Graphite electrode defect detection system

Abstract

The invention discloses a graphite electrode defect detection system, and relates to the technical field of graphite electrode detection.The graphite electrode defect detection system comprises a product detection unit, a production detection unit, a detection output unit and a database; the method comprises the following steps: determining a quality data threshold value of a graphite electrode product in each scene, selecting a detection technology of a produced graphite electrode, detecting defects of the graphite electrode, analyzing a graphite electrode product processing scheme, and detecting the defects of the graphite electrode by using multiple detection technologies, so that the detection is more detailed and accurate; and meanwhile, the processing scheme of the graphite electrode is analyzed, so that the utilization rate of the graphite electrode is improved, the maximum utilization value of the graphite electrode is exerted, and the qualification of the quality of the graphite electrode in actual use is guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of graphite electrode detection, and in particular to a graphite electrode defect detection system. Background Art

[0002] Graphite electrode is a high-temperature resistant graphite conductive material. Graphite electrodes work in high-temperature and high-pressure environments such as steelmaking. If there are defects or cracks in the graphite electrodes, it may cause the electrodes to break, thus affecting the production process and even causing safety accidents. Therefore, defect detection can timely understand the quality of graphite electrodes, and process them into products for different production scenarios according to the quality to ensure production safety.

[0003] Prior art, such as a graphite electrode defect detection system disclosed in an application with publication number CN111157622A, comprises: a cloud server, a first processor module, a second processor module, a knocking mechanism and a sound detection module connected to the second processor module, and a transmission module electrically connected to the first processor module; the second processor module is suitable for controlling the knocking mechanism to knock on the graphite electrode; the sound detection module is suitable for detecting the sound made by the knocking mechanism knocking on the graphite electrode, and sending the sound signal to the second processor module, so as to be forwarded by the second processor module to the first processor module; the first processor module is suitable for sending the sound signal to the cloud server through the transmission module; the cloud server is suitable for judging the degree of defect of the graphite electrode according to the sound signal, thereby realizing automatic detection of the degree of defect of the graphite electrode, avoiding the shortcomings of strong subjectivity and high missed detection rate caused by methods such as manual detection, and reducing labor costs at the same time.

[0004] In the above scheme, the graphite electrode is only knocked to detect the knocking sound and analyze the degree of defect of the graphite electrode. There are many defects in the graphite electrode. According to the size of the graphite electrode and the type of defect, different detection technologies need to be selected for comprehensive detection to make the detection more detailed and accurate. The above scheme only uses a single detection technology for defect detection, which cannot improve the comprehensiveness and accuracy of the detection and increase the defect detection effect. At the same time, the quality requirements of graphite electrodes in different application scenarios are different, and the produced graphite electrodes can be processed into a variety of graphite electrode products. Different products are suitable for different scenarios according to their quality. In the above scheme, the defects of the graphite electrode are only detected according to the knocking sound, and the subsequent processing of the graphite electrode is not analyzed. It is impossible to improve the utilization rate of the graphite electrode, nor to maximize the utilization value of the graphite electrode, and it is impossible to guarantee the quality of the graphite electrode in actual use. Summary of the invention

[0005] In view of the above-mentioned technical deficiencies, an object of the present invention is to provide a graphite electrode defect detection system.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solution: The present invention provides a graphite electrode defect detection system, including: a product detection unit, a production detection unit and a detection output unit.

[0007] The product inspection unit includes a product defect detection module and a product loss analysis module.

[0008] The product detection module is used to detect the graphite electrode products in each scene and obtain the detection data of the graphite electrode products in each scene.

[0009] The product loss analysis module is used to obtain the initial production inspection data and comprehensive environmental data corresponding to the graphite electrode products in each scenario, and extract the historical application data of each scenario. At the same time, the inspection data of the graphite electrode products in each scenario is used to analyze the quality data threshold corresponding to the graphite electrode products in each scenario.

