Steel coil surface defect prediction method based on knowledge graph and related equipment

Abstract

The invention discloses a steel coil surface defect prediction method based on a knowledge graph and related equipment, and relates to the technical field of defect detection.The method comprises the steps that surface problem prediction is conducted before steel coil production based on a prediction model and the knowledge graph of steel coil surface problems, and a first prediction result is obtained, the prediction model is a prediction model of the steel coil surface problem constructed based on the knowledge graph and the influence factor information of the steel coil surface problem; on the basis of the prediction model and the real-time process value, surface problem prediction in steel coil production is carried out, and a second prediction result is obtained; and calculating the prediction accuracy of the surface problem based on the first prediction result and the second prediction result.

Description

Technical Field

[0001] The present application relates to the field of defect detection technology, and in particular to a steel coil surface defect prediction method based on a knowledge graph and related equipment. Background Art

[0002] Galvanizing is an important surface treatment process for steel, which aims to improve the corrosion resistance, service life and appearance of steel. However, after galvanizing, surface problems such as dezincification may occur. At present, the detection of these surface problems mainly depends on subsequent processes, especially dezincification, which is usually detected through laboratory adhesive experiments. The detection process is not only laborious and resource-intensive, but also unable to achieve in-process prevention, and lacks an effective prediction method based on production processes.

[0003] The existing technology mainly uses manual methods to inspect the surface of steel coils. However, manual inspection cannot achieve pre-prevention. Inspection is often carried out after the steel coils are produced. Once the problem is discovered, losses have already occurred, and potential quality problems cannot be discovered in time during the production process. Secondly, manual inspection consumes a lot of human resources. Inspectors need to observe and judge one by one, which not only increases labor costs, but also prolongs the production cycle and reduces overall output capacity. In addition, when surface problems occur, manual inspection often cannot quickly locate the cause, resulting in slow problem solving, lack of effective rectification measures, and inability to avoid the recurrence of similar problems.

[0004] Therefore, there is an urgent need for a steel coil surface defect prediction method based on knowledge graph to improve detection efficiency and accuracy, achieve in-process prevention, reduce manpower investment, and quickly locate the cause of the problem. Summary of the invention

[0005] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further described in detail in the Detailed Description of the Invention section. The Summary of the Invention section of this application does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the scope of protection of the claimed technical solution.

[0006] In a first aspect, the present application provides a method for predicting surface defects of a steel coil based on a knowledge graph, comprising:

[0007] Based on the prediction model and the knowledge graph of the surface problems of the steel coil, the surface problems of the steel coil are predicted before production to obtain a first prediction result, wherein the prediction model is a prediction model of the surface problems of the steel coil constructed based on the knowledge graph and the information of influencing factors of the surface problems of the steel coil;

[0008] Based on the above prediction model and real-time process values, surface problems in steel coil production are predicted to obtain a second prediction result;

[0009] Based on the first prediction result and the second prediction result, the prediction accuracy of the surface problem is calculated.

[0010] In a feasible implementation manner, the specific steps of constructing the knowledge graph of the surface problems of the steel coils include:

[0011] Based on the production process, the ontology, attributes and relationships of the above production process are abstracted to form the knowledge graph rules of the surface problems of the steel coil, wherein the above production process includes galvanizing process plan, galvanizing plan, primary galvanizing and secondary galvanizing;

[0012] Based on the above knowledge graph rules, data mapping is performed, wherein the above data mapping includes ontology-attribute mapping and relationship mapping;

[0013] Based on the above data mapping, dynamically generate SQL statements to query data in the production data source, extract knowledge, and form a triple file;

[0014] Import the above triple files into the RDF graph database to form a knowledge graph of the surface problems of the steel coil.

[0015] In a feasible implementation manner, the specific steps of constructing the above prediction model include:

[0016] The SPARQL language is used to query the process performance data of normal steel coils and abnormal steel coils from the above knowledge graph, wherein the above abnormal steel coils are steel coils with surface problems;

[0017] Calculate the average values ​​of normal process performance data and abnormal process performance data of each production process parameter respectively to obtain the normal average value and the abnormal average value;

[0018] Based on the above normal average value and the above abnormal average value, determine the influencing factors of the surface problems of the steel coil and the positive and negative correlations of the above influencing factors;

[0019] Based on the above positive and negative correlations, the process value intervals of each of the above influencing factors are divided;

[0020] Calculate the weight of each of the above influencing factors based on the factor weight algorithm;

[0021] Based on the names, weights, standard rules and determination results of the above-mentioned influencing factors, influencing factor information of the surface problems of the steel coil is formed, wherein the above-mentioned determination results are determined based on the above-mentioned process value range and the above-mentioned standard rules;

[0022] The above impact factor information is entered into the prediction model.

