. Verification results: After sintering, the spur gear tooth tip circle diameter was 12.17mm and the thickness was 5.28mm; the deviation rates from the design dimensions were 24.97% and 12.67%, respectively. The dimensions were seriously out of tolerance, making assembly impossible. This proves that the uncompensated scheme cannot meet the molding requirements of aerospace precision spare parts.

[0125] Comparative Example 2: Traditional Global Linear Pre-compensation Scheme

[0126] The only difference between this comparative example and Example 1 is that it adopts the mainstream global linear compensation scheme of existing technology. Based on the measured shrinkage rates of 25.61% in the XY direction and 13.33% in the Z direction of the E group reference annular sample, the spur gear model is globally linearly scaled with a uniform magnification factor in the XY direction. =1 / (1-0.2561)=1.344, uniform magnification in the Z direction =1 / (1-0.1333)=1.154; the remaining raw material formulations, printing processes, and sintering processes are consistent with those in Example 1. Verification results: Based on the simulation model calibrated according to this invention, under the traditional linear compensation scheme, the spur gear tip circle diameter after sintering is 16.84mm and the thickness is 5.93mm, with deviation rates of 3.82% and 1.17% respectively from the design dimensions; although the Z-axis dimensional deviation is smaller, the XY-axis deviation far exceeds the design requirements and is prone to contour distortion, making it unsuitable for the forming requirements of aerospace precision parts.

[0127] Summary of the Examples: Through the verification of the above examples and comparative examples, the lunar soil-like DLP printing size compensation method constructed by this invention can accurately predict the anisotropic sintering shrinkage behavior of complex lunar soil-like components. The optimal compensation scheme (Example 1) can control the size deviation rate within 1.5%, which is significantly improved in size accuracy compared with the no-compensation scheme (deviation rate 24.97%) and the traditional linear compensation scheme (deviation rate 3.82%). Among them, the conservative compensation scheme of Example 2 takes into account both size accuracy and process tolerance, with a size deviation rate of ≤1% in the XY direction, meeting the assembly requirements of aerospace precision transmission parts; the compressive strength of the sintered part is ≥60Mpa, which has the potential for engineering applications. The method of this invention does not require a large number of trial and error experiments, significantly reducing process costs and iteration cycles, adapting to the scenario of in-situ resource utilization on the moon, and providing technical support for the on-site manufacturing of deep space exploration equipment.

[0128] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for size compensation in lunar soil-like DLP printing based on sintering inversion, characterized in that, Includes the following steps: S1: Select silicate-based lunar soil simulant powder and mechanically ball-mill it to obtain fine powder that meets the requirements of photocuring layup. S2: Using photosensitive resin as the matrix, add the powder obtained from S1 and additives to prepare a high solids content lunar soil-like photosensitive slurry; S3: Use a DLP printer to form a reference sample with standard geometric features, and perform staged gradient debinding and high-temperature sintering on the green blank of the reference sample. S4: Measure the anisotropic linear shrinkage rate of the reference sample after sintering. Construct a finite element simulation model based on the Skorhod-Olevsky viscous sintering theory. This model treats the green body as a nonlinear viscoelastic body and assumes that the deformation during the sintering process is driven by both external load and internal pore surface tension. Identify the viscosity parameters and anisotropy factors of the material through parameter scanning inversion. S5: Apply the viscosity parameters and anisotropy factors obtained from the inversion identification in S4 to the finite element simulation model of the target component, apply gravity and bottom friction boundary conditions, and calculate the displacement field during the sintering process. S6: Based on the displacement field obtained in S5, extract the anisotropic shrinkage rate of the target component, perform reverse geometric compensation on the CAD model of the target component, and export the corrected model for final printing and sintering.

2. The method according to claim 1, characterized in that: In S1, the main phase of the silicate-based lunar soil analog powder is plagioclase, pyroxene, or olivine silicate minerals. The ball milling refining process employs a staged grinding strategy to reduce the median particle size d of the powder. 50 Refine to 4-6 μm.

3. The method according to claim 1, characterized in that: In S2, the photosensitive resin matrix is ​​a ternary system, comprising polyurethane acrylate, 1,6-hexanediol diacrylate, and trimethylolpropane triacrylate, with a mass ratio of (40-55): (25-35): (15-25); the mass ratio of the photosensitive resin matrix to the lunar soil analog powder is (2.5-3.5): (6.5-7.5); and the content of the lunar soil analog powder in the photosensitive slurry is controlled at 55-65 wt.%.

