Aluminum-based material design method based on physical constraint state space reachability determination

Abstract

The application relates to an aluminum-based material design method based on physical constraint state space reachability determination, and belongs to the technical field of aluminum-based material design. The method constructs a high-dimensional material state space formed by a composition space, a heat treatment state space, a processing state space, a material state agent variable space and a performance evidence state space, constructs a material state reachability network subjected to physical constraints, and establishes performance projection mapping; a target performance region is received, a reachability closure is generated from an initial available node set, a high-dimensional promotion set of the target performance region is constructed, according to the intersection condition of the reachability closure and the high-dimensional promotion set, it is determined that there is a reachable state, there is a potential compensation state or a real unreachable state, and a determination result is output. The application can distinguish between projected unreachable and real unreachable, and output traceable unreachable proof and a minimum relaxation direction.

Description

Technical Field

[0001] This invention belongs to the field of aluminum-based material design technology, and relates to an aluminum-based material design method based on physical constraint state space reachability determination. Background Technology

[0002] With the widespread application of data-driven methods in materials science, new methods for composition design and performance recommendation of aluminum-based materials are constantly emerging. Current mainstream methods include: machine learning-based composition-performance prediction methods, which use alloy composition and process parameters as input to establish a black-box mapping relationship between composition, process, and performance through neural networks or tree models; reverse design methods, which search for the optimal composition combination in the composition space after setting target performance, such as composition reverse design methods based on Bayesian optimization, genetic algorithms, or latent space optimization; material performance mapping methods, which plot material properties in a two-dimensional or multi-dimensional performance space and use performance index lines to assist in material selection, represented by Ashby graphs; and knowledge graph and graph search methods, which express material knowledge as a graph structure and use graph reasoning to assist in material recommendation.

[0003] However, the existing methods described above have the following shortcomings in the design process of aluminum-based materials: First, existing methods directly equate the lack of information in the low-dimensional performance space with physical inaccessibility. The true state of aluminum-based materials is determined by the high-dimensional space, with multiple factors such as composition, heat treatment state, processing path, precipitation state, and performance evidence working together to ultimately project the performance onto the low-dimensional performance space. This is a typical high-dimensional to low-dimensional projection process, inevitably accompanied by information loss. Existing methods uniformly classify the gaps in the low-dimensional projection as inaccessible or skip them directly, failing to distinguish between the two scenarios of "sparse samples but accessible in the high-dimensional space" and "no accessible path in the high-dimensional state space." Taking the regressive reaging (RRA) state of 7055 aluminum alloy as an example, the tensile strength of this state can reach over 660 MPa, but direct experimental data on its thermal conductivity are extremely scarce in published literature. This region appears blank in the tensile strength-thermal conductivity two-dimensional performance space, and conventional methods would classify this region as inaccessible or skip it directly; however, the 7055-RRA state does not actually not exist physically, but its thermal conductivity data has not yet been fully measured.

[0004] Second, the graph structures constructed by existing methods lack constraints on the physical evolution of materials, making them data graphs rather than physically reachable graphs. Existing knowledge graph or graph search methods construct graph structures from material data, but the edges in these graphs are usually based on data co-occurrence or similarity, rather than on the laws governing the physical evolution of materials. For example, there is a real process evolution relationship between the T6 state and the RRA state, where RRA is achieved by adding regression and re-aging steps to the T6 state; however, the transition from the T73 state to the T6 state is physically irreversible, and the coarse precipitates caused by over-aging cannot be restored to the peak aging state through simple heat treatment. Ordinary knowledge graphs cannot distinguish the physical properties of these two types of edges, and the graph structure lacking physical constraints leads to incorrect state transition determinations.

