A screening method, device, equipment and medium for ceramic metal brazing filler

Abstract

The application provides a ceramic metal brazing filler metal screening method, device, equipment and medium, and relates to the technical field of brazing filler metal screening. The method comprises the following steps: based on initial parameters of brazing filler metal, adopting a solution model, determining Gibbs free energy in a main brazing reaction and activity of each component; according to the Gibbs free energy and the activity, constructing a nonlinear equation of equilibrium conditions and solving to obtain an equilibrium process; based on the initial parameters of the brazing filler metal, determining the number of moles of each component in the equilibrium process, and determining effective components of the brazing filler metal according to the number of moles; according to the effective components of the brazing filler metal, determining a component descriptor, and constructing a descriptor constraint target space; based on the descriptor constraint target space, extending to generate a candidate component combination, and screening to obtain an optimal component combination. The application can realize high-throughput screening of brazing filler metal components with consideration of screening reliability, efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to the field of brazing filler metal technology, and more specifically, to a method, apparatus, equipment, and medium for screening ceramic-metal brazing filler metals. Background Technology

[0002] The bonding of ceramics and metals is a key material composite technology, its core significance lying in overcoming the performance limitations of single materials and achieving complementary advantages. By combining the high-temperature resistance, corrosion resistance, and high hardness of ceramics with the toughness, electrical and thermal conductivity, and ease of processing of metals, composite components can be manufactured that can withstand extreme environments (such as high temperatures, strong corrosion, and high wear) while possessing good mechanical load-bearing and sealing capabilities. This technology directly drives progress in aerospace, new energy, electronics, and high-end manufacturing, playing an irreplaceable role in fields such as hot-end components of aero-engines, power semiconductor substrates, new energy battery seals, and biomedical implants.

[0003] Due to the significant differences in physical and chemical properties between ceramics and metals, direct bonding presents core challenges such as difficulty in wetting, uncontrolled reactions, and susceptibility to cracking upon cooling. Therefore, brazing filler metal is needed as an intermediate "bridge" for connection. However, different brazing filler metal compositions result in varying bonding effects. If the brazing filler metal composition is inappropriate, it may not only fail to achieve a strong metallurgical bond but may also exacerbate harmful interfacial reactions, generating a large amount of brittle phases, or cause residual stress concentration and cracking failure due to poor thermal expansion matching, ultimately preventing the ceramic and metal properties from synergistically functioning. Conversely, if the brazing filler metal composition is appropriate, these problems can be avoided, resulting in a strong bond between the two materials.

[0004] In high-throughput screening and performance evaluation of solder components, related technologies use the nominal composition initially designed for the solder for calculation, or employ computational models such as physical calculation models (first-principles / molecular dynamics point-to-point calculation models) and dataset-dependent machine learning models to directly calculate or predict screening results. However, during soldering, interfacial reactions significantly consume the active elements of the solder and may introduce non-metallic species (such as Si and C) from the ceramic side. Continuing to use the initially designed nominal composition for calculations can lead to significant deviations in results. Furthermore, in active solder design scenarios involving multiple coupled factors such as interfacial reactions, diffusion, and thermal stress, physical calculation models perform high-precision thermodynamic and kinetic calculations across all components and reaction pathways. This involves numerous parameters, high computational dimensionality, and requires substantial computing power and time, resulting in high costs. Machine learning models, on the other hand, suffer from low accuracy due to data gaps and a disconnect from physical mechanisms. Summary of the Invention

[0005] The present invention aims to solve at least one of the above-mentioned problems.

[0006] To address the above problems, this invention provides a method, apparatus, equipment, and medium for screening ceramic metal brazing filler metals.

[0007] In a first aspect, the present invention provides a method for screening ceramic metal brazing filler metals, comprising: Based on the initial parameters of the brazing filler metal, a solution model was used to determine the Gibbs free energy and the activity of each component in the dominant brazing reaction. Based on the Gibbs free energy and the activity, a nonlinear equation for the equilibrium condition is constructed and solved to obtain the equilibrium process. Based on the initial parameters of the solder, the molar number of each component in the balancing process is determined, and the effective component of the solder is determined according to the molar number. Based on the effective components of the solder, component descriptors are determined, and a descriptor-constrained target space is constructed; Based on the descriptor constraining the target space, candidate component combinations are expanded and generated, and the optimal component combination is obtained by screening.

[0008] Optionally, the determination of the Gibbs free energy and the activity of each component in the dominant brazing reaction based on the initial parameters of the brazing filler metal and using a solution model includes: Based on the initial parameters of the brazing filler metal, the Gibbs free energy in the dominant brazing reaction is determined using the melt model. Based on the mole fraction constraint, the partial derivative of the expression for the Gibbs free energy is used to obtain the expression for the chemical potential of the component. The activity is determined based on the chemical potential expression.

[0009] Optionally, the composition descriptor includes a formation tendency descriptor, a valence electron concentration descriptor, and a coefficient of thermal expansion; The step of determining component descriptors based on the effective components of the solder and constructing a descriptor-constrained target space includes: Based on the effective components of the solder, and based on the physical definition and quantification relationship of descriptors, the formation tendency descriptor, the valence electron concentration descriptor, and the coefficient of thermal expansion are determined respectively. Based on the formation tendency descriptor, the valence electron concentration descriptor, and the thermal expansion coefficient, the descriptor-constrained target space is constructed.

