Digital twin driven alloy hot working control method and system

Abstract

This application relates to the field of alloy hot working control technology, and discloses a digital twin-driven alloy hot working control method and system. The method includes: acquiring the hot working process data and the end state data of the current pass of the alloy billet; generating the microstructure evolution result of the current pass through a digital twin microstructure evolution model; comparing the microstructure state quantity with the microstructure judgment boundary in the target microstructure condition corresponding to the current pass item by item to determine the microstructure offset result of the current pass; generating the microstructure inheritance control constraint of the next pass based on the microstructure evolution result and the microstructure offset result of the current pass; screening and combining the candidate control parameters of the next pass according to the microstructure inheritance control constraint of the next pass, generating the control parameter result of the next pass, and outputting it to the hot working execution equipment to execute the next hot working; this method can improve the continuity of multi-pass hot working microstructure inheritance control and the consistency of product microstructure.

Description

Technical Field

[0001] This application relates to the field of alloy hot working control technology, and more specifically, to a digital twin-driven alloy hot working control method and system. Background Technology

[0002] The hot working process of alloy billets is usually completed in multiple consecutive passes. In each pass, the billet undergoes heating, deformation, holding, cooling, or transfer. Each pass not only changes the size and shape of the billet, but also alters its internal recrystallization state, grain size, deformation accumulation state, residual stress state, and microstructure distribution between different regions. These microstructures are not formed independently in a single pass, but rather accumulate gradually over multiple processing passes and continuously affect the machinability of subsequent passes.

[0003] With the application of digital twin technology in thermal processing control, existing methods can predict the forming state, equipment load state, or final quality trend of the current pass based on temperature, load, displacement, speed, and dimensional change data during the thermal processing, and adjust the process parameters of the current pass or subsequent passes accordingly. However, existing control methods usually focus more on whether the temperature, load, size, or equipment execution parameters are within the allowable range of the process, and lack the processing to carry over the control parameters of the next pass to the actual microstructure evolution state formed in the current pass.

[0004] Specifically, in multi-pass hot processing, the microstructure formed after the current pass may have deviated from the target microstructure conditions corresponding to that pass. This deviation may manifest as insufficient local recrystallization, grain size exceeding the allowable range, insufficient deformation accumulation, or uneven distribution of residual stress. If subsequent passes only generate the starting temperature, deformation amount, deformation rate, heat preservation and cooling conditions, or equipment execution sequence according to the pre-set basic process plan, without transforming the microstructure evolution results and microstructure deviation of the current pass into control constraints for the next pass, then subsequent passes cannot inherit and correct the microstructure deviation formed in the previous pass.

[0005] For example, if the dimensions of a certain alloy billet are within the allowable range after the current pass, but the local calculated area has insufficient recrystallization or differences in microstructure distribution due to temperature drop, insufficient deformation distribution, or abnormal load response, if the next pass is still performed according to the original reduction, deformation speed, and heat preservation and cooling conditions, even if the equipment load and dimensional changes continue to be within the allowable range, the microstructure shift formed in the previous pass may continue to accumulate, resulting in a decrease in the consistency of microstructure and mechanical properties in different areas of the final product.

[0006] Therefore, the problem with the existing technology is that in the multi-pass alloy hot working process, the digital twin control results usually remain at the level of current pass state prediction or parameter adjustment, and fail to convert the actual microstructure evolution results and microstructure offset results formed in the current pass into the constraint basis for the generation of parameters in the next pass. This results in the discontinuity of the microstructure control chain between adjacent passes, making it difficult to achieve the microstructure state continuity control throughout the entire multi-pass hot working process. Summary of the Invention

[0007] To overcome the problem in existing multi-pass alloy hot working processes that fail to convert the actual microstructure evolution results of the current pass into constraints for generating parameters for the next pass, leading to discontinuities in the microstructure control chain between adjacent passes, this invention performs microstructure evolution modeling, target microstructure condition comparison, microstructure offset identification, generation of microstructure inheritance control constraints for the next pass, and candidate control parameter screening and combination on the hot working process data and end state data of the current pass of the alloy billet. This forms a processing chain that continuously inherits the microstructure state of the current pass from the control parameters of the next pass, thereby improving the continuity of microstructure state control, the accuracy of parameter generation, and the consistency of the final product microstructure in multi-pass alloy hot working processes. To achieve the above objectives, this invention provides the following technical solution: A digital twin-driven method for controlling alloy hot working includes: Acquire the hot working process data and the end status data of the current pass of the alloy billet; Based on the heat treatment process data and the current pass end status data, the current pass organization evolution result is generated using a digital twin organization evolution model; Compare the organization state quantities in the current track organization evolution result with the organization determination boundaries in the target organization conditions corresponding to the current track item by item to determine the organization offset result of the current track. Based on the current course organization evolution result and the current course organization offset result, the next course organization succession control constraint is generated; Based on the next-stage organizational acceptance control constraints, the candidate control parameters for the next stage are screened and combined to generate the control parameter results for the next stage. The control parameter results for the next pass are output to the heat processing execution device to execute the next heat processing pass. After the next pass is completed, the next pass is used as the new current pass to continue processing until the multi-pass heat processing process is completed.

[0008] Furthermore, methods for obtaining hot processing process data and current pass completion status data include: Temperature process data, deformation execution data, and load response data are collected between the start and end times of the current pass, and the settling time is determined based on the time of change of stroke position or the time of change of equipment control command in the deformation execution data. The temperature data between adjacent settling times is determined as the representative temperature value, the load data between adjacent settling times is determined as the representative load value, and the change in stroke position, change in compression, or change in speed between adjacent settling times is determined as the deformation execution amount. After the current pass is completed, the end status data of the current pass is collected, and a correspondence is established between the end status data of the current pass and the hot working process data according to the same alloy billet and the same current pass.

[0009] Furthermore, methods for generating current track evolution results using digital twin organizational evolution models include: Based on the geometric description of the alloy billet and the deformation contact area of ​​the current pass, the alloy billet is divided into several calculation zones, and the representative temperature value, representative load value, deformation execution amount and deformation speed corresponding to each calculation zone are determined. According to the time period sequence, the temperature representative value, deformation execution amount, deformation speed and load representative value corresponding to each calculation partition are converted into segmented thermal deformation state, and the cumulative thermal deformation state is determined based on the segmented thermal deformation state of each time period. By calculating the organizational state quantities in the digital twin organizational evolution model, the cumulative thermal deformation state is converted into the organizational state quantities corresponding to each calculation partition, thus forming the organizational evolution result for the current pass.

[0010] Furthermore, methods for determining the current track shift organization result include: Based on the alloy material type, target product microstructure requirements, multi-pass hot working process plan, and current pass sequence position, determine the target microstructure conditions corresponding to the current pass, and determine the microstructure judgment boundary for each microstructure state quantity in the microstructure evolution result of the current pass. When the organization's determination boundary is within the allowable value range, the organization's state quantity is compared with the lower and upper boundaries of the allowable value range to determine the individual comparison results for items that are below the boundary, above the boundary, or have not formed an offset. When the organizational judgment boundary is within the allowable level range, the organizational state quantity is converted into the corresponding level and compared with the allowable level range to determine the individual comparison result of insufficient level, excessive level, or no offset. Based on the comparison results of each item, generate the current track organization offset result, including the organization offset direction and the organization offset degree.

[0011] Furthermore, methods for generating the next level of organizational control constraints include: Read the organization status quantity in the organization evolution result of the current track, and read the type of offset organization status quantity, offset calculation partition, organization offset direction and organization offset degree in the organization offset result of the current track; Based on the preset organizational succession relationship, find the succession relationship item corresponding to the organizational offset result of the current pass. The succession relationship item includes organizational state quantity type, organizational offset direction, organizational offset degree range, next pass constraint item, constraint adjustment direction, and constraint adjustment magnitude formation method. Read the next basic control boundary, and modify the next basic control boundary according to the next constraint item, constraint adjustment direction and constraint adjustment range formation method in the inheritance relationship item, and generate the next organizational inheritance control constraint.

