Artificial intelligence-based additive manufacturing process optimization method and system

Abstract

The present invention relates to the field of additive manufacturing technology, and specifically discloses an artificial intelligence-based additive manufacturing process optimization method and system. The method comprises: obtaining material characteristic parameters for additive manufacturing, evaluating the material's thermophysical properties and photophysical properties using the material characteristic parameters, and comprehensively analyzing to obtain material characteristic evaluation parameters; obtaining an evaluation threshold for abnormal porosity changes in additively manufactured components and an evaluation threshold for abnormal temperature changes in additively manufactured components based on the material characteristic evaluation parameters, and comprehensively analyzing to obtain an evaluation index for the degree of abnormality of the additively manufactured components; and adjusting additive manufacturing process parameters based on the evaluation index for the degree of abnormality of the additively manufactured components. The present invention realizes automated control and decision-making of the additive manufacturing process, improves production efficiency, and simultaneously regulates process parameters in real time based on the material characteristics, porosity, and temperature of the additively manufactured components, ensuring the stability of part forming and reducing post-processing workload.

Description

Technical Field

[0001] The present invention relates to the field of additive manufacturing technology, and specifically to an artificial intelligence-based additive manufacturing process optimization method and system. Background Art

[0002] Currently, additive manufacturing technology, with its unique layer-by-layer manufacturing approach, offers advantages such as complex structures, personalized customization, and rapid prototyping, and is widely used in a variety of fields, including aerospace, healthcare, and consumer products. However, the additive manufacturing process involves numerous variables, such as material properties, equipment parameters, and environmental conditions. These factors interact with each other and significantly impact the quality of the final product and production efficiency. Precisely controlling and optimizing these complex processes is key to improving the practicality and competitiveness of additive manufacturing technology. Therefore, there is a need for an artificial intelligence-based additive manufacturing process optimization method and system to optimize the additive manufacturing process.

[0003] For example, the invention patent with the announcement number CN113435670B is a quantitative prediction method for the offset of the cladding layer in additive manufacturing, which belongs to the field of precision welding technology. After completing the base work of the first four layers of additive manufacturing, the molten pool visual sensor is used to collect the molten pool image of the offset of the fifth layer of additive cladding layer, and the ROI area is selected for the molten pool image; the offset corresponding to each collected molten pool image is calculated according to the Pythagorean theorem, and a data set is made, and the arc starting point and arc extinction point of the fifth layer of additive manufacturing are determined according to the data set; classification experiments are carried out under different offsets of cladding layers and classification is carried out using a deep residual network; prediction experiments are carried out on the offset of the linearly changing cladding layer, and the classification task of different offsets of the weld is converted into a regression task; and the generalization ability of the network is verified. The present invention uses molten pool visual information to quantitatively predict the offset of the cladding layer in the additive process, adjusts the actual position of the welding gun during the additive manufacturing process, thereby obtaining a good molten pool morphology and improving welding quality.

[0004] Based on the above solution, it was found that there are still some shortcomings in additive manufacturing. In traditional methods, material selection and process parameter setting are often carried out independently, lacking deep integration and collaborative optimization, which limits the space and speed of process optimization and cannot achieve the global optimal performance of additively manufactured parts. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides an artificial intelligence-based additive manufacturing process optimization method and system, which can effectively solve the problems involved in the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: The first aspect of the present invention provides an artificial intelligence-based additive manufacturing process optimization method, including: obtaining material characteristic parameters for additive manufacturing, evaluating the material's thermal physical properties and material's photophysical properties through the material characteristic parameters, and comprehensively analyzing to obtain material characteristic evaluation parameters.

[0007] The additive manufacturing process is monitored and analyzed to obtain the evaluation values ​​of abnormal porosity changes and abnormal temperature changes of additively manufactured components. The evaluation thresholds of abnormal porosity changes and abnormal temperature changes of additively manufactured components are obtained based on the matching of material property evaluation parameters. Comprehensive analysis is then conducted to obtain the evaluation index of the abnormality degree of additively manufactured components.

[0008] The additive manufacturing process parameters are adjusted according to the additive manufacturing component abnormality evaluation index.

[0009] As a further method, the comprehensive analysis obtains material property evaluation parameters. The specific analysis process is: based on the material property parameters, including the viscosity, thermal conductivity and thermal expansion coefficient of various metal materials and the absorptivity and reflectivity of various metal materials, the critical viscosity of metal materials, the reference standard thermal conductivity of metal materials, the critical thermal expansion coefficient of metal materials, the critical absorptivity of metal materials and the critical reflectivity of metal materials are extracted from the additive manufacturing database, and the material thermophysical property evaluation parameters and material photophysical property evaluation parameters are obtained after processing.

[0010] Based on the material thermophysical property evaluation parameters and the material photophysical property evaluation parameters, the material characteristic evaluation parameters are obtained through comprehensive analysis.

[0011] As a further method, the analysis obtains an evaluation value of abnormal porosity changes of additively manufactured components and an evaluation value of abnormal temperature changes of additively manufactured components. The specific analysis process is: deploying several time monitoring points, collecting the porosity of the additively manufactured components at each time monitoring point, and at the same time extracting the critical porosity and critical porosity growth rate of the additively manufactured components from the additive manufacturing database, and obtaining the evaluation value of abnormal porosity changes of the additively manufactured components after processing.

[0012] The temperature of each printing layer of the additively manufactured component is monitored, and the temperature of each printing layer at each time monitoring point is collected. At the same time, the reference standard cooling rate of the printing layer and the reference standard temperature difference between layers are extracted from the additive manufacturing database. After processing, the evaluation value of the abnormal temperature change of the additively manufactured component is obtained.

