Workstation generation method, workstation generation system, storage medium and electronic equipment

Abstract

The invention provides a workstation generation method, a workstation generation system, a storage medium and electronic equipment, and the method comprises the steps: receiving demand information, and when the demand information is a three-dimensional model, processing the three-dimensional model to obtain target product information, the demand information comprising the category and parameter information of a target product; under the condition that the demand information is text information, processing the text information to obtain target product information; and workstation information for processing the target product is determined based on the product knowledge graph, the equipment knowledge graph and the target product information, the workstation information comprises multiple pieces of equipment information for processing the target product, the product knowledge graph comprises multiple products, and the equipment knowledge graph comprises multiple pieces of candidate equipment. The method solves the problem that in the prior art, equipment selection efficiency is low due to the fact that equipment selection of a workstation needs to be manually matched one by one according to parameters of a target product.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of motor control, in particular to a workstation generation method, a workstation generation system, a computer readable storage medium and an electronic device. BACKGROUND

[0002] At present, in the workstation configuration process in the manufacturing field, the equipment selection usually depends on manual experience, and the engineering and technical personnel need to compare and select one by one in the existing equipment information according to the size, material, structure and other parameters of the target product, so as to determine the adaptive equipment combination. This method not only needs strong professional experience, but also is often tedious and time-consuming in processing complex products or multi-device matching, resulting in low equipment selection efficiency, and even may cause configuration deviation due to inconsistent manual judgment. With the diversification of product types and production processes, the traditional manual method is difficult to meet the demand of rapid response, and there is an urgent need for a workstation generation technology that can improve the equipment selection efficiency and accuracy. SUMMARY

[0003] The main purpose of the present application is to provide a workstation generation method, a workstation generation system, a computer readable storage medium and an electronic device, so as to at least solve the problem that the equipment selection of the workstation in the prior art needs to manually match the equipment according to the parameters of the target product, resulting in low equipment selection efficiency.

[0004] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a workstation generation method is provided, comprising: receiving demand information, in the case that the demand information is a three-dimensional model, processing the three-dimensional model to obtain target product information, the demand information including the category and parameter information of the target product; in the case that the demand information is text information, processing the text information to obtain the target product information; determining the workstation information for processing the target product based on the product knowledge graph, the equipment knowledge graph and the target product information, the workstation information including a plurality of equipment information for processing the target product, the product knowledge graph including a plurality of products, and the equipment knowledge graph including a plurality of candidate equipment.

[0005] Optionally, in the case that the demand information is text information, processing the text information to obtain the target product information, comprising: adopting a pre-trained neural network model to process the text information to extract key information in the demand information, the key information at least including the category and parameters of the target product; filling a preset template according to the key information to obtain the target product information.

[0006] Optionally, after determining the workstation information for processing the target product based on the product knowledge graph, the equipment knowledge graph and the target product information, the method further comprises: generating simulation parameters according to the workstation information, and constructing a workstation simulation model based on the simulation parameters; running the workstation simulation model to obtain simulation results, the simulation results comprising running state data of the equipment in the workstation information, size data of the target product and efficiency data; scoring the simulation results according to a preset scoring rule to obtain a score of the simulation results, and obtaining decision information according to the score, the decision information comprising selection result information of the equipment for processing the target product.

[0007] Optionally, after obtaining the decision information according to the score, the method further comprises: outputting the decision information in a case where confirmation of the decision information is received; and correcting the decision information according to correction information of the decision information in a case where the correction information of the decision information is received.

[0008] Optionally, the equipment knowledge graph comprises fixed relationship nodes and non-fixed relationship nodes, the fixed relationship nodes being one-to-one corresponding nodes of the target product and the equipment for processing the target product, and the non-fixed relationship nodes comprising a preset function for determining a plurality of the equipment for processing the target product corresponding to the target product, the determining the workstation information for processing the target product based on the preset product family, the preset equipment family and the target product information comprising: traversing the equipment knowledge graph, determining, in a case where the fixed relationship nodes are traversed, that the set of the target product information and the equipment information for processing the target product is the workstation information; and determining, in a case where the non-fixed relationship nodes are traversed, a matching value of the target product information and the plurality of the equipment for processing the target product according to the preset function corresponding to the non-fixed relationship nodes, and generating the workstation information comprising the matching value.

[0009] According to another aspect of the present application, a workstation generation system is provided, comprising: an intelligent unit configured to receive demand information, process the demand information to obtain target product information in a case where the demand information is a three-dimensional model, and process the demand information to obtain the target product information in a case where the demand information is text information; a retrieval unit configured to determine workstation information for processing the target product based on a preset knowledge graph and the target product information; and an interaction unit configured to output the workstation information.

[0010] Optionally, the intelligent unit comprises a demand analysis intelligent agent, an inference decision intelligent agent, a simulation intelligent agent and an artificial teaching intelligent agent, the demand analysis intelligent agent is configured to extract the demand information to obtain target product information, the inference decision intelligent agent is configured to generate workstation information, the simulation intelligent agent is configured to generate simulation parameters, and the artificial teaching intelligent agent is configured to correct the workstation information according to received correction information.

[0011] Optionally, the system further comprises a verification unit comprising a simulation module and a decision module, the simulation module is configured to perform simulation verification on the simulation parameters sent by the intelligent unit to obtain simulation results, and the decision module is configured to generate decision information according to the simulation results.

