Cement-based composite material design method and device for guiding machine learning based on knowledge graph, electronic equipment and storage medium

Abstract

The invention is suitable for the technical field of machine learning, and provides a cement-based composite material design method and device for guiding machine learning based on a knowledge graph, electronic equipment and a storage medium, and the method comprises the steps: obtaining the knowledge graph of a cement-based composite material, the knowledge graph comprises a plurality of nodes and directed edges connected with different nodes, the nodes comprise design nodes and attribute nodes; searching an attribute node corresponding to a design target of the cement-based composite material in the knowledge graph as a target attribute node; selecting a target design node from the candidate design nodes connected with the target attribute node; taking the target design node as a variable, and constructing a cement-based composite material design model based on machine learning; and calculating a cement-based composite material design scheme by using the cement-based composite material design model.

Description

Technical Field

[0001] The present application belongs to the field of machine learning, and in particular, relates to a cement-based composite material design method and device, electronic device, and computer-readable storage medium based on knowledge graph-guided machine learning. Background Art

[0002] Traditional methods for designing cement-based composites usually rely on labor-intensive laboratory experiments, resulting in inefficiencies in time and cost. In recent years, the efficiency of designing cement-based composites using machine learning methods has been high, but cement-based composite design models based on machine learning often rely only on experimental data training, which may lead to violations of scientific principles and limited reasoning ability of the model, thus affecting the design effect of cement-based composites. Summary of the invention

[0003] The embodiments of the present application provide a cement-based composite material design method and device, electronic device and computer-readable storage medium based on knowledge graph-guided machine learning, which can solve the problem in the related art that the reasoning ability of the cement-based composite material design model based on machine learning is limited, affecting the design effect of the cement-based composite material.

[0004] In a first aspect, an embodiment of the present application provides a cement-based composite material design method based on knowledge graph-guided machine learning, the method comprising: obtaining a knowledge graph of cement-based composite materials, the knowledge graph comprising multiple nodes and directed edges connecting different nodes, the nodes comprising design nodes and attribute nodes; searching the knowledge graph for attribute nodes corresponding to the design goals of cement-based composite materials as target attribute nodes; selecting a target design node from candidate design nodes connected to the target attribute nodes; using the target design node as a variable to construct a cement-based composite material design model based on machine learning; and using the cement-based composite material design model to calculate a cement-based composite material design solution.

[0005] In the second aspect, an embodiment of the present application provides a cement-based composite material design device based on knowledge graph-guided machine learning, the device comprising: an acquisition module, used to acquire a knowledge graph of cement-based composite materials, the knowledge graph comprising multiple nodes and directed edges connecting different nodes, the nodes comprising design nodes and attribute nodes; a search module, used to search the knowledge graph for attribute nodes corresponding to the design goals of cement-based composite materials as target attribute nodes; a selection module, used to select the target design node from the candidate design nodes connected to the target attribute node; a construction module, used to use the target design node as a variable to construct a cement-based composite material design model based on machine learning; and a calculation module, used to use the cement-based composite material design model to calculate a cement-based composite material design scheme.

[0006] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein when the processor executes the computer program, the cement-based composite material design method described in the first aspect above is implemented.

[0007] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the cement-based composite material design method described in the first aspect above.

[0008] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes the cement-based composite material design method described in the first aspect above.

[0009] Compared with the prior art, the embodiments of the present application have the following beneficial effects: by obtaining a knowledge graph of cement-based composite materials, the knowledge graph includes multiple nodes and directed edges connecting different nodes, and the nodes include design nodes and attribute nodes; searching the knowledge graph for attribute nodes corresponding to the design target of cement-based composite materials as target attribute nodes; selecting target design nodes from candidate design nodes connected to the target attribute nodes; using the target design nodes as variables to construct a cement-based composite material design model based on machine learning; and using the cement-based composite material design model to calculate the cement-based composite material design solution. The knowledge graph is used to guide the construction of the cement-based composite material design model and assist in the selection of variables in the model, thereby improving the reasoning ability of the model and further improving the design results of the cement-based composite material. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0011] Figure 1 is a structural schematic diagram of an electronic device provided by an embodiment of the present application;

[0012] Figure 2 It is a flow chart of a cement-based composite material design method based on knowledge graph-guided machine learning provided in one embodiment of the present application;

[0013] Figure 3 is a partial schematic diagram of a knowledge graph of cement-based composite materials in a specific example of the present application;

[0014] Figure 4It is a structural schematic diagram of a cement-based composite material design device based on knowledge graph-guided machine learning provided in one embodiment of the present application. DETAILED DESCRIPTION

[0015] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0016] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0017] It should also be understood that the term “and / or” used in the specification and appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0018] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0019] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0020] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0021] The cement-based composite material design method based on knowledge graph-guided machine learning provided in the embodiment of the present application can be applied to electronic devices, including but not limited to servers, server clusters, mobile phones, tablet computers, laptops, desktop computers, personal digital assistants, wearable devices and other electronic devices with computing functions. The embodiment of the present application does not impose any restrictions on the specific type of electronic devices.