[0010] The production inspection unit includes a production quality inspection module and a product processing analysis module.

[0011] The production quality detection module is used to detect the graphite electrodes after production and obtain quality data corresponding to the graphite electrodes.

[0012] The product processing analysis module is used to analyze the processing plan corresponding to the graphite electrode by using the quality data corresponding to the graphite electrode and the quality data threshold corresponding to the graphite electrode product in each scene.

[0013] The detection output unit is used to output the processing plan corresponding to the graphite electrode.

[0014] The beneficial effects of the present invention are as follows: the present invention provides a graphite electrode defect detection system, which detects the quality and environment of graphite electrode products in different scenarios, confirms the quality data threshold of graphite electrode products in each scenario, selects the detection technology for the graphite electrodes after production, and then detects the defects of the graphite electrodes, and then analyzes the processing plan of the graphite electrode products, and uses a variety of detection technologies to detect the defects of the graphite electrodes, so that the detection is more detailed and accurate, and the comprehensiveness of the defect detection is guaranteed. At the same time, the processing plan of the graphite electrodes is analyzed to improve the utilization rate of the graphite electrodes, maximize the utilization value of the graphite electrodes, and ensure the quality of the graphite electrodes in actual use. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a schematic diagram of the system structure connection of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] See also Figure 1 As shown, a graphite electrode defect detection system includes: a product detection unit, a production detection unit, a detection output unit and a database.

[0019] The product inspection unit includes a product defect detection module and a product loss analysis module.

[0020] The product detection module is used to detect the graphite electrode products in each scene and obtain the detection data of the graphite electrode products in each scene.

[0021] It should be added that the various scenarios include electric arc furnace smelting scenarios, metal welding scenarios and the electronics industry; the graphite electrode products of electric arc furnace smelting are rough electrodes, the graphite electrode products of metal welding scenarios are fine-particle electrodes, and the graphite electrode products of the electronics industry are high-purity electrodes.

[0022] The specific process of testing the graphite electrode products in each scene is as follows: obtaining the size data corresponding to the graphite electrode products in each scene from the database, and using environmental detection equipment to collect environmental data of each scene, and at the same time extracting the appropriate size data interval and appropriate environmental size data corresponding to each detection technology from the database, inputting them into the technology selection evaluation model, and outputting the compatibility results of the graphite electrode products in each scene and each detection technology.

[0023] In the above, the expression of the technology selection evaluation model is:

[0024] In the formula represents the compatibility result of the graphite electrode product in the fth scenario and the jth detection technology, C f , H f They represent the size data of the graphite electrode product in the fth scene and the environmental data of the fth scene, respectively. j , H jThey respectively represent the appropriate size data interval and the appropriate environment size data corresponding to the j-th detection technology, f represents the number of each scene, j represents the number of each detection technology, and both f and j are positive integers.

[0025] The adaptability result includes data of 1 and 0. When the adaptability result is 1, it indicates adaptability, otherwise it indicates incompatibility.

[0026] The detection technologies whose corresponding adaptability results for the graphite electrode products in each scene are 1 are selected as the target detection technologies for the graphite electrode products in each scene, and the graphite electrode products in each scene are detected using the target detection technologies to obtain the detection data of the graphite electrode products in each scene.

[0027] It should be noted that the dimensional data includes length, width and thickness, etc. It is designed by product designers before the graphite electrode products are processed and stored in the database.

[0028] Environmental detection equipment includes temperature sensors and magnetic field measuring instruments, etc.; environmental data includes temperature and electromagnetic intensity, etc.

[0029] Various detection technologies include visual detection, infrared detection, ultrasonic detection, X-ray detection, electromagnetic detection and acoustic emission detection.

[0030] The appropriate size data range and appropriate environmental size data corresponding to each detection technology are jointly discussed and formulated by multiple experts based on historical experience and a large amount of historical data, and stored in the database.

