A method and system for residual stress detection and process traceability of main beam steel
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-10
- Publication Date
- 2026-08-14
AI Technical Summary
目前大梁钢质量检测仍以成品抽检为主，虽有企业采用磁测法检测残余应力作为辅助手段，但检测数据易受表面状态干扰，精度不足，且未建立工艺缺陷与残余应力特征的关联体系
Smart Images
Figure CN122572001A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method and system for detecting residual stress and tracing the process of main beam steel, belonging to the cross-technical field of metal material testing and manufacturing methods and equipment. Background Technology
[0002] As a core structural material for equipment such as automobiles and construction machinery, the consistency of product quality and the stability of the production process of beam steel directly determine the safety and service life of the equipment, and are also a core requirement for enterprises to enhance their market competitiveness. In the industrial production of beam steel, process fluctuations such as abnormal coiling temperature, uneven cooling, and roll wear can easily lead to batch quality defects, which not only increase rework and scrap costs, but also disrupt the production rhythm. Therefore, it is urgent to rely on scientific testing methods to achieve accurate identification of process deviations, rapid traceability of problems and defects, and targeted optimization of process parameters.
[0003] Residual stress is highly correlated with production process parameters, and process deviations directly manifest as abnormal distributions of residual stress. Therefore, residual stress is an important indicator reflecting process rationality and assisting in quality control. Currently, quality inspection of main beam steel still relies primarily on finished product sampling. Although some companies use magnetic testing to detect residual stress as an auxiliary method, the test data is easily affected by surface conditions, resulting in insufficient accuracy, and a correlation system between process defects and residual stress characteristics has not been established. When sampling reveals substandard finished product quality, tracing the root cause of the process often relies on manual experience, which is inefficient and inaccurate. Process optimization lacks precise data support, feedback control is lagging, and it is impossible to achieve a reverse traceability closed loop from substandard results to process defects and then to process optimization. Similar quality problems recur, making it difficult to meet the high-quality supply requirements of high-end equipment manufacturing. Summary of the Invention
[0004] The purpose of this invention is to provide a method and system for residual stress detection and process traceability of main beam steel. By obtaining the two-dimensional residual stress distribution of finished products based on magnetic measurement, employing diagnostic technology to trace the root cause, and implementing hierarchical feedback control, it achieves residual stress detection of non-conforming products, precise location of process defects, and targeted optimization of the production process. It constructs a reverse traceability closed loop from non-conforming results to process optimization, ensuring the consistency of main beam steel product quality, improving the stability and controllability of the production process, and realizing closed-loop quality control from "stress detection" to "precise diagnosis" to "offline optimization." This effectively solves the problems of high false alarm rate, inaccurate traceability, and delayed feedback in traditional methods, significantly improving the production quality and process controllability of main beam steel, and effectively addressing the aforementioned problems existing in the background technology.
[0005] The technical solution of this invention is: a method for detecting residual stress and tracing processes in main beam steel, comprising the following steps:
[0006] S1. Database and Standards Construction Phase: Based on historical production measurement data, laboratory simulation defect data, and finite element simulation data, construct a dynamically evolving "process defect-stress characteristics" database; formulate quality judgment standards for residual stress under different processes, and establish an online learning and updating mechanism for the database;
[0007] S2. Inspection and Comparison Stage: Residual stress is tested on the main beam steel products with unsatisfactory performance. After compensation and correction, the two-dimensional residual stress distribution is obtained. The test results are compared with the quality judgment standards to identify areas of abnormal stress.
[0008] S3. Report Generation and Process Optimization Stage: Generate an inspection report based on the comparison results, clarifying the type and extent of the anomaly; based on the inspection report, integrate the real-time production process parameters of the batch, use a Bayesian inference network to calculate the posterior probability of each process step causing the anomaly, and trace the most likely root cause of the process; based on the root cause type, execute offline feedback control: generate process optimization suggestions for the next batch for systematic deviations in the finished product stage.
