Steelmaking process optimization based on smelting temperature data: methods and systems

Abstract

This invention discloses a method and system for optimizing the steelmaking process based on smelting temperature data, comprising the following steps: collecting smelting temperature data corresponding to each process stage of the steelmaking process; performing preprocessing; constructing a stage thermal history feature set; performing cross-stage correlation to generate a temperature memory chain; performing causal decomposition on the temperature sequence of the current process stage; identifying heat offset, heat decay, heat amplification, and heat transfer misalignment, generating stage transfer distortion results; correcting the temperature control path of the current process stage; and performing continuous temperature optimization control on subsequent process stages of the steelmaking process. This invention uses the temperature memory chain method to achieve optimized temperature control in steelmaking, possessing the advantages of precise temperature control and stable continuity.

Description

Technical Field

[0001] This invention relates to the field of steelmaking process control, and more particularly to a steelmaking process method and system based on smelting temperature data optimization. Background Technology

[0002] In the steelmaking process, smelting temperature is a crucial parameter reflecting changes in the thermal state and the effectiveness of process execution at each stage. Current technologies typically involve collecting temperature data from stages such as converter, refining, and continuous casting pretreatment, and combining this data with empirical rules, stage setpoints, or single-stage control models to adjust the heating, holding, or stage switching processes of the corresponding stages, thereby achieving temperature control in the steelmaking process.

[0003] However, existing technologies mostly focus on the independent analysis of the temperature state at the current process stage, lacking a systematic characterization of the thermal history of previous process stages. It is difficult to accurately identify the differences between historical influences and current effects in the temperature changes at the current stage, and it is also difficult to make targeted judgments on heat offset, heat decay, heat amplification, and heat transfer misalignment. This results in insufficient accuracy of temperature control path correction and limited continuous optimization control effects in subsequent process stages. Summary of the Invention

[0004] One objective of this invention is to propose a method and system for optimizing the steelmaking process based on smelting temperature data. This invention uses a temperature memory chain method to achieve optimized temperature control in steelmaking, which has the advantages of precise temperature control and stable connection.

[0005] The steelmaking process method based on smelting temperature data optimization according to embodiments of the present invention includes the following steps: S1. Collect smelting temperature data corresponding to each process stage of the steelmaking process, and identify the process stage and time index to form the original temperature dataset of the stage. S2. Perform preprocessing on the original temperature dataset of the stage to generate a standardized stage temperature sequence set; S3. Based on the standardized stage temperature sequence set, extract the temperature change trajectory features within each process stage, the temperature residual features at the end of each process stage, and the temperature transfer features between adjacent process stages to generate a stage thermal history feature set. S4. Based on the stage thermal history feature set, perform cross-stage correlation to generate temperature memory chain; S5. Based on the temperature memory chain, perform causal decomposition on the temperature sequence of the current process stage to generate the causal decomposition result of the temperature of the current stage. S6. Based on the current stage temperature cause decomposition results, identify heat offset, heat decay, heat amplification and heat transfer misalignment, and generate stage transfer distortion results; S7. Based on the stage transmission distortion results, perform correction on the temperature control path of the current process stage and generate the temperature correction result of the current process stage. S8. Based on the temperature correction results of the current process stage, perform continuous temperature optimization control on the subsequent process stages of the steelmaking process, and re-extract the thermal history feature set of the updated stage according to the temperature sequence of the updated stage, update the temperature memory chain, and generate the continuous evolutionary temperature optimization results of the steelmaking process.

[0006] Optionally, S2 specifically includes: S21. Read the smelting temperature data corresponding to each process stage in the original temperature dataset of the stage, and arrange the smelting temperature data in the same process stage in sequence according to the time index to form the initial stage temperature sequence corresponding to each process stage. S22. Perform abnormal temperature measurement record removal on the temperature sequence of each initial stage to generate a stage abnormal removal temperature sequence. S23. Based on the temperature continuity relationship corresponding to adjacent time index identifiers in the stage anomaly removal temperature sequence, fill in the missing temperature measurement records at the missing time index identifier positions to form a stage complete temperature sequence. S24. Map the stage completion temperature sequence corresponding to each process stage to a unified time base, align the time index start point and time interval between different process stages, and generate a stage aligned temperature sequence. S25. Based on the continuity of temperature changes before and after the boundary position of adjacent process stages in the stage-aligned temperature sequence, identify stage switching points and generate stage switching point results. S26. Based on the stage switching point results, re-perform stage segmentation on the stage-aligned temperature sequence to obtain a standardized stage temperature sequence set corresponding to each process stage.

[0007] Optionally, S3 specifically includes: S31. Read the standardized stage temperature sequence corresponding to each process stage in the standardized stage temperature sequence set, and extract the continuously arranged temperature data entries within each process stage according to the process stage identifier to form a stage internal temperature analysis sequence. S32. For the temperature changes corresponding to adjacent time indexes in the internal temperature analysis sequence of each stage, identify the temperature rise segment, temperature fall segment, temperature fluctuation segment and temperature stability segment in sequence, and collect them according to the arrangement order of each segment in the corresponding process stage to form the temperature change trajectory characteristics corresponding to each process stage. S33. At the end of the standardized stage temperature sequence corresponding to each process stage, continuously extract the temperature data entries within the end segment of the stage according to the time index identifier, and merge the end temperature level, end temperature change direction and end temperature fluctuation of the temperature data entries to form the temperature residual characteristics corresponding to each process stage. S34. For the standardized stage temperature sequence corresponding to adjacent process stages, extract the continuous temperature data entries between the end of the previous process stage and the beginning of the next process stage, and extract the continuous characteristics of temperature connection before and after the boundary position, the characteristics of temperature connection change type and the characteristics of temperature connection amplitude to form the temperature connection characteristics between adjacent process stages. S35. Collect the temperature change trajectory characteristics, temperature residual characteristics, and temperature transfer characteristics between adjacent process stages according to the process stage sequence to form a stage thermal history feature set.

[0008] Optionally, the temperature data entry is a temperature record in the smelting temperature data corresponding to a single time index identifier, which includes the temperature measurement value at a time point and the corresponding process stage identifier and time index identifier.

[0009] Optionally, S4 specifically includes: S41. The temperature residual features, temperature change trajectory features and temperature continuity features corresponding to adjacent process stages are collected in the reading stage thermal history feature set and paired in the order of the preceding process stage first and the subsequent process stage last to form thermal history association units of adjacent process stages. S42. In the thermal history correlation unit of each adjacent process stage, extract the temperature residual features of the preceding process stage and the temperature change trajectory features of the preceding segment of the subsequent process stage, and match them to form the temperature continuity correspondence results between the preceding and following stages. S43. For each adjacent process stage thermal history correlation unit, extract the temperature continuity characteristics between the preceding process stage and the subsequent process stage, organize the temperature continuity relationship before and after the boundary position of adjacent process stages, and form the temperature continuity correlation result at the boundary position. S44. Combine the corresponding results of temperature continuity between the preceding and following stages with the results of temperature connection at the boundary to generate a heat history transfer chain between stages, describing the heat history of the preceding process stage to the subsequent process stage. S45. According to the execution order of each process stage in the steelmaking process, the heat history transfer chain segments between each stage are connected in sequence, and the subsequent process stages of the heat history transfer chain segment between the previous stage are sequentially connected with the preceding process stages of the heat history transfer chain segment between the next stage to form a temperature memory chain.

[0010] Optionally, S5 specifically includes: S51. Read the standardized stage temperature sequence corresponding to the current process stage and the temperature memory chain node corresponding to the current process stage in the temperature memory chain. Extract the stage temperature data entries corresponding to the current process stage and the inter-stage heat history transfer chain segment that continues to be transferred from the previous process stage to the current process stage to form the current stage temperature cause analysis data. S52. Based on the current stage temperature cause analysis data, extract the initial time period temperature data in the standardized stage temperature sequence of the current process stage, and extract the temperature horizontal transmission result, temperature directional transmission result, and temperature fluctuation transmission result in the inter-stage heat history transmission chain segment. S53. Based on the temperature level transfer results, the temperature data portion formed by the continuous transfer of the temperature level from the end of the previous process stage is calibrated in the temperature data at the beginning of the current process stage to form a temperature level decomposition result. S54. Based on the temperature direction transmission results, extract the temperature change portion in the temperature sequence of the current process stage standardization stage that is consistent with the temperature change direction at the end of the previous process stage, and record it as the historical influence direction component. Record the remaining temperature change portion as the current action direction component to form the temperature direction decomposition result. S55. Based on the temperature fluctuation transmission results, extract the temperature fluctuation part in the temperature sequence of the current process stage standardization stage that matches the temperature fluctuation state at the end of the previous process stage and record it as the historical influence fluctuation component. Record the remaining temperature fluctuation parts as the current action fluctuation component to form the temperature fluctuation decomposition result. S56. The historical influence temperature level component, the historical influence direction component, and the historical influence fluctuation component in the temperature level decomposition result are aggregated to form the historical influence component. The current influence temperature level component, the current influence direction component, and the current influence fluctuation component in the temperature fluctuation decomposition result are aggregated to form the current influence component. S57. Record the historical influence components and the current effect components according to the same current process stage, and generate the temperature cause decomposition results for the current stage.

