A method for optimizing a forging heat treatment process based on a temperature control curve

Abstract

The application discloses a kind of based on temperature control curve's forging heat treatment process optimization method, it is related to data processing technical field, including, based on process task data packet, temperature control performance prediction model is constructed and reverse solution is executed, generates core temperature control reference curve;To core temperature control reference curve executes heat response mapping, obtains furnace temperature control trajectory, and obtains online core tracking control scheme by heat treatment risk timing evaluation;Online core tracking control scheme is executed, process data is recorded synchronously, and combined furnace temperature control trajectory is passed through heat history inversion operation, obtains execution temperature control curve;Based on execution temperature control curve, temperature control performance prediction model is optimized, and obtains process quality evaluation result by performance deviation evaluation.The application is by constructing temperature control performance prediction model and executing reverse solution and feedback optimization, improves the calculation consistency of process parameter design and the quantifiable and iterative optimization ability of result evaluation.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to an optimization method for the heat treatment process of forgings based on temperature control curves. Background Technology

[0002] With the continuous development of data processing, industrial informatization, and complex process modeling methods, process parameter modeling, prediction, and optimization based on multi-source data have gradually become an important research direction for digital upgrading. Existing technologies, for complex industrial processes, collect multi-dimensional parameters related to process execution, construct process performance prediction models, and evaluate and adjust parameter configuration schemes to achieve continuous optimization of process quality. In recent years, with the improvement of model computing power and data processing capabilities, some technologies have begun to introduce predictive models to support the intelligentization of parameter configuration processes, providing a new technical path for the digital management of complex processes.

[0003] However, existing data processing techniques based on process data still have certain limitations in practical applications. On the one hand, they focus on forward prediction or post-event analysis of process execution results, lacking a backward reasoning mechanism centered on target performance constraints, resulting in the generation of process parameters still relying on empirical rules. On the other hand, the utilization of process feedback data is relatively crude, often only used for simple corrections and offline analysis, lacking time-dimensional deviation assessment and model self-consistent update mechanisms, which limits the adaptability of data processing in complex process optimization scenarios and the reliability of decision-making. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method for optimizing the heat treatment process of forgings based on temperature control curves, which solves the difficulties in reverse generation of temperature control curves and quantitative optimization of process risks.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] This invention provides a method for optimizing the heat treatment process of forgings based on temperature control curves. The method includes: collecting heat treatment process task parameters of the forgings and preprocessing them to obtain a process task data package; constructing a temperature control performance prediction model based on the process task data package and performing inverse solving to generate a core temperature control reference curve; performing thermal response mapping on the core temperature control reference curve to obtain the furnace temperature control trajectory, and obtaining an online core tracking control scheme through heat treatment risk time-series assessment; executing the online core tracking control scheme, synchronously recording process data, and obtaining the executed temperature control curve through thermal history inversion calculation combined with the furnace temperature control trajectory; optimizing the temperature control performance prediction model based on the executed temperature control curve, and obtaining process quality assessment results through performance deviation assessment.

[0008] As a preferred embodiment of the optimization method for the heat treatment process of forgings based on temperature control curves described in this invention, the heat treatment process parameters of the forgings include forging material, geometric dimensions, furnace loading method, hardness requirements, and residual stress requirements.

[0009] The preprocessing includes integrity checks, unit unification, and structured organization.

[0010] As a preferred embodiment of the optimization method for the heat treatment process of forgings based on temperature control curves described in this invention, the process task data package includes forging information data and performance requirement data.

[0011] As a preferred embodiment of the forging heat treatment process optimization method based on temperature control curves described in this invention, the specific steps for constructing a temperature control performance prediction model based on process task data packages are as follows:

[0012] The process task data package is parsed and then broken down into a forging information data set and a performance requirement data set through classification operations.

[0013] Based on the forging information dataset, a temperature control feature extraction algorithm is used to extract and map attributes related to temperature and time to generate a set of temperature control variables.

[0014] Based on the performance requirement data set and the temperature control variable set, a temperature control performance mapping structure is generated through constraint-driven mapping.

[0015] Based on the process task data package, the predicted values ​​of hardness and residual stress are calculated, and the parameters in the temperature control performance mapping structure are adjusted to obtain the temperature control performance prediction model.

[0016] As a preferred embodiment of the optimization method for the heat treatment process of forgings based on temperature control curves described in this invention, the specific steps for generating the core temperature control reference curve are as follows:

[0017] Based on the temperature control performance prediction model, the target set is obtained by reconstructing the target performance and solving it in reverse.

[0018] The target set is solved in reverse to search for temperature control variables, generating an initial sequence of temperature control variables. Then, the performance deviation direction is analyzed to generate a sequence of temperature control variable correction directions.

[0019] The temperature control variable correction direction sequence and the initial temperature control variable sequence are solved in reverse. The temperature time axis reconstruction operation is performed to obtain the core temperature control reference curve.

[0020] As a preferred embodiment of the optimization method for the heat treatment process of forgings based on temperature control curves described in this invention, the step of performing thermal response mapping on the core temperature control reference curve to obtain the furnace temperature control trajectory refers to converting the core temperature control reference curve into a time-varying furnace target temperature sequence through a thermal response mapping algorithm to generate the furnace temperature control trajectory.