[0009] The "process defect-stress characteristic" database in step S1 contains typical residual stress distribution patterns and their characteristic parameters corresponding to process defects such as abnormal winding temperature, uneven cooling, roll wear, or heating temperature difference. The dynamic evolution mechanism refers to the automatic marking and push for manual review when the system detects a new unmatched stress pattern. After confirmation, the new pattern and its corresponding solution are automatically entered into the database.
[0010] The quality judgment criteria established in step S1 include the allowable range and calculation formula of the following characteristic parameters:
[0011] (1) Residual stress amplitude σ M Characterized by the difference between the maximum and minimum residual stress values.
[0012]
[0013] Where σ max and σ min These are the maximum and minimum residual stress values, respectively, in MPa.
[0014] (2) Residual stress root mean square deviation σ S Characterizes the degree of deviation of residual stress.
[0015]
[0016] Where σ i and σ j The residual stress at each test point is expressed in MPa; i and j are the test point numbers, and n is the total number of test points.
[0017] (3) Residual stress symmetry σsym Characterized by the difference between the residual stress in the middle and the residual stress at the edge.
[0018]
[0019] Where σ center σ ds and σ os These are the residual stresses in the middle section, OS side, and DS side, respectively, in MPa.
[0020] (4) Residual stress asymmetry σ asy Characterizing the difference between the two sides of the residual stress
[0021] .
[0022] The compensation correction in step S2 involves using a differential algorithm to calculate the thickness and roughness of the oxide scale on the steel plate surface by introducing laser ranging or a vision sensor. The magnetic signal is then corrected according to a preset attenuation coefficient to eliminate the influence of the surface condition on stress measurement.
[0023] The detection report in step S3 includes the location of the anomaly, the type of anomaly, the degree of deviation from the standard deviation, the top three most likely potential process root causes, and recommended remedial measures.
[0024] The offline feedback control in step S3 refers to:
[0025] Batch optimization: If a systematic deviation is found during the finished product inspection stage, a process parameter adjustment report is automatically generated, and the initial settings for the next batch of production are locked.
[0026] It also includes a feedback optimization mechanism that inputs process optimization suggestions into the production system to adjust the process parameters of subsequent batches, thereby achieving closed-loop process control.
[0027] A residual stress detection and process traceability system for main beam steel includes a database and standards management module, a magnetic testing platform, a data analysis and report generation module, and a process traceability and optimization engine. These modules are sequentially connected. The database and standards management module stores "process defect-stress characteristic" data and quality judgment standards, and performs online learning and updates. The magnetic testing platform includes a magnetic meter and probe sensors for performing full-area residual stress scanning. The data analysis and report generation module performs data preprocessing, feature extraction, and intelligent comparison algorithms, and generates structured inspection reports. The process traceability and optimization engine integrates real-time production data, performs Bayesian inference diagnosis, and outputs process problem analysis and optimization suggestions.
[0028] The beneficial effects of this invention are as follows: By obtaining the two-dimensional residual stress distribution of finished products based on magnetic measurement, tracing back to the root cause using diagnostic technology, and implementing hierarchical feedback control, it achieves residual stress detection of non-conforming products, precise location of process defects, and targeted optimization of production processes. It constructs a reverse traceability closed loop from non-conforming results to process optimization, ensuring the consistency of beam steel product quality, improving the stability and controllability of production processes, and realizing closed-loop quality control from "stress detection" to "precise diagnosis" and then to "offline optimization." It effectively solves the problems of high false alarm rate, inaccurate traceability, and delayed feedback in traditional methods, and significantly improves the production quality and process controllability of beam steel.
[0029] Instruction manual illustrations
[0030] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation
[0031] To make the purpose, technical solutions, and advantages of the invention's embodiments clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only a small part of the embodiments of the present invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.
[0032] A method for detecting residual stress and tracing processes in main beam steel includes the following steps:
[0033] S1. Database and Standards Construction Phase: Based on historical production measurement data, laboratory simulation defect data, and finite element simulation data, construct a dynamically evolving "process defect-stress characteristics" database; formulate quality judgment standards for residual stress under different processes, and establish an online learning and updating mechanism for the database;
[0034] S2. Inspection and Comparison Stage: Residual stress is tested on the main beam steel products with unsatisfactory performance. After compensation and correction, the two-dimensional residual stress distribution is obtained. The test results are compared with the quality judgment standards to identify areas of abnormal stress.
