Online resistance testing method and system for oxygen-free copper rod

Abstract

The invention relates to the technical field of resistance testing, in particular to an on-line resistance testing method and system for an oxygen-free copper rod, and the method comprises the steps: monitoring the electromagnetic field change in the production process of the oxygen-free copper rod, and recognizing the resistance abnormality caused by the internal structure change of a material and the machining stress; applying currents with different intensities to the copper rods with abnormal resistance, performing real-time measurement and preliminary screening on instantaneous resistance of the copper rods, and identifying the copper rods with possible abnormal resistance; based on a feature extraction algorithm, performing feature analysis on the preliminarily screened abnormal copper rod; and comparing the source and the property of the abnormal resistance with a preset quality standard to obtain an online resistance test result. According to the invention, the problem that the traditional off-line resistance test cannot monitor the resistance change in real time in the production process is effectively solved, the phenomenon of inconsistent product quality caused by the fact that the process fluctuation is not detected in time is avoided, and the high-precision and real-time on-line resistance test is realized.

Description

Technical Field

[0001] The invention relates to the technical field of resistance testing, and in particular to an online resistance testing method and system for an oxygen-free copper rod. Background Art

[0002] Oxygen-free copper rods are widely used in electrical and electronic fields. Due to their superior conductivity, their resistance characteristics are crucial to the performance of the final product. In order to ensure the consistency and quality of oxygen-free copper rods during the production process, resistance testing becomes a necessary link. However, traditional resistance testing methods are usually performed offline, which has obvious shortcomings, especially in terms of real-time performance and quality control.

[0003] In traditional offline resistance testing, the test is to take the sample out of the production line after production is completed and then measure it. A major problem with this method is that it is impossible to monitor the changes in the resistance of the copper rod in real time during the production process. Due to the lack of real-time data support, any fluctuations in process parameters during the production process may not be detected in time, resulting in the copper rods of the entire production batch may have resistance that does not meet the specifications.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present disclosure, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The present invention provides an online resistance testing method and system for an oxygen-free copper rod, thereby effectively solving the problems pointed out in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: An online resistance testing method for an oxygen-free copper rod comprises: Monitor the electromagnetic field changes during the production of oxygen-free copper rods to identify resistance anomalies caused by changes in the internal structure of the material and processing stress; Applying currents of different strengths to the copper rods with abnormal resistance, using a nonlinear response analysis algorithm to perform real-time measurement and preliminary screening of the instantaneous resistance of the copper rods, and identifying the copper rods that may have abnormal resistance; Based on the feature extraction algorithm, the characteristics of the abnormal copper rods initially screened out are analyzed to obtain the resistance characteristic data and the source and nature of the resistance anomaly. The source and nature of the resistance anomaly are compared with a preset quality standard to obtain an online resistance test result.

[0007] Furthermore, the electromagnetic field changes during the production of oxygen-free copper rods are monitored to identify resistance anomalies caused by changes in the internal structure of the material and processing stress, including: Collect electromagnetic field data generated during the copper rod production process and construct an electromagnetic field data set; Preprocessing the data in the electromagnetic field data set and performing feature extraction to identify electromagnetic field features related to changes in the internal structure of the material and processing stress; The extracted electromagnetic field characteristic signal is compared with a preset normal state benchmark, and the abnormal fluctuation in the electromagnetic field is detected using a pattern recognition algorithm; Through the correlation analysis algorithm, abnormal fluctuations in the electromagnetic field are matched with possible anomalies, and abnormal resistance areas caused by changes in the internal structure of the material and processing stress are identified.

[0008] Furthermore, the abnormal fluctuations in the electromagnetic field are matched with possible anomalies through the correlation analysis algorithm, including: Based on historical production data and known resistance anomaly cases, a correlation model between electromagnetic field characteristics and resistance anomaly is constructed; The electromagnetic field features extracted in real time are input into the constructed correlation model, and the similarity and correlation between the electromagnetic field features and the resistance anomaly pattern are calculated using an algorithm.

[0009] Furthermore, the association model includes: The input layer receives the feature vectors extracted from the electromagnetic field data during the copper rod production process; The hidden layer, consisting of multiple nonlinear transformation layers, captures the relationship between electromagnetic field characteristics and resistance anomalies through linear mapping; An output layer maps the high-level features extracted by the hidden layer into specific resistance anomaly types; The association rule part learns and identifies the association rules between the specific electromagnetic field feature combination and the resistance anomaly type by analyzing the historical production data and the resistance anomaly cases; The updating mechanism updates the model parameters during the production process through online learning.