[0011] Optionally, S6 specifically includes: S61. Obtain the current stage temperature cause decomposition results and extract historical influence components and current action components; S62. Read the historical influence temperature level component in the historical influence component and compare it with the temperature data of the starting time corresponding to the current process stage. Determine the temperature difference after the temperature level at the end of the previous process stage is transferred to the starting temperature state of the current process stage, and form the heat offset analysis result. S63. Count the number of consecutive temperature data entries corresponding to the starting position of the temperature sequence of the standardized stage in the current process stage, determine the decay state of the thermal history of the previous process stage in the current process stage, and form the heat decay analysis results. S64. Extract the distribution position and distribution length of the current action component and the historical influence component in the temperature sequence of the current process stage standardization stage, determine the degree of enhancement of the current action component to the historical influence component, and form the heat amplification analysis results. S65. Extract the starting position of the historical influence component in the temperature sequence of the current process stage standardization stage. Based on the positional difference between the starting position and the starting position of the temperature sequence of the current process stage standardization stage, determine the positional deviation of the thermal history in the current process stage and form the heat transfer misalignment analysis results. S66. The results of heat offset analysis, heat decay analysis, heat amplification analysis, and heat transfer misalignment analysis are aggregated according to the same current process stage to generate the stage transfer distortion result.

[0012] Optionally, S7 specifically includes: S71. Read the distortion results during the reading stage and extract the heat offset analysis results, heat attenuation analysis results, heat amplification analysis results and heat transfer misalignment analysis results; S72. Based on the heat offset analysis results, the heating control path corresponding to the current process stage is corrected to form the heating control path correction result. S73. Based on the heat decay analysis results, adjust the heat preservation control path corresponding to the current process stage to form a heat preservation control path correction result. S74. Based on the results of heat amplification analysis and heat transfer misalignment analysis, the stage connection control path corresponding to the current process stage is corrected to form the stage connection control path correction result. S75. The temperature rise control path correction results, the heat preservation control path correction results, and the stage transition control path correction results are aggregated according to the same current process stage to generate the temperature correction results for the current process stage.

[0013] Optionally, S8 specifically includes: S81. Read the temperature correction result of the current process stage and apply it to the subsequent process stages adjacent to the current process stage according to the process execution order to form the temperature control data of the subsequent process stages. S82. Based on the temperature control data of the subsequent process stage, the smelting temperature data corresponding to the subsequent process stage is continuously adjusted, and the adjusted temperature data entries are recorded in the order of time index identification to form the temperature sequence of the subsequent process stage update stage. S83. Based on the temperature sequence of the subsequent process stage update stage, form a thermal history feature set of the update stage and form an update temperature memory chain; S84. Following the execution sequence of each process stage in the steelmaking process, based on the updated temperature memory chain, continue to perform temperature regulation on the next subsequent process stage, and repeatedly form the updated stage temperature sequence and the updated stage thermal history feature set until the continuous temperature optimization control of the subsequent process stages of the steelmaking process is completed, generating the continuous evolutionary temperature optimization result of the steelmaking process.

[0014] A steelmaking process system optimized based on smelting temperature data according to an embodiment of the present invention includes: The data acquisition and identification module is used to collect smelting temperature data corresponding to each process stage of the steelmaking process, and to identify the process stage and time index to form the original temperature dataset of the stage. The preprocessing module is used to perform preprocessing and generate a standardized set of stage temperature sequences; The thermal history feature extraction module is used to generate a stage thermal history feature set based on a standardized stage temperature sequence set. The temperature memory chain generation module is used to generate temperature memory chains by performing cross-stage associations based on the stage thermal history feature set. The causal decomposition module is used to perform causal decomposition on the temperature sequence of the current process stage based on the temperature memory chain, and generate the causal decomposition results of the temperature of the current stage. The stage transfer distortion identification module is used to identify heat offset, heat decay, heat amplification, and heat transfer misalignment based on the current stage temperature cause decomposition results, and generate stage transfer distortion results. The temperature control path correction module is used to correct the temperature control path of the current process stage based on the stage transmission distortion results, and generate the temperature correction result of the current process stage. The continuous temperature optimization control module is used to perform continuous temperature optimization control on subsequent process stages of steelmaking based on the temperature correction results of the current process stage, and to re-extract and update the stage thermal history feature set, update the temperature memory chain, and generate continuous evolutionary temperature optimization results for the steelmaking process.

[0015] The beneficial effects of this invention are: This invention uses smelting temperature data corresponding to each process stage in steelmaking as its foundation. By preprocessing the original temperature datasets for each stage, a standardized set of stage temperature sequences is generated. Furthermore, temperature change trajectory features, residual temperature features, and temperature continuity features are extracted to form a stage thermal history feature set. A temperature memory chain is then constructed based on this feature set, thereby continuously organizing the temperature change information that was originally scattered within each process stage and at the boundaries between adjacent stages. This allows for a clear characterization of the continuous influence of the thermal state of preceding process stages on the temperature changes of subsequent process stages. Based on this, the invention avoids the problem of thermal history omissions caused by isolated analysis based solely on the temperature data of the current process stage in existing technologies, improving the coherence of temperature analysis in the steelmaking process and the accuracy of identifying stage connections.

[0016] Building upon this foundation, the present invention further decomposes the temperature sequence of the current process stage based on the temperature memory chain, generating a causal decomposition result for the current stage temperature. Based on this, it identifies heat offset, heat decay, heat amplification, and heat transfer misalignment, generating stage transfer distortion results. Then, based on these distortion results, it performs targeted corrections on the heating control path, the holding control path, and the stage transition control path. Finally, it performs continuous temperature optimization control on subsequent process stages, and re-extracts the updated stage thermal history feature set and updates the temperature memory chain from the updated stage temperature sequence, generating a continuously evolving temperature optimization result for the steelmaking process. Therefore, the present invention can not only more accurately distinguish between historical and current influence components in the temperature changes of the current process stage, but also implement corresponding adjustments around different distortion types, improving the targeting, continuity, and reliability of temperature control path correction, enhancing the temperature optimization control effect of subsequent process stages, and ultimately improving the temperature control accuracy, process transition stability, and overall temperature regulation level of the steelmaking process. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of the steelmaking process optimization method based on smelting temperature data proposed in this invention; Figure 2 This is a schematic diagram illustrating the generation of the temperature memory chain in the steelmaking process optimization method based on smelting temperature data proposed in this invention. Detailed Implementation

[0018] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0019] refer to Figures 1-2 The steelmaking process optimization method based on smelting temperature data includes the following steps: S1. Collect smelting temperature data corresponding to each process stage of the steelmaking process, and identify the process stage and time index to form the original temperature dataset of the stage. S2. Perform preprocessing on the original temperature dataset of the stage to generate a standardized stage temperature sequence set; S3. Based on the standardized stage temperature sequence set, extract the temperature change trajectory features within each process stage, the temperature residual features at the end of each process stage, and the temperature transfer features between adjacent process stages to generate a stage thermal history feature set. S4. Based on the stage thermal history feature set, perform cross-stage correlation to generate temperature memory chain; S5. Based on the temperature memory chain, perform causal decomposition on the temperature sequence of the current process stage to generate the causal decomposition result of the temperature of the current stage. S6. Based on the current stage temperature cause decomposition results, identify heat offset, heat decay, heat amplification and heat transfer misalignment, and generate stage transfer distortion results; S7. Based on the stage transmission distortion results, perform correction on the temperature control path of the current process stage and generate the temperature correction result of the current process stage. S8. Based on the temperature correction results of the current process stage, perform continuous temperature optimization control on the subsequent process stages of the steelmaking process, and re-extract the thermal history feature set of the updated stage according to the temperature sequence of the updated stage, update the temperature memory chain, and generate the continuous evolutionary temperature optimization results of the steelmaking process.