[0005] Third, existing methods only output "not found" when the search fails, which cannot prove that there is no reachable path in the constrained state space. Existing reverse engineering methods usually output "no component that meets the conditions was found" when the search fails, but this only indicates that the search algorithm has not converged or that there is insufficient data, and cannot prove that the target performance is indeed physically unreachable; users cannot determine whether "not found" is due to algorithm problems, insufficient data, or whether the performance combination does not actually exist in the constrained manufacturable state space. Summary of the Invention

[0006] To address the technical problem that existing methods directly equate the lack of information in the low-dimensional performance space with physical inaccessibility and fail to distinguish between projected inaccessibility and actual inaccessibility, this invention provides an aluminum-based material design method based on the reachability determination of the physical constraint state space. This method formally determines whether there is a reachable material state in the target performance region within the constrained manufacturable state space and outputs a traceable determination result.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A design method for aluminum-based materials based on physical constraint state-space reachability determination includes the following steps: A high-dimensional material state space is constructed, which is spanned by the composition space, heat treatment state space, processing state space, material state proxy variable space, and performance evidence state space of aluminum-based materials. The composition space is spanned by the mass percentage of each alloying element in the aluminum-based material. Constructing physically constrained material state reachable networks Where V is the set of material state nodes, Let Ω be the set of state transition edges, and Ω be the set of state transition constraints; the state transition edges are generated by the material state evolution operator. Generate, where u is a material processing action, and the state transition edge is subject to the activation conditions of composition constraints, heat treatment constraints, and physical constraints; Establish performance projection mapping The high-dimensional material state is mapped to a low-dimensional performance projection space Y, which is spanned by at least two performance indices of the aluminum-based material. Receive target performance area The target performance region is defined by the target value of the performance metric; from the initial set of available nodes. Starting from this point, search along the state transition edges that satisfy the state transition constraint set Ω to generate a reachable closure. ; Construct a high-dimensional enhancement set for the target performance region. The high-dimensional enhancement set is a set of high-dimensional material states that can be projected onto the target performance region; based on the intersection of the reachability closure and the high-dimensional enhancement set, it is determined to be one of the following: a reachable state, a potential compensation state, or a truly unreachable state. The output state space can reach the decision result.

[0008] Furthermore, the performance evidence state space is defined by the data confidence type and the source reliability.

[0009] Furthermore, the material processing action u includes at least one of composition adjustment action, heat treatment action, and deformation treatment action; the heat treatment action includes at least one of solution treatment, aging treatment, regression treatment, and re-aging treatment.

[0010] Furthermore, the state transition edge includes at least one of composition transition edge, heat treatment transition edge, and state completion edge; the state completion edge transforms the material state node from an incomplete state to a complete state by supplementing direct performance measurement data.

[0011] Furthermore, the state transition constraint set Ω includes at least two of the following: composition constraints, heat treatment constraints, and physical constraints; the heat treatment constraints include the requirement that the state transition direction must conform to the heat treatment process evolution direction; the physical constraints are based on the material state evolution consistency function. It is confirmed that monotonous performance is not required.

[0012] Furthermore, the reachable closure To start from the initial set of available nodes Starting from the state transition edge that satisfies the state transition constraint set Ω, we find the set of all material state nodes reachable from it.

[0013] Furthermore, the condition for determining the existence of a reachable state is: .

[0014] Furthermore, the criteria for determining the existence of a potential compensation state are as follows: Furthermore, there exists a reachable node that meets the target requirements in at least one performance dimension, and can be transformed into a usable state through state-complete edges.

[0015] Furthermore, the criteria for determining the truly unreachable state are as follows: Furthermore, there are no material state nodes that can be transformed into usable states through state completion edges; the determination result of the true unreachable state also includes the unreachability proof and the minimum relaxation direction.

[0016] Furthermore, the present invention also provides an aluminum-based material design system, comprising: The high-dimensional state space construction module is used to construct the high-dimensional material state space X; The reachable network construction module, connected to the high-dimensional state space construction module, is used to receive the high-dimensional material state space output by the high-dimensional state space construction module and construct a physically constrained material state reachable network. ; The projection mapping construction module, connected to the high-dimensional state space construction module and the reachable network construction module, is used to establish a performance projection mapping. ; The input module is used to receive user input regarding the target performance area. ; The reachability determination module, connected to the reachable network construction module, the projection mapping construction module, and the input module, is used to perform state space reachability determination, including generating reachable closures. Constructing a high-dimensional enhanced set And output the determination result based on the intersection of the reachable closure and the high-dimensional lifting set; An output module, connected to the reachability determination module, is used to receive and output the determination result output by the reachability determination module.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention constructs a physically constrained material state reachability network and generates reachability closures, thereby enabling a systematic determination of whether reachable material states exist in a target performance region. This overcomes the shortcomings of existing methods that indiscriminately classify low-dimensional performance voids as unreachable or skip them directly.