[0031] The thickness of the graphite electrode may make it difficult for X-rays to penetrate, affecting the image quality. High temperature will affect the results of infrared detection of graphite electrodes. High temperature may cause the sensitivity of infrared detectors to decrease, affecting the accuracy of temperature measurement. In addition, changes in the radiation characteristics of graphite at high temperatures may cause signal distortion, thereby affecting defect identification and imaging quality. Different detection technologies are suitable for different detection sizes and environments. Therefore, it is necessary to detect the size and environmental data of graphite electrode products in different scenarios, and then select appropriate detection technologies for defect detection to improve the accuracy of the detection results.

[0032] The product loss analysis module is used to obtain the initial production inspection data and comprehensive environmental data corresponding to the graphite electrode products in each scenario, and extract the historical application data of each scenario. At the same time, the inspection data of the graphite electrode products in each scenario is used to analyze the quality data threshold corresponding to the graphite electrode products in each scenario.

[0033] In a specific embodiment, the quality data threshold corresponding to the graphite electrode product in each scene is analyzed, and the specific process is as follows: during the use of the graphite electrode product in each scene, a number of collection moments are arranged, and at each collection moment, environmental detection equipment is used to collect environmental data of the graphite electrode product in each scene, and the environmental data of the graphite electrode product at each collection moment in each scene is obtained as the comprehensive environmental data of each scene.

[0034] The environmental type of the graphite electrode products in each scenario is analyzed using the comprehensive environmental data corresponding to each scenario. At the same time, the loss characteristic values ​​of the graphite electrode products in each scenario are analyzed using the initial production inspection data and inspection data corresponding to the graphite electrode products in each scenario.

[0035] In the above, the environmental type of the graphite electrode product in each scene is analyzed, and the specific process is as follows: the environmental data of the graphite electrode product at each collection time is extracted from the comprehensive environmental data corresponding to each scene, and recorded as H ft , t represents the number of each collection moment, and t is a positive integer.

[0036] The environmental data at each collection time in the graphite electrode products of each scenario are compared with each other to obtain the environmental data difference between each collection time in the graphite electrode products of each scenario, and the mean is calculated to obtain the average environmental data difference, which is recorded as ΔH.

[0037] Input to the environmental assessment model: Get the environmental assessment characteristic value δ of the graphite electrode product in the fth scenario f , where p represents the total number of acquisition moments, H f(t-1) It represents the environmental data of the graphite electrode product in the fth scenario at the t-1th collection moment, and κ1 and κ2 are the lower limit and upper limit of the preset environmental assessment coefficient respectively.

[0038] It should be noted that the lower and upper limits of the environmental assessment coefficient are critical values ​​for assessing whether the environment is stable. They are jointly discussed and formulated by a number of experts based on historical experience and a large amount of historical data. When the value is greater than κ2, it indicates that the environmental volatility is large and the environment is unstable. When the value is between κ1 and κ2, it indicates that the environment fluctuates, but the volatility is small; when the value is less than κ1, it indicates that the environment has almost no fluctuations and is stable. For example, κ1 and κ2 are 2 and 6 respectively. The calculated value of is 3, which is between 2 and 6, indicating that there is fluctuation in the environment of graphite electrode products in the fth scenario, but the volatility is small.

[0039] When the environmental assessment characteristic value is 1, it indicates that the environment type is stable; when the environmental assessment characteristic value is 0, it indicates that the environment type is small fluctuation type; when the environmental assessment characteristic value is -1, it indicates that the environment type is unstable; x is 1, 2 and 3, 1, 2 and 3 represent stable type, small fluctuation type and unstable type respectively, and y is 3.

[0040] Preferably, the loss characteristic value of the graphite electrode product in each scene is analyzed, and the specific process is as follows: extracting the initial detection time of the initial detection data of the production of the graphite electrode product in each scene and the detection time of the detection data, respectively recorded as T0 f and T f .