[0035] S3. Report Generation and Process Optimization Stage: Generate an inspection report based on the comparison results, clarifying the type and extent of the anomaly; based on the inspection report, integrate the real-time production process parameters of the batch, use a Bayesian inference network to calculate the posterior probability of each process step causing the anomaly, and trace the most likely root cause of the process; based on the root cause type, execute offline feedback control: generate process optimization suggestions for the next batch for systematic deviations in the finished product stage.
[0036] The "process defect-stress characteristic" database in step S1 contains typical residual stress distribution patterns and their characteristic parameters corresponding to process defects such as abnormal winding temperature, uneven cooling, roll wear, or heating temperature difference. The dynamic evolution mechanism refers to the automatic marking and push for manual review when the system detects a new unmatched stress pattern. After confirmation, the new pattern and its corresponding solution are automatically entered into the database.
[0037] The quality judgment criteria established in step S1 include the allowable range and calculation formula of the following characteristic parameters:
[0038] (1) Residual stress amplitude σ M Characterized by the difference between the maximum and minimum residual stress values.
[0039]
[0040] Where σ max and σ min These are the maximum and minimum residual stress values, respectively, in MPa.
[0041] (2) Residual stress root mean square deviation σ S Characterizes the degree of deviation of residual stress.
[0042]
[0043] Where σ i and σ j The residual stress at each test point is expressed in MPa; i and j are the test point numbers, and n is the total number of test points.
[0044] (3) Residual stress symmetry σ sym Characterized by the difference between the residual stress in the middle and the residual stress at the edge.
[0045]
[0046] Where σ center σ ds and σ os These are the residual stresses in the middle section, OS side, and DS side, respectively, in MPa.
[0047] (4) Residual stress asymmetry σ asy Characterizing the difference between the two sides of the residual stress
[0048] .
[0049] The compensation correction in step S2 involves using a differential algorithm to calculate the thickness and roughness of the oxide scale on the steel plate surface by introducing laser ranging or a vision sensor. The magnetic signal is then corrected according to a preset attenuation coefficient to eliminate the influence of the surface condition on stress measurement.
[0050] The detection report in step S3 includes the location of the anomaly, the type of anomaly, the degree of deviation from the standard deviation, the top three most likely potential process root causes, and recommended remedial measures.
[0051] The offline feedback control in step S3 refers to:
[0052] Batch optimization: If a systematic deviation is found during the finished product inspection stage, a process parameter adjustment report is automatically generated, and the initial settings for the next batch of production are locked.
[0053] It also includes a feedback optimization mechanism that inputs process optimization suggestions into the production system to adjust the process parameters of subsequent batches, thereby achieving closed-loop process control.
[0054] A residual stress detection and process traceability system for main beam steel includes a database and standards management module, a magnetic testing platform, a data analysis and report generation module, and a process traceability and optimization engine. These modules are sequentially connected. The database and standards management module stores "process defect-stress characteristic" data and quality judgment standards, and performs online learning and updates. The magnetic testing platform includes a magnetic meter and probe sensors for performing full-area residual stress scanning. The data analysis and report generation module performs data preprocessing, feature extraction, and intelligent comparison algorithms, and generates structured inspection reports. The process traceability and optimization engine integrates real-time production data, performs Bayesian inference diagnosis, and outputs process problem analysis and optimization suggestions.
[0055] Example:
[0056] The present invention performs the following steps:
[0057] S1 Database and Standards Construction Phase
[0058] 1. Data Acquisition and Integration: Collect historical production measurement data of beam steel production (including residual stress detection data under various process parameters, process defect records, and finished product quality judgment results), laboratory simulation defect data (artificially simulating typical process defects such as abnormal coiling temperature, uneven cooling, roll wear, and heating temperature difference to obtain corresponding residual stress distribution data), and finite element simulation data (simulating the evolution process of residual stress in beam steel under different process deviations through simulation software to supplement the deficiencies of the measured data). Standardize and clean the three types of data, and unify indicators such as stress units, process parameter dimensions, and detection point coordinates.