[0012] Furthermore, methods for generating the control parameter results for the next pass include: Based on the next heat treatment process plan and the executable range of the heat treatment equipment, multiple candidate control parameter groups for the next process are generated. Each candidate control parameter group for the next process includes at least one of the following: candidate starting temperature, candidate deformation amount, candidate deformation rate, candidate heat preservation and cooling conditions, and candidate equipment execution sequence. Compare each parameter item in the next candidate control parameter group with the corresponding final constraint boundary in the next organizational acceptance control constraint, and eliminate any next candidate control parameter group whose parameter item is not within the corresponding final constraint boundary. Among the retained candidate control parameters for the next pass, the control parameter results for the next pass are determined based on whether they meet the final constraint boundary generated by the current pass's organization offset results, the equipment execution margin, and the degree of deviation from the next pass's thermal processing plan.

[0013] Furthermore, the method for continuing processing the next pass as a new current pass includes: During the next execution, collect the heat treatment process data of the next step, and after the next execution is completed, collect the end status data of the next step. The hot working process data of the next pass is used as the hot working process data of the new current pass, the end state data of the next pass is used as the end state data of the new current pass, the state of the alloy billet after the completion of the next pass is used as the billet state before the start of the new current pass, and the subsequent passes after the next pass are defined as the new next pass. Continue generating new current pass organization evolution results, new current pass organization offset results, new next pass organization acceptance control constraints, and new next pass control parameter results, until the number of completed passes reaches the preset number of passes, and the dimensional state of the final pass is within the dimensional judgment boundary corresponding to the target product dimensional requirements, and the organization state quantity of the final pass is within the organization judgment boundary in the final target organization conditions.

[0014] Furthermore, the method for establishing the digital twin organizational evolution model includes: Acquire alloy material type, initial geometric data of alloy billet, multi-pass hot working process plan, range of hot working equipment, historical hot working process data, historical pass end status data and historical microstructure detection data; A geometric description of the billet is established based on the initial geometric data of the alloy billet and the multi-pass hot working process plan; a material thermophysical property relationship is established based on the alloy material type and the historical hot working process data; a hot deformation calculation relationship is established based on the historical hot working process data; and a microstructure state quantity calculation relationship is established based on the historical hot working process data, the historical pass end state data, and the historical microstructure detection data. The input-output relationship between the billet geometry description, the material thermophysical property relationship, the thermal deformation calculation relationship, and the microstructure state quantity calculation relationship is established to form the digital twin microstructure evolution model.

[0015] Furthermore, methods for determining the final constraint boundaries also include: When the current course organization offset result includes multiple organization offset bases, the corresponding successor relationship item is found according to each organization offset base, and the corresponding next course constraint boundary is formed respectively. When multiple next-level constraint boundaries correspond to the same next-level constraint term, the intersection of the multiple next-level constraint boundaries is processed, and the range after the intersection process is determined as the final constraint boundary of the corresponding next-level constraint term. When there is no executable range for the hot processing equipment after the intersection processing, the next pass constraint boundary formed by the organization offset corresponding to the target product organization requirements in the current pass organization offset result is intersected with the executable range of the hot processing equipment, and the range after the intersection processing is determined as the final constraint boundary of the corresponding next pass constraint item.

[0016] A digital twin-driven alloy hot working control system, used to implement the aforementioned digital twin-driven alloy hot working control method, the system comprising: The data acquisition module is used to acquire the hot working process data and the end status data of the current pass of the alloy billet. The organization generation module is used to generate the organization evolution result of the current pass based on the heat treatment process data and the current pass end status data, using a digital twin organization evolution model. The offset determination module is used to compare the organization state quantity in the current track organization evolution result with the organization determination boundary in the target organization condition corresponding to the current track item by item to determine the organization offset result of the current track. The constraint generation module is used to generate the next course organization succession control constraint based on the current course organization evolution result and the current course organization offset result. The parameter generation module is used to filter and combine candidate control parameters for the next pass based on the next pass organization acceptance control constraints, and generate the control parameter results for the next pass. The loop control module is used to output the control parameter results of the next pass to the heat processing execution device to execute the next heat processing pass. After the next pass is executed, the next pass is used as the new current pass to continue processing until the multi-pass heat processing process is completed.

[0017] Compared with related technologies, this application has the following advantages: First, this application obtains the hot working process data and the end state data of the current pass of the alloy billet, and generates the microstructure evolution result of the current pass through a digital twin microstructure evolution model. This allows temperature, deformation, load and end state of the current pass to participate in the determination of microstructure, avoiding the judgment of the current pass processing state based solely on equipment setting parameters or dimensional results, thereby improving the completeness of the characterization of the current pass microstructure.

[0018] Second, this application compares the microstructure state quantity in the current pass microstructure evolution result with the microstructure determination boundary in the target microstructure condition corresponding to the current pass item by item to determine the microstructure offset result of the current pass, so that the offsets such as recrystallization state, grain size, deformation accumulation, residual stress or microstructure distribution difference can be clearly identified, and a processing result including the offset object, offset direction and offset degree is formed, providing a callable basis for subsequent pass parameter adjustment.

[0019] Third, this application generates the next pass's organizational transition control constraints based on the current pass's organizational evolution results and current pass's organizational offset results. This allows the organizational state already formed in the current pass to be converted into constraints for the next pass's starting temperature, deformation amount, deformation speed, heat preservation and cooling conditions, or equipment execution timing. This establishes a transition relationship between organizational evolution results and parameter generation constraints between adjacent passes, reducing organizational control breakpoints between passes.

[0020] Fourth, this application selects and combines candidate control parameters for the next pass based on the control constraints of the next pass's organization, and generates control parameter results for the next pass. This ensures that the control parameters for the next pass are not only limited by the basic process plan and the executable range of the equipment, but also by the organization offset of the current pass. This helps to suppress the continued accumulation of organization offset in subsequent passes and improve the organization consistency control capability of multi-pass hot processing.

[0021] Fifth, after the next pass is completed, this application will treat the next pass as the new current pass and continue processing until the multi-pass hot working process is completed. This ensures that each pass is executed in a cyclical manner according to the processing chain of "generating microstructure evolution results - determining microstructure offset - generating microstructure acceptance control constraints - filtering and outputting control parameters". This forms a closed-loop acceptance control of the microstructure state for the entire process, improving the continuity, stability and consistency of the microstructure of the alloy billet in multi-pass hot working control. Attached Figure Description

[0022] Figure 1 A schematic flowchart of a digital twin-driven alloy hot working control method provided in this application; Figure 2 A schematic diagram of a digital twin-driven alloy hot working control system module provided in this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] Example 1 Please see Figure 1 As shown, this embodiment provides a digital twin-driven alloy hot working control method, which is applicable to parameter control of alloy billets in multi-pass hot working processes such as forging, rolling, extrusion, ring rolling or die forging; the alloy billet is a metal billet that undergoes heating, deformation, heat preservation, cooling or transfer treatment in multiple passes; the current pass is the single hot working process that is being collected, calculated and controlled, and the next pass is the hot working process that needs to be continued after the current pass.

[0025] In this embodiment, by collecting the heat treatment process data and the end state data of the current pass, the organization evolution result of the current pass is generated using a digital twin organization evolution model. Then, the organization state quantity in the organization evolution result of the current pass is compared item by item with the organization determination boundary in the target organization condition of the current pass to determine the organization offset result of the current pass. Then, the organization acceptance control constraint of the next pass is generated based on the organization evolution result and the organization offset result of the current pass, and the candidate control parameters of the next pass are screened using the organization acceptance control constraint of the next pass to form the control parameter result of the next pass, so that the heat treatment parameters of the next pass can accept the organization state already formed in the current pass.

[0026] Step 10, the specific steps for obtaining the hot working process data and the end status data of the current pass of the alloy billet include: Step 101: After the start of hot processing in the current pass, collect the hot processing process data of the alloy billet in the current pass; the hot processing process data includes temperature process data, deformation execution data, and load response data; the temperature process data is used to characterize the thermal state changes of the alloy billet and its processing environment in the current pass, including at least one of billet surface temperature data, furnace temperature data, die temperature data, temperature data of the area adjacent to the rolls, and the corresponding acquisition time; the deformation execution data is used to characterize the execution process of the hot processing equipment applying deformation to the alloy billet, including at least one of reduction amount, stroke position, deformation speed, holding time, rolling speed, extrusion speed, or die closing position; the load response data is used to characterize the response of the alloy billet to external loads during deformation, including at least one of pressure data, torque data, rolling force data, extrusion force data, load change sequence, or the time when the load peak occurs.