[0010] Optionally, the step of expanding and generating candidate component combinations based on the descriptor-constrained target space includes: Based on the descriptor-constrained target space, the Pareto front method is used to expand and generate the candidate component combinations.

[0011] Optionally, the screening to obtain the optimal combination of components includes: Based on the additional hard constraints, a non-dominated sorting method is used to screen the candidate component combinations to obtain the optimal solution set; The optimal component combination is determined based on the frequency of occurrence of each element in the optimal solution set.

[0012] Optionally, the additional hard constraints include: The value of the valence electron concentration descriptor is greater than a first preset threshold, and the value of the formation tendency descriptor is greater than a second preset threshold.

[0013] Optionally, after obtaining the optimal combination of components through screening, the method further includes: The optimal combination of components was experimentally verified.

[0014] Secondly, the present invention provides a screening device for ceramic metal brazing filler metal, comprising: The reaction module is used to determine the Gibbs free energy and the activity of each component in the brazing-dominant reaction based on the initial parameters of the brazing filler metal and using a solution model. The process module is used to construct and solve the equilibrium condition nonlinear equation based on the Gibbs free energy and the activity to obtain the equilibrium process. An effective module is used to determine the molar number of each component in the balancing process based on the initial parameters of the solder, and to determine the effective component of the solder based on the molar number. The description module is used to determine the component descriptors based on the effective components of the solder, and to construct the descriptor constraint target space; The filtering module is used to expand and generate candidate component combinations based on the target space constrained by the descriptor, and filter to obtain the optimal component combination.

[0015] Thirdly, the present invention provides an electronic device, including a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the ceramic metal brazing filler metal screening method as described in the first aspect when executing the computer program.

[0016] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for screening ceramic metal brazing filler metal as described in the first aspect.

[0017] The beneficial effects of the screening method, apparatus, equipment, and medium for ceramic metal brazing filler metal of the present invention are: First, based on the initial parameters of the brazing filler metal, a solution model was used to determine the Gibbs free energy and the activity of each component in the dominant brazing reaction. This provided core thermodynamic data support for the subsequent accurate solution of the reaction equilibrium state, avoiding subsequent calculation deviations caused by the lack of effective capture of key thermodynamic parameters of the reaction. Next, based on the aforementioned Gibbs free energy and component activities, a nonlinear equation for the equilibrium condition was constructed and the equilibrium process was solved. This accurately reflects the actual situation of active element consumption and the introduction of nonmetallic species on the ceramic side in the interfacial reaction, laying a precise foundation for the subsequent calculation of the effective components. Subsequently, based on the initial parameters of the brazing filler metal and the equilibrium process, the molar number of each component was determined, and the effective components of the brazing filler metal were obtained. The effective components that actually participate in joint formation replaced the initial design parameters (nominal components) commonly used in related technologies, fundamentally eliminating the design input bias caused by component offset and solving the problem of calculation errors caused by the use of initial design parameters in related technologies. To address the issue of significant deviations, the system first determines component descriptors based on the effective components of the solder and constructs a descriptor-constrained target space. This establishes the correlation between components and performance based on real components and clear physical logic, avoiding the low accuracy problem caused by machine learning models being disconnected from physical mechanisms. This provides a clear performance guide for subsequent candidate component screening. Finally, candidate component combinations are generated and the optimal combination is screened based on the descriptor-constrained target space. Unlike physical calculation models, which require high-precision thermodynamic and kinetic calculations of all components and reaction paths, this system efficiently expands and screens within a targeted constrained target space, significantly reducing computational power and time consumption. This solves the problem of high cost of physical calculation models. At the same time, relying on the accurate effective components and constraints that fit the physical mechanisms, the accuracy of screening the optimal component combination is ensured. Ultimately, this achieves high-throughput screening of ceramic-metal solders that balances reliability, efficiency, and accuracy. Attached Figure Description

[0018] Figure 1 A schematic flowchart illustrating the method for screening ceramic metal brazing filler metals provided in an embodiment of the present invention; Figure 2 A contour diagram of the Co-Ti-Si-C quaternary system at 1300K provided in an embodiment of the present invention; Figure 3 A schematic diagram illustrating the multi-objective Pareto optimization results provided in an embodiment of the present invention; Figure 4 This is a schematic diagram showing the morphology of brazed joints obtained by different brazing filler metals according to embodiments of the present invention; Figure 5 for Figure 4 Table of EDS element analysis results for each feature point in the data; Figure 6 A schematic diagram of the structure of the ceramic metal brazing filler metal screening device provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0020] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0021] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; and the term "optionally" means "optional embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first," "second," etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0022] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0023] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0024] like Figure 1 As shown in the figure, an embodiment of the present invention provides a method for screening ceramic metal brazing filler metal, comprising: Based on the initial parameters of the brazing filler metal, a solution model is used to determine the Gibbs free energy and the activity of each component in the dominant brazing reaction. According to the Gibbs free energy and the activity, a nonlinear equation for the equilibrium condition is constructed and solved to obtain the equilibrium process. Based on the initial parameters of the brazing filler metal, the molar number of each component in the equilibrium process is determined, and the effective component of the brazing filler metal is determined according to the molar number.

