Annealing furnace temperature intelligent control system and method based on big data analysis

Abstract

The invention discloses an annealing furnace temperature intelligent control system and method based on big data analysis, and relates to the technical field of annealing furnace temperature control. Historical data of annealing treatment of a blank is collected, the annealing process is divided into annealing stages according to the temperature, the annealing rate of the annealing stages is counted, and the annealing rate interval is divided; the method comprises the following steps: selecting an optimal annealing rate interval according to the relationship between the annealing rate interval and the qualified rate of a finished product, analyzing the influence of the annealing rate on the qualified rate of the product, selecting an annealing stage influenced by the annealing rate, analyzing a temperature adjustment process of a blank, and adjusting temperature parameters of an annealing furnace; according to the method, the qualified rate of a next finished product is predicted, a temperature parameter log related to the qualified rate is established, a temperature parameter log in an existing monitoring period is called, the optimal temperature parameter is selected as a standard temperature parameter, and the temperature parameter is dynamically adjusted in a predictive mode by capturing the temperature change rate of each key stage.

Description

Technical Field

[0001] The present invention relates to the technical field of annealing furnace temperature control, and specifically to an annealing furnace temperature intelligent control system and method based on big data analysis. Background Art

[0002] Traditional annealing processes in the field of heat treatment of metal materials have long faced the problem of insufficient precision in controlling process parameters. Due to the lack of systematic data support, the setting of process parameters mainly relies on the operator's experience and judgment, making it difficult to make differentiated adjustments for blanks of different materials and specifications. In the multi-stage annealing process, the temperature adjustment rate of each stage often uses a fixed value, which cannot be dynamically optimized according to the actual phase change characteristics of the material, resulting in large fluctuations in the microstructure uniformity and mechanical properties of the finished parts. This experience-driven control mode not only makes it difficult to achieve a deep match between process parameters and material properties, but is also prone to quality defects due to human judgment errors, making the improvement of the qualified rate of finished products of the traditional annealing process a bottleneck.

[0003] The temperature control mechanism of the existing annealing process has significant limitations when dealing with complex production environments. Traditional control systems usually use a preset fixed temperature curve for regulation and lack the ability to dynamically respond to real-time temperature changes. When unexpected conditions such as uneven temperature distribution in the annealing furnace and decreased thermal efficiency due to equipment aging occur, the system is unable to perceive and adjust the parameters in time, which can easily cause the annealing rate to deviate from the ideal range, thereby causing quality problems such as coarse grains and excessive residual stress. During the vacuum annealing process, the traditional radiation heating method causes a significant temperature difference between the inside and outside of the material, and the cooling stage relies on natural cooling, which is not only time-consuming, but also difficult to ensure the consistency of different batches of products. This passive temperature control mode makes the production process less adaptable to environmental interference and increases the risk of defective products.

[0004] The optimization of annealing process parameters has long relied on periodic manual experiments and experience summaries, lacking a systematic iterative mechanism. Traditional methods obtain limited process data through small-scale experiments, which makes it difficult to fully cover the influence of multi-dimensional variables such as material properties, equipment status, and environmental conditions. As a result, the optimized parameter combinations often cannot reproduce the ideal results in actual production. In addition, the adjustment of process parameters lacks continuity, and the parameter optimization experience between different batches of production is difficult to effectively inherit and accumulate. This requires companies to repeatedly invest a lot of resources in trial and error experiments when responding to new materials and new process requirements. This static optimization model not only restricts the improvement of production efficiency, but also makes it difficult for process control solutions to adapt to rapidly changing market demands. The performance of the insulation materials of traditional vacuum annealing furnaces deteriorates after long-term use, but due to the lack of a data feedback mechanism, the parameters cannot be optimized in a timely manner, further exacerbating quality fluctuations. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent temperature control system and method for an annealing furnace based on big data analysis to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solution: an annealing furnace temperature intelligent control method based on big data analysis, comprising the following steps: S1. Collect historical data of annealing of blanks and divide the annealing process into annealing stages according to temperature; S2. Calculate the annealing rate in the annealing stage, divide the annealing rate intervals, and select the optimal annealing rate interval based on the relationship between the annealing rate interval and the qualified rate of finished products; S3. Analyze the influence of annealing rate on product qualification rate and select the annealing stage affected by annealing rate; S4. Analyze the temperature adjustment process of the blank and adjust the temperature parameters of the annealing furnace; S5. After the annealing process of the blank is completed, the qualified rate of the next finished product is predicted and a temperature parameter log related to the qualified rate is established; S6. Call a temperature parameter log within a pre-set monitoring period and select the best temperature parameter as the standard temperature parameter.