[0020] In this embodiment, S2 specifically includes: S21. Read the smelting temperature data corresponding to each process stage in the original temperature dataset of the stage, and arrange the smelting temperature data in the same process stage in sequence according to the time index to form the initial stage temperature sequence corresponding to each process stage. S22. Perform abnormal temperature measurement record removal on the temperature sequence of each initial stage to generate a stage abnormal removal temperature sequence. S23. Based on the temperature continuity relationship corresponding to adjacent time index identifiers in the stage anomaly removal temperature sequence, fill in the missing temperature measurement records at the missing time index identifier positions to form a stage complete temperature sequence. S24. Map the stage completion temperature sequence corresponding to each process stage to a unified time base, align the time index start point and time interval between different process stages, and generate a stage aligned temperature sequence. S25. Based on the continuity of temperature changes before and after the boundary position of adjacent process stages in the stage-aligned temperature sequence, identify stage switching points and generate stage switching point results. When generating the stage switching point, extract the consecutive temperature data entries at the end of the previous process stage and the consecutive temperature data entries at the beginning of the next process stage from the adjacent process stages, and splice them according to the time index identifier order to form the temperature data before and after the boundary position; calculate the temperature difference between adjacent time indices based on the temperature measurement values ​​in the temperature data before and after the boundary position. The positive or negative attribute of the temperature difference is determined as the direction of temperature change, and the absolute value of the temperature difference is determined as the amplitude of temperature change. When the temperature change is within a preset stable range, the corresponding temperature data entry is marked as a stable change; when the temperature change is not within a preset stable range, and three or more consecutive adjacent temperature differences within the same temperature change segment are all greater than 0, the corresponding continuous temperature data entry is marked as a temperature rise; when the temperature change is not within a preset stable range, and three or more consecutive adjacent temperature differences within the same temperature change segment are all less than 0, the corresponding continuous temperature data entry is marked as a temperature fall; when the temperature change is not within a preset stable range, and three or more consecutive adjacent temperature differences within the same temperature change segment switch from positive to negative or from negative to positive, the corresponding continuous temperature data entry is marked as a fluctuating change. The change marker with the largest proportion at the end of the previous process stage is used as the corresponding change marker for the previous process stage, and the change marker with the largest proportion at the beginning of the next process stage is used as the corresponding change marker for the next process stage. The change markers of temperature data before and after the boundary position are retrieved sequentially along the time index identifier. The position where the change marker changes from the corresponding change marker of the previous process stage to the corresponding change marker of the next process stage is determined as the candidate switching position, and the temperature transition deviation corresponding to the candidate switching position is recorded as... When there is When there are candidate switching positions that are not greater than a preset continuous threshold, the position with the earliest time index among the corresponding candidate switching positions is selected as the stage switching point; when there are no such candidate switching positions... When the candidate switching position is not greater than a preset continuous threshold, The smallest candidate switching position is used as the stage switching point. The temperature transition deviation corresponding to the candidate switching position is the absolute value of the difference between the temperature measurement value at the time index before the candidate switching position and the temperature measurement value at the current time index of the candidate switching position, with the candidate switching position as the boundary. This value is used to represent the magnitude of the temperature jump before and after the candidate switching position. S26. Based on the stage switching point results, re-perform stage segmentation on the stage-aligned temperature sequence to obtain a standardized stage temperature sequence set corresponding to each process stage.

[0021] In this embodiment, S3 specifically includes: S31. Read the standardized stage temperature sequence corresponding to each process stage in the standardized stage temperature sequence set, and extract the continuously arranged temperature data entries within each process stage according to the process stage identifier to form a stage internal temperature analysis sequence. S32. For the temperature changes corresponding to adjacent time indexes in the internal temperature analysis sequence of each stage, identify the temperature rise segment, temperature fall segment, temperature fluctuation segment and temperature stability segment in sequence, and collect them according to the arrangement order of each segment in the corresponding process stage to form the temperature change trajectory characteristics corresponding to each process stage. Temperature rise segment, temperature fall segment, temperature fluctuation segment, and temperature stability segment are different temperature change segments formed by temperature data entries arranged continuously according to time index within the same process stage. Among them, temperature rise segment corresponds to the segment where adjacent temperature measurement values ​​continuously increase, temperature fall segment corresponds to the segment where adjacent temperature measurement values ​​continuously decrease, temperature fluctuation segment corresponds to the segment where the direction of change of adjacent temperature measurement values ​​changes, thus forming fluctuations, and temperature stability segment corresponds to the segment where the temperature measurement value remains unchanged or the change amplitude of adjacent temperature measurement values ​​is within a preset stable range. S33. At the end of the standardized stage temperature sequence corresponding to each process stage, continuously extract the temperature data entries within the end segment of the stage according to the time index identifier, and merge the end temperature level, end temperature change direction and end temperature fluctuation of the temperature data entries to form the temperature residual characteristics corresponding to each process stage. The terminal temperature level is quantified by the average of all temperature measurements within the end of the process stage. The direction of terminal temperature change is quantified by the positive or negative attribute of the difference between the last temperature measurement and the first temperature measurement within the end stage. The terminal temperature fluctuation is quantified by the sum of the absolute values ​​of the differences between adjacent temperature measurements within the end stage. S34. For the standardized stage temperature sequence corresponding to adjacent process stages, extract the continuous temperature data entries between the end of the previous process stage and the beginning of the next process stage, and extract the continuous characteristics of temperature connection before and after the boundary position, the characteristics of temperature connection change type and the characteristics of temperature connection amplitude to form the temperature connection characteristics between adjacent process stages. The generation of temperature continuity features specifically includes: extracting consecutive temperature data entries from the end period of the previous process stage and the beginning period of the next process stage, and concatenating the extracted temperature data entries according to the time index to form a boundary temperature continuity analysis sequence; reading the last temperature measurement value from the end period of the previous process stage and the first temperature measurement value from the beginning period of the next process stage in the boundary temperature continuity analysis sequence, calculating the temperature difference between the two, and forming a temperature continuity feature; extracting all consecutive temperature data entries from the end period of the previous process stage and all consecutive temperature data entries from the beginning period of the next process stage, respectively, and calculating the change in the difference between adjacent temperature measurement values ​​in each period; when the difference between adjacent temperature measurement values ​​is consistently positive, the temperature change in the corresponding period is determined to be increasing; when the difference between adjacent temperature measurement values ​​is consistently negative, the temperature change in the corresponding period is determined to be decreasing; when adjacent... When the positive or negative attribute of the temperature measurement difference changes at least once, the temperature change in the corresponding time period is determined to be fluctuating; when the difference between adjacent temperature measurement values ​​is zero or within a preset stable range, the temperature change in the corresponding time period is determined to be stable; the determination results of the end period of the previous process stage and the beginning period of the next process stage are recorded accordingly to form the temperature transition change type feature; the absolute value of the difference between the last temperature measurement value and the first temperature measurement value in the end period of the previous process stage is calculated to obtain the temperature change amplitude of the previous period; the absolute value of the difference between the last temperature measurement value and the first temperature measurement value in the beginning period of the next process stage is calculated to obtain the temperature change amplitude of the next period; the absolute value of the difference between the temperature change amplitude of the previous period and the temperature change amplitude of the next period is recorded as the temperature change transition amplitude, forming the temperature transition amplitude feature; the temperature transition continuity feature, the temperature transition change type feature, and the temperature transition amplitude feature are combined to form the temperature continuity feature between adjacent process stages. The upper limit of the preset stability range is calculated by adding the standard deviation to the average absolute value of the temperature difference between adjacent heats in the historical qualified heats, and the lower limit of the preset stability range is 0. The historical qualified heats are records of past heats that have been completed and judged as qualified by the inspection of tapping temperature, composition and billet quality under the same steel grade and the same process route. S35. Collect the temperature change trajectory characteristics, temperature residual characteristics, and temperature transfer characteristics between adjacent process stages according to the process stage sequence to form a stage thermal history feature set.