[0025] Specifically, the initial parameters of the solder include core parameters such as the nominal composition of the solder, the types of components (e.g., Cu, Ni, Mo, Ti, etc.), and thermodynamic temperature. The types of components can be flexibly selected according to the requirements of the ceramic-metal bonding system. For example, in the SiCf / SiC-GH5188 system, Cu, Ni, and Mo can be selected as the matrix and alloying components, while Ti is selected as the active component. The preferred solution model is the substitution-solution model, which can be replaced by other solution models with multi-component liquid-phase thermodynamic description capabilities (such as sublattice models, as long as they can accurately characterize the interactions between components and the thermodynamic properties of the system). First, through thermodynamic calculations and literature review, the dominant brazing reaction is identified. For example, the dominant reaction between SiCf / SiC and Ti-based solder is [Ti] + SiC → TiC + [Si]. Then, the initial parameters of the solder are substituted into the solution model to solve for the equilibrium process of the dominant reaction. Finally, based on the equilibrium process, the nominal composition is corrected to obtain the effective composition of the solder that actually participates in the joint formation.

[0026] More specifically, the initial parameters of the solder are first determined, including the nominal composition of the solder (e.g., the mass fraction ratio of Cu-20Ni-10Ti-10Mo), the types of components (e.g., Cu, Ni, Ti, Mo, etc.), the thermodynamic temperature (e.g., 1300K), the Gibbs free energy data of the liquid phase of pure elements, and the binary interaction parameters (taken from thermodynamic databases or literature). A displacement-solution model is adopted, using formula G... m =G ref +G id +G ex Calculate the Gibbs free energy (where G is the dominant reaction in brazing) in the brazing process. m For Gibbs free energy, G ref For reference to Gibbs free energy, G id For the contribution of mixed entropy, G ex The excess Gibbs free energy is used, and the activity of each component is determined based on the correlation between the mole fraction and activity coefficient of the components. Then, the reaction quotient Q is calculated based on the component activities. There may be multiple potential reaction pathways at the brazing interface, including reactions between the active element and the ceramic, or reactions between the active element and the metal matrix. These reactions can be uniformly represented as: ; in, For the i-th species or phase, is the stoichiometric coefficient. It should be noted that this model, based on thermodynamic calculations and literature review, only identifies the dominant interfacial reactions, rather than exhaustively listing all reaction pathways. For the reactions shown above, the reaction quotient Q can be defined as: ; in, For activity, The reaction progress is then determined. Subsequently, based on the calculated Gibbs free energy, the standard Gibbs free energy change ΔG°(T) of the reaction is calculated using the difference between the standard Gibbs free energies of the products and reactants. Combined with the temperature parameter, the equilibrium constant K(T) = exp(-ΔG°(T) / (RT)) is obtained, where T is the temperature, R is the gas constant (molar gas constant), R = 8.314 J / mol, and the equilibrium condition is defined as Q( * ) = K(T), where, * This is the equilibrium process, also called the equilibrium reaction progress. Based on the reaction quotient and the equilibrium constant, a nonlinear equation for the equilibrium condition, f( )=Q( When K(T) = 0, the equilibrium process is solved using the Newton-Raphson method. * This solution method can be replaced by the bisection method or other equivalent single-variable nonlinear equation solvers. Finally, to facilitate subsequent high-throughput screening, the total number of moles in the system was normalized to 1 mol. , The initial composition (molar ratio). The initial molar number is determined based on the nominal composition in the initial solder parameters, and the molar number of each component during the balancing process is then used as the initial molar number. ,in, denoted as the number of moles of the i-th species or phase.

[0027] Then, the effective composition of the solder is determined based on the stated number of moles: ,in, The active components of the solder exist in a ratio form. denoted as the number of moles of the j-th species or phase.

[0028] In this embodiment, the initial parameters of the solder are the fundamental input for thermodynamic calculations, the Gibbs free energy and component activities are the core parameters for constructing the equilibrium equation, and the equilibrium process is the key basis for correcting the nominal composition. Each feature is progressively defined, forming a complete logic for calculating the effective composition. The beneficial effect is that by quantifying the thermodynamic parameters and equilibrium process step by step, the effective composition of the solder after the reaction is accurately obtained, solving the screening bias problem caused by using the nominal composition as input in traditional methods, and providing high-quality data support for subsequent descriptor calculations and screening.

[0029] Based on the effective components of the solder, component descriptors are determined, and a descriptor-constrained target space is constructed.

[0030] Specifically, based on the obtained effective components of the solder, and according to the physical definition and quantitative relationship of each component descriptor, each component descriptor is determined. These descriptors can be used to evaluate the formation tendency of brittle intermetallic compounds in the solder, reflect the deformability of the solder, and characterize the thermal expansion matching degree between the solder and the ceramic / metal substrate. The component descriptors can be expanded and replaced according to actual application requirements. For example, for scenarios with high thermal conductivity requirements, thermal conductivity descriptors can be introduced; when electrical conductivity is required, resistivity descriptors can be added. Subsequently, a descriptor constraint target space is constructed, which clarifies the optimization direction of each descriptor, i.e., the screening constraints. In this embodiment, the effective components of the solder provide a precise compositional basis for descriptor calculation. The descriptors correspond to the core performance requirements of the solder, and the constraint target space transforms the performance requirements into quantifiable screening criteria. These three are interconnected, ensuring that the screening direction is consistent with the actual requirements of ceramic-metal bonding. The beneficial effect is that by constructing a constraint space through multi-dimensional descriptors, multiple objectives such as reaction control, toughness improvement, and thermal mismatch mitigation are achieved, avoiding overall performance imbalance caused by single performance optimization.