[0021] As a preferred embodiment of the optimization method for the heat treatment process of forgings based on temperature control curves described in this invention, the specific steps for obtaining the online core tracking control scheme are as follows:

[0022] By combining the furnace temperature control trajectory and process task data package, the heat treatment risk level is calculated through the heat treatment risk time series assessment algorithm, and a heat treatment risk time series curve is generated.

[0023] Based on the heat treatment risk time series curve and furnace temperature control trajectory, the combination of adjustment variables is determined by the online control strategy synthesis method to generate an online core tracking control scheme.

[0024] As a preferred embodiment of the forging heat treatment process optimization method based on temperature control curves described in this invention, the online core tracking control scheme is implemented, process data is recorded synchronously, and the execution temperature control curve is obtained through thermal history inversion calculation combined with the furnace temperature control trajectory. The specific steps are as follows.

[0025] Based on the online core tracking control scheme, the furnace temperature is adjusted and the furnace temperature, surface temperature and core temperature are recorded simultaneously to generate a process data sequence;

[0026] By comparing the cardiac temperature control reference curve with the process data sequence, and by performing thermal history inversion calculations, the actual cardiac temperature control curve is obtained.

[0027] The actual temperature control curve of the core is combined with the temperature control trajectory of the furnace temperature control trajectory to obtain the execution temperature control curve.

[0028] As a preferred embodiment of the optimization method for the heat treatment process of forgings based on temperature control curves described in this invention, the optimization of the temperature control performance prediction model based on the executed temperature control curves includes the following specific steps.

[0029] Based on the execution temperature control curve, a set of performance prediction time series is generated by replaying the temperature time history of the execution temperature control curve;

[0030] Based on the performance prediction time series set and performance requirement data, a two-dimensional performance deviation distribution of time indicators is constructed and the deviation peak and deviation accumulation are extracted to generate a performance deviation field.

[0031] Based on the performance deviation field, the parameters of the temperature control performance prediction model are optimized by self-consistent write-back to generate an optimized temperature control performance prediction model.

[0032] As a preferred embodiment of the optimization method for the heat treatment process of forgings based on temperature control curves described in this invention, the process quality assessment result is obtained by comprehensively quantifying and classifying the performance achievement degree, deviation distribution characteristics and model convergence state based on the optimized temperature control performance prediction model.

[0033] The beneficial effects of this invention are as follows: by constructing a temperature control performance prediction model and performing inverse solving and feedback optimization, the generation and optimization of temperature control curves based on target performance constraints are realized, which improves the calculation consistency of process parameter design and the quantifiable and iterative optimization capabilities of result evaluation. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart of an optimization method for the heat treatment process of forgings based on temperature control curves.

[0036] Figure 2 A flowchart for generating a cardiac temperature control reference curve.

[0037] Figure 3 A flowchart for obtaining an online cardiac tracking and control scheme.

[0038] Figure 4 A flowchart for obtaining process quality assessment results. Detailed Implementation

[0039] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0040] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0041] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0042] Reference Figures 1-4 This is one embodiment of the present invention, which provides a method for optimizing the heat treatment process of forgings based on temperature control curves, including the following steps:

[0043] S1: Collect the process parameters of the forging heat treatment process and obtain the process task data package through preprocessing.

[0044] S1.1: The process parameters for heat treatment of forgings include forging material, geometric dimensions, furnace loading method, hardness requirements, and residual stress requirements;

[0045] Specifically, a spectrometer is used to perform spectral analysis on the surface of the forging, and the results are used to collect material information of the forging.

[0046] The forging is scanned as a whole using a coordinate measuring machine or laser scanning equipment, and the external dimensions and key cross-sectional dimensions are extracted from the scanning results to obtain geometric dimensional information.

[0047] During the furnace loading process, position sensors and hanging force sensors are used to record the spatial coordinates, number of layers, and hanging status of the forgings in the furnace, and to collect information on the loading method.

[0048] By using a hardness tester to perform multi-point hardness measurements at predetermined test locations on representative qualified forgings during the process verification phase, and statistically obtaining the target hardness range (representing the range obtained from multi-point hardness measurements on representative qualified forgings during the process verification phase), for each control location in the residual stress requirement data, the time interval from the start of heating to when the residual stress is stable is recorded as the control time range field, generating hardness requirement information.

[0049] During the process verification phase, residual stress was measured in the stress control area of ​​representative qualified forgings, and the allowable residual stress range was statistically obtained to acquire residual stress requirement information.

[0050] S1.2: Preprocessing includes integrity checks, unit unification, and structured organization;

[0051] Specifically, each item was checked to confirm that the forging material, geometric dimensions, and furnace loading method were filled in; the strength grade related to the forging material, the length tolerance related to the geometric dimensions, and the quantity and layer related to the furnace loading method were unified into standard units; the forging material, geometric dimensions, and furnace loading method, after integrity checks and unit unification, were merged into the same data structure and marked to obtain the forging information data.

[0052] Each hardness requirement and residual stress requirement was verified to provide numerical ranges and control location descriptions; the hardness scale and numerical range expressions related to hardness requirements and the stress unit expressions related to residual stress requirements were unified into standard unit expressions; the hardness requirements and residual stress requirements after integrity checks and unit unification were merged into the same data structure to obtain performance requirement data.