[0022] In this embodiment, a temperature data entry is a temperature record in the smelting temperature data corresponding to a single time index identifier, which includes the temperature measurement value at a time point and the corresponding process stage identifier and time index identifier.

[0023] In this embodiment, S4 specifically includes: S41. The temperature residual features, temperature change trajectory features and temperature continuity features corresponding to adjacent process stages are collected in the reading stage thermal history feature set and paired in the order of the preceding process stage first and the subsequent process stage last to form thermal history association units of adjacent process stages. S42. In the thermal history correlation unit of each adjacent process stage, extract the temperature residual features of the preceding process stage and the temperature change trajectory features of the preceding segment of the subsequent process stage, and match them to form the temperature continuity correspondence results between the preceding and following stages. The generation of the corresponding results for the temperature continuity between preceding and subsequent stages specifically includes: reading the residual temperature characteristics corresponding to the preceding process stage and extracting the end temperature level, end temperature change direction, and end temperature fluctuation to form the end temperature characterization data of the preceding process stage; reading the temperature change trajectory characteristics corresponding to the preceding stage and extracting the starting temperature measurement value and temperature change type corresponding to each temperature change segment to form the temperature trajectory data of the preceding stage; comparing the end temperature level with the starting temperature measurement value corresponding to each temperature change segment, selecting the candidate temperature change segment with the smallest absolute value of the difference with the end temperature level, and then determining the target temperature change segment based on the temperature change type corresponding to the candidate temperature change segment to form the temperature level correspondence result; reading the temperature change type corresponding to the target temperature change segment and comparing it with the end temperature change direction. When the end temperature change direction is increasing and the target temperature change segment is a temperature rising segment, it is recorded as consistent direction; when the end temperature change direction is decreasing and the target temperature change segment is a temperature falling segment, it is recorded as consistent direction; when the end temperature change direction is decreasing and the target temperature change segment is a temperature falling segment, it is recorded as consistent direction. When the end temperature change direction is stable and the target temperature change segment is a stable temperature segment, it is recorded as consistent direction; otherwise, it is recorded as inconsistent direction, forming a temperature direction correspondence result. Based on the comparison between the end temperature fluctuation and the preset fluctuation threshold, the temperature fluctuation state at the end of the preceding process stage is determined. When the end temperature fluctuation is greater than the preset fluctuation threshold, the temperature fluctuation state at the end of the preceding process stage is recorded as fluctuating; when the end temperature fluctuation is not greater than the preset fluctuation threshold, the temperature fluctuation state at the end of the preceding process stage is recorded as stable, forming an end temperature fluctuation determination result. The temperature change type corresponding to the target temperature change segment is read. When the end temperature fluctuation determination result is fluctuating and the target temperature change segment is a fluctuating temperature segment, it is recorded as consistent fluctuation; when the end temperature fluctuation determination result is stable and the target temperature change segment is a stable temperature segment, it is recorded as consistent fluctuation; otherwise, it is recorded as inconsistent fluctuation, forming a temperature fluctuation correspondence result. The temperature level correspondence result, the temperature direction correspondence result, and the temperature fluctuation correspondence result are then combined to form a temperature continuity correspondence result between the preceding and following stages. The preset fluctuation threshold is calculated by adding the standard deviation to the average of the cumulative absolute values ​​of the temperature difference between adjacent temperature measurements at the end of the stage in historical qualified furnace cycles. S43. For each adjacent process stage thermal history correlation unit, extract the temperature continuity characteristics between the preceding process stage and the subsequent process stage, organize the temperature continuity relationship before and after the boundary position of adjacent process stages, and form the temperature continuity correlation result at the boundary position. The generation of temperature continuity correlation results at the boundary positions specifically includes: extracting temperature continuity features between preceding and subsequent process stages, and reading the temperature continuity features, temperature change type features, and temperature amplitude features to form basic data for temperature continuity at the boundary positions; determining the continuity status between temperature measurements before and after the boundary positions of adjacent process stages based on the temperature difference corresponding to the temperature continuity features; recording a seamless connection when the temperature difference is zero, an adjacent connection when the temperature difference is greater than zero and not greater than a preset continuity threshold, and a jump connection when the temperature difference is greater than the preset continuity threshold, thus forming a temperature continuity determination result; and analyzing the temperature change type features to determine the relationship between the end of the preceding process stage and the start of the subsequent process stage. The corresponding temperature change type is recorded in sequence as temperature rise segment-temperature rise segment, temperature fall segment-temperature fall segment, temperature stability segment-temperature stability segment, temperature fluctuation segment-temperature fluctuation segment, or other combinations, forming a temperature transition change type determination result; based on the magnitude of the amplitude difference corresponding to the temperature transition amplitude characteristics, the difference in temperature change amplitude before and after the boundary position of adjacent process stages is determined. When the amplitude difference is not greater than the preset amplitude threshold, it is recorded as amplitude close; when the amplitude difference is greater than the preset amplitude threshold, it is recorded as amplitude deviation, forming a temperature transition amplitude determination result; the temperature transition continuity determination result, the temperature transition change type determination result, and the temperature transition amplitude determination result are aggregated according to the boundary position of the same adjacent process stage to form the boundary position temperature connection association result; The preset continuous threshold is calculated by adding the standard deviation to the average absolute value of the temperature difference before and after the boundary position of adjacent process stages in historical qualified furnaces; the preset amplitude threshold is calculated by adding the standard deviation to the average absolute value of the temperature change amplitude before and after the boundary position in historical qualified furnaces. S44. Combine the corresponding results of temperature continuity between the preceding and following stages with the results of temperature connection at the boundary to generate a heat history transfer chain between stages, describing the heat history of the preceding process stage to the subsequent process stage. The generation of inter-stage heat history transfer segments specifically includes: reading the temperature continuity correspondence results of preceding and following stages and the temperature transition correlation results at the boundary positions corresponding to the heat history association units of the same adjacent process stages, and pairing them in the order of preceding process stages first and subsequent process stages last; extracting temperature level correspondence results, temperature direction correspondence results, and temperature fluctuation correspondence results from the temperature continuity correspondence results of preceding and following stages, and extracting temperature connection continuity judgment results, temperature connection change type judgment results, and temperature connection amplitude judgment results from the temperature transition correlation results at the boundary positions; making judgments based on the temperature connection continuity judgment results, recording it as continuous temperature level transfer when the temperature connection continuity judgment result is a seamless connection or an adjacent connection; recording it as a temperature level deviation transfer when the temperature connection continuity judgment result is a jump connection, forming a temperature level transfer result; and determining the temperature direction correspondence results and the temperature connection change type. The results are jointly determined. When the temperature direction is consistent and the temperature change type at the end of the preceding process stage is the same as the temperature change type at the beginning of the subsequent process stage, it is recorded as consistent temperature direction transmission. Otherwise, it is recorded as temperature direction deviation transmission, forming a temperature direction transmission result. The temperature fluctuation result and the temperature connection amplitude determination result are jointly determined. When the temperature fluctuation result is consistent and the temperature connection amplitude determination result is close, it is recorded as temperature fluctuation matching transmission. Otherwise, it is recorded as temperature fluctuation deviation transmission, forming a temperature fluctuation transmission result. The temperature horizontal transmission result, temperature direction transmission result, and temperature fluctuation transmission result are aggregated according to the same adjacent process stage thermal history association unit to form an inter-stage thermal history transmission chain segment that represents the continuation of the thermal history of the preceding process stage to the subsequent process stage. S45. According to the execution order of each process stage in the steelmaking process, the heat history transfer chain segments between each stage are connected in sequence, and the subsequent process stages of the heat history transfer chain segment between the previous stage are sequentially connected with the preceding process stages of the heat history transfer chain segment between the next stage to form a temperature memory chain. By sequentially pairing, corresponding, and connecting the residual temperature characteristics of preceding process stages, the temperature change trajectory characteristics of subsequent process stages, and the temperature transfer characteristics between adjacent process stages, the temperature change information that was originally scattered within each process stage and at the boundary of adjacent process stages is unified into a temperature memory chain that can characterize the thermal history continuity relationship. This allows for a more accurate reflection of the continuous influence of the thermal state of preceding process stages on the temperature changes of subsequent process stages, avoiding the thermal history omission problem caused by isolated judgments based solely on the temperature sequence of the current process stage. Simultaneously, by constructing inter-stage thermal history transfer chains and further forming temperature memory chains, a continuous, clear, and traceable data foundation is provided for subsequent execution of current stage temperature cause decomposition, stage transfer distortion identification, and temperature control path correction. This improves the coherence of temperature analysis in the steelmaking process, the accuracy of stage connection identification, and the reliability of subsequent continuous evolutionary temperature optimization control. The thermal history continuity relationship is the correspondence between the temperature state formed in the preceding process stage and the temperature state formed in the preceding process stage, which is continuously transferred to the subsequent process stage through the boundary of adjacent process stages and reflected in the temperature changes of the subsequent process stage.