[0031] Based on the descriptor constraining the target space, candidate component combinations are expanded and generated, and the optimal component combination is obtained by screening.

[0032] Specifically, based on the constructed descriptor-constrained target space, the Pareto front method is used to expand and generate candidate component combinations. This method can be replaced by other multi-objective optimization expansion methods with the ability to generate high-quality samples in high-dimensional spaces (such as the NSGA-III algorithm, MOEA / D algorithm, etc., as long as they can efficiently cover the non-dominated regions within the descriptor-constrained space). During the expansion process, it is necessary to ensure that the candidate component combinations uniformly cover the high-dimensional component space. For example, in the XYZ-Ti quaternary design space, 3×10^6 sets of candidate components can be generated, ensuring the comprehensiveness of the screening. Subsequently, based on the optimization direction of the descriptor-constrained target space, the optimal component combination is selected from the candidate component combinations. In this step, the descriptor-constrained target space defines the generation range and performance boundary of candidate component combinations, the expansion and generation process ensures the representativeness and comprehensiveness of the samples, and the screening process focuses on the component combinations with the best performance. The three form a logical chain of "range definition - sample generation - selection of the best among the best". The beneficial effects are that it can efficiently cover the high-dimensional composition space, avoid the blindness of traditional trial and error, and quickly identify the candidate combinations of brazing filler metals with the characteristics of "low reactivity, low thermal mismatch, and high plasticity", thus significantly improving screening efficiency and reliability.

[0033] In this embodiment, firstly, based on the initial parameters of the brazing filler metal, a solution model is used to determine the Gibbs free energy and the activity of each component in the brazing-dominant reaction. This provides core thermodynamic data support for the subsequent accurate solution of the reaction equilibrium state, avoiding subsequent calculation deviations caused by the lack of effective capture of key thermodynamic parameters of the reaction. Next, based on the aforementioned Gibbs free energy and component activities, a nonlinear equation for the equilibrium condition is constructed and the equilibrium process is solved. This accurately reflects the actual situation of active element consumption and the introduction of nonmetallic species on the ceramic side in the interfacial reaction, laying a precise foundation for the subsequent calculation of the effective components. Subsequently, based on the initial parameters of the brazing filler metal and the equilibrium process, the molar number of each component is determined, and the effective components of the brazing filler metal are obtained. The effective components that actually participate in joint formation replace the initial design parameters (nominal components) commonly used in related technologies, fundamentally eliminating the design input bias caused by component offset and solving the problem of calculation errors caused by the use of initial design parameters in related technologies. This process addresses the issue of significant result deviations. Subsequently, component descriptors are determined based on the effective components of the solder, and a descriptor-constrained target space is constructed. This establishes the correlation between components and performance based on real components and clear physical logic, avoiding the low accuracy problem caused by machine learning models being disconnected from physical mechanisms. This provides a clear performance guide for subsequent candidate component screening. Finally, candidate component combinations are generated and the optimal combination is screened based on the descriptor-constrained target space. Unlike physical calculation models, which require high-precision thermodynamic and kinetic calculations across all components and reaction paths, this method efficiently expands and screens within a targeted constrained target space, significantly reducing computational power and time consumption. This solves the problem of high cost associated with physical calculation models. Furthermore, relying on the accurate effective components and constraints aligned with physical mechanisms ensures the accuracy of the optimal component combination screening, ultimately achieving high-throughput screening of ceramic-metal solders that balances reliability, efficiency, and accuracy.

[0034] Optionally, the determination of the Gibbs free energy and the activity of each component in the dominant brazing reaction based on the initial parameters of the brazing filler metal and using a solution model includes: Based on the initial parameters of the brazing filler metal, the Gibbs free energy in the dominant brazing reaction is determined using the melt model. Based on the mole fraction constraint, the partial derivative of the expression for the Gibbs free energy is obtained to obtain the expression for the chemical potential of the component. The activity is determined based on the chemical potential expression.

[0035] Specifically, the initial parameters of the solder include the initial mole fraction of the components, the Gibbs free energy of the liquid phase of pure elements Gliq,i, the binary interaction parameters Ak,ij (k=0,1,2), and the thermodynamic temperature T. A displacement-solution model is adopted, according to formula G... m =G ref +G id +Gex The Gibbs free energy in the dominant brazing reaction is determined, where the excess Gibbs free energy Gex is expressed using a Redlich-Kister polynomial (ignoring ternary and higher-order interaction parameters; it can be replaced by an extended polynomial considering some lower-order ternary interaction parameters depending on the system complexity, as long as computational cost is controlled and accuracy is guaranteed). The excess Gibbs free energy G... ex Reference Gibbs free energy G ref and the mixed entropy contribution term G id They are represented as follows: ; ; ; in, For the reaction progress The instantaneous mole fraction of the lower component i, For the reaction progress The instantaneous mole fraction of the lower element j, Let i be the liquid phase Gibbs free energy. Here, represents the binary interaction parameters (from thermodynamic databases or literature), and k is the polynomial order (k = 0, 1, 2). Based on the mole fraction constraint (Σxi=1), the Gibbs free energy G is... m Taking the partial derivative of the expression, we obtain the chemical potential expression for each component. Then, based on the chemical potential expression and the thermodynamic definition of activity, we derive the correlation between component activity, mole fraction, and activity coefficient, thereby determining the activity of each component. The chemical potential expression, correlation, and activity of each component are as follows: ; and, ; ; in, Let i be the chemical potential of component i. For the excess chemical potential of component i, Let be the activity coefficient of component i. Let i represent the activity of component i. In this embodiment, the initial solder parameters provide the necessary basic data for calculating the Gibbs free energy, the mole fraction constraint ensures the rationality of the thermodynamic calculation, and the chemical potential expression bridges the gap between the Gibbs free energy and the activity. All features are interdependent and logically coherent. The beneficial effect is that, through rigorous thermodynamic derivation, the activity of the component is accurately quantified, providing accurate thermodynamic parameters for the subsequent construction and solution of the equilibrium equations, thus ensuring the reliability of the equilibrium process calculation.