[0013] As a further method, the material property evaluation parameters are matched to obtain the abnormal porosity change evaluation threshold of the additive manufacturing component and the abnormal temperature change evaluation threshold of the additive manufacturing component, and the abnormality degree evaluation index of the additive manufacturing component is obtained by comprehensive analysis. The specific analysis process is: matching the material property evaluation parameters with the abnormal porosity change evaluation threshold of the additive manufacturing component and the abnormal temperature change evaluation threshold of the additive manufacturing component corresponding to each material property evaluation parameter interval stored in the additive manufacturing database to obtain the abnormal porosity change evaluation threshold of the additive manufacturing component and the abnormal temperature change evaluation threshold of the additive manufacturing component, and based on the abnormal porosity change evaluation value of the additive manufacturing component and the abnormal temperature change evaluation value of the additive manufacturing component, a comprehensive analysis is performed to obtain the abnormality degree evaluation index of the additive manufacturing component.

[0014] As a further method, the additive manufacturing process parameters are adjusted according to the additive manufacturing component abnormality evaluation index. The specific analysis process is: matching the additive manufacturing component abnormality evaluation index with the laser power and printing layer thickness corresponding to the abnormality evaluation index interval of each additive manufacturing component stored in the additive manufacturing database to obtain the appropriate laser power and appropriate printing layer thickness of the additive manufacturing component, and adjusting the laser power and printing layer thickness of the additive manufacturing in real time according to the appropriate laser power and appropriate printing layer thickness.

[0015] As a further method, the material property evaluation parameter is obtained by analyzing the material thermophysical property evaluation parameter and the material photophysical property evaluation parameter to obtain a quantitative evaluation index, which is used to quantitatively evaluate the suitability of the material properties for additive manufacturing.

[0016] As a further method, the additive manufacturing component abnormality evaluation index is a quantitative indicator obtained by comprehensively analyzing the abnormal changes in porosity and temperature of the additive manufacturing component, which is used to evaluate the abnormality of the additive manufacturing component.

[0017] As a further method, the material property evaluation parameter has a specific numerical expression as follows:

[0018]

[0019] In the formula, γ represents the material property evaluation parameter, e represents the natural constant, β R Represents the evaluation parameter of the material's thermophysical properties, β G represents the material photophysical property evaluation parameter, ω1 represents the material property evaluation impact factor corresponding to the set material thermophysical property evaluation parameter, and ω2 represents the material property evaluation impact factor corresponding to the set material photophysical property evaluation parameter.

[0020] As a further method, the additive manufacturing component abnormality evaluation index has a specific numerical expression as follows:

[0021]

[0022] Where δ represents the abnormality evaluation index of additive manufacturing parts, e represents the natural constant, and ε K represents the evaluation value of abnormal porosity change of additively manufactured parts, ε W represents the evaluation value of abnormal temperature change of additively manufactured parts, ε K0 represents the evaluation threshold of abnormal porosity change of additively manufactured parts, ε W0 represents the threshold for evaluating abnormal temperature changes of additively manufactured components, ψ1 represents the influencing factor of the degree of abnormality of the additively manufactured components corresponding to the set evaluation value of abnormal porosity changes of the additively manufactured components, and ψ2 represents the influencing factor of the degree of abnormality of the additively manufactured components corresponding to the set evaluation value of abnormal temperature changes of the additively manufactured components.

[0023] The second aspect of the present invention provides an artificial intelligence-based additive manufacturing process optimization system, including: a material property analysis module for obtaining material property parameters for additive manufacturing, evaluating the material's thermal physical properties and material's photophysical properties through the material property parameters, and comprehensively analyzing to obtain material property evaluation parameters.

[0024] The additive manufacturing monitoring module is used to monitor the additive manufacturing process, analyze and obtain the evaluation value of abnormal porosity change and abnormal temperature change of additive manufacturing components, and obtain the evaluation threshold of abnormal porosity change and abnormal temperature change of additive manufacturing components based on the matching of material property evaluation parameters, and comprehensively analyze and obtain the abnormality degree evaluation index of additive manufacturing components.

[0025] The process parameter adjustment module is used to adjust the additive manufacturing process parameters according to the abnormality evaluation index of the additive manufacturing component.

[0026] The additive manufacturing database is used to store additive manufacturing related data, including the critical viscosity of metal materials, the reference standard thermal conductivity of metal materials, the critical thermal expansion coefficient of metal materials, the critical absorptivity of metal materials, the critical reflectivity of metal materials, the critical porosity and critical porosity growth rate of additively manufactured parts, the reference standard cooling rate of the printing layer and the reference standard temperature difference between layers, the abnormal porosity change assessment threshold of the additively manufactured parts and the abnormal temperature change assessment threshold of the additively manufactured parts corresponding to each material property evaluation parameter interval, and the laser power and printing layer thickness corresponding to each additively manufactured part abnormality assessment index interval.

[0027] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0028] (1) The present invention provides an artificial intelligence-based additive manufacturing process optimization method and system to analyze the material properties of additively manufactured components and monitor the porosity and temperature of the additively manufactured components during the additive manufacturing process, thereby realizing automated control and decision-making of the additive manufacturing process, reducing the burden of manual intervention, and improving production efficiency. At the same time, the process parameters are adjusted in real time according to the material properties, porosity and temperature of the additively manufactured components to ensure the stability of part forming and reduce the workload of post-processing.

