Process / structure / performance analytical model verification method and device

Abstract

The invention provides a process / structure / performance analytical model verification method and device. The method provided by the invention comprises the following steps: determining a to-be-tested process condition of a process / structure / performance analytical model; determining a laser power grade and a scanning speed grade based on the to-be-detected process conditions; making a to-be-scanned sample; screening a candidate scanning area from the to-be-scanned sample; grain feature information is extracted based on the candidate scanning area image; calculating the material hardness according to the grain size information; and combining the material hardness, the grain hardness information and corresponding to-be-tested process conditions to obtain an evaluation data set, and evaluating the performance of the analytical model based on the evaluation data set. The process / structure / performance analytical model verification method and device provided by the invention are used for carrying out performance verification on the process / structure / performance analytical model, so that the reliability and accuracy of the process / structure / performance analytical model in actual process production application are ensured.

Description

Technical Field

[0001] The present application relates to the field of additive manufacturing technology, and in particular to a method and device for verifying an analytical model of a process / structure / performance. Background Art

[0002] In the fields of manufacturing and materials science, PSP (Process-Structure-Property) analytical models are widely used to understand and predict the relationship between the processing technology, microstructure and final performance of materials. Traditional PSP analytical models often rely on complex numerical models and data-driven models to simulate and optimize these processes. First, numerical models require a lot of computing resources and time, especially when dealing with complex multi-physics field problems. They are not suitable for real-time optimization and online control when process parameters are frequently adjusted. Secondly, data-driven models rely on a large amount of historical data, but the reliability and coverage of the data may be insufficient, and there is an overfitting problem, making it difficult to cope with unknown process conditions. In addition, traditional PSP analytical models are usually static and cannot adapt to dynamically changing production environments.

[0003] In order to solve these problems, a PSP analytical model constructed based on continuous growth limiting factors, interdependence models and Hall-Petch relationships provides a new solution. Compared with traditional numerical models, the analytical model provided in this application greatly improves the computational efficiency, can quickly predict under different process parameters, and is suitable for real-time adjustment and optimization in online production. In addition, the combination of the interdependence model further improves the prediction accuracy, can better simulate the interaction between process parameters, and is applicable to a variety of materials and process conditions. However, although the PSP analytical model of this application has significant advantages in computational efficiency and adaptability, in order to ensure its reliability and accuracy in practical applications, its performance must be verified, especially in new processes and material systems.

[0004] Therefore, there is an urgent need for a method to verify the performance of the analytical model of process / structure / performance to ensure the reliability and accuracy of the analytical model of process / structure / performance in actual process production applications. Summary of the invention

[0005] In view of this, the present application provides a method and device for verifying an analytical model of process / structure / performance, which is used to perform performance verification on the analytical model of process / structure / performance to ensure the reliability and accuracy of the analytical model of process / structure / performance in actual process production applications.

[0006] Specifically, the present application is implemented through the following technical solutions: In a first aspect, the present application provides a method for verifying an analytical model of a process / structure / performance, the method comprising: Determine the process conditions to be tested of the analytical model of process / structure / performance, wherein there are multiple process conditions to be tested, and the analytical model is used to output the material hardness based on the process conditions to be tested; Determine the laser power level and the scanning speed level based on the process conditions to be tested, wherein the laser power level and the scanning speed level of each process condition to be tested are different, predict the microstructure change value under each process condition to be tested based on the analytical model, and match the corresponding laser power level and the scanning speed level based on the difference of the microstructure change values ​​under different process conditions to be tested; Prepare samples to be scanned; Screening candidate scanning areas from the sample to be scanned, wherein the candidate molten pool boundary area is located according to the microstructure change value predicted by the analytical model, and the scanning step length and the observation area corresponding to the scanning point are determined based on the influence of each of the process conditions to be tested on the sample to be scanned; Extracting grain feature information based on the candidate scanning area image; the grain feature information at least includes grain size information and grain hardness information; calculating the material hardness based on the grain size information; The material hardness, the grain hardness information and the corresponding process conditions to be tested are combined to obtain an evaluation data set, and the performance of the analytical model is evaluated based on the evaluation data set.

[0007] The second aspect of the present application provides a device for verifying an analytical model of a process / structure / performance, the device comprising a determination module, an acquisition module, a screening module, a calculation module and an evaluation module; wherein: The determination module is used to determine the process conditions to be tested of the analytical model of process / structure / performance, wherein there are multiple process conditions to be tested, and the analytical model is used to output the material hardness based on the process conditions to be tested; The determination module is further used to determine the laser power level and the scanning speed level based on the process conditions to be tested, wherein the laser power level and the scanning speed level of each process condition to be tested are different, and the microstructure change value under each process condition to be tested is predicted based on the analytical model, and the corresponding laser power level and the scanning speed level are matched based on the difference of the microstructure change values ​​under different process conditions to be tested; The acquisition module is used to prepare a sample to be scanned; The screening module is used to screen candidate scanning areas from the sample to be scanned, wherein the candidate area of ​​the molten pool boundary is located according to the microstructure change value predicted by the analytical model, and the scanning step length and the observation area corresponding to the scanning point are determined based on the influence of each of the process conditions to be tested on the sample to be scanned; The acquisition module is further used to extract grain feature information based on the candidate scanning area image; the grain feature information at least includes grain size information and grain hardness information; The calculation module is used to calculate the material hardness according to the grain size information; The evaluation module is used to combine the material hardness, the grain hardness information and the corresponding process conditions to be tested to obtain an evaluation data set, and evaluate the performance of the analytical model based on the evaluation data set.

[0008] The analytical model verification method and device for process / structure / performance provided in the present application, in the first aspect, by comparing and analyzing the material hardness predicted by the analytical model of process / structure / performance with the grain hardness information obtained by experiment, the performance of the analytical model of process / structure / performance can be fully evaluated. First, the material hardness predicted by the analytical model of process / structure / performance is calculated based on the grain size information through the Hall-Page law, which reflects the modeling ability of the analytical model for the relationship between microscopic features and macroscopic performance, while the grain hardness information is the experimental data directly measured, which has high authenticity and accuracy. The comparison between the two can intuitively reflect the deviation and applicability of the analytical model in predicting material properties, and further verify the rationality of the analytical model by analyzing whether the trend predicted by the analytical model is consistent with the experimental results. This evaluation method can not only verify the accuracy of the analytical model, but also discover the potential limitations of the analytical model, provide a scientific basis for process optimization and model improvement, thereby improving the reliability and wide applicability of the analytical model in practical applications. On the second aspect, since the sample to be scanned usually contains multiple different regions, the grain structure and performance of these regions may be significantly different due to factors such as processing conditions, heat conduction, and material composition. This application locates the candidate area of ​​the molten pool boundary according to the microstructure change value predicted by the analytical model, and can focus on the representative and significantly changed area, so that the observation range is concentrated in the key area, and can fully capture the influence of the process conditions on the microstructure, so as to clearly reflect the overall picture of the structural change. In this way, the data deviation caused by the interference of irrelevant areas can be effectively avoided, the accuracy of the observation results can be improved, and the high adaptability of the experimental data to the analytical model verification can be ensured. Secondly, by determining the scanning step length and the position of the scanning point based on the different degrees of influence of each process condition to be tested on the sample, efficient and accurate scanning can be achieved. The optimized setting of the scanning step length and the point position can reduce the invalid area of ​​the scan while ensuring the accuracy of the experimental data, thereby greatly improving the experimental efficiency. In addition, this method of precise positioning can avoid waste of resources, reduce the complexity of experimental data processing, and make the entire verification process more efficient and reliable. By selecting a representative candidate scanning area, data distortion caused by local special phenomena can be avoided, ensuring that the experimental data can reflect the typical characteristics of the entire sample or processing process, thereby enhancing the reliability of analytical model verification. At the same time, according to the verification requirements of the analytical model of process / structure / performance, it is only necessary to focus on certain key areas, such as the center area of ​​the weld, the heat-affected zone or the inter-grain connection area, which are usually sensitive points of material performance changes and the focus of process optimization. Selecting these areas as candidate scanning areas can not only effectively reduce the experimental workload and improve the experimental efficiency, but also ensure that the collected data better matches the analytical model verification target and avoids the interference of irrelevant data on the experimental results.Thirdly, verification data is obtained based on the laser power level and scanning speed level determined by the process conditions to be tested, ensuring that the analytical model verification process is highly relevant to the actual production environment. This avoids the gap between laboratory conditions and industrial applications that may exist in traditional experimental methods, making the verification of analytical models closer to real production scenarios. In the process of preparing samples to be scanned and screening candidate scanning areas, the reliability and accuracy of experimental data are further enhanced by ensuring the actual representativeness of the samples and selecting appropriate areas for testing. This process not only improves the accessibility of data by extracting grain feature information (such as grain size and hardness information) in combination with images, but also provides a real and efficient basis for subsequent hardness calculations. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A flowchart of the analytical model verification method for process / structure / performance provided in Example 1 of the present application; Figure 2 This is a schematic diagram of the structure of the analytical model verification device for process / structure / performance provided in Example 2 of the present application. DETAILED DESCRIPTION