[0041] Among them, when the graphite electrode product is tested, a timestamp can be automatically generated and recorded and stored in the test record, thereby extracting the initial detection time and the detection time of the detection data corresponding to the initial production detection data of the graphite electrode product in each scene from the test record.

[0042] The initial values ​​of various test data are obtained from the initial production test data corresponding to the graphite electrode products in each scenario, and are recorded as W0 fa , obtain the detection values ​​of various detection data from the detection data corresponding to the graphite electrode products in each scene, denoted as W fa , set the proportional coefficient of each type of detection data, denoted as γ a , a represents the serial number of each type of detection data, and a is a positive integer.

[0043] Among them, the maximum and minimum values ​​of each type of detection data are obtained from the detection values ​​of each type of detection data in the detection data corresponding to the graphite electrode products in each scene, and the maximum value of each type of detection data is subtracted from the minimum value, and then divided by the minimum value of each type of detection data to obtain the difference rate of each type of detection data, and the difference rate of each type of detection data is divided by the sum of the difference rates of each type of detection data to obtain the proportional coefficient of each type of detection data.

[0044] Using analytical formula Get the loss characteristic value of the graphite electrode product in the fth scenario q represents the total number of detected data types.

[0045] The historical comprehensive environmental data and loss characteristic values ​​corresponding to each historical graphite electrode product are obtained from the historical application data of each scenario, and the environmental type corresponding to each historical graphite electrode product in each scenario is analyzed.

[0046] It should be noted that the analysis method for the environmental type corresponding to each historical graphite electrode product in each scenario is the same as the analysis method for the environmental type of the graphite electrode products in each scenario, which will not be repeated here.

[0047] Based on the environmental types of graphite electrode products in each scenario, the loss characteristic values ​​of graphite electrode products, the environmental types and loss characteristic values ​​of historical graphite electrode products, the characteristic weights of each environmental type in each scenario and the set of loss characteristic values ​​corresponding to each link type are analyzed.

[0048] Preferably, the characteristic weights of each environment type in each scene are analyzed, and the specific process is as follows: using the environment type of the graphite electrode product in each scene and the environment type corresponding to each historical graphite electrode product in each scene, the number of occurrences of each environment type in each scene is counted, recorded as D fx According to the environmental type of graphite electrode products in each scenario, the loss characteristic value of graphite electrode products, the environmental type and loss characteristic value of each historical graphite electrode product, the loss characteristic value set corresponding to each environmental type in each scenario is statistically obtained, and the loss characteristic value volatility corresponding to each environmental type in each scenario and the average loss characteristic value volatility are calculated, which are recorded as and

[0049] Using analytical formula Get the feature weight ε of the xth environment type in the fth scene fx , where e represents a natural constant.

[0050] The minimum loss characteristic value is selected from the loss characteristic value set corresponding to each environmental type in each scene, and the initial production inspection data of the graphite electrode product corresponding to the minimum loss characteristic value is extracted as the initial quality data threshold corresponding to each environmental type in each scene.

[0051] According to the analysis formula Get the quality data threshold α corresponding to the graphite electrode product in the fth scene f , where ε fx , Z fx They respectively represent the feature weight and initial quality data threshold of the xth environment type in the fth scene, f represents the number of each scene, x represents the number corresponding to each link type, y represents the total number of environment types, f is a positive integer, x is 1, 2 and 3, 1, 2 and 3 represent stable, small fluctuation and unstable types respectively, and y is 3.

[0052] Different scenarios have different quality requirements for graphite electrodes. For example, for applications such as electric arc furnace smelting and aluminum electrolysis, the conductivity of the electrode is required to be high, while the high-precision electronics industry has a low tolerance for surface defects. By testing the quality of graphite electrode products in different scenarios, and then testing the quality of the completed graphite electrodes, and then performing targeted processing based on the graphite electrodes, the loss caused by processing a single product of graphite electrodes can be reduced, and the utilization rate of graphite electrodes can be effectively improved.