[0012] According to another aspect of the present application, a computer readable storage medium is provided, the computer readable storage medium comprises a stored program, wherein the computer readable storage medium controls the device where the computer readable storage medium is located to execute any one of the methods when the program runs.

[0013] According to another aspect of the present application, an electronic device is provided, comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs comprise a program for executing any one of the methods.

[0014] According to the technical scheme of the present application, different types of demand information are analyzed to automatically extract the category, geometric parameters, material and process requirements of the target product, avoiding the inefficient way of relying on manual reading of drawings or instructions to extract information in the prior art, and improving the accuracy and consistency of information acquisition. Through the retrieval of the product knowledge graph and the device knowledge graph, the target product category can be quickly located, and multiple candidate devices corresponding to the product can be found in the device knowledge graph, so as to generate workstation information. The tedious process of manually comparing product parameters and device parameters is avoided, and the efficiency of device selection and workstation configuration is improved. The generated workstation information includes multiple device information for processing the target product, which can cover different processes and process links required for the production of the target product, ensuring that the configured workstation can completely support the manufacturing process of the target product. The present scheme does not need to manually compare target product parameters and candidate device parameters, but can directly generate standardized target product information through automatic analysis of three-dimensional models or text information. In combination with the product knowledge graph and the device knowledge graph, retrieval and matching are realized, so as to automatically determine the workstation information for processing the target product, solving the problem of low efficiency of device selection for the workstation in the prior art, which needs to manually match devices according to the parameters of the target product. BRIEF DESCRIPTION OF DRAWINGS

[0015] The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the application, and are incorporated in and constitute a part of this specification, the illustrations of the embodiments of the application, and to explain the application without imposing on it any unnecessary limitations. In the drawings:

[0016] Figure 1 A flowchart of a workstation generation method according to an embodiment of the application is shown;

[0017] Figure 2 A structural block diagram of a workstation generation system according to an embodiment of the application is shown;

[0018] Figure 3 A structural block diagram of yet another workstation generation system according to an embodiment of the application is shown;

[0019] Figure 4 A flowchart of yet another workstation generation method according to an embodiment of the application is shown. DETAILED DESCRIPTION

[0020] It should be noted that the embodiments and features of the application herein disclosed can be combined with each other unless specifically stated otherwise. The embodiments of the application will be described in greater detail below with reference to the accompanying drawings, in which:

[0021] In order to make the personnel in the art better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor should be within the scope of protection of the present application.

[0022] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described accompanying drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not necessarily limit to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0023] As introduced in the background, the device selection of the workstation in the prior art needs to manually match the device according to the parameters of the target product, which leads to low efficiency of device selection. To solve the above problem, the embodiments of the present application provide a workstation generation method, a workstation generation system, a computer readable storage medium and an electronic device.

[0024] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.

[0025] Figure 1 is a flowchart of the workstation generation method according to the embodiments of the present application. As shown in Figure 1 , the method comprises the following steps:

[0026] Step S101, receiving demand information, in the case of the demand information being a three-dimensional model, processing the three-dimensional model to obtain target product information, the demand information including the category and parameter information of the target product;

[0027] Specifically, the demand information includes the category, name and geometric characteristics of the target product, the parameter information of the target product, such as size, material and thickness; process demand information, such as target processing method, process precision, processing path or task type; constraint condition information, such as production cycle, cost limit, resource availability, safety requirement, etc. When the demand information is a three-dimensional model, the three-dimensional model is analyzed and processed, and the shape characteristics, size data and material label of the three-dimensional model are analyzed to obtain the category information and parameter information related to the target product.

[0028] Step S102, in the case of the demand information being text information, processing the text information to obtain the target product information;

[0029] Specifically, the target product information is a data set of the target product category and parameters.

[0030] Step S103, determining the workstation information for processing the target product based on the product knowledge graph, the device knowledge graph and the target product information, the workstation information including a plurality of device information for processing the target product, the product knowledge graph including a plurality of products, and the device knowledge graph including a plurality of candidate devices.

[0031] Specifically, the product knowledge graph is used to describe the classification relationship and attribute information between different products, such as product category nodes such as “pressure vessel”, “ship component”, “engineering machinery” and corresponding attribute parameters. By retrieving the product node corresponding to the target product information in the product knowledge graph, the standardized expression of the target product can be determined. The above-mentioned equipment knowledge graph is used to describe the category, performance parameters and association relationship with product category of the candidate equipment, such as equipment nodes such as “automatic welding machine”, “cutting machine” and “assembly robot”. By retrieving the equipment node associated with the target product node in the equipment knowledge graph, the information of multiple candidate equipment can be obtained.

[0032] Through the above embodiment, according to different types of demand information, the category, geometric parameters, material and process requirements of the target product are automatically extracted, avoiding the inefficient way of relying on manual reading of drawings or instructions to extract information in the prior art, and improving the accuracy and consistency of information acquisition. Through the retrieval of the product knowledge graph and the equipment knowledge graph, the category to which the target product belongs can be quickly located, and multiple candidate equipment corresponding to the product can be found in the equipment knowledge graph, so as to generate the workstation information. Avoiding the tedious process of manually comparing product parameters and equipment parameters one by one, the efficiency of equipment selection and workstation configuration is improved. The generated workstation information includes multiple equipment information for processing the target product, which can cover different processes and process links required for the production of the target product, ensuring that the configured workstation can completely support the manufacturing process of the target product. The present scheme does not need to manually compare the target product parameters and candidate equipment parameters one by one, but can directly generate standardized target product information through automatic analysis of three-dimensional models or text information; combined with the product knowledge graph and the equipment knowledge graph, retrieval and matching are realized, so as to automatically determine the workstation information for processing the target product, solving the problem of low efficiency of equipment selection in the prior art, which requires manual matching of equipment according to the parameters of the target product.