[0023] In a feasible implementation manner, the influencing factors of the surface problems of the steel coil and the positive and negative correlations of the influencing factors are determined based on the normal average value and the abnormal average value, including:

[0024] Determine whether the above production process parameters are the above influencing factors;

[0025] In the case where the above production process parameters are the above influencing factors, determining the positive and negative correlation of the above influencing factors;

[0026] In a feasible implementation manner, the process value interval of each of the above-mentioned influencing factors is divided based on the above-mentioned positive and negative correlations, including:

[0027] In the case where the above-mentioned influencing factors are positively correlated, the unqualified interval is when the above-mentioned process value is less than the first preset value, the hidden danger interval is when the above-mentioned process value is greater than or equal to the first preset value and less than or equal to the second preset value, and the qualified interval is when the above-mentioned process value is greater than the second preset value;

[0028] When the above-mentioned influencing factors are negatively correlated, the unqualified interval is when the above-mentioned process value is greater than the third preset value, the hidden danger interval is when the above-mentioned process value is greater than or equal to the fourth preset value and less than or equal to the third preset value, and the qualified interval is when the above-mentioned process value is less than the fourth preset value.

[0029] In a feasible implementation, it also includes:

[0030] The process value of the influencing factor is adjusted based on the first prediction result and the second prediction result.

[0031] In a feasible implementation, it also includes:

[0032] The above prediction accuracy is determined based on the following formula:

[0033] Forecast accuracy = (total production quantity - the number of actual surface problems that did not appear but were not predicted - the number of surface problems that were predicted but were not actually found) / total production quantity;

[0034] Among them, the number of steel coils that actually had surface problems but were not predicted to have surface problems is the number of steel coils that actually had surface problems but were not identified by the first prediction result or the second prediction result; the number of steel coils that were predicted to have surface problems but did not actually have surface problems is the number of steel coils that were predicted to have surface problems by the first prediction result or the second prediction result but did not actually have surface problems.

[0035] In the second aspect, the present application proposes a device for predicting surface defects of a steel coil based on a knowledge graph, comprising:

[0036] A first prediction unit is used to predict the surface problems of the steel coil before production based on the prediction model and the knowledge graph of the surface problems of the steel coil, and obtain a first prediction result, wherein the above-mentioned prediction model is a prediction model of the surface problems of the steel coil constructed based on the above-mentioned knowledge graph and the information of influencing factors of the surface problems of the steel coil;

[0037] A second prediction unit is used to predict surface problems in steel coil production based on the prediction model and real-time process values ​​to obtain a second prediction result;

[0038] The statistical result unit is used to calculate the prediction accuracy of the surface problem based on the first prediction result and the second prediction result.

[0039] In a third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is used to implement the steps of a method for predicting surface defects of steel coils based on a knowledge graph as described in any one of claims 1 to 7 when executing the computer program stored in the memory.

[0040] In a fourth aspect, the present application further proposes a computer-readable storage medium on which a computer program is stored, characterized in that when the above-mentioned computer program is executed by a processor, a steel coil surface defect prediction method based on a knowledge graph as described in any one of claims 1 to 7 is implemented.

[0041] In summary, the embodiment of the present application can more accurately identify potential defect patterns by integrating historical data, production process and real-time monitoring information, analyzing multidimensional data, thereby improving the accuracy of defect prediction and reducing the rate of missed detection and misjudgment; by real-time monitoring of the state of the steel coil during the production process, the system can promptly detect abnormal conditions and warn of potential surface defects. This in-process prevention mechanism can effectively reduce later rework and losses and improve production efficiency; this solution can realize the automation of the detection process, reduce manual intervention, and thus save labor costs. By automatically identifying and analyzing surface defects, the subjective errors caused by manual detection are reduced, and the overall detection efficiency is improved; using the semantic relationship of the knowledge graph, the system can quickly analyze historical defect data, automatically locate the cause of the problem, and provide corresponding rectification suggestions. This feature can significantly shorten the problem solving time and improve the response speed; by constructing a knowledge graph, the system can realize the effective storage and sharing of defect knowledge. This not only promotes the accumulation of knowledge within the enterprise, but also provides a basis for subsequent technical improvements and process optimization; the prediction model can provide data support for management to help it make more scientific decisions. For example, during the production process, process parameters can be adjusted according to real-time data to further reduce the defect rate and improve product quality. By detecting and correcting surface defects at an early stage, the overall quality of the steel coil can be improved, the frequency and severity of defects can be reduced, and the rework and scrap rates can be reduced, thereby directly reducing production costs and improving the company's economic benefits.

