Steel rolling defect tracing method and device and storage medium

Abstract

The embodiment of the invention provides a steel rolling defect tracing method and device and a storage medium, and the method comprises the steps: obtaining the production process data, inspection information and after-sales data of a target steel rolling, and determining data nodes which comprise a production node, an inspection node and an objection node; determining a node association relationship between the two data nodes according to the data semantic features and the key identifiers of the data nodes; determining a node association type of the node association relationship according to node data timeliness and node process correlation; determining a graph structure data model according to the data nodes, the node association relationship and the node association type; and performing defect tracing on the target steel rolling according to the graph structure data model. According to the method, the steel rolling defect tracing efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of iron and steel smelting, and more specifically to a method, apparatus and storage medium for tracing defects in rolled steel. Background Technology

[0002] The entire iron and steel smelting process, from ironmaking to steelmaking and rolling, is closely interconnected. Differences in raw materials, process fluctuations, or operational deviations at any stage can lead to product defects. Therefore, accurately locating the causes of defects through full-process production traceability and quality analysis is crucial for improving product quality, optimizing processes, and reducing costs, and is of great significance for enhancing enterprise competitiveness.

[0003] The core of this type of analysis is the integration of multi-source data, including raw material procurement information from the ERP system, testing and analysis results from the quality inspection system, process parameters from the production execution system, and product composition data at each stage. By statistically correlating these data, common characteristics of different production stages and product quality can be summarized, providing data support for process optimization and quality control.

[0004] Currently, the industry commonly employs BI reports and manual analysis. The specific process involves professionals and developers collaborating to determine the wide table structure, developers completing data integration and report functionality development, and finally, analysts manually filtering data, creating charts, and summarizing common defect factors based on experience. However, with increasingly complex production processes, diversified products, and higher quality requirements, the inherent shortcomings of this approach are becoming increasingly apparent, failing to meet the needs for accurate end-to-end traceability and efficient defect localization. Specifically: 1. High relationship maintenance costs: It relies on preset wide tables and fixed reports. Adding new analysis scenarios requires redesign and development, resulting in slow response and high costs.

[0005] 2. Lack of dynamic transmission: It is impossible to automatically transmit quality problems in previous processes (such as excessive sulfur content in ironmaking raw materials) to subsequent processes, making it difficult to capture the impact of cross-process delays.

[0006] 3. Over-reliance on manual intervention: The analysis process requires manual triggering, and the conclusions rely on experience-based judgment, resulting in low efficiency, unstable accuracy, and significant interference from objective factors.

[0007] 4. Hidden relationships are difficult to uncover: It is impossible to discover non-explicit relationships across processes (such as the effect of BOF tapping temperature on billet inclusions), and new influencing factors need to be redeveloped before they can be analyzed.

[0008] In summary, existing solutions suffer from problems such as unstable accuracy, high maintenance costs, and low efficiency, making them difficult to match the industry's needs for high-quality development and intelligent transformation. Summary of the Invention

[0009] The purpose of this invention is to provide a method, apparatus, and storage medium for tracing defects in steel rolling mills, which improves the efficiency of tracing defects in steel rolling mills.

[0010] To achieve the above objectives, embodiments of the present invention provide a method for tracing defects in steel rolling, the method comprising: The production process data, inspection information and after-sales data of the target steel roll are obtained to determine data nodes, including production nodes, inspection nodes and objection nodes; The node association relationship between two data nodes is determined based on the data semantic features and key identifiers of the data nodes. The node association type is determined based on the timeliness of node data and the relevance of node processes. The graph structure data model is determined based on the data nodes, node relationships, and node relationship types. Defect tracing is performed on the target steel roll based on the graph structure data model.

[0011] Optionally, the key identifiers include the furnace number that runs through steelmaking, rolling, billet to finished product, the iron number that connects ironmaking and steelmaking, the finished product number that connects finished product and after-sales quality objection, and the billet number that connects billet and steel billet. The node association relationships include furnace number association, iron number association, finished product number association, and billet number association.