[0022] Figure 1 FIG. 1 is a block diagram of a partial structure of an electronic device provided in an embodiment of the present application. Figure 1 , the electronic device includes: a processor 10, a memory 20, a bus 30, an input device 40, an output device 50 and a communication device 60. The processor 10 and the memory 20 are connected to each other through the bus 30, and the input device 40, the output device 50 and the communication device 60 are also connected to the bus 30. Those skilled in the art can understand that Figure 1 The structure of the electronic device shown in the figure does not constitute a limitation of the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0023] Combine the following Figure 1 A detailed introduction to the various components of electronic equipment:

[0024] The processor 10 is the control center of the electronic device, and can run the program stored in the memory 20 to perform various functions and process data. The processor 10 can be a central processing unit (CPU), and the processor 10 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In some embodiments, the processor 10 may include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0025] The memory 20 is used to store an operating system, an application program, a boot loader, data, and other programs, such as program codes of computer programs. The memory 20 can also be used to temporarily store data required for executing the program and generated. The memory 20 may include a high-speed random access memory, and may also include a non-volatile memory, such as a flash memory, a hard disk, a multimedia card, a card-type memory, etc. The memory 20 may include a storage unit disposed inside the electronic device, such as a hard disk of the electronic device, and / or a removable external storage unit, such as a mobile hard disk, a USB flash drive, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, etc.

[0026] The input device 40 may include at least one of a keyboard, a mouse, a touch panel, a joystick, etc., and is used to collect user input operations to generate corresponding operation instructions.

[0027] The output device 50 is used to output information to be provided to the user. The output device 50 generally includes a display, and optionally, a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc. can be used. In addition, the output device can further include a speaker.

[0028] The communication device 60 may include a modem, a network card, etc., for establishing a network connection with other electronic devices and communicating with each other.

[0029] The cement-based composite material design method based on knowledge graph-guided machine learning provided in the embodiment of the present application can be implemented as a computer software program. For example, an embodiment of the present application provides a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 60, and / or installed from a removable external storage unit. When the computer program is executed by the processor 10, the various functions defined in the cement-based composite material design method based on knowledge graph-guided machine learning provided in the embodiment of the present application are implemented.

[0030] Figure 2 A schematic flowchart of a cement-based composite material design method based on knowledge graph-guided machine learning provided in one embodiment of the present application is shown. As an example but not a limitation, the method can be applied to the above-mentioned electronic device.

[0031] S1: Construct a knowledge graph based on the public knowledge of cement-based composites.

[0032] Knowledge graph is a semantic network used to describe the relationship between entities. It is a storage and representation form of structured knowledge. It transforms the information in the real world into a graphical knowledge base, using nodes to represent entities and edges to represent the relationship between entities, thereby revealing the dynamic development laws of the knowledge field and providing practical and valuable references for subject research.

[0033] Specifically, relevant information can be extracted from public knowledge that conforms to objective laws and is provided by public publications such as books, papers, and engineering reports to construct a knowledge graph. The same method can be applied to expand the knowledge graph on the basis of the original one. The specific construction method is not limited here.

[0034] Cement-based composite materials may include paste, mortar, concrete, etc. For ease of description, concrete is taken as an example below.

[0035] In an embodiment of the present application, the knowledge graph is stored and applied in the form of a directed graph, specifically including multiple nodes and directed edges connecting different nodes. The starting point of a directed edge is called its starting node, and the end point of the directed edge is called its terminating node. Nodes include design nodes and attribute nodes. Design nodes are used to represent entities that need to be determined in the design process of cement-based composite materials, which may specifically include raw materials and design parameters. Raw materials refer to raw materials for preparing cement-based composite materials, such as cement, water, aggregates, admixtures, admixtures, etc.; design parameters refer to other parameters in the design of cement-based composite materials other than raw materials, such as the ratio of raw materials, production process, etc., which may specifically include mix ratio, curing method, ambient temperature and humidity, etc. Attribute nodes are used to represent the performance indicators of the prepared cement-based composite materials, such as rheology, work, mechanics, durability and other indicators.