[0010] Furthermore, currents of different strengths are applied to the copper rods with abnormal resistance, and the instantaneous resistance of the copper rods is measured and preliminarily screened in real time using a nonlinear response analysis algorithm to identify copper rods that may have abnormal resistance, including: During the copper rod production process, currents of different strengths are applied to measure the instantaneous resistance of the copper rod in real time, and the resistance response data of the copper rod under different current conditions are obtained; Extracting key features from the resistance response data, wherein the key features reflect the conductive properties of the copper rod under different currents; Applying the nonlinear response analysis algorithm to analyze the extracted key features, and obtaining abnormal resistance characteristic analysis results; According to the abnormal resistance characteristic analysis result and the normal resistance response mode, the copper rods showing abnormal nonlinear characteristics are screened out.

[0011] Furthermore, based on the feature extraction algorithm, the abnormal copper rods initially screened out are analyzed for their characteristics, and the resistance characteristic data is obtained to obtain the source and nature of the resistance anomaly, including: Extract key resistance characteristic data from the abnormal copper rods initially screened; Applying a feature extraction algorithm to perform a multi-dimensional analysis on the key resistance characteristic data, and identifying characteristics that are different from normal copper rod resistance characteristics; According to the process data and material characteristics, and by comparing the abnormal features in the resistance characteristic data with the features of the normal copper rod resistance characteristics, the source and nature of the resistance anomaly are obtained.

[0012] Furthermore, a clustering algorithm is used to group similar abnormal resistance data and refine the abnormal properties, which include abnormal abnormalities and occasional abnormalities.

[0013] Furthermore, the method further comprises: By performing time series analysis on resistance data collected over a long period of time, the occurrence trend of resistance anomalies can be identified, and prediction algorithms can be used to determine possible abnormal situations and pre-adjust production parameters.

[0014] An online resistance testing system for an oxygen-free copper rod, the system comprising: The resistance anomaly identification module monitors the changes in the electromagnetic field during the production of oxygen-free copper rods and identifies resistance anomalies caused by changes in the internal structure of the material and processing stress; Applying currents of different strengths to the copper rods with abnormal resistance, using a nonlinear response analysis algorithm to perform real-time measurement and preliminary screening of the instantaneous resistance of the copper rods, and identifying the copper rods that may have abnormal resistance; Based on the feature extraction algorithm, the characteristics of the abnormal copper rods initially screened out are analyzed to obtain the resistance characteristic data and the source and nature of the resistance anomaly. The source and nature of the resistance anomaly are compared with a preset quality standard to obtain an online resistance test result.

[0015] Furthermore, the resistance abnormality identification module includes: A data set construction unit collects electromagnetic field data generated during the copper rod production process and constructs an electromagnetic field data set; A magnetic field feature recognition unit pre-processes the data in the electromagnetic field data set, performs feature extraction, and recognizes electromagnetic field features related to changes in the internal structure of the material and processing stress; The abnormal fluctuation identification unit compares the extracted electromagnetic field characteristic signal with a preset normal state reference and detects abnormal fluctuations in the electromagnetic field using a pattern recognition algorithm; The abnormal area identification unit matches the abnormal fluctuations in the electromagnetic field with possible anomalies through a correlation analysis algorithm, and identifies the abnormal resistance area caused by changes in the internal structure of the material and processing stress.

[0016] The technical solution of the present invention can achieve the following technical effects: The present invention effectively solves the problem that traditional off-line resistance testing cannot monitor resistance changes in real time during the production process, avoids inconsistent product quality due to untimely detection of process fluctuations, and realizes high-precision, real-time online resistance testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] 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 recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A schematic diagram of a process for online resistance testing of an oxygen-free copper rod; Figure 2 Schematic diagram of the process for identifying resistance anomalies caused by internal structural changes and processing stress of materials; Figure 3 A schematic diagram of a process for matching abnormal fluctuations in the electromagnetic field with possible anomalies; Figure 4 Schematic diagram of the process for identifying copper rods that may have abnormal resistance; Figure 5 Schematic diagram of the process for obtaining the source and nature of resistance anomalies. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0021] Embodiment 1 like Figure 1 As shown, the present invention provides an online resistance testing method for an oxygen-free copper rod, the method comprising: S1: Monitor the changes in electromagnetic fields during the production of oxygen-free copper rods and identify resistance anomalies caused by changes in the internal structure of the material and processing stress; Specifically, this step can capture changes in the internal structure and stress state of the material during processing through real-time monitoring of the electromagnetic field. These changes often lead to abnormal fluctuations in resistance. By identifying these anomalies, it can help discover potential quality problems at an early stage, thereby avoiding the production of unqualified products. The core role of this step is to provide key data support for subsequent testing and analysis, ensuring that the copper rods in the production process maintain stable resistance performance.