[0033] In an optional solution, when the above demand information is text information, the above text information is processed to obtain the above target product information, including: using a pre-trained neural network model to process the above text information to extract key information in the above demand information, the above key information at least including the category and parameters of the above target product; filling a preset template according to the above key information to obtain the above target product information.

[0034] In the above embodiments, the pre-trained neural network model can understand synonyms, sequence changes, and complex sentence patterns, thereby ensuring that product categories and parameter information can be correctly extracted even if the description methods of the text information are different. Compared with methods based on fixed rules or simple keywords, the present scheme can still maintain high accuracy when facing ambiguous descriptions or diversified expressions. The extracted key information is filled into the preset template to form structured target product information, so that text inputs of different sources and formats can be finally converted into unified parameterized expressions, facilitating subsequent matching with product knowledge graphs and equipment knowledge graphs, improving the standardization of data, and avoiding retrieval and reasoning biases caused by non-uniform information expression.

[0035] Specifically, the neural network model is a semantic analysis model based on natural language processing technology, which can perform word segmentation, vector representation, and context semantic analysis on text information, thereby identifying the product categories and parameter descriptions involved in the text. For example, when the text contains "a pressure vessel end cover with a thickness of 10 mm", the neural network model can identify "pressure vessel end cover" as the target product category and "thickness 10 mm" as the product parameter. Then, the key information identified by the neural network model is extracted and summarized. The key information at least includes the category information of the target product, such as "pressure vessel" and "ship component", and the parameter information, such as "thickness 10 mm".

[0036] In another optional scheme, after determining the workstation information for processing the target product based on the product knowledge graph, the equipment knowledge graph, and the above target product information, the above method further includes: generating simulation parameters according to the above workstation information, and constructing a workstation simulation model based on the above simulation parameters; running the above workstation simulation model to obtain simulation results, the simulation results including the running state data of the equipment in the above workstation information, the size data of the target product, and the efficiency data; scoring the simulation results according to a preset scoring rule to obtain a score of the simulation results, and obtaining decision information according to the score, the decision information including selection result information of the equipment for processing the target product.

[0037] In the above embodiments, after determining the workstation information, the simulation parameters are generated and the workstation simulation model is constructed, so that the candidate equipment combination can be simulated and run in a virtual environment, thereby verifying the feasibility of the equipment combination in advance before actual production, and avoiding high cost and time waste caused by direct physical test. The simulation results obtained after running the simulation model not only include the running state data of the equipment, such as load, energy consumption, running stability, but also include the processing size data of the target product, such as geometric size deviation, shape and position tolerance, and efficiency data, such as production rhythm, output quantity, which provides an objective and quantitative basis for equipment selection. Compared with the traditional experience-based judgment method, it is more scientific and reliable. The simulation results are scored by pre-set scoring rules, so as to comprehensively consider the running performance of the equipment, the processing accuracy of the product and the overall efficiency of the workstation, and generate an adaptation score. Based on the score, the decision information can directly point out the optimal equipment combination scheme, and the final decision information can avoid inefficient, unstable or unsatisfactory equipment configuration, significantly reduce the risk of quality problems, low efficiency or equipment overload in actual production, and improve the reliability of workstation configuration.

[0038] Specifically, the workstation information includes a plurality of candidate equipment for processing the target product. The equipment model, process capacity, processing sequence and layout constraint information are extracted and converted into simulation parameters. For example, the current, voltage and welding speed of the welding equipment, the feed speed and cutting path of the cutting equipment, which are part of the simulation parameters. Then, a workstation simulation model is constructed based on the simulation parameters, so as to simulate the running of the entire workstation in a virtual environment, including the action range of the equipment, the process connection relationship and the production rhythm, so as to virtually verify the selected workstation scheme before actual deployment. Then, the workstation simulation model is run to obtain simulation results. The simulation results include a plurality of aspects of data, for example: equipment running state data, including running time, load condition, energy consumption data, failure rate prediction, etc.; target product size data, including key dimensions of machined parts, shape and position tolerances, and deviations from design dimensions; efficiency data, including output quantity per unit time, rhythm time and overall production line efficiency. Then, the simulation results are evaluated according to the pre-set scoring rules. The scoring rules can comprehensively consider the stability of the equipment, the processing accuracy of the target product and the overall efficiency of the workstation, etc., to quantitatively score the simulation results and obtain a score of the simulation results. Finally, decision information is obtained according to the score. The decision information includes equipment selection result information for processing the target product, i.e., clearly indicating which equipment combination can optimally meet the processing requirements of the target product. The decision information can be directly used as a basis for subsequent workstation configuration and production deployment, thereby avoiding the uncertainty of manual experience-based judgment.

[0039] In an alternative, after obtaining the decision information according to the score, the method further comprises: outputting the decision information upon receiving a confirmation of the decision information; and correcting the decision information according to correction information of the decision information upon receiving the correction information.