[0059] 2. Database Construction: Based on the cleaned data, a dynamic "process defect-stress characteristic" database is constructed. The core of the database stores typical process defect types (abnormal winding temperature, uneven cooling, etc.), corresponding residual stress distribution patterns (two-dimensional stress cloud map characteristics), key stress characteristic parameters (amplitude, root mean square error, symmetry, asymmetry) and parameter thresholds generated by process defects, and establishes a one-to-one correspondence between process defects and stress characteristics.
[0060] 3. Development of Multidimensional Quality Judgment Standards: Based on the distribution law of stress characteristic parameters of qualified beam steel in the database, the allowable ranges and calculation formulas of residual stress amplitude σM, root mean square deviation σS, symmetry σsym, and asymmetry σasy are formulated. It is clear that any parameter exceeding the allowable range is judged as stress anomaly, which serves as the core basis for subsequent testing and comparison.
[0061] 4. Deployment of online learning and update mechanism: Configure an intelligent recognition module for the database. When the system detects a new stress mode that does not match, it will automatically mark the mode and push it to the background manual review end. After the manual completes the process root cause tracing and feature parameter extraction of the new stress mode, the new stress mode-process root cause correlation data will be entered into the database to realize the dynamic evolution of the database and adapt to the new process defect types added in production.
[0062] S2 Detection and Comparison Stage
[0063] 1. Determination of testing objects: In the sampling inspection of finished steel beams, finished products that are determined to be substandard are included in the residual stress testing process of this method as testing objects.
[0064] 2. Full-width magnetic measurement and scanning: The magnetic measuring instrument and probe sensor of the magnetic measurement and testing platform are used to perform a full-width two-dimensional residual stress scan on the unqualified finished products. The detection point spacing is set to 30mm to ensure the integrity of stress distribution data. The surface data of the finished products (oxide scale thickness, roughness) are collected simultaneously during the scanning process.
[0065] 3. The differential algorithm is used to calculate the original magnetic measurement signal. Combined with the surface oxide thickness and roughness data, the original signal is surface compensated and corrected according to the preset magnetic signal attenuation coefficient. The result is accurate two-dimensional residual stress distribution data and stress characteristic parameters of each detection point.
[0066] 4. Stress Anomaly Identification: The pre-processed stress characteristic parameters (σ) are then used to identify anomalies. M σ S σ sym σ asyThe stress distribution pattern is compared with the multi-dimensional quality judgment criteria established in the S1 stage one by one. Through intelligent algorithms, the stress distribution pattern is compared with the typical pattern in the database to accurately identify the stress anomaly area (such as the edge and the middle), the anomaly type (such as the amplitude exceeding the standard, the symmetry not meeting the standard), and the degree of anomaly deviation.
[0067] S3 Report Generation and Process Optimization Phase
[0068] 1. Structured Inspection Report Generation: The data analysis and report generation module automatically generates standardized inspection reports based on the comparison results. The core of the report includes basic information on non-conforming products, location, type, and degree of stress anomalies, similar process defects matched in the database, the top 3 most likely potential process root causes, and preliminary treatment measures, providing data support for process traceability.
[0069] 2. Precise traceability of process defects: The process traceability and optimization engine retrieves the real-time production process parameters of the batch of main beam steel, integrates the stress anomaly characteristics in the inspection report with the real-time process parameters, and inputs them into the Bayesian inference network; the Bayesian inference network calculates the posterior probability of stress anomalies caused by each process step, sorts them from high to low probability, and identifies the most likely process root cause.
[0070] 3. Offline process optimization and closed-loop control: For the core process defects identified through tracing, the system automatically generates optimization suggestions for the next batch of process parameters and forms a process parameter adjustment report.