[0018] This invention binds state transitions to physically executable material processing actions through a material state evolution operator, making the state transition edges different from those constructed based on data co-occurrence or similarity in existing knowledge graphs. This results in the construction of a physically reachable graph rather than a data graph, avoiding erroneous state transition determinations caused by graph structures lacking physical constraints.

[0019] This invention distinguishes between projected inaccessibility and true inaccessibility by determining the intersection of a high-dimensional lifting set and an accessible closure. It reveals that low-dimensional performance voids may be caused only by the loss of projected information, thereby avoiding misjudging material states that are sparse in samples but accessible in high dimensions as inaccessible.

[0020] This invention overcomes the shortcomings of existing methods that only output "not found" when the search fails, without proving that the target performance is indeed unreachable in the confined manufacturable state space, by outputting proof of unreachability in the confined space and the minimum relaxation direction. This provides a traceable basis for material design decisions.

[0021] This invention constrains state transitions through a state evolution consistency function and does not require performance monotonicity. It only requires state transitions to satisfy constraints on composition continuity, process continuity, and performance variation range, which is more in line with the actual physical properties of materials and makes the state transition determination results more accurate and reliable.

[0022] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will now be described in detail with reference to the accompanying drawings, wherein: Figure 1 This is an overall flowchart of the aluminum-based material design method based on physical constraint state space reachability determination of the present invention; Figure 2 For high-dimensional material state space A schematic diagram of the structure; Figure 3 This is a schematic diagram of the physically constrained material state reachability network in Example 2, showing the material state nodes and the material state evolution operator. The relationship between the generated state transition edges and the set of state transition constraints Ω; Figure 4 For reachable closure With higher-dimensional lifting sets Intersection determination diagram; Figure 5 A diagram illustrating the distinction between projection inaccessibility and actual inaccessibility; Figure 6 A schematic diagram showing the connection relationships between the modules of the system designed for aluminum-based materials. Detailed Implementation

[0024] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0025] Example 1 Existing methods treat material data as independent sample points, failing to establish a structured representation of the high-dimensional material state space, and the relationships between different dimensions are not systematically modeled. Therefore, this embodiment provides a method for constructing a high-dimensional material state space and mapping samples.

[0026] High-dimensional material state space Composition space C, heat treatment state space Processing state space P, material state proxy variable space And the performance evidence state space Q is formed.

[0027] The composition space C is defined by the mass percentage of each alloying element in the aluminum-based material. Let the aluminum-based material contain n alloying elements, and the mass percentage of each element be... (j=1,2,…,n), then the component space is The mass percentage of aluminum is determined by the balance: .

[0028] Taking 7xxx series aluminum alloys as an example, the composition space is defined by the mass percentages of four elements: zinc, magnesium, copper, and zirconium. In the sample library, the zinc content ranges from 5.0% to 9.0%, magnesium from 1.5% to 3.0%, copper from 1.0% to 2.8%, and zirconium from 0.05% to 0.20%. For example, 7055 aluminum alloy typically contains 7.6-8.4% zinc, 2.0-2.6% magnesium, 2.0-2.6% copper, and 0.08-0.15% zirconium. 7075 aluminum alloy typically contains 5.1-6.1% zinc, 2.1-2.9% magnesium, and 1.2-2.0% copper, partially overlapping with but not entirely the same composition range as 7055. 6082 aluminum alloy, belonging to the 6xxx series, primarily contains magnesium and silicon, and its composition space is significantly different from the 7xxx series.