[0040] In the above embodiments, after generating the decision information, the final decision information is outputted only when a confirmation signal is received, so as to avoid that the incorrect or unverified result directly enters the actual execution link, thereby improving the reliability and safety of the output result. When the correction information is received, the existing decision information is modified or adjusted. For example, when a candidate device is unavailable due to failure in the actual environment, the device can be excluded from the candidate result through the correction information, and the system automatically corrects the decision result accordingly, so as to flexibly cope with the dynamically changing production environment and ensure the implementability of the workstation generated scheme.

[0041] In some example schemes of the present application, the device knowledge graph includes fixed relationship nodes and non-fixed relationship nodes, the fixed relationship nodes are one-to-one corresponding nodes of the target product and the device for processing the target product, and the non-fixed relationship nodes include a preset function for determining a plurality of devices for processing the target product corresponding to the target product, and the workstation information for processing the target product is determined based on a preset product family, a preset device family and the target product information, including: traversing the device knowledge graph, determining that the set of the target product information and the device information for processing the target product is the workstation information when the fixed relationship node is traversed; and determining a matching value of the target product information and a plurality of devices for processing the target product according to the preset function corresponding to the non-fixed relationship node, and generating the workstation information including the matching value when the non-fixed relationship node is traversed.

[0042] In the above embodiments, by introducing fixed relationship nodes and non-fixed relationship nodes in the equipment knowledge graph, when there is a fixed relationship node, the corresponding equipment of the target product can be directly determined, and the workstation information can be quickly generated; when there is a non-fixed relationship node, the preset function is used for calculation, and a plurality of candidate equipment and matching values thereof are dynamically determined. The non-fixed relationship node introduces a dynamic matching mechanism through function calculation, and the candidate equipment can be flexibly selected according to different product parameters, equipment capabilities and process requirements. In this way, the stability of the configuration can be maintained, and the needs of diversified and complex production scenes can be met. In the case of a non-fixed relationship node, a matching value between the target product and a plurality of candidate equipment is generated, and the matching value can represent the adaptability of different equipment to the target product, thereby providing a quantitative basis for subsequent scoring, simulation and optimization. Compared with the traditional manual device screening method, the present scheme can more scientifically compare the advantages and disadvantages of the equipment, and ensure the rationality of the final selection. In addition, the traditional method often relies on manual comparison of product parameters and equipment parameters, which is not only low in efficiency, but also easy to miss or misjudge. The present scheme combines the fixed relationship node and the function matching mechanism to quickly and accurately output the workstation information and the equipment matching value, thereby improving the efficiency and accuracy of the workstation configuration.

[0043] Exemplarily, after obtaining the target product information, nodes associated with the target product are traversed layer by layer in the equipment knowledge graph. Each node in the equipment knowledge graph represents an association relationship, and different processing strategies are adopted when different types of nodes are traversed. When a fixed relationship node is traversed, it indicates that there is a stable one-to-one correspondence between the target product and the equipment, for example, a sheet part is processed by a laser cutting machine, and a pressure vessel end cover is processed by an automatic welding machine. In this case, the correspondence between the target product information and the equipment information is directly determined, and the target product information and the corresponding equipment information are collected as part of the candidate workstation information. When a non-fixed relationship node is traversed, it indicates that there is a dynamic and adjustable correspondence between the target product and a plurality of equipment. For example, the "steel plate welding process" can be completed by a robot welding machine or a manual welding workstation. At this time, a preset function bound to the non-fixed relationship node is called, and the specific parameters of the target product, such as thickness, material, size requirements, and the performance parameters of the candidate equipment, such as load range, machining precision and running speed, are matched and calculated to obtain the matching values between the target product and a plurality of candidate equipment.

[0044] In this embodiment, the aforementioned preset function is the result of statistical analysis and fitting of historical production case data. Based on recorded data from previous product processing, including product parameters and processing performance data of different equipment, a correspondence between product parameters and equipment performance is established through regression modeling and curve fitting, and this relationship is solidified into a preset function. This preset function is used to calculate the target product information and candidate equipment information during the traversal of the equipment knowledge graph, obtaining the fit between the target product and multiple candidate equipment, thereby providing a quantitative basis for generating workstation information.

[0045] This application also provides a workstation generation system, such as Figure 2 As shown, it includes: an intelligent unit 10, used to receive demand information, process the three-dimensional model to obtain target product information when the demand information is a three-dimensional model, and process the text information to obtain target product information when the demand information is text information; a retrieval unit 20, used to determine workstation information for processing the target product based on a preset knowledge graph and the target product information; and an interaction unit 30, used to output the workstation information.

[0046] Through the above embodiments, the intelligent unit of this solution can simultaneously support two input methods: 3D models and text information. When the requirement information is a 3D model, it automatically parses information such as geometric dimensions, structural features, and material properties. The intelligent unit processes different input sources uniformly and ultimately outputs structured target product information, which can be directly used as input for knowledge graph retrieval, avoiding inconsistencies and error rates associated with manual parameter extraction, thereby improving the accuracy and standardization of information parsing. Through the retrieval unit, based on preset product and equipment knowledge graphs, it quickly matches candidate equipment related to the target product and combines them into workstation information, improving the efficiency of equipment selection and workstation generation. The interaction unit can output workstation information to users or upper-level systems, making the generated results visible.

