Method and apparatus for generating and utilizing a knowledge graph of a manufacturing simulation model

Abstract

A method for generating and utilizing a knowledge graph of a manufacturing simulation model, comprising: obtaining a manufacturing simulation model in a factory and generating model representation data based on the manufacturing simulation model; obtaining preset graph semantics for the manufacturing simulation model, where the graph semantics provide a semantic description of the knowledge graph of the manufacturing simulation model; and extracting data corresponding to the graph semantics from the model representation data as graph data of the knowledge graph to form a knowledge graph. Saving and managing the knowledge of the manufacturing simulation model in a unified format helps with the digitalization of discrete manufacturing factories. The knowledge graph also provides a unified way for machines and humans to acquire knowledge, improving the efficiency of knowledge acquisition. Converting manufacturing simulation models in different formats into model representation data in a unified format and then into graph data of the knowledge graph enables the easy conversion between manufacturing simulation models and knowledge graphs.

Description

Technical Field

[0001] The present disclosure relates to the technical field of industrial manufacturing, and more particularly, to methods, apparatuses, computing devices, computer-readable storage media, and program products for generating and utilizing knowledge graphs of manufacturing simulation models. Background Art

[0002] With the widespread use of simulation technology in industrial digitization, simulation models play a very important role in industry. For discrete manufacturing plants, simulation models can simulate the warehousing, transportation, and production processes in the plant, thereby assisting in the management of material storage and transportation, identifying production bottlenecks, and improving production capacity.

[0003] The work of establishing or maintaining a manufacturing simulation model (such as changing the model structure or parameters) is usually done by simulation engineers who are familiar with simulation software and model formats. When other plant personnel who are not familiar with simulation software want to obtain the simulation results of a certain manufacturing simulation model, they need to collaborate with simulation engineers. Summary of the Invention

[0004] Different manufacturing simulation models usually have different formats or are established via different simulation software, so it is difficult to manage them uniformly. In addition, machines (such as AI programs) and other plant personnel who are not familiar with simulation software cannot directly obtain the required model knowledge from manufacturing simulation models. Moreover, the development and maintenance of models can only be done by simulation engineers, which not only consumes a large amount of time and energy of simulation engineers, but also makes it impossible for other plant personnel to modify the simulation model in a timely manner and obtain simulation results, thereby reducing the work efficiency of some work that depends on simulation results.

[0005] A first embodiment of the present disclosure proposes a method for generating and utilizing a knowledge graph of a manufacturing simulation model, including: obtaining a manufacturing simulation model in a factory and generating model representation data based on the manufacturing simulation model; obtaining a preset graph semantics for the manufacturing simulation model, where the graph semantics provides a semantic description of the knowledge graph of the manufacturing simulation model; and extracting data corresponding to the graph semantics from the model representation data as graph data of the knowledge graph to form a knowledge graph.

[0006] In this embodiment, by means of the preset graph semantics, the manufacturing simulation model is converted into a knowledge graph, which can save and manage the knowledge of the manufacturing simulation model in a unified format, and is helpful for the digitization of discrete manufacturing plants. The knowledge graph of the manufacturing simulation model also provides a unified knowledge acquisition method for machines and humans, improving the efficiency of knowledge acquisition. In addition, by converting manufacturing simulation models in different formats into model representation data in a unified format and then into graph data of the knowledge graph, it is easy to realize the conversion between manufacturing simulation models and knowledge graphs.

[0007] The second embodiment of the present disclosure proposes an apparatus for generating and utilizing a knowledge graph of a manufacturing simulation model, including: a model data generation unit configured to obtain a manufacturing simulation model in a factory and generate model representation data based on the manufacturing simulation model; a graph semantic acquisition unit configured to obtain preset graph semantics for the manufacturing simulation model, where the graph semantics provide a semantic description of the knowledge graph of the manufacturing simulation model; and a graph data extraction unit configured to extract data corresponding to the graph semantics from the model representation data as the graph data of the knowledge graph to form the knowledge graph.

[0008] The third embodiment of the present disclosure proposes a computing device, which includes: a processor; and a memory for storing computer-executable instructions, which, when executed, cause the processor to execute the method in the first embodiment.

[0009] The fourth embodiment of the present disclosure proposes a computer-readable storage medium having computer-executable instructions stored thereon, and the computer-executable instructions are used to execute the method in the first embodiment.

[0010] The fifth embodiment of the present disclosure proposes a computer program product, which is tangibly stored on a computer-readable storage medium and includes computer-executable instructions, and the computer-executable instructions, when executed, cause at least one processor to execute the method in the first embodiment. Description of the Drawings

[0011] In combination with the accompanying drawings and with reference to the following detailed description, the features, advantages and other aspects of the embodiments of the present disclosure will become more apparent. Several embodiments of the present disclosure are shown herein in an exemplary rather than restrictive manner. In the drawings:

[0012] Figure 1 A method for generating and utilizing a knowledge graph of a manufacturing simulation model according to some embodiments of the present disclosure is shown;

[0013] Figure 2 A system architecture diagram for implementing the method in Figure 1 is shown;