[0053] The forging information data and performance requirement data are linked one by one and encapsulated into a unified data structure to generate a process task data package.

[0054] S2: Based on the process task data package, construct a temperature control performance prediction model and perform reverse solving to generate a core temperature control reference curve.

[0055] S2.1: Parse the fields of the process task data package and decompose it into a forging information data set and a performance requirement data set through classification operations;

[0056] It should be noted that, based on the process task data package, the forging material data, geometric dimension data, furnace loading method data, hardness requirement data, and residual stress requirement data in each forging heat treatment process task parameter record are read; the forging material data, geometric dimension data, and furnace loading method data are marked as forging information related fields, and the hardness requirement data and residual stress requirement data are marked as performance requirement related fields.

[0057] The data corresponding to all forging information fields are summarized and arranged sequentially into the same data structure to form a forging information data set; the data corresponding to all performance requirement fields are summarized and arranged sequentially into another data structure to form a performance requirement data set.

[0058] S2.2: Based on the forging information data set, use the temperature control feature extraction algorithm to extract and map attributes related to temperature and time, and generate a set of temperature control variables;

[0059] It should be noted that forging material data, geometric dimension data, and furnace loading method data are sequentially read from the forging information data set. Temperature-related attributes and time-related attributes are identified according to field names and their meanings in the process task data package. Attributes reflecting heating temperature level, holding interval temperature, and cooling stage temperature are organized into temperature-related attributes, and attributes reflecting heating time, holding time, and cooling time are organized into time-related attributes. For each temperature-related attribute, the corresponding upper and lower temperature limits are recorded, and these upper and lower temperature limits are used as the temperature range. For each time-related attribute, the corresponding start and end times or durations are recorded, and these start and end times or durations are used as the control time range or duration. Attributes that do not reflect temperature and time changes are removed during the temperature control feature extraction process.

[0060] According to the method of one-to-one correspondence between temperature-related attributes and time-related attributes, each group of temperature-related attributes and its corresponding time-related attributes are combined into a group of temperature control variables. All temperature control variables are summarized, organized and recorded as a set of temperature control variables in the order of combination.

[0061] Furthermore, the temperature control feature extraction algorithm identifies and extracts attributes directly related to temperature and time from the forging information data set, and pairs temperature-related attributes and time-related attributes into a set of temperature control variables; it also eliminates redundant information unrelated to temperature control, so that the variables used by the temperature control performance mapping structure can centrally reflect the key temperature control features of the heating, holding and cooling processes.

[0062] S2.3: Based on the performance requirement data set and the temperature control variable set, generate a temperature control performance mapping structure through constraint-driven mapping;

[0063] It should be noted that all hardness requirement data and residual stress requirement data are read from the performance requirement data set and organized into performance constraint records; the temperature control variable records in the temperature control variable set are read one by one in the generation order, and the temperature-related attributes and time-related attributes contained in each temperature control variable record are compared with the performance constraint records.

[0064] By comparing the relationship between the temperature range corresponding to the temperature-related attribute and the target hardness range in the hardness requirement data, for example, when the temperature range corresponding to the temperature-related attribute is 820℃~860℃ and the target hardness range in the hardness requirement data is 30HRC~35HRC, the temperature range of 820℃~860℃ and the target hardness range of 30HRC~35HRC are recorded as a set of matched temperature-hardness constraints. Similarly, by comparing the relationship between the duration corresponding to the time-related attribute and the control time range in the residual stress requirement data, for example, when the duration corresponding to the time-related attribute is 90 minutes and the control time range in the residual stress requirement data is 60 minutes~120 minutes, the duration of 90 minutes and the control time range of 60 minutes~120 minutes are recorded as a set of matched time-residual stress constraints. Matching temperature-hardness constraints and time-residual stress constraints are determined for each temperature control variable record, and the temperature control variable record, hardness constraint, and residual stress constraint are combined into a single mapping record.

[0065] Following the reading order of temperature control variable records, all mapping records are summarized and the performance requirement data set and constraints are organized to generate a temperature control performance mapping structure.

[0066] S2.4: Based on the process task data package, calculate the predicted values ​​of hardness and residual stress, and adjust the parameters in the temperature control performance mapping structure to obtain the temperature control performance prediction model;

[0067] It should be noted that the hardness requirement data and residual stress requirement data are read from the performance requirement data set in the process task data package; and the hardness prediction value and residual stress prediction value associated with each temperature control variable record are calculated according to the correspondence between each temperature control variable record in the temperature control variable set and the mapping record in the temperature control performance mapping structure.

[0068] The expression for calculating the predicted hardness value is:

[0069] ;

[0070] in, This is the predicted hardness value. Upper limit of the target hardness range and the lower limit of the target hardness range The target hardness center value is obtained by averaging. This is the hardness-temperature effect coefficient. This is the hardness-duration effect coefficient. For the upper limit of the temperature range and lower limit of temperature range The average temperature obtained after averaging For duration.

[0071] The hardness-temperature influence coefficient is obtained through linear regression of temperature-related properties and hardness measurement results during the process verification stage. An exemplary value range is [-1, 1], which is used to convert temperature into hardness dimensions.