[0024] In this embodiment, S5 specifically includes: S51. Read the standardized stage temperature sequence corresponding to the current process stage and the temperature memory chain node corresponding to the current process stage in the temperature memory chain. Extract the stage temperature data entries corresponding to the current process stage and the inter-stage heat history transfer chain segment that continues to be transferred from the previous process stage to the current process stage to form the current stage temperature cause analysis data. A temperature memory chain node is a recording unit in the temperature memory chain corresponding to the boundary position of adjacent process stages, used to record the thermal history continuity relationship corresponding to that boundary position; S52. Based on the current stage temperature cause analysis data, extract the initial time period temperature data in the standardized stage temperature sequence of the current process stage, and extract the temperature horizontal transmission result, temperature directional transmission result, and temperature fluctuation transmission result in the inter-stage heat history transmission chain segment. The initial temperature data consists of consecutive temperature data entries within a preset initial time period after the start of the current process stage, used to characterize the temperature state in the initial stage of the current process stage. S53. Based on the temperature level transfer results, the temperature data portion formed by the continuous transfer of the temperature level from the end of the previous process stage is calibrated in the temperature data at the beginning of the current process stage to form a temperature level decomposition result. The generation of temperature level decomposition results specifically includes: reading the initial temperature data corresponding to the current process stage, and extracting consecutively arranged temperature data entries according to the time index to form the initial temperature analysis data; reading the temperature level transfer result corresponding to the current process stage; when the temperature level transfer result is a continuous temperature level transfer, determining all temperature data entries in the initial temperature analysis data as the temperature data to be calibrated, forming the temperature data to be calibrated; when the temperature level transfer result is a temperature level deviation transfer, reading the difference between the first temperature measurement value and each subsequent temperature measurement value in the initial temperature analysis data, and sequentially extracting consecutive temperature data entries whose difference from the first temperature measurement value is not greater than a preset level threshold, and... The extracted continuous temperature data entries are identified as the temperature data to be calibrated, forming the temperature data to be calibrated. Historical influence labels are assigned to each temperature data entry in the temperature data to be calibrated, forming historical influence temperature level data. The remaining temperature data entries in the initial temperature analysis data that were not assigned historical influence labels are extracted and assigned current influence temperature level labels, forming current influence temperature level data. The historical influence temperature level data is recorded as a historical influence temperature level component, and the current influence temperature level data is recorded as a current influence temperature level component. The corresponding positions of the historical influence temperature level data and the current influence temperature level data in the initial temperature data of the current process stage are then combined to form the temperature level decomposition result. The preset level threshold is calculated by adding the standard deviation to the average of the absolute values ​​of the differences between the first temperature measurement value and subsequent temperature measurement values ​​at the start of the current process stage in historical qualified furnace batches. S54. Based on the temperature direction transmission results, extract the temperature change portion in the temperature sequence of the current process stage standardization stage that is consistent with the temperature change direction at the end of the previous process stage, and record it as the historical influence direction component. Record the remaining temperature change portion as the current action direction component to form the temperature direction decomposition result. The generation of temperature direction decomposition results specifically includes: reading the standardized stage temperature sequence corresponding to the current process stage, and extracting consecutively arranged temperature data entries according to the time index to form the temperature direction analysis data for the current process stage; reading the temperature direction transfer result corresponding to the current process stage, and when the temperature direction transfer result is consistent temperature direction transfer, extracting consecutive temperature data entries in the temperature direction analysis data of the current process stage whose direction of the difference between adjacent temperature measurements is the same as the direction of temperature change at the end of the previous process stage, to form consistent temperature data; when the temperature direction transfer result is deviating temperature direction transfer, starting from the first temperature data entry in the temperature direction analysis data of the current process stage, extracting consecutively arranged temperature data entries to form consistent temperature data; The first continuous temperature data entry with the same direction of difference between adjacent temperature measurements as the end temperature change direction of the previous process stage is selected to form locally consistent temperature data. Each temperature data entry in the consistent temperature data or locally consistent temperature data is assigned a historical influence direction identifier to form historical influence direction data. The remaining temperature data entries in the current process stage temperature direction analysis data that have not been assigned a historical influence direction identifier are extracted and assigned a current action direction identifier to form current action direction data. The historical influence direction data is recorded as historical influence direction components, and the current action direction data is recorded as current action direction components. They are then aggregated according to the same current process stage to form temperature direction decomposition results. S55. Based on the temperature fluctuation transmission results, extract the temperature fluctuation part in the temperature sequence of the current process stage standardization stage that matches the temperature fluctuation state at the end of the previous process stage and record it as the historical influence fluctuation component. Record the remaining temperature fluctuation parts as the current action fluctuation component to form the temperature fluctuation decomposition result. The generation of temperature fluctuation decomposition results specifically includes: reading the standardized stage temperature sequence corresponding to the current process stage, and extracting consecutively arranged temperature data entries according to the time index to form the temperature fluctuation analysis data for the current process stage; reading the temperature fluctuation transmission result corresponding to the current process stage, and when the temperature fluctuation transmission result is a temperature fluctuation matching transmission, extracting consecutive temperature data entries from the current process stage temperature fluctuation analysis data that are the same as the temperature fluctuation state at the end of the previous process stage to form fluctuation matching temperature data; when the temperature fluctuation transmission result is a temperature fluctuation deviation transmission, starting from the first temperature data entry in the current process stage temperature fluctuation analysis data, sequentially extracting... Take the first continuous temperature data entry that is the same as the temperature fluctuation state at the end of the previous process stage to form local fluctuation matching temperature data; assign historical influence fluctuation labels to each temperature data entry in the fluctuation matching temperature data or local fluctuation matching temperature data to form historical influence fluctuation data; extract the remaining temperature data entries in the current process stage temperature fluctuation analysis data that have not been assigned historical influence fluctuation labels, and assign current effect fluctuation labels to them to form current effect fluctuation data; record the historical influence fluctuation data as historical influence fluctuation components and the current effect fluctuation data as current effect fluctuation components, and aggregate them according to the same current process stage to form temperature fluctuation decomposition results; S56. The historical influence temperature level component, the historical influence direction component, and the historical influence fluctuation component in the temperature level decomposition result are aggregated to form the historical influence component. The current influence temperature level component, the current influence direction component, and the current influence fluctuation component in the temperature fluctuation decomposition result are aggregated to form the current influence component. S57. Record the historical influence components and the current effect components according to the same current process stage, and generate the temperature cause decomposition results for the current stage.