[0036] For the design node of raw materials, multiple candidate materials for the node can be stored in the form of a database, and each candidate material has multiple attributes. These attributes can also be added to the knowledge graph as intermediate nodes, so that multiple levels are formed in the knowledge graph. An example of a local knowledge graph of concrete constructed in this way is as follows: Figure 3 shown.

[0037] For example, for a cement node, the attributes of each candidate cement may include but are not limited to: cement name, cement number, cement composition, cement manufacturer, cement strength, cement performance (such as frost resistance, impermeability, resistance to chloride ion corrosion, etc.), cement density, cement hydration heat, cement setting time (initial setting and final setting time), cement particle size. For a water node, the attributes of each candidate water may include but are not limited to: water name, water number, water composition, water source (such as river, sea or tap water, etc.). For an aggregate node, the attributes of each candidate aggregate may include but are not limited to: aggregate name, aggregate number, aggregate composition, aggregate particle size, aggregate shape (which can be quantitatively characterized by sphericity and angularity), aggregate density, aggregate water absorption, aggregate compressive strength, aggregate mud content. For an admixture node, the attributes of each candidate admixture may include but are not limited to: admixture name, admixture number, admixture composition, admixture function, and admixture manufacturer. For the admixture node, the attributes of each candidate admixture may include but are not limited to: admixture name, admixture number, admixture composition, admixture function, admixture manufacturer, admixture particle size, and admixture microscopic morphology.

[0038] There is a large amount of experimental data in public knowledge, and each piece of experimental data generally includes the raw materials, design parameters and performance of cement-based composite materials. Since a single piece of experimental data is generally part of an experiment designed for a specific field / goal / scenario, the entities involved may not necessarily cover all the nodes in the knowledge graph. To this end, you can try to fill in the uncovered nodes by extracting information from the context of the experimental data and / or based on common sense. For nodes that cannot be filled, they can be recorded as unknown. The experimental data can be stored in the nodes involved in the knowledge graph, or it can be stored in a database independent of the knowledge graph.

[0039] S2: Assign values ​​to the directed edges in the knowledge graph.

[0040] In the knowledge graph provided in the embodiment of the present application, a directed edge from a design node to an attribute node means that the design node has an influence on the attribute node, and the value of the edge indicates the magnitude of the influence. To facilitate subsequent quantitative calculations, it is generally normalized to a preset interval, such as [-1,1]. Positive values ​​indicate positive correlations, and negative values ​​indicate negative correlations. Directed edges between different design nodes usually indicate an inclusion relationship. For example, a certain raw material may include multiple components, and the influence of the raw material on the design node may be the influence of some of the components. To enrich the knowledge graph, these components can be added to the graph as intermediate nodes. In this case, the edge from the raw material to the component indicates inclusion, and the edge from the component to the attribute node indicates the existence of an influence.

[0041] For directed edges between different design nodes, their values ​​can be the content of the terminating node in the starting node; for directed edges from design nodes to attribute nodes, their values ​​represent the magnitude of the impact, and the value of the directed edge can be determined by relevant experimental data (i.e., experimental data involving the starting node and terminating node of the directed edge).

[0042] For example, the correlation between the starting node and the ending node of a directed edge can be calculated using relevant experimental data in public knowledge as the value of the directed edge. For example, the knowledge graph contains cement content, water-cement ratio, coarse aggregate content, sand content, and compressive strength. To calculate the size of the edge of cement content to compressive strength, the seaborn library in python can be used to draw a heat map of the correlation between database variables and obtain the correlation coefficient between the nodes "cement content" and "compressive strength" (for example, +0.6). The size of this edge is +0.6. The same method is used to calculate the edge "water-cement ratio-compressive strength" (-0.7), the edge "coarse aggregate content-compressive strength" (+0.3), and the edge "sand content-compressive strength" (+0.2).

[0043] Alternatively, the influence of the starting nodes of all directed edges connecting the same terminal node and the influence of the starting nodes on the terminal nodes can be determined based on public knowledge. Specifically, relevant experimental data in public knowledge can be used to train decision tree-based models, such as decision trees, random forests, etc. Then, the influence can be determined based on the level of the nodes in the decision tree-based model. Specifically, the root node has the strongest influence, and the farther the level from the root node, the weaker the influence. The influence of leaf nodes at the same level can be considered to be at the same level.