[0012] Optionally, the node association types include production relationships, inspection relationships, objection relationships, and cross-node relationships; The node association type, which determines the node association relationship based on the timeliness of node data and the relevance of node processes, includes: If there is an upstream and downstream production flow relationship between two data nodes, then the association type between the two data nodes is a production relationship; If there is a correspondence between the test results and the corresponding production batches between two data nodes, then the association type between the two data nodes is a test relationship. If there is a correspondence between customer quality complaints and specific products between two data nodes, then the association type between the two data nodes is a dispute relationship. If two data nodes are associated across nodes through numbering, then the association type between the two data nodes is a cross-node relationship.

[0013] Optionally, the graph structure data model includes a forward tracing path and a reverse tracing path; The forward traceability path is the complete chain from raw materials to finished products according to the production process; The reverse tracing path is a complete link that traces back from the quality objection to the source of the problem.

[0014] Optionally, the step of tracing defects in the target steel roll based on the graph structure data model includes: The alarm data of the target steel roll is detected, and the alarm data includes the defect type, location, and severity. The alarm data is input into the graph structure data model to obtain its corresponding data nodes. Based on the data nodes, the defect tracing results are determined, including defect node clusters, defect levels, recurrence status, and distribution characteristics.

[0015] Optionally, the method further includes: determining the scheduling cycle, scheduling time and triggering conditions based on the defect tracing results, wherein the comprehensive weight value of the defect node cluster is calculated using the analytic hierarchy process (AHP), and the scheduling cycle is determined based on the basic scheduling cycle and the comprehensive weight value.

[0016] Optionally, the production node, inspection node, and objection node are each provided with a node type, node ID, and node name field.

[0017] On the other hand, this application also proposes a device for tracing defects in steel rolling, the device comprising: The first processing module is used to acquire production process data, inspection information and after-sales data of the target steel roll, and to determine data nodes, including production nodes, inspection nodes and objection nodes. The second processing module is used to determine the node association relationship between two data nodes based on the data semantic features and key identifiers of the data nodes; The third processing module is used to determine the node association type of the node association relationship based on the timeliness of node data and the relevance of node process. The fourth processing module is used to determine the graph structure data model based on the data nodes, node association relationships and node association types, and to perform defect tracing on the target steel roll based on the graph structure data model.

[0018] Optionally, the key identifiers include the furnace number that runs through steelmaking, rolling, billet to finished product, the iron number that connects ironmaking and steelmaking, the finished product number that connects finished product and after-sales quality disputes, and the billet number that connects billet and steel billet. The node association relationships include furnace number association, iron number association, finished product number association, and billet number association.

[0019] On the other hand, this application also proposes a machine-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the steel rolling defect tracing method described above.

[0020] A method for tracing defects in steel rolling mills according to the present invention includes: acquiring production process data, inspection information, and after-sales data of the target steel rolling mill to determine data nodes, including production nodes, inspection nodes, and dispute nodes; determining the node association relationship between two data nodes based on the data semantic features and key identifiers of the data nodes; determining the node association type of the node association relationship based on the timeliness of the node data and the relevance of the node to the process; determining a graph structure data model based on the data nodes, node association relationships, and node association types; and tracing defects in the target steel rolling mill based on the graph structure data model. This method uses relational modeling in a graph database, enabling rapid tracing of the root causes of quality problems, reducing the traditional problem investigation time of several hours to minutes, and achieving automatic identification of the correlation between equipment failures, abnormal process parameters, and product quality defects.

[0021] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0022] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating a method for tracing defects in steel rolling according to the present invention; Figure 2 This is a schematic diagram of one embodiment of the present invention; Figure 3 This is a schematic diagram of another embodiment of the present invention; Figure 4 This invention provides a steel production quality traceability diagram. Figure 5 This is a schematic diagram of a steel rolling defect tracing device according to the present invention.