[0047] In one alternative embodiment, the aforementioned intelligent unit includes a demand analysis agent, a reasoning and decision-making agent, a simulation agent, and a human teaching agent. The demand analysis agent is used to extract the aforementioned demand information to obtain target product information. The reasoning and decision-making agent is used to generate workstation information. The simulation agent is used to generate simulation parameters. The human teaching agent is used to correct the aforementioned workstation information based on the received correction information.

[0048] In the above embodiments, the demand analysis agent can automatically parse demand information and extract the category, parameters, and process requirements of the target product, ensuring the accuracy and standardization of the input information. This provides a reliable data foundation for subsequent workstation generation. The reasoning and decision-making agent can combine product knowledge graphs and equipment knowledge graphs to automatically generate candidate equipment combination schemes using a reasoning mechanism. After generating workstation information, the simulation agent further generates simulation parameters for constructing a workstation simulation model and performing simulation verification. The human teaching agent can receive external feedback correction information and revise the generated workstation information. This human-machine collaborative mechanism provides a feedback channel for the continuous optimization of the knowledge graph and decision-making model.

[0049] In another alternative embodiment, the system further includes a verification unit, comprising a simulation module and a decision module. The simulation module is used to perform simulation verification on the simulation parameters sent by the intelligent unit to obtain simulation results, and the decision module is used to generate decision information based on the simulation results.

[0050] In the above embodiments, by setting up a verification unit, the simulation parameters generated by the intelligent unit can be virtually simulated before actual production to obtain simulation results. This allows for early verification of the rationality of equipment combinations and process flows, avoiding the high costs and potential risks associated with direct testing on a real production line. The simulation results output by the simulation module include multi-dimensional data such as equipment operating status, processing accuracy, energy consumption, and efficiency, comprehensively reflecting the workstation's performance and providing objective and quantitative evaluation criteria for the decision-making module. Compared to the traditional method of relying on subjective judgment based on human experience, this solution significantly improves the scientific nature of equipment selection and process configuration. The decision-making module can generate decision information based on the simulation results and clearly indicate which equipment combinations are optimal, shortening the configuration cycle and ensuring the rationality and reliability of the output solution.

[0051] To enable those skilled in the art to better understand the technical solution of this application, the implementation process of the workstation generation method of this application will be described in detail below with reference to specific embodiments.

[0052] This embodiment relates to a method applied to a specific workstation generation system, such as... Figure 3As shown, the generation system includes an intelligent unit, a retrieval unit, an interaction unit, and a verification unit. The intelligent unit receives text information from the interaction unit, sends requirement information to the retrieval unit and receives retrieval information, and sends simulation data and decision data to the verification unit. Text information is information sent by the user and transformed by the input / output unit, consisting of characters, numbers, and punctuation marks. Requirement information is objects or concepts with specific categories or meanings automatically identified and extracted from the text information by the intelligent unit. Retrieval information is formatted text retrieved by the retrieval unit that has a structured relationship with the requirement information and has been transformed. Simulation data refers to all data and programs used in scenario generation and use; the aforementioned decision data refers to the decision-making reasoning process of the intelligent unit. The retrieval unit includes product families, device families, and a database. It receives entity information sent by the intelligent unit, performs correlation retrieval to generate a retrieval subgraph, and converts it into text-based retrieval information, which is then sent to the intelligent unit. The input and output units send text information to the intelligent unit and receive simulation information and decision information sent by the verification unit. The verification unit includes a simulation generation unit and a decision generation unit. The simulation generation unit receives simulation data and decision data sent by the intelligent unit and converts them into simulation information. The decision generation unit generates decision information. The verification unit sends the simulation information and decision information to the input / output unit. Simulation information refers to data and programs that can be displayed through a display screen. Decision information refers to the text-based reasoning and decision-making process information of the intelligent unit. The intelligent unit includes a large model, a requirements analysis agent, a reasoning and decision-making agent, a simulation generation agent, and a human teaching agent. The requirements analysis agent receives text information sent by the interaction unit, extracts requirements information, sends it to the retrieval unit, matches the reasoning chain template, sends it to the reasoning and decision-making agent, and receives retrieval information from the retrieval unit. The reasoning and decision-making agent receives the reasoning chain template sent by the requirements analysis agent and the retrieval information sent by the retrieval unit. Relying on mapping tools, it ultimately outputs decision data, simulation data, and equipment information, and sends the equipment information to the simulation generation agent. The simulation generation agent receives the equipment information sent by the reasoning and decision-making agent and generates simulation data. The human teaching agent decides whether to send modification information to the reasoning and decision-making agent based on operator feedback.The retrieval unit includes product families, equipment families, and a database. Product families refer to end products or structural components with welding as a key manufacturing process, including pressure vessels, ship components, engineering machinery, and automotive parts. Equipment families refer to various tools, devices, and systems used to realize the welding process, including welding robots, auxiliary shaft equipment, welding torches, welding machines, fume extraction devices, torch cleaning devices, testing devices, and protective devices. The database refers to the storage unit used to store enterprise processing and production cases. Both product families and equipment families are clusters in the form of knowledge graphs. The relationships between adjacent nodes include fixed relationships and non-fixed relationships. Fixed relationships are used to construct the basic classification framework. For example, pressure vessels, ship components, and engineering machinery all belong to product families. Non-fixed relationships are achieved through function tools to realize adjustable associations between nodes, such as Y=F. C11 (X), where X is the input matrix and Y is the selection result of the node output. The input / output unit includes an input module and an output module. The input module receives information such as 3D models and text requirements sent by the user, and the output module displays the simulation information and decision information sent by the intelligent unit. The verification unit includes a simulation generation module and a decision generation module. The simulation generation module converts the simulation data sent by the intelligent unit into simulation information, and the decision generation module converts the decision data sent by the intelligent unit into decision information. Using the above generation system, a method for generation on a specific workstation is provided, such as... Figure 4 As shown, it includes:

[0053] Step S201: The intelligent unit receives text information sent by the input / output unit, and the demand analysis intelligent agent converts the text information into demand information.