[0027] Step 102: After the current pass is completed, collect the end-of-pass status data; the end-of-pass status data is used to characterize the actual state of the alloy billet after the current pass, including at least one of the following: billet size data, billet surface thermal state data, billet shape deviation data, cooling dwell time data, or transfer waiting time data; the billet size data is used to characterize the length, width, thickness, diameter, or cross-sectional dimensions after the current pass; the billet surface thermal state data is used to characterize the surface temperature or temperature distribution of the billet after the current pass; the billet shape deviation data is used to characterize the deviation position and amount of the actual shape relative to the target shape after the current pass.

[0028] Step 103: Using the start and end times of the current pass as boundaries, limit the temperature process data, deformation execution data, and load response data to the same current pass; then, based on the acquisition time, perform time alignment on the temperature process data, deformation execution data, and load response data. If the sampling intervals of different data are consistent, then directly form data groups corresponding to each sampling time according to the same sampling time; if the sampling intervals of different data are inconsistent, then use the time of change of stroke position or change of equipment control command in the deformation execution data as the alignment time.

[0029] Step 104: For the time period between two adjacent settling moments, the temperature data within this time period is determined as the representative temperature value for the corresponding time period, the load data within this time period is determined as the representative load value for the corresponding time period, and the change in stroke position, change in compression, or change in speed within this time period is determined as the deformation execution amount for the corresponding time period. For the time period between two adjacent settling moments, first determine whether there is a deformation contact moment within this time period. If there is a deformation contact moment, determine the temperature value corresponding to the deformation contact moment as the representative temperature value for this time period, and determine the load value corresponding to the deformation contact moment as the representative load value for this time period. If there is no deformation contact moment, determine the average value of multiple temperature acquisition values ​​within this time period as the representative temperature value, and determine the average value of multiple load acquisition values ​​within this time period as the representative load value. If the difference between the temperature acquisition value at the end of this time period and the average temperature value exceeds the temperature representative value switching boundary, then determine the temperature acquisition value at the end of this time period as the representative temperature value. If the difference between the peak load value and the average load value within this time period exceeds the load representative value switching boundary, then determine the peak load value within this time period as the representative load value. The switching boundary for the temperature representative value is determined based on the temperature measurement accuracy and the allowable deviation of the temperature control of the heat processing equipment; the switching boundary for the load representative value is determined based on the load detection accuracy and the allowable deviation of the load of the heat processing equipment; the deformation execution amount is determined based on the actual displacement change, compression change, or speed change of the heat processing execution equipment within the specified time period. This forms the current pass's heat processing process data, arranged sequentially from multiple time periods.

[0030] Step 105: Based on the same alloy billet and the same current pass, establish a correspondence between the end state data of the current pass and the hot working process data; if the end state data of the current pass includes billet size data, then use the billet size data to verify the actual deformation result of the current pass; if the end state data of the current pass includes billet surface thermal state data, then use the billet surface thermal state data to verify the thermal state at the end of the current pass; if the end state data of the current pass includes billet shape deviation data, then use the billet shape deviation data to verify the deformation distribution result of the current pass; the hot working process data and the end state data of the current pass after the correspondence are completed are used in step 20 to generate the microstructure evolution result of the current pass.

[0031] In some implementations, the following example is used to illustrate the formation process of the settling time, representative temperature value, representative load value, and deformation execution amount.

[0032] For example, if the start time of the current pass is 0s and the end time is 60s, and the stroke position change times in the deformation execution data are 0s, 15s, 35s, and 60s respectively, then 0s, 15s, 35s, and 60s are determined as the settling times, forming three time periods: 0s to 15s, 15s to 35s, and 35s to 60s. If the billet surface temperature collected during the 0s to 15s time period is 1080℃, 1076℃, and 1072℃ respectively, and there is a deformation contact time of 10s within this time period, then 1076℃ corresponding to 10s is determined as the representative temperature value for this time period. If the rolling force collected during this time period is 8.1MN, 8.6MN, and 8.4MN respectively, and the rolling force corresponding to 10s is 8.6MN, then 8.6MN is determined as the representative load value for this time period. If the reduction changes from 0mm to 8mm during this time period, then 8mm is determined as the deformation execution amount for this time period. For the 15s to 35s time period, if no new deformation contact moment occurs within this period, multiple temperature acquisition values ​​within this period are read and averaged to obtain the representative temperature value for this period; multiple load acquisition values ​​within this period are read and averaged to obtain the representative load value for this period. Through the above processing, each time period corresponds to a representative temperature value, a representative load value, and a deformation execution amount, enabling the thermal processing data to be input into the subsequent digital twin tissue evolution model in time period order.

[0033] Step 20, the implementation steps for generating the microstructure evolution result of the current pass based on the hot working process data and the current pass end status data, include: Step 201: Establish or invoke a digital twin organizational evolution model. Specific methods for establishing such a model include: Step 2011: Obtain basic data for model building. This basic data includes alloy material type, initial geometric data of the alloy billet, multi-pass hot working process plan, the executable range of the hot working equipment, historical hot working process data, historical pass completion status data, and historical microstructure detection data. The initial geometric data of the alloy billet is used to determine the size, shape, and deformable area of ​​the alloy billet before the start of the current pass; the multi-pass hot working process plan is used to determine the processing sequence, deformation contact area, and process control range for each pass; the historical hot working process data provides information on temperature changes, deformation execution changes, and load response changes; and the historical microstructure detection data provides a reference for the microstructure state after the completion of historical passes.

[0034] Step 2012: Based on the initial geometric data of the alloy billet and the multi-pass hot working process plan, establish the geometric description of the billet. Specifically, first, determine the basic geometric shape of the billet based on its length, width, thickness, diameter, cross-sectional shape, or contour shape before the start of the current pass; then, determine the deformation contact area based on the die contact area, roll contact area, pressing area, extrusion area, or heated area corresponding to the current pass; subsequently, based on the positional relationship between the deformation contact area and the basic geometric shape of the billet, divide the alloy billet into several calculation zones, and determine the zone position, zone size, and correspondence between the zone and the deformation contact area for each calculation zone, thereby forming the geometric description of the billet.

[0035] Step 2013: Based on the alloy material type and historical hot working process data, establish the material thermophysical property relationships. Specifically, first, read the thermal conductivity, thermal softening characteristics, and deformation resistance variation patterns corresponding to the alloy material type; then, combine the process change results under different temperature conditions, different deformation execution conditions, and different load response conditions in the historical hot working process data to determine the correspondence between temperature change and heat transfer state, temperature change and thermal softening state, and deformation execution change and deformation resistance state. This forms the material thermophysical property relationships, which are used in subsequent calculations to determine the thermal state and deformation resistance state of the corresponding calculation zone based on representative temperature values, deformation execution amounts, deformation rates, and representative load values.

[0036] Step 2014: Establish thermal deformation calculation relationships based on historical hot processing data. Specifically, first, divide the historical hot processing data into passes according to the start and end times of each pass. Then, use the stroke position change time or equipment control command change time in the historical deformation execution data as the historical settling time. Determine the historical temperature data between adjacent historical settling times as the historical temperature representative value, and the historical load data between adjacent historical settling times as the historical load representative value. Determine the historical stroke position change, historical reduction change, or historical speed change within the corresponding time period as the historical deformation execution amount. Subsequently, establish a correspondence between the historical temperature representative value, historical deformation execution amount, historical deformation speed, and historical load representative value and the corresponding calculation partition to form a thermal deformation calculation relationship used to generate segmented thermal deformation states.

[0037] Step 2015: Determine the composition of segmented thermal deformation states based on the thermal deformation calculation relationship. For any time period and any calculation zone, the thermal deformation calculation relationship converts the representative temperature value corresponding to that time period into segmented thermal states, the deformation execution amount and deformation speed corresponding to that time period into segmented deformation states, and the representative load value corresponding to that time period into segmented deformation resistance states. The segmented thermal states include the representative temperature value, the direction of temperature change, and the duration of temperature variation; the segmented deformation states include the deformation execution amount, deformation speed, and the duration of deformation variation; and the segmented deformation resistance states include the representative load value, the direction of load change, and the duration of load variation. Thus, the thermal deformation calculation relationship can convert the thermal processing data within each time period of the current pass into segmented thermal deformation states for the corresponding calculation zone.