[0023] Explanation of reference numerals in the attached figures 100 - A device for tracing defects in steel rolling mills; 200 - First Processing Module; 300 - Second processing module; 400 - Third Processing Module; 500 - Fourth processing module. Detailed Implementation

[0024] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0025] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0026] Example 1 Figure 1 This is a flowchart illustrating a method for tracing defects in steel rolling according to the present invention, as shown below. Figure 1 As shown, a method for tracing defects in steel rolling according to the present invention includes: Step S101 involves obtaining production process data, inspection information, and after-sales data of the target steel roll to determine data nodes, including production nodes, inspection nodes, and objection nodes.

[0027] Specifically, the production process data spans the entire steel rolling process from raw materials to finished products, and is used to trace the production process and optimize process parameters. Specifically, it includes: raw material data (including basic raw material information and raw material pretreatment data), pre-rolling process data (including steelmaking process data and heating process data), and core rolling process data (including equipment parameters, process execution data, and process material data).

[0028] The inspection information includes incoming material inspection data, raw material physicochemical test reports, raw material conformity judgment, process inspection data (including inter-process dimensional inspection data, process performance sampling data, and surface defect online detection data), and finished product inspection data (including dimensional accuracy inspection data, physicochemical performance inspection data, final surface quality inspection data, and inspection conclusions and judgment data).

[0029] The after-sales data includes delivery logistics data, shipping information, acceptance data, customer usage data (including application scenario data and performance feedback data), after-sales issue data (including quality objection records and after-sales processing data), and product traceability and recall data (including quality issue traceability records and recall data).

[0030] Production nodes are determined based on the production process data, inspection nodes are determined based on the inspection information, and objection nodes are determined based on the after-sales data. Each production node, inspection node, and objection node has a node type, node ID, and node name field.

[0031] The production nodes correspond to key processes in steel rolling production. Node IDs are assigned according to the process type + sequence number rule for easy traceability. For example, if the node type is raw material pretreatment, the node ID is SC-001, and the node name is scrap steel crushing and mixing. Inspection nodes correspond to key inspection stages in incoming materials, in-process, and finished products. Node IDs are assigned according to the inspection stage + sequence number rule to distinguish different inspection types. For example, if the node type is incoming material inspection, the node ID is IN-001, and the node name is scrap steel composition spectral analysis. Objection nodes correspond to key handling stages for after-sales quality issues. Node IDs are assigned according to the objection type + sequence number rule, covering the entire process from issue feedback to closed-loop management. For example, if the node type is customer feedback, the node ID is CO-001, and the node name is customer objection upon arrival and acceptance.

[0032] According to a specific implementation method, the production nodes include lime kiln, sintering, ironing, furnace cycles, billet rolling, steel billet, and finished product nodes; the inspection nodes include lime kiln inspection, sintering inspection, analysis of raw materials entering the furnace for ironing, iron composition inspection, molten steel inspection, billet rolling inspection, and steel billet inspection nodes; and the objection nodes include after-sales quality objections.

[0033] Step S102 is to determine the node association relationship between two data nodes based on the data semantic features and key identifiers of the data nodes.

[0034] The key identifiers include furnace numbers that run through steelmaking, rolling, billet production, and finished products; iron batch numbers that connect ironmaking and steelmaking; finished product numbers that connect finished products and those with after-sales quality disputes; and billet numbers that connect rolled billets and steel billets. The node relationships include furnace number associations, iron batch number associations, finished product number associations, and billet number associations. Specifically, the furnace number association includes core identifiers that run through steelmaking → rolling → billet → finished products; the iron batch number association includes iron batch traceability connecting ironmaking → steelmaking; the finished product number association includes traceability for quality disputes connecting finished products to after-sales services; and the billet number association includes billet traceability connecting rolled billets to steel billets.