4. The method according to claim 3, characterized in that: The additives include a dispersant, a silane coupling agent, a photoinitiator, and a cosolvent. The dispersant is BYK-9077, added at 1-4 wt.% of the photosensitive resin matrix; the silane coupling agent is KH-570, added at 5-10 wt.% of the photosensitive resin matrix; the photoinitiator is diphenyl(2,4,6-trimethylbenzoyl)phosphine oxide, added at 3-6 wt.% of the photosensitive resin matrix; and the cosolvent is PEG-300, added at 30-45 wt.% of the photosensitive resin matrix. The slurry is degassed using an ultrasonic degassing process to remove air bubbles.

5. The method according to claim 1, characterized in that: In S3, the wavelength of the light source for DLP printing is 405nm, the single-layer exposure time is 2-4s, the layer thickness is set to 0.05-0.1mm, and the light intensity is controlled at 0-20mW / cm². The reference sample is a ring or cylinder structure, with an outer diameter of 10-15mm, an inner diameter of 3-5mm, and a thickness of 1-3mm for the ring, and a diameter of 8-12mm and a height of 5-10mm for the cylinder.

6. The method according to claim 1, characterized in that: In S3, the staged gradient degreasing and high-temperature sintering includes: heating from room temperature to 200℃ at a rate of 1-2℃ / min; continuing to heat up, performing multi-platform heat preservation degreasing in the 200-600℃ range, i.e., controlling the heating rate within the range to be less than 1℃ / min, and holding at multiple set heat preservation temperatures for 1-2 hours respectively; heating from 600-800℃ at a rate of 1-1.5℃ / min, and continuing to heat up from 800℃ at a rate of 1-2℃ / min until the target maximum sintering temperature, wherein the maximum sintering temperature is 1100-1200℃, and performing high-temperature densification sintering at the maximum sintering temperature for 4-8 hours; and then naturally cooling with the furnace.

7. The method according to claim 1, characterized in that: S4 includes: S41: Measure the radial and axial dimensions of the sintered reference sample and calculate the linear shrinkage rate in the XY plane and Z direction as target data for finite element simulation model calibration. S42: Establish a three-dimensional model according to the original size of the reference sample, set the material constitutive relationship based on the Skorhod-Olevsky viscous sintering theory, and define core parameters including initial porosity and sintering stress. S43: The material viscosity is divided into the XY plane and the Z-axis direction, and an anisotropy factor is introduced to characterize the difference in flow resistance between the two directions; S44: Add a grain growth equation to simulate the evolution of grain size during sintering and feed it back into the sintering stress calculation to correct the densification rate in the later stage. S45: Apply the temperature curves of the phased gradient degreasing and high-temperature sintering in S3, apply gravity and bottom friction constraints, and perform transient solution; S46: Scan the reference viscosity and anisotropy factor within the set range, compare the simulated shrinkage rate with the measured value, find the parameter combination with the smallest error, obtain the viscosity parameters and anisotropy factor of the material, and complete the material parameter inversion and model calibration.

8. The method according to claim 1, characterized in that: S5 includes: S51: Import the 3D CAD model of the target component into the finite element simulation software to construct the finite element simulation model of the target component; S52: Assign the reference viscosity and anisotropy factor obtained from the inversion identification in step S4 to the finite element simulation model of the target component; S53: Apply gravity load to simulate the effect of self-weight during sintering, and set friction constraints on the bottom surface of the model to simulate the contact behavior between the component and the sintering plate; S54: Load the temperature curves of the phased gradient degreasing and high-temperature sintering in S3, perform transient solution of multi-field coupling of thermo-mechanical-microstructure, and calculate the evolution of displacement field, stress field and density field at each position of the component during the entire sintering process. S55: After the solution is completed, extract the displacement of the corresponding part of the target component, calculate the anisotropic shrinkage rate distribution, identify the deformation law and stress concentration area, and obtain the displacement field during the sintering process.

9. The method according to claim 1, characterized in that: S6 includes: From the displacement field in step S5, the shrinkage rate of the target component at key positions in the XY plane and Z-axis direction is extracted; based on the shrinkage rate, the amplification factor in each direction is calculated according to the reverse compensation principle, so that the green billet model shrinks to the design size after sintering, and the compensated green billet 3D CAD model is obtained; using the printing parameters determined in S3, the compensated model is DLP printed to obtain the green billet component; using the gradient degreasing and high-temperature sintering conditions determined in S3, the final sintered part is obtained.

Images