[0029] Heat treatment state space It is composed of parameters such as heat treatment type, temperature, and time. Assume the heat treatment process consists of m stages, with each stage having a parameter of temperature. With time The combination of these methods. The sample library in this embodiment includes heat treatment types such as T6, T73, regression re-aging, and natural aging. Taking regression re-aging as an example, its state vector includes eight parameters: solution temperature, solution time, aging temperature, aging time, regression temperature, regression time, re-aging temperature, and re-aging time. The typical process for 7055-T6 is solution treatment at approximately 470℃ and aging at approximately 120℃ for 24 hours; 7055-RRA adds a regression treatment of approximately 180-200℃ for 15-30 minutes, followed by re-aging at approximately 120℃ for 24 hours, on top of T6.

[0030] The processing state space P is spanned by parameters such as processing type and deformation amount. Processing types include cold working, hot working, equal channel angle extrusion, high-pressure torsion, rolling, and forging. Material state proxy variable space. The state space Q, representing the performance evidence, is defined by indicators such as precipitation state proxy, hardness, and electrical conductivity, which reflect the internal state of a material. The data confidence type and source reliability define the state space Q. Approximately 60% of the samples in the database are directly measured, about 30% are indirectly sourced, and about 10% are estimated.

[0031] Taking a 7055-T6 state sample as an example, its complete material state includes: the mass percentages of the main elements in the alloy composition are approximately 8.0% zinc, 2.3% magnesium, 2.2% copper, and 0.12% zirconium; the heat treatment type is T6, with a solution temperature of approximately 470℃, a solution time of approximately 1-2 hours, an aging temperature of approximately 120℃, and an aging time of approximately 24 hours; the target properties include a tensile strength of approximately 620 MPa and a thermal conductivity of approximately 140 W / mK; the evidence type is direct measurement. This sample is mapped to a state point in a high-dimensional material state space. Taking a 7075-T73 state sample as an example, its composition ranges from 5.1-6.1% zinc, 2.1-2.9% magnesium, and 1.2-2.0% copper; the heat treatment is T73; the tensile strength is approximately 510-560 MPa, and the thermal conductivity is approximately 150-160 W / mK. Taking a sample of 6082-T6 condition as an example, it belongs to the 6xxx series, with magnesium and silicon as the main alloying elements, a tensile strength of about 290-340MPa, and a thermal conductivity of about 160-180W / mK.

[0032] Existing knowledge graph or graph search methods construct material data into graph structures, but the edges in these graphs are typically built based on data co-occurrence or similarity, rather than on the physical evolution of materials. Therefore, this embodiment provides a method for constructing a physically constrained material state reachability network.

[0033] I. Aggregation of Material State Nodes Based on the material states in the sample library, aggregation is performed according to alloy system, main composition range, heat treatment state, and property projection range to generate the following material state nodes: First node The corresponding composition range is 6.5% to 7.5% zinc, 2.0% to 2.5% magnesium, and 1.2% to 1.8% copper, corresponding to the 7075 alloy category. The heat treatment state is T6, and the projected properties are tensile strength of 560 to 620 MPa and thermal conductivity of 130 to 150 W / mK. The condition completeness is high.

[0034] Second node The corresponding composition range is 7.5% to 8.5% zinc, 2.2% to 2.8% magnesium, and 1.6% to 2.2% copper, corresponding to the 7055 alloy category. The heat treatment condition is reversion and reaging, and the projected properties are tensile strength of 620 to 670 MPa, thermal conductivity of 140 to 165 W / mK, and medium condition completeness.

[0035] Third node The corresponding composition range is the same as the first node, namely class 7075, the heat treatment state is T73, the performance projection is tensile strength of 500 to 580 MPa, thermal conductivity of 145 to 170 W / mK, and the state completeness is high.

[0036] Fourth node The corresponding composition range is 0.5% to 1.5% magnesium and 0.4% to 0.8% silicon, corresponding to alloy type 6082, with a heat treatment state of T6, and projected performance of tensile strength of 290 to 340 MPa, thermal conductivity of 160 to 180 W / mK, and high condition completeness.