[0042] The present invention proposes a method for predicting surface defects of steel coils based on a knowledge graph. Other advantages, objectives and features of the present invention will be partially reflected through the following description, and will also be partially understood by technicians in this field through research and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present specification. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0044] Figure 1 A schematic diagram of a method for predicting surface defects of a steel coil based on a knowledge graph provided in an embodiment of the present application;

[0045] Figure 2 A schematic diagram of the structure of a steel coil surface defect prediction device based on a knowledge graph provided in an embodiment of the present application;

[0046] Figure 3 A schematic diagram of a knowledge graph of impact factors provided in an embodiment of the present application;

[0047] Figure 4A schematic diagram of the structure of an electronic device for predicting surface defects of steel coils based on a knowledge graph provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0049] See also Figure 1 , which is a schematic flow chart of a method for predicting surface defects of a steel coil based on a knowledge graph provided in an embodiment of the present application, which may specifically include:

[0050] S110, based on the prediction model and the knowledge graph of the surface problems of the steel coil, predict the surface problems of the steel coil before production to obtain a first prediction result, wherein the prediction model is a prediction model of the surface problems of the steel coil constructed based on the knowledge graph and the information of influencing factors of the surface problems of the steel coil;

[0051] Exemplarily, data sources related to steel coil surface problems are collected, including historical production records, failure cases, process parameters, raw material characteristics, etc., and knowledge graph technology is used to associate information in different data sources to form a network of nodes (such as production processes, environmental factors, material characteristics) and edges (such as cause-effect relationships, degree of influence), and the main influencing factors of steel coil surface problems are identified, including but not limited to temperature, humidity, equipment status, operator experience, etc. These influencing factors are quantitatively and qualitatively analyzed to determine the degree of their influence on surface problems during the production process. The node and influencing factor information in the knowledge graph is used as feature input to predict the probability of occurrence of surface problems, and the constructed prediction model is used to input the process parameters before steel coil production to obtain a first prediction result indicating potential surface problems.

[0052] S120, predicting surface problems in steel coil production based on the prediction model and real-time process values ​​to obtain a second prediction result;

[0053] Exemplarily, during the steel coil production process, process values ​​such as temperature, pressure, speed, etc. are monitored in real time through sensors and data acquisition systems, and the real-time process values ​​are input into the constructed prediction model to perform dynamic surface problem prediction, and obtain a second prediction result, which indicates the surface problems occurring in the current production process.

[0054] S130: Calculate the prediction accuracy of the surface problem based on the first prediction result and the second prediction result.

[0055] Exemplarily, actual occurrences of the first prediction result and the second prediction result are collected, a data set is constructed, the results predicted by the model are compared with the actual observed results, and cases of correct and incorrect predictions are identified.

[0056] In summary, the embodiment of the present application can more accurately identify potential defect patterns by integrating historical data, production process and real-time monitoring information, analyzing multidimensional data, thereby improving the accuracy of defect prediction and reducing the rate of missed detection and misjudgment; by real-time monitoring of the state of the steel coil during the production process, the system can promptly detect abnormal conditions and warn of potential surface defects. This in-process prevention mechanism can effectively reduce later rework and losses and improve production efficiency; this solution can realize the automation of the detection process, reduce manual intervention, and thus save labor costs. By automatically identifying and analyzing surface defects, the subjective errors caused by manual detection are reduced, and the overall detection efficiency is improved; using the semantic relationship of the knowledge graph, the system can quickly analyze historical defect data, automatically locate the cause of the problem, and provide corresponding rectification suggestions. This feature can significantly shorten the problem solving time and improve the response speed; the prediction model can provide data support for management to help it make more scientific decisions. For example, in the production process, process parameters can be adjusted according to real-time data to further reduce the defect rate and improve product quality; by early detection and correction of surface defects, the overall quality of the steel coil can be improved and the frequency and severity of defects can be reduced.