[0007] Furthermore, in step S1, historical data of annealing treatment of any type of blank is counted, the number of blanks processed in history is N, and after the blanks in the annealing furnace are kept at temperature A1, the temperature of the blanks needs to be lowered to A I+1 , and then become a finished product. The annealing process is divided into I annealing stages according to the temperature, where the i-th annealing stage means that the blank is annealed from temperature A i Adjust to temperature A i+1 In the process of processing the nth blank, i=1,2,…,I, the annealing rate in the i-th annealing stage is T n_i , T n_i =(A i+1 -A i ) / t n_i , where n=1,2,…,N, t n_iThe dwell time of the nth blank in the i-th annealing stage; by statistically analyzing the historical data of the blank annealing treatment, the annealing process is scientifically divided into multiple stages according to temperature, and the correlation between the annealing rate and dwell time in each stage can be systematically sorted out. This operation can extract the characteristics of process parameters from historical experience, laying a data foundation for the subsequent analysis of the impact of the annealing rate of each stage on the quality of the finished product. The statistical analysis of the annealing rate in stages can refine the process control, make the process parameters of each temperature adjustment link more traceable, and facilitate the discovery of the changing patterns of the annealing rate in different stages. It provides an objective basis for accurately locating key process stages and optimizing the annealing rate range, improves the systematicness and accuracy of the annealing process analysis, and assists in the subsequent process optimization work to improve the qualified rate of finished products.

[0008] Furthermore, in step S2, the annealing rates of N blanks in the i-th annealing stage are counted, and X annealing rate intervals are set according to the annealing rates, where the x-th annealing rate interval is (V x ,V x+1 ], where V x represents the lower limit of the annealing rate in the xth annealing rate interval, where V x+1 The annealing rate upper limit of the x-th annealing rate interval is the same for each annealing rate interval. The annealing rate difference represents the difference between the upper limit and the lower limit of the annealing rate interval, thereby determining the qualified rate B of the finished product corresponding to the blank whose annealing rate in the i-th annealing stage is in the x-th annealing rate interval. i_x , thus selecting the optimal annealing rate interval of the i-th annealing stage from high to low according to the qualified rate, the optimal annealing rate interval is (V i_α ,V i_α+1 By dividing the annealing rate in the annealing stage into intervals and calculating the qualified rate of finished products corresponding to each interval, the degree of influence of different annealing rates on product quality can be systematically analyzed. This operation classifies and integrates discrete annealing rate data, and accurately locates the rate range with high qualified rate in units of equidistant intervals, avoiding the problem of unclear process rules caused by data dispersion. Screening the optimal annealing rate interval according to the qualified rate can provide a clear parameter optimization direction for the annealing process, so that the subsequent annealing process can be executed with reference to the optimal interval, effectively improving the stability of the qualified rate of finished products, and at the same time laying the foundation for identifying key process stages and establishing refined annealing control standards, helping the annealing process to transform from experience-driven to data-driven precision management.

[0009] Furthermore, in step S3, the qualified rate β1 of the blanks whose annealing rate is within the optimal annealing rate range in the i-th annealing stage, and the qualified rate β2 of the blanks whose annealing rate is not within the optimal annealing rate range in the i-th annealing stage are counted. If β1-β2>β, the i-th annealing stage is judged to be the annealing stage affected by the annealing rate, β represents the established qualified rate judgment threshold, and the predicted defect coefficient k is set for the p-th annealing stage affected by the annealing rate. p , k p =1-B p_y , where p represents the pth annealing stage affected by the annealing rate, y represents the blank in the yth annealing rate interval during the pth annealing stage affected by the annealing rate, B p_y The yield rate of finished parts corresponding to the yth annealing rate interval for a blank in the pth annealing stage affected by the annealing rate is expressed as p = 1, 2, …, P, where P is the number of annealing rate intervals, and y = 1, 2, …, Y, where Y is the number of annealing rate intervals. By comparing the yield rates of finished parts in the optimal and suboptimal annealing rate intervals within the annealing stage, critical process stages sensitive to annealing rate can be accurately identified, avoiding the diversion of resources to non-critical links. This identification mechanism uses yield rate differences as a basis for screening annealing stages that truly impact product quality, enabling more targeted process control. By setting a predicted defect coefficient for the affected stages, the degree of quality risk within different annealing rate intervals can be quantified, providing data support for subsequent real-time monitoring of the impact of annealing rate fluctuations on finished product quality. This helps operators quickly identify risk areas during the annealing process and take proactive adjustments, effectively reducing the probability of defective products and improving the refined management of the annealing process and the stability of finished product quality.