[0022] S2: Apply currents of different strengths to the copper rods with abnormal resistance, use a nonlinear response analysis algorithm to measure and preliminarily screen the instantaneous resistance of the copper rods in real time, and identify the copper rods that may have abnormal resistance; Specifically, by applying currents of different intensities, the resistance response characteristics of the copper rod under different current conditions can be observed. Under normal circumstances, the resistance of the copper rod should show a consistent, linear response as the current changes. However, when there are structural defects, stress concentration or other factors that cause abnormal resistance inside the copper rod, its resistance response often shows nonlinear changes or abnormal fluctuations. Using a nonlinear response analysis algorithm to measure these instantaneous resistances in real time can accurately capture these abnormal resistance characteristics. The algorithm can identify abnormal points, mutations or nonlinear areas in the resistance response curve. These features are often signs of abnormal resistance. Through preliminary screening, the system can quickly screen out copper rods that may have abnormal resistance and mark these abnormal copper rods for further detection or processing.

[0023] S3: Based on the feature extraction algorithm, the feature analysis is performed on the abnormal copper rods initially screened out to obtain resistance characteristic data and the source and nature of the resistance anomaly. Specifically, based on the feature extraction algorithm, the abnormal copper rods preliminarily screened out are analyzed for characteristics, resistance characteristic data is obtained, and the source and nature of the resistance anomaly are determined. The main purpose of this step is to deeply analyze and diagnose the specific cause and nature of the resistance anomaly. This process not only confirms the existence of the copper rod resistance anomaly, but also further reveals the root cause of the anomaly, such as microstructural defects inside the material, stress concentration during processing, or the influence of environmental factors. Through feature analysis, the specific patterns and characteristics of abnormal resistance characteristics can be identified. These characteristics can help determine the specific type of anomaly. For example, a certain anomaly may be related to the unevenness of the grain structure, while another anomaly may be caused by excessive external stress. This in-depth analysis allows problems in the production process to be accurately located, thereby providing a basis for taking targeted improvement measures and ultimately ensuring the quality stability of the copper rod.

[0024] S4: Compare the source and nature of the resistance anomaly with the preset quality standard to obtain the online resistance test result.

[0025] Specifically, first, the preset quality standards are usually based on historical production data, industry specifications, and technical requirements of specific products. These standards define the resistance characteristics and allowable fluctuation range that copper rods should have under normal production conditions. These standards may include upper and lower limits of resistance values, the rate of resistance change, and other key parameters related to conductivity. During the comparison process, the system will match the extracted abnormal resistance characteristic data with these standards one by one. For example, if the resistance change exceeds the preset threshold, or the nonlinear characteristics of the resistance are significantly different from the normal mode in the standard, then the system will determine that the copper rod may not meet the quality requirements. Through this comparison, the test results can be clearly drawn: if the resistance characteristics are consistent with the quality standards, the copper rod is considered qualified; conversely, if there are abnormalities beyond the standards, the system will mark the copper rod as unqualified and may generate an alarm or report to prompt further processing.

[0026] The present invention effectively solves the problem that traditional off-line resistance testing cannot monitor resistance changes in real time during the production process, avoids inconsistent product quality due to untimely detection of process fluctuations, and realizes high-precision, real-time online resistance testing.

[0027] As a preferred embodiment of the above, Figure 2 As shown, step S1 monitors the change of electromagnetic field during the production process of oxygen-free copper rods and identifies the resistance anomaly caused by the change of internal structure of the material and processing stress, including: S11: Collect electromagnetic field data generated during the copper rod production process and construct an electromagnetic field data set; S12: preprocessing the data in the electromagnetic field data set and performing feature extraction to identify electromagnetic field features related to changes in the internal structure of the material and processing stress; S13: comparing the extracted electromagnetic field characteristic signal with a preset normal state reference, and detecting abnormal fluctuations in the electromagnetic field using a pattern recognition algorithm; S14: Through the correlation analysis algorithm, the abnormal fluctuations in the electromagnetic field are matched with possible anomalies to identify the abnormal resistance areas caused by the internal structure changes of the material and the processing stress.