[0057] This application is described in detail with the prediction of dezincification problem on the surface of high-strength steel:

[0058] In some examples, the specific steps of constructing the knowledge graph of the surface problems of the steel coils include:

[0059] Based on the production process, the ontology, attributes and relationships of the above production process are abstracted to form the knowledge graph rules of the surface problems of the steel coil, wherein the above production process includes galvanizing process plan, galvanizing plan, primary galvanizing and secondary galvanizing;

[0060] Based on the above knowledge graph rules, data mapping is performed, wherein the above data mapping includes ontology-attribute mapping and relationship mapping;

[0061] Based on the above data mapping, dynamically generate SQL statements to query data in the production data source, extract knowledge, and form a triple file;

[0062] Import the above triple files into the RDF graph database to form a knowledge graph of the surface problems of the steel coil.

[0063] Exemplarily, a surface knowledge graph is constructed, and according to the galvanizing production process of steel coils, the rules (Schema) of the surface knowledge graph are designed, which mainly include entities such as galvanizing DAS, galvanizing QDR hydrogen concentration, galvanizing QDR zinc liquid temperature, galvanizing QDR galvanizing process performance, galvanizing QDR pyrometer P7 temperature furnace nose, galvanizing process, galvanizing plan, etc., as well as the attributes of the entity. Some of the contents are shown in Table 1, where DAS and QDR represent different single-value tables stored during data collection, CGL1 is galvanizing line 1, OS and DS are the operating side and the driving side respectively, P7 and P9 represent the number of a collection point, GA is a zinc-iron alloy product, and GI is a pure zinc product.

[0064] Schema design refers to a set of rules used to define the data structure when building a knowledge graph, describing the entities, attributes, and relationships between them in the knowledge graph. Schema design is the metadata of the knowledge graph, which is used to define the format of the data to be added to the knowledge graph. Through Schema design, the consistency and readability of the knowledge graph are ensured, making subsequent data mapping, query, and reasoning more efficient.

[0065] In the prediction of surface problems of steel coils, Schema design involves abstracting the corresponding ontologies, relationships and attributes according to the actual production processes (such as galvanizing process plan, galvanizing plan, primary galvanizing and secondary galvanizing, etc.), thereby forming a surface knowledge graph Schema suitable for this application scenario, which helps to clearly define the various types of information contained in the knowledge graph and their relationships, and facilitates subsequent data fusion and utilization.

[0066] Table 1 Ontology attribute table

[0067]

[0068]

[0069] Table 2 Ontology attribute table

[0070] subject predicate object subject predicate object contract Contract production line Contract production line Galvanizing Plan Galvanizing plan hydrogen content Galvanizing plan hydrogen content Contract production line Contract galvanizing process Galvanizing process design Galvanizing Plan Galvanizing plan heating temperature Galvanizing plan heating temperature Galvanizing process design Hydrogen content in galvanizing process Hydrogen content in galvanizing process Contract production line Contract cold rolled materials Cold rolled materials Galvanizing process design Galvanizing process annealing speed Galvanizing process annealing speed Cold rolled materials Galvanized DAS performance Galvanized DAS single value table Contract production line Contract Galvanizing Program Galvanizing Plan Cold rolled materials Galvanizing QDR performance Galvanized QDR

[0071] After the Schema design is completed, it is connected to the production data source. The knowledge graph is used to build the platform to map the ontology, attributes, and relationships with the tables and fields in the data source. Data mapping is divided into four steps:

[0072] Connect to the production data source and select the table that needs to be mapped.

[0073] Map the business table where the ontology is located and select the unique primary key of the ontology.

[0074] Map the attributes of the ontology to the fields in the table and confirm the source of the attributes, including:

[0075] Ontology and attribute mapping: This process accurately maps the ontology and its attributes defined in the knowledge graph schema to the corresponding tables and fields in the actual production database. It requires careful matching of the table structure of the entity represented by each ontology in the database, while ensuring that the value of each attribute can be accurately extracted or stored from the database field. Data mapping not only ensures the accuracy of the data, but also promotes the accessibility and consistency of the data, laying a solid foundation for subsequent data analysis, query and knowledge reasoning.

[0076] The relationship between ontologies is mapped through the relationship between tables, including:

[0077] Relationship mapping, as a supplement to data mapping, focuses on parsing and expressing the relationships between tables in the database, which are usually implemented through association fields. In the knowledge graph, these relationships are abstracted as edges connecting different ontologies, thereby building a complex and rich knowledge network. Relationship mapping requires a deep understanding of the design logic of the database, accurately identifying which tables are associated, and how these associations affect the overall structure and meaning of the data. Through relationship mapping, we can reproduce the inherent logic and hierarchical structure of the data in the knowledge graph, providing strong support for complex data analysis and knowledge discovery.