[0072] The hardness-duration influence coefficient is obtained by performing linear regression on the duration-related attributes and hardness measurement results, and requires that the predicted change in hardness under different duration conditions remain within the target hardness range. An exemplary value range is [-1, 1], which is used to convert the duration into hardness units.

[0073] The expression for calculating the predicted residual stress is:

[0074] ;

[0075] in, This is the predicted value of residual stress. It is determined by the upper limit of the allowable residual stress range and the lower limit of the allowable residual stress range The target residual stress center value is obtained by averaging. This is the temperature-residual stress influence coefficient. The duration-residual stress influence coefficient.

[0076] The temperature-residual stress influence coefficient is obtained by linear regression of temperature-related properties and residual stress measurement results, with an exemplary range of [-1, 1], used to convert temperature difference into stress dimensions.

[0077] The duration-residual stress influence coefficient is obtained by linear regression of duration-related properties and residual stress measurements, with an exemplary range of [-1, 1], used to convert the time difference into stress dimensions.

[0078] The difference between the predicted hardness value and the hardness requirement data in the performance requirement dataset is taken as the hardness deviation; the difference between the predicted residual stress value and the residual stress requirement data in the performance requirement dataset is taken as the residual stress deviation; based on the sign of the hardness deviation, it is determined whether the parameters corresponding to temperature-related attributes in the temperature control performance mapping structure need to be increased or decreased, and the parameters corresponding to temperature-related attributes are adjusted according to the absolute value of the hardness deviation; based on the sign of the residual stress deviation, it is determined whether the parameters corresponding to time-related attributes in the temperature control performance mapping structure need to be extended or shortened, and the parameters corresponding to time-related attributes are adjusted according to the absolute value of the residual stress deviation. The adjusted temperature control performance mapping structure is then fixed to generate the temperature control performance prediction model.

[0079] A superior approach involves calculating the predicted hardness and residual stress values ​​line by line using the performance requirement data set in the process task data package. The hardness deviation is then used to adjust the parameters corresponding to temperature-related attributes in the temperature control performance mapping structure, and the residual stress deviation is used to adjust the parameters corresponding to time-related attributes in the same structure. This achieves closed-loop correction of the temperature control performance prediction model parameters around the target hardness range and the allowable residual stress range. This improves the consistency between the core temperature control reference curve obtained from subsequent inverse solving and the performance requirement data, thereby enhancing the accuracy of temperature control curve optimization and the efficiency of forging heat treatment process development.

[0080] S2.5: Based on the temperature control performance prediction model, the target set is obtained by reconstructing the target performance;

[0081] It should be noted that all hardness requirement data and residual stress requirement data are reorganized, and each set of hardness requirement data and residual stress requirement data is combined to form a target performance record. These are then summarized into a target performance record sequence according to the reorganization order. A set of temperature control variable records is selected from the temperature control variable set as a reference temperature control variable record sequence. The reference temperature control variable record sequence is input into the temperature control performance prediction model one by one to obtain the hardness prediction value and residual stress prediction value corresponding to each reference temperature control variable record. The target performance record, the corresponding hardness deviation, the corresponding residual stress deviation, and the reference temperature control variable record associated with the target performance record are combined to form a reverse solution target record. All reverse solution target records are summarized and organized according to the arrangement order of the target performance record sequence to obtain the reverse solution target set.

[0082] S2.6: Search for temperature control variables in the inverse solution target set to generate an initial temperature control variable sequence, and perform performance deviation direction analysis to generate a temperature control variable correction direction sequence;

[0083] It should be noted that the reference temperature control variable records associated with each inverse solution target record in the inverse solution target set are recorded within the temperature range and time range given by the temperature control variable set. By comparing the deviations of the hardness prediction value and the residual stress prediction value with the hardness requirement data and residual stress requirement data of the target performance record in the inverse solution target set, the temperature control variable records with gradually decreasing deviations are selected and arranged in chronological order to generate the initial temperature control variable sequence.

[0084] Then, based on the hardness deviation and residual stress deviation recorded in the target set obtained by reverse solving, determine whether the temperature control variable records in the initial temperature control variable sequence are adjusted in the direction of increasing or decreasing in terms of temperature-related attributes; and whether they are adjusted in the direction of extending or shortening in terms of time-related attributes. The adjustment directions obtained by continuous judgment are sorted and recorded in order to generate a temperature control variable correction direction sequence.

[0085] S2.7: Perform temperature time axis reconstruction operation on the temperature control variable correction direction sequence and the inverse solution of the initial temperature control variable sequence to obtain the core temperature control reference curve.

[0086] It should be noted that, according to the temperature control variable correction direction sequence, the temperature-related attributes and time-related attributes in the initial temperature control variable sequence obtained through reverse solving are adjusted one by one by increasing or decreasing, and extending or shortening. For example, when the temperature-related attribute of the temperature control variable record in the initial temperature control variable sequence obtained through reverse solving is 850℃ and the time-related attribute is 60 minutes, and the temperature adjustment direction given by the temperature control variable correction direction sequence for this temperature control variable record is increasing and the time adjustment direction is extending, then the temperature-related attribute is adjusted from 850℃ to 860℃ and the time-related attribute is adjusted from 60 minutes to 70 minutes. Each adjusted temperature control variable record is mapped onto a unified time axis in chronological order, and continuously spliced ​​to generate a continuous temperature-time trajectory describing the relationship between the core temperature of the forging and time. This continuous temperature-time trajectory is used as the core temperature control reference curve.