[0028] Specifically, first, electromagnetic field data in the copper rod production process is collected in real time through electromagnetic field sensors or other monitoring equipment arranged on the production line. These data may include the magnetic field strength, frequency changes and other related electromagnetic signals generated around the copper rod during different processing stages, such as drawing, cooling and annealing. These signals reflect the internal structure and stress state of the copper rod material during processing and are the basis for identifying resistance anomalies; the collected electromagnetic field data is usually raw and unprocessed, so it must be preprocessed to ensure the accuracy and availability of the data. The preprocessing step includes removing noise signals, such as removing irrelevant noise caused by external interference or equipment vibration through filters, and reducing random fluctuations in the data through smoothing algorithms. The data is standardized so that the electromagnetic field data at different production stages can be compared at the same scale; next, the preprocessed data will enter the feature extraction stage. At this stage, signal processing techniques such as fast Fourier transform (FFT) are used to extract the feature data. ) or wavelet transform to extract key features from the electromagnetic field data. These features may include changes in frequency components, sudden changes in amplitude, or significant increases in energy in certain specific frequency bands. These changes are usually related to changes in the internal structure of the material or stress concentration during processing. Once these characteristic signals are extracted, they will be compared with a preset normal state benchmark. The normal state benchmark is usually based on historical data or established through ideal production process data under laboratory conditions. It reflects the electromagnetic field characteristics of the internal structure and stress state of the material under normal circumstances. By comparison, abnormal fluctuations that deviate from the normal benchmark can be identified; in order to further determine whether these abnormal fluctuations correspond to resistance anomalies, the system uses pattern recognition algorithms to analyze the extracted electromagnetic field features. Pattern recognition algorithms may include machine learning methods such as support vector machines (SVM), neural networks, or decision trees. These algorithms can learn the difference between normal and abnormal electromagnetic field characteristics and can automatically detect possible abnormal patterns in new data; finally, through the association analysis algorithm, the identified abnormal electromagnetic field fluctuations are matched with the known patterns of resistance anomalies. This process involves building a correlation model that associates specific electromagnetic field characteristics with known types of resistance anomalies (such as resistance increases due to changes in grain structure or resistance fluctuations due to stress concentration). By inputting the real-time detected electromagnetic field data into the association model, the system can accurately locate and identify specific resistance abnormality areas.

[0029] As a preferred embodiment of the above, Figure 3 As shown, step S14, matching abnormal fluctuations in the electromagnetic field with possible abnormalities through a correlation analysis algorithm, includes: S141: Based on historical production data and known resistance anomaly cases, a correlation model between electromagnetic field characteristics and resistance anomaly is constructed; S142: Inputting the electromagnetic field features extracted in real time into the constructed correlation model, and using an algorithm to calculate the similarity and correlation between the electromagnetic field features and the resistance anomaly pattern.

[0030] Specifically, first, the system will collect a large amount of historical production data, including electromagnetic field feature data collected in different production batches and the corresponding resistance measurement results. These data may cover the electromagnetic field changes generated by the copper rod under different process conditions (such as temperature, pressure, drawing speed, etc.), and the resistance anomalies corresponding to these changes. Among these historical data, known resistance anomaly cases (such as abnormal resistance caused by internal structural defects, stress concentration, etc.) will be used as key references for training and building association models; next, these extracted electromagnetic field features are associated with the corresponding resistance anomaly data for analysis. By using machine learning algorithms (such as support vector machines, decision trees, neural networks, etc.), the system can learn the relationship between electromagnetic field features and resistance anomalies and build an association model that can identify and capture the correlation between specific electromagnetic field feature patterns and known resistance anomalies. For example, if there is a high correlation between electromagnetic field fluctuations of a certain frequency and a sudden increase in the resistance of a copper rod, the model will be able to identify this correlation and apply it in future real-time detection. Once the correlation model is built, the system will continuously extract the electromagnetic field features in the current production during the real-time monitoring process and input these real-time features into the constructed correlation model. The model calculates the similarity and correlation between these real-time features and known resistance anomaly patterns through algorithms. Specifically, the algorithm will evaluate the degree of match between the current electromagnetic field features and the resistance anomaly patterns recorded in the model. If the similarity or correlation exceeds a preset threshold, the system will determine that there may be a resistance anomaly and mark the copper rods of this batch or location as potential anomalies.