[0037] II. Material State Evolution Operators and State Transition Edges The state transition edge is determined by the material state evolution operator. generate, Indicates the state of the material Under the action of material processing, it evolves into The applicability of the state evolution operator depends on the action u on the current state. Whether it is physically feasible, and the new state that arises after the action is performed. Whether the constraints of composition, heat treatment and physical constraints are met, as well as whether the state completeness Cpl after the action is executed is not lower than the preset threshold, are jointly determined.

[0038] Define a heat treatment transfer edge from the first node to the second node, corresponding to the material processing action u=RRA, i.e., regression re-aging treatment. State evolution operator. This indicates that 7075-T6, after regression and re-aging treatment, evolves into the 7055-RRA state. This transition satisfies composition constraints (composition range within feasible limits), heat treatment constraints, T6→RRA being an allowable direction, physical constraints, and a consistency function. This side is now enabled.

[0039] Define a heat treatment transition edge from the first node to the third node, corresponding to the material processing action u=overaging, i.e., over-aging treatment. State evolution operator. The transition is enabled by all constraints.

[0040] Define a heat treatment transfer edge from the third node to the first node, corresponding to the material processing action u=revert, i.e., attempting to recover from over-aging. State evolution operator. This is physically unworkable because the heat treatment constraint fails, T73→T6 is an unallowed direction, and this edge is not enabled.

[0041] Define the composition transfer edge from the first node to the fourth node, corresponding to the material processing action. This refers to switching from the 7xxx series to the 6xxx series. The state evolution operator is physically inoperable because the component constraints fail, the component interval change exceeds the threshold, and the edge is not enabled.

[0042] III. Definition of State-Complete Edges The state-completeness edge indicates that the second node can be transformed from an incomplete state to a complete state by supplementing it with direct measurement data. The tensile strength of the second node has been confirmed, but the thermal conductivity data is only available from indirect sources. By using the state-completeness edge to perform direct thermal conductivity measurements on the 7055-RRA, the state completeness of this node can be improved from medium to high.

[0043] This embodiment verifies that a physically constrained material state reachability network can distinguish between reversible and irreversible state transitions, as well as compositional barriers between different alloy systems. State transition edges are triggered by material processing actions u, unlike data co-occurrence edges in ordinary knowledge graphs.

[0044] Example 2 This embodiment demonstrates the core innovation of the present invention: voids in two-dimensional performance space are judged as unreachable by traditional methods, while the present invention distinguishes them as projected unreachability rather than actual unreachability.

[0045] The user inputs target performance requirements: tensile strength not less than 500 MPa, and thermal conductivity not less than 180 W / mK. In the tensile strength-thermal conductivity 2D graph, this region has zero sample points. Taking 7055-RRA as an example, its tensile strength is approximately 660 MPa, but thermal conductivity data is extremely scarce, appearing as a blank area in the 2D graph. Conventional Ashby graph methods skip this blank area, and ordinary machine learning reverse engineering methods output "no component found that meets the conditions." The user cannot determine whether the blank area is due to insufficient data or because the performance combination is physically impossible to achieve.

[0046] In this invention, the first available node is the first node. This refers to the 7075-T6 state, with a high state completeness. From Starting from the material state evolution operator Reaching the second node That is, the 7055-RRA state, along Reaching the third node That is, the 7075-T73 state generates a reachable closure. .

[0047] Build target area higher-dimensional enhancement set The intersection of the performance projection of each node and the target region is determined as follows: That is, the tensile strength of 7055-RRA is 620-670MPa, which meets the target, while the thermal conductivity is 140-165W / mK, which does not meet the target. However, there is a state completeness boundary, and the actual value can be confirmed by direct thermal conductivity measurement.

[0048] System judgment: Although However, it exists Make the consistency function Less than the threshold, and There exists a state-complete edge. The objective is projectively unreachable but not actually unreachable, belonging to a potential compensated state.