[0035] The production node, inspection node, and objection node can achieve full-process traceability through associated fields (such as batch number and product ID). For example, the production node SC-003 (continuous casting) can be associated with the inspection node IP-001 (slab flaw detection), and finally associated with the objection node CO-002 (batch traceability).

[0036] Step S103 is to determine the node association type based on the timeliness of node data and the relevance of node processes.

[0037] The node association types include production relationships, inspection relationships, objection relationships, and cross-node relationships.

[0038] The production relationship is used to represent the production flow of upstream and downstream processes; the inspection relationship is used to associate quality inspection results with corresponding production batches; the objection relationship is used to connect customer quality complaints with specific products; and the cross-node relationship is used to realize cross-node data association analysis application scenarios through numbering. The node association types also include quality traceability relationship and path relationship. The quality traceability relationship is used to quickly locate the entire production process based on the finished product number and analyze the root cause of quality problems. The path relationship is used to support shortest path search and complete production link analysis.

[0039] The node association type, determined based on the timeliness of node data and the relevance of node processes, includes: if there is an upstream and downstream production flow relationship between two data nodes, the association type between the two data nodes is a production relationship; if there is a correspondence between inspection results and corresponding production batches between two data nodes, the association type between the two data nodes is an inspection relationship; if there is a correspondence between customer quality complaints and specific products between two data nodes, the association type between the two data nodes is an objection relationship; if two data nodes achieve cross-node data association through numbering, the association type between the two data nodes is a cross-node relationship.

[0040] Step S104 is to determine the graph structure data model based on the data nodes, node association relationships, and node association types.

[0041] Based on the production nodes, inspection nodes, and objection nodes in steel rolling production, as well as the relationships and types between nodes, a graph data model is constructed. This model achieves full-process data visualization and traceability through a three-layer structure of nodes, edges, and attributes. In the graph data model, nodes carry entity information, edges carry relationships, and attributes supplement the characteristics of entities and relationships.

[0042] Specifically, such as Figure 4 As shown, the steps for constructing a graph-structured data model include: cleaning raw production, inspection, and after-sales data; extracting nodes (including node type, node ID field, and node name field); standardizing the format of key association fields; creating three types of nodes in batches in the graph database based on the preprocessed data; establishing directed edges between nodes according to association rules; verifying the integrity of the core traceability link; if the association relationship of a certain type of node is too complex, the node can be split to improve query efficiency; when a new production batch is added, the corresponding production node is automatically created and the association edge is executed; after the inspection task is completed, the corresponding production node is automatically associated, and an inspection association edge is generated; when a quality objection is reported, a traceability query is automatically triggered, a traceability association edge is generated, and a rectification association edge is added after rectification is completed.

[0043] The graph-structured data model includes forward tracing paths and reverse tracing paths. The forward tracing path is the complete link from raw materials to finished products according to the production process. The reverse tracing path is the complete link that traces back to the source of the problem from a quality objection. For example, starting from the "customer objection report" node, a reverse query along the traceability edges should be able to accurately locate the corresponding nodes such as "finished product inspection," "finish rolling," and "slab heating." like Figure 3 As shown, specific graph analysis tasks are created based on the data nodes, node relationships, and node relationship types. Users can select different task types, including graph statistics, path analysis, community discovery, centrality analysis, or custom queries. Depending on the selected task type, the graph structure data model will automatically generate or allow users to edit Cypher query statements, and query parameters can be set. After configuration, the analysis task is executed, analysis results are obtained, and task execution status is monitored through the task status management function.

[0044] The attributes of nodes and node relationships in this method can be dynamically expanded, and new fields can be added without modifying the table structure, which can directly meet the business requirements of intelligent manufacturing and quality control.

[0045] Step S105 involves tracing defects in the target steel roll based on the graph structure data model.