[0031] As a preferred embodiment of the above, the association model includes: The input layer receives the feature vectors extracted from the electromagnetic field data during the copper rod production process; The hidden layer, consisting of multiple nonlinear transformation layers, captures the relationship between electromagnetic field characteristics and resistance anomalies through linear mapping; The output layer maps the high-level features extracted by the hidden layer into specific resistance anomaly types; In the association rule part, by analyzing historical production data and resistance anomaly cases, the association rules between specific electromagnetic field feature combinations and resistance anomaly types are learned and identified; The updating mechanism updates the model parameters during the production process through online learning.

[0032] Specifically, the input layer is responsible for receiving the electromagnetic field feature vectors extracted from the copper rod production process. These features may include the amplitude, frequency, phase, etc. of the electromagnetic field. As the basic data for model analysis, these feature vectors provide key information related to the resistance of the copper rod and provide necessary input for subsequent processing; the hidden layer is composed of multiple nonlinear transformation layers, which process the input features through linear mapping and nonlinear activation functions (such as ReLU, sigmoid, etc.). The multi-level structure of the hidden layer can capture the complex nonlinear relationship between the electromagnetic field characteristics and the resistance anomaly, and reveal the potential correlation; the output layer maps the high-level features extracted by the hidden layer to specific resistance anomaly types, providing clear The classification results are used to determine whether there is resistance anomaly in the copper rod and identify the specific type of anomaly; the association rule part analyzes historical data and resistance anomaly cases to learn and identify the association rules between the electromagnetic field feature combination and the resistance anomaly type. These rules are generated through data mining technology and machine learning algorithms, so that the model can effectively associate input features with specific anomalies; the update mechanism continuously updates the model parameters through online learning, and uses new data obtained in real time to further train the model to ensure that it adapts to changes in the production environment and maintains the ability to detect new anomaly types. This continuous optimization enables the model to provide high-precision anomaly identification under different production conditions, ensuring the stability of the production process and product quality.

[0033] As a preferred embodiment of the above, Figure 4 As shown, step S2 applies currents of different strengths to the copper rods with abnormal resistance, uses a nonlinear response analysis algorithm to measure and preliminarily screen the instantaneous resistance of the copper rods in real time, and identifies the copper rods that may have abnormal resistance, including: S21: During the copper rod production process, currents of different strengths are applied to measure the instantaneous resistance of the copper rod in real time, and the resistance response data of the copper rod under different current conditions are obtained; S22: Extract key features from the resistance response data, which reflect the conductivity characteristics of the copper rod under different currents; S23: applying a nonlinear response analysis algorithm to analyze the extracted key features and obtain abnormal resistance characteristic analysis results; S24: Based on the abnormal resistance characteristic analysis results and the normal resistance response mode, the copper rods showing abnormal nonlinear characteristics are screened out.

[0034] Specifically, during the copper rod production process, currents of different intensities are applied to the copper rods, and their instantaneous resistance is measured in real time. This process gradually increases or decreases the current intensity to observe the resistance response characteristics of the copper rods under different current conditions. The measured resistance data not only includes the instantaneous resistance value, but also records the dynamic response of the resistance with the change of current. These data can reveal the conductive properties of the copper rod material under different current environments, especially when the internal structure or stress state of the material changes, the resistance may show an abnormal response; key features are extracted from the collected resistance response data, including the rate of change of resistance, the degree of nonlinearity, and the resistance fluctuation under a specific current. These features are further analyzed and extracted through signal processing techniques, such as Fourier transform or wavelet analysis, so as to convert the raw data into more meaningful feature information, thereby better reflecting the conductivity and potential resistance anomalies of the copper rod; The extracted key features are deeply analyzed using a nonlinear response analysis algorithm that can identify complex patterns in resistance response, especially those nonlinear changes that are inconsistent with linear models. For example, if the resistance of a copper rod shows an abnormal nonlinear increase at a specific current, this may indicate that there is some defect or abnormality inside the material. Through analysis, the algorithm generates abnormal resistance characteristic analysis results and identifies copper rods that may have problems. These analysis results are compared in detail with the preset normal resistance response pattern. The normal pattern is usually constructed from historical production data or ideal response data of standard samples. Through comparison, copper rods that show abnormal nonlinear characteristics can be screened out. The resistance characteristics of these copper rods are significantly different from normal conductive characteristics, and there may be potential quality problems. The screened abnormal copper rods will be marked as requiring further inspection or processing to ensure the quality and consistency of the final product.