[0078] After performing the data mapping in the above four steps, the knowledge graph construction platform will automatically generate SQL statements to query data in the data source, extract knowledge to form a triple NT file, and finally import the triple NT file into the RDF graph database to form a knowledge graph.

[0079] In some examples, the specific steps of constructing the above prediction model include:

[0080] The SPARQL language is used to query the process performance data of normal steel coils and abnormal steel coils from the above knowledge graph, wherein the above abnormal steel coils are steel coils with surface problems;

[0081] Calculate the average values ​​of normal process performance data and abnormal process performance data of each production process parameter respectively to obtain the normal average value and the abnormal average value;

[0082] Based on the above normal average value and the above abnormal average value, determine the influencing factors of the surface problems of the steel coil and the positive and negative correlations of the above influencing factors;

[0083] Based on the above positive and negative correlations, the process value intervals of each of the above influencing factors are divided;

[0084] Calculate the weight of each of the above influencing factors based on the factor weight algorithm;

[0085] Based on the names, weights, standard rules and determination results of the above-mentioned influencing factors, influencing factor information of the surface problems of the steel coil is formed, wherein the above-mentioned determination results are determined based on the above-mentioned process value range and the above-mentioned standard rules;

[0086] The above impact factor information is entered into the prediction model.

[0087] Exemplarily, the SPARQL language is used to query the process performance data of normal steel coils and abnormal steel coils respectively. The average values ​​of the queried normal and abnormal process performance data are calculated, and the average values ​​of each parameter in the query results are counted using tools such as Python. Using the normal average value and the abnormal average value, the influencing factor of each process parameter on the surface problems of the steel coil is analyzed. By calculating the difference between the normal and abnormal values, the positive and negative correlations are determined. For example, if the abnormal value of a parameter is higher than the normal value, it may be positively correlated with the surface problem. Based on the positive and negative correlations, the value ranges of the process parameters are divided to better understand the impact of each parameter on the quality of the steel coil. For example, a threshold can be set. When the parameter value exceeds a certain range, it is considered to cause surface problems. The factor weight algorithm is used to calculate the weight of each influencing factor. The weight of each factor reflects its relative importance to the surface problems of the steel coil. Expert experience further accurately determines on this basis, reasonably adjusts the weight, removes factors without influencing, and finally obtains the influencing factors and risk control rules table of high-strength steel dezincification, as shown in Table 2. The knowledge graph of the influencing factor information is shown in Figure 2 shown.

[0088] Table 2 Influencing factors and risk control rules for high-strength steel dezincification

[0089]

[0090]

[0091] Note: ppm is the unit of low concentration content

[0092] Exemplarily, the determination result is established under the standard rule; the weight of each influencing factor is calculated by comparing the normal mean value of each process parameter with the abnormal mean value, and converting the probability of occurrence of the influencing factor with a statistical change rate greater than 15% into a percentage to obtain the weight.

[0093] The calculation formula for the evaluation result is: (determination result of impact factor 1 × correlation weight of impact factor 1 + determination result of impact factor 2 × correlation weight of impact factor 2 + ... + determination result of impact factor n × correlation weight of impact factor n) / sum of correlation weights.

[0094] In some examples, the determining of the influencing factors of the surface problems of the steel coil and the positive and negative correlation of the influencing factors based on the normal average value and the abnormal average value includes:

[0095] Determine whether the above production process parameter is the above influencing factor; if the change rate between the above normal average value and the above abnormal average value is less than or equal to the preset change rate, the above production process parameter is not the above influencing factor; if the change rate between the above normal average value and the above abnormal average value is greater than the preset change rate, the above production process parameter is the above influencing factor;

[0096] When the above production process parameters are the above influencing factors, the positive and negative correlation of the above influencing factors is determined, including:

[0097] Determine whether the above-mentioned normal average value of the above-mentioned influencing factors is greater than the above-mentioned abnormal average value of the above-mentioned influencing factors; if the above-mentioned normal average value of the above-mentioned influencing factors is greater than the above-mentioned abnormal average value of the above-mentioned influencing factors, the above-mentioned influencing factors are positively correlated; if the above-mentioned normal average value of the above-mentioned influencing factors is less than the above-mentioned abnormal average value of the above-mentioned influencing factors, the above-mentioned influencing factors are negatively correlated.