[0025] In this embodiment, S6 specifically includes: S61. Obtain the current stage temperature cause decomposition results and extract historical influence components and current action components; S62. Read the historical influence temperature level component in the historical influence component and compare it with the temperature data of the starting time corresponding to the current process stage. Determine the temperature difference after the temperature level at the end of the previous process stage is transferred to the starting temperature state of the current process stage, and form the heat offset analysis result. S63. Count the number of consecutive temperature data entries corresponding to the starting position of the temperature sequence of the standardized stage in the current process stage, determine the decay state of the thermal history of the previous process stage in the current process stage, and form the heat decay analysis results. The generation of heat decay analysis results specifically includes: reading historical influence components and extracting temperature data entries arranged continuously in time index order from the standardized stage temperature sequence corresponding to the current process stage to form the basic data for decay analysis; identifying temperature data entries labeled as historical influence components in the basic data for decay analysis to form historical influence temperature data; starting from the first temperature data entry in the standardized stage temperature sequence of the current process stage, sequentially determining whether each temperature data entry belongs to historical influence temperature data, and cumulatively counting the temperature data entries that continuously belong to historical influence temperature data to form a continuous corresponding quantity result; when the first temperature that does not belong to historical influence temperature appears... When accumulating temperature data entries, stop the cumulative counting and record the consecutive corresponding results as the number of temperature data entries. Determine the decay status based on the number of temperature data entries. When the number of temperature data entries is zero, it is recorded as complete decay. When the number of temperature data entries is greater than zero and less than the total number of temperature data entries corresponding to the temperature sequence of the current process stage's standardized stage, it is recorded as decay. When the number of temperature data entries is equal to the total number of temperature data entries corresponding to the temperature sequence of the current process stage's standardized stage, it is recorded as no decay, forming a thermal history decay determination result. Record the thermal history decay determination results according to the same current process stage to form a heat decay analysis result. S64. Extract the distribution position and distribution length of the current action component and the historical influence component in the temperature sequence of the current process stage standardization stage, determine the degree of enhancement of the current action component to the historical influence component, and form the heat amplification analysis results. The generation of heat amplification analysis results specifically includes: reading the current active component and historical influence components, and extracting temperature data entries arranged sequentially according to time index from the standardized stage temperature sequence corresponding to the current process stage to form the basic data for amplification analysis; identifying the temperature data entries labeled as the current active component and the temperature data entries labeled as historical influence components from the basic data for amplification analysis, forming the current active temperature data and historical influence temperature data respectively; determining the distribution position of the current active component based on the first and last positions of the current active temperature data in the standardized stage temperature sequence, and determining the distribution length of the current active component based on the number of temperature data entries corresponding to the current active temperature data, forming the current active distribution result; and determining the distribution of the historical influence components based on the first and last positions of the historical influence temperature data in the standardized stage temperature sequence. The distribution location is determined, and the distribution length of the historical influence component is determined based on the number of temperature data entries corresponding to the historical influence temperature data, forming a historical influence distribution result. The current influence distribution result is compared with the historical influence distribution result. When the distribution location of the current influence component overlaps with the distribution location of the historical influence component, and the distribution length of the current influence component is greater than the distribution length of the historical influence component, it is recorded as enhanced transmission. When the distribution location of the current influence component overlaps with the distribution location of the historical influence component, and the distribution length of the current influence component is not greater than the distribution length of the historical influence component, it is recorded as local enhanced transmission. When the distribution location of the current influence component does not overlap with the distribution location of the historical influence component, it is recorded as no enhanced transmission, forming a heat amplification judgment result. The heat amplification judgment result is recorded according to the same current process stage, forming a heat amplification analysis result. S65. Extract the starting position of the historical influence component in the temperature sequence of the current process stage standardization stage. Based on the positional difference between the starting position and the starting position of the temperature sequence of the current process stage standardization stage, determine the positional deviation of the thermal history in the current process stage and form the heat transfer misalignment analysis results. The generation of heat transfer misalignment analysis results specifically includes: reading historical influence components and extracting temperature data entries arranged sequentially according to time index in the standardized stage temperature sequence corresponding to the current process stage to form misalignment analysis basic data; identifying temperature data entries identified as historical influence components in the misalignment analysis basic data and extracting the position of the first historical influence temperature data entry in the standardized stage temperature sequence of the current process stage to form the historical influence starting position result; extracting the position of the first temperature data entry in the standardized stage temperature sequence of the current process stage to form the current process stage starting position result; comparing the historical influence starting position result with the current process stage starting position result to determine the position difference value between the two to form the position difference result; judging based on the position difference result: when the position difference value is equal to zero, it is recorded as no misalignment; when the position difference value is greater than zero and not greater than a preset misalignment threshold, it is recorded as slight misalignment; when the position difference value is greater than the preset misalignment threshold, it is recorded as misalignment, forming a heat transfer misalignment judgment result; recording the heat transfer misalignment judgment result according to the same current process stage to form the heat transfer misalignment analysis result; The preset misalignment threshold is calculated by rounding up the average number of indexes of the lag time between the starting position of the historical influence component and the starting position of the current process stage in the historical qualified furnace batches. S66. The results of heat offset analysis, heat decay analysis, heat amplification analysis, and heat transfer misalignment analysis are aggregated according to the same current process stage to generate the stage transfer distortion result.

[0026] In this embodiment, S7 specifically includes: S71. Read the distortion results during the reading stage and extract the heat offset analysis results, heat attenuation analysis results, heat amplification analysis results and heat transfer misalignment analysis results; S72. Based on the heat offset analysis results, the heating control path corresponding to the current process stage is corrected. Specifically, when the temperature difference between the end temperature level of the previous process stage and the starting temperature state of the current process stage, as indicated by the heat offset analysis results, is greater than zero, the heating amplitude corresponding to the heating control path is reduced; when the temperature difference is less than zero, the heating amplitude corresponding to the heating control path is increased; when the temperature difference is equal to zero, the heating control path is kept unchanged, thus forming the heating control path correction result. S73. Based on the heat decay analysis results, adjust the insulation control path corresponding to the current process stage. Specifically, when the heat decay analysis result is complete decay, extend the insulation time corresponding to the insulation control path; when the heat decay analysis result is decay, increase the insulation strength corresponding to the insulation control path; when the heat decay analysis result is no decay, keep the insulation control path unchanged, and form the insulation control path correction result. S74. Based on the results of heat amplification analysis and heat transfer misalignment analysis, the stage connection control path corresponding to the current process stage is corrected. Specifically, when the heat amplification analysis result is enhanced transfer or locally enhanced transfer, the stage connection time corresponding to the current process stage is shortened; when the heat transfer misalignment analysis result is slight misalignment or misalignment, the starting position of the stage connection corresponding to the current process stage is adjusted to form the stage connection control path correction result. The heating control path is the adjustment path used to control the temperature rise process in the current process stage; the heat preservation control path is the adjustment path used to control the temperature holding process in the current process stage; and the stage transition control path is the adjustment path used to control the stage switching and temperature transition process between adjacent process stages. S75. The temperature rise control path correction results, the heat preservation control path correction results, and the stage transition control path correction results are aggregated according to the same current process stage to generate the temperature correction results for the current process stage.

[0027] In this embodiment, S8 specifically includes: S81. Read the temperature correction result of the current process stage and apply it to the subsequent process stages adjacent to the current process stage according to the process execution order to form the temperature control data of the subsequent process stages. S82. Based on the temperature control data of the subsequent process stage, the smelting temperature data corresponding to the subsequent process stage is continuously adjusted, and the adjusted temperature data entries are recorded in the order of time index identification to form the temperature sequence of the subsequent process stage update stage. S83. Based on the temperature sequence of the subsequent process stage update stage, form a thermal history feature set of the update stage and form an update temperature memory chain; S84. Following the execution sequence of each process stage in the steelmaking process, based on the updated temperature memory chain, continue to perform temperature regulation on the next subsequent process stage, and repeatedly form the updated stage temperature sequence and the updated stage thermal history feature set until the continuous temperature optimization control of the subsequent process stages of the steelmaking process is completed, generating the continuous evolutionary temperature optimization result of the steelmaking process.

[0028] A steelmaking process system optimized based on smelting temperature data includes: The data acquisition and identification module is used to collect smelting temperature data corresponding to each process stage of the steelmaking process, and to identify the process stage and time index to form the original temperature dataset of the stage. The preprocessing module is used to perform preprocessing and generate a standardized set of stage temperature sequences; The thermal history feature extraction module is used to generate a stage thermal history feature set based on a standardized stage temperature sequence set. The temperature memory chain generation module is used to generate temperature memory chains by performing cross-stage associations based on the stage thermal history feature set. The causal decomposition module is used to perform causal decomposition on the temperature sequence of the current process stage based on the temperature memory chain, and generate the causal decomposition results of the temperature of the current stage. The stage transfer distortion identification module is used to identify heat offset, heat decay, heat amplification, and heat transfer misalignment based on the current stage temperature cause decomposition results, and generate stage transfer distortion results. The temperature control path correction module is used to correct the temperature control path of the current process stage based on the stage transmission distortion results, and generate the temperature correction result of the current process stage. The continuous temperature optimization control module is used to perform continuous temperature optimization control on subsequent process stages of steelmaking based on the temperature correction results of the current process stage, and to re-extract and update the stage thermal history feature set, update the temperature memory chain, and generate continuous evolutionary temperature optimization results for the steelmaking process.