[0010] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0011] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items.

[0012] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0013] Specific embodiments are given below to introduce the technical solution of the present application in detail.

[0014] Figure 1This is a flow chart of the analytical model verification method for the process / structure / performance provided in Example 1 of the present application. Figure 1 , the method provided in this embodiment may include: S101. Determine a process condition to be tested of an analytical model of process / structure / performance, wherein there are multiple process conditions to be tested, and the analytical model is used to output material hardness based on the process conditions to be tested.

[0015] Specifically, the process / structure / performance analytical model is a multi-level, multi-variable analytical framework used to explore and understand the relationship between the process conditions, the resulting micro or macro structure, and the final performance of materials or products during processing. The process / structure / performance analytical model is used to optimize the processing conditions to obtain the desired product performance. The process / structure / performance analytical model establishes a quantitative relationship between process, structure, and performance through a mathematical model. Since this quantitative relationship is usually complex and involves the interaction of multiple factors, a large amount of experimental data is required to verify the performance of the process / structure / performance analytical model.

[0016] Furthermore, the analytical model of process / structure / performance uses a continuous growth limiting factor to replace the traditional grain growth assumption, uses an interdependence model to characterize the growth relationship between grains, and uses the Hall-Page law to replace the traditional performance prediction method; the analytical model is used to determine the material hardness based on the grain size information corresponding to different process conditions to be tested, and adjust the process conditions to be tested based on the difference between the material hardness and the target hardness.

[0017] The analytical model of process / structure / performance associates the process conditions to be tested with the grain size information, and the continuous growth limiting factor is inversely proportional to the grain size information; the Hall-Page law associates the grain size information with the material hardness, and the reciprocal square root of the grain size information varies linearly with the material hardness.

[0018] Specifically, the analytical model of process / structure / performance is constructed based on continuous growth limiting factors, interdependence model and Hall-Page law. The analytical model of process / structure / performance predicts the material hardness based on the grain size information under different process conditions to be tested, and adjusts and optimizes the process conditions to be tested based on the difference between the material hardness and the target hardness corresponding to the processing requirements. Continuous growth limiting factors refer to limiting factors that affect the growth of grains under specific process conditions, including temperature, time, stress state, etc., and conditions that limit grain growth during heat treatment or other processing. Continuous growth limiting factors will affect the final size of the material grains, usually by limiting the grain growth rate to control the grain size, thereby affecting the final material properties. Traditional grain growth assumptions are usually based on simple grain growth laws or empirical formulas, and do not fully consider the dynamic effects of external conditions and multiple limiting factors on grain growth.

[0019] The continuous growth limiting factor describes the change of cooling rate on the solidification path. The tangent of the continuous growth limiting factor curve represents the initial development rate of cooling rate at different nucleation locations. The continuous growth limiting factor can be expressed as: ; ; ; Among them, the is a continuous growth limiting factor; is the liquidus slope; is the solute concentration; is the distribution coefficient; is the solidification path; is the diffusion coefficient of the solute in the liquid; is the solidification speed; is the solute diffusion length before the S / L interface; is the laser scanning speed; is the angle between the laser scanning direction and the grain growth direction.

[0020] Furthermore, since the traditional Q model (traditional growth limiting factor Q model) is limited by the following factors, there are limitations when constructing a process-structure module based on the traditional Q model. First, the temperature field and cooling rate are assumed to be uniform: the traditional Q model assumes that the temperature and cooling rate are uniform throughout the molten pool, ignoring the differences in local temperature gradients and cooling rates, and cannot reflect the local solidification characteristics of the molten pool. Second, the influence of alloy composition changes and multiphase structures is ignored: the influence of factors such as alloy composition, phase change and grain morphology on the solidification process is not considered. Third, oversimplified grain growth description: the traditional model simplifies grain growth into a single limiting factor Q, ignoring the interaction of multiple factors such as nucleation rate and cooling rate. Fourth, the dynamic changes of the solidification process are not considered: the traditional Q model does not capture the impact of dynamic changes such as temperature and flow on grain growth. In order to overcome the limitations of the traditional Q model, a continuous growth limiting factor is used. Instead of Q in the traditional Q model, the average grain size of the molten pool solidification structure is developed as Varying grain size, i.e.: ; ; Among them, the is the average grain size; is the diffusion coefficient of the solute in the liquid; is the solidification speed; is a continuous growth limiting factor; The increment of supercooling required to activate subsequent nucleation events; is the critical nucleation undercooling of the material; is the solute concentration; is the distribution coefficient; is the additional distance to the next most effective nucleation particle; is the solute distribution on the liquid side before the solid-liquid interface ( less than 1).

[0021] It should be noted that in the traditional Q model, Q as a growth limiting factor mainly relies on the assumption of uniformity of cooling rate, but it fails to accurately capture the dynamic changes of process parameters during solidification, especially the influence of factors such as local temperature gradient and molten pool flow. In order to overcome this limitation, the introduction of Qc as a function of cooling rate is an effective improvement based on the same theoretical basis. The core similarity between Qc and Q is that they both reflect the influence of cooling rate on the grain growth process, but Qc considers the characteristics of cooling rate changing with solidification path, rather than a fixed growth limiting factor. This makes Qc more flexible and accurate in describing the dynamic evolution of grain growth during solidification. In addition, there is a significant linear relationship between grain size and 1 / Qc, and Qc can more accurately reflect the change of grain size, avoiding the oversimplified assumptions in the traditional Q model. The adaptability of Qc is further enhanced by introducing the interdependence theory and the description of the continuous solidification path. The interdependence theory shows that in the solidification process, factors such as temperature, flow, and composition interact with each other, and changing any factor will affect the behavior of grain growth. By taking these dynamic changes into account, Qc can more realistically capture the changes in grain size under different process conditions, rather than just the effect of a single cooling rate. Compared to the traditional Q model, Qc can not only more accurately describe the effect of cooling rate on grain size, but also take into account the non-uniformity of local temperature and composition of the molten pool, further improving the prediction accuracy. By building a "PS" module through Qc, it is possible to optimize grain size and microstructure according to different process parameters to achieve higher performance material preparation. As an improved form of growth limiting factor, Qc not only inherits the functions of the traditional Q model in theory, but also provides a more accurate tool by considering dynamic cooling rate, changes in solidification path and the interaction of multiple factors. It is particularly suitable for modeling and predicting grain growth behavior under complex process conditions.