[0014] Figure 3 Shown in Figure 2 is a flowchart of a method for generating a knowledge graph of a production line simulation model in the embodiment of

[0015] Figure 4 Shown in Figure 2 is a flowchart of a method for improving a production line simulation model in the embodiment of

[0016] Figure 5 shows a method flowchart for querying a knowledge graph of a production line simulation model in an Figure 2 embodiment;

[0017] Figure 6 shows a method flowchart for modifying a manufacturing simulation model by using a knowledge graph of a production line simulation model in an Figure 2 embodiment;

[0018] Figure 7 shows a schematic diagram of a production line simulation model in an Figure 2 embodiment;

[0019] Figure 8 shows a schematic diagram of a knowledge graph of a production line simulation model in Figure 7 ;

[0020] Figure 9 shows an apparatus for generating and using a knowledge graph of a manufacturing simulation model according to an embodiment of the present disclosure; and

[0021] Figure 10 shows a block diagram of a computing device for generating and using a knowledge graph of a manufacturing simulation model according to an embodiment of the present disclosure. Detailed Description of Specific Embodiments

[0022] The following describes in detail various exemplary embodiments of the present disclosure with reference to the accompanying drawings. Although the exemplary methods and apparatuses described below include software and / or firmware executed on hardware among other components, it should be noted that these examples are merely illustrative and should not be considered restrictive. For example, it is contemplated that any or all of the hardware, software, and firmware components may be implemented exclusively in hardware, exclusively in software, or in any combination of hardware and software. Thus, although the exemplary methods and apparatuses have been described below, those skilled in the art will readily appreciate that the examples provided are not intended to limit the manner in which these methods and apparatuses may be implemented.

[0023] In addition, the flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of methods and systems according to various embodiments of the present disclosure. It should be noted that the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the flowchart and / or block diagram, and combinations of blocks in the flowchart and / or block diagram, may be implemented using a dedicated hardware-based system that performs the specified functions or operations, or may be implemented using a combination of dedicated hardware and computer instructions.

[0024] As used herein, the terms "comprising", "including" and similar terms are open-ended terms, i.e., "including / including but not limited to", indicating that other content may also be included. 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" and so on.

[0025] Figure 1 A method for generating and utilizing a knowledge graph of a manufacturing simulation model according to some embodiments of the present disclosure is shown. Referring to Figure 1 , method 100 starts at step 101. In step 101, a manufacturing simulation model in a factory is obtained and model representation data is generated based on the manufacturing simulation model. The manufacturing simulation model can be any type of simulation model involved in the production and manufacturing activities of the factory, including but not limited to production line models, warehousing models, in-plant logistics models, etc. These manufacturing simulation models can be established by different simulation software and have different formats, and they can be stored at any storage device (such as a server) within the factory. The manufacturing simulation model is used to simulate the corresponding actual manufacturing process. For example, the production line model is used to simulate the production process of the actual production line, the warehousing model is used to simulate the processes of inbound, storage, and outbound of goods (such as materials, workpieces, etc.), and the in-plant logistics model is used to simulate the transportation process of goods.

[0026] The internal data of the manufacturing simulation model generally includes the data used in the corresponding actual manufacturing process, including but not limited to the types, names, attribute values of the various components of the manufacturing simulation model, the connection order between the various components, and so on. For example, for a production line simulation model, its internal data includes the types and names of the various components of the actual production line, such as production workstations (such as assembly stations, test stations, packaging stations, processing stations, etc.), transfer workstations (such as conveyor belts, AGVs, robots, manual labor, etc.), buffer areas, supply stations, recycling stations, etc., their attribute values (such as spatial location, preparation time, processing time, processing capacity, additional tools or materials, personnel requirements, etc.), and the connection order between them. The internal data of these production line simulation models is converted into model representation data in a low-level format. The model representation data also includes the content of the above internal data. The low-level format can be, for example, JSON format or XML format.

[0027] In step 102, the preset graph semantics for manufacturing the simulation model are obtained. The graph semantics provide a semantic description of the knowledge graph of the manufacturing simulation model. A unified graph database can be established for one or more factories to centralize all the knowledge within the factory. Common graph semantics are defined for this graph database. For each different domain, the graph semantics provide a semantic description of the knowledge graph of that domain. The graph semantics include all entity categories, sub-categories, and attributes related to that domain, and the relationship categories and attributes between entities. A domain can be a collection of several similar objects. For example, a collection of multiple similar production lines can be called a domain. Entities serve as the vertices of the knowledge graph, and the relationships between entities serve as the edges connecting the vertices. Entities are associated through relationships to form the hierarchical structure of the knowledge graph. It should be noted that for different domains, the entity categories, sub-categories, and attributes, as well as the relationship categories and attributes between entities, may all be different.

[0028] When generating the knowledge graph of the manufacturing simulation model, it is necessary to first obtain the preset graph semantics for this manufacturing simulation model. For the production line simulation model, the entity categories in the knowledge graph can include production workstations, conveyor workstations, buffer areas, supply stations, recycling stations, etc. in the production line. The attributes of entities can vary according to different entity categories. For example, the attributes of a production workstation can include name, spatial location, preparation time, and processing time; the attributes of a conveyor workstation can include name, spatial location, and conveyor speed; the attributes of a buffer area can include name, spatial location, and storage capacity; the attributes of a supply station can include name, spatial location, and material supply interval time; the attributes of a recycling station can include name, spatial location, and processing time. The relationship categories between entities can include the connection relationships between entities, such as the previous entity and the subsequent entity in the connection. At the same time, the connection relationship can represent the production order of the products to be produced.