[0035] As a preferred embodiment of the above, Figure 5 As shown, step S3, based on the feature extraction algorithm, performs feature analysis on the abnormal copper rods preliminarily screened out, obtains resistance characteristic data, and obtains the source and nature of the resistance anomaly, including: S31: extract key resistance characteristic data from the abnormal copper rods initially screened; S32: Apply feature extraction algorithm to perform multi-dimensional analysis on key resistance characteristic data and identify characteristics that are different from normal copper rod resistance characteristics; S33: According to the process data and the material characteristics, and by comparing the abnormal characteristics in the resistance characteristic data with the characteristics of the normal copper rod resistance characteristics, the source and nature of the resistance abnormality are obtained.

[0036] Specifically, firstly, key resistance characteristic data are extracted from the abnormal copper rods that have been preliminarily screened. These data include the time variation of the resistance value, the response characteristics under different current intensities, and other related parameters that may affect the resistance. By extracting these key data, the system can capture the electrical performance characteristics that may cause resistance abnormality. Then, a feature extraction algorithm is applied to perform multi-dimensional analysis on these key resistance characteristic data. Multi-dimensional analysis means that the system will conduct in-depth analysis of the resistance characteristics from multiple angles (such as time, frequency, spatial distribution, etc.). This analysis process helps to identify the key differences between abnormal copper rods and normal copper rods, such as nonlinear response under specific current conditions, abnormal resistance fluctuation mode, or abnormal resistance characteristics within a specific frequency band. In step S33, the extracted abnormal resistance characteristics are compared in detail with the resistance characteristics of normal copper rods in combination with process data and material characteristics. The process data may include the temperature, pressure, drawing speed, etc. used in the production process, while the material characteristics involve the purity and grain structure of the copper rod. Through this comparison, the specific source of the resistance abnormality is identified, such as internal defects caused by excessive drawing stress, or resistance increase caused by insufficient material purity.

[0037] As a preferred embodiment of the above embodiment, a clustering algorithm is used to group similar abnormal resistance data and refine the abnormal properties, which include abnormal abnormalities and occasional abnormalities.

[0038] Specifically, clustering algorithms group these extracted abnormal resistance characteristic data. Clustering algorithms such as K-means clustering, hierarchical clustering or DBSCAN can cluster similar abnormal data points together to form several categories or clusters. Each of these categories represents a group of copper rods with similar resistance abnormal characteristics. Through cluster analysis, the nature of the abnormality can be further refined. Specifically, systematic abnormalities are usually manifested as abnormal characteristics that appear repeatedly in multiple copper rod samples. Such abnormalities are often caused by persistent problems in the production process, such as improper equipment parameter settings or material batch quality problems. On the other hand, sporadic abnormalities are abnormalities that occur occasionally under specific conditions. They may be caused by random external factors or short-term process fluctuations, such as occasional current fluctuations or temporary temperature changes. This refinement process not only determines the root cause of the resistance abnormality more accurately, but also provides clearer guidance for operators. For example, for systematic abnormalities, it may be necessary to check each step of the production process or the status of the equipment, while for sporadic abnormalities, it may only be necessary to focus on specific operating conditions or make some temporary adjustments.

[0039] As a preference of the above embodiment, the method further includes: By performing time series analysis on resistance data collected over a long period of time, the occurrence trend of resistance anomalies can be identified, and prediction algorithms can be used to determine possible abnormal situations and pre-adjust production parameters.

[0040] Specifically, first, the resistance data generated in the copper rod production process are continuously collected. These data include the resistance values ​​and their changes in different production batches and time periods. These long-term data form a time series, recording the trend and pattern of resistance changes over time; next, through time series analysis, the potential trends and periodic changes in the resistance data are identified. For example, if the resistance value gradually increases over time or shows some periodic fluctuations, these trends can be identified to determine whether they indicate the possibility of resistance anomalies. Time series analysis methods may include moving average, exponential smoothing, ARIMA models, etc., which are used to reveal long-term trends, seasonal patterns or sudden abnormal changes in the data; after identifying these trends, prediction algorithms are further used to determine possible abnormal situations that may occur in the future. Prediction algorithms can use linear regression, support vector regression (SVR), long short-term memory network (LSTM), etc., which can predict future resistance changes based on existing data trends. If the prediction results show that the resistance is about to exceed the normal range or there is a significant risk of abnormality, the system will issue an early warning. Based on these prediction results, production parameters can be adjusted in advance to avoid potential resistance abnormalities. For example, if the prediction algorithm shows that resistance may increase due to continued process stress, process parameters such as drawing speed, cooling rate or current intensity can be automatically adjusted to reduce stress and prevent abnormalities.