[0098] Exemplarily, the calculation of the rate of change is expressed as:

[0099]

[0100] In the embodiment of the present application, a preset change rate of 15% is selected to determine whether the production process parameter is an influencing factor; if the change rate is less than or equal to the preset change rate, the production process parameter is not an influencing factor. If the change rate is greater than the preset change rate, the production process parameter is identified as an influencing factor.

[0101] For production process parameters determined to be influencing factors, it is necessary to further judge their positive and negative correlation. If the normal average value is greater than the abnormal average value, the influencing factor is considered to be positively correlated; if the normal average value is less than the abnormal average value, the influencing factor is considered to be negatively correlated.

[0102] In some examples, the process value interval of each of the above-mentioned influencing factors is divided based on the above-mentioned positive and negative correlations, including:

[0103] In the case where the above-mentioned influencing factors are positively correlated, the unqualified interval is when the above-mentioned process value is less than the first preset value, the hidden danger interval is when the above-mentioned process value is greater than or equal to the first preset value and less than or equal to the second preset value, and the qualified interval is when the above-mentioned process value is greater than the second preset value;

[0104] When the above-mentioned influencing factors are negatively correlated, the unqualified interval is when the above-mentioned process value is greater than the third preset value, the hidden danger interval is when the above-mentioned process value is greater than or equal to the fourth preset value and less than or equal to the third preset value, and the qualified interval is when the above-mentioned process value is less than the fourth preset value.

[0105] Exemplarily, according to the positive and negative correlation of the influencing factors, the process value interval can be divided into three categories: unqualified interval, potential danger interval and qualified interval, which can help real-time monitoring and problem warning in the production process. In the embodiment of the present application, the first preset value can be set to 0.85μ, the second preset value can be set to 0.975μ, the third preset value can be set to 1.15μ, and the fourth preset value can be set to 1.025μ, where μ is the normal average value.

[0106] In some examples, this also includes:

[0107] The process value of the influencing factor is adjusted based on the first prediction result and the second prediction result.

[0108] For example, to obtain a prediction model from a knowledge graph, the SPARQL is as follows:

[0109] SELECT*WHERE{? pjbz a<high-strength steel dezincification risk assessment standard>.? pjbz<factor>? yz.? yz<rule description>? gzm s.? yz<rule condition>? gztj.? yz<determination result>? pdjg.? yz<correlation>? xgx OPTIONAL{? yz<solution>? jjcs.}}; This code is a query statement written in SPARQL.

[0110] SELECT*WHERE{?pjbz a<High-strength steel dezincification risk assessment standard>., where SELECT*: indicates selecting all matching results, WHERE{...}: is the subject of the query, which defines the query conditions, ?pjbz a<High-strength steel dezincification risk assessment standard>: indicates the subject to be queried, (?pjbz) is an instance of "High-strength steel dezincification risk assessment standard", and a indicates the type of an entity.

[0111] ?pjbz<factor>?yz. indicates that there is a relationship between the subject?pjbz and the factor (?yz), and <factor> is an attribute pointing to the factor.

[0112] ?yz<rule description>?gzms. indicates that the factor ?yz has an attribute named "rule description" whose value is ?gzms.

[0113] ?yz<rule condition>?gztj. indicates that the factor ?yz also has an attribute named "rule condition", whose value is ?gztj.

[0114] ?yz<decision result>?pdjg indicates that the factor ?yz has an attribute of "decision result", whose value is ?pdjg.

[0115] ?yz<correlation>?xgx represents the "correlation" attribute of factor ?yz, whose value is ?xgx.

[0116] OPTIONAL{...}: This content indicates an optional query, which means that even if no "solution" information is found, the query will return other results.

[0117] ? yz<solution>? jjcs: If there is a "solution", its value? jjcs is returned.

[0118] Pre-prediction: After formulating the galvanizing production plan and before the actual operation, the design values ​​of production process parameters such as dew point, hydrogen concentration, oxygen concentration, furnace zone speed, zinc liquid temperature, furnace pressure, etc. are obtained from the surface knowledge graph, and the rule information of each influencing factor is obtained from the prediction model. The rule engine is executed in the Java code to predict surface problems and obtain the first prediction result. It is determined whether the first prediction result has risks. If there are risks, the process values ​​of the influencing factors are adjusted; if there are no risks, steel coil production is carried out.