[0022] Furthermore, the interdependence model describes the complex relationship between grains, including grain competition, co-growth, spatial confinement, etc., and can more accurately reflect the growth behavior of grains in a complex process environment. Most traditional models regard the influence between grains as independent or only model them in a simple competitive manner.

[0023] The Hall-Page law characterizes the quantitative relationship between material hardness and grain size. There is an inverse relationship between material hardness and grain size, that is, the smaller the grain size, the greater the material hardness. Traditional performance prediction methods are mostly based on linear or empirical formulas to estimate the relationship between grain size and performance, ignoring the nonlinear effect of microstructure on macroscopic performance. The Hall-Page law can be expressed as: ; Among them, the is the material hardness; is the inherent hardness; is the Hall-Page coefficient; is the grain size; , Determined by experiments, it is a fixed value (the Hall-Page coefficient and fixed hardness corresponding to each material remain unchanged).

[0024] In specific implementation, the process / structure / performance analytical model first determines the corresponding grain size information according to different process conditions (such as temperature, cooling rate, time, etc.). Based on the Hall-Page law, the process / structure / performance model uses the calculated grain size information to calculate the material hardness, and adjusts the process conditions (such as heating time, cooling rate, pressure, etc.) through feedback on the material hardness prediction to achieve control of the material microstructure and ultimately achieve the desired material properties. The process / structure / performance analytical model predicts the material microstructure evolution and material hardness under the process conditions to be tested, and optimizes the production process.

[0025] Furthermore, there are multiple process conditions to be tested, including material properties, process parameters, structural characteristics, performance requirements, etc. Among them, material properties include the selected material type (such as metal, polymer, ceramic, etc.), the initial properties of the material (such as hardness, elastic modulus, thermal conductivity, etc.), and the material processing method (such as heat treatment, surface treatment, etc.). Process parameters include the type of processing process (such as milling, laser cutting, 3D printing), the setting of process conditions (such as temperature, pressure, time, etc.), processing speed, feed rate, cutting force, etc. Structural characteristics include the geometric shape of the product (such as size, thickness, surface roughness, etc.), the complexity of the structure (such as multilayer structure, pore structure, mesh structure, etc.). Performance requirements include target performance indicators (such as mechanical properties, thermal properties, electrical properties, optical properties, etc.), strength, wear resistance, corrosion resistance, thermal stability and other specific requirements.

[0026] In specific implementation, when determining the process conditions to be tested, it is necessary to comprehensively consider multiple aspects such as material properties, process parameters, structural properties, performance requirements, etc. Therefore, multiple process conditions to be tested are determined. When collecting experimental data (grain characteristic information) based on the process conditions to be tested in subsequent embodiments, it is also necessary to conduct experiments separately under multiple process conditions to be tested, that is, control the experimental environment to be carried out under multiple groups of process conditions to be tested, to ensure a comprehensive analysis of the performance of the analytical model of process / structure / performance under different process conditions to be tested.

[0027] It should be noted that, since there are many process conditions involved in additive manufacturing, in this embodiment, only the laser power level and the scanning speed level are used as the main research objects for explanation, and other process conditions are controlled to remain consistent.

[0028] The method provided in this embodiment effectively opens up the relationship between process conditions, grain size and material properties by introducing the analytical model of process / structure / performance constructed by continuous growth limiting factors, and this relationship is based on the same physical basis, which makes the analytical model have higher accuracy and applicability in practical applications. First, the fixed growth limiting factor in the traditional Q model is replaced by the continuous growth limiting factor, and the dynamic characteristics of grain growth are described by the continuous growth limiting factor, which enables the analytical model to more accurately capture the influence of factors such as temperature, time, and stress state on grain growth. By considering the changes in cooling rate and solidification path, the continuous growth limiting factor can adapt to grain growth under different process conditions, ensuring the precise connection between process, organization and performance. Secondly, the interdependence model further enhances the practical adaptability of the analytical model. It takes into account the mutual influence between grains and the dynamic behavior of grain boundaries. This refined modeling method can effectively reflect the complex interaction of multiple factors in the grain growth process. The Hall-Page law provides reliable theoretical support for material performance prediction by quantifying the relationship between grain size and material hardness. This physical relationship makes hardness prediction more accurate and avoids the limitations of traditional empirical formulas. The analytical model can adjust the process conditions based on the difference between the material hardness and the target hardness, thereby optimizing the process parameters so that the material performance meets the expected requirements. By accurately predicting the grain size under different process conditions, the analytical model can flexibly adjust the processing conditions according to the material hardness prediction results, avoiding the fixed assumptions and over-simplification in traditional methods. Unlike the traditional grain growth hypothesis, the analytical model is based on a more dynamic and detailed theory, which can capture the comprehensive impact of process conditions on grain growth and provide more accurate material performance regulation. In summary, the analytical model of process / structure / performance not only breaks the limitations of traditional methods by introducing a dynamic model based on physics, but also provides a more flexible and accurate process optimization method, which enables it to have higher predictive ability and practical guidance in practical applications.

[0029] S102, determining a laser power level and a scanning speed level based on the process conditions to be measured.

[0030] Among them, the laser power level and scanning speed level of each of the process conditions to be tested are different, and the microstructure change value under each of the process conditions to be tested is predicted based on the analytical model, and the corresponding laser power level and scanning speed level are matched based on the difference in the microstructure change values ​​under different process conditions to be tested.

[0031] Specifically, the laser power level and the scanning speed level are different ranges or categories of two important process parameters in the laser processing process, which directly affect the processing effect of the material. Among them, the laser power level refers to the power output of the laser. In laser processing, the power determines the energy density of the laser beam, which in turn affects the heating, melting, cutting or surface treatment of the material. The laser power level is usually characterized by being divided into several levels based on different ranges of laser power. For example, the laser power can be divided into low, medium and high levels according to different power ranges. The scanning speed level refers to the speed at which the laser beam moves along the processing path, which is usually divided into different speed levels. The scanning speed level is usually classified according to different ranges of scanning speed. For example, the scanning speed is divided into low, medium and high speed levels, which indicates the distance the laser beam moves per unit time. The speed of the scanning speed directly affects the interaction time between the laser and the material (i.e., the action time), which in turn affects the processing effect, the morphology of the molten pool, the size of the heat-affected zone, etc.

[0032] It should be noted that, combined with the above description, since there are multiple process conditions to be tested, the laser power levels and scanning speed levels determined for the process conditions to be tested should also be multiple groups, and each process condition to be tested corresponds to a different laser power level and scanning speed level.