[0054] Step S202: The retrieval unit receives the demand information and converts it into retrieval information; the reasoning and decision-making agent receives the retrieval information and converts it into the aforementioned decision data and equipment information; the device information is sent to the simulation generation agent; the simulation generation agent receives the device information sent by the reasoning and decision-making agent and generates simulation data.

[0055] Step S203: The verification unit generates simulation information and decision information based on the simulation data and decision data sent by the intelligent unit. The input / output unit receives and displays the simulation information and decision information. The operator judges whether the task is completed based on the simulation information and decision information.

[0056] Step S204: If the operator considers the task complete, the work ends; if the operator considers it incomplete, the manual teaching phase begins, where the information is modified by the human teaching agent, and the case is added to the database; the retrieval unit completes the map incremental update task based on the newly added data.

[0057] This application also provides a specific implementation scenario for traversing a knowledge graph. In the scenario of generating a welding workstation for ship components, fixed relationship nodes represent steel plates of certain thicknesses, such as those thicker than 20mm. In the shipbuilding industry, a specific model of high-power submerged arc welding machine must be used; this is an industry standard constraint and represents a stable one-to-one correspondence. When traversing the knowledge graph, encountering this fixed relationship node allows for the direct addition of the thick plate to the workstation information using a submerged arc welding machine. Non-fixed relationship nodes, such as for medium-thickness steel plates (6 to 12mm), may allow for the use of either a gas-shielded welding robot or a manual welding workstation, depending on the product length, weld complexity, and precision requirements. This situation is handled by a preset function. The function input includes target product parameters (plate thickness, weld length, required precision) and candidate equipment parameters (number of robots, rated current, processing efficiency). The function outputs the matching value between each candidate device and the target product; for example, a gas-shielded welding robot scores 0.82, and a manual welding robot scores 0.67. Finally, the matching value is appended to the candidate equipment information to form workstation information containing priorities. By quantifying non-fixed relationship nodes using preset functions, the subjectivity of relying solely on human experience in equipment selection is avoided, making the selection process more scientific and repeatable. Fixed relationships ensure the execution of standardized, rigid constraints, while non-fixed relationships provide the possibility of dynamic selection, flexibly responding to changing needs of different products and processes. The entire process is implemented through knowledge graph traversal and function calculation, avoiding manual comparison of parameters one by one, significantly shortening equipment selection time and improving configuration efficiency.

[0058] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0059] This invention provides a computer-readable storage medium including a stored program, wherein the program, when running, controls the device containing the computer-readable storage medium to execute the workstation generation method.

[0060] Specifically, the workstation generation methods include:

[0061] Step S101: Receive demand information. If the demand information is a three-dimensional model, process the three-dimensional model to obtain target product information. The demand information includes the category and parameter information of the target product.

[0062] Specifically, the requirement information includes the target product's category, name, and geometric features; parameter information such as dimensions, material, and thickness; process requirement information such as target processing method, process precision, processing path, or task type; and constraint information such as production cycle, cost limits, resource availability, and safety requirements. When the requirement information is a 3D model, the 3D model is parsed and processed, and the shape features, dimensional data, and material labels of the 3D model are analyzed to obtain the category and parameter information related to the target product.

[0063] Step S102: If the above-mentioned demand information is text information, process the above-mentioned text information to obtain the above-mentioned target product information.

[0064] Specifically, the aforementioned target product information is a data set of target product categories and parameters.

[0065] Step S103: Based on the product knowledge graph, the equipment knowledge graph, and the aforementioned target product information, determine the workstation information for processing the target product. The workstation information includes information on multiple devices for processing the target product. The product knowledge graph includes multiple products, and the equipment knowledge graph includes multiple candidate devices.

[0066] Specifically, the product knowledge graph describes the classification relationships and attribute information between different products, such as product category nodes like "pressure vessel," "ship component," and "construction machinery," along with their corresponding attribute parameters. By retrieving product nodes corresponding to the target product information from the product knowledge graph, a standardized representation of the target product can be determined. The aforementioned equipment knowledge graph describes the categories, performance parameters, and relationships with product categories of candidate equipment, such as equipment nodes like "automatic welding machine," "cutting machine," and "assembly robot." By retrieving equipment nodes associated with the target product node from the equipment knowledge graph, information on multiple candidate devices can be obtained.

[0067] In one embodiment of this application, when the above-mentioned demand information is text information, the above-mentioned text information is processed to obtain the above-mentioned target product information, including: processing the above-mentioned text information using a pre-trained neural network model to extract key information from the above-mentioned demand information, the above-mentioned key information including at least the category and parameters of the above-mentioned target product; filling a preset template according to the above-mentioned key information to obtain the above-mentioned target product information.