[0010] Furthermore, in step S4, after the blank is kept warm, the temperature adjustment process of the blank is monitored in real time. When the blank is in the pth annealing stage affected by the annealing rate, the real-time temperature of the blank is monitored to be A, that is, the real-time annealing rate of the blank in the pth annealing stage affected by the annealing rate is T, T=(AA p ) / t, where t represents the time duration of the blank entering the pth annealing stage affected by the annealing rate. According to the real-time annealing rate, the qualified rate B of the finished product corresponding to the yth annealing rate interval when the blank is in the pth annealing stage affected by the annealing rate is called p , if B p / B best >γ, then the current annealing rate is considered stable, where B best It indicates the qualified rate of finished products when the blank is in the optimal annealing rate range at the pth annealing stage affected by the annealing rate. The optimal annealing rate range is expressed as (V p_α ,V p_α+1], γ represents the qualified rate stability judgment coefficient; otherwise, the current annealing rate is judged to be abnormal, and the annealing furnace temperature parameters are adjusted; The adjustment of the annealing furnace temperature parameters includes: if the real-time annealing rate T≤V p_α , the temperature of the annealing furnace in the pth annealing stage affected by the annealing rate is increased by F1=(V p_α -T) / V p_α , that is, the temperature C of the annealing furnace in the pth annealing stage affected by the annealing rate is increased to (1+F1)*C; if the real-time annealing rate T>V p_α+1 , the temperature of the annealing furnace in the pth annealing stage affected by the annealing rate is reduced by F2=(TV p_α+1 ) / V p_α+1 , that is, the temperature C of the annealing furnace in the pth annealing stage affected by the annealing rate is reduced to (1-F2)*C; by real-time monitoring of the temperature adjustment process of the blank after insulation, the annealing rate changes can be dynamically captured in the key annealing stage, and the stability of the current rate can be judged in combination with the historical pass rate data, thereby realizing precise control of the annealing process. When the annealing rate is abnormal, the system automatically and adaptively adjusts the annealing furnace temperature parameters based on the deviation between the optimal rate range and the real-time rate, so that the annealing rate returns to the high pass rate range in time, effectively avoiding quality defects of the finished product caused by rate fluctuations. This mechanism combines real-time monitoring, intelligent judgment and dynamic adjustment, which can not only reduce the cost of manual intervention, but also improve the stability of the annealing process through closed-loop control, laying the foundation for automated control for continuously optimizing annealing process parameters and ensuring the pass rate of finished products.

[0011] Furthermore, in step S5, after the annealing process of the blank is completed, the adjusted temperature parameters are called to predict the next annealing process and calculate the predicted finished product qualification rate G: ; Analyze the annealing process of the blank in the pth annealing stage affected by the annealing rate. The qualified rate corresponding to the annealing rate range is T p, save the predicted finished product qualification rate and the adjusted temperature parameters as a temperature parameter log; after the annealing process is completed, use the adjusted temperature parameters to predict the finished product qualification rate of the next annealing treatment, and analyze the rate interval qualification rate of the key annealing stage at the same time, and save the predicted results and temperature parameters synchronously as a log. This operation can form a dynamically optimized parameter library through the accumulation of historical data, so that the subsequent annealing process can continuously adjust the temperature parameters based on the actual production results, thereby improving the accuracy of the prediction model. The establishment of a temperature parameter log provides a quantitative reference for process optimization, making it easier for operators to trace the impact of different parameter combinations on the quality of finished products. Through multiple rounds of data iteration, the optimal annealing parameter range is gradually clarified, forming a closed-loop management of "prediction-execution-recording-optimization", which helps the annealing process upgrade from experience-driven to data-driven precision control.

[0012] In step S6, a temperature parameter log within a predetermined monitoring period is called, the temperature parameter log corresponding to the highest value of the predicted finished product qualification rate is selected, and the temperature parameters in the temperature parameter log are called as the standard temperature parameters for the next annealing process.

[0013] The intelligent control system for annealing furnace temperature based on big data analysis includes: historical data collection and stage division module, annealing rate analysis and interval screening module, annealing rate impact analysis module, annealing rate stability control module, qualified rate prediction and log generation module, and temperature parameter log optimization module; The historical data collection and stage division module is used to collect historical data of annealing of blanks and divide the annealing process into annealing stages according to temperature; The annealing rate analysis and interval screening module is used to count the annealing rate in the annealing stage, divide the annealing rate intervals, and select the optimal annealing rate interval based on the relationship between the annealing rate interval and the qualified rate of finished products; The annealing rate impact analysis module is used to analyze the impact of the annealing rate on the product qualification rate and select the annealing stage affected by the annealing rate; The annealing rate stability control module is used to analyze the temperature adjustment process of the blank and adjust the temperature parameters of the annealing furnace; The pass rate prediction and log generation module is used to predict the pass rate of the next finished product after the annealing process of the blank is completed, and to create a temperature parameter log related to the pass rate; The temperature parameter log optimization module is used to call the temperature parameter log within a pre-set monitoring period and select the best temperature parameter as the standard temperature parameter.