[0049] The output judgment result is as follows: a potential compensated state exists; the compensated state is 7055-RRA; the composition corridor is zinc 7.5-8.5%, magnesium 2.2-2.8%, copper 1.6-2.2%, zirconium 0.08-0.15%; the material processing sequence is T6 → regression and re-aging, specifically solution treatment at approximately 470℃ → aging at approximately 120℃ for 24 hours → regression at approximately 180-200℃ for 15-30 minutes → re-aging at approximately 120℃ for 24 hours; the projected performance is tensile strength of 620-670 MPa and thermal conductivity of 140-165 W / mK, which needs to be directly verified; the state completion suggestion is to perform direct thermal conductivity measurement on 7055-RRA; the most recent high-confidence candidate is 7075-T7451, with a tensile strength of approximately 510 MPa and a thermal conductivity of approximately 155 W / mK.

[0050] This embodiment verifies that the present invention can distinguish between projected inaccessibility and actual inaccessibility, correcting the erroneous judgment of traditional methods that equates low-dimensional voids with physical inaccessibility. 7055-RRA appears as a void in the tensile strength-thermal conductivity two-dimensional diagram, but through high-dimensional state reachability network search and high-dimensional enhancement set determination, it was found that this state has a clear physically reachable path in terms of composition and heat treatment dimensions.

[0051] Example 3 This example demonstrates the process of determining and recommending the existence of reachable states.

[0052] User-input target performance requirements: tensile strength not less than 580MPa, yield strength not less than 500MPa, thermal conductivity not less than 130W / mK.

[0053] The first available node is the first node. That is, 7075-T6. From Starting from the edges that satisfy the physical constraints, perform a graph search to generate a reachable closure. The high-dimensional uplifted set of the target region It contains all high-dimensional states that can be projected onto this region. First node. The projected performance values ​​are tensile strength of 560 to 620 MPa, yield strength of approximately 500 MPa, and thermal conductivity of 130 to 150 W / mK, which meet the target requirements. It is determined that there is a reachable state.

[0054] The output judgment result is as follows: an attainable state exists; the recommended state is 7075-T6; the composition corridor is zinc 5.1-6.1%, magnesium 2.1-2.9%, copper 1.2-2.0%, and chromium 0.18-0.28%; the heat treatment state is T6, i.e., solution treatment at approximately 480℃ → aging at approximately 120℃ for 24 hours; the projected properties are tensile strength of 560-620MPa and thermal conductivity of 130-150W / mK; the level of evidence is high.

[0055] Example 4 This embodiment demonstrates the determination of a true unreachable state and outputs the proof of unreachability and the minimum relaxation direction.

[0056] User-input target performance requirements: tensile strength not less than 650MPa, thermal conductivity not less than 220W / mK.

[0057] reachable closure .

[0058] The performance projection of each node is as follows: That is, 7075-T6 has a tensile strength of 560-620MPa and a thermal conductivity of 130-150W / mK; That is, 7055-RRA has a tensile strength of 620-670MPa and a thermal conductivity of 140-165W / mK; That is, 7075-T73 has a tensile strength of 500-580MPa and a thermal conductivity of 145-170W / mK. It was determined to be real but unreachable.

[0059] Proof of unreachability: Reachable closure The target region has a tensile strength ≥ 650 MPa and a thermal conductivity ≥ 220 W / mK; the performance dimension that makes it unattainable is the thermal conductivity dimension; the upper bound of the thermal conductivity of all reachable nodes is lower than the target value of 220 W / mK, i.e. =150、 =165、 =170, all less than 220. The minimum relaxation direction is: Option 1, maintain tensile strength ≥650MPa, relax the thermal conductivity target from 220W / mK to 155-165W / mK; Option 2, maintain thermal conductivity ≥220W / mK, relax the tensile strength target from 650MPa to 330-350MPa.

[0060] This embodiment verifies that the unreachability proof and minimum relaxation direction method can output traceable unreachability determination criteria and clear adjustment directions.