[0119] In-process prediction: In the galvanizing operation process, the actual production process data such as production line dew point, hydrogen concentration, oxygen concentration, furnace zone speed, zinc liquid temperature, furnace pressure, etc. are collected in real time for different positions of the steel coil, and the rule information of each influencing factor is obtained from the prediction model. The rule engine is executed in the Java code to predict the surface dezincification problem at each position of the steel coil in real time, and obtain the second prediction result. It is judged whether the second prediction result has risks. If there are risks, the process values ​​of the influencing factors are adjusted in real time; if there are no risks, the process values ​​of the influencing factors are kept unchanged.

[0120] For the unqualified influencing factors, they are adjusted by 5% of the original value based on the positive and negative correlation. If the unqualified influencing factors are positively correlated, the process value of the unqualified influencing factors is adjusted to 105% of the original process value; if the unqualified influencing factors are negatively correlated, the process value of the unqualified influencing factors is adjusted to 95% of the original process value.

[0121] In some examples, this also includes:

[0122] The above prediction accuracy is determined based on the following formula:

[0123] Forecast accuracy = (total production quantity - the number of actual surface problems that did not appear but were not predicted - the number of surface problems that were predicted but were not actually found) / total production quantity;

[0124] Among them, the number of steel coils that actually had surface problems but were not predicted to have surface problems is the number of steel coils that actually had surface problems but were not identified by the first prediction result or the second prediction result; the number of steel coils that were predicted to have surface problems but did not actually have surface problems is the number of steel coils that were predicted to have surface problems by the first prediction result or the second prediction result but did not actually have surface problems.

[0125] Exemplarily, the present application counts the surface problem prediction accuracy on a monthly basis, and queries the steel coil surface problem prediction results and the actual surface problem coils by monthly grouping in the knowledge graph, and calculates the surface problem prediction accuracy based on this result. When the influencing factor prediction is unqualified, the rule condition scope and weight are readjusted.

[0126] See also Figure 3 , which is a schematic diagram of the structure of a steel coil surface defect prediction device based on a knowledge graph provided in an embodiment of the present application, comprising:

[0127] A first prediction unit 21 is used to predict the surface problems of the steel coil before production based on the prediction model and the knowledge graph of the surface problems of the steel coil, and obtain a first prediction result, wherein the above-mentioned prediction model is a prediction model of the surface problems of the steel coil constructed based on the above-mentioned knowledge graph and the influencing factor information of the surface problems of the steel coil;

[0128] A second prediction unit 22 is used to predict surface problems in the production of steel coils based on the above prediction model and real-time process values ​​to obtain a second prediction result;

[0129] The statistical result unit 23 is used to calculate the prediction accuracy of the surface problem based on the first prediction result and the second prediction result.

[0130] See also Figure 4 The embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any method for predicting surface defects of steel coils based on the knowledge graph are implemented.

[0131] Since the electronic device introduced in this embodiment is a device used to implement the steel coil surface defect prediction device based on the knowledge graph in the embodiment of the present application, based on the method introduced in the embodiment of the present application, the technical personnel in this field can understand the specific implementation mode of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present application is not introduced in detail here. As long as the equipment adopted by the technical personnel in this field to implement the method in the embodiment of the present application, it belongs to the scope of protection of this application.

[0132] During the specific implementation process, when the computer program 311 is executed by a processor, any implementation method in the embodiments corresponding to the first aspect can be implemented.

[0133] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0134] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-readable program code.

[0135] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0136] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0138] The present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes Figure 1 The process of the steel coil surface defect prediction method based on knowledge graph in the corresponding embodiment.

[0139] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integration. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)), etc.

[0140] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0141] In the several embodiments provided in the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0142] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0143] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0144] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0145] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

[0146] Although the preferred embodiments of this specification have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of this specification.

[0147] Obviously, those skilled in the art can make various changes and modifications to this specification without departing from the spirit and scope of this specification. Thus, if these modifications and variations of this specification fall within the scope of the claims of this specification and their equivalents, this specification is also intended to include these modifications and variations.

Claims

1. A method for predicting surface defects of steel coils based on knowledge graph, characterized in that: The method comprises: Based on the prediction model and the knowledge graph of the surface problems of the steel coil, the surface problems of the steel coil are predicted before production to obtain a first prediction result, wherein the prediction model is a prediction model of the surface problems of the steel coil constructed based on the knowledge graph and the influencing factor information of the surface problems of the steel coil; Based on the prediction model and the real-time process value, predict the surface problems in the steel coil production to obtain a second prediction result; Based on the first prediction result and the second prediction result, the prediction accuracy of the surface problem is calculated.