[0046] The defect tracing of the target steel roll based on the graph structure data model includes: detecting alarm data of the target steel roll, wherein the alarm data includes defect type, location of occurrence and severity; inputting the alarm data into the graph structure data model to obtain its corresponding data node; and determining the defect tracing result based on the data node, including defect node cluster, defect level, recurrence status and distribution characteristics.

[0047] Based on the graph structure data model, graph statistical analysis, path analysis, community detection, centrality analysis, and custom queries are performed on the target steel roll to obtain the final quality influencing factors. Specifically, path analysis involves finding and analyzing paths between nodes in the graph, including complex path calculation tasks such as shortest path, all paths, and path length constraints. Community detection identifies community structures in the graph, discovering tightly connected node groups for clustering and community partitioning analysis. Centrality analysis calculates the importance indicators of nodes in the graph, including various centrality measures such as degree centrality, betweenness centrality, and proximity centrality. Custom queries support user-defined Cypher query statements to execute specific graph analysis needs, providing maximum flexibility. Each existing modeled node and its relationships are further linked according to different algorithms to obtain the final quality influencing factors.

[0048] The method further includes: determining the scheduling cycle, scheduling time and triggering conditions based on the defect tracing results, wherein the comprehensive weight value of the defect node cluster is calculated using the analytic hierarchy process (AHP), and the scheduling cycle is determined based on the basic scheduling cycle and the comprehensive weight value.

[0049] Specifically, starting from the objection node in the graph structure model, the associated inspection and production nodes are located in reverse along the traceability edges to form a defect node cluster (including production nodes and inspection nodes that directly cause quality problems, as well as upstream and downstream nodes that indirectly affect the generation of defects). The comprehensive weight of the defect node cluster is calculated using the Analytic Hierarchy Process (AHP) to determine the scheduling cycle, scheduling time, and triggering conditions.

[0050] Figure 2 This is a schematic diagram of one embodiment of the present invention, as shown below. Figure 2 As shown in the diagram, this is a specific implementation of the present application. The application proposes a layered technical architecture for node relationship management, encompassing data configuration, relationship management, and information flow. This architecture includes a front-end application layer, middleware layer, gateway layer, service layer, and data storage layer. The front-end application layer serves as the operation entry point, providing message listening configuration, data source configuration, relational configuration, batch processing configuration, and relationship querying, covering the entire process from configuration to query. The middleware layer provides basic technical support (Kafka / RabbitMQ: responsible for asynchronous message communication and queue management, achieving service decoupling; Redis: caching high-frequency data, improving query performance; Nacos: providing service registration and configuration management, achieving dynamic service scheduling). The gateway layer serves as the communication entry point between the front-end and back-end, undertaking basic gateway responsibilities such as request forwarding and access control. The service layer includes message listening services, data source services, graph data services, and batch processing services. The data storage layer includes relational storage and graph storage. This architecture achieves full-link management of node relationships through front-end configuration, service processing, and graph storage, while leveraging middleware components to ensure the system's asynchronous communication, caching, and service governance capabilities.

[0051] This method uses relational modeling with a graph database to quickly trace the root causes of quality problems, reducing troubleshooting time from hours to minutes. It automatically identifies the correlation between equipment malfunctions, abnormal process parameters, and product quality defects. Compared to traditional wide-table analysis, it can uncover hidden data patterns, helping quality engineers identify previously undiscovered influencing factors and reducing system maintenance costs.

[0052] Example 2 Figure 5 This is a schematic diagram of a steel rolling defect tracing device according to the present invention, as shown below. Figure 5As shown, the steel rolling defect tracing device 100 includes: a first processing module 200, used to acquire production process data, inspection information, and after-sales data of the target steel rolling, and to determine data nodes, including production nodes, inspection nodes, and objection nodes; a second processing module 300, used to determine the node association relationship between two data nodes based on the data semantic features and key identifiers of the data nodes; a third processing module 400, used to determine the node association type of the node association relationship based on the timeliness of the node data and the relevance of the node process; and a fourth processing module 500, used to determine a graph structure data model based on the data nodes, node association relationships, and node association types, and to perform defect tracing of the target steel rolling based on the graph structure data model.