[0087] S3: Perform thermal response mapping on the core temperature control reference curve to obtain the furnace temperature control trajectory, and obtain an online core tracking control scheme through heat treatment risk timing assessment.

[0088] S3.1: Based on the core temperature control reference curve, the furnace temperature control trajectory is generated by converting it into a time-varying target temperature sequence of the furnace through a thermal response mapping algorithm.

[0089] It should be noted that, based on the core temperature control reference curve, the core temperature values ​​are read sequentially from the core temperature control reference curve at fixed time intervals and arranged into a core temperature time series; at the same time, the forging geometric dimension data and furnace loading method data are read from the forging information data set in the process task data package; based on the forging geometric dimension data and furnace loading method data, the core temperature time series is used as the target temperature response, and the time of occurrence of the furnace temperature change corresponding to each target core temperature is obtained; and the time difference between the time of occurrence of the furnace temperature change and the time of the target core temperature is used as the advance amount of the furnace temperature to take effect.

[0090] In the furnace loading method data, the heating position closest to the heat source is selected as the reference heating position, and the ratio of the distance between other heating positions and the reference heating position is used as the furnace temperature correction coefficient. In the core temperature time series, the rise value of the reference furnace temperature is read at each time point, and the product of the rise value of the reference furnace temperature and the furnace temperature correction coefficient is used as the furnace temperature rise amplitude. The core temperature value at each time point is added to the corresponding furnace temperature rise amplitude and shifted on the time axis according to the time advance to generate the furnace target temperature value. All furnace target temperature values ​​are arranged in chronological order to obtain the furnace target temperature sequence, which is used as the furnace temperature control trajectory.

[0091] Furthermore, the thermal response mapping algorithm combines the core temperature control reference curve with the forging geometry data and furnace loading method data to convert the heating advance and temperature rise required for each core target temperature time point into a furnace target temperature time series, thereby generating a furnace temperature control trajectory that corresponds one-to-one with the core temperature control reference curve.

[0092] The superior thermal response mapping algorithm enables temperature control to move beyond simply relying on empirically set furnace temperature curves. Instead, it incorporates feedforward compensation and dynamic adjustments based on the core temperature requirements, improving the matching degree between the furnace temperature control trajectory and the actual thermal response of the core. This reduces hardness deviations and residual stress anomalies caused by core temperature lag or overshoot, providing a more reasonable target furnace temperature basis for online core tracking control schemes.

[0093] S3.2: Combining the furnace temperature control trajectory and process task data package, calculate the heat treatment risk level through the heat treatment risk time series assessment algorithm and generate the heat treatment risk time series curve;

[0094] It should be noted that, based on the furnace temperature control trajectory, the target furnace temperature time series is read sequentially, and the forging geometric dimension data, hardness requirement data, and residual stress requirement data are read from the forging information data set and performance requirement data set in the process task data package. Based on the target furnace temperature time series and forging geometric dimension data, and using the hardness requirement data and residual stress requirement data as calculation constraints, a heat treatment risk level value is calculated for each time point, reflecting the combined level of hardness failure risk (the risk that the predicted hardness value deviates from the hardness requirement data, resulting in the forging hardness performance failing to meet the service requirements) and residual stress exceeding the limit risk (the risk that the predicted residual stress value exceeds the residual stress requirement data, resulting in the forging cracking, deformation, or reduced service life). All heat treatment risk level values ​​are arranged in chronological order, and the time series formed by the continuous arrangement is used as the heat treatment risk time series curve.

[0095] The expression for calculating the numerical level of heat treatment risk is:

[0096] ;

[0097] in, This represents the numerical value for the risk level of heat treatment.

[0098] Furthermore, the heat treatment risk time series assessment algorithm combines the furnace temperature control trajectory with the forging geometric data, hardness requirement data, and residual stress requirement data to convert the complex temperature time history into a heat treatment risk level signal that changes over time, thereby providing a comprehensive characterization of hardness failure risk and residual stress exceedance risk at each time point.

[0099] Ideally, a heat treatment risk timing assessment algorithm can automatically identify time periods with concentrated risks throughout the entire heat treatment process, distinguishing between relatively safe and risk-sensitive sections. This provides a quantitative basis for selecting and adjusting control variables in the online core tracking control scheme. Comparative analysis of the risk timing assessment results from multiple heat treatment cycles can also help identify long-standing process weaknesses, reducing the number of trial-and-error adjustments based on experience and improving the stability, controllability, and traceability of the forging heat treatment process.

[0100] S3.3: Based on the heat treatment risk time series curve and furnace temperature control trajectory, the combination of adjustment variables is determined by the online control strategy synthesis method to generate an online core tracking control scheme.