[0029] (2) The present invention helps to accurately set the process parameters of additive manufacturing by analyzing the material properties of additively manufactured parts. Appropriate process parameters can ensure that the material is evenly melted and well solidified during the forming process, reduce the generation of defects, and improve the density and overall performance of the parts. At the same time, the material properties can assist in the optimization of the process parameters in the additive manufacturing process, achieve global optimization of additive manufacturing, and thus improve product quality and reliability.

[0030] (3) The present invention monitors the porosity and temperature of additively manufactured parts during the additive manufacturing process, analyzes abnormal changes in porosity and temperature, reduces the probability of defects, and ensures that the manufactured parts meet the specified requirements. At the same time, the monitoring of porosity and temperature helps to timely feedback and adjust process parameters, thereby improving product production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The present invention is further described with reference to the accompanying drawings. However, the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative effort.

[0032] Figure 1 Schematic diagram of the method of the present invention.

[0033] Figure 2 This is a schematic diagram of system module connections of the present invention. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0035] Reference Figure 1As shown, the first aspect of the present invention provides an artificial intelligence-based additive manufacturing process optimization method, including: obtaining material characteristic parameters of additive manufacturing, evaluating the thermal physical properties and photophysical properties of the material through the material characteristic parameters, and comprehensively analyzing to obtain material characteristic evaluation parameters.

[0036] Specifically, a comprehensive analysis is performed to obtain material property evaluation parameters. The specific analysis process is: based on the material property parameters, including the viscosity, thermal conductivity and thermal expansion coefficient of various metal materials, as well as the absorptivity and reflectivity of various metal materials, the critical viscosity of metal materials, the reference standard thermal conductivity of metal materials, the critical thermal expansion coefficient of metal materials, the critical absorptivity of metal materials and the critical reflectivity of metal materials are extracted from the additive manufacturing database, and the material thermophysical property evaluation parameters and material photophysical property evaluation parameters are obtained after processing.

[0037] It should be understood that the material thermophysical property evaluation parameters in this embodiment are quantitative indicators obtained by comprehensively analyzing the viscosity, thermal conductivity and thermal expansion coefficient of additively manufactured metal materials. They are used to quantitatively evaluate the additive manufacturing adaptability level of the thermophysical properties of additively manufactured metal materials and provide a data basis for material property evaluation.

[0038] In a specific embodiment, the material thermophysical property evaluation parameter is expressed as follows:

[0039]

[0040] Where, β R represents the evaluation parameter of the material's thermophysical properties, e represents the natural constant, and ND i Indicates the viscosity of the i-th type of metal material, RD i Represents the thermal conductivity of the i-th type of metal material, RP i represents the thermal expansion coefficient of the i-th type of metal material, ND0 represents the critical viscosity of the metal material, RD0 represents the reference standard thermal conductivity of the metal material, RP0 represents the critical thermal expansion coefficient of the metal material, ΔRD represents the set allowable deviation of the thermal conductivity of the metal material, χ1 represents the influence factor of the suitability of the thermophysical properties corresponding to the set viscosity, χ2 represents the influence factor of the suitability of the thermophysical properties corresponding to the set thermal conductivity, χ3 represents the influence factor of the suitability of the thermophysical properties corresponding to the set thermal expansion coefficient, i represents the number of each type of metal material, i = 1, 2, 3, ..., n, n represents the total number of types of metal materials.

[0041] In a specific embodiment, the evaluation parameters of the material's thermophysical properties can not only be obtained through the above calculation method, but also through computer simulation, using finite element analysis (FEA), molecular dynamics simulation and other methods to predict the thermal behavior of the material during the additive manufacturing process, including temperature distribution, thermal stress formation, phase change and other phenomena. It is also possible to obtain experimentally verified material thermophysical property data by consulting existing material databases (such as MatWeb, ASM Alloy Center, etc.). These data can be directly used to preliminarily evaluate the material's adaptability to additive manufacturing and quantify the material's thermophysical property evaluation parameters.

[0042] It should be understood that the material photophysical property evaluation parameters in this embodiment are quantitative indicators obtained by analyzing the absorptivity and reflectivity of additively manufactured metal materials, and are used to quantitatively evaluate the additive manufacturing adaptability level of the photophysical properties of additively manufactured metal materials, providing a data basis for material property evaluation.

[0043] In a specific embodiment, the material photophysical property evaluation parameter is expressed as follows:

[0044]

[0045] Where, β G Indicates the evaluation parameter of the photophysical properties of the material, XS i Indicates the absorption rate of the i-th type metal material, FS i represents the reflectivity of the i-th type of metal material, XS0 represents the critical absorptivity of the metal material, FS0 represents the critical reflectivity of the metal material, φ1 represents the suitability factor of the photophysical properties corresponding to the set absorptivity of the metal material, and φ2 represents the suitability factor of the photophysical properties corresponding to the set reflectivity of the metal material.

[0046] In a specific embodiment, the evaluation parameters of the photophysical properties of the material can be obtained not only by the above calculation method, but also by consulting academic papers, technical reports, patents and other literature to understand the photophysical behavior of known materials in the additive manufacturing process, such as light absorption, scattering, fluorescence, nonlinear optical response and other characteristics and their influence on the forming process, theoretically analyze the interaction mechanism between the photophysical properties of the material and the additive manufacturing process parameters (such as laser power, scanning speed, layer thickness, etc.), and predict their influence on the forming quality (such as accuracy, surface roughness, internal microstructure) and the final product function (such as optical transparency, color stability, photocatalytic activity, etc.). It is also possible to use light propagation models (such as Fresnel integrals and Maxwell equations solvers) to simulate the propagation path, focusing effect and light intensity distribution of the laser in multi-layer structures and complex geometric samples, evaluate the material's utilization of laser energy and possible optical non-uniformity problems, and obtain the material photophysical property evaluation parameters through comprehensive analysis.