[0014] Compared with existing technologies, the present invention achieves the following beneficial effects: On the one hand, through systematic statistics and analysis of historical annealing data, it can accurately locate the optimal annealing rate range that significantly affects the finished product yield in each annealing stage. Based on this, temperature control can be implemented, effectively improving the process accuracy of the production process. This method changes the traditional empirical control method and uses big data to mine the optimal process parameter range for different stages, making the annealing process more closely aligned with the material properties of the blank, reducing quality defects caused by improper annealing rates at the source, and significantly improving the yield and consistency of the finished product.

[0015] On the one hand, real-time monitoring and dynamic adjustment mechanisms give the production process greater stability and anti-interference capabilities. During the annealing process of blanks, the system can capture the temperature change rate at each key stage in real time. If the annealing rate deviates from the optimal range, it can quickly make targeted adjustments to the annealing furnace temperature parameters based on preset logic to promptly correct process deviations. This proactive control mode can effectively address unexpected situations such as temperature fluctuations and equipment wear and tear that may arise during production, ensuring that the annealing process is always carried out under ideal conditions, reducing the occurrence of defective products and mitigating quality risks in the production process.

[0016] On the other hand, the accumulation and optimization mechanism of temperature parameter logs forms a closed-loop system for continuous improvement. After each annealing process, the system records and analyzes the adjusted temperature parameters and the predicted pass rate. By comparing the parameter logs within a certain period, the temperature parameters that give the highest predicted pass rate are selected as the standard. This process is like giving the system learning capabilities, enabling it to continuously optimize control strategies in long-term production and adapt to the differences in characteristics of different batches of blanks. This not only improves production efficiency, but also makes the process control solution more accurate and efficient through continuous iteration, providing strong support for enterprises to achieve refined production and quality improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is a structural diagram of the annealing furnace temperature intelligent control system based on big data analysis of the present invention; Figure 2 It is a flow chart of the annealing furnace temperature intelligent control method based on big data analysis of the present invention. DETAILED DESCRIPTION

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0019] See also Figure 1 and Figure 2 The present invention provides a technical solution: an intelligent temperature control method for an annealing furnace based on big data analysis, comprising the following steps: S1. Collect historical data of annealing of blanks and divide the annealing process into annealing stages according to temperature; S2. Calculate the annealing rate in the annealing stage, divide the annealing rate intervals, and select the optimal annealing rate interval based on the relationship between the annealing rate interval and the qualified rate of finished products; S3. Analyze the influence of annealing rate on product qualification rate and select the annealing stage affected by annealing rate; S4. Analyze the temperature adjustment process of the blank and adjust the temperature parameters of the annealing furnace; S5. After the annealing process of the blank is completed, the qualified rate of the next finished product is predicted and a temperature parameter log related to the qualified rate is established; S6. Call a temperature parameter log within a pre-set monitoring period and select the best temperature parameter as the standard temperature parameter.

[0020] In step S1, the historical data of annealing treatment of any type of blank is counted. The number of blanks processed in the history is N. After the blanks in the annealing furnace are kept at temperature A1, the temperature of the blanks needs to be lowered to A1. I+1 , and then become a finished product. The annealing process is divided into I annealing stages according to the temperature, where the i-th annealing stage means that the blank is annealed from temperature A i Adjust to temperature A i+1 In the process of processing the nth blank, i=1,2,…,I, the annealing rate in the i-th annealing stage is T n_i , T n_i =(A i+1 -A i ) / t n_i , where n=1,2,…,N, t n_iThe dwell time of the nth blank in the i-th annealing stage; by statistically analyzing the historical data of the blank annealing treatment, the annealing process is scientifically divided into multiple stages according to temperature, and the correlation between the annealing rate and dwell time in each stage can be systematically sorted out. This operation can extract the characteristics of process parameters from historical experience, laying a data foundation for the subsequent analysis of the impact of the annealing rate of each stage on the quality of the finished product. The statistical analysis of the annealing rate in stages can refine the process control, make the process parameters of each temperature adjustment link more traceable, and facilitate the discovery of the changing patterns of the annealing rate in different stages. It provides an objective basis for accurately locating key process stages and optimizing the annealing rate range, improves the systematicness and accuracy of the annealing process analysis, and assists in the subsequent process optimization work to improve the qualified rate of finished products.