[0038] Step 2016: Based on historical heat treatment process data, historical pass end state data, and historical microstructure detection data, establish the calculation relationship for microstructure state quantities. Specifically, firstly, convert the historical heat treatment process data into historical cumulative thermal deformation states according to calculation zones and time periods; then, establish a correspondence between the historical cumulative thermal deformation states and the historical pass end state data, and use the historical pass end state data to verify the dimensional changes, surface thermal state changes, and shape distribution changes of the historical passes; subsequently, establish a correspondence between the verified historical cumulative thermal deformation states and the historical microstructure detection data, and determine the calculation relationship between recrystallization state quantities, grain size characterization quantities, deformation accumulation quantities, residual stress characterization quantities, and microstructure distribution difference quantities corresponding to different temperature change processes, deformation accumulation processes, deformation rate change processes, and load response change processes, thereby forming the calculation relationship for microstructure state quantities.

[0039] Step 2017 involves establishing an input-output relationship between the billet geometric description, material thermophysical property relationships, hot deformation calculation relationships, and microstructure state quantity calculation relationships to form a digital twin microstructure evolution model. This input-output relationship includes: determining calculation zones based on the billet geometric description; determining the thermal state and deformation resistance state of each calculation zone under different temperatures and deformation conditions based on the material thermophysical property relationships; converting the current hot working process data into the segmented hot deformation state and cumulative hot deformation state of each calculation zone based on the hot deformation calculation relationships; and converting the cumulative hot deformation state into the corresponding microstructure state quantity for each calculation zone based on the microstructure state quantity calculation relationships. After completing the above input-output relationship, the digital twin microstructure evolution model is obtained.

[0040] Step 2018: During the current pass processing, the digital twin microstructure evolution model is invoked. Specifically, the hot working process data obtained from the first title and the current pass end state data are input into the digital twin microstructure evolution model. The digital twin microstructure evolution model determines the calculation partition of the current pass based on the billet geometry description, generates the segmented hot deformation state and cumulative hot deformation state corresponding to each calculation partition based on the hot deformation calculation relationship, and then generates the microstructure evolution result of the current pass based on the microstructure state quantity calculation relationship.

[0041] Step 202: Based on the billet geometry description and the deformation contact area of ​​the current pass, the alloy billet is divided into several calculation zones. Specifically, the billet shape type, billet size, billet center position, outer surface position, and deformation contact area corresponding to the current pass are read from the billet geometry description. The billet shape type includes one of plate billet, bar billet, ring billet, or irregularly shaped billet. The deformation contact area is the area in the current pass where the die, rolls, pressure head, extrusion cylinder, or heating area interacts with the alloy billet by force or heat.

[0042] When the alloy billet is a plate-shaped billet, the length direction and thickness direction of the billet are used as the partitioning direction. First, the length partitioning boundary is determined along the length direction according to the change position of the equipment control command, the temperature measurement position and the boundary of the deformation contact area. Then, the thickness partitioning boundary is determined along the thickness direction according to the surface area, the middle area and the center area. The calculation partition is formed by the intersection of the length partitioning boundary and the thickness partitioning boundary.

[0043] When the alloy billet is a rod-shaped billet, the axial and radial directions are used as the partitioning directions. First, the axial partitioning boundary is determined along the axial direction based on the start and end positions of the deformation contact area, the temperature measurement position, and the load application position. Then, the radial partitioning boundary is determined along the radial direction based on the surface area, the mid-radius area, and the axial center area. The calculation partition is formed by the intersection of the axial partitioning boundary and the radial partitioning boundary.

[0044] When the alloy billet is an annular billet, at least two of the circumferential, radial, and thickness directions are used as partitioning directions. First, the circumferential partitioning boundary is determined based on the contact position between the roll or die and the annular billet. Then, the radial partitioning boundary or thickness partitioning boundary is determined based on the position of the inner diameter region, outer diameter region, and wall thickness direction. The calculated partitions are formed by the circumferential partitioning boundary, radial partitioning boundary, or thickness partitioning boundary.

[0045] When the alloy billet is an irregularly shaped billet, it is first divided into contact zones adjacent to the deformation contact area, transition zones adjacent to the contact zones, and non-contact zones far away from the deformation contact area according to the outer contour of the billet; then, the contact zones, transition zones and non-contact zones are further subdivided according to the temperature measurement position, load application position and shape change position to form calculation zones.

[0046] Each calculation partition includes a partition boundary, a partition location, a correspondence between the partition and the deformation contact area, and a correspondence between the partition and the data acquisition location. The partition boundary is used to define the spatial range of the calculation partition in the alloy billet. The partition location is used to determine the position of the calculation partition relative to the surface, center, or deformation contact area of ​​the billet. The correspondence between the partition and the deformation contact area is used to determine whether the calculation partition is directly deformed by the current pass. The correspondence between the partition and the data acquisition location is used to determine the representative temperature value, deformation amount, deformation speed, and load value corresponding to the calculation partition.

[0047] For calculation zones located within the deformation contact area, the deformation execution amount, deformation rate, and load representative value within the time period corresponding to the deformation contact area are used as the input data for that calculation zone. For calculation zones located outside the deformation contact area but adjacent to it, the estimated temperature representative value and estimated load representative value for that calculation zone are determined based on the spatial distance between the calculation zone and the deformation contact area, the temperature representative value of the adjacent calculation zone, and the load representative value. For calculation zones far from the deformation contact area, their corresponding temperature representative value is mainly determined based on the temperature representative value of the adjacent calculation zone and the billet geometry description, and their load representative value is mainly determined based on the load representative value of the adjacent calculation zone and the spatial distance between them and the deformation contact area.

[0048] If a certain calculation partition cannot directly obtain a representative temperature value or a representative load value, then the calculation partition adjacent to that partition and having acquired values ​​is used as a reference partition. If a certain calculation partition cannot directly obtain a representative temperature value or a representative load value, then the calculation partition adjacent to that partition and having acquired values ​​is selected as a reference partition. The spatial distance between the center position of each reference partition and the center position of the calculation partition is determined, and each spatial distance is converted into its reciprocal. The reciprocal distance corresponding to each reference partition is divided by the sum of the reciprocals of all reference partition distances to obtain the estimated weight corresponding to that reference partition. Then, the representative temperature value of each reference partition is multiplied by its corresponding estimated weight and summed to determine the estimated representative temperature value corresponding to that calculation partition. The representative load value of each reference partition is multiplied by its corresponding estimated weight and summed to determine the estimated representative load value corresponding to that calculation partition. If only one reference partition exists, then the representative temperature value or representative load value of that reference partition is determined as the estimated representative temperature value or estimated representative load value of that calculation partition. The estimated representative temperature value and estimated representative load value continue to serve as input data for generating the segmented thermal deformation state of that calculation partition within the current time period.

[0049] Step 203: Read the thermal processing data in the order of time periods, and determine the segmented thermal state, segmented deformation state, and segmented deformation resistance state of each calculation partition in each time period; specifically, first, according to the time period order formed in steps 103 and 104, read the representative temperature value, representative load value, deformation execution amount, deformation speed, and duration of each time period one by one; then, according to the correspondence between the calculation partition and the data acquisition location determined in step 202, determine the representative temperature value, representative load value, deformation execution amount, and deformation speed of each calculation partition in the current time period.

[0050] For calculation partitions located within the deformation contact area, the representative values ​​of temperature, load, deformation execution amount, and deformation speed corresponding to the deformation contact area in the current time period are used as the representative values ​​of temperature, load, deformation execution amount, and deformation speed of that calculation partition. For calculation partitions located outside the deformation contact area, if the estimated representative value of temperature or estimated representative value of load has been determined in step 202, then the estimated representative value of temperature or estimated representative value of load is read as the representative value of temperature or load of load in the current time period. For calculation partitions located outside the deformation contact area, the spatial distance between the calculation partition and the boundary of the deformation contact area is first determined, and adjacent to the calculation partition and whose deformation execution amount has been determined are selected. The calculation partitions for row volume and partition deformation speed are used as reference partitions; the spatial distance between the center position of each reference partition and the center position of the calculation partition are determined, and the estimated weights corresponding to each reference partition are obtained according to the estimation weight determination method in step 202; the partition deformation execution volume of each reference partition is multiplied by the corresponding estimated weight and summed to determine the partition deformation execution volume of the calculation partition; the partition deformation speed of each reference partition is multiplied by the corresponding estimated weight and summed to determine the partition deformation speed of the calculation partition; if the calculation partition is far from the deformation contact area and the deformation execution volume of all adjacent reference partitions does not exceed the execution identification boundary, then the partition deformation execution volume of the calculation partition in the current time period is determined to be zero, and the partition deformation speed is determined to be zero.