[0029] In step 103, data corresponding to the graph semantics is extracted from the model representation data as the graph data of the knowledge graph to form the knowledge graph. As described above, the graph semantics include all entity categories, sub-categories and attributes related to the domain, and the relationship categories and attributes between entities. Before forming the knowledge graph, corresponding data can be extracted from the model representation data according to the preset graph semantics used for manufacturing the simulation model to generate the graph data. The graph semantics and the model representation data have different definitions for data categories. That is to say, for an entity or relationship of a certain category in the graph semantics, it is identified by another category in the model representation data. Therefore, the mapping relationship between the entity categories and relationship categories of the graph semantics and the data categories in the model representation data can be established in advance, and then the corresponding data can be extracted from the model representation data according to the mapping relationship as the graph data for generating the knowledge graph. By establishing the mapping relationship between the entity categories and relationship categories of the graph semantics and the data categories in the model representation data, it is possible to easily search and extract the graph data for forming the knowledge graph from the model representation data. The generated graph data can be used to form and save the knowledge graph through any existing graph database.

[0030] In some embodiments, in addition to the knowledge graph of the manufacturing simulation model, the graph database can also save other knowledge graphs related to one or more factories, such as the knowledge graph of production records, the knowledge graph related to production line maintenance, and so on. By establishing a factory-level or enterprise-level graph database, the scattered data from different sources in the entire factory or enterprise can be centralized and uniformly saved in the form of a knowledge graph, so that it is easier to achieve data acquisition and utilization.

[0031] In some embodiments, method 100 further includes ( Figure 1(not shown in the figure): Obtain an updated knowledge graph for the manufacturing simulation model; read the graph data of the updated knowledge graph and generate updated model representation data based on the read graph data; and update the manufacturing simulation model using the updated model representation data. In these embodiments, in addition to being able to generate the knowledge graph of the manufacturing simulation model, the manufacturing simulation model can also be updated according to the updated knowledge graph. When it is necessary to update the manufacturing simulation model, the graph data in the knowledge graph can be directly modified. For example, change the attribute values of the entities and the relationships between entities in the knowledge graph, add entities or relationships, delete entities or relationships, and so on. After updating the knowledge graph, read the graph data of the updated knowledge graph and generate updated model representation data. Similar to generating graph data from model representation data, the required data can be extracted from the updated graph data as the updated model representation data according to the mapping relationship between the entity categories and relationship categories of the pre-established graph semantics and the data categories in the model representation data. Then, update the manufacturing simulation model using the updated model representation data. The manufacturing simulation model can be regenerated according to the updated model representation data, or only the data that has changed can be modified in the manufacturing simulation model. By updating or maintaining the manufacturing simulation model at the level of the knowledge graph of the manufacturing simulation model, there is no need for specialized simulation engineers to update and maintain the model anymore, which not only saves the time and effort of the simulation engineers, but also improves the efficiency of model update and maintenance, enabling other factory personnel to conveniently modify the model and perform simulations as needed.

[0032] In some embodiments, updating the manufacturing simulation model using the updated model representation data further includes: comparing the updated model representation data with the previous model representation data; determining the difference data between the updated model representation data and the previous model representation data based on the comparison result; and modifying the manufacturing simulation model according to the difference data. The previous model representation data is the graph data before the knowledge graph is updated. The difference data between the updated model representation data and the previous model representation data can be entities, the relationships between entities, and / or the attribute values of entities or relationships. For example, for a production line simulation model, the difference data can be that one or more attribute values (such as setup time and processing time, etc.) of a production workstation have changed, a transfer workstation and its connection relationship with other workstations have been added, or the connection relationship between two production workstations has changed (which means that the processing procedure of the product to be produced has changed), and so on. After determining the difference data, the corresponding internal data in the manufacturing simulation model can be modified to update the manufacturing simulation model. Only modifying the internal data that has changed in the manufacturing simulation model can avoid potential errors that may occur when regenerating the manufacturing simulation model, increase the accuracy of model update, and also improve the model update speed without the need for cross-departmental collaboration.

[0033] In some embodiments, before obtaining the updated knowledge graph of the manufacturing simulation model, method 100 further includes ( Figure 1 not shown in the figure): judging whether there is model update data associated with the manufacturing simulation model according to the graph data of the knowledge graph; and when there is model update data, using the model update data to modify the graph data to form an updated knowledge graph. The entities and the relationships between the entities in the graph data can be read, and the model update data, that is, new entities, relationships, and / or their new attribute values, can be searched in other knowledge sources according to the entities and relationships. After the model update data is found, the corresponding entities, relationships, and / or their attribute values in the graph data can be modified to update the graph data for forming the updated knowledge graph. The model update data can be data stored in any storage device in the factory from other external data sources, or can be the graph data of other knowledge graphs stored in the same graph database as the knowledge graph of the manufacturing simulation model. The model update data can be any data that needs to be updated in the manufacturing simulation model. For example, it can be the actual manufacturing data corresponding to the manufacturing simulation model, the expected modification data in the manufacturing simulation model (such as adding, deleting, and / or modifying components, connection orders, and / or attribute values in the manufacturing simulation model), and so on.