[0053] The production inspection unit includes a production quality inspection module and a product processing analysis module.

[0054] The production quality detection module is used to detect the graphite electrodes after production and obtain quality data corresponding to the graphite electrodes.

[0055] In a specific embodiment, the graphite electrode after production is inspected, and the specific process is as follows: obtaining a set of inspection technologies corresponding to various types of inspection data from a database; and simultaneously obtaining historical inspection records from the database; thereby obtaining the appearance data and production data of the historical graphite electrodes corresponding to each historical inspection from the historical inspection records.

[0056] Among them, the detection technology set corresponding to each type of detection data is set by experts. For example, if the detection data is the apparent crack size, the detection technologies in the detection technology set include visual detection technology, ultrasonic detection, infrared thermal imaging, magnetic particle detection and coating penetration detection.

[0057] It should be noted that appearance data includes shape and size, etc., and production data includes production data corresponding to each production stage, for example: each production stage includes raw material calcination, crushing and grinding, batching, kneading, molding, roasting and impregnation, etc.; the production data of raw material calcination is calcination temperature and calcination time, etc.; the production data of crushing and grinding is crushing particle size and grinding particle size, etc.

[0058] The appearance data and manufacturing data corresponding to the graphite electrode are extracted, and compared with the appearance data and manufacturing data of the historical graphite electrodes corresponding to each historical test, and the historical graphite electrodes corresponding to each historical test with the same appearance data and manufacturing data are selected as the reference electrodes of the graphite electrodes.

[0059] The detection technology and repeated detection times corresponding to each type of detection data of each reference electrode are obtained from the historical detection records, and the detection technology and repeated detection times corresponding to each type of detection data are counted. The repeated detection times corresponding to each type of detection data of each detection technology are averaged, and the detection technology with the smallest mean is selected as the detection technology corresponding to each type of detection data.

[0060] The graphite electrode is divided into grids to obtain partitions; each partition is then tested using a test technology corresponding to each type of test data to obtain quality data for each analysis as quality data corresponding to the graphite electrode.

[0061] The quality data includes various test data, including apparent crack size, internal crack size and internal corrosion size.

[0062] The product processing analysis module is used to analyze the processing plan corresponding to the graphite electrode by using the quality data corresponding to the graphite electrode and the quality data threshold corresponding to the graphite electrode product in each scene.

[0063] In a specific embodiment, the processing scheme corresponding to the graphite electrode is analyzed, and the specific analysis process is as follows: using the quality data of each partition and the quality data threshold corresponding to the graphite electrode product in each scene, the adaptability of each partition to each scene is analyzed, and the priority adaptation scene set corresponding to each partition is selected.

[0064] Preferably, the priority adaptation scene set corresponding to each partition is selected, and the specific process is as follows: the quality data of each partition and the quality data threshold corresponding to the graphite electrode product in each scene are respectively recorded as Z b and Z f , b is the number of each partition, b is a positive integer, according to the analysis formula Get the degree of fit θ between the bth partition and the fth scene bf , where v1 represents the preset quality data difference rate threshold.

[0065] When the degree of adaptation is greater than 1, it indicates adaptation, otherwise it indicates incompatibility. The scenes with a degree of adaptation of 1 in each partition are selected as the priority adaptation scenes, and the priority adaptation scene sets corresponding to each partition are obtained by integration.

[0066] The priority adaptation scene sets corresponding to adjacent partitions are compared. When there is at least one priority adaptation scene in the priority adaptation scene sets corresponding to adjacent partitions that is the same, the adjacent partition is taken as a processing partition, and the same priority adaptation scenes are taken as preferred scenes; thereby obtaining each processing partition and each preferred scene corresponding to each processing partition.

[0067] Obtain the size data of the graphite electrode in each preferred scenario corresponding to each processing zone, predict the loss value of the graphite electrode product in each preferred scenario processed in each processing zone, select the graphite electrode product in the preferred scenario corresponding to the minimum loss value as the graphite electrode product processed in each processing zone, and use the graphite electrode product processed in each processing zone as the processing plan corresponding to the graphite electrode.