[0036] Optionally, the composition descriptor includes a formation tendency descriptor, a valence electron concentration descriptor, and a coefficient of thermal expansion; The step of determining component descriptors based on the effective components of the solder and constructing a descriptor-constrained target space includes: Based on the effective components of the solder, and based on the physical definition and quantification relationship of descriptors, the formation tendency descriptor, the valence electron concentration descriptor, and the coefficient of thermal expansion are determined respectively. Based on the formation tendency descriptor, the valence electron concentration descriptor, and the thermal expansion coefficient, the descriptor-constrained target space is constructed.

[0037] Specifically, the component descriptors include formation tendency descriptors ( The descriptors include the valence electron concentration (VEC) and the coefficient of thermal expansion (CTE). The formation tendency descriptor is used to assess the formation tendency of intermetallic compounds in high-entropy alloys and can be replaced by other thermodynamic criteria that assess the formation tendency of brittle intermetallic compounds. The valence electron concentration descriptor can be adjusted by combining the toughness correlation formula of different alloy systems. The coefficient of thermal expansion can be replaced by a multi-component nonlinear superposition model instead of a linear superposition model. Based on the effective composition of the solder, and based on the physical definition and quantification relationship of the descriptors, each descriptor is calculated separately. For example, for... Value, through formula We obtained, among which, For configurational entropy, For complementary entropy, The excess entropy is used; the VEC value is calculated by weighting the valence electron count and mole fraction of each component; the CTE value is calculated by linear superposition based on the CTE data of each component's pure substance and the mole fraction of the effective component. Subsequently, with the optimization objectives of "maximizing the φ value, maximizing the VEC value, and minimizing the CTE value," and combining the physical meaning of each descriptor, a descriptor-constrained target space is constructed, such as the φ-VEC-CTE three-dimensional space. ; Where n is the number of groups; for example, in a 4-element system, n is 4. Let represent the mole fraction of the i-th element. In this embodiment, the effective component of the solder is the direct input for descriptor calculation. The quantification relationship between each descriptor is crucial to ensuring calculation accuracy. The descriptor-constrained target space integrates multi-dimensional performance requirements, with each feature interconnected, achieving a quantification transformation of performance requirements. The beneficial effect is that it comprehensively covers the core performance indicators in the solder joining process, and the constructed constraint target space clarifies the screening direction, providing clear judgment criteria for the subsequent generation and screening of candidate component combinations.

[0038] For example, configurational entropy Complementary entropy and excess entropy It can be obtained through the following formula: ; ; ; ; in, Let i be the enthalpy of mixing between the i-th and j-th elements. The total mixing entropy is a dimensionless parameter related to mole fraction, atomic radius, and atomic packing. Let i be the mole fraction of the i-th element. Where is the atomic radius. For dimensionless parameters related to atomic stacking, such as the FCC atomic stacking mode. The value is 0.74, representing the BCC atomic packing method. It is 0.68. Melting point The expression is as follows: ; ; ; ; ; ; ; ; in, , as well as All are structural parameters. , Let be the atomic diameters of the i-th and j-th atoms, respectively. Let the diameter of the k-th (k=1~n) atom be the atomic diameter. Z is the number density, and Z is the compressibility. For the dimensionless parameter associated with the stacking of the i-th atom, , These are the mole fractions corresponding to the i-th and j-th atoms, respectively.

[0039] Optionally, the step of expanding and generating candidate component combinations based on the descriptor-constrained target space includes: Based on the descriptor-constrained target space, the Pareto front method is used to expand and generate the candidate component combinations.

[0040] Specifically, the descriptor constrains the target space as -VEC-CTE three-dimensional space, optimization objective is to maximize The goal is to maximize the VEC value and minimize the CTE value. Based on this constrained target space, a Pareto front method is used to expand and generate candidate component combinations. Specifically, in a high-dimensional component space (such as an XYZ-Ti quaternary space), an improved LHS+Gaussian perturbation sampling strategy is used to generate initial candidate samples (which can be replaced by Latin hypercube sampling, Monte Carlo sampling, or other sampling methods that can achieve uniform coverage of the component space). After mapping the initial samples to the descriptor-constrained target space, the Pareto front method is used to screen out non-dominated samples. Based on the distribution characteristics of these non-dominated samples, more candidate component combinations covering high-quality regions are expanded and generated, ensuring that the candidate combinations are both uniformly distributed and focused on regions with superior performance. In this embodiment, the descriptor-constrained target space defines the performance boundary of the candidate component combinations, and the Pareto front method ensures that the expanded candidate combinations have performance balance. The two work together to achieve efficient exploration of the high-dimensional component space. The beneficial effects are that it avoids blindly generating a large number of invalid samples, ensures the quality of candidate combinations while controlling the number of samples, significantly improves the efficiency of high-throughput screening, and balances computational cost and screening effect.