[0034] In some embodiments, the model update data is actual manufacturing data, and using the model update data to modify the graph data further includes: judging whether the actual manufacturing data is the same as the corresponding data in the graph data; and when the actual manufacturing data is different from the corresponding data in the graph data, replacing the corresponding data in the graph data with the actual manufacturing data. The actual manufacturing data is the data generated during the actual manufacturing process corresponding to the manufacturing simulation model. For example, for a production line simulation model, the actual manufacturing data can be one or more actual attribute values of the production workstations on the production line (such as actual setup time, processing time, equipment failure rate, etc.). The actual manufacturing data can come from external data sources such as production records, for example. The actual manufacturing data can be automatically read and compared with the corresponding data in the graph data to judge whether they are the same. If the actual manufacturing data is different from the corresponding data in the graph data, the corresponding data in the graph data is replaced with the actual manufacturing data, that is, the entities, relationships, and / or their attribute values are replaced with the actual values. Automatically updating the manufacturing simulation model at the level of the knowledge graph using the actual manufacturing data can keep the manufacturing simulation model in the latest state, thereby improving the model update speed and enabling more accurate simulation of the actual manufacturing process.

[0035] In some embodiments, the model update data is expected modification data, and method 100 further includes ( Figure 1(not shown in the figure): performing simulation using an updated manufacturing simulation model; and determining whether to apply the expected modification data in the factory based on the simulation results. The expected modification data may be components, connection sequences, and / or attribute values added, deleted, and / or modified in the manufacturing simulation model. For example, for a production line simulation model, the expected modification data may be a production workstation added in the actual production line, a modified connection sequence, and / or one or more attribute values of the production workstation (such as setup time, processing time, equipment failure rate, etc.). The expected modification data may be automatically obtained, for example, by an AI program from an external data source such as the Internet or other documents. After updating the manufacturing simulation model by updating the knowledge graph, the updated manufacturing simulation model is used to perform simulation again, so as to obtain new simulation results of the manufacturing simulation model after applying the expected modification data. It can be determined whether to apply the expected modification data in the factory based on the new simulation results and some other conditions (such as the difficulty of applying the expected modification data in the factory, the number of products that may affect production, the time and economic cost required, etc.). For example, for a production line simulation model, if a certain production workstation in the expected production line is replaced with another production workstation with a shorter processing time, the expected modification data is the processing time of the other production workstation. After performing simulation on the updated manufacturing simulation model, it is determined based on the simulation results whether to replace the original production workstation with the other production workstation in the actual production line. By using the expected modification data to update the knowledge graph of the manufacturing simulation model and re-simulating the updated manufacturing simulation model, it is possible to provide a decision on updating equipment for the factory, thereby saving costs and improving the efficiency of the entire factory.

[0036] In some embodiments, method 100 further includes ( Figure 1(not shown in the figure): Obtain query conditions for the manufacturing simulation model; and read graph data matching the query conditions from the knowledge graph. After generating the knowledge graph of the manufacturing simulation model, any knowledge related to the manufacturing simulation model can also be queried from the knowledge graph. For example, for a production line simulation model, the number of production workstations in the production line, the number of process steps required to produce a product, whether the production of the product will be affected when a certain production workstation fails, etc. can be queried from its knowledge graph. When other knowledge graphs related to the factory (such as the knowledge graph related to products, the knowledge graph of production records, etc.) are also included in the graph database storing the knowledge graph of the manufacturing simulation model, more knowledge can also be queried using multiple knowledge graphs. For example, for a production line simulation model, the products and their quantities that will be affected during the maintenance period of a certain production workstation can be queried using the knowledge graph related to products and the knowledge graph of the production line simulation model, and the products and their quantities that have been affected during the failure period of a certain production workstation can be queried using the knowledge graph of production records and the knowledge graph of the production line simulation model, etc. Decompose the content to be queried into query conditions associated with the graph data in the knowledge graph, and then read the graph data matching the query conditions from the knowledge graph. For example, when querying the number of test stations in the production line from the knowledge graph of the production line simulation model, the query conditions can be the entity category (workstation) of the test station and the entity name (including the word "test"). By counting the queried entities, the number of test stations can be obtained. By generating the knowledge graph of the manufacturing simulation model, the model knowledge can be converted into a form that can be understood by machines, so that model-related knowledge can be queried at the level of the knowledge graph, and the value of the model can be better explored and utilized.

[0037] In some embodiments, method 100 can be executed by a server device communicating with a client device used by an engineer. A request is sent from the client device, and in response to the request, a knowledge graph of the manufacturing simulation model, querying the knowledge graph, or updating the manufacturing simulation model is generated at the server device. In other embodiments, method 100 can also be directly executed by the client device.