[0021] In step S2, the annealing rates of N blanks in the i-th annealing stage are counted, and X annealing rate intervals are set according to the annealing rates, where the x-th annealing rate interval is (V x ,V x+1 ], where V x represents the lower limit of the annealing rate in the xth annealing rate interval, where V x+1 The annealing rate upper limit of the x-th annealing rate interval is the same for each annealing rate interval. The annealing rate difference represents the difference between the upper limit and the lower limit of the annealing rate interval, thereby determining the qualified rate B of the finished product corresponding to the blank whose annealing rate in the i-th annealing stage is in the x-th annealing rate interval. i_x , thus selecting the optimal annealing rate interval of the i-th annealing stage from high to low according to the qualified rate, the optimal annealing rate interval is (V i_α ,V i_α+1 By dividing the annealing rate in the annealing stage into intervals and calculating the qualified rate of finished products corresponding to each interval, the degree of influence of different annealing rates on product quality can be systematically analyzed. This operation classifies and integrates discrete annealing rate data, and accurately locates the rate range with high qualified rate in units of equidistant intervals, avoiding the problem of unclear process rules caused by data dispersion. Screening the optimal annealing rate interval according to the qualified rate can provide a clear parameter optimization direction for the annealing process, so that the subsequent annealing process can be executed with reference to the optimal interval, effectively improving the stability of the qualified rate of finished products, and at the same time laying the foundation for identifying key process stages and establishing refined annealing control standards, helping the annealing process to transform from experience-driven to data-driven precision management.

[0022] In step S3, the qualified rate β1 of the blanks whose annealing rate is within the optimal annealing rate range in the i-th annealing stage, and the qualified rate β2 of the blanks whose annealing rate is not within the optimal annealing rate range in the i-th annealing stage are counted. If β1-β2>β, the i-th annealing stage is judged to be the annealing stage affected by the annealing rate, β represents the established qualified rate judgment threshold, and the predicted defect coefficient k is set for the p-th annealing stage affected by the annealing rate. p , k p =1-B p_y , where p represents the pth annealing stage affected by the annealing rate, y represents the blank in the yth annealing rate interval during the pth annealing stage affected by the annealing rate, B p_y The yield rate of finished parts corresponding to the yth annealing rate interval for a blank in the pth annealing stage affected by the annealing rate is expressed as p = 1, 2, …, P, where P is the number of annealing rate intervals, and y = 1, 2, …, Y, where Y is the number of annealing rate intervals. By comparing the yield rates of finished parts in the optimal and suboptimal annealing rate intervals within the annealing stage, critical process stages sensitive to annealing rate can be accurately identified, avoiding the diversion of resources to non-critical links. This identification mechanism uses yield rate differences as a basis for screening annealing stages that truly impact product quality, enabling more targeted process control. By setting a predicted defect coefficient for the affected stages, the degree of quality risk within different annealing rate intervals can be quantified, providing data support for subsequent real-time monitoring of the impact of annealing rate fluctuations on finished product quality. This helps operators quickly identify risk areas during the annealing process and take proactive adjustments, effectively reducing the probability of defective products and improving the refined management of the annealing process and the stability of finished product quality.

[0023] In step S4, after the blank is kept warm, the temperature adjustment process of the blank is monitored in real time. When the blank is in the pth annealing stage affected by the annealing rate, the real-time temperature of the blank is monitored to be A, that is, the real-time annealing rate of the blank in the pth annealing stage affected by the annealing rate is T, T=(AA p ) / t, where t represents the time duration of the blank entering the pth annealing stage affected by the annealing rate. According to the real-time annealing rate, the qualified rate B of the finished product corresponding to the yth annealing rate interval when the blank is in the pth annealing stage affected by the annealing rate is called p , if B p / B best >γ, then the current annealing rate is considered stable, where B best It indicates the qualified rate of finished products when the blank is in the optimal annealing rate range at the pth annealing stage affected by the annealing rate. The optimal annealing rate range is expressed as (V p_α ,V p_α+1], γ represents the qualified rate stability judgment coefficient; otherwise, the current annealing rate is judged to be abnormal, and the annealing furnace temperature parameters are adjusted; The adjustment of the annealing furnace temperature parameters includes: if the real-time annealing rate T≤V p_α , the temperature of the annealing furnace in the pth annealing stage affected by the annealing rate is increased by F1=(V p_α -T) / V p_α , that is, the temperature C of the annealing furnace in the pth annealing stage affected by the annealing rate is increased to (1+F1)*C; if the real-time annealing rate T>V p_α+1 , the temperature of the annealing furnace in the pth annealing stage affected by the annealing rate is reduced by F2=(TV p_α+1 ) / V p_α+1 , that is, the temperature C of the annealing furnace in the pth annealing stage affected by the annealing rate is reduced to (1-F2)*C; by real-time monitoring of the temperature adjustment process of the blank after insulation, the annealing rate changes can be dynamically captured in the key annealing stage, and the stability of the current rate can be judged in combination with the historical pass rate data, thereby realizing precise control of the annealing process. When the annealing rate is abnormal, the system automatically and adaptively adjusts the annealing furnace temperature parameters based on the deviation between the optimal rate range and the real-time rate, so that the annealing rate returns to the high pass rate range in time, effectively avoiding quality defects of the finished product caused by rate fluctuations. This mechanism combines real-time monitoring, intelligent judgment and dynamic adjustment, which can not only reduce the cost of manual intervention, but also improve the stability of the annealing process through closed-loop control, laying the foundation for automated control for continuously optimizing annealing process parameters and ensuring the pass rate of finished products.