[0053] The key identifiers include the furnace number that runs through steelmaking, rolling, billet production to finished product, the iron number that connects ironmaking and steelmaking, the finished product number that connects finished product and after-sales quality disputes, and the billet number that connects billet and steel billet; the node association relationships include furnace number association, iron number association, finished product number association, and billet number association.

[0054] This device, by adapting to the characteristics of multi-dimensional entity associations, clearly expresses the complex relationships between raw materials, equipment, process parameters, and product quality. It also constructs a time-series relationship graph and supports multi-level associations, thus achieving dynamic tracking and penetrating analysis across the entire steel rolling production chain, laying a solid foundation for quality traceability. Furthermore, by optimizing the path query algorithm and avoiding the cumbersome logic of traditional multi-table associations, while supporting real-time relationship traversal and pattern matching, it enables rapid location of quality problem propagation paths, significantly reducing query latency and meeting the core needs of real-time quality monitoring in the production process. Through graph traversal algorithms, it traces related links, analyzes the scope of impact, and uncovers association patterns, thereby achieving precise location of the root causes of quality problems, rapid assessment of affected batches, and early identification of potential quality risk patterns, facilitating proactive prevention and control.

[0055] A method for tracing defects in steel rolling mills according to the present invention includes: acquiring production process data, inspection information, and after-sales data of the target steel rolling mill to determine data nodes, including production nodes, inspection nodes, and dispute nodes; determining the node association relationship between two data nodes based on the data semantic features and key identifiers of the data nodes; determining the node association type of the node association relationship based on the timeliness of the node data and the relevance of the node to the process; determining a graph structure data model based on the data nodes, node association relationships, and node association types; and tracing defects in the target steel rolling mill based on the graph structure data model. This method uses relational modeling in a graph database, enabling rapid tracing of the root causes of quality problems, reducing the traditional problem investigation time of several hours to minutes, and achieving automatic identification of the correlation between equipment failures, abnormal process parameters, and product quality defects.

[0056] The steel rolling defect tracing device 100 includes a processor and a memory. The first processing module 200, the second processing module 300, the third processing module 400, the fourth processing module 500, etc. are all stored in the memory as program units. The processor executes the program units stored in the memory to realize the corresponding functions.

[0057] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and adjusting kernel parameters can improve the efficiency of steel rolling defect tracing.

[0058] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0059] This invention provides a storage medium storing a program that, when executed by a processor, implements the method for tracing steel rolling defects.

[0060] This invention provides a processor for running a program, wherein the program executes the method for tracing steel rolling defects.

[0061] This invention provides a device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: acquiring production process data, inspection information, and after-sales data of a target steel roll to determine data nodes, including production nodes, inspection nodes, and objection nodes; determining the node association relationship between two data nodes based on the data semantic features and key identifiers of the data nodes; determining the node association type of the node association relationship based on the timeliness of the node data and the relevance of the node process; determining a graph structure data model based on the data nodes, node association relationships, and node association types; and performing defect tracing on the target steel roll based on the graph structure data model. The device described herein can be a server, PC, PAD, mobile phone, etc.

[0062] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program with the following method steps: acquiring production process data, inspection information, and after-sales data of the target steel roll to determine data nodes, the data nodes including production nodes, inspection nodes, and objection nodes; determining the node association relationship between two data nodes based on the data semantic features and key identifiers of the data nodes; determining the node association type of the node association relationship based on the timeliness of the node data and the relevance of the node process; determining a graph structure data model based on the data nodes, node association relationships, and node association types; and performing defect tracing on the target steel roll based on the graph structure data model.

[0063] 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.

[0064] 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.

[0065] 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.

[0066] 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.

[0067] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0068] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0069] 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.

[0070] 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.