[0038] In the above embodiments, by means of the preset graph semantics, converting the manufacturing simulation model into a knowledge graph can save and manage the knowledge of the manufacturing simulation model in a unified format, which is helpful for the digitization of discrete manufacturing factories. The knowledge graph of the manufacturing simulation model also provides a unified way of knowledge acquisition for machines and humans, improving the efficiency of knowledge acquisition. In addition, converting manufacturing simulation models in different formats into model representation data in a unified format and then into graph data of the knowledge graph can easily achieve the conversion between the manufacturing simulation model and the knowledge graph.

[0039] The method for generating and utilizing a knowledge graph of a manufacturing simulation model according to the present disclosure will be described below with reference to a specific embodiment. In this embodiment, the manufacturing simulation model is a production line simulation model. Figure 2 shows a system architecture diagram for implementing the Figure 1 method in accordance with an embodiment of the present disclosure. Figure 3 shows a flowchart of a method for generating a knowledge graph of a production line simulation model in the Figure 2 embodiment. Figure 4 shows a flowchart of a method for improving a production line simulation model in the Figure 2 embodiment. Figure 5 shows a flowchart of a method for querying a knowledge graph of a production line simulation model in the Figure 2 embodiment. Figure 6 shows a flowchart of a method for modifying a production line simulation model by utilizing a knowledge graph of the production line simulation model in the Figure 2 embodiment. Refer to Figure 2 - Figure 6 for illustration of this embodiment.

[0040] First, refer to Figure 2 and Figure 3 . In method 300 of Figure 3 , step 301 includes defining common graph semantics 202 for graph database 203 by graph semantic definition unit 201. Graph database 203 is used to store all knowledge graphs of the entire factory or enterprise, for example, knowledge graphs of production records, knowledge graphs related to products, knowledge graphs of production line simulation models, and so on. Common graph semantics 202 provides semantic descriptions of the knowledge graph for the production line simulation model. Common graph semantics 202 includes entity categories, sub - categories and attributes, and relationship categories and attributes between entities. Tables 1 and 2 below show examples of a part of the graph semantics for the production line simulation model.

[0041] Table 1

[0042] Entity:

[0043]

[0044] Table 2

[0045] Relationship:

[0046] Category Attribute Description Connection Production Path

[0047] As shown in Table 1 and Table 2, the entity categories for the graph semantics of the production line simulation model include production workstations, supply stations, and recycling stations, and the relationship category includes connections. The attributes of a production workstation include name, x coordinate, y coordinate, preparation time, and processing time. The attributes of a supply station include name, x coordinate, y coordinate, and material supply interval time. The attributes of a recycling station include name, x coordinate, y coordinate, and processing time.

[0048] Next, in step 302, the model connector 205 obtains the production line simulation model 204 and generates model representation data 206 based on the production line simulation model 204. The model connector 205 can be implemented as a plug-in in the simulation software. Figure 7 A schematic diagram of a production line simulation model is shown. In the production line simulation model 700, between the supply station 701 and the recycling station 707, there are 5 production workstations in sequence according to the product processing order: assembly station 702, pre-test station 703, first test station 704, second test station 705, and packaging station 706. The first test station 704 and the second test station 705 are test stations for simultaneous testing, indicating that the product to be produced can pass through one of them. In this embodiment, the internal data is read from the production line simulation model 204 and converted into model representation data 206 in the JSON file format. In other embodiments, other formats can also be used to save the model representation data 206, such as the XML format. Table 3 below lists a partial example of the model representation data 206.

[0049] Table 3

[0050]

[0051] In step 303, the common graph semantics 202 for the production line simulation model 204 are obtained. As described above, the common graph semantics 202 provide a semantic description of the knowledge graph of the production line simulation model 204. Continuing with step 304, the graph connector 207 extracts the data corresponding to the graph semantics 202 from the model representation data 206 as the graph data of the knowledge graph 208. A mapping relationship is pre-established in the graph connector 207 between the entity categories of the graph semantics 202 and the data categories in the model representation data 206. For the exemplary production line simulation model 700, the entity categories "supply station" and "recycling station" defined in the graph semantics 202 respectively correspond to "supply station" and "recycling station" in the model representation data 206, the entity category "production workstation" corresponds to "single processing station" in the model representation data 206, and the relationship category "connection" corresponds to "connection" in the model representation data 206. The graph connector 207 extracts the corresponding data from the model representation data 206 as the graph data according to the mapping relationship. In this embodiment, TinkerPop is used as the graph database, and the graph data can be a script file for execution by TinkerPop. The graph database 203 forms and saves the knowledge graph 208 of the production line simulation model 204 using the graph data. Figure 8 FIG. shows a schematic diagram of the knowledge graph 800 of the production line simulation model 700. The knowledge graph 800 includes a supply station 801, an assembly station 802, a pre-test station 803, a first test station 804, a second test station 805, a packaging station 806, and a recycling station 807. The arrows between these workstations represent the connection relationships between them and also represent the production order of the products.