[0101] It should be noted that the heat treatment risk level value at each time point is read from the heat treatment risk time series curve in chronological order and paired one-to-one with the furnace target temperature value at the same time point in the furnace temperature control trajectory to form a pairing sequence of heat treatment risk level value and furnace target temperature value. The pairing sequence of heat treatment risk level value and furnace target temperature value is traversed along the time sequence. By comparing the magnitude of the heat treatment risk level value at each time point with the heat treatment risk level value at the previous time point, it is determined whether the heat treatment risk level value is in an increasing state, a decreasing state, or a basically stable state.

[0102] When the heat treatment risk level is rising, reduce the furnace target temperature and decrease the furnace target temperature correction, while simultaneously reducing the heating conditions and shortening the core temperature monitoring interval. When the heat treatment risk level is falling, maintain the furnace target temperature or slightly increase the furnace target temperature correction, while simultaneously maintaining the heating conditions and extending the core temperature monitoring interval. When the heat treatment risk level is basically stable, maintain the furnace target temperature correction at the same rate, while simultaneously maintaining the heating conditions and keeping the core temperature monitoring interval constant.

[0103] The furnace target temperature correction, heating condition adjustment mode, and core temperature monitoring rhythm at each time point are recorded as a combination of adjustment variables. All combinations of adjustment variables are arranged in chronological order to obtain the online core tracking control scheme.

[0104] Furthermore, the online control strategy synthesis method refers to analyzing the changes in the heat treatment risk level point by point in time sequence after obtaining the furnace temperature control trajectory and the heat treatment risk time-series curve. At each time point, a combination of adjustment variables is selected collaboratively from the furnace target temperature correction, heating condition adjustment mode, and core temperature monitoring rhythm to ensure that the core temperature control reference curve is consistent with the performance requirement data and to suppress the excessively rapid increase in the heat treatment risk level. By sequentially splicing the combination of adjustment variables obtained from continuous time points, an online core tracking control scheme covering the entire heat treatment process is formed. This makes the furnace temperature control process more stable, safer, and easier for subsequent process optimization while meeting hardness and residual stress requirements.

[0105] S4: Execute the online core tracking control scheme, synchronously record process data, and obtain the execution temperature control curve by combining the furnace temperature control trajectory and thermal history inversion calculation.

[0106] S4.1: Based on the online core tracking control scheme, the furnace temperature is adjusted and the furnace temperature, surface temperature and core temperature are recorded simultaneously to generate a process data sequence;

[0107] It should be noted that, based on the online core tracking control scheme, at the start of heat treatment, each temperature control record is read sequentially according to the time sequence recorded in the online core tracking control scheme. The furnace target temperature value, heating condition adjustment mode, and core temperature monitoring rhythm in each temperature control record are used as furnace temperature adjustment commands and temperature measurement commands, respectively. The furnace temperature adjustment commands and temperature measurement commands are executed, for example, by adjusting the output power, adjusting the operating status of the circulating fan, or controlling the opening and closing of the spray cooling to make the furnace temperature follow the change of the furnace target temperature value. At the core temperature monitoring time point corresponding to each temperature control record, a temperature measurement action is triggered, and the furnace temperature, surface temperature, and core temperature are recorded simultaneously.

[0108] After each temperature measurement is completed, the current time, furnace temperature, surface temperature, and core temperature are combined and organized into a process record; all process records are summarized and organized to generate a process data sequence.

[0109] S4.2: Compare the cardiac temperature control reference curve with the process data sequence, and obtain the actual cardiac temperature control curve through thermal history inversion calculation;

[0110] It should be noted that the core temperature control reference curve is read and the core temperature control reference curve is resampled according to the time interval in the process data sequence to obtain the reference core temperature value at the corresponding time point of each process record; each process record in the process data sequence is read in chronological order, and the current time, furnace temperature and surface temperature in each process record are combined with the reference core temperature value at the same time point to form a comparison record;

[0111] Between two adjacent measurement time points corresponding to each control record, the time between the start time point and the end time point is divided into equally spaced small time steps. At each small time step, the distance parameter corresponding to the thickness position of the core is selected. The ratio of the difference between the furnace temperature reading and the surface temperature reading at the start time point and the end time point to the distance parameter is taken as the core temperature value. The core temperature values ​​of consecutive small time steps are connected in chronological order to obtain the core temperature change curve.

[0112] All the core temperature change curves obtained in adjacent measurement time periods are spliced ​​together in chronological order to form a continuous core temperature time curve covering the entire heat treatment process. This continuous core temperature time curve is used as the actual core temperature control curve.

[0113] S4.3: Combine the actual core temperature control curve with the temperature control trajectory of the furnace temperature control trajectory to obtain the execution temperature control curve.

[0114] It should be noted that, based on the actual core temperature control curve and the furnace temperature control trajectory, the core temperature time series in the actual core temperature control curve is aligned with the furnace target temperature time series in the furnace temperature control trajectory. At each time step, the core temperature value is extracted from the actual core temperature control curve, and the furnace target temperature value is extracted from the furnace temperature control trajectory. The core temperature value and the furnace target temperature value are paired one-to-one to form a temperature comparison record containing time, furnace target temperature value, and actual core temperature value. All temperature comparison records are arranged continuously in chronological order to obtain the execution temperature control curve.

[0115] S5: Based on the temperature control curve, optimize the temperature control performance prediction model and obtain the process quality assessment results through performance deviation evaluation.