[0071] The above are merely embodiments of this application and are not intended to limit the scope of 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 scope of the claims of this application.

Claims

1. A method for tracing defects in steel rolling, characterized in that, The method includes: The production process data, inspection information and after-sales data of the target steel roll are obtained to determine data nodes, including production nodes, inspection nodes and objection nodes; The node association relationship between two data nodes is determined based on the data semantic features and key identifiers of the data nodes. The node association type is determined based on the timeliness of node data and the relevance of node processes. The graph structure data model is determined based on the data nodes, node relationships, and node relationship types. Defect tracing is performed on the target steel roll based on the graph structure data model.

2. The method according to claim 1, characterized in that, The key identifiers include the furnace number that runs through steelmaking, rolling, billet production to finished products, the iron number that connects ironmaking and steelmaking, the finished product number that connects finished products and after-sales quality disputes, and the billet number that connects rolled billets and steel billets. The node association relationships include furnace number association, iron number association, finished product number association, and billet number association.

3. The method according to claim 1, characterized in that, The node association types include production relationships, inspection relationships, objection relationships, and cross-node relationships; The node association type, which determines the node association relationship based on the timeliness of node data and the relevance of node processes, includes: If there is an upstream and downstream production flow relationship between two data nodes, then the association type between the two data nodes is a production relationship; If there is a correspondence between the test results and the corresponding production batches between two data nodes, then the association type between the two data nodes is a test relationship. If there is a correspondence between customer quality complaints and specific products between two data nodes, then the association type between the two data nodes is a dispute relationship. If two data nodes are associated across nodes through numbering, then the association type between the two data nodes is a cross-node relationship.

4. The method according to claim 1, characterized in that, The graph structure data model includes forward tracing paths and reverse tracing paths; The forward traceability path is the complete chain from raw materials to finished products according to the production process; The reverse tracing path is a complete link that traces back from the quality objection to the source of the problem.

5. The method according to claim 1, characterized in that, The defect tracing of the target steel roll based on the graph structure data model includes: The alarm data of the target steel roll is detected, and the alarm data includes the defect type, location, and severity. The alarm data is input into the graph structure data model to obtain its corresponding data nodes. Based on the data nodes, the defect tracing results are determined, including defect node clusters, defect levels, recurrence status, and distribution characteristics.

6. The method according to claim 5, characterized in that, The method also includes: The scheduling cycle, scheduling time, and triggering conditions are determined based on the defect tracing results. The comprehensive weight value of the defect node cluster is calculated using the analytic hierarchy process (AHP), and the scheduling cycle is determined based on the basic scheduling cycle and the comprehensive weight value.

7. The method according to claim 1, characterized in that, The production node, inspection node, and objection node all have a node type, node ID, and node name field.

8. A device for tracing defects in steel rolling, characterized in that, The device includes: The first processing module is used to acquire production process data, inspection information and after-sales data of the target steel roll, and to determine data nodes, including production nodes, inspection nodes and objection nodes. The second processing module is used to determine the node association relationship between two data nodes based on the data semantic features and key identifiers of the data nodes; The third processing module is used to determine the node association type of the node association relationship based on the timeliness of node data and the relevance of node process. The fourth processing module is used to determine the graph structure data model based on the data nodes, node association relationships and node association types, and to perform defect tracing on the target steel roll based on the graph structure data model.

9. The apparatus according to claim 8, characterized in that, The key identifiers include the furnace number that runs through steelmaking, rolling, billet production to finished products, the iron number that connects ironmaking and steelmaking, the finished product number that connects finished products and after-sales quality disputes, and the billet number that connects rolled billets and steel billets. The node association relationships include furnace number association, iron number association, finished product number association, and billet number association.

10. A machine-readable storage medium storing instructions thereon, characterized in that, When executed by a processor, this instruction causes the processor to be configured to perform the method for tracing steel rolling defects according to any one of claims 1 to 7.

Images