[0041] Embodiment 2 Based on the same inventive concept as the online resistance testing method of an oxygen-free copper rod in the aforementioned embodiment, the present invention further provides an online resistance testing system for an oxygen-free copper rod, the system comprising: The resistance anomaly identification module monitors the changes in the electromagnetic field during the production of oxygen-free copper rods and identifies resistance anomalies caused by changes in the internal structure of the material and processing stress; Apply currents of different strengths to copper rods with abnormal resistance, use nonlinear response analysis algorithms to measure and preliminarily screen the instantaneous resistance of the copper rods in real time, and identify copper rods that may have abnormal resistance; Based on the feature extraction algorithm, the characteristics of the abnormal copper rods initially screened out are analyzed to obtain the resistance characteristic data and the source and nature of the resistance anomaly. The source and nature of the resistance anomaly are compared with the preset quality standards to obtain the online resistance test results.

[0042] The above-mentioned testing system in the present invention can effectively implement the online resistance testing method of the oxygen-free copper rod, and the technical effects that can be achieved are as described in the above-mentioned embodiments, which will not be repeated here.

[0043] Furthermore, the resistance abnormality identification module includes: A data set construction unit collects electromagnetic field data generated during the copper rod production process and constructs an electromagnetic field data set; The magnetic field feature recognition unit pre-processes the data in the electromagnetic field data set and performs feature extraction to identify the electromagnetic field features related to the internal structure changes of the material and the processing stress; The abnormal fluctuation identification unit compares the extracted electromagnetic field characteristic signal with a preset normal state reference and detects abnormal fluctuations in the electromagnetic field using a pattern recognition algorithm; The abnormal area identification unit matches the abnormal fluctuations in the electromagnetic field with possible anomalies through a correlation analysis algorithm, and identifies the abnormal resistance area caused by changes in the internal structure of the material and processing stress.

[0044] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the method in Example 1, which will not be repeated here.

[0045] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined therein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. An online resistance testing method for an oxygen-free copper rod, characterized in that: include: Monitor the electromagnetic field changes during the production of oxygen-free copper rods to identify resistance anomalies caused by changes in the internal structure of the material and processing stress; Applying currents of different strengths to the copper rods with abnormal resistance, using a nonlinear response analysis algorithm to perform real-time measurement and preliminary screening of the instantaneous resistance of the copper rods, and identifying the copper rods that may have abnormal resistance; Based on the feature extraction algorithm, the characteristics of the abnormal copper rods initially screened out are analyzed to obtain the resistance characteristic data and the source and nature of the resistance anomaly. The source and nature of the resistance anomaly are compared with a preset quality standard to obtain an online resistance test result.

2. The online resistance testing method of oxygen-free copper rod according to claim 1, characterized in that: Monitor the changes in electromagnetic fields during the production of oxygen-free copper rods and identify resistance anomalies caused by changes in the internal structure of the material and processing stress, including: Collect electromagnetic field data generated during the copper rod production process and construct an electromagnetic field data set; Preprocessing the data in the electromagnetic field data set and performing feature extraction to identify electromagnetic field features related to changes in the internal structure of the material and processing stress; The extracted electromagnetic field characteristic signal is compared with a preset normal state benchmark, and the abnormal fluctuation in the electromagnetic field is detected using a pattern recognition algorithm; Through the correlation analysis algorithm, abnormal fluctuations in the electromagnetic field are matched with possible anomalies, and abnormal resistance areas caused by changes in the internal structure of the material and processing stress are identified.

3. The online resistance testing method of oxygen-free copper rod according to claim 2, characterized in that: Through the correlation analysis algorithm, abnormal fluctuations in the electromagnetic field are matched with possible anomalies, including: Based on historical production data and known resistance anomaly cases, a correlation model between electromagnetic field characteristics and resistance anomaly is constructed; The electromagnetic field features extracted in real time are input into the constructed correlation model, and the similarity and correlation between the electromagnetic field features and the resistance anomaly pattern are calculated using an algorithm.