[0044] When the influence of different starting nodes can be divided into multiple levels, the directed edges can be assigned values ​​according to the levels. For example, the three design nodes ABC have a positive influence on the attribute node D, and the influence of A is significantly stronger than that of B and C. In other words, the influence can be divided into two levels, and [0,1] can be divided into two intervals [0.5,1] and [0,0.5]. The midpoint of the interval is taken to assign values ​​to the edges, that is, the size of the edge of AD is 0.75, and the size of the edge of BD and the edge of CD is 0.25.

[0045] When the influence of different starting nodes cannot be divided into multiple levels, the directed edges are assigned uniform values, that is, the edge size can be directly set to a fixed value, such as 0.5. If there is a conflict with the subinterval in the middle of the odd level, you can consider setting it to other values.

[0046] S3: Obtain the knowledge graph of cement-based composites.

[0047] The aforementioned constructed knowledge graph can be obtained. S1-S2 and S3-S7 can be executed by the same electronic device or by different electronic devices.

[0048] S4: Search the knowledge graph for the attribute node corresponding to the design goal of cement-based composite materials as the target attribute node.

[0049] S5: Select a target design node from candidate design nodes connected to the target attribute node.

[0050] The knowledge weight of the candidate design node can be determined according to the size of the edge between the candidate design node and the target attribute node. Specifically, if there is only one edge between the candidate design node and the target attribute node, that is, there is no intermediate node, the knowledge weight is the absolute value of the edge; if there are multiple edges between the candidate design node and the target attribute node, that is, there is an intermediate node, the knowledge weight is the absolute value of the product of all edges between the candidate design node and the target attribute node. Then, a specified number or a specified proportion of candidate design nodes with the largest knowledge weight can be selected as the target design nodes.

[0051] Different candidate design nodes may be associated. In the subsequent model training and use process, selecting two candidate design nodes that are too closely associated will cause one of the candidate design nodes to be wasted, thereby adversely affecting the effect of the model. To solve this problem, optionally, before selecting the target design node, the similarity between the two candidate design nodes can be calculated. Specifically, the nodes can be vectorized and then projected onto the hyperplane of the relationship, and the transH model can be trained by using the relationship between existing entities. The trained transH model can then be used to calculate the Euclidean distance between the projections of the two candidate design nodes on the hyperplane of the relationship. The shorter the distance, the higher the similarity. When the similarity is greater than the first threshold, a candidate design node is removed, and generally the candidate design node with a smaller knowledge weight is removed.

[0052] For example, the transH method can be used to calculate the similarity between two candidate design nodes. The similarity can be represented by cosine similarity, which ranges from [-1,1]. The larger the value, the more similar it is. When the cosine similarity is 1, it means that the two vectors are exactly the same, when it is 0, it means orthogonal (no similarity), and when it is -1, it means it is completely opposite. If the similarity between two nodes is greater than or equal to 0.8, then remove the candidate design node with a smaller knowledge weight and retain the candidate design node with a larger knowledge weight.

[0053] S6: Taking the target design nodes as variables, a cement-based composite design model is constructed based on machine learning.

[0054] The initial weights of the variables in the cement-based composite material design model can be determined according to the knowledge weights of the target design nodes. Specifically, the knowledge weights of all target design nodes can be normalized to obtain the initial weights, that is, the knowledge weights of all target design nodes are geometrically scaled to satisfy the sum of the knowledge weights of all target design nodes to be 1.

[0055] The specific architecture and training process of the cement-based composite material design model are not limited here, and the architecture can be random forest, LightGBM, XGBoost, CatBoost, etc. There can be more than one cement-based composite material design model. After the training is completed, the performance index of the model is calculated to select the cement-based composite material design model with the best performance index for subsequent calculations.

[0056] S7: Use the cementitious composite design model to calculate cementitious composite design options.

[0057] Substituting the design objectives and constraints of cement-based composite materials into the cement-based composite material design model, optimizing the calculation to obtain the value of each variable, the design scheme of cement-based composite materials can be obtained.

[0058] Through the implementation of this embodiment, a knowledge graph of cement-based composite materials is obtained, the knowledge graph includes multiple nodes and directed edges connecting different nodes, and the nodes include design nodes and attribute nodes; the attribute node corresponding to the design target of the cement-based composite material is searched in the knowledge graph as the target attribute node; the target design node is selected from the candidate design nodes connected to the target attribute node; the target design node is used as a variable to build a cement-based composite material design model based on machine learning; and the cement-based composite material design model is used to calculate the cement-based composite material design scheme. The knowledge graph is used to guide the construction of the cement-based composite material design model and assist in the selection of variables in the model, thereby improving the reasoning ability of the model and further improving the design results of the cement-based composite material.