[0047] It should be understood that in this embodiment, viscosity describes the flow resistance of the metal material in a molten state, that is, the ability of the liquid metal to resist shear deformation. In the additive manufacturing process, metal powder is heated above the melting point to form a liquid. At this time, the viscosity determines the fluidity, spreadability, and ability of the molten metal pool to fill complex geometric structures. Low viscosity is conducive to the molten metal quickly and evenly filling the printed layer, forming a good metallurgical bond, and improving forming efficiency and precision. High viscosity may lead to poor fluidity, uneven spreading, or difficulty in filling the molten pool, thereby affecting printing quality.

[0048] It should be understood that in this embodiment, thermal conductivity is a measure of the ability of a metal material to transfer heat per unit time, per unit area, and per unit temperature gradient. In the additive manufacturing process, thermal conductivity directly affects the heat distribution of the metal molten pool, the cooling rate, and the heat exchange between the molten area and the solid matrix. Metal materials with high thermal conductivity can quickly dissipate heat, causing the molten pool to solidify quickly, which helps to reduce the heat-affected zone, but may make it difficult to control the temperature of the molten pool, and may require higher power energy input to maintain the molten state. Conversely, the molten pool of low thermal conductivity materials cools more slowly, which is beneficial to prolong the metallurgical reaction time in the molten state, but may increase the heat-affected zone and increase the risk of residual stress and deformation. Appropriate thermal management strategies, including adjusting laser power, scanning speed, and path planning, often need to consider the thermal conductivity characteristics of metal materials.

[0049] It should be understood that the thermal expansion coefficient (usually referring to the linear thermal expansion coefficient) in this embodiment represents the ratio of the length of the metal material to the linear increase in a certain direction as the temperature increases when the temperature changes. Standard thermal expansion coefficient data for various types of metal materials can be obtained by querying in professional books such as the "Materials Handbook" and the "Metal Materials Handbook". These data are based on a large number of experimental measurements and industrial practice verification. During the printing process, the metal layer will generate internal stress when it cools and shrinks. If the thermal expansion coefficients between the layers do not match or the cooling rate is uneven, it may lead to a large accumulation of thermal stress, causing deformation or cracking of the parts. Therefore, it is crucial to design a reasonable cooling strategy, select a material combination with similar thermal expansion coefficients, and perform post-processing (such as heat treatment) for stress release to reduce thermal stress.

[0050] It should be understood that in this embodiment, the reflectivity and absorptivity of various metal materials are detected by spectrometer equipment. Reflectivity refers to the percentage of light energy reflected back by the surface of a metal material when a laser beam is irradiated on the surface of the metal material to the incident light energy. Metals generally have a high reflectivity, especially for lasers of a specific wavelength. The reflectivity depends on factors such as the type of metal, the wavelength of the laser, the surface condition (such as roughness, oxide layer, coating, etc.) and the angle of incidence. High reflectivity means that most of the laser energy is not absorbed by the material, but is reflected back to the system or scattered into the environment, which is not conducive to the laser processing process. Absorption refers to the proportion of light energy absorbed by the material and converted into heat energy after the laser beam is incident on the surface of the metal material to the incident light energy. In additive manufacturing, an ideal metal material should have a high absorptivity to ensure that the laser can quickly and effectively melt or heat the target area. The reflectivity and absorptivity of metal materials in additive manufacturing are key indicators to measure their response to laser energy, which directly affect processing efficiency, forming quality and process stability. By rationally selecting materials, adjusting laser parameters, and optimizing surface conditions, the utilization rate of laser energy can be effectively improved, promoting the generation of high-quality additive manufacturing parts.

[0051] Based on the material thermophysical property evaluation parameters and the material photophysical property evaluation parameters, the material characteristic evaluation parameters are obtained through comprehensive analysis.

[0052] Specifically, the material property evaluation parameters are quantitative evaluation indicators obtained by analyzing the material thermophysical property evaluation parameters and the material photophysical property evaluation parameters. They are used to quantitatively evaluate the suitability of material properties for additive manufacturing and provide a data basis for the adjustment of additive manufacturing process parameters.

[0053] Furthermore, the material property evaluation parameters are expressed as follows:

[0054]

[0055] In the formula, γ represents the material property evaluation parameter, e represents the natural constant, β R Represents the evaluation parameter of the material's thermophysical properties, β G represents the material photophysical property evaluation parameter, ω1 represents the material property evaluation impact factor corresponding to the set material thermophysical property evaluation parameter, and ω2 represents the material property evaluation impact factor corresponding to the set material photophysical property evaluation parameter.

[0056] In a specific embodiment, the material property evaluation parameters can be obtained not only through the above calculation method, but also by using specially designed software tools. Users can make judgments based on product information and material properties in preset dimensions such as materials and processes, structures and characteristics, performance and functions, production and costs. The system will automatically calculate and generate an evaluation report, giving a score or grade of the material's adaptability to additive manufacturing. Laboratory tests can also be conducted according to the procedures specified in the standard to obtain performance data of the material under additive manufacturing processes such as molten deposition, laser sintering, and electron beam melting, thereby quantifying its adaptability and obtaining material property evaluation parameters.