[0061] Example 5 This embodiment demonstrates the calculation process of state-complete edges and state completeness scores.

[0062] A state-completeness edge represents transforming a node with incomplete but potentially existing information into a complete state by supplementing missing state information. The tensile strength of 7055-RRA has been confirmed, measured at approximately 660-668 MPa, but thermal conductivity data is only available from indirect sources. A state-completeness edge refers to completing the state through direct thermal conductivity measurements, such as laser flash.

[0063] For the 7055-RRA node =15, meaning direct experimental samples, mainly tensile strength data. =50, meaning samples from highly reliable sources. =0.6, which is the information completeness score. Points were deducted because thermal conductivity data was only obtained from indirect sources. Let a=0.05, b=0.02, c=0.3, then the calculation is as follows: Below the threshold of 0.7, it is judged as medium completeness. After supplementing the direct thermal conductivity measurement through state completeness edge supplementation, It increased from 0.6 to 0.9. It improved to approximately 0.67, approaching high completeness.

[0064] The recommended state completion for the output is as follows: the state to be tested is 7055-RRA; the performance index to be tested is thermal conductivity; the recommended experimental method is direct thermal conductivity measurement, i.e., laser flash method; the expected performance range is 140-165 W / mK; and the expected level of evidence after completion is medium to high.

[0065] Example 6 This embodiment uses the 2xxx series 2024-T8 alloy as an example to demonstrate the determination of the potential reachable state.

[0066] 2xxx series aluminum alloys, i.e., Al-Cu-Mg series, possess high strength-thermal conductivity balance properties, but their thermal conductivity data vary considerably across different literatures. In the 2024-T8 condition (solution treatment + cold working + artificial aging), there exists a sparse region in the range of tensile strength approximately 450-500 MPa and thermal conductivity approximately 150-170 W / mK; traditional methods classify this region as unreachable.

[0067] User-input targets: tensile strength not less than 470 MPa, thermal conductivity not less than 165 W / mK. Initially available nodes are in the 2024-T3 state. Its tensile strength is approximately 430-480 MPa, and its thermal conductivity is approximately 140-160 W / mK. From... Starting from the heat treatment transfer edge Reaching the 2024-T8 state node Its composition ranges from 3.8-4.9% copper, 1.2-1.8% magnesium, and 0.3-0.9% manganese, with projected properties of tensile strength of 450-520 MPa and thermal conductivity of 150-170 W / mK. It can generate closure. Construct a high-dimensional enhanced set of the target region and discover... The performance projection partially overlaps with the target area. Although The thermal conductivity has high confidence data but few, but there are state-complete edges. The system determines that the target belongs to a potential compensation state.

[0068] Example 7 This embodiment uses the 6xxx series 6082-T6 alloy as an example to demonstrate proof of a truly unattainable state.

[0069] 6xxx series aluminum alloys have excellent thermal conductivity, but relatively low tensile strength. The thermal conductivity of 6082-T6 can reach 170-190 W / mK, but the tensile strength is usually 290-340 MPa.

[0070] User-input targets: tensile strength not less than 400 MPa, thermal conductivity not less than 180 W / mK. Initially available nodes are in state 6082-T6. Its composition ranges from 0.6-1.2% magnesium, 0.7-1.3% silicon, and 0.4-1.0% manganese, with projected properties of tensile strength of 290-340 MPa and thermal conductivity of 170-190 W / mK.