[0051] After determining the representative temperature value for each zone, the representative temperature value for the current time period is compared with that of the previous time period. If the difference between the current and previous representative temperature values ​​exceeds a preset temperature change detection boundary, the temperature change direction is determined to be heating. If the difference exceeds the preset temperature change detection boundary, the temperature change direction is determined to be cooling. If neither difference exceeds the preset temperature change detection boundary, the temperature change direction is determined to be maintaining the current temperature. The temperature change direction for the first time period is determined by comparing the current representative temperature value with the initial temperature value at the start of the current pass. The preset temperature change detection boundary is set based on the temperature measurement accuracy, the temperature control accuracy of the hot processing equipment, and the temperature control requirements of the alloy material, and is not limited to a specific value.

[0052] The segmented thermal state of the calculated partition within the current time period is determined by the representative value of the partition temperature, the direction of temperature change, and the duration of the current time period. The segmented deformation state of the calculated partition within the current time period is determined by the partition deformation execution amount, the partition deformation rate, and the duration of the current time period. The representative value of the partition load in the current time period is compared with the representative value of the partition load in the previous time period, and the load change direction is determined using the same comparison method as the temperature change direction. The segmented deformation resistance state of the calculated partition within the current time period is determined by the representative value of the partition load, the load change direction, and the duration of the current time period. The preset load change identification boundary is set based on the load detection accuracy, equipment load control accuracy, and the load fluctuation range of historical qualified batches, without limiting specific values.

[0053] Step 204: Match the segmented thermal state, segmented deformation state, and segmented deformation resistance state corresponding to the same calculation partition within the same time period to form segmented thermal deformation states. Specifically, for each time period and each calculation partition, read the segmented thermal state, segmented deformation state, and segmented deformation resistance state of that calculation partition within the current time period, and match them according to the same time period identifier and the same calculation partition identifier. The time period identifier is used to indicate the sequential position of the current time period in the current pass; the calculation partition identifier is used to indicate the billet space region corresponding to the segmented thermal deformation state.

[0054] After the mapping is completed, the segmented thermal deformation state of the calculation partition within the current time period is formed. The segmented thermal deformation state includes the time period identifier, the calculation partition identifier, the representative value of the partition temperature, the direction of temperature change, the duration of temperature change, the partition deformation execution amount, the partition deformation speed, the duration of deformation, the representative value of the partition load, the direction of load change, and the duration of load. The segmented thermal deformation state is used to represent the thermal effects, deformation effects, and load effects that the same calculation partition is subjected to simultaneously within the same time period, and serves as the input object for determining the cumulative thermal deformation state in step 205.

[0055] Step 205: For the same calculation partition, process the segmented thermal deformation state according to the chronological order of each time period within the current trace to determine the cumulative thermal deformation state of the calculation partition within the current trace. Specifically, first read the segmented thermal deformation states corresponding to each time period within the current trace for the same calculation partition, and then arrange them in order from first to last according to the time period identifiers. Subsequently, arrange the partition temperature representative value, temperature change direction, and temperature duration in each time period in chronological order to form the temperature change process of the calculation partition; accumulate the partition deformation execution amount in each time period in chronological order to form the deformation accumulation process of the calculation partition; arrange the partition deformation speed and deformation duration in each time period in chronological order to form the deformation speed change process of the calculation partition; arrange the partition load representative value, load change direction, and load duration in chronological order in each time period to form the load response change process of the calculation partition.

[0056] When determining the deformation accumulation process, if the deformation execution amount in the current time period corresponds to a change in compression, a change in stroke position, or a change in extrusion displacement, then the deformation execution amount for each time period is accumulated according to the deformation direction of the current pass. If there is no effective deformation execution amount in the current time period, then the deformation execution amount for that time period is treated as zero, while retaining the temperature change process and load response change process for that time period. The effective deformation execution amount is determined based on whether the changes in equipment control commands, stroke position, or compression exceed a preset execution identification boundary. The preset execution identification boundary is set based on the equipment control resolution and historical qualified batch execution records, and is not limited to a specific value.

[0057] The cumulative thermal deformation state of the calculation zone in the current pass is formed by the temperature change process, deformation accumulation process, deformation rate change process, and load response change process. The cumulative thermal deformation state includes the calculation zone identifier, the current pass identifier, the temperature change process, the deformation accumulation process, the deformation rate change process, and the load response change process, and serves as the input object for calculating the microstructure state in step 207.

[0058] In some implementations, the following example is used to illustrate how the segmented thermal deformation state forms the cumulative thermal deformation state: For example, a plate-shaped billet in the current pass is divided into a surface calculation zone, an intermediate calculation zone, and a center calculation zone. For the surface calculation zone, the segmented thermal deformation states formed during the 0s to 15s time period include a representative zone temperature of 1076℃, a cooling direction, a temperature duration of 15s, a deformation amount of 8mm, a deformation speed of 0.53mm / s, a representative zone load of 8.6MN, an increasing load direction, and a load duration of 15s; the segmented thermal deformation states formed during the 15s to 35s time period include a representative zone temperature of 1062℃, a cooling direction, a temperature duration of 20s, a deformation amount of 6mm, a deformation speed of 0.30mm / s, a representative zone load of 8.9MN, an increasing load direction, and a load duration of 20s; and the segmented thermal deformation shapes formed during the 35s to 60s time period... The state includes a representative temperature value of 1050℃ for the zone, a decreasing temperature change direction, a temperature duration of 25s, a zone deformation amount of 0mm, a zone deformation rate of 0mm / s, a representative load value of 3.0MN for the zone, a decreasing load change direction, and a load duration of 25s. After processing in time intervals, the representative temperature values, temperature change directions, and temperature durations of the three time intervals are combined to form the temperature change process of the calculated zone for the surface layer. The 8mm, 6mm, and 0mm values ​​are accumulated in time sequence to form the deformation accumulation process of the calculated zone for the surface layer. The zone deformation rates and deformation durations of the three time intervals are combined to form the deformation rate change process. The zone load rates, load change directions, and load durations of the three time intervals are combined to form the load response change process. This forms the cumulative thermal deformation state of the calculated zone for the surface layer within the current pass, which serves as the input for subsequent calculations of the microstructure state.

[0059] Step 206: Establish the calculation relationship for microstructure state based on alloy material type, historical hot working process data, and historical microstructure testing data. Specifically, first, classify the historical hot working process data and historical microstructure testing data according to alloy material type, so that the same type of historical data corresponds to the same or similar alloy material type. Then, following the same processing method as steps 203 to 205, divide the historical hot working process data into historical time periods and historical calculation zones, and generate the historical cumulative thermal deformation state of each historical calculation zone within the historical passes. The historical cumulative thermal deformation state includes the historical temperature change process, the historical deformation accumulation process, the historical deformation rate change process, and the historical load response change process.

[0060] Subsequently, the microstructure detection results corresponding to each historical pass and each historical calculation partition are read from the historical microstructure detection data. These microstructure detection results include at least one of the following: recrystallization detection results, grain size detection results, cumulative deformation evaluation results, residual stress detection results, or partition microstructure difference detection results. Using the historical cumulative thermal deformation state as input samples and the corresponding microstructure detection results as output samples, a corresponding sample relationship between the cumulative thermal deformation state and the microstructure state quantity is established.

[0061] When establishing calculation relationships for microstructure states, calculation relationships are established according to the type of microstructure state. For recrystallization states, a correspondence is established between historical temperature changes, historical deformation accumulation, historical deformation rate changes, historical heat preservation and cooling processes, and recrystallization detection results. For grain size characterization quantities, a correspondence is established between historical recrystallization detection results, historical temperature changes, historical deformation accumulation, and grain size detection results. For deformation accumulation quantities, a correspondence is established between the deformation execution quantities within a historical time period and the deformation accumulation evaluation results. For residual stress characterization quantities, a correspondence is established between historical load response changes, historical deformation accumulation, historical pass end size deviation, and residual stress detection results. For microstructure distribution difference quantities, a correspondence is established between the differences in detection results of the same type of microstructure in different historical calculation zones and the zone-specific microstructure difference detection results.