[0068] Among them, the size of each processing zone is counted, and the size of each processing zone is divided by the size data of the graphite electrode in each preferred scenario. If there is a remainder, the remainder is used as the loss value of the graphite electrode product in each processing zone in each preferred scenario. If there is no remainder, the loss value of the graphite electrode product in each processing zone in each preferred scenario is recorded as 0.

[0069] The detection output unit is used to output the processing plan corresponding to the graphite electrode.

[0070] The database is used to store the dimensional data corresponding to the graphite electrode products in each scenario, store the appropriate dimensional data interval and appropriate environmental dimensional data corresponding to each detection technology, store the detection technology set corresponding to each type of detection data, and store historical detection records.

[0071] The embodiment of the present invention detects the quality and environment of graphite electrode products in different scenarios, confirms the quality data threshold of graphite electrode products in each scenario, selects the detection technology for graphite electrodes after production, detects defects in the graphite electrodes, and then analyzes the processing plan of the graphite electrode products. A variety of detection technologies are used to detect defects in the graphite electrodes, making the detection more detailed and accurate, ensuring the comprehensiveness of defect detection, and at the same time analyzing the processing plan of the graphite electrodes, improving the utilization rate of the graphite electrodes, maximizing the utilization value of the graphite electrodes, and ensuring the quality of the graphite electrodes in actual use.

[0072] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they shall all fall within the protection scope of the present invention.

Claims

1. A graphite electrode defect detection system, characterized in that: include: Product testing unit, production testing unit and testing output unit; The product inspection unit includes a product defect detection module and a product loss analysis module; The product detection module is used to detect the graphite electrode products in each scene and obtain the detection data of the graphite electrode products in each scene; The product loss analysis module is used to obtain the initial production inspection data and comprehensive environmental data corresponding to the graphite electrode products in each scenario, and extract the historical application data of each scenario, and use the inspection data of the graphite electrode products in each scenario to analyze the quality data threshold corresponding to the graphite electrode products in each scenario; The production inspection unit includes a production quality inspection module and a product processing analysis module; The production quality detection module is used to detect the graphite electrode after production and obtain the quality data corresponding to the graphite electrode; The product processing analysis module is used to analyze the processing scheme corresponding to the graphite electrode by using the quality data corresponding to the graphite electrode and the quality data threshold corresponding to the graphite electrode product in each scene; The detection output unit is used to output the processing plan corresponding to the graphite electrode.

2. A graphite electrode defect detection system according to claim 1, characterized in that: The specific process of testing the graphite electrode products in each scenario is as follows: Obtain the size data corresponding to the graphite electrode products in each scenario from the database, and use environmental detection equipment to collect environmental data for each scenario. At the same time, extract the appropriate size data interval and appropriate environmental size data corresponding to each detection technology from the database, input them into the technology selection evaluation model, and output the compatibility results of the graphite electrode products in each scenario and each detection technology; The adaptability result contains data of 1 and 0. When the adaptability result is 1, it indicates adaptability, otherwise it indicates incompatibility. The detection technologies whose corresponding adaptability results for the graphite electrode products in each scene are 1 are selected as the target detection technologies for the graphite electrode products in each scene, and the graphite electrode products in each scene are detected using the target detection technologies to obtain the detection data of the graphite electrode products in each scene.

3. A graphite electrode defect detection system according to claim 2, characterized in that: The expression of the technology selection evaluation model is: In the formula represents the compatibility result of the graphite electrode product in the fth scenario and the jth detection technology, C f , H f They represent the size data of the graphite electrode product in the fth scene and the environmental data of the fth scene, respectively. j , H j They respectively represent the appropriate size data interval and the appropriate environment size data corresponding to the j-th detection technology, f represents the number of each scene, j represents the number of each detection technology, and both f and j are positive integers.