[0052] After generating the knowledge graph 208 of the production line simulation model 204, optionally, method 400 can also be executed to improve the production line simulation model 204. Referring to Figure 4 , in step 401, the model improvement module 209 determines whether there is actual manufacturing data related to the production line simulation model 204 according to the graph data of the knowledge graph 208, for example, the latest production records of the production line corresponding to the production line simulation model 204 stored in the graph database 203. The production records usually include the actual attribute values of each workstation, such as the actual preparation time, actual processing time, actual failure rate, etc. In step 402, when there is actual manufacturing data, the model improvement module 209 determines whether the actual manufacturing data is the same as the corresponding data in the graph data, for example, comparing the actual preparation time of a certain processing workstation with the corresponding preparation time in the knowledge graph 208. In step 403, when the actual manufacturing data is different from the corresponding data in the graph data, the model improvement module 209 replaces the corresponding data in the graph data with the actual manufacturing data, thereby updating the knowledge graph 208.

[0053] In step 404, the updated knowledge graph 208 of the production line simulation model 204 is obtained by the graph connector 207, the graph data is read, and the model representation data 206 is updated based on the graph data. Similar to generating graph data according to the model representation data 206, the required data is extracted from the updated graph data according to the mapping relationship between the entity categories of the graph semantics 202 and the data categories in the model representation data 206. Next, in step 405, the updated model representation data 206 is compared with the previous model representation data 206 by the model connector 205. In step 406, the model connector 205 determines the difference data between the updated model representation data 206 and the previous model representation data 206 based on the comparison result. For example, different attribute values of a certain production workstation. In step 407, the model connector 205 modifies the production line simulation model 204 according to the difference data, that is, modifies the internal data of the model.

[0054] In the system 200, the knowledge graph 208 can also be queried through the graph query module 210. Figure 5 The method flow chart for querying the knowledge graph of the production line simulation model is shown. Also refer to Figure 2 and Figure 5 . In Figure 5 's method 500, step 501 includes obtaining the query condition for the production line simulation model 204 by the graph query module 210. The query condition can be obtained by decomposing the content to be queried. Taking the production line simulation model 700 in Figure 7 as an example, the content to be queried may include the number of production workstations in the actual production line, the number of process steps required to produce products, and so on. The query condition varies according to the content to be queried. In step 502, the graph query module 210 reads the graph data that matches the query condition from the knowledge graph. After that, in step 503, the graph query module 210 post-processes the read graph data to generate a query result. The post-processing can be, for example, merging the read graph data, further processing the read graph data, and so on.

[0055] As Figure 2 shown, the system 200 also includes a model modification module 211. Figure 6 The method flow chart for modifying the production line simulation model using the knowledge graph of the production line simulation model is shown. Also refer to Figure 2 and Figure 6。In method 600, step 601 includes the model modification module 211 determining whether there is expected modification data related to the production line simulation model 204 according to the graph data of the knowledge graph 208. The expected modification data may be added production workstations in the production line, modified connection sequences, and / or one or more attribute values of the production workstations. The expected modification data can be obtained from external data sources such as the Internet or documents. In step 602, when there is expected modification data, the model modification module 211 updates the knowledge graph 208 with the expected modification data, that is, replaces the corresponding data in the knowledge graph 208 with the expected modification data. Taking Figure 7 the production line simulation model 700 in

[0056] as an example, if the model modification module 211 determines that a new test station with a processing time of 120 seconds can be used to replace the second test station in the production line, then the original processing time (140 seconds) of the second test station in the knowledge graph 208 can be replaced with the new processing time (120 seconds).

[0057] Steps 603-606 in method 600 are the same as steps 404-407 in method 400. In step 603, the updated knowledge graph 208 of the production line simulation model 204 is obtained by the graph connector 207, the graph data is read, and the model representation data 206 is updated based on the graph data. In step 604, the updated model representation data 206 is compared with the previous model representation data 206 by the model connector 205. In step 605, the model connector 205 determines the difference data between the updated model representation data 206 and the previous model representation data 206 based on the comparison result. In step 606, the model connector 205 modifies the production line simulation model 204 according to the difference data.

[0057] Next, in step 607, the simulation software ( Figure 2 not shown in Figure 2 ) performs a simulation using the updated production line simulation model 204. In step 608, the application judgment module ( Figure 2 not shown in Figure 7 ) determines whether to apply the expected modification data in the actual production line according to the simulation results. Still taking Figure 7 the production line simulation model 700 in

[0058]

[0058] In the above embodiments, by means of the preset graph semantics, the manufacturing simulation model is converted into a knowledge graph, which can save and manage the knowledge of the manufacturing simulation model in a unified format, contributing to the digitization of discrete manufacturing factories. The knowledge graph of the manufacturing simulation model also provides a unified knowledge acquisition method for machines and humans, improving the efficiency of knowledge acquisition. In addition, by converting manufacturing simulation models in different formats into model representation data in a unified format and then into graph data of the knowledge graph, the conversion between the manufacturing simulation model and the knowledge graph can be easily achieved.