[0062] The relationship for calculating the organizational state quantity is established using a historical sample difference matching method. During establishment, the historical temperature change process, historical deformation accumulation process, historical deformation rate change process, and historical load response change process in each historical cumulative thermal deformation state are first converted into comparable historical input quantities. Then, the temperature change process, deformation accumulation process, deformation rate change process, and load response change process in the current cumulative thermal deformation state are converted into current input quantities. Subsequently, the current input quantities are compared item by item with each historical input quantity to obtain the temperature process difference quantity, deformation accumulation difference quantity, deformation rate difference quantity, and load response difference quantity, respectively. These differences are then weighted according to their respective weighting relationships to obtain the matching difference quantity between the current cumulative thermal deformation state and each historical cumulative thermal deformation state. The weighting relationships for each difference quantity are determined based on the degree of influence of each input quantity on the organizational test results in historical qualified batches. The matching difference quantities are sorted from smallest to largest, and the historical organizational test result corresponding to the historical cumulative thermal deformation state with the smallest ranking is used as the basis for calculating the current organizational state quantity. If there are multiple historical cumulative thermal deformation states with higher ranking matching difference quantities, their corresponding historical organizational test results are weighted according to the reciprocal of the matching difference quantity to determine the current organizational state quantity.

[0063] Step 207: For each calculation zone, read the cumulative thermal deformation state and calculate the microstructure state quantity using the microstructure state quantity calculation relationship. Specifically, first read the cumulative thermal deformation state corresponding to the calculation zone formed in step 205, and then extract the temperature change process, deformation accumulation process, deformation rate change process, and load response change process from the cumulative thermal deformation state according to the input objects required for the microstructure state quantity calculation relationship. Simultaneously, read at least one of the following from the current pass end state data: billet size data, billet surface thermal state data, billet shape deviation data, cooling residence time data, or transfer waiting time data, as the end state input object for the microstructure state quantity calculation.

[0064] When calculating the recrystallization state quantity, the temperature change process, deformation accumulation process, deformation rate change process, and heat preservation and cooling process corresponding to the calculation partition are input into the calculation relationship corresponding to the recrystallization state quantity to obtain the recrystallization state quantity of the calculation partition. The heat preservation and cooling process is determined based on the heat preservation time, cooling dwell time, transfer waiting time, and temperature change process in the current pass. When calculating the grain size characterization quantity, the recrystallization state quantity, temperature change process, and deformation accumulation quantity are input into the calculation relationship corresponding to the grain size characterization quantity to obtain the grain size characterization quantity of the calculation partition. When calculating the deformation accumulation quantity, the partition deformation execution quantity of each time period in the current pass is accumulated in chronological order to obtain the deformation accumulation quantity of the calculation partition. When calculating the residual stress characterization quantity, the load response change process, deformation accumulation quantity, and dimensional deviation after the end of the current pass are input into the calculation relationship corresponding to the residual stress characterization quantity to obtain the residual stress characterization quantity of the calculation partition.

[0065] At least one of the above recrystallization state quantity, grain size characterization quantity, deformation accumulation quantity and residual stress characterization quantity, together with the corresponding calculation partition identifier and the current pass identifier, constitutes the microstructure state quantity of the calculation partition; the microstructure state quantity is used to determine the microstructure distribution difference quantity in step 208, and is used to compare it item by item with the microstructure determination boundary in the target microstructure condition in step 30.

[0066] Step 208: Determine the difference in microstructure distribution based on the microstructure state quantities corresponding to different calculation partitions. Specifically, first, select the same type of microstructure state quantity from the current pass's microstructure evolution results. The same type of microstructure state quantity is one of the recrystallization state quantity, grain size characterization quantity, deformation accumulation quantity, or residual stress characterization quantity that has been calculated in different calculation partitions. Then, determine the calculation partition pairs to be compared according to the preset partition comparison order. The preset partition comparison order is determined based on the billet geometry description, deformation contact area location, and target product microstructure requirements, and may include at least one of the following: comparison of adjacent partitions, comparison of surface partitions and center partitions, and comparison of contact partitions and non-contact partitions.

[0067] For each computational partition pair, the difference between the same type of organizational state variables in the two computational partitions is calculated to form a partition difference value. If the same type of organizational state variable is a numerical organizational state variable, the partition difference value is determined based on the difference between the two organizational state variables; if the same type of organizational state variable is a hierarchical organizational state variable, the partition difference value is determined based on the hierarchical difference between the corresponding hierarchical levels of the two organizational state variables. After comparing all computational partition pairs, the partition difference values ​​are arranged from largest to smallest, and the partition difference value with the largest value is determined as the maximum partition difference value. The computational partition pair corresponding to the maximum partition difference value is determined as the maximum difference partition pair.

[0068] The organization distribution difference quantity is formed by the same type of organization state quantity, the preset partition comparison order, the difference value of each partition, the maximum partition difference value, and the maximum difference partition pair. The organization distribution difference quantity is used to characterize the degree of inconsistency between the organization states of different calculation partitions after the current pass is completed, and is used to compare the organization determination boundary corresponding to the organization distribution difference quantity in step 30.

[0069] Step 209: Correct the microstructure state based on the current pass's end-of-pass data. Specifically, the digital twin microstructure evolution model first calculates the predicted dimensions, predicted surface thermal state, and predicted shape distribution at the end of the current pass based on the hot working process data and cumulative hot deformation state. The predicted dimensions are determined based on the cumulative deformation process and billet geometry of each calculation zone, representing the predicted dimensional state of the alloy billet at the end of the current pass. The predicted surface thermal state is determined based on the temperature change process of each surface calculation zone, representing the predicted thermal state of the billet surface at the end of the current pass. The predicted shape distribution is determined based on the cumulative deformation process and microstructure distribution differences of each calculation zone, representing the predicted shape distribution of different regions of the billet at the end of the current pass.

[0070] Then, the predicted size results are compared with the measured billet size data in the current pass end state data to obtain the size difference results; the predicted surface thermal state results are compared with the measured billet surface thermal state data in the current pass end state data to obtain the thermal state difference results; the predicted shape distribution results are compared with the measured billet shape deviation data in the current pass end state data to obtain the shape distribution difference results; the size difference results include the size difference direction, size difference amount, and size difference corresponding region; the thermal state difference results include the thermal state difference direction, thermal state difference amount, and thermal state difference corresponding region; the shape distribution difference results include the shape deviation direction, shape deviation amount, and shape deviation zone.

[0071] The final state consistency comparison result is formed by at least one of the size difference result, thermal state difference result, and shape distribution difference result; the final state consistency comparison result is used to correct the tissue state quantity in step 210.

[0072] Step 210: Based on the consistency comparison results of the final state, correct the microstructure state quantities of the corresponding calculated partitions and form the corrected microstructure evolution results for the current pass. Specifically, first read the dimensional difference results, thermal state difference results, and shape distribution difference results from the consistency comparison results of the final state, and then compare them with the corresponding preset consistency boundaries. The preset dimensional consistency boundary is used to determine whether the dimensional difference between the predicted dimensional result and the measured billet dimensional data needs to be corrected; the preset thermal state consistency boundary is used to determine whether the thermal state difference between the predicted surface thermal state result and the measured billet surface thermal state data needs to be corrected; the preset shape consistency boundary is used to determine whether the shape distribution difference between the predicted shape distribution result and the measured billet shape deviation data needs to be corrected. The above preset consistency boundaries are set based on the detection accuracy, equipment control accuracy, and allowable deviation of the target product, and are not limited to specific values.

[0073] If the dimensional difference exceeds the preset dimensional consistency boundary, the calculation zone that needs correction is determined based on the area corresponding to the dimensional difference, and the correction direction for the cumulative deformation is determined based on the direction of the dimensional difference. Then, the correction amount for the cumulative deformation is determined based on the dimensional difference and the preset dimensional correction relationship, and the cumulative deformation of the corresponding calculation zone is corrected using the correction amount. The preset dimensional correction relationship is determined based on the correspondence between historical dimensional difference results and historical cumulative deformation correction records.

[0074] If the thermal state difference results exceed the preset thermal state consistency boundary, the calculation partitions that need correction are determined based on the corresponding regions of the thermal state difference, and the correction direction for the temperature change process is determined based on the direction of the thermal state difference. Then, the correction amount for the temperature change process is determined based on the thermal state difference amount and the preset thermal state correction relationship, and this correction amount is used to correct the temperature change process of the corresponding calculation partition. The preset thermal state correction relationship is determined based on the correspondence between historical thermal state difference results and historical temperature change process correction records.

[0075] If the shape distribution difference exceeds the preset shape consistency boundary, the calculation partition that needs correction is determined based on the partition where the shape deviation is located, and the correction direction of the tissue distribution difference is determined based on the direction of the shape deviation. Then, the correction amount of the tissue distribution difference is determined based on the shape deviation amount and the preset shape correction relationship, and the tissue distribution difference amount of the corresponding calculation partition is corrected using the tissue distribution difference correction amount. The preset shape correction relationship is determined based on the correspondence between historical shape distribution difference results and historical tissue distribution difference correction records.