4. The online resistance testing method of oxygen-free copper rod according to claim 3, characterized in that: The association model includes: The input layer receives the feature vectors extracted from the electromagnetic field data during the copper rod production process; The hidden layer, consisting of multiple nonlinear transformation layers, captures the relationship between electromagnetic field characteristics and resistance anomalies through linear mapping; An output layer maps the high-level features extracted by the hidden layer into specific resistance anomaly types; The association rule part learns and identifies the association rules between the specific electromagnetic field feature combination and the resistance anomaly type by analyzing the historical production data and the resistance anomaly cases; The updating mechanism updates the model parameters during the production process through online learning.

5. The online resistance testing method of oxygen-free copper rod according to claim 1, characterized in that: Apply currents of different strengths to the copper rod with abnormal resistance, use a nonlinear response analysis algorithm to measure and preliminarily screen the instantaneous resistance of the copper rod in real time, and identify the copper rod that may have abnormal resistance, including: During the copper rod production process, currents of different strengths are applied to measure the instantaneous resistance of the copper rod in real time, and the resistance response data of the copper rod under different current conditions are obtained; Extracting key features from the resistance response data, wherein the key features reflect the conductive properties of the copper rod under different currents; Applying the nonlinear response analysis algorithm to analyze the extracted key features, and obtaining abnormal resistance characteristic analysis results; According to the abnormal resistance characteristic analysis result and the normal resistance response mode, the copper rods showing abnormal nonlinear characteristics are screened out.

6. The online resistance testing method of oxygen-free copper rod according to claim 1, characterized in that: Based on the feature extraction algorithm, the abnormal copper rods initially screened out are analyzed for their characteristics, and the resistance characteristic data is obtained to obtain the source and nature of the resistance anomaly, including: Extract key resistance characteristic data from the abnormal copper rods initially screened; Applying a feature extraction algorithm to perform a multi-dimensional analysis on the key resistance characteristic data, and identifying characteristics that are different from normal copper rod resistance characteristics; According to the process data and material characteristics, and by comparing the abnormal features in the resistance characteristic data with the features of the normal copper rod resistance characteristics, the source and nature of the resistance anomaly are obtained.

7. The online resistance testing method of oxygen-free copper rod according to claim 1, characterized in that: A clustering algorithm is used to group similar abnormal resistance data and refine the abnormal properties, which include abnormal abnormalities and occasional abnormalities.

8. The online resistance testing method of oxygen-free copper rod according to claim 1, characterized in that: The method further comprises: By performing time series analysis on resistance data collected over a long period of time, the occurrence trend of resistance anomalies can be identified, and prediction algorithms can be used to determine possible abnormal situations and pre-adjust production parameters.

9. An online resistance test system for oxygen-free copper rod, characterized in that: The system comprises: The resistance anomaly identification module monitors the changes in the electromagnetic field during the production of oxygen-free copper rods and identifies resistance anomalies caused by changes in the internal structure of the material and processing stress; Applying currents of different strengths to the copper rods with abnormal resistance, using a nonlinear response analysis algorithm to perform real-time measurement and preliminary screening of the instantaneous resistance of the copper rods, and identifying the copper rods that may have abnormal resistance; Based on the feature extraction algorithm, the characteristics of the abnormal copper rods initially screened out are analyzed to obtain the resistance characteristic data and the source and nature of the resistance anomaly. The source and nature of the resistance anomaly are compared with a preset quality standard to obtain an online resistance test result.

10. The online resistance testing system of oxygen-free copper rod according to claim 9, characterized in that: The resistance anomaly identification module includes: A data set construction unit collects electromagnetic field data generated during the copper rod production process and constructs an electromagnetic field data set; A magnetic field feature recognition unit pre-processes the data in the electromagnetic field data set, performs feature extraction, and recognizes electromagnetic field features related to changes in the internal structure of the material and processing stress; The abnormal fluctuation identification unit compares the extracted electromagnetic field characteristic signal with a preset normal state reference and detects abnormal fluctuations in the electromagnetic field using a pattern recognition algorithm; The abnormal area identification unit matches the abnormal fluctuations in the electromagnetic field with possible anomalies through a correlation analysis algorithm, and identifies the abnormal resistance area caused by changes in the internal structure of the material and processing stress.

Images