[0041] Optionally, the screening to obtain the optimal combination of components includes: Based on the additional hard constraints, a non-dominated sorting method is used to screen the candidate component combinations to obtain the optimal solution set; The optimal component combination is determined based on the frequency of occurrence of each element in the optimal solution set.

[0042] Optionally, the additional hard constraints include: The value of the valence electron concentration descriptor is greater than a first preset threshold, and the value of the formation tendency descriptor is greater than a second preset threshold.

[0043] Specifically, the additional hard constraint condition is that the value of the valence electron concentration descriptor is greater than a first preset threshold, and the value of the formation tendency descriptor is greater than a second preset threshold. For example, the first preset threshold is set to 7, and the second preset threshold is set to 0, that is, the additional hard constraint condition is set to VEC > 7. The constraint condition >0 can be flexibly adjusted according to the specific performance requirements of the ceramic-metal bonding system (e.g., VEC>6.8). >0.05, etc.). Based on this additional hard constraint, the candidate component combinations generated by the Pareto front method are non-dominated and sorted. Dominated solutions that are comprehensively dominated by other samples in all three objectives are eliminated, and non-dominated solutions that simultaneously satisfy the hard constraint and have the characteristics of "high φ, high VEC, and low CTE" are retained to form the optimal solution set (Pareto optimal set). Subsequently, statistical analysis is performed on all candidate components in the optimal solution set. The occurrence frequency of each single element (such as Cu, Ni, Mo, W, etc.) and element combinations (such as Cu-Mo-Ni, Cu-Ni-W, etc.) is counted, and the element combination with the highest occurrence frequency is selected as the optimal component combination. For example, in the SiCf / SiC-GH5188 system, Cu-Mo-Ni, Cu-Ni-W, and Cr-Cu-Ni are high-frequency and high-quality combinations. In this embodiment, the additional hard constraint ensures the basic performance baseline of the solder, the non-dominated sorting achieves the initial screening of high-quality samples, and the element occurrence frequency statistics converge to the stable and reliable optimal component combination. Each feature forms a logical closed loop of "bottom-line constraint - high-quality screening - precise convergence". The beneficial effect is that it not only ensures that the basic performance of the optimal component combination meets the standards, but also improves the stability and repeatability of the combination through frequency statistics, thereby reducing the risk and cost of subsequent experimental verification.

[0044] Optionally, after obtaining the optimal combination of components through screening, the method further includes: The optimal combination of components was experimentally verified.

[0045] Specifically, in this embodiment, after selecting the optimal composition combination (such as Cu-20Ni-10Ti-10Mo), experimental verification was conducted to confirm its practical application effect. The materials to be joined in the experiment were SiCf / SiC composite material (10mm×10mm×3mm) and GH5188 high-temperature alloy (10mm×10mm×3mm). A brazing filler metal (50μm thickness, adjustable to 40-60μm according to connection requirements) was prepared based on the optimal composition combination. The brazing process was performed in a vacuum environment (vacuum degree ≤5×10^-3Pa), with a heating rate of 10℃ / min, a holding temperature of 1175℃ for 10min, and furnace cooling. These process parameters can be replaced with a heating rate of 8-12℃ / min, a holding temperature of 1125-1225℃, and a holding time of 5-15min. During the experiment, the macro / micro morphology of the joint was observed, and the mechanical performance stability of the joint was evaluated through shear testing. The experimental verification results need to be compared with the screening targets to confirm whether the optimal component combination achieves the connection effect of "crack-free, controllable interface reaction, and stable mechanical properties". In this embodiment, the optimal component combination is the core object of experimental verification, and the experimental process and characterization method are the key means to verify the effect. The two work together to complete the practical feasibility verification of the screening results. The beneficial effect is that the effectiveness of the screening method and the engineering application value of the optimal component combination are intuitively verified through experimental data, avoiding the limitations of purely theoretical screening, and providing a reliable basis for the practical promotion and application of brazing filler metals.

[0046] For example, the purpose of this embodiment is to verify the effectiveness of the "screening method for ceramic metal brazing filler metals" and to screen active brazing filler metal formulations suitable for heterogeneous joining of SiCf / SiC composite material (CMC) and GH5188 high-temperature alloy (Co-based), so as to achieve the joint goal of "crack-free, controllable interface reaction, and stable mechanical properties". The materials to be joined are SiCf / SiC composite material (used in extreme environments such as engine tail nozzles and high heat flux components of advanced nuclear energy systems) with dimensions of 10mm×10mm×3mm and GH5188 high-temperature alloy (Co-based, with excellent high-temperature stability and mechanical properties). The brazing filler metal design space is an XYZ-Ti quaternary system, where X, Y, and Z are matrix and alloying elements (such as Cu, Ni, Mo, W, Cr, etc.), and Ti is the active element. The binary interaction parameters of the core components are taken from thermodynamic databases and literature. The experimental equipment includes a vacuum brazing machine (vacuum degree ≤5×10^-3Pa), a scanning electron microscope (SEM), an energy dispersive spectroscopy (EDS) analyzer, and a universal shear testing machine.