[0057] In a specific embodiment, analyzing the material properties of additively manufactured parts helps to accurately set the process parameters of additive manufacturing. Appropriate process parameters can ensure that the material melts evenly and solidifies well during the forming process, reduce the generation of defects, and improve the density and overall performance of the parts. At the same time, material properties can assist in optimizing the process parameters during additive manufacturing, achieve global optimization of additive manufacturing, and thus improve product quality and reliability.

[0058] The additive manufacturing process is monitored and analyzed to obtain the evaluation values ​​of abnormal porosity changes and abnormal temperature changes of additively manufactured components. The evaluation thresholds of abnormal porosity changes and abnormal temperature changes of additively manufactured components are obtained based on the matching of material property evaluation parameters. Comprehensive analysis is then conducted to obtain the evaluation index of the abnormality degree of additively manufactured components.

[0059] Specifically, the analysis obtains the evaluation value of abnormal porosity change of additively manufactured components and the evaluation value of abnormal temperature change of additively manufactured components. The specific analysis process is: deploying several time monitoring points, collecting the porosity of additively manufactured components at each time monitoring point, and extracting the critical porosity and critical porosity growth rate of additively manufactured components from the additive manufacturing database. After processing, the evaluation value of abnormal porosity change of additively manufactured components is obtained.

[0060] It should be understood that the evaluation value of abnormal porosity change of the additively manufactured component in this embodiment is a quantitative indicator obtained by analyzing the porosity and the porosity change rate during the printing process of the additively manufactured component. It is used to quantitatively evaluate the degree of abnormality of the porosity change of the additively manufactured component and provide a data basis for the evaluation of the degree of abnormality of the additively manufactured component.

[0061] In a specific embodiment, the porosity of the additively manufactured component at two adjacent time monitoring points is obtained, and the interval between the time monitoring points is extracted. A comprehensive analysis is performed to obtain an evaluation value of the abnormal change in the porosity of the additively manufactured component. The specific numerical expression is:

[0062]

[0063] Where, ε K K represents the evaluation value of abnormal porosity change of additively manufactured parts. j represents the porosity of the additive manufacturing component at the jth time monitoring point, K0 represents the critical porosity of the additive manufacturing component, and K m represents the porosity of the additively manufactured component at the mth time monitoring point, K m-1 represents the porosity of the additively manufactured component at the m-1th time monitoring point, T represents the interval between the time monitoring points, K(V)0 represents the critical porosity growth rate, υ represents the porosity abnormal change influencing factor corresponding to the set porosity, j represents the number of each time monitoring point, j = 1, 2, 3, ..., m, and m represents the total number of time monitoring points.

[0064] In a specific embodiment, the evaluation value of abnormal porosity changes in additively manufactured parts can be obtained not only by the above calculation method, but also by installing pressure, temperature or strain sensors during the printing process, collecting process parameter data, and combining machine learning algorithms to predict or diagnose porosity abnormalities in real time. The morphology and cooling rate of the molten pool during the printing process can also be monitored. Abnormal thermal behavior may indicate the formation of pores, and a comprehensive analysis can be performed to obtain the evaluation value of abnormal porosity changes in additively manufactured parts.

[0065] It should be understood that in this embodiment, the printed part is scanned non-destructively in three dimensions by X-rays to generate high-resolution images of the internal structure. The software algorithm can perform three-dimensional reconstruction and pore analysis on the CT data, accurately measure the number, size, shape and distribution of the pores, and thus obtain the porosity of the additively manufactured part. As the number of layers increases, the heat accumulation effect may cause the heat-affected zone to expand, and the difference in thermal expansion between the molten metal and the solidified layer may cause stress concentration, inducing microcracks or pores. In addition, repeated heating and cooling of the powder bed may also cause changes in thermal stress between powder particles, causing periodic fluctuations in porosity between different layers. By monitoring the porosity of additively manufactured parts during the additive manufacturing process, it can help optimize process parameters and improve the final quality of additively manufactured parts.

[0066] The temperature of each printing layer of the additively manufactured component is monitored, and the temperature of each printing layer at each time monitoring point is collected. At the same time, the reference standard cooling rate of the printing layer and the reference standard temperature difference between layers are extracted from the additive manufacturing database. After processing, the evaluation value of the abnormal temperature change of the additively manufactured component is obtained.

[0067] It should be understood that the evaluation value of abnormal temperature changes of additively manufactured components in this embodiment is a quantitative indicator obtained by analyzing the temperature change conditions of each printed layer, which is used to quantitatively evaluate the degree of abnormality of temperature changes of additively manufactured components and provide a data basis for the evaluation of the degree of abnormality of additively manufactured components.

[0068] In a specific embodiment, the temperature of each printing layer at adjacent time monitoring points is obtained, and a comprehensive analysis is performed to obtain an evaluation value of abnormal temperature change of the additive manufacturing component. The specific numerical expression is:

[0069]

[0070] Where, ε W Indicates the evaluation value of abnormal temperature change of additively manufactured parts, W(V) r represents the cooling rate of the rth printing layer, W r→m Indicates the temperature of the rth printing layer at the mth time monitoring point, W r→m-1 represents the temperature of the rth printing layer at the m-1th time monitoring point, W(V)0 represents the reference standard cooling rate of the printing layer, W r+1 represents the temperature of the r+1th printing layer at the adjacent time monitoring point, W r represents the temperature of the rth printing layer at the adjacent time monitoring point, W0 represents the reference standard interlayer temperature difference, ΔW represents the set interlayer temperature difference allowable deviation value, τ1 represents the temperature abnormal change impact factor corresponding to the set interlayer temperature difference, τ2 represents the temperature abnormal change impact factor corresponding to the set cooling rate, r represents the number of each printing layer, r = 1, 2, 3, ..., h, and h represents the total number of printing layers.