[0059] Figure 4 A schematic diagram of the structure of a cement-based composite material design device based on knowledge graph-guided machine learning provided in one embodiment of the present application is shown. The device includes an acquisition module 11, a search module 12, a selection module 13, a construction module 14 and a calculation module 15.

[0060] The acquisition module 11 is used to obtain the knowledge graph of cement-based composite materials. The knowledge graph includes multiple nodes and directed edges connecting different nodes. The nodes include design nodes and attribute nodes.

[0061] The search module 12 is used to search the knowledge graph for an attribute node corresponding to the design target of the cement-based composite material as a target attribute node.

[0062] The selection module 13 is used to select a target design node from candidate design nodes connected to the target attribute node.

[0063] The construction module 14 is used to construct a cement-based composite material design model based on machine learning by taking the target design node as a variable.

[0064] The calculation module 15 is used to calculate the cement-based composite material design solution using the cement-based composite material design model.

[0065] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / modules / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0066] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0067] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0068] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0069] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the camera / electronic device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0070] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0071] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0072] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0073] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0074] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A cement-based composite material design method based on knowledge graph-guided machine learning, characterized in that: The method comprises: Acquire a knowledge graph of cement-based composite materials, wherein the knowledge graph includes a plurality of nodes and directed edges connecting different nodes, wherein the nodes include design nodes and attribute nodes; Searching the knowledge graph for an attribute node corresponding to a design target of the cement-based composite material as a target attribute node; Selecting a target design node from candidate design nodes connected to the target attribute node; Using the target design node as a variable, constructing a cement-based composite material design model based on machine learning; The cement-based composite material design model is used to calculate the cement-based composite material design solution.

2. The method according to claim 1, characterized in that The selecting a target design node from candidate design nodes connected to the target attribute node comprises: Determine the knowledge weight of the candidate design node according to the size of the edge between the candidate design node and the target attribute node; A specified number or a specified proportion of the candidate design nodes having the largest knowledge weights are selected as the target design nodes.

3. The method according to claim 2, characterized in that Before selecting a target design node from candidate design nodes connected to the target attribute node, the method includes: Calculating the similarity between two candidate design nodes; When the similarity is greater than a first threshold, one of the candidate design nodes is removed.

4. The method according to claim 2, characterized in that The method of using the target design node as a variable and constructing a cement-based composite material design model based on machine learning includes: The initial weights of the variables in the cement-based composite material design model are determined according to the knowledge weights of the target design nodes.

5. The method according to claim 1, characterized in that The step of obtaining the knowledge graph of cement-based composite materials also includes: Constructing the knowledge graph according to public knowledge of cement-based composite materials; Assign values ​​to the directed edges in the knowledge graph.

6. The method according to claim 5, characterized in that The assigning of values ​​to the edges in the knowledge graph includes: The correlation between the start node and the end node of the directed edge is calculated using the relevant experimental data in the public knowledge as the value of the directed edge.

7. The method according to claim 5, characterized in that The assigning of values ​​to the edges in the knowledge graph includes: Determine the starting nodes of all the directed edges connected to the same terminating node and the strength of the influence of the starting nodes on the terminating node according to the public knowledge; When the influence strength of different starting nodes can be divided into multiple levels, assigning values ​​to the directed edges according to the levels; When the influence strengths of different starting nodes cannot be divided into multiple levels, the directed edges are uniformly assigned values.

8. The method according to claim 7, characterized in that The determining, based on the public knowledge, of the starting nodes of all the directed edges connected to the same terminating node and the strength of the influence of the starting nodes on the terminating node comprises: Training a decision tree-based model using relevant experimental data from the public knowledge; The strength of the influence is determined according to the level of the nodes in the decision tree-based model.

9. A cement-based composite material design device based on knowledge graph-guided machine learning, characterized in that: The device comprises: An acquisition module, used for acquiring a knowledge graph of cement-based composite materials, wherein the knowledge graph includes a plurality of nodes and directed edges connecting different nodes, wherein the nodes include design nodes and attribute nodes; A search module, used to search the knowledge graph for an attribute node corresponding to a design target of cement-based composite materials as a target attribute node; A selection module, used for selecting a target design node from candidate design nodes connected to the target attribute node; A construction module is used to construct a cement-based composite material design model based on machine learning by taking the target design node as a variable; The calculation module is used to calculate the cement-based composite material design solution using the cement-based composite material design model.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

11. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images