[0116] S5.1: Based on the execution temperature control curve, a set of performance prediction timing sequences is generated by replaying the temperature time history of the execution temperature control curve;

[0117] It should be noted that, in the execution of the temperature control curve, the time marker and temperature value corresponding to each time point are read sequentially according to the time sequence. The temperature value of each time point is combined with the forging information data in the process task data package to form an input data record. Based on the input data record, the hardness prediction value and residual stress prediction value corresponding to the current time marker are read from the temperature control performance prediction model, and the current time marker, hardness prediction value, and residual stress prediction value are organized into a performance prediction record. According to the time sequence of the execution of the temperature control curve, all performance prediction records are obtained sequentially and appended sequentially. The performance prediction records are completed at all time points in the entire heat treatment process to generate a performance prediction time sequence set.

[0118] S5.2: Based on the performance prediction time series set and performance requirement data, construct a two-dimensional performance deviation distribution of time indicators and extract the deviation peak and deviation accumulation to generate a performance deviation field;

[0119] It should be noted that each performance prediction record is read sequentially from the performance prediction time series set in chronological order. In each performance prediction record, the time stamp, hardness prediction value, and residual stress prediction value are obtained. These are then compared with the hardness requirement data and residual stress requirement data in the performance requirement data to obtain the hardness deviation and residual stress deviation corresponding to the current time stamp. Using the time stamp as the time index and the hardness deviation and residual stress deviation as the performance index, the hardness deviation and residual stress deviation corresponding to each time stamp are filled into the intersection of time and performance index to form a two-dimensional performance deviation distribution of time index.

[0120] Traverse the two-dimensional performance deviation distribution of time indexes, record the maximum deviation amplitude of hardness deviation and residual stress deviation as the deviation peak value; sum the deviation peak value of hardness deviation and residual stress deviation at all time positions as the deviation accumulation; unify and organize the two-dimensional performance deviation distribution of time indexes, deviation peak value and deviation accumulation to generate a performance deviation field.

[0121] S5.3: Based on the performance deviation field, perform self-consistent write-back optimization on the parameters of the temperature control performance prediction model to generate an optimized temperature control performance prediction model;

[0122] It should be noted that the hardness deviation and residual stress deviation corresponding to each time position are read sequentially from the performance deviation field in chronological order, and each time position is matched one-to-one with the temperature-related attributes and time-related attributes in the temperature control variable set used in the process of constructing the temperature control performance prediction model; the positive and negative directions of the hardness deviation and residual stress deviation are used as the direction of increasing or decreasing the parameters corresponding to the temperature-related attributes and time-related attributes.

[0123] Using the peak deviation and accumulated deviation in the performance deviation field as references for adjustment intensity, at each time point, the parameters corresponding to temperature-related attributes in the temperature control performance prediction model are adjusted by increasing or decreasing according to the magnitude of hardness deviation and residual stress deviation. For example, when the performance deviation field shows that the target hardness is too low, the temperature-related attribute parameter of the corresponding temperature platform in the temperature control performance prediction model is adjusted from 600 degrees Celsius to 620 degrees Celsius. The parameters corresponding to time-related attributes in the temperature control performance prediction model are adjusted by extending or shortening. For example, when the performance deviation field shows that the residual stress control is insufficient, the time-related attribute parameter of the corresponding heat preservation stage in the temperature control performance prediction model is adjusted from 30 minutes to 40 minutes. Based on the executed temperature control curve, process task data package, and temperature control performance prediction model, a new performance prediction time series set is generated and the performance deviation field is updated to obtain the optimized temperature control performance prediction model.

[0124] S5.4: Based on the optimized temperature control performance prediction model, the process quality assessment results are generated by comprehensively quantifying and classifying the performance achievement degree, deviation distribution characteristics and model convergence state.

[0125] It should be noted that, based on the optimized temperature control performance prediction model, the predicted hardness and residual stress values ​​are read line by line from the performance prediction time series set and compared with the hardness and residual stress requirement data in the performance requirement data to calculate the performance achievement degree. The two-dimensional performance deviation distribution, peak deviation, and accumulated deviation are read from the performance deviation field, and the model convergence state and deviation distribution characteristics are recorded in conjunction with the changing trend of the performance deviation field. Several process quality level intervals are divided according to the performance achievement degree. For example, a performance achievement degree higher than 0.90 can be classified as the first level, a performance achievement degree between 0.80 and 0.90 as the second level, and a performance achievement degree lower than 0.80 as the third level. After completing the quality level division, the quality level, performance achievement degree, and deviation distribution characteristics are uniformly organized to generate the process quality assessment results.

[0126] The expression for calculating performance achievement is:

[0127] ;

[0128] in, To achieve performance targets, As the normalization factor, This represents the total number of time points in the performance prediction time series set. For at a certain point in time The numerical value of the risk level of heat treatment.

[0129] A superior approach involves calculating the performance achievement degree based on an optimized temperature control performance prediction model. This is combined with time-based indicators such as two-dimensional performance deviation distribution, peak deviation, accumulated deviation, and model convergence status. This allows for the unified quantification and grading of the differences between the predicted hardness and residual stress values ​​at each time point in the performance prediction time series and the performance requirement data. Consequently, a single process quality assessment result simultaneously reflects the performance achievement degree, deviation distribution characteristics, and heat treatment process stability. This expands process quality judgment from a single final inspection result to a comprehensive evaluation covering the entire execution temperature control curve. This facilitates rapid identification of problematic time segments and sensitive temperature control variables, improving the targeted nature and traceability of forging heat treatment process optimization.