[0033] In the specific implementation, the parameter information of all the process conditions to be tested is collected as input conditions. The analytical model of process / structure / performance is used to simulate and calculate each process condition to be tested, and the corresponding microstructure change value under each process condition to be tested is obtained. The difference calculation is performed on the corresponding microstructure change values ​​under different process conditions to be tested, and the difference in microstructure change values ​​between different process conditions to be tested is obtained. Further, based on the calculated microstructure change value difference, according to the preset mapping relationship between the microstructure change value difference and the laser power level and the scanning speed level, the corresponding laser power level and the scanning speed level are matched one by one to obtain the corresponding laser power level and the scanning speed level under each process condition to be tested. It should be noted that the mapping relationship between the microstructure change value difference and the laser power level and the scanning speed level is determined by experiment, and the analytical model of process / structure / performance is used to simulate the microstructure change values ​​under different laser power levels and scanning speed level combinations to generate a simulated data set. Through regression analysis or fitting method, the corresponding relationship between the microstructure change value difference and the laser power level and the scanning speed level is determined, and a mapping relationship is established.

[0034] The method provided in this embodiment takes into account that due to the different process conditions to be tested, each process condition to be tested has different effects on the microstructure of the material. Therefore, different laser power levels and scanning speed levels are matched for different process conditions to be tested to meet the needs of these changing conditions. By adjusting the process parameters in a targeted manner, the actual microstructure changes under different process conditions to be tested can be better displayed, ensuring the diversity and representativeness of the experimental data, thereby providing more comprehensive support for the analytical model for verifying the process / structure / performance. This differentiated setting can reflect the sensitivity of the process parameters to the microstructure and performance, and provide a more powerful basis for the optimization and improvement of the analytical model. Secondly, when the analytical model has not yet achieved high accuracy, directly using the model prediction value may cause inaccurate matching due to errors. By matching with the relative change (difference) of the microstructure change value, the inaccurate parts that may exist in the model prediction can be effectively offset. The relative change pays more attention to the trend difference under different process conditions to be tested, rather than the accuracy of the absolute value. This processing method reduces the impact of the model prediction error on the matching result and improves the reliability and robustness of the matching. Using this method, reasonable parameter matching can be achieved even when model prediction is not yet perfect, providing a feasible experimental basis and result verification for further optimization of the analytical model.

[0035] In specific implementation, the laser power level and the scanning speed level are determined based on the process conditions to be tested, including: predicting the microstructure change values ​​under the process conditions to be tested based on the analytical model; the microstructure change values ​​include grain size and grain hardness; based on the influence of the laser power level on the microstructure change value, the influence of the scanning speed level on the microstructure change value, and the microstructure change values ​​under the process conditions to be tested, the laser power level and the scanning speed level under the process conditions to be tested are determined; the laser power level is proportional to the grain size and inversely proportional to the hardness; the scanning speed level is inversely proportional to the grain size and proportional to the grain hardness.

[0036] Specifically, the microstructural changes include grain size and grain hardness. Both the laser power level and the scanning speed level will affect the grain size and grain hardness in the microstructural changes. Specifically, as the laser power level increases, the molten pool temperature increases, the cooling rate slows down, and larger grains are formed, the grain size increases, and the grain hardness decreases. As the scanning speed level increases, the cooling rate increases, the molten pool temperature decreases, and smaller grains are formed, the grain size decreases, and the grain hardness increases.

[0037] In the specific implementation, the analytical model of process / structure / performance is used to predict the microstructure change value under each process condition to be tested, including grain size and grain hardness. Based on the experimental data, the influence of laser power level on the microstructure change value (grain size and grain hardness) is analyzed to confirm that it is proportional to the grain size and inversely proportional to the grain hardness; the influence of scanning speed level on the microstructure change value is analyzed to confirm that it is inversely proportional to the grain size and proportional to the grain hardness. The mapping relationship between laser power level and grain size and grain hardness, as well as the mapping relationship between scanning speed level and grain size and grain hardness are constructed. For each process condition to be tested, the corresponding laser power level and scanning speed level are calculated based on the predicted grain size and grain hardness and the mapping relationship. According to the grain size information, select the laser power level that is proportional to it, and select the scanning speed level that is inversely proportional to it. According to the grain hardness information, select the laser power level that is inversely proportional to it, and select the scanning speed level that is proportional to it. Taking into account the changes in grain size and grain hardness, the laser power level and scanning speed level under each process condition to be tested are output.

[0038] It should be noted that, with the different process conditions to be measured, the laser power level and the scanning speed level change dynamically within a range. For example, in this embodiment, the laser power level changes dynamically within the range of 1000W-3000W, and the scanning speed level changes dynamically within the range of 4 mm / s-6 mm / s. In this embodiment, the laser power level is set to three levels of 1400W, 2100W and 2600W, and the scanning speed level is set to three levels of 4.8 mm / s, 5.3 mm / s and 5.9 mm / s.

[0039] S103, preparing a sample to be scanned.

[0040] Specifically, the sample to be scanned is a material sample that has been processed under specific process conditions (such as laser power level, scanning speed level, etc.). The sample to be scanned will be used for subsequent scanning and analysis to extract characteristic information such as grain size and grain hardness, and verify the performance of the analytical model of process / structure / performance through these characteristic information.

[0041] In a specific implementation, the preparation of the sample to be scanned includes: using a concentric powder feeding method, based on the laser power level and the scanning speed level, the powder material is deposited layer by layer to obtain a multi-layer sample; based on the laser scanning direction, the central area of ​​the weld in the multi-layer sample is vertically cut to obtain a sample distributed along a two-dimensional plane; the weld is the connecting area between each layer of the sample; the sample is mechanically polished and ground, and chemically corroded using aqua regia to obtain the sample to be scanned.

[0042] Specifically, distribution along a two-dimensional plane refers to distribution along an xy plane, that is, a plane perpendicular to the laser scanning path and the substrate. A weld refers to the connection area between each layer of a multi-layer sample. Aqua regia refers to a mixed solution obtained by mixing 3 parts of hydrochloric acid and 1 part of nitric acid, which has strong chemical corrosiveness.

[0043] In the specific implementation, a suitable powder material is selected, and the working parameters of the laser are set according to the determined laser power level and scanning speed level. Using the concentric powder feeding method, the powder material is deposited layer by layer under the irradiation of the laser beam. The thickness, laser power level and scanning speed level of each layer are controlled according to the process requirements to form a multi-layer sample containing a multi-layer structure. Each layer melts the powder by laser and combines during the solidification process to form a weld area. Further, the laser scanning direction is determined, and the weld center area in the multi-layer sample is cut according to the laser scanning path. When cutting, the cutting is performed in a direction perpendicular to the laser scanning direction to obtain a sample distributed along a two-dimensional plane. The cut sample exposes the weld area between each layer. Finally, the cut sample is mechanically polished, and the rough material on the surface of the sample is removed using appropriate polishing tools and polishing paste to make the surface smooth and bright. Further, grinding is continued to further improve the smoothness of the sample surface to ensure that the microstructural characteristics of the sample can be clearly observed. Finally, the ground sample is placed in aqua regia for chemical corrosion to remove unnecessary oxide layers on the surface of the metal sample to obtain a sample to be scanned.

[0044] For example, when conducting the experiment in this embodiment, Deloro60 nickel-based alloy powder was selected as the powder material. The main components of the powder are Ni and Cr, which contains 4.3% Si as a solute, which is simplified to a binary alloy system. The substrate is a 42CrMo steel plate, whose chemical composition includes carbon, chromium, silicon, molybdenum, manganese and iron, and has good heat resistance and strength characteristics. The laser in the experiment uses a continuous wave CO2 laser, and uses a concentric powder feeding method to deposit the Deloro60 nickel-based alloy powder on the 42CrMo steel substrate to form a single-grain, single-layer sample layer by layer. After the deposition is completed, the central area of ​​the weld is vertically cut along the laser scanning direction to obtain a sample along the xy plane. The sample is finely mechanically polished and ground, and chemically corroded with aqua regia to enhance the microscopic features of the microstructure to obtain a sample to be scanned.