[0059] Figure 9 shows an apparatus for generating and utilizing a knowledge graph of a manufacturing simulation model according to an embodiment of the present disclosure. Referring to Figure 9 , the apparatus 900 includes a model data generation unit 901, a graph semantics acquisition unit 902, and a graph data extraction unit 903. The model data generation unit 901 is configured to obtain a manufacturing simulation model in a factory and generate model representation data based on the manufacturing simulation model. The graph semantics acquisition unit 902 is configured to obtain preset graph semantics for the manufacturing simulation model, and the graph semantics provides a semantic description of the knowledge graph of the manufacturing simulation model. The graph data extraction unit 903 is configured to extract data corresponding to the graph semantics from the model representation data as graph data of the knowledge graph to form a knowledge graph. Figure 9 Each unit in

[0060] can be implemented by using software, hardware (such as integrated circuits, FPGAs, etc.), or a combination of software and hardware. Figure 9 In some embodiments, the apparatus 900 further includes a graph acquisition unit, a model data update unit, and a model update unit (

[0061] not shown in

[0062] ). The graph acquisition unit is configured to obtain an updated knowledge graph of the manufacturing simulation model. The model data update unit is configured to read the graph data of the updated knowledge graph and generate updated model representation data based on the read graph data. The model update unit is configured to update the manufacturing simulation model by using the updated model representation data.

[0061] In some embodiments, the model update unit is further configured to: compare the updated model representation data with the previous model representation data; determine difference data between the updated model representation data and the previous model representation data based on the comparison result; and modify the manufacturing simulation model according to the difference data.

[0062] In some embodiments, the apparatus 900 further includes an update data judgment unit and a graph data modification unit ( Figure 9(not shown in the figure). The updated data determination unit is configured to determine whether there is model update data associated with the manufacturing simulation model according to the graph data of the knowledge graph. The graph data modification unit is configured to modify the graph data by using the model update data to form an updated knowledge graph.

[0063] In some embodiments, the model update data is actual manufacturing data, and the graph data modification unit is further configured to: determine whether the actual manufacturing data is the same as the corresponding data in the graph data; and when the actual manufacturing data is different from the corresponding data in the graph data, replace the corresponding data in the graph data with the actual manufacturing data.

[0064] In some embodiments, the model update data is expected modification data, and the apparatus 900 further includes a model simulation unit and an application determination unit ( Figure 9 (not shown in the figure). The model simulation unit is configured to perform simulation by using the updated manufacturing simulation model. The application determination unit is configured to determine whether to apply the expected modification data in the factory based on the simulation result.

[0065] In some embodiments, the apparatus 900 further includes a query condition obtaining unit and a graph data reading unit ( Figure 9 (not shown in the figure). The query condition obtaining unit is configured to obtain a query condition for the manufacturing simulation model. The graph data reading unit is configured to read graph data matching the query condition from the knowledge graph.

[0066] In some embodiments, the knowledge graph of the manufacturing simulation model is stored in a graph database, and the graph database is further used to store other knowledge graphs related to the factory.

[0067] Figure 10 FIG. shows a block diagram of a computing device for generating and utilizing a knowledge graph of a manufacturing simulation model according to an embodiment of the present disclosure. As Figure 10 can be seen, the computing device 1000 for generating and utilizing the knowledge graph of the manufacturing simulation model includes a processor 1001 and a memory 1002 coupled to the processor 1001. The memory 1002 is used to store computer-executable instructions, and when the computer-executable instructions are executed, the processor 1001 executes the methods in the above embodiments.

[0068] In addition, alternatively, the above method can be implemented via a computer-readable storage medium. A computer-readable program instruction for executing various embodiments of the present disclosure is uploaded on the computer-readable storage medium. The computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. The computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (a non-exhaustive list) of the computer-readable storage medium include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, a mechanically encoded device, such as a punched card or raised structures in grooves storing instructions thereon, and any suitable combination of the above. The computer-readable storage medium used herein is not construed as an instantaneous signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated through a waveguide or other transmission medium (e.g., an optical pulse through an optical fiber cable), or an electrical signal transmitted through a wire.

[0069] Accordingly, in another embodiment, the present disclosure provides a computer-readable storage medium having computer-executable instructions stored thereon for performing the methods in various embodiments of the present disclosure.

[0070] In another embodiment, the present disclosure provides a computer program product tangibly stored on a computer-readable storage medium and including computer-executable instructions that, when executed, cause at least one processor to perform the methods in various embodiments of the present disclosure.

[0071] Generally, the various example embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software executable by a controller, a microprocessor, or other computing devices. When aspects of the embodiments of the present disclosure are illustrated or described as block diagrams, flowcharts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented as non-limiting examples in hardware, software, firmware, dedicated circuits or logic, general hardware or controllers or other computing devices, or some combination thereof.

[0072] The computer-readable program instructions or computer program products for implementing the various embodiments of the present disclosure can also be stored in the cloud. When needed, users can access the computer-readable program instructions stored in the cloud for implementing an embodiment of the present disclosure through the mobile Internet, fixed network, or other networks, so as to implement the technical solutions disclosed in accordance with the various embodiments of the present disclosure.

[0073] Although the embodiments of the present disclosure have been described with reference to several specific embodiments, it should be understood that the embodiments of the present disclosure are not limited to the specific embodiments disclosed. The embodiments of the present disclosure are intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims. The scope of the claims is to be accorded the broadest interpretation so as to encompass all such modifications and equivalent structures and functions.