[0071] In a specific embodiment, the evaluation value of abnormal temperature changes of additively manufactured components can not only be obtained through the above calculation method, but also by continuously tracking the changes in process parameters such as laser power, scanning path, layer thickness, cooling conditions, and analyzing their impact on temperature. When the parameters deviate from the set range, there may be a risk of causing temperature anomalies. It is also possible to combine historical data with real-time monitoring data to train machine learning models (such as neural networks, support vector machines, etc.) to predict temperature change trends, and to provide early warning of possible temperature anomalies, and to obtain the evaluation value of abnormal temperature changes of additively manufactured components through analysis.

[0072] It should be understood that in this embodiment, a print layer is the basic unit used in the 3D printing process to construct an object by stacking layers from the bottom up according to the design model. Each layer represents a 2D projection of the 3D model at a specific height. Its thickness is one of the additive manufacturing process parameters and can be precisely set based on actual needs and equipment capabilities.

[0073] It should be understood that in this embodiment, infrared thermal imaging is used to monitor the temperature of each printed layer of an AM component. The cooling rate of each printed layer can affect the quality of the AM component. If a printed layer cools too quickly (e.g., due to excessive cooling air velocity, low ambient temperature, or high material thermal conductivity), it can lead to large temperature gradients within and around the melt pool, resulting in high thermal stresses and increasing the risk of component cracking or warping. Rapid cooling can also lead to insufficient microstructural refinement, affecting material properties. Conversely, if the cooling rate is too slow (e.g., due to poor heat dissipation, high material heat capacity, or excessive heat accumulation between layers), the melt pool can remain at a high temperature for too long, potentially resulting in an excessively large heat-affected zone between adjacent layers, affecting interlayer metallurgical bonding and promoting element diffusion, altering the alloy composition distribution. Furthermore, abnormal thermal gradients within the printed layer can also affect the quality of the AM component. Differences in thermal expansion coefficients between different metal materials or the same material in different directions can lead to uneven thermal expansion forces during printing, causing component deformation. As the number of printed layers increases, accumulated thermal stresses can cause component warping, distortion, or overall deformation.

[0074] In a specific embodiment, by monitoring the porosity and temperature of additively manufactured parts during the additive manufacturing process, analyzing abnormal changes in porosity and temperature, reducing the probability of defects, and ensuring that the manufactured parts meet specified requirements, the monitoring of porosity and temperature helps to provide timely feedback and adjustment of process parameters, thereby improving product production efficiency.

[0075] Specifically, the abnormal porosity change assessment threshold of the additively manufactured component and the abnormal temperature change assessment threshold of the additively manufactured component are obtained according to the matching of the material property evaluation parameters, and the abnormal degree assessment index of the additively manufactured component is obtained by comprehensive analysis. The specific analysis process is: matching the material property evaluation parameters with the abnormal porosity change assessment threshold of the additively manufactured component and the abnormal temperature change assessment threshold of the additively manufactured component corresponding to each material property evaluation parameter interval stored in the additive manufacturing database to obtain the abnormal porosity change assessment threshold of the additively manufactured component and the abnormal temperature change assessment threshold of the additively manufactured component, and according to the abnormal porosity change assessment value of the additively manufactured component and the abnormal temperature change assessment value of the additively manufactured component, a comprehensive analysis is performed to obtain the abnormal degree assessment index of the additively manufactured component.

[0076] Specifically, the additive manufacturing component abnormality assessment index is a quantitative indicator obtained by comprehensively analyzing the abnormal changes in porosity and temperature of additive manufacturing components. It is used to evaluate the abnormality of additive manufacturing components and provide a data basis for the adjustment of additive manufacturing process parameters.

[0077] Furthermore, the specific numerical expression of the additive manufacturing component abnormality evaluation index is:

[0078]

[0079] Where δ represents the abnormality evaluation index of additive manufacturing parts, e represents the natural constant, and ε K represents the evaluation value of abnormal porosity change of additively manufactured parts, ε W represents the evaluation value of abnormal temperature change of additively manufactured parts, ε K0 represents the evaluation threshold of abnormal porosity change of additively manufactured parts, ε W0 represents the threshold for evaluating abnormal temperature changes of additively manufactured components, ψ1 represents the influencing factor of the degree of abnormality of the additively manufactured components corresponding to the set evaluation value of abnormal porosity changes of the additively manufactured components, and ψ2 represents the influencing factor of the degree of abnormality of the additively manufactured components corresponding to the set evaluation value of abnormal temperature changes of the additively manufactured components.

[0080] In a specific embodiment, the abnormality assessment index of an additively manufactured component can not only be obtained by the above calculation method, but also by using advanced additive manufacturing process simulation software (such as ANSYS Additive Suite, ExaSIM, etc.) to numerically simulate the entire forming process before printing, predict possible defects (such as deformation caused by thermal stress, voids caused by insufficient melting, poor interlayer bonding caused by uneven powder bed, etc.), and provide an expected abnormality assessment value as a reference in the actual manufacturing process. In addition, non-destructive testing technologies (such as ultrasonic testing, X-ray computed tomography (CT), magnetic particle inspection, eddy current testing, etc.) can be used during the printing process to perform real-time or near-real-time internal defect detection on the formed part. These technologies can reveal hidden defects such as pores, cracks, inclusions, etc., and quantify the abnormality of the component based on information such as the number, size, location and type of defects to obtain the abnormality assessment index of the additively manufactured component.