[0045] S104: Screen candidate scanning areas from the sample to be scanned.

[0046] The candidate area of ​​the molten pool boundary is located according to the microstructure change value predicted by the analytical model, and the scanning step length and the observation area corresponding to the scanning point are determined based on the influence of each of the process conditions to be tested on the sample to be scanned.

[0047] Specifically, candidate scanning areas refer to areas selected on the sample to be scanned, which are representative and suitable for subsequent scanning analysis to extract relevant grain feature information (such as grain size information, grain hardness information, etc.). Candidate scanning areas are selected from the entire sample to be scanned by specific criteria to ensure the accuracy and representativeness of the subsequent analytical model verification results for process / structure / performance.

[0048] It should be noted that since the sample to be scanned usually contains multiple different areas, the grain structure and performance of each area may be different. By selecting representative candidate scanning areas, it can be ensured that the scanning results can reflect the typical characteristics of the entire sample or processing process, rather than local special phenomena. In addition, according to the verification requirements of the analytical model of process / structure / performance, it may be necessary to focus on certain specific areas (such as the center area of ​​the weld, the heat-affected zone, or the connection area between grains). Therefore, selecting these areas as candidate scanning areas can better match the model verification goals and ensure the relevance of the experiment.

[0049] In the specific implementation, the analytical model of process / structure / performance is used to predict the sample to be scanned, and the distribution of microstructure change values ​​under different process conditions to be tested is obtained. According to the predicted distribution of microstructure change values, the demarcation points with significant microstructure change values ​​are identified, and the candidate area of ​​the molten pool boundary is determined (the candidate area of ​​the molten pool boundary is usually the location where the grain characteristics change significantly, such as the center area of ​​the weld or the heat-affected zone). According to the degree of influence of each process condition to be tested on the microstructure of the sample to be scanned, the influence range within the candidate area of ​​the molten pool boundary is calculated, including the gradient and distribution characteristics of the microstructure change. Based on the microstructure change gradient within the candidate area of ​​the molten pool boundary, combined with the resolution requirements of the experiment, the appropriate scanning step size is calculated (a smaller scanning step size is used for areas with larger change gradients, and the scanning step size can be appropriately increased for areas with smaller change gradients). In the candidate area of ​​the molten pool boundary, the scanning points are evenly distributed according to the set scanning step size, and the observation area is divided according to the position of each scanning point to ensure that each observation area covers the characteristic change range of the microstructure. According to the determined scanning points and observation areas, a complete candidate scanning area is generated.

[0050] The method provided in this embodiment takes into account that since the sample to be scanned usually contains multiple different areas, the grain structure and performance of these areas may be significantly different due to factors such as processing conditions, heat conduction, and material composition. The present application locates the candidate area of ​​the molten pool boundary according to the microstructure change value predicted by the analytical model, and can focus on representative and significantly changed areas, so that the observation range is concentrated in the key area, and can fully capture the influence of process conditions on the microstructure, thereby clearly reflecting the overall picture of structural changes. In this way, data deviations caused by interference from irrelevant areas can be effectively avoided, the accuracy of observation results can be improved, and the high adaptability of experimental data to analytical model verification can be ensured. Secondly, by determining the scanning step length and the position of the scanning point based on the different degrees of influence of each process condition to be tested on the sample, efficient and accurate scanning can be achieved. The optimized setting of the scanning step length and point position can reduce the invalid area of ​​the scan while ensuring the accuracy of the experimental data, thereby greatly improving the experimental efficiency. In addition, this precise positioning method can avoid waste of resources, reduce the complexity of experimental data processing, and make the entire verification process more efficient and reliable. By selecting representative candidate scanning areas, data distortion caused by local special phenomena can be avoided, ensuring that the experimental data can reflect the typical characteristics of the entire sample or processing process, thereby enhancing the reliability of analytical model verification.

[0051] Optionally, screening the candidate scanning area from the sample to be scanned includes: observing the microstructure of the sample to be scanned to determine the molten pool morphology and grain distribution; determining the molten pool boundary based on the change demarcation point of the molten pool morphology and the grain distribution; determining the size of the candidate scanning area based on the coverage of the molten pool boundary; determining the scanning step length based on the grain size and resolution requirements of the molten pool boundary; determining the candidate scanning area based on the molten pool boundary, the size of the candidate scanning area, and the scanning step length.

[0052] Specifically, an electron backscatter diffractometer (EBSD) and a scanning electron microscope (SEM) are used to observe the sample to be scanned and obtain its microstructure image. Through the microstructure image, the molten pool morphology and grain distribution in the sample to be scanned are analyzed, and the molten pool area and the grain boundary are identified. Through the analysis of the microstructure image, the morphological characteristics of the molten pool area are determined, and the molten pool edge is marked. At the same time, the distribution of the grains is observed, and the places where the grain size changes greatly, that is, the areas where the grains are unevenly distributed, are identified, and these change points are used as the identification and reference points of the molten pool boundary. Based on the determined change points, the molten pool boundary is drawn. By analyzing the molten pool boundary, the area covered by the molten pool boundary is calculated, and the size of the candidate scanning area is determined. The size of the candidate scanning area needs to cover the area covered by the entire molten pool boundary and the surrounding parts that may affect the microstructure. It should be noted that the size of the candidate scanning area can be appropriately adjusted according to the experimental needs and process requirements to ensure that it contains enough areas to extract effective grain features. Further, according to the grain size within the molten pool boundary and the resolution requirements of the scanning equipment, the appropriate scanning step length is calculated. Scanning compensation is dynamically changing. If the grain size is large, the scanning step can be slightly larger; if the grain size is small, the scanning step needs to be reduced to improve resolution and accuracy. The final candidate scanning area is determined by comprehensively considering the molten pool boundary, the size of the candidate scanning area, and the scanning step. The candidate scanning area can completely cover the molten pool area, and the selection of the scanning step can also meet the requirements for accurate extraction of grain features.

[0053] In a specific implementation, the scanning step length is determined based on the grain size and resolution requirements of the molten pool boundary, including: determining an initial scanning step length based on the grain size of the molten pool boundary, wherein the grain size is positively correlated with the scanning step length; adjusting the initial scanning step length based on the resolution requirement, and determining a scanning compensation; wherein the resolution requirement is negatively correlated with the scanning step length.

[0054] S105, extracting grain feature information based on the candidate scanning area image; the grain feature information at least includes grain size information and grain hardness information.

[0055] Specifically, grain characteristic information characterizes the microstructure of the material, reflecting the overall performance of the material and the impact of the process on the material structure. Grain characteristic information includes grain size information and grain hardness information, which respectively characterize different material properties and quality. Among them, grain size information refers to the size of the grain (such as diameter, side length or area, etc.), which is an important parameter for measuring the microstructure of crystalline materials. Grain size has a significant impact on material performance. Grain hardness information refers to the hardness characteristics of the grains in the material.

[0056] In a specific implementation, the extracting of grain feature information based on the candidate scanning area image includes: identifying the candidate scanning area image by an electron backscatter diffractometer to extract grain size information; analyzing the candidate scanning area to determine the solidification path; determining the depth to be measured based on the solidification path, and determining the grain hardness information of the candidate scanning area at the depth to be measured by micro-Vickers hardness measurement.