[0076] After the above corrections are completed, if the cumulative deformation or temperature change process is corrected, the microstructure state calculation relationship in step 207 is called back to update the recrystallization state, grain size characterization, and residual stress characterization of the corresponding calculation partition; if the microstructure distribution difference is corrected, the corresponding microstructure distribution difference in the current pass microstructure evolution result is updated; finally, the corrected current pass microstructure evolution result is formed. The corrected current pass microstructure evolution result includes the microstructure state, microstructure distribution difference, calculation partition identifier, and current pass identifier corresponding to each calculation partition, and is used to determine the current pass microstructure offset result in the third heading.

[0077] In some implementations, the following example may be used to illustrate the composition of the revised current track organization evolution result: For example, the current pass microstructure evolution results include the microstructure states corresponding to the surface calculation zone, intermediate calculation zone, and center calculation zone. For the surface calculation zone, based on the calculation relationship between the cumulative thermal deformation state and the microstructure state, the recrystallization state value is 0.72, the grain size characterization value is 18μm, the cumulative deformation value is 14mm, and the residual stress characterization value is level 2. For the intermediate calculation zone, the recrystallization state value is 0.68, the grain size characterization value is 20μm, the cumulative deformation value is 12mm, and the residual stress characterization value is level 2. For the center calculation zone, the recrystallization state value is 0.55, the grain size characterization value is 26μm, the cumulative deformation value is 9mm, and the residual stress characterization value is level 3. If the difference between the measured surface thermal state and the predicted surface thermal state in the current pass end state data exceeds the preset thermal state consistency boundary, then the calculation zone that needs to be corrected is determined based on the region corresponding to the thermal state difference, and the temperature change process of the calculation zone is corrected according to the preset thermal state correction relationship. After correction, the microstructure state calculation relationship is re-invoked to update the recrystallization state, grain size characterization, and residual stress characterization of the corresponding calculation partition. The final corrected microstructure evolution result for the current pass includes the current pass identifier, the identifiers of each calculation partition, the microstructure state corresponding to each calculation partition, and the microstructure distribution difference, which is used for subsequent item-by-item comparison with the target microstructure conditions corresponding to the current pass.

[0078] Step 30: Compare the current track organization evolution result with the target organization conditions item by item and determine the current track organization offset result; specific implementation steps include: Step 301: Determine the target microstructure conditions for the current pass before execution. Specifically, first read the alloy material type, target product microstructure requirements, multi-pass hot working process plan, and current pass sequence position. The alloy material type is used to determine the evaluation object of the microstructure quantity. The target product microstructure requirements are used to determine the allowable microstructure range of the final product. The multi-pass hot working process plan is used to determine the processing purpose of the current pass in the entire hot working process. The current pass sequence position is used to determine the microstructure control role undertaken by the current pass.

[0079] When the current pass is used to establish the initial deformation structure, the cumulative deformation and the difference in structure distribution are used as the criteria for determining the current pass. When the current pass is used to promote structure adjustment, the recrystallization state, grain size characterization, and residual stress characterization are used as the criteria for determining the current pass. When the current pass is used to correct the structure shift of the previous pass, the type of the shifted structure state value in the previous pass's structure shift result is read, and the structure state value corresponding to the type of shifted structure state value is used as the correction criteria for the current pass. When the same current pass corresponds to multiple structure control effects, the structure state values ​​corresponding to each structure control effect are all used as the criteria for determining the current pass. This forms the set of criteria for determining the target structure conditions corresponding to the current pass.

[0080] Step 302: For each organizational state quantity in the set of judgment objects, an organizational judgment boundary is formed. Specifically, the target organizational range corresponding to each organizational state quantity in the organizational requirements of the target product is read first, and then the organizational test results of the same alloy material type and the same or similar pass sequence position in the historical qualified batches are read. Combined with the executable range of the hot processing equipment and the current process purpose, the organizational judgment boundary corresponding to each organizational state quantity is determined.

[0081] For recrystallization state quantities, grain size characterization quantities, deformation accumulation quantities, and residual stress characterization quantities, the microstructure determination boundary is expressed as an allowable numerical range or an allowable grade range; the allowable numerical range includes a lower boundary and an upper boundary, and the allowable grade range includes the lowest allowable grade and the highest allowable grade; for microstructure distribution difference quantities, the microstructure determination boundary is expressed as an allowable partition difference range, which is used to limit the difference boundary of the same microstructure state quantity between different calculation partitions; a one-to-one correspondence is established between each microstructure state quantity, microstructure determination boundary, current pass identifier, and calculation partition identifier to form the target microstructure condition corresponding to the current pass.

[0082] Step 303: Read the microstructure state quantities from the current pass microstructure evolution results formed in Step 20. Specifically, first read the current pass microstructure evolution results according to the current pass identifier, and then read at least one of the recrystallization state quantity, grain size characterization quantity, deformation accumulation quantity, residual stress characterization quantity, and microstructure distribution difference quantity corresponding to each calculation partition according to the calculation partition identifier. Subsequently, read the microstructure judgment boundary corresponding to each microstructure state quantity from the target microstructure conditions corresponding to the current pass, and establish a comparison correspondence according to the microstructure state quantity type and calculation partition identifier. After completing the correspondence, compare each microstructure state quantity with the corresponding microstructure judgment boundary item by item to form a single comparison result corresponding to each microstructure state quantity.

[0083] Step 304: When the tissue judgment boundary is within the allowable numerical range, determine the individual comparison result according to the numerical range comparison method; specifically, read the tissue state quantity, the lower boundary of the allowable numerical range, and the upper boundary of the allowable numerical range; if the tissue state quantity is lower than the lower boundary of the allowable numerical range, the individual comparison result is determined to be below the tissue judgment boundary, the tissue offset direction is determined to be below the boundary, and the degree of tissue offset is determined based on the difference between the lower boundary of the allowable numerical range and the tissue state quantity; if the tissue state quantity exceeds the upper boundary of the allowable numerical range, the individual comparison result is determined to be above the tissue judgment boundary, the tissue offset direction is determined to be above the boundary, and the degree of tissue offset is determined based on the difference between the tissue state quantity and the upper boundary of the allowable numerical range; if the tissue state quantity is between the lower boundary and the upper boundary of the allowable numerical range, the individual comparison result is determined to be no offset.

[0084] Step 305: When the organization determination boundary is within the allowable level range, determine the individual comparison result according to the level range comparison method; specifically, first read the organization state quantity and the preset level boundary, and convert the organization state quantity into the corresponding level of the organization state quantity according to the preset level boundary; the preset level boundary is determined based on the organization test results of historical qualified batches, the organization requirements of the target product and the purpose of the current process, and is used to convert the organization state quantity into a comparable level.

[0085] Subsequently, the minimum and maximum allowable levels within the allowed level range are read. If the level corresponding to the organizational state quantity is lower than the minimum allowable level, the single comparison result is determined as insufficient level, the organizational offset direction is determined as insufficient level, and the degree of organizational offset is determined based on the level difference between the minimum allowable level and the level corresponding to the organizational state quantity. If the level corresponding to the organizational state quantity exceeds the maximum allowable level, the single comparison result is determined as excessive level, the organizational offset direction is determined as excessive level, and the degree of organizational offset is determined based on the level difference between the level corresponding to the organizational state quantity and the maximum allowable level. If the level corresponding to the organizational state quantity is within the allowed level range, the single comparison result is determined as no offset has occurred.

[0086] Step 306: When the organizational determination boundary is within the allowable partition difference range, determine the individual comparison result according to the partition difference comparison method. Specifically, first, read the organizational distribution difference amount from the current trace organizational evolution result. The organizational distribution difference amount includes the partition difference value corresponding to the same type of organizational state quantity, the corresponding calculated partition pair, and the preset partition comparison order. Then, compare each partition difference value with the allowable partition difference range. If the partition difference value exceeds the allowable partition difference range, the individual comparison result is determined as partition difference exceeding the limit, the organizational offset direction is determined as partition difference exceeding the limit, and the degree of organizational offset is determined based on the magnitude by which the partition difference value exceeds the allowable partition difference range. If the partition difference value is within the allowable partition difference range, the individual comparison result corresponding to the organizational distribution difference amount is determined as no offset has occurred.