[0081] The additive manufacturing process parameters are adjusted according to the additive manufacturing component abnormality evaluation index.

[0082] Specifically, the additive manufacturing process parameters are adjusted according to the additive manufacturing component abnormality evaluation index. The specific analysis process is: matching the additive manufacturing component abnormality evaluation index with the laser power and printing layer thickness corresponding to the additive manufacturing component abnormality evaluation index interval stored in the additive manufacturing database to obtain the appropriate laser power and appropriate printing layer thickness of the additive manufacturing component, and adjusting the additive manufacturing laser power and printing layer thickness in real time according to the appropriate laser power and appropriate printing layer thickness.

[0083] It should be understood that in this embodiment, the additive manufacturing laser power is adjusted. Laser power directly affects the ability of the metal powder to fully melt and the melting rate. Higher power can heat and melt the powder more quickly, improving forming efficiency. However, excessive power may lead to problems such as excessive melting, spattering, and increased porosity. Appropriate power adjustment helps to achieve the ideal melt pool size and morphology (such as depth, width, and shape), ensuring good fusion and solidification.

[0084] It should be understood that in this embodiment, the thickness of the printing layer of additive manufacturing is adjusted, and the thickness of each layer of powder deposition is set in the equipment software, usually in microns. This thickness is determined by the metal powder layer that the powder spreading system evenly spreads on the melted layer before each layer is built. The printing layer thickness should be matched with the laser scanning strategy (such as scanning speed, scanning spacing, filling mode, etc.) and the laser power to ensure that each layer can be fully melted and well combined with the next layer. Smaller layer thickness can improve the surface quality and dimensional accuracy of the formed part, because thinner layers allow the laser to shape details and contours more finely and reduce step effects. However, layers that are too thin may result in a significant increase in manufacturing time.

[0085] Reference Figure 2 As shown, the second aspect of the present invention provides an artificial intelligence-based additive manufacturing process optimization system, including: a material property analysis module, an additive manufacturing monitoring module, a process parameter adjustment module and an additive manufacturing database.

[0086] The material property analysis module is used to obtain material property parameters for additive manufacturing, evaluate the material's thermal physical properties and material's photophysical properties through the material property parameters, and obtain material property evaluation parameters through comprehensive analysis.

[0087] The additive manufacturing monitoring module is used to monitor the additive manufacturing process, analyze and obtain an assessment value of abnormal porosity changes and an assessment value of abnormal temperature changes of additively manufactured components, and obtain an assessment threshold value of abnormal porosity changes and an assessment threshold value of abnormal temperature changes of additively manufactured components based on material property assessment parameters, and comprehensively analyze and obtain an assessment index of the degree of abnormality of additively manufactured components.

[0088] The process parameter adjustment module is used to adjust the additive manufacturing process parameters according to the additive manufacturing component abnormality degree evaluation index.

[0089] The additive manufacturing database is used to store additive manufacturing related data, including critical viscosity of metal materials, reference standard thermal conductivity of metal materials, critical thermal expansion coefficient of metal materials, critical absorptivity of metal materials, critical reflectivity of metal materials, critical porosity and critical porosity growth rate of additively manufactured parts, reference standard cooling rate of printed layers and reference standard interlayer temperature difference, abnormal porosity change assessment thresholds and abnormal temperature change assessment thresholds of additively manufactured parts corresponding to each material property evaluation parameter interval, and laser power and printing layer thickness corresponding to each additively manufactured part abnormality assessment index interval.

[0090] In a specific embodiment, the present invention provides an artificial intelligence-based additive manufacturing process optimization method and system to analyze the material properties of additively manufactured components and monitor the porosity and temperature of the additively manufactured components during the additive manufacturing process, thereby realizing automated control and decision-making of the additive manufacturing process, reducing the burden of manual intervention, and improving production efficiency. At the same time, the process parameters are adjusted in real time according to the material properties, porosity and temperature of the additively manufactured components to ensure the stability of part forming and reduce the workload of post-processing.