[0024] In step S5, after the annealing process of the blank is completed, the adjusted temperature parameters are called to predict the next annealing process and calculate the predicted finished product qualification rate G: ; Analyze the annealing process of the blank in the pth annealing stage affected by the annealing rate. The qualified rate corresponding to the annealing rate range is T p, save the predicted finished product qualification rate and the adjusted temperature parameters as a temperature parameter log; after the annealing process is completed, use the adjusted temperature parameters to predict the finished product qualification rate of the next annealing treatment, and analyze the rate interval qualification rate of the key annealing stage at the same time, and save the predicted results and temperature parameters synchronously as a log. This operation can form a dynamically optimized parameter library through the accumulation of historical data, so that the subsequent annealing process can continuously adjust the temperature parameters based on the actual production results, thereby improving the accuracy of the prediction model. The establishment of a temperature parameter log provides a quantitative reference for process optimization, making it easier for operators to trace the impact of different parameter combinations on the quality of finished products. Through multiple rounds of data iteration, the optimal annealing parameter range is gradually clarified, forming a closed-loop management of "prediction-execution-recording-optimization", which helps the annealing process upgrade from experience-driven to data-driven precision control.

[0025] In step S6, a temperature parameter log within a predetermined monitoring period is called, the temperature parameter log corresponding to the highest value of the predicted finished product qualification rate is selected, and the temperature parameters in the temperature parameter log are called as the standard temperature parameters for the next annealing process.

[0026] An intelligent annealing furnace temperature control system based on big data analysis, comprising: a historical data acquisition and stage division module, an annealing rate analysis and interval screening module, an annealing rate impact analysis module, an annealing rate stability control module, a pass rate prediction and log generation module, and a temperature parameter log optimization module; The historical data collection and stage division module is used to collect historical data of annealing of blanks and divide the annealing process into annealing stages according to temperature; The annealing rate analysis and interval screening module is used to count the annealing rate in the annealing stage, divide the annealing rate interval, and select the optimal annealing rate interval based on the relationship between the annealing rate interval and the qualified rate of finished products; The annealing rate impact analysis module is used to analyze the impact of annealing rate on product qualification rate and select the annealing stages affected by annealing rate; The annealing rate stability control module is used to analyze the temperature adjustment process of the blank and adjust the temperature parameters of the annealing furnace; The qualification rate prediction and log generation module is used to predict the qualification rate of the next finished product after the annealing process of the blank is completed, and to establish a temperature parameter log related to the qualification rate; The temperature parameter log optimization module is used to call the temperature parameter log within a set monitoring period and select the best temperature parameter as the standard temperature parameter.

[0027] Example 1: In any annealing process, first collect annealing history data for a certain type of blank. These blanks are held at an initial temperature in an annealing furnace and then cooled to a target temperature to form the finished product. The entire annealing process is divided into multiple stages based on temperature changes, with each stage corresponding to a temperature adjustment process. For example, in one stage, the blank must be cooled from a higher temperature to the adjacent lower temperature. The duration of each blank's stay in each stage is recorded. Based on the relationship between the temperature change and the dwell time, the annealing rate at each stage is calculated, laying the data foundation for subsequent analysis.

[0028] Next, the annealing rates for all blanks in each annealing stage are divided into several equal intervals. The corresponding finished product pass rate for the blanks within each interval is calculated. By comparing the pass rates of each interval, the annealing rate interval that achieves the highest pass rate for that stage is selected as the optimal operating range for that stage, providing a precise parameter reference range for subsequent process control.