[0047] During implementation, a high-dimensional composition space uniform sampling strategy of "improved LHS + Gaussian perturbation" was first adopted to generate 3×10^6 candidate nominal compositions in the XYZ-Ti quaternary design space to ensure uniform coverage of the composition space. Subsequently, a Si-C-Co-Ti quaternary thermodynamic model was established to identify the types of reactions that Ti and the substrate are more likely to undergo. The results show that Ti has a more significant affinity for the SiC side, and Ti tends to diffuse towards the SiC side (e.g., ...). Figure 2 As shown in the figure, this is a contour plot of ΔGmix and μTi ternary system of Co-Ti-Si-C at 1300K. (ac) is the distribution of ΔGmix and (df) is the distribution of μTi, which intuitively presents the reaction trend of Ti in each subsystem. Finally, it was determined that [Ti]+SiC→TiC+[Si] is the dominant interface reaction with the strongest thermodynamic trend. The initial parameters of the brazing filler metal were used to solve the equilibrium process of the dominant brazing reaction using a solution model to determine the effective composition of the brazing filler metal after welding.

[0048] Next, the 3×10^6 candidate nominal components were used as initial parameters for the solder to determine the effective components. Then, based on the effective components, the component descriptors were calculated respectively. Values, VEC value, and CTE value, set to "maximize". The optimization objectives are "to maximize the VEC value and minimize the CTE value", with the hard constraints VEC>7 and φ>0. The three-dimensional descriptive subspace of "-VEC-CTE" is sorted using the Pareto front method, and non-dominated solutions are retained to form a Pareto optimal set (e.g., ...). Figure 3 As shown, Figure (a) is - A 3D scatter plot in the VEC-CTE space, with blue dots representing all candidate samples and red dots representing the Pareto optimal set; (a1-a3) represents the 2D projection of the 3D space onto the φ-CTE, φ-VEC, and VEC-CTE planes, with green dots representing the Pareto optimal set; (b1-b3) represents the linear fitting results of the 2D projection), and the frequency of occurrence of elements and combinations in the Pareto optimal set is statistically analyzed (e.g., Figure 3 (c) As shown, high-frequency combinations Cu-Mo-Ni, Cu-Ni-W, and Cr-Cu-Ni were screened out, and the composition design principle of "using Cu-Ni as the matrix and adding high-melting-point transition metals such as Mo / W" was established. Finally, experimental verification was carried out. Based on the selected composition window, three candidate solders with different alloying strategies (mass fraction) were prepared: Cu-20Ni-10Ti-10Co, Cu-20Ni-10Ti-10Fe, and Cu-20Ni-10Ti-10Mo (all solder thickness was 50μm). The solders were heated to 1175℃ in a vacuum environment at a heating rate of 10℃ / min, held for 10min, and then cooled with the furnace. The macro / micro morphology of the joint was observed, EDS elemental analysis was performed, and shear strength was tested.

[0049] Experimental results show that, in terms of the macroscopic morphology of the joint, such as Figure 4 As shown, Figure 4 (a) Cu-20Ni-10Ti-10Co brazing connector and Figure 4 (b) The Cu-20Ni-10Ti-10Fe brazing filler joints all showed obvious cracks, while Figure 4 (cf) The Cu-20Ni-10Ti-10Mo brazing filler joint showed no obvious cracks and excellent forming quality; the microstructure and EDS analysis results are as follows: Figure 5 As shown in the table, controlled TiC formation occurs at the Cu-20Ni-10Ti-10Mo joint interface. x The reaction layer (P1 point: Ti 18.65%, C 66.47%, Mo 4.08%) and the finely dispersed Ni-Mo-Si second phase (P2 point: Ni 19.62%, Mo 30.95%, Si 13.48%) in the weld have favorable microstructure. In contrast, the reaction layer at the joint interface of the control group is uneven, and a large number of brittle intermetallic compounds appear in the weld, leading to joint cracking. Mechanical property tests show that the shear strength of the Cu-20Ni-10Ti-10Mo brazing filler joint is more than 30% higher than that of the control group, and the mechanical property stability is significantly improved, meeting the connection strength requirements of the SiCf / SiC-GH5188 system.

[0050] like Figure 6 As shown, an embodiment of the present invention provides a ceramic metal brazing filler metal screening device, comprising: The reaction module is used to determine the Gibbs free energy and the activity of each component in the brazing-dominant reaction based on the initial parameters of the brazing filler metal and using a solution model. The process module is used to construct and solve the equilibrium condition nonlinear equation based on the Gibbs free energy and the activity to obtain the equilibrium process. An effective module is used to determine the molar number of each component in the balancing process based on the initial parameters of the solder, and to determine the effective component of the solder based on the molar number. The description module is used to determine the component descriptors based on the effective components of the solder, and to construct the descriptor constraint target space; The filtering module is used to expand and generate candidate component combinations based on the target space constrained by the descriptor, and filter to obtain the optimal component combination.

[0051] like Figure 7 As shown, an electronic device 700 provided in this embodiment of the invention includes a memory 710 and a processor 720; the memory 710 is used to store a computer program; the processor 720 is used to implement the ceramic metal brazing filler metal screening method as described above when the computer program is executed.