[0091] The above content is merely an example and explanation of the structure of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. An artificial intelligence-based additive manufacturing process optimization method, characterized in that: include: Obtain material characteristic parameters for additive manufacturing, including viscosity, thermal conductivity, thermal expansion coefficient, absorptivity, and reflectivity of various metal materials. Extract the critical viscosity, reference standard thermal conductivity, critical thermal expansion coefficient, critical absorptivity, and critical reflectivity of metal materials from the additive manufacturing database. After processing, obtain material thermophysical property evaluation parameters and material photophysical property evaluation parameters respectively. According to the material thermophysical property evaluation parameters and the material photophysical property evaluation parameters, the material characteristic evaluation parameters are obtained through comprehensive analysis; Monitor the additive manufacturing process, collect the porosity of additively manufactured components at each monitoring point, and obtain the abnormal change assessment value ε of the additively manufactured components through processing. K , specifically: Where K j represents the porosity of the additive manufacturing component at the jth time monitoring point, K0 represents the critical porosity of the additive manufacturing component, and K m represents the porosity of the additively manufactured component at the mth time monitoring point, K m-1 represents the porosity of the additive manufacturing component at the m-1th time monitoring point, T represents the interval between time monitoring points, K(V)0 represents the critical porosity growth rate, υ represents the porosity abnormal change influencing factor corresponding to the set porosity, j represents the number of each time monitoring point, j = 1, 2, 3, ..., m, m represents the total number of time monitoring points; K0 and K(V)0 are extracted from the additive manufacturing database; Monitor the temperature of each printing layer of the additive manufacturing component, collect the temperature of each printing layer at each monitoring point, and obtain the abnormal temperature change assessment value ε of the additive manufacturing component after processing. W , specifically: Where W(V) r represents the cooling rate of the rth printing layer, W r→m Indicates the temperature of the rth printing layer at the mth time monitoring point, W r→m-1 represents the temperature of the rth printing layer at the m-1th time monitoring point, W(V)0 represents the reference standard cooling rate of the printing layer, W r+1 represents the temperature of the r+1th printing layer at the adjacent time monitoring point, W r represents the temperature of the rth printing layer at the adjacent time monitoring point, represents the reference standard interlayer temperature difference, ΔW represents the set interlayer temperature difference allowable deviation value, τ1 represents the temperature abnormal change influence factor corresponding to the set interlayer temperature difference, τ2 represents the temperature abnormal change influence factor corresponding to the set cooling rate, r represents the number of each printing layer, r=1,2,3,...,h, h represents the total number of printing layers; W(V)0 and Extracted from the Additive Manufacturing Database; Matching the material property evaluation parameters with the porosity abnormal change evaluation threshold and the temperature abnormal change evaluation threshold corresponding to each material property evaluation parameter interval stored in the additive manufacturing database to obtain the porosity abnormal change evaluation threshold and the temperature abnormal change evaluation threshold. Based on the porosity abnormal change evaluation value and the temperature abnormal change evaluation value, a comprehensive analysis is performed to obtain the additive manufacturing component abnormality degree evaluation index; The additive manufacturing process parameters are adjusted according to the abnormality evaluation index.

2. The artificial intelligence-based additive manufacturing process optimization method according to claim 1, characterized in that: The additive manufacturing process parameters are adjusted according to the abnormality evaluation index, and the specific analysis process is as follows: The abnormality evaluation index of the additive manufacturing component is matched with the laser power and printing layer thickness corresponding to the abnormality evaluation index interval of each additive manufacturing component stored in the additive manufacturing database to obtain the appropriate laser power and appropriate printing layer thickness of the additive manufacturing component, and the laser power and printing layer thickness of the additive manufacturing are adjusted in real time according to the appropriate laser power and appropriate printing layer thickness.

3. The artificial intelligence-based additive manufacturing process optimization method according to claim 1, characterized in that: The material property evaluation parameters are quantitative evaluation indicators obtained by analyzing the material thermophysical property evaluation parameters and the material photophysical property evaluation parameters, and are used to quantitatively evaluate the suitability of the material properties for additive manufacturing.

4. The artificial intelligence-based additive manufacturing process optimization method according to claim 1, characterized in that: The additive manufacturing component abnormality evaluation index is a quantitative indicator obtained by comprehensively analyzing abnormal changes in porosity and temperature of the additive manufacturing component, and is used to evaluate the abnormality degree of the additive manufacturing component.

5. The artificial intelligence-based additive manufacturing process optimization method according to claim 1, characterized in that: The material property evaluation parameter has the following specific numerical expression: In the formula, γ represents the material property evaluation parameter, e represents the natural constant, β R Represents the evaluation parameter of the material's thermophysical properties, β G represents the material photophysical property evaluation parameter, ω1 represents the material property evaluation impact factor corresponding to the set material thermophysical property evaluation parameter, and ω2 represents the material property evaluation impact factor corresponding to the set material photophysical property evaluation parameter.

6. The artificial intelligence-based additive manufacturing process optimization method according to claim 1, characterized in that: The specific numerical expression of the additive manufacturing component abnormality evaluation index is: Where δ represents the abnormality evaluation index of additive manufacturing parts, e represents the natural constant, and ε K represents the evaluation value of abnormal porosity change of additively manufactured parts, ε W represents the evaluation value of abnormal temperature change of additively manufactured parts, ε K0 represents the evaluation threshold of abnormal porosity change of additively manufactured parts, ε W0 represents the threshold for evaluating abnormal temperature changes of additively manufactured components, ψ1 represents the influencing factor of the degree of abnormality of the additively manufactured components corresponding to the set evaluation value of abnormal porosity changes of the additively manufactured components, and ψ2 represents the influencing factor of the degree of abnormality of the additively manufactured components corresponding to the set evaluation value of abnormal temperature changes of the additively manufactured components.

7. An artificial intelligence-based additive manufacturing process optimization system, applying the artificial intelligence-based additive manufacturing process optimization method according to any one of claims 1 to 6, characterized in that: include: The material property analysis module is used to obtain the material property parameters of additive manufacturing, evaluate the material's thermal and photophysical properties through the material property parameters, and obtain the material property evaluation parameters through comprehensive analysis; The additive manufacturing monitoring module is used to monitor the additive manufacturing process, analyze and obtain the assessment value of abnormal porosity change and abnormal temperature change of additive manufacturing components, and obtain the assessment threshold value of abnormal porosity change and abnormal temperature change of additive manufacturing components based on the matching of material characteristic assessment parameters. Comprehensive analysis is then conducted to obtain the assessment index of the abnormality degree of additive manufacturing components. The process parameter adjustment module is used to adjust the additive manufacturing process parameters according to the abnormality evaluation index of the additive manufacturing component.

Images