Claims

1. A method for generating and utilizing a knowledge graph of a manufacturing simulation model, comprising: Obtaining a manufacturing simulation model in a factory and generating model representation data based on the manufacturing simulation model; wherein, the model representation data includes production workstations, transfer workstations, staging areas, supply stations, and recycling stations; Obtaining preset graph semantics for the manufacturing simulation model, the graph semantics providing a semantic description of the knowledge graph of the manufacturing simulation model; and Extracting data corresponding to the graph semantics from the model representation data as graph data of the knowledge graph to form the knowledge graph; Wherein, further comprising: Obtaining an updated knowledge graph of the manufacturing simulation model; Reading the graph data of the updated knowledge graph and generating updated model representation data based on the read graph data; and Updating the manufacturing simulation model using the updated model representation data.

2. The method according to claim 1, wherein Updating the manufacturing simulation model using the updated model representation data further comprises: Comparing the updated model representation data with the previous model representation data; Determining difference data between the updated model representation data and the previous model representation data based on the comparison result; and Modifying the manufacturing simulation model according to the difference data.

3. The method according to claim 1, wherein, Before obtaining the updated knowledge graph of the manufacturing simulation model, the method further comprises: Judging whether there is model update data associated with the manufacturing simulation model according to the graph data of the knowledge graph; and When there is the model update data, modifying the graph data using the model update data to form the updated knowledge graph.

4. The method according to claim 3, wherein The model update data is actual manufacturing data, and modifying the graph data using the model update data further comprises: Judging whether the actual manufacturing data is the same as the corresponding data in the graph data; and When the actual manufacturing data is different from the corresponding data in the graph data, replacing the corresponding data in the graph data with the actual manufacturing data.

5. The method according to claim 3, wherein The model update data is expected modification data, and the method further comprises: Performing a simulation using the updated manufacturing simulation model; and Judging whether to apply the expected modification data in the factory based on the simulation result.

6. The method according to claim 1, further comprising: Obtaining a query condition for the manufacturing simulation model; And Reading graph data matching the query condition from the knowledge graph.

7. The method according to claim 1, wherein, The knowledge graph of the manufacturing simulation model is stored in a graph database, and the graph database is further used to store other knowledge graphs related to the factory.

8. An apparatus for generating and utilizing a knowledge graph of a manufacturing simulation model, comprising: A model data generation unit configured to obtain a manufacturing simulation model in a factory and generate model representation data based on the manufacturing simulation model; wherein, the model representation data includes production workstations, transfer workstations, staging areas, supply stations, and recycling stations; A graph semantic acquisition unit, configured to acquire preset graph semantics for the manufacturing simulation model, where the graph semantics provide a semantic description of the knowledge graph of the manufacturing simulation model; and A graph data extraction unit, configured to extract data corresponding to the graph semantics from the model representation data as graph data of the knowledge graph to form the knowledge graph; Wherein, it further includes: A graph acquisition unit, configured to acquire an updated knowledge graph of the manufacturing simulation model; A model data update unit, configured to read the graph data of the updated knowledge graph and generate updated model representation data based on the read graph data; and A model update unit, configured to update the manufacturing simulation model with the updated model representation data.

9. The device according to claim 8, wherein, The model update unit is further configured to: Compare the updated model representation data with the previous model representation data; Determine difference data between the updated model representation data and the previous model representation data based on the comparison result; And Modify the manufacturing simulation model according to the difference data.

10. The apparatus according to claim 8, further including: An update data judgment unit, configured to judge whether there is model update data associated with the manufacturing simulation model according to the graph data of the knowledge graph; And A graph data modification unit, configured to modify the graph data with the model update data to form the updated knowledge graph.

11. The apparatus according to claim 10, wherein, The model update data is actual manufacturing data, and the graph data modification unit is further configured to: Judge whether the actual manufacturing data is the same as the corresponding data in the graph data; and When the actual manufacturing data is different from the corresponding data in the graph data, replace the corresponding data in the graph data with the actual manufacturing data.

12. The apparatus according to claim 10, wherein, The model update data is expected modification data, and the apparatus further includes: A model simulation unit, configured to perform simulation using the updated manufacturing simulation model; and An application judgment unit, configured to judge whether to apply the expected modification data in the factory based on the simulation result.

13. The apparatus according to claim 8, further including: A query condition acquisition unit, configured to acquire a query condition for the manufacturing simulation model; And A graph data reading unit, configured to read graph data matching the query condition from the knowledge graph.

14. The apparatus according to claim 8, wherein, The knowledge graph of the manufacturing simulation model is stored in a graph database, and the graph database is further used to store other knowledge graphs related to the factory.

15. A computing device, including: A processor; And A memory, configured to store computer-executable instructions, which when executed cause the processor to execute the method according to any one of claims 1-7.

16. A computer-readable storage medium having computer-executable instructions stored thereon for performing the method according to any one of claims 1-7.

17. A computer program product tangibly stored on a computer-readable storage medium and comprising computer-executable instructions that, when executed, cause at least one processor to perform the method according to any one of claims 1-7.

Images