[0052] Alternatively, an electronic device 700 includes a memory 710 and a processor 720 coupled to the memory 710; the memory 710 is configured to store a computer program; and the processor 720 is configured to perform the following operations when the computer program is executed: Based on the initial parameters of the brazing filler metal, a solution model was used to determine the Gibbs free energy and the activity of each component in the dominant brazing reaction. Based on the Gibbs free energy and the activity, a nonlinear equation for the equilibrium condition is constructed and solved to obtain the equilibrium process. Based on the initial parameters of the solder, the molar number of each component in the balancing process is determined, and the effective component of the solder is determined according to the molar number. Based on the effective components of the solder, component descriptors are determined, and a descriptor-constrained target space is constructed; Based on the descriptor constraining the target space, candidate component combinations are expanded and generated, and the optimal component combination is obtained by screening.

[0053] This invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the ceramic metal brazing filler metal screening method described above.

[0054] Alternatively, a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the following operations: Based on the initial parameters of the brazing filler metal, a solution model was used to determine the Gibbs free energy and the activity of each component in the dominant brazing reaction. Based on the Gibbs free energy and the activity, a nonlinear equation for the equilibrium condition is constructed and solved to obtain the equilibrium process. Based on the initial parameters of the solder, the molar number of each component in the balancing process is determined, and the effective component of the solder is determined according to the molar number. Based on the effective components of the solder, component descriptors are determined, and a descriptor-constrained target space is constructed; Based on the descriptor constraining the target space, candidate component combinations are expanded and generated, and the optimal component combination is obtained by screening.

[0055] The present invention will now be described an electronic device 700 that can serve as a server or client of the present invention, which is an example of a hardware device that can be applied to various aspects of the present invention. Electronic device 700 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 700 can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0056] Electronic device 700 includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) or a computer program loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0057] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. In this application, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention according to actual needs. Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units can be implemented in hardware or as software functional units.

[0058] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A method for screening ceramic metal brazing filler metal, characterized in that, include: Based on the initial parameters of the brazing filler metal, a solution model was used to determine the Gibbs free energy and the activity of each component in the dominant brazing reaction. Based on the Gibbs free energy and the activity, a nonlinear equation for the equilibrium condition is constructed and solved to obtain the equilibrium process. Based on the initial parameters of the solder, the molar number of each component in the balancing process is determined, and the effective component of the solder is determined according to the molar number. Based on the effective components of the solder, component descriptors are determined, and a descriptor-constrained target space is constructed; Based on the descriptor constraining the target space, candidate component combinations are expanded and generated, and the optimal component combination is obtained by screening.

2. The method for screening ceramic metal brazing filler metal according to claim 1, characterized in that, Based on the initial parameters of the brazing filler metal, a solution model is used to determine the Gibbs free energy and the activity of each component in the dominant brazing reaction, including: Based on the initial parameters of the brazing filler metal, the Gibbs free energy in the dominant brazing reaction is determined using the melt model. Based on the mole fraction constraint, the partial derivative of the expression for the Gibbs free energy is used to obtain the expression for the chemical potential of the component. The activity is determined based on the chemical potential expression.

3. The method for screening ceramic metal brazing filler metal according to claim 1, characterized in that, The composition descriptors include formation tendency descriptors, valence electron concentration descriptors, and coefficients of thermal expansion; The step of determining component descriptors based on the effective components of the solder and constructing a descriptor-constrained target space includes: Based on the effective components of the solder, and based on the physical definition and quantification relationship of descriptors, the formation tendency descriptor, the valence electron concentration descriptor, and the coefficient of thermal expansion are determined respectively. Based on the formation tendency descriptor, the valence electron concentration descriptor, and the thermal expansion coefficient, the descriptor-constrained target space is constructed.

4. The method for screening ceramic metal brazing filler metal according to claim 1, characterized in that, The step of expanding and generating candidate component combinations based on the descriptor-constrained target space includes: Based on the descriptor-constrained target space, the Pareto front method is used to expand and generate the candidate component combinations.

5. The method for screening ceramic metal brazing filler metal according to claim 1, characterized in that, The screening process to obtain the optimal component combination includes: Based on the additional hard constraints, a non-dominated sorting method is used to screen the candidate component combinations to obtain the optimal solution set; The optimal component combination is determined based on the frequency of occurrence of each element in the optimal solution set.

6. The method for screening ceramic metal brazing filler metal according to claim 5, characterized in that, The additional hard constraints include: The value of the valence electron concentration descriptor is greater than a first preset threshold, and the value of the formation tendency descriptor is greater than a second preset threshold.

7. The method for screening ceramic metal brazing filler metal according to claim 1, characterized in that, After obtaining the optimal combination of components through screening, the process further includes: The optimal combination of components was experimentally verified.

8. A screening device for ceramic metal brazing filler metal, characterized in that, include: The reaction module is used to determine the Gibbs free energy and the activity of each component in the brazing-dominant reaction based on the initial parameters of the brazing filler metal and using a solution model. The process module is used to construct and solve the equilibrium condition nonlinear equation based on the Gibbs free energy and the activity to obtain the equilibrium process. An effective module is used to determine the molar number of each component in the balancing process based on the initial parameters of the solder, and to determine the effective component of the solder based on the molar number. The description module is used to determine the component descriptors based on the effective components of the solder, and to construct the descriptor constraint target space; The filtering module is used to expand and generate candidate component combinations based on the target space constrained by the descriptor, and filter to obtain the optimal component combination.

9. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to, when executing the computer program, implement the method for screening ceramic metal brazing filler metal as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method for screening ceramic metal brazing filler metal as described in any one of claims 1 to 7.

Images