[0057] Specifically, the solidification path refers to the direction and shape of crystal growth when liquid metal (or other materials) gradually transforms into a solid state during the solidification process of the material. The solidification path is usually affected by factors such as temperature, cooling rate, and external force field (such as electromagnetic field or laser irradiation direction). In the molten pool, the solidification process of the metal is not uniform, but forms a regular structure along certain specific directions. The solidification path reflects information such as the growth direction of the grains, the distribution of grain boundaries, and the arrangement of the grains.

[0058] In specific implementation, the candidate scanning area is observed by electron backscatter diffractometer (EBSD) and scanning electron microscope (SEM). Through the image of the candidate scanning area, the grain boundaries in the sample are identified, and the grain boundaries usually correspond to the areas with different crystal orientations. According to the distribution and shape of the grain boundaries, the grain size information (such as grain diameter, area, aspect ratio, etc.) is extracted. Based on the solidification characteristics of the material, the solidification path on the surface of the sample is analyzed, and the solidification direction is determined by the texture, morphology and other features observed in the image of the candidate scanning area. Combined with the processing parameters (such as laser scanning direction, temperature gradient, etc.) and the stripes or linear patterns along the laser scanning direction, the position and morphology of the solidification path are determined. Further, according to the hierarchical structure of the solidification path and the sample, the depth to be measured is determined based on the thermal gradient and temperature change during the solidification process (the solidification path is usually closely related to the thermal history of the material and the grain growth direction). In addition, when determining the depth to be measured, it is also necessary to select an area perpendicular or parallel to the solidification path. Finally, at the determined depth to be measured, the sample is tested for hardness using a micro-Vickers hardness tester, a certain load is applied, the indentation is observed using a microscope, and the hardness value is calculated. The hardness value of each grain is recorded and extracted, and the grain hardness information is obtained based on the hardness value distribution.

[0059] By analyzing the curvature of the solidification path, the directional characteristics of grain growth can be identified, and then the process conditions and process parameters (such as laser power level and scanning speed level) to be tested can be adjusted to achieve grain refinement and hardness optimization. A larger curvature may indicate the presence of a strong temperature gradient or non-uniformity at the molten pool boundary, which is closely related to the orientation and morphology of the grains. Combined with the information on the molten pool temperature gradient, the grain growth rate and potential grain refinement area can be further deduced to determine the microstructural evolution under different process conditions to be tested.

[0060] S106, calculating material hardness according to the grain size information.

[0061] In a specific implementation, the calculation of the material hardness based on the grain size information includes: extracting a preset number of grain size information and grain hardness information from the grain feature information; fitting the preset number of grain size information and the grain hardness information based on the Hall-Page law to determine the Hall-Page coefficient and the inherent hardness in the Hall-Page law; and determining the material hardness based on the remaining number of grain size information in the grain feature information using the fitted Hall-Page law.

[0062] Specifically, in combination with the above description, the Hall-Page law characterizes the quantitative relationship between material hardness and grain size. There is an inverse relationship between material hardness and grain size, that is, the smaller the grain size, the greater the material hardness. When the Hall-Page coefficient and inherent hardness are determined, the grain size information is input into the Hall-Page law to calculate the material hardness. The preset number is set according to actual needs, and in this embodiment, it is not limited to this. It should be noted that the preset number of grain size information and grain hardness information is used to determine the Hall-Page coefficient and inherent hardness in the Hall-Page law.

[0063] In specific implementation, a certain number of grains are selected as samples from the grain feature information extracted from the candidate scanning area. For each selected grain, its grain size information and grain hardness information are extracted to obtain a database containing grain size information and grain hardness information. Furthermore, the extracted grain size information and the corresponding grain hardness information are substituted into the Hall-Page law, and the Hall-Page coefficient and intrinsic hardness are fitted using the least squares method or other fitting methods. During each fitting, the Hall-Page coefficient and inherent hardness are repeatedly adjusted based on the results of each fitting until the number of fittings reaches a maximum value. Finally, the Hall-Page law obtained by fitting (the Hall-Page coefficient and inherent hardness are determined) is used, combined with the grain size information corresponding to the remaining number of grains that have not been selected, and the grain size information is substituted into the Hall-Page law to calculate the corresponding material hardness.

[0064] S107, combining the material hardness, the grain hardness information and the corresponding process conditions to be tested to obtain an evaluation data set, and evaluating the performance of the analytical model based on the evaluation data set.

[0065] Specifically, material hardness is the overall hardness value of the material predicted by the analytical model of process / structure / performance based on the process conditions to be tested. It is an estimate of the material's ability to resist plastic deformation at the macro level by the analytical model of process / structure / performance. Grain hardness information is obtained by directly measuring the hardness of a single grain through microhardness testing (such as Vickers hardness). In order to ensure the reliability and accuracy of the analytical model of process / structure / performance in practical applications, it is necessary to compare the material hardness predicted by the process / structure / performance model with the grain hardness information obtained through direct experiments to verify the performance of the analytical model of process / structure / performance.

[0066] In a specific implementation, the material hardness, the grain hardness information and the corresponding process conditions to be tested are combined to obtain an evaluation data set, and the performance of the analytical model is evaluated based on the evaluation data set, including: drawing a first hardness curve and a second hardness curve based on the material hardness and the grain hardness information under different process conditions to be tested in the evaluation data set; wherein the first hardness curve represents a changing trend of the material hardness with the grain size information, and the second hardness curve represents a changing trend of the grain hardness information with the grain size information; the material hardness is predicted data obtained based on the analytical model, and the grain hardness information is measured data obtained by scanning an image of the candidate area; the first hardness curve and the second hardness curve are compared, and the performance of the analytical model is determined based on the changing trends of the first hardness curve and the second hardness curve and the difference between the first hardness curve and the second hardness curve under any grain size information.

[0067] Specifically, in combination with the above description, different process conditions to be tested correspond to different material hardnesses (predicted by the analytical model of process / structure / performance) and different grain hardness information (directly measured by microhardness test). The first hardness curve is drawn based on different process conditions to be tested and the corresponding material hardness, and the second hardness curve is drawn based on different process conditions to be tested and the corresponding grain hardness information.

[0068] In specific implementation, for different process conditions to be tested, the process conditions to be tested are used as the horizontal coordinates. Under different process conditions to be tested, the corresponding material hardness is recorded, and it is matched with the process conditions to be tested to form a series of points, and these points are connected to draw a first hardness curve. For different process conditions to be tested, the process conditions to be tested are used as the horizontal coordinates. Under different process conditions to be tested, the corresponding grain hardness information is recorded, and it is matched with the grain size information to form a series of points, and these points are connected to draw a second hardness curve. Compare the first hardness curve with the second hardness curve to view the changing trend of the two curves. Compare the changing rules of the two curves under different process conditions to be tested, including the slope and shape of the curves. Compare the hardness value difference between the first hardness curve and the second hardness curve under each corresponding process condition to be tested. For each process condition to be tested, calculate the difference between the two curves. Based on the changing trend and difference of the two curves, evaluate the performance of the analytical model of process / structure / performance.