4. A graphite electrode defect detection system according to claim 1, characterized in that: The specific process of analyzing the quality data thresholds corresponding to graphite electrode products in each scenario is as follows: During the use of the graphite electrode products in each scene, several collection moments are arranged, and environmental detection equipment is used at each collection moment to collect environmental data of the graphite electrode products in each scene, so as to obtain the environmental data of the graphite electrode products in each scene at each collection moment as the comprehensive environmental data of each scene; The environmental type of the graphite electrode products in each scenario is analyzed using the comprehensive environmental data corresponding to each scenario; at the same time, the loss characteristic values ​​of the graphite electrode products in each scenario are analyzed using the initial production test data and test data corresponding to the graphite electrode products in each scenario; Obtain the historical comprehensive environmental data and loss characteristic values ​​corresponding to each historical graphite electrode product from the historical application data of each scenario, and analyze the environmental type corresponding to each historical graphite electrode product in each scenario; Based on the environmental type of graphite electrode products in each scenario, the loss characteristic value of graphite electrode products, the environmental type and loss characteristic value of each historical graphite electrode product, the characteristic weight of each environmental type in each scenario and the set of loss characteristic values ​​corresponding to each link type are analyzed; Select the minimum loss characteristic value from the loss characteristic value set corresponding to each environment type in each scene, and extract the initial production inspection data of the graphite electrode product corresponding to the minimum loss characteristic value as the initial quality data threshold corresponding to each environment type in each scene; According to the analysis formula Get the quality data threshold α corresponding to the graphite electrode product in the fth scene f , where ε fx , Z fx They respectively represent the feature weight and initial quality data threshold of the xth environment type in the fth scene, f represents the number of each scene, x represents the number corresponding to each link type, y represents the total number of environment types, f is a positive integer, x is 1, 2 and 3, 1, 2 and 3 represent stable, small fluctuation and unstable types respectively, and y is 3.

5. A graphite electrode defect detection system according to claim 4, characterized in that: The specific process of analyzing the environmental types of graphite electrode products in each scenario is as follows: The environmental data of graphite electrode products at each collection time are extracted from the comprehensive environmental data corresponding to each scene, and recorded as H ft , t represents the number of each acquisition moment, and t is a positive integer; The environmental data at each collection time of the graphite electrode products in each scenario are compared with each other to obtain the environmental data difference between each collection time of the graphite electrode products in each scenario, and the mean is calculated to obtain the average environmental data difference, which is recorded as ΔH; Input to the environmental assessment model: Get the environmental assessment characteristic value δ of the graphite electrode product in the fth scenario f , where p represents the total number of acquisition moments, H f ( t-1 ) represents the environmental data of the graphite electrode product in the fth scenario at the t-1th collection time, κ1 and κ2 are the lower limit and upper limit of the preset environmental assessment coefficient respectively; When the environmental assessment characteristic value is 1, it indicates that the environment type is stable; when the environmental assessment characteristic value is 0, it indicates that the environment type is small fluctuation type; when the environmental assessment characteristic value is -1, it indicates that the environment type is unstable; x is 1, 2 and 3, 1, 2 and 3 represent stable type, small fluctuation type and unstable type respectively, and y is 3.

6. A graphite electrode defect detection system according to claim 4, characterized in that: The specific process of analyzing the loss characteristic values ​​of graphite electrode products in each scenario is as follows: Extract the initial detection time and detection time of the initial detection data of the production of graphite electrode products corresponding to each scene, and record them as T0 and T1 respectively. f and T f ; The initial values ​​of various test data are obtained from the initial production test data corresponding to the graphite electrode products in each scenario, and are recorded as W0 fa , obtain the detection values ​​of various detection data from the detection data corresponding to the graphite electrode products in each scene, denoted as W fa , set the proportional coefficient of each type of detection data, denoted as γ a , a represents the number of each type of test data, and a is a positive integer; Using analytical formula Get the loss characteristic value of the graphite electrode product in the fth scenario q represents the total number of detected data types.