[0029] Example 1: To verify the feasibility of this invention in practice, it was applied to the actual smelting process in a steelmaking workshop of a steel company. The application was conducted within a continuous production cycle, and the application locations were the production areas corresponding to the converter station, refining station, and subsequent continuous casting pretreatment station. In this scenario, the steelmaking process includes multiple closely linked process stages. Although each process stage has independent operational tasks, temperature changes are not isolated from each other. The thermal state formed in the preceding process stage continuously affects subsequent process stages along the process connection relationship. Existing temperature control methods typically focus on judging and adjusting the temperature state of the current process stage, easily overlooking the continuous effect of residual temperature from preceding process stages on subsequent stages. Especially under production conditions with frequent stage switching, continuous changes in thermal state, and significant local fluctuations, it is often difficult to identify the source of temperature deviation caused by the continuation of thermal history in a timely manner, resulting in insufficient smoothness of stage connection control, unclear basis for temperature adjustment in the current process stage, and insufficient continuity of temperature optimization control in subsequent process stages. This invention is deployed in this production scenario. By continuously collecting smelting temperature data corresponding to each process stage of the steelmaking process and synchronously writing process stage identifiers and time index identifiers, a stage original temperature dataset is formed, so that temperature changes throughout the entire steelmaking process can be continuously recorded and tracked on a unified data basis.

[0030] This invention is applied to a continuous steel production process, which sequentially passes through a converter tapping stage, a refining stage, and a continuous casting pretreatment stage. The temperature measurement system collects smelting temperature data at 1-minute sampling intervals. In this embodiment, the preset horizontal threshold is 6℃, the preset fluctuation threshold is 8℃, and the preset continuous threshold is 4℃. Taking one heat as an example, the temperature measurements at the end of the converter tapping stage are 1568℃, 1572℃, 1570℃, and 1573℃, respectively. The system calculates the final temperature level as 1570.75℃. The cumulative absolute value of the adjacent temperature differences is 9℃, and the positive / negative attribute of the temperature difference switches twice. Therefore, the end of the converter tapping stage is determined to be in a fluctuating state.

[0031] The temperature measurements at the beginning of the refining stage were 1575℃, 1578℃, 1581℃, and 1584℃, respectively. The absolute difference between the first temperature measurement of 1575℃ and the final temperature level of 1570.75℃ at the end of the converter tapping stage was 4.25℃, which is less than the preset threshold of 6℃. Therefore, the system determined that the thermal state at the end of the converter tapping stage had a historical influence on the starting temperature of the refining stage. The absolute difference between the last temperature measurement of 1573℃ at the end of the converter tapping stage and the first temperature measurement of 1575℃ at the beginning of the refining stage was 2℃, which is less than the preset continuous threshold of 4℃. Therefore, the boundary position was recorded as adjacent connection. During the initial period of the refining stage, the adjacent temperature differences were all greater than 0, which was recorded as a temperature rise. The direction of temperature change at the end of the converter tapping stage was also a temperature rise, thus generating a consistent temperature direction transmission. Therefore, the system formed an inter-stage heat history transmission chain from the converter tapping stage to the refining stage, containing continuous temperature level transmission, consistent temperature direction transmission, and temperature fluctuation deviation transmission.

[0032] To verify the performance of the present invention, it was compared with the traditional method. The comparison results are shown in Table 1.

[0033] Table 1. Comparison of Comprehensive Performance of Continuous Temperature Optimization Control in Steelmaking Processes

[0034] As shown in Table 1, the method of this invention outperforms the traditional method in all key indicators. Specifically, the average initial temperature deviation of the process stage decreased from 18.6℃ to 7.9℃, indicating that this invention can more accurately inherit the thermal state of the preceding process stage and reduce temperature deviation after stage switching. The temperature recovery stabilization time after stage switching was shortened from 11.4 min to 6.8 min, and the average temperature fluctuation amplitude of subsequent process stages decreased from 24.1℃ to 12.7℃, demonstrating that this invention can improve the continuity and stability of temperature control.

[0035] Meanwhile, the accuracy rate of stage transition anomaly identification increased from 81.3% to 93.8%, indicating that the present invention is more accurate in identifying heat offset, heat attenuation, heat amplification, and heat transfer misalignment. The number of temperature control adjustments decreased from 5.6 times / furnace to 3.2 times / furnace, indicating that the present invention can reduce redundant adjustments and improve control efficiency. The furnace temperature compliance rate increased from 88.7% to 96.4%, indicating that the present invention can improve the overall temperature optimization control effect of the steelmaking process. The main reason for this is that the present invention constructs a temperature memory chain, combines the current stage temperature cause decomposition results and stage transfer distortion results, and performs targeted corrections on the heating control path, holding control path, and stage transition control path, thereby improving the temperature control accuracy and process transition stability.

[0036] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A steelmaking process optimization method based on smelting temperature data, characterized in that, Includes the following steps: S1. Collect smelting temperature data corresponding to each process stage of the steelmaking process, and identify the process stage and time index to form the original temperature dataset of the stage. S2. Perform preprocessing on the original temperature dataset of the stage to generate a standardized stage temperature sequence set; S3. Based on the standardized stage temperature sequence set, extract the temperature change trajectory features within each process stage, the temperature residual features at the end of each process stage, and the temperature transfer features between adjacent process stages to generate a stage thermal history feature set. S4. Based on the stage thermal history feature set, perform cross-stage correlation to generate a temperature memory chain; S5. Based on the temperature memory chain, perform causal decomposition on the temperature sequence of the current process stage to generate the causal decomposition result of the temperature of the current stage. S6. Based on the current stage temperature cause decomposition results, identify heat offset, heat decay, heat amplification and heat transfer misalignment, and generate stage transfer distortion results; S7. Based on the stage transmission distortion results, perform correction on the temperature control path of the current process stage and generate the temperature correction result of the current process stage. S8. Based on the temperature correction results of the current process stage, perform continuous temperature optimization control on the subsequent process stages of the steelmaking process, and re-extract the thermal history feature set of the updated stage according to the temperature sequence of the updated stage, update the temperature memory chain, and generate the continuous evolution temperature optimization results of the steelmaking process. S4 specifically includes: S41. The temperature residual features, temperature change trajectory features and temperature continuity features corresponding to adjacent process stages are collected in the reading stage thermal history feature set and paired in the order of the preceding process stage first and the subsequent process stage last to form thermal history association units of adjacent process stages. S42. In the thermal history correlation unit of each adjacent process stage, extract the temperature residual features of the preceding process stage and the temperature change trajectory features of the preceding segment of the subsequent process stage, and match them to form the temperature continuity correspondence results between the preceding and following stages. S43. For each adjacent process stage thermal history correlation unit, extract the temperature continuity characteristics between the preceding process stage and the subsequent process stage, organize the temperature continuity relationship before and after the boundary position of adjacent process stages, and form the temperature continuity correlation result at the boundary position. S44. Combine the corresponding results of temperature continuity between the preceding and following stages with the results of temperature connection at the boundary to generate a heat history transfer chain between stages, describing the heat history of the preceding process stage to the subsequent process stage. S45. According to the execution order of each process stage in the steelmaking process, the heat history transfer chain segments between each stage are connected in sequence, and the subsequent process stages of the heat history transfer chain segment between the previous stage are sequentially connected with the preceding process stages of the heat history transfer chain segment between the next stage to form a temperature memory chain. S5 specifically includes: S51. Read the standardized stage temperature sequence corresponding to the current process stage and the temperature memory chain node corresponding to the current process stage in the temperature memory chain. Extract the stage temperature data entries corresponding to the current process stage and the inter-stage heat history transfer chain segment that continues to be transferred from the previous process stage to the current process stage to form the current stage temperature cause analysis data. S52. Based on the current stage temperature cause analysis data, extract the initial time period temperature data in the standardized stage temperature sequence of the current process stage, and extract the temperature horizontal transmission result, temperature directional transmission result, and temperature fluctuation transmission result in the inter-stage heat history transmission chain segment. S53. Based on the temperature level transfer results, the temperature data portion formed by the continuous transfer of the temperature level from the end of the previous process stage is calibrated in the temperature data at the beginning of the current process stage to form a temperature level decomposition result. S54. Based on the temperature direction transmission results, extract the temperature change portion in the temperature sequence of the current process stage standardization stage that is consistent with the temperature change direction at the end of the previous process stage, and record it as the historical influence direction component. Record the remaining temperature change portion as the current action direction component to form the temperature direction decomposition result. S55. Based on the temperature fluctuation transmission results, extract the temperature fluctuation part in the temperature sequence of the current process stage standardization stage that matches the temperature fluctuation state at the end of the previous process stage and record it as the historical influence fluctuation component. Record the remaining temperature fluctuation parts as the current action fluctuation component to form the temperature fluctuation decomposition result. S56. The historical influence temperature level component, the historical influence direction component, and the historical influence fluctuation component in the temperature level decomposition result are aggregated to form the historical influence component. The current influence temperature level component, the current influence direction component, and the current influence fluctuation component in the temperature fluctuation decomposition result are aggregated to form the current influence component. S57. Record the historical influence components and the current effect components according to the same current process stage, and generate the temperature cause decomposition results for the current stage.