[0068] In one embodiment of this application, after determining the workstation information for processing the target product based on the product knowledge graph, the equipment knowledge graph, and the aforementioned target product information, the method further includes: generating simulation parameters based on the workstation information, and constructing a workstation simulation model based on the simulation parameters; running the workstation simulation model to obtain simulation results, the simulation results including the operating status data of the equipment, the size data of the target product, and the efficiency data in the workstation information; scoring the simulation results according to a preset scoring rule to obtain a score for the simulation results, and obtaining decision information based on the score, the decision information including selection result information of the equipment used to process the target product.

[0069] In one embodiment of this application, after obtaining decision information based on the above score, the method further includes: outputting the decision information upon receiving confirmation of the decision information; and correcting the decision information based on the correction information upon receiving correction information for the decision information.

[0070] In one embodiment of this application, the device knowledge graph includes fixed relationship nodes and non-fixed relationship nodes. The fixed relationship nodes are nodes that correspond one-to-one with the target product and the equipment used to process the target product. The non-fixed relationship nodes include a preset function for determining multiple equipment used to process the target product. Based on a preset product family, a preset equipment family, and the target product information, the function determines workstation information for processing the target product. This includes: traversing the device knowledge graph; when encountering a fixed relationship node, determining the set of the target product information and the equipment information for processing the target product as the workstation information; when encountering a non-fixed relationship node, determining the matching value between the target product information and the multiple equipment used to process the target product according to the preset function corresponding to the non-fixed relationship node, and generating workstation information including the matching value.

[0071] This invention provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs at least the following steps: Step S101, receiving demand information, and if the demand information is a three-dimensional model, processing the three-dimensional model to obtain target product information, wherein the demand information includes the category and parameter information of the target product.

[0072] Specifically, the requirement information includes the target product's category, name, and geometric features; parameter information such as dimensions, material, and thickness; process requirement information such as target processing method, process precision, processing path, or task type; and constraint information such as production cycle, cost limits, resource availability, and safety requirements. When the requirement information is a 3D model, the 3D model is parsed and processed, and the shape features, dimensional data, and material labels of the 3D model are analyzed to obtain the category and parameter information related to the target product.

[0073] Step S102: If the above-mentioned demand information is text information, process the above-mentioned text information to obtain the above-mentioned target product information.

[0074] Specifically, the aforementioned target product information is a data set of target product categories and parameters.

[0075] Step S103: Based on the product knowledge graph, the equipment knowledge graph, and the aforementioned target product information, determine the workstation information for processing the target product. The workstation information includes information on multiple devices for processing the target product. The product knowledge graph includes multiple products, and the equipment knowledge graph includes multiple candidate devices.

[0076] Specifically, the product knowledge graph describes the classification relationships and attribute information between different products, such as product category nodes like "pressure vessel," "ship component," and "construction machinery," along with their corresponding attribute parameters. By retrieving product nodes corresponding to the target product information from the product knowledge graph, a standardized representation of the target product can be determined. The aforementioned equipment knowledge graph describes the categories, performance parameters, and relationships with product categories of candidate equipment, such as equipment nodes like "automatic welding machine," "cutting machine," and "assembly robot." By retrieving equipment nodes associated with the target product node from the equipment knowledge graph, information on multiple candidate devices can be obtained.

[0077] In one embodiment of this application, when the above-mentioned demand information is text information, the above-mentioned text information is processed to obtain the above-mentioned target product information, including: processing the above-mentioned text information using a pre-trained neural network model to extract key information from the above-mentioned demand information, the above-mentioned key information including at least the category and parameters of the above-mentioned target product; filling a preset template according to the above-mentioned key information to obtain the above-mentioned target product information.

[0078] In one embodiment of this application, after determining the workstation information for processing the target product based on the product knowledge graph, the equipment knowledge graph, and the aforementioned target product information, the method further includes: generating simulation parameters based on the workstation information, and constructing a workstation simulation model based on the simulation parameters; running the workstation simulation model to obtain simulation results, the simulation results including the operating status data of the equipment, the size data of the target product, and the efficiency data in the workstation information; scoring the simulation results according to a preset scoring rule to obtain a score for the simulation results, and obtaining decision information based on the score, the decision information including selection result information of the equipment used to process the target product.

[0079] In one embodiment of this application, after obtaining decision information based on the above score, the method further includes: outputting the decision information upon receiving confirmation of the decision information; and correcting the decision information based on the correction information upon receiving correction information for the decision information.

[0080] In one embodiment of this application, the device knowledge graph includes fixed relationship nodes and non-fixed relationship nodes. The fixed relationship nodes are nodes that correspond one-to-one with the target product and the equipment used to process the target product. The non-fixed relationship nodes include a preset function for determining multiple equipment used to process the target product. Based on a preset product family, a preset equipment family, and the target product information, the function determines workstation information for processing the target product. This includes: traversing the device knowledge graph; when encountering a fixed relationship node, determining the set of the target product information and the equipment information for processing the target product as the workstation information; when encountering a non-fixed relationship node, determining the matching value between the target product information and the multiple equipment used to process the target product according to the preset function corresponding to the non-fixed relationship node, and generating workstation information including the matching value.

[0081] The devices mentioned in this article can be servers, PCs, tablets, mobile phones, etc.