[0029] Next, the yield rates of finished products in the optimal and non-optimal annealing rate ranges are compared within each annealing stage. If the difference exceeds a preset standard, the stage is identified as a critical stage affected by the annealing rate. For each critical stage, a predicted defect coefficient is set based on the yield rates corresponding to different rate ranges to reflect quality risk. This coefficient is used to quantify the probability of defective products in each rate range, providing a risk assessment basis for real-time monitoring.

[0030] Next, once the blank has completed its insulation and entered the critical annealing phase, its temperature changes are monitored in real time. The real-time annealing rate is calculated based on the current temperature and the duration since the phase began. The finished product pass rate corresponding to this rate in historical data is compared with the pass rate in the optimal range. If the difference exceeds the allowable range, the rate is considered abnormal. At this point, the annealing furnace temperature is automatically adjusted based on the deviation between the real-time rate and the optimal range. This includes increasing the furnace temperature if the annealing rate is too low, and decreasing it if it is too high, to return the annealing rate to the optimal range.

[0031] After each annealing process is completed, the yield rate of the finished product for the next annealing process is predicted based on the adjusted temperature parameters. The annealing rate ranges and corresponding yield rates for key stages are analyzed, and the predicted yield rates are saved along with the adjusted temperature parameters in a temperature parameter log, forming a traceable historical data record.

[0032] Finally, set a fixed monitoring cycle (such as weekly or monthly), call all temperature parameter logs within the cycle, filter out the log records with the highest predicted finished product qualification rate, extract the temperature parameters from them, and use them as the standard temperature parameters for the next annealing treatment, thereby realizing dynamic optimization and continuous iteration of annealing process parameters.

[0033] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. An intelligent temperature control method for an annealing furnace based on big data analysis, characterized by: The method comprises the following steps: S1. Collect historical data of annealing of blanks and divide the annealing process into annealing stages according to temperature; S2. Calculate the annealing rate in the annealing stage, divide the annealing rate intervals, and select the optimal annealing rate interval based on the relationship between the annealing rate interval and the qualified rate of finished products; S3. Analyze the influence of annealing rate on product qualification rate and select the annealing stage affected by annealing rate; S4. Analyze the temperature adjustment process of the blank and adjust the temperature parameters of the annealing furnace; S5. After the annealing process of the blank is completed, the qualified rate of the next finished product is predicted and a temperature parameter log related to the qualified rate is established; S6. Call a temperature parameter log within a pre-set monitoring period and select the best temperature parameter as the standard temperature parameter.

2. The annealing furnace temperature intelligent control method based on big data analysis according to claim 1 is characterized in that: In step S1, the historical data of annealing treatment of any type of blank is counted. The number of blanks processed in the history is N. After the blanks in the annealing furnace are kept at temperature A1, the temperature of the blanks needs to be lowered to A1. I+1 , and then become a finished product. The annealing process is divided into I annealing stages according to the temperature, where the i-th annealing stage means that the blank is annealed from temperature A i Adjust to temperature A i+1 In the process of processing the nth blank, i=1,2,…,I, the annealing rate in the i-th annealing stage is T n_i , T n_i =(A i+1 -A i ) / t n_i , where n=1,2,…,N, t n_i is the residence time of the nth blank in the i-th annealing stage.

3. The intelligent temperature control method for an annealing furnace based on big data analysis according to claim 2, characterized in that: In step S2, the annealing rates of N blanks in the i-th annealing stage are counted, and X annealing rate intervals are set according to the annealing rates, where the x-th annealing rate interval is (V x ,V x+1 ], where V x represents the lower limit of the annealing rate in the xth annealing rate interval, where V x+1 The annealing rate upper limit of the x-th annealing rate interval is the same for each annealing rate interval. The annealing rate difference represents the difference between the upper limit and the lower limit of the annealing rate interval, thereby determining the qualified rate B of the finished product corresponding to the blank whose annealing rate in the i-th annealing stage is in the x-th annealing rate interval. i_x , thus selecting the optimal annealing rate interval of the i-th annealing stage from high to low according to the qualified rate, the optimal annealing rate interval is (V i_α ,V i_α+1 ].