[0130] In summary, this invention achieves the generation and optimization of temperature control curves based on target performance constraints by constructing a temperature control performance prediction model and performing inverse solving and feedback optimization, thereby improving the computational consistency of process parameter design and the quantifiable and iterative optimization capabilities of result evaluation.

[0131] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for optimizing the heat treatment process of forgings based on temperature control curves, characterized in that: include, Collect the process parameters for the heat treatment of forgings and obtain the process task data package through preprocessing; Based on the process task data package, a temperature control performance prediction model is constructed and inversely solved to generate a core temperature control reference curve. The specific steps are as follows. Based on the temperature control performance prediction model, the target set is obtained by reconstructing the target performance and solving it in reverse. The target set is solved in reverse to search for temperature control variables, generating an initial sequence of temperature control variables. Then, the performance deviation direction is analyzed to generate a sequence of temperature control variable correction directions. The temperature control variable correction direction sequence and the initial temperature control variable sequence are solved in reverse. The temperature time axis reconstruction operation is performed to obtain the core temperature control reference curve. Thermal response mapping was performed on the core temperature control reference curve to obtain the furnace temperature control trajectory. An online core tracking control scheme was then obtained through a heat treatment risk time series assessment. The specific steps are as follows: Based on the core temperature control reference curve, the furnace temperature control trajectory is generated by converting it into a time-varying target temperature sequence of the furnace through a thermal response mapping algorithm. By combining the furnace temperature control trajectory and process task data package, the heat treatment risk level is calculated through the heat treatment risk time series assessment algorithm, and a heat treatment risk time series curve is generated. Based on the heat treatment risk time series curve and furnace temperature control trajectory, the combination of adjustment variables is determined by the online control strategy synthesis method to generate an online core tracking control scheme. An online core tracking control scheme is implemented, process data is recorded synchronously, and the execution temperature control curve is obtained through thermal history inversion calculation combined with the furnace temperature control trajectory. The specific steps are as follows. Based on the online core tracking control scheme, the furnace temperature is adjusted and the furnace temperature, surface temperature and core temperature are recorded simultaneously to generate a process data sequence; By comparing the cardiac temperature control reference curve with the process data sequence, and by performing thermal history inversion calculations, the actual cardiac temperature control curve is obtained. The actual temperature control curve of the core is combined with the temperature control trajectory of the furnace temperature control trajectory to obtain the execution temperature control curve; Based on the temperature control curve, the temperature control performance prediction model is optimized, and the process quality assessment results are obtained through performance deviation evaluation. The specific steps are as follows. Based on the execution temperature control curve, a set of performance prediction time series is generated by replaying the temperature time history of the execution temperature control curve; Based on the performance prediction time series set and performance requirement data, a two-dimensional performance deviation distribution of time indicators is constructed and the deviation peak and deviation accumulation are extracted to generate a performance deviation field. Based on the performance deviation field, the parameters of the temperature control performance prediction model are optimized by self-consistent write-back to generate the optimized temperature control performance prediction model. Based on the optimized temperature control performance prediction model, the process quality assessment results are generated by comprehensively quantifying and classifying the performance achievement, deviation distribution characteristics, and model convergence status.

2. The method for optimizing the heat treatment process of forgings based on temperature control curves as described in claim 1, characterized in that: The heat treatment process parameters for the forgings include the forging material, geometric dimensions, furnace loading method, hardness requirements, and residual stress requirements. The preprocessing includes integrity checks, unit unification, and structured organization.

3. The method for optimizing the heat treatment process of forgings based on temperature control curves as described in claim 2, characterized in that: The process task data package includes forging information data and performance requirement data.

4. The method for optimizing the heat treatment process of forgings based on temperature control curves as described in claim 3, characterized in that: The specific steps for constructing a temperature control performance prediction model based on process task data are as follows. The process task data package is parsed and then broken down into forging information data set and performance requirement data set through classification operations; Based on the forging information dataset, a temperature control feature extraction algorithm is used to extract and map attributes related to temperature and time to generate a set of temperature control variables. Based on the performance requirement data set and the temperature control variable set, a temperature control performance mapping structure is generated through constraint-driven mapping. The process of generating a temperature control performance mapping structure through constraint-driven mapping refers to comparing the relationship between the temperature range corresponding to the temperature-related attributes and the target hardness range in the hardness requirement data; comparing the relationship between the duration corresponding to the time-related attributes and the control time range in the residual stress requirement data; determining matching temperature, hardness constraints, time, and residual stress constraints for each temperature control variable record; and combining the temperature control variable record, hardness constraints, and residual stress constraints into a mapping record. According to the reading order of the temperature control variable records, all mapping records are summarized and the performance requirement data set and constraints are organized to generate the temperature control performance mapping structure; Based on the process task data package, the predicted values ​​of hardness and residual stress are calculated, and the parameters in the temperature control performance mapping structure are adjusted to obtain the temperature control performance prediction model.

Images