2. The steelmaking process optimization method based on smelting temperature data according to claim 1, characterized in that, S2 specifically includes: S21. Read the smelting temperature data corresponding to each process stage in the original temperature dataset of the stage, and arrange the smelting temperature data in the same process stage in sequence according to the time index to form the initial stage temperature sequence corresponding to each process stage. S22. Perform abnormal temperature measurement record removal on the temperature sequence of each initial stage to generate an abnormal temperature sequence of each stage. S23. Based on the temperature continuity relationship corresponding to adjacent time index identifiers in the stage anomaly removal temperature sequence, fill in the missing temperature measurement records at the missing time index identifier positions to form a stage complete temperature sequence. S24. Map the stage completion temperature sequence corresponding to each process stage to a unified time base, align the time index start point and time interval between different process stages, and generate a stage aligned temperature sequence. S25. Based on the continuity of temperature changes before and after the boundary position of adjacent process stages in the stage-aligned temperature sequence, identify stage switching points and generate stage switching point results. S26. Based on the stage switching point results, re-perform stage segmentation on the stage-aligned temperature sequence to obtain a standardized stage temperature sequence set corresponding to each process stage.

3. The steelmaking process optimization method based on smelting temperature data according to claim 1, characterized in that, S3 specifically includes: S31. Read the standardized stage temperature sequence corresponding to each process stage in the standardized stage temperature sequence set, and extract the continuously arranged temperature data entries within each process stage according to the process stage identifier to form a stage internal temperature analysis sequence. S32. For the temperature changes corresponding to adjacent time indexes in the internal temperature analysis sequence of each stage, identify the temperature rise segment, temperature fall segment, temperature fluctuation segment and temperature stability segment in sequence, and collect them according to the arrangement order of each segment in the corresponding process stage to form the temperature change trajectory characteristics corresponding to each process stage. S33. At the end of the standardized stage temperature sequence corresponding to each process stage, continuously extract the temperature data entries within the end segment of the stage according to the time index identifier, and merge the end temperature level, end temperature change direction and end temperature fluctuation of the temperature data entries to form the temperature residual characteristics corresponding to each process stage. S34. For the standardized stage temperature sequence corresponding to adjacent process stages, extract the continuous temperature data entries between the end of the previous process stage and the beginning of the next process stage, and extract the continuous characteristics of temperature connection before and after the boundary position, the characteristics of temperature connection change type and the characteristics of temperature connection amplitude to form the temperature connection characteristics between adjacent process stages. S35. Collect the temperature change trajectory characteristics, temperature residual characteristics, and temperature transfer characteristics between adjacent process stages according to the process stage sequence to form a stage thermal history feature set.

4. The steelmaking process optimization method based on smelting temperature data according to claim 3, characterized in that, The temperature data entry is a temperature record in the smelting temperature data corresponding to a single time index identifier, which includes the temperature measurement value at a time point and the corresponding process stage identifier and time index identifier.

5. The steelmaking process optimization method based on smelting temperature data according to claim 1, characterized in that, S6 specifically includes: S61. Obtain the current stage temperature cause decomposition results and extract historical influence components and current action components; S62. Read the historical influence temperature level component in the historical influence component and compare it with the temperature data of the starting time corresponding to the current process stage. Determine the temperature difference after the temperature level at the end of the previous process stage is transferred to the starting temperature state of the current process stage, and form the heat offset analysis result. S63. Count the number of consecutive temperature data entries corresponding to the starting position of the temperature sequence of the standardized stage in the current process stage, determine the decay state of the thermal history of the previous process stage in the current process stage, and form the heat decay analysis results. S64. Extract the distribution position and distribution length of the current action component and the historical influence component in the temperature sequence of the current process stage standardization stage, determine the degree of enhancement of the current action component to the historical influence component, and form the heat amplification analysis results. S65. Extract the starting position of the historical influence component in the temperature sequence of the current process stage standardization stage. Based on the positional difference between the starting position and the starting position of the temperature sequence of the current process stage standardization stage, determine the positional deviation of the thermal history in the current process stage and form the heat transfer misalignment analysis results. S66. The results of heat offset analysis, heat decay analysis, heat amplification analysis, and heat transfer misalignment analysis are aggregated according to the same current process stage to generate the stage transfer distortion result.

6. The steelmaking process optimization method based on smelting temperature data according to claim 1, characterized in that, Specifically, S7 includes: S71. Read the distortion results during the reading stage and extract the heat offset analysis results, heat attenuation analysis results, heat amplification analysis results and heat transfer misalignment analysis results; S72. Based on the heat offset analysis results, the heating control path corresponding to the current process stage is corrected to form the heating control path correction result. S73. Based on the heat decay analysis results, adjust the heat preservation control path corresponding to the current process stage to form a heat preservation control path correction result. S74. Based on the results of heat amplification analysis and heat transfer misalignment analysis, the stage connection control path corresponding to the current process stage is corrected to form the stage connection control path correction result. S75. The temperature rise control path correction results, the heat preservation control path correction results, and the stage transition control path correction results are aggregated according to the same current process stage to generate the temperature correction results for the current process stage.

7. The steelmaking process optimization method based on smelting temperature data according to claim 1, characterized in that, S8 specifically includes: S81. Read the temperature correction result of the current process stage and apply it to the subsequent process stages adjacent to the current process stage according to the process execution order to form the temperature control data of the subsequent process stages. S82. Based on the temperature control data of the subsequent process stage, the smelting temperature data corresponding to the subsequent process stage is continuously adjusted, and the adjusted temperature data entries are recorded in the order of time index identification to form the temperature sequence of the subsequent process stage update stage. S83. Based on the temperature sequence of the subsequent process stage update stage, form a thermal history feature set of the update stage and form an update temperature memory chain; S84. Following the execution sequence of each process stage in the steelmaking process, based on the updated temperature memory chain, continue to perform temperature regulation on the next subsequent process stage, and repeatedly form the updated stage temperature sequence and the updated stage thermal history feature set until the continuous temperature optimization control of the subsequent process stages of the steelmaking process is completed, generating the continuous evolutionary temperature optimization result of the steelmaking process.

8. A steelmaking process system optimized based on smelting temperature data, executing the steelmaking process method optimized based on smelting temperature data as described in any one of claims 1 to 7, characterized in that, include: The data acquisition and identification module is used to collect smelting temperature data corresponding to each process stage of the steelmaking process, and to identify the process stage and time index to form the original temperature dataset of the stage. The preprocessing module is used to perform preprocessing and generate a standardized set of stage temperature sequences; The thermal history feature extraction module is used to generate a stage thermal history feature set based on a standardized stage temperature sequence set. The temperature memory chain generation module is used to generate temperature memory chains by performing cross-stage associations based on the stage thermal history feature set. The causal decomposition module is used to perform causal decomposition on the temperature sequence of the current process stage based on the temperature memory chain, and generate the causal decomposition results of the temperature of the current stage. The stage transfer distortion identification module is used to identify heat offset, heat decay, heat amplification, and heat transfer misalignment based on the current stage temperature cause decomposition results, and generate stage transfer distortion results. The temperature control path correction module is used to correct the temperature control path of the current process stage based on the stage transmission distortion results, and generate the temperature correction result of the current process stage. The continuous temperature optimization control module is used to perform continuous temperature optimization control on subsequent process stages of steelmaking based on the temperature correction results of the current process stage, and to re-extract and update the stage thermal history feature set, update the temperature memory chain, and generate continuous evolutionary temperature optimization results for the steelmaking process.

Images