4. The intelligent temperature control method for an annealing furnace based on big data analysis according to claim 3, characterized in that: In step S3, the qualified rate β1 of the blanks whose annealing rate is within the optimal annealing rate range in the i-th annealing stage, and the qualified rate β2 of the blanks whose annealing rate is not within the optimal annealing rate range in the i-th annealing stage are counted. If β1-β2>β, the i-th annealing stage is judged to be the annealing stage affected by the annealing rate, β represents the established qualified rate judgment threshold, and the predicted defect coefficient k is set for the p-th annealing stage affected by the annealing rate. p , k p =1-B p_y , where p represents the pth annealing stage affected by the annealing rate, y represents the blank in the yth annealing rate interval during the pth annealing stage affected by the annealing rate, B p_y It represents the qualified rate of the finished product corresponding to the blank in the yth annealing rate interval at the pth annealing stage affected by the annealing rate, p = 1, 2, ..., P, P is the number of annealing rate intervals, y = 1, 2, ..., Y, Y is the number of annealing rate intervals.

5. The annealing furnace temperature intelligent control method based on big data analysis according to claim 4 is characterized in that: In step S4, after the blank is kept warm, the temperature adjustment process of the blank is monitored in real time. When the blank is in the pth annealing stage affected by the annealing rate, the real-time temperature of the blank is monitored to be A, that is, the real-time annealing rate of the blank in the pth annealing stage affected by the annealing rate is T, T=(AA p ) / t, where t represents the time duration of the blank entering the pth annealing stage affected by the annealing rate. According to the real-time annealing rate, the qualified rate B of the finished product corresponding to the yth annealing rate interval when the blank is in the pth annealing stage affected by the annealing rate is called p , if B p / B best >γ, then the current annealing rate is considered stable, where B best It indicates the qualified rate of finished products when the blank is in the optimal annealing rate range at the pth annealing stage affected by the annealing rate. The optimal annealing rate range is expressed as (V p_α ,V p_α+1 ], γ represents the qualified rate stability judgment coefficient; otherwise, the current annealing rate is judged to be abnormal, and the annealing furnace temperature parameters are adjusted.

6. The intelligent temperature control method for an annealing furnace based on big data analysis according to claim 5, characterized in that: The adjustment of the annealing furnace temperature parameters includes: if the real-time annealing rate T≤V p_α , the temperature of the annealing furnace in the pth annealing stage affected by the annealing rate is increased by F1=(V p_α -T) / V p_α , that is, the temperature C of the annealing furnace in the pth annealing stage affected by the annealing rate is increased to (1+F1)*C; if the real-time annealing rate T>V p_α+1 , the temperature of the annealing furnace in the pth annealing stage affected by the annealing rate is reduced by F2=(TV p_α+1 ) / V p_α+1 , that is, the temperature C of the annealing furnace in the pth annealing stage affected by the annealing rate is reduced to (1-F2)*C.

7. The intelligent temperature control method for an annealing furnace based on big data analysis according to claim 6, characterized in that: In step S5, after the annealing process of the blank is completed, the adjusted temperature parameters are called to predict the next annealing process and calculate the predicted finished product qualification rate G: ； Analyze the annealing process of the blank in the pth annealing stage affected by the annealing rate. The qualified rate corresponding to the annealing rate range is T p , save the predicted finished product qualification rate and the adjusted temperature parameters as a temperature parameter log.

8. The intelligent temperature control method for an annealing furnace based on big data analysis according to claim 6, characterized in that: In step S6, a temperature parameter log within a predetermined monitoring period is called, the temperature parameter log corresponding to the highest value of the predicted finished product qualification rate is selected, and the temperature parameters in the temperature parameter log are called as the standard temperature parameters for the next annealing process.

9. An intelligent annealing furnace temperature control system based on big data analysis, wherein the system is applied to the intelligent annealing furnace temperature control method based on big data analysis according to any one of claims 1 to 8, characterized in that: The system includes: a historical data collection and stage division module, an annealing rate analysis and interval screening module, an annealing rate impact analysis module, an annealing rate stability control module, a pass rate prediction and log generation module, and a temperature parameter log optimization module; The historical data collection and stage division module is used to collect historical data of annealing of blanks and divide the annealing process into annealing stages according to temperature; The annealing rate analysis and interval screening module is used to count the annealing rate in the annealing stage, divide the annealing rate intervals, and select the optimal annealing rate interval based on the relationship between the annealing rate interval and the qualified rate of finished products; The annealing rate impact analysis module is used to analyze the impact of the annealing rate on the product qualification rate and select the annealing stage affected by the annealing rate; The annealing rate stability control module is used to analyze the temperature adjustment process of the blank and adjust the temperature parameters of the annealing furnace; The pass rate prediction and log generation module is used to predict the pass rate of the next finished product after the annealing process of the blank is completed, and to create a temperature parameter log related to the pass rate; The temperature parameter log optimization module is used to call the temperature parameter log within a pre-set monitoring period and select the best temperature parameter as the standard temperature parameter.

Images