[0071] The above embodiments are only used to illustrate and not limit the technical solutions of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention without departing from the spirit and scope of the present invention. Any modifications or partial substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A method for detecting residual stress and tracing the process of main beam steel, characterized in that... Includes the following steps: S1. Database and Standards Construction Phase: Based on historical production measurement data, laboratory simulation defect data, and finite element simulation data, construct a dynamically evolving "process defect-stress characteristics" database; Establish quality judgment standards for residual stress under different processes, and create an online learning and updating mechanism for the database; S2. Inspection and Comparison Stage: Residual stress is tested on the main beam steel products with unsatisfactory performance. After compensation and correction, the two-dimensional residual stress distribution is obtained. The test results are compared with the quality judgment standards to identify areas of abnormal stress. S3. Report Generation and Process Optimization Stage: Generate a test report based on the comparison results to clarify the type and extent of the anomaly; based on the test report, integrate the real-time production process parameters of the batch, use a Bayesian inference network to calculate the posterior probability of each process step causing the anomaly, and trace the most likely process root cause; Based on the root cause type, implement offline feedback control: generate process optimization suggestions for the next batch based on systematic deviations in the finished product stage.
2. The method for detecting residual stress and tracing processes in main beam steel according to claim 1, characterized in that: The "process defect-stress characteristic" database in step S1 contains typical residual stress distribution patterns and their characteristic parameters corresponding to process defects such as abnormal winding temperature, uneven cooling, roll wear, or heating temperature difference. The dynamic evolution mechanism refers to the automatic marking and manual review when the system detects a new stress pattern that does not match. After confirmation, the new pattern and its corresponding solution are automatically entered into the database.
3. The method for detecting residual stress and tracing processes in main beam steel according to claim 1, characterized in that: The quality judgment criteria established in step S1 include the allowable range and calculation formula of the following characteristic parameters: (1) Residual stress amplitude σ M Characterized by the difference between the maximum and minimum residual stress values. Where σ max and σ min These are the maximum and minimum residual stress values, respectively, in MPa. (2) Residual stress root mean square deviation σ S Characterizes the degree of deviation of residual stress. Where σ i and σ j The residual stress at each test point is expressed in MPa; i and j are the test point numbers, and n is the total number of test points. (3) Residual stress symmetry σ sym Characterized by the difference between the residual stress in the middle and the residual stress at the edge. Where σ center σ ds and σ os These are the residual stresses in the middle section, OS side, and DS side, respectively, in MPa. (4) Residual stress asymmetry σ asy Characterizing the difference between the two sides of the residual stress 。 4. The method for detecting residual stress and tracing processes in main beam steel according to claim 1, characterized in that: The compensation correction in step S2 involves using a differential algorithm to calculate the thickness and roughness of the oxide scale on the steel plate surface by introducing laser ranging or a vision sensor. The magnetic signal is then corrected according to a preset attenuation coefficient to eliminate the influence of the surface condition on stress measurement.
5. The method for detecting residual stress and tracing processes in main beam steel according to claim 1, characterized in that: The detection report in step S3 includes the location of the anomaly, the type of anomaly, the degree of deviation from the standard deviation, the top three most likely potential process root causes, and recommended remedial measures.
6. The method for detecting residual stress and tracing processes in main beam steel according to claim 1, characterized in that: The offline feedback control in step S3 refers to: Batch optimization: If a systematic deviation is found during the finished product inspection stage, a process parameter adjustment report is automatically generated, and the initial settings for the next batch of production are locked.
7. The method for detecting residual stress and tracing processes in main beam steel according to claim 1, characterized in that: It also includes a feedback optimization mechanism that inputs process optimization suggestions into the production system to adjust the process parameters of subsequent batches, thereby achieving closed-loop process control.
8. A residual stress detection and process traceability system for main beam steel, characterized in that: The system comprises a database and standards management module, a magnetic testing platform, a data analysis and report generation module, and a process traceability and optimization engine. These modules are sequentially connected. The database and standards management module stores "process defect-stress characteristic" data and quality judgment standards, and performs online learning and updates. The magnetic testing platform includes a magnetic meter and probe sensors, used to perform full-area residual stress scanning. The data analysis and report generation module performs data preprocessing, feature extraction, and intelligent comparison algorithms, and generates structured inspection reports. The process traceability and optimization engine integrates real-time production data, performs Bayesian inference diagnosis, and outputs process problem analysis and optimization suggestions.