[0082] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0083] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0084] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0085] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0086] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process.Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0087] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0088] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, like read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0089] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0090] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0091] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0092] As can be seen from the above description, the embodiments of this application achieve the following technical effects:

[0093] 1) The workstation generation method of this application parses different types of requirement information and automatically extracts the category, geometric parameters, material, and process requirements of the target product. This avoids the inefficient method of relying on manual reading of drawings or instructions to extract information in existing technologies, improving the accuracy and consistency of information acquisition. Through product knowledge graph and equipment knowledge graph retrieval, the category of the target product can be quickly located, and multiple candidate devices corresponding to the product can be found in the equipment knowledge graph, thereby generating workstation information. This avoids the tedious process of manually comparing product parameters with equipment parameters one by one, improving the efficiency of equipment selection and workstation configuration. The generated workstation information includes information on multiple devices used to process the target product, covering different processes and technological steps required for the production of the target product. This ensures that the configured workstations can fully support the manufacturing process of the target product. This solution eliminates the need for manual comparison of target product parameters with candidate device parameters. It can directly generate standardized target product information through automatic parsing of 3D models or text information. Furthermore, by combining product knowledge graphs and device knowledge graphs, it enables retrieval and matching, thereby automatically determining the workstation information used to process the target product. This solves the problem in existing technologies where equipment selection for workstations requires manual matching of devices based on the parameters of the target product, resulting in low equipment selection efficiency.

[0094] 2) The workstation generation system of this application features an intelligent unit that simultaneously supports both 3D models and text information as input. When the requirement information is a 3D model, it automatically parses information such as geometric dimensions, structural features, and material properties. The intelligent unit processes different input sources uniformly, ultimately outputting structured target product information, which can be directly used as input for knowledge graph retrieval. This avoids inconsistencies and error rates associated with manual parameter extraction, thereby improving the accuracy and standardization of information parsing. Through the retrieval unit, based on preset product and equipment knowledge graphs, it quickly matches candidate equipment related to the target product and combines them into workstation information, improving the efficiency of equipment selection and workstation generation. The interaction unit can output workstation information to users or upper-level systems, making the generated results visual.

[0095] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A workstation generation method, characterized in that, include: Receive demand information; if the demand information is a three-dimensional model, process the three-dimensional model to obtain target product information; the demand information includes the category and parameter information of the target product. If the demand information is text information, the text information is processed to obtain the target product information; Based on the product knowledge graph, the equipment knowledge graph, and the target product information, workstation information for processing the target product is determined. The workstation information includes information on multiple devices for processing the target product. The product knowledge graph includes multiple products, and the equipment knowledge graph includes multiple candidate devices.

2. The method according to claim 1, characterized in that, When the demand information is text information, the text information is processed to obtain the target product information, including: The text information is processed using a pre-trained neural network model to extract key information from the demand information, including at least the category and parameters of the target product. The target product information is obtained by filling the preset template with the key information.

3. The method according to claim 1, characterized in that, After determining the workstation information for processing the target product based on the product knowledge graph, the equipment knowledge graph, and the target product information, the method further includes: Simulation parameters are generated based on the workstation information, and a workstation simulation model is constructed based on the simulation parameters. Run the workstation simulation model to obtain simulation results, which include the operating status data of the equipment, the size data of the target product, and the efficiency data in the workstation information. The simulation results are scored according to a preset scoring rule to obtain a score for the simulation results, and decision information is obtained based on the score. The decision information includes selection result information of the equipment used to process the target product.

4. The method according to claim 3, characterized in that, After obtaining decision information based on the scores, the method further includes: Upon receiving confirmation of the decision information, the decision information is output. Upon receiving correction information for the decision information, the decision information is corrected according to the correction information.

5. The method according to claim 1, characterized in that, The device knowledge graph includes fixed relationship nodes and non-fixed relationship nodes. The fixed relationship nodes are nodes that correspond one-to-one between the target product and the equipment used to process the target product. The non-fixed relationship nodes include preset functions used to determine multiple pieces of equipment used to process the target product, and to determine workstation information used to process the target product based on preset product families, preset equipment families, and the target product information, including: Traverse the device knowledge graph, and when the fixed relationship node is reached, determine the set of the target product information and the device information used to process the target product as the workstation information; When the non-fixed relationship node is encountered, the matching value between the target product information and the multiple devices used to process the target product is determined according to the preset function corresponding to the non-fixed relationship node, and the workstation information including the matching value is generated.

6. A workstation generation system, characterized in that, include: The intelligent unit is used to receive demand information, process the three-dimensional model to obtain target product information when the demand information is a three-dimensional model, and process the text information to obtain the target product information when the demand information is text information. The retrieval unit is used to determine the workstation information for processing the target product based on a preset knowledge graph and the target product information; An interactive unit is used to output the workstation information.

7. The system according to claim 6, characterized in that, The intelligent unit includes a demand analysis agent, a reasoning and decision-making agent, a simulation agent, and a human teaching agent. The demand analysis agent is used to extract the demand information to obtain target product information. The reasoning and decision-making agent is used to generate workstation information. The simulation agent is used to generate simulation parameters. The human teaching agent is used to correct the workstation information based on the received correction information.

8. The system according to claim 6, characterized in that, The system also includes: The verification unit includes a simulation module and a decision module. The simulation module is used to perform simulation verification on the simulation parameters sent by the intelligent unit to obtain simulation results. The decision module is used to generate decision information based on the simulation results.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method according to any one of claims 1 to 5.

10. An electronic device, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising methods for performing any one of claims 1 to 5.

Images