[0069] Combined with the above description, the analytical model of process / structure / performance is used to predict the hardness of the material based on the process conditions to be tested, and the material hardness is inversely proportional to the process conditions to be tested (corresponding to the grain size information). When the change trends of the two curves are both reversed, that is, as the grain size information decreases, the hardness increases. In addition, based on the comparison of the grain size information corresponding to the analytical model of process / structure / performance under different process conditions to be tested and the actual grain size information, the accuracy of the analytical model of process / structure / performance is determined based on the difference in grain size information.

[0070] For example, if the trends of the two curves are similar and the difference is small under different grain size information, it indicates that the analytical model of process / structure / performance has high prediction accuracy and good performance. If there is a significant difference or a large difference between the trends of the two curves, it may indicate that the prediction results of the analytical model of process / structure / performance have a large deviation from the experimental data, the performance is poor, and adjustment or optimization is required.

[0071] The method provided in this embodiment can fully evaluate the performance of the analytical model of process / structure / performance by comparing and analyzing the material hardness predicted by the analytical model of process / structure / performance with the grain hardness information obtained from the experiment. First, the material hardness predicted by the analytical model of process / structure / performance is calculated based on the grain size information through the Hall-Page law, which reflects the modeling ability of the analytical model for the relationship between microscopic features and macroscopic performance, while the grain hardness information is experimental data obtained by direct measurement, which has high authenticity and accuracy. The comparison between the two can intuitively reflect the deviation and applicability of the analytical model in predicting material properties, and further verify the rationality of the analytical model by analyzing whether the trend predicted by the analytical model is consistent with the experimental results. This evaluation method can not only verify the accuracy of the analytical model, but also discover the potential limitations of the analytical model, provide a scientific basis for process optimization and model improvement, and thus improve the reliability and wide applicability of the analytical model in practical applications.

[0072] The analytical model verification method for process / structure / performance provided in this embodiment, in the first aspect, can fully evaluate the performance of the analytical model of process / structure / performance by comparing and analyzing the material hardness predicted by the analytical model of process / structure / performance with the grain hardness information obtained by experiment. First, the material hardness predicted by the analytical model of process / structure / performance is calculated based on the grain size information through the Hall-Page law, which reflects the modeling ability of the analytical model for the relationship between microscopic features and macroscopic performance, while the grain hardness information is the experimental data directly measured, which has high authenticity and accuracy. The comparison between the two can intuitively reflect the deviation and applicability of the analytical model in predicting material properties, and further verify the rationality of the analytical model by analyzing whether the trend predicted by the analytical model is consistent with the experimental results. This evaluation method can not only verify the accuracy of the analytical model, but also discover the potential limitations of the analytical model, provide a scientific basis for process optimization and model improvement, thereby improving the reliability and wide applicability of the analytical model in practical applications. On the second aspect, since the sample to be scanned usually contains multiple different regions, the grain structure and performance of these regions may be significantly different due to factors such as processing conditions, heat conduction, and material composition. This application locates the candidate area of ​​the molten pool boundary according to the microstructure change value predicted by the analytical model, and can focus on the representative and significantly changed area, so that the observation range is concentrated in the key area, and can fully capture the influence of the process conditions on the microstructure, so as to clearly reflect the overall picture of the structural change. In this way, the data deviation caused by the interference of irrelevant areas can be effectively avoided, the accuracy of the observation results can be improved, and the high adaptability of the experimental data to the analytical model verification can be ensured. Secondly, by determining the scanning step length and the position of the scanning point based on the different degrees of influence of each process condition to be tested on the sample, efficient and accurate scanning can be achieved. The optimized setting of the scanning step length and the point position can reduce the invalid area of ​​the scan while ensuring the accuracy of the experimental data, thereby greatly improving the experimental efficiency. In addition, this method of precise positioning can avoid waste of resources, reduce the complexity of experimental data processing, and make the entire verification process more efficient and reliable. By selecting a representative candidate scanning area, data distortion caused by local special phenomena can be avoided, ensuring that the experimental data can reflect the typical characteristics of the entire sample or processing process, thereby enhancing the reliability of analytical model verification. At the same time, according to the verification requirements of the analytical model of process / structure / performance, it is only necessary to focus on certain key areas, such as the center area of ​​the weld, the heat-affected zone or the inter-grain connection area, which are usually sensitive points of material performance changes and the focus of process optimization. Selecting these areas as candidate scanning areas can not only effectively reduce the experimental workload and improve the experimental efficiency, but also ensure that the collected data better matches the analytical model verification target and avoids the interference of irrelevant data on the experimental results.Thirdly, verification data is obtained based on the laser power level and scanning speed level determined by the process conditions to be tested, ensuring that the analytical model verification process is highly relevant to the actual production environment. This avoids the gap between laboratory conditions and industrial applications that may exist in traditional experimental methods, making the verification of analytical models closer to real production scenarios. In the process of making samples to be scanned and screening candidate scanning areas, the reliability and accuracy of experimental data are further enhanced by ensuring the actual representativeness of the samples and selecting appropriate areas for testing. This process not only improves the accessibility of data by extracting grain feature information (such as grain size and hardness information) from images, but also provides a real and efficient basis for subsequent hardness calculations. Fourthly, grain size information and grain hardness information are used as important input parameters for analytical model prediction, and combined with the Hall-Page law, an effective way to predict material hardness is provided. This transformation from microscopic grain characteristics to macroscopic hardness helps to reveal the intrinsic relationship between grain size and material properties, thereby enhancing the predictive ability of analytical models. Moreover, by comparing the material hardness calculated by the analytical model with the actual measured grain hardness information, the applicability and accuracy of the analytical model under different process conditions to be tested can be systematically evaluated. This comparative analysis not only provides a feedback mechanism for the optimization of the model, but also helps to identify and adjust potential deviations in the model and improve its prediction accuracy. In addition, the flexible setting of process parameters (such as laser power level, scanning speed level, etc.) enables the analytical model verification to cover a variety of process scenarios, which enhances the adaptability and wide application possibility of the analytical model. Through image analysis and data extraction, the entire analytical model verification process can be automated, which improves the verification efficiency and reduces manual intervention and errors, thereby ensuring the efficiency and accuracy of the experiment. Combined with these advantages, this application not only ensures the reliability of the analytical model by systematically combining and verifying the relationship between process, structure and performance, but also promotes the efficient implementation of process optimization, and can provide practical technical support in actual manufacturing.

[0073] Corresponding to the aforementioned embodiment of a method for verifying an analytical model of a process / structure / performance, the present application also provides an embodiment of an apparatus for verifying an analytical model of a process / structure / performance.

[0074] Figure 2 This is a schematic diagram of the structure of the analytical model verification device for the process / structure / performance provided in Example 2 of this application. Figure 2 The device provided in this embodiment includes a determination module 210, an acquisition module 220, a screening module 230, a calculation module 240 and an evaluation module 250; wherein, The determination module 210 is used to determine the process conditions to be tested of the analytical model of the process / structure / performance, wherein the process conditions to be tested are multiple, and the analytical model is used to output the material hardness based on the process conditions to be tested; The determination module 210 is further used to determine the laser power level and the scanning speed level based on the process conditions to be tested; wherein the laser power level and the scanning speed level of each process condition to be tested are different, and the microstructure change value under each process condition to be tested is predicted based on the analytical model, and the corresponding laser power level and the scanning speed level are matched based on the difference of the microstructure change values ​​under different process conditions to be tested; The acquisition module 220 is used to prepare a sample to be scanned; The screening module 230 is used to screen candidate scanning areas from the sample to be scanned; wherein, the candidate area of ​​the molten pool boundary is located according to the microstructure change value predicted by the analytical model, and the scanning step length and the observation area corresponding to the scanning point are determined based on the influence of each of the process conditions to be tested on the sample to be scanned; The acquisition module 220 is further used to extract grain feature information based on the candidate scanning area image; the grain feature information at least includes grain size information and grain hardness information; The calculation module 240 is used to calculate the material hardness according to the grain size information; The evaluation module 250 is used to combine the material hardness, the grain hardness information and the corresponding process conditions to be tested to obtain an evaluation data set, and evaluate the performance of the analytical model based on the evaluation data set.