2. According to claim 1, a method for predicting surface defects of steel coils based on knowledge graph is characterized in that: The specific steps of constructing the knowledge graph of the steel coil surface problem include: Based on the production process, the ontology, attributes and relationships of the production process are abstracted to form the knowledge graph rules of the surface problems of the steel coil, wherein the production process includes galvanizing process plan, galvanizing plan, primary galvanizing and secondary galvanizing; Based on the knowledge graph rules, data mapping is performed, wherein the data mapping includes ontology-attribute mapping and relationship mapping; Based on the data mapping, dynamically generate SQL statements to query data in the production data source, perform knowledge extraction, and form a triple file; The triple file is imported into an RDF graph database to form a knowledge graph of the surface problems of the steel coil.

3. The method for predicting surface defects of a steel coil based on a knowledge graph according to claim 1, characterized in that: The specific steps of constructing the prediction model include: Using SPARQL language, querying the process performance data of normal steel coils and abnormal steel coils from the knowledge graph respectively, wherein the abnormal steel coils are steel coils with surface problems; Calculate the average values ​​of normal process performance data and abnormal process performance data of each production process parameter respectively to obtain the normal average value and the abnormal average value; Based on the normal average value and the abnormal average value, determining the influencing factors of the surface problems of the steel coil and the positive and negative correlations of the influencing factors; Based on the positive and negative correlation, dividing the process value interval of each of the influencing factors; Calculating the weight of each of the influencing factors based on a factor weight algorithm; Based on the name, weight, standard rule and determination result of the influencing factor, the influencing factor information of the surface problem of the steel coil is formed, wherein the determination result is determined based on the process value range and the standard rule; The impact factor information is entered into the prediction model.

4. The method for predicting surface defects of a steel coil based on a knowledge graph according to claim 3 is characterized in that: The determining, based on the normal average value and the abnormal average value, the influencing factors of the surface problems of the steel coil and the positive and negative correlation of the influencing factors comprises: Determining whether the production process parameter is the influencing factor; When the production process parameter is the influencing factor, the positive and negative correlation of the influencing factor is determined.

5. The method for predicting surface defects of a steel coil based on a knowledge graph according to claim 3 is characterized in that: The dividing of the process value interval of each of the influencing factors based on the positive and negative correlation includes: In the case where the influencing factors are positively correlated, the unqualified interval is when the process value is less than the first preset value, the potential danger interval is when the process value is greater than or equal to the first preset value and less than or equal to the second preset value, and the qualified interval is when the process value is greater than the second preset value; When the influencing factors are negatively correlated, the unqualified interval is when the process value is greater than the third preset value, the hidden danger interval is when the process value is greater than or equal to the fourth preset value and less than or equal to the third preset value, and the qualified interval is when the process value is less than the fourth preset value.

6. The method for predicting surface defects of steel coils based on knowledge graph according to claim 1 is characterized in that: Also includes: The process value of the influencing factor is adjusted based on the first prediction result and the second prediction result.

7. The method for predicting surface defects of steel coils based on knowledge graph according to claim 1 is characterized in that: Also includes: The prediction accuracy is determined based on the following formula: Forecast accuracy = (total production quantity - the number of actual surface problems that did not appear but were not predicted - the number of surface problems that were predicted but were not actually found) / total production quantity; The number of steel coils that actually have surface problems but are not predicted to have surface problems is the number of steel coils that actually have surface problems but are not identified by the first prediction result or the second prediction result. The number of steel coils predicted to have surface problems but not actually having surface problems is the number of steel coils predicted to have surface problems by the first prediction result or the second prediction result but not actually having surface problems.

8. A device for predicting surface defects of steel coils based on knowledge graph, characterized in that: include: A first prediction unit is used to predict the surface problems of the steel coil before production based on the prediction model and the knowledge graph of the surface problems of the steel coil, and obtain a first prediction result, wherein the prediction model is a prediction model of the surface problems of the steel coil constructed based on the knowledge graph and the influencing factor information of the surface problems of the steel coil; A second prediction unit is used to predict surface problems in steel coil production based on the prediction model and real-time process values ​​to obtain a second prediction result; The statistical result unit is used to calculate the prediction accuracy of the surface problem based on the first prediction result and the second prediction result.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is used to implement the steps of a method for predicting surface defects of a steel coil based on a knowledge graph as described in any one of claims 1 to 7 when executing the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a method for predicting surface defects of a steel coil based on a knowledge graph as described in any one of claims 1 to 7 is implemented.

Images