[0071] reachable closure Construct a high-dimensional enhanced set for the target region. The thermal conductivity of 170-190 W / mK can meet the requirement of 180 W / mK, but the tensile strength of 290-340 MPa cannot reach 400 MPa. Starting from 6082-T6, without heat treatment transfer edge, the tensile strength can be increased to 400 MPa, that is, the upper limit of the peak aging tensile strength of 6xxx series is about 360 MPa. Without composition transfer edge, 6082 can be transformed into 7xxx series. It was determined to be real but unreachable.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A design method for aluminum-based materials based on physical constraint state-space reachability determination, characterized in that, Includes the following steps: A high-dimensional material state space is constructed, which is spanned by the composition space, heat treatment state space, processing state space, material state proxy variable space, and performance evidence state space of aluminum-based materials. The composition space is spanned by the mass percentage of each alloying element in the aluminum-based material. Constructing physically constrained material state reachable networks: Where V is the set of material state nodes, Let Ω be the set of state transition edges, and Ω be the set of state transition constraints; the state transition edges are generated by the material state evolution operator. Generate, where u is a material processing action, and the state transition edge is subject to the activation conditions of composition constraints, heat treatment constraints, and physical constraints; Establish performance projection mapping The high-dimensional material state is mapped to a low-dimensional performance projection space Y, which is spanned by at least two performance indices of the aluminum-based material; the target performance region is received. The target performance region is defined by the target value of the performance metric; from the initial set of available nodes. Starting from this point, search along the state transition edges that satisfy the state transition constraint set Ω to generate a reachable closure. ; Construct a high-dimensional enhancement set for the target performance region. The high-dimensional enhancement set is a set of high-dimensional material states that can be projected onto the target performance region; based on the intersection of the reachability closure and the high-dimensional enhancement set, it is determined to be one of the following: a reachable state, a potential compensation state, or a truly unreachable state. The output state space can reach the decision result.

2. The aluminum-based material design method according to claim 1, characterized in that, The performance evidence state space is defined by the data confidence type and the source reliability.

3. The aluminum-based material design method according to claim 1, characterized in that, The material processing action u includes at least one of composition adjustment action, heat treatment action and deformation treatment action; The heat treatment process includes at least one of solution treatment, aging treatment, reversion treatment, and re-aging treatment.

4. The aluminum-based material design method according to claim 1, characterized in that, The state transition edge includes at least one of the following: composition transition edge, heat treatment transition edge, and state completion edge; The state completion edge transforms the material state node from an incomplete state to a complete state by supplementing it with direct performance measurement data.

5. The aluminum-based material design method according to claim 1, characterized in that, The state transition constraint set Ω includes at least two of the following: composition constraints, heat treatment constraints, and physical constraints. The heat treatment constraints include the requirement that the state transition direction must conform to the heat treatment process evolution direction; The physical constraints are based on the material state evolution consistency function. It is confirmed that monotonous performance is not required.

6. The aluminum-based material design method according to claim 1, characterized in that, The reachable closure To start from the initial set of available nodes Starting from the state transition edge that satisfies the state transition constraint set Ω, we find the set of all material state nodes reachable from it.

7. The aluminum-based material design method according to claim 1, characterized in that, The condition for determining the existence of a reachable state is: .

8. The aluminum-based material design method according to claim 1, characterized in that, The condition for determining the existence of a potential compensation state is as follows: Furthermore, there exists a reachable node that meets the target requirements in at least one performance dimension, and can be transformed into a usable state through state-complete edges.

9. The aluminum-based material design method according to claim 1, characterized in that, The criteria for determining the truly unreachable state are: Furthermore, there are no material state nodes that can be transformed into usable states through state completion edges; the determination result of the true unreachable state also includes the unreachability proof and the minimum relaxation direction.

10. An aluminum-based material design system employing the method described in any one of claims 1 to 9, characterized in that, include: The high-dimensional state space construction module is used to construct the high-dimensional material state space X; The reachable network construction module, connected to the high-dimensional state space construction module, is used to receive the high-dimensional material state space output by the high-dimensional state space construction module and construct a physically constrained material state reachable network. ; The projection mapping construction module, connected to the high-dimensional state space construction module and the reachable network construction module, is used to establish a performance projection mapping. ; The input module is used to receive user input regarding the target performance area. ; The reachability determination module, connected to the reachable network construction module, the projection mapping construction module, and the input module, is used to perform state space reachability determination, including generating reachable closures. Constructing a high-dimensional enhanced set And output the determination result based on the intersection of the reachable closure and the high-dimensional lifting set; An output module, connected to the reachability determination module, is used to receive and output the determination result output by the reachability determination module.

Images