7. A graphite electrode defect detection system according to claim 4, characterized in that: The specific process of analyzing the feature weights of each environment type in each scene is as follows: Using the environmental type of the graphite electrode products in each scenario and the environmental type corresponding to each historical graphite electrode product in each scenario, the number of occurrences of each environmental type in each scenario is counted, recorded as D fx According to the environmental type of graphite electrode products in each scenario, the loss characteristic value of graphite electrode products, the environmental type and loss characteristic value of each historical graphite electrode product, the loss characteristic value set corresponding to each environmental type in each scenario is statistically obtained, and the loss characteristic value volatility corresponding to each environmental type in each scenario and the average loss characteristic value volatility are calculated, which are recorded as and Using analytical formula Get the feature weight ε of the xth environment type in the fth scene fx , where e represents a natural constant.

8. A graphite electrode defect detection system according to claim 1, characterized in that: The specific process of testing the graphite electrode after production is as follows: Obtain a set of detection technologies corresponding to various types of detection data from the database; and simultaneously obtain historical detection records from the database; thereby obtaining appearance data and manufacturing data of historical graphite electrodes corresponding to each historical detection from the historical detection records; Extracting appearance data and manufacturing data corresponding to the graphite electrode, and comparing them with the appearance data and manufacturing data of historical graphite electrodes corresponding to each historical test, and selecting the historical graphite electrodes corresponding to each historical test with the same appearance data and manufacturing data as the reference electrodes of the graphite electrode; Obtaining the detection technology and repeated detection times corresponding to each type of detection data of each reference electrode from the historical detection records, counting the detection technology and repeated detection times corresponding to each type of detection data, calculating the mean of the repeated detection times corresponding to each type of detection data, and selecting the detection technology with the smallest mean as the detection technology corresponding to each type of detection data; The graphite electrode is divided into grids to obtain partitions; each partition is then tested using a test technology corresponding to each type of test data to obtain quality data for each analysis as quality data corresponding to the graphite electrode.

9. A graphite electrode defect detection system according to claim 8, characterized in that: The processing scheme corresponding to the analysis of graphite electrode, the specific analysis process is as follows: Using the quality data of each partition and the quality data threshold corresponding to the graphite electrode products in each scene, analyze the adaptability of each partition to each scene, and select the priority adaptation scene set corresponding to each partition; The priority adaptation scene sets corresponding to adjacent partitions are compared. When at least one priority adaptation scene in the priority adaptation scene sets corresponding to adjacent partitions is the same, the adjacent partition is taken as a processing partition, and the same priority adaptation scenes are taken as preferred scenes; thereby obtaining each processing partition and each preferred scene corresponding to each processing partition; Obtain the size data of the graphite electrode in each preferred scenario corresponding to each processing zone, predict the loss value of the graphite electrode product in each preferred scenario processed in each processing zone, select the graphite electrode product in the preferred scenario corresponding to the minimum loss value as the graphite electrode product processed in each processing zone, and use the graphite electrode product processed in each processing zone as the processing plan corresponding to the graphite electrode.

10. A graphite electrode defect detection system according to claim 1, characterized in that: The specific process of selecting the priority adaptation scene set corresponding to each partition is as follows: The quality data of each partition and the quality data threshold corresponding to the graphite electrode product in each scene are respectively recorded as Z b and Z f , b is the number of each partition, b is a positive integer, according to the analysis formula Get the degree of fit θ between the bth partition and the fth scene bf , where v1 represents the preset quality data difference rate threshold; When the degree of adaptation is greater than 1, it indicates adaptation, otherwise it indicates incompatibility. The scenes with a degree of adaptation of 1 in each partition are selected as the priority adaptation scenes, and the priority adaptation scene sets corresponding to each partition are obtained by integration.

Images