[0075] The device of this embodiment can be used to perform Figure 1 The steps, specific implementation principles and implementation processes of the method embodiment shown are similar and will not be repeated here.

[0076] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0077] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0078] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A process / structure / performance analytical model verification method, characterized in that: The method comprises: Determine the process conditions to be tested of the analytical model of process / structure / performance, wherein there are multiple process conditions to be tested, and the analytical model is used to output the material hardness based on the process conditions to be tested; Determine the laser power level and the scanning speed level based on the process conditions to be tested, wherein the laser power level and the scanning speed level of each process condition to be tested are different, predict the microstructure change value under each process condition to be tested based on the analytical model, and match the corresponding laser power level and the scanning speed level based on the difference of the microstructure change values ​​under different process conditions to be tested; Prepare samples to be scanned; Screening candidate scanning areas from the sample to be scanned, wherein the candidate molten pool boundary area is located according to the microstructure change value predicted by the analytical model, and the scanning step length and the observation area corresponding to the scanning point are determined based on the influence of each of the process conditions to be tested on the sample to be scanned; Extracting grain feature information based on the candidate scanning area image; the grain feature information at least includes grain size information and grain hardness information; calculating the material hardness based on the grain size information; The material hardness, the grain hardness information and the corresponding process conditions to be tested are combined to obtain an evaluation data set, and the performance of the analytical model is evaluated based on the evaluation data set.

2. The method according to claim 1, characterized in that The step of combining the material hardness, the grain hardness information and the corresponding process conditions to be tested to obtain an evaluation data set, and evaluating the performance of the analytical model based on the evaluation data set, comprises: Based on the material hardness and the grain hardness information under different process conditions to be tested in the evaluation data set, a first hardness curve and a second hardness curve are drawn; wherein the first hardness curve represents the change trend of the material hardness with the process conditions to be tested, and the second hardness curve represents the change trend of the grain hardness information with the process conditions to be tested; the material hardness is the predicted data obtained based on the analytical model, and the grain hardness information is the measured data obtained by scanning the candidate area image; The first hardness curve and the second hardness curve are compared, and the performance of the analytical model is determined based on the change trends of the first hardness curve and the second hardness curve and the difference between the first hardness curve and the second hardness curve under any process condition to be tested.

3. The method according to claim 1, characterized in that The calculating the material hardness according to the grain size information comprises: Extracting a preset amount of grain size information and grain hardness information from the grain characteristic information; Fitting the preset number of grain size information and the grain hardness information based on the Hall-Page law to determine the Hall-Page coefficient and the intrinsic hardness in the Hall-Page law; The material hardness is determined by using the Hall-Page law obtained by fitting and based on the remaining grain size information in the grain characteristic information.

4. The method according to claim 1, characterized in that The step of extracting grain feature information based on the candidate scanning area image comprises: Identifying the candidate scanning area image by electron backscatter diffractometer to extract grain size information; Analyzing the candidate scanning area to determine a coagulation path; A depth to be measured is determined based on the solidification path, and grain hardness information of the candidate scanning area at the depth to be measured is determined by micro-Vickers hardness measurement.

5. The method according to claim 1, characterized in that The step of screening a candidate scanning area from the sample to be scanned includes: Observe the microstructure of the sample to be scanned to determine the molten pool morphology and grain distribution; Determining the molten pool boundary based on the change demarcation point of the molten pool morphology and the grain distribution; Determining the size of the candidate scanning area based on the coverage of the molten pool boundary; Determine the scanning step length based on the grain size and resolution requirements of the molten pool boundary; A candidate scanning area is determined based on the molten pool boundary, the size of the candidate scanning area, and the scanning step size.

6. The method according to claim 1, characterized in that The step of preparing the sample to be scanned comprises: Using a concentric powder feeding method, based on the laser power level and the scanning speed level, the powder material is deposited layer by layer to obtain a multi-layer sample; Based on the laser scanning direction, the central area of ​​the weld in the multi-layer sample is vertically cut to obtain a sample distributed along a two-dimensional plane; the weld is the connection area between each layer of samples; The sample is mechanically polished and ground, and chemically etched using aqua regia to obtain a sample to be scanned.

7. The method according to claim 1, characterized in that The step of determining the laser power level and the scanning speed level based on the process conditions to be measured includes: Predicting the microstructure change value under each process condition to be tested based on the analytical model; the microstructure change value includes grain size and grain hardness; Based on the influence of laser power level on microstructure change value, the influence of scanning speed level on microstructure change value, and the microstructure change value under each process condition to be tested, the laser power level and scanning speed level under each process condition to be tested are determined; the laser power level is directly proportional to the grain size and inversely proportional to the hardness; the scanning speed level is inversely proportional to the grain size and directly proportional to the grain hardness.

8. The method according to claim 1, characterized in that The analytical model of process / structure / performance uses a continuous growth limiting factor to replace the traditional grain growth assumption, uses an interdependence model to characterize the growth relationship between grains, and uses the Hall-Page law to replace the traditional performance prediction method; the analytical model is used to determine the material hardness based on the grain size information corresponding to different process conditions to be tested, and adjust the process conditions to be tested based on the difference between the material hardness and the target hardness.

9. The method according to claim 8, characterized in that The analytical model of process / structure / performance associates the process conditions to be tested with the grain size information, and the continuous growth limiting factor is inversely proportional to the grain size information; the Hall-Page law associates the grain size information with the material hardness, and the reciprocal square root of the grain size information varies linearly with the material hardness.

10. A process / structure / performance analytical model verification device, characterized in that: The device includes a determination module, an acquisition module, a screening module, a calculation module and an evaluation module; wherein, The determination module is used to determine the process conditions to be tested of the analytical model of process / structure / performance, wherein there are multiple process conditions to be tested, and the analytical model is used to output the material hardness based on the process conditions to be tested; The determination module is further used to determine the laser power level and the scanning speed level based on the process conditions to be tested; wherein the laser power level and the scanning speed level of each process condition to be tested are different, and the microstructure change value under each process condition to be tested is predicted based on the analytical model, and the corresponding laser power level and the scanning speed level are matched based on the difference of the microstructure change values ​​under different process conditions to be tested; The acquisition module is used to prepare a sample to be scanned; The screening module is used to screen candidate scanning areas from the sample to be scanned; wherein, the candidate area of ​​the molten pool boundary is located according to the microstructure change value predicted by the analytical model, and the scanning step length and the observation area corresponding to the scanning point are determined based on the influence of each of the process conditions to be tested on the sample to be scanned; The acquisition module is further used to extract grain feature information based on the candidate scanning area image; the grain feature information at least includes grain size information and grain hardness information; The calculation module is used to calculate the material hardness according to the grain size information; The evaluation module is used to combine the material hardness, the grain hardness information and the corresponding process conditions to be tested to obtain an evaluation data set, and evaluate the performance of the analytical model based on the evaluation data set.

Images