[0087] Step 307: Based on the individual comparison results corresponding to each organizational state variable, form candidate organizational offset results for the current track. Specifically, read each individual comparison result, filter out those below the organizational judgment boundary, above the organizational judgment boundary, with insufficient level, exceeding the level, or exceeding the partition difference limit, and remove individual comparison results that do not form an offset. For each filtered individual comparison result, extract the corresponding organizational state variable type, calculate the partition identifier, organizational offset direction, organizational offset degree, and corresponding organizational judgment boundary to form an organizational offset candidate. One or more organizational offset candidates form the candidate organizational offset results for the current track.

[0088] Step 308: When the current pass organization offset candidate results include multiple organization offset candidates, determine the main offset basis and auxiliary offset basis according to the preset organization state processing order. Specifically, first read the performance correlation level corresponding to each organization state quantity in the target product performance requirements, then read the control order of each organization state quantity in the current pass process objective, and read the number of times or the magnitude of the impact of each organization state quantity offset on the control parameters of subsequent passes in the historical process data.

[0089] Subsequently, it is first determined whether each organizational offset candidate corresponds to the acceptance index in the target product performance requirements, and the organizational offset candidates corresponding to the acceptance index are ranked first. When multiple organizational offset candidates correspond to the acceptance index or do not correspond to the acceptance index, it is determined whether each organizational offset candidate belongs to the control object in the current process objective, and the organizational offset candidates belonging to the control object in the current process objective are ranked first. If the above conditions are still the same, the historical number of times each organizational state quantity caused the subsequent control parameter adjustment in the historical process data to form an offset is read, and sorted according to the historical number from most to least. If the historical number is the same, the historical control parameter adjustment amount corresponding to each organizational state quantity is read, and sorted according to the historical control parameter adjustment amount from largest to smallest. The organizational offset candidate ranked first is determined as the primary offset basis, and the remaining organizational offset candidates are determined as auxiliary offset basis. If only one organizational offset candidate exists, it is determined as the primary offset basis, and the auxiliary offset basis is set to empty. If no organizational offset candidate exists, it is determined that no organizational offset has been formed in the current process, and the organizational offset result of the current process is set to no offset result.

[0090] Step 309: Generate the current track organization offset result based on the current track organization offset candidate result, the main offset basis, and the auxiliary offset basis. Specifically, the current track organization offset result includes the current track identifier, the offset organization state type, the offset calculation partition, the organization offset direction, the organization offset degree, the corresponding organization judgment boundary, the main offset basis, and the auxiliary offset basis. The current track organization offset result is used in step 40 to generate the next track organization acceptance control constraint.

[0091] In some implementations, the following example can be used to illustrate how the current track shift result is formed: For example, in the target microstructure conditions corresponding to the current pass, the microstructure determination boundary for recrystallization state is 0.65 to 0.85, the microstructure determination boundary for grain size characterization is 16 μm to 24 μm, and the allowable level range for residual stress characterization is level 1 to level 2. After reading the corrected microstructure evolution results for the current pass, the recrystallization state of the surface calculation partition is 0.72, which is within the corresponding microstructure determination boundary, and the single-item comparison result is no offset; the grain size characterization of the middle calculation partition is 20 μm, which is within the corresponding microstructure determination boundary, and the single-item comparison result is no offset. The following parameters are considered: The recrystallization state value of the central calculation partition is 0.55, which is lower than the lower boundary of the microstructure determination boundary corresponding to the recrystallization state value. The single-item comparison result is lower than the microstructure determination boundary, the microstructure shift direction is below the boundary, and the degree of microstructure shift is determined based on the difference between 0.65 and 0.55. The grain size characterization value of the central calculation partition is 26 μm, which exceeds the upper boundary of the microstructure determination boundary corresponding to the grain size characterization value. The single-item comparison result is higher than the microstructure determination boundary, the microstructure shift direction is above the boundary, and the degree of microstructure shift is determined based on the difference between 26 μm and 24 μm. Subsequently, based on the preset processing order of organizational states, the recrystallization state quantity offset, which corresponds to the performance requirements of the target product and has the highest impact on the control parameters of subsequent passes, is determined as the primary offset basis, and the grain size characterization quantity offset is determined as the secondary offset basis. This generates the current pass organizational offset result, which includes the current pass identifier, the type of offset organizational state quantity, the offset calculation partition, the organizational offset direction, the organizational offset degree, the corresponding organizational judgment boundary, the primary offset basis, and the secondary offset basis. This result serves as the input object for the organizational control constraints of the next pass.

[0092] Step 40: Generate the next course organization succession control constraints based on the current course organization evolution results and the current course organization offset results. Specific implementation steps include: Step 401: Read the current pass microstructure evolution result and the current pass microstructure offset result. Specifically, first, read the current pass identifier, calculation zone identifier, and microstructure state quantity corresponding to each calculation zone from the current pass microstructure evolution result to determine the microstructure state of the alloy billet before entering the next pass. Then, read the offset microstructure state quantity type, offset calculation zone, microstructure offset direction, microstructure offset degree, main offset basis, and auxiliary offset basis from the current pass microstructure offset result to determine the microstructure offset items that need to be corrected in the next pass. After reading, use the current pass microstructure evolution result and the current pass microstructure offset result as input objects to generate the microstructure acceptance control constraints for the next pass.

[0093] Step 402: Form or invoke a preset organizational acceptance relationship; specifically, the preset organizational acceptance relationship consists of multiple acceptance relationship items, each of which represents the corresponding processing relationship between the current pass organizational offset situation and the next pass control constraint; each acceptance relationship item includes organizational state quantity type, organizational offset direction, organizational offset degree range, offset calculation partition type, next pass constraint item, constraint adjustment direction, and constraint adjustment magnitude formation method.

[0094] Among them, the organization state type is used to match the offset organization state type in the current pass organization offset result; the organization offset direction is used to match the organization offset direction in the current pass organization offset result; the organization offset degree range is used to match the organization offset degree in the current pass organization offset result; the offset calculation partition type is determined based on the position of the offset calculation partition relative to the deformation contact area, billet surface area, billet center area, or non-contact area, and is used to match the offset calculation partition in the current pass organization offset result; the next pass constraint item is used to determine the control object that needs to be corrected; the constraint adjustment direction and constraint adjustment amplitude formation method are used to determine the correction method of the next pass constraint boundary relative to the basic control boundary.

[0095] Step 403: Establish a preset microstructure relationship. Specifically, first, read the alloy material type, target product microstructure requirements, and multi-pass hot working process plan to determine the control objects that can be adjusted for each pass. The control objects that can be adjusted include at least one of the following: starting temperature, deformation amount, deformation speed, heat preservation and cooling conditions, and equipment execution sequence. Then, read the historical current pass microstructure offset results, historical subsequent pass control parameters, and historical final microstructure detection results of the same historical billet between adjacent passes from the historical process data, and establish the correspondence between the three according to the same historical billet identifier, adjacent pass identifier, and historical calculation zone identifier. Classify the historical current pass microstructure offset results according to the microstructure state type, microstructure offset direction, microstructure offset degree range, and offset calculation zone type to form multiple historical offset categories.

[0096] For each historical offset category, the corresponding historical subsequent pass control parameters and historical final organization detection results are read. First, it is determined whether the organization detection quantity in the historical final organization detection result is within the organization determination boundary corresponding to the target product's organization requirements. If it is within the organization determination boundary, the corresponding historical subsequent pass control parameters are used as valid subsequent control parameters. If it is not within the organization determination boundary, the corresponding historical subsequent pass control parameters are not used as valid subsequent control parameters for that historical offset category. Then, the valid subsequent control parameters are compared item by item with the corresponding historical pass basic control parameters to determine at least one of the following: initial temperature change direction, deformation change direction, deformation speed change direction, heat preservation and cooling condition change direction, or equipment execution timing change direction. These change directions are determined as the subsequent pass control parameter change directions corresponding to that historical offset category, and the subsequent pass control parameter change directions are converted into the next pass constraint item and constraint adjustment direction, forming corresponding inheritance relationship items. Multiple inheritance relationship items together form a preset organization inheritance relationship.

[0097] Step 404: Determine the constraint adjustment range formation method. Specifically, first, read the historical organization offset degree in each historical offset category, and read the effective subsequent control parameters and historical track basic control parameters corresponding to that historical offset category. Then, compare the effective subsequent control parameters with the historical track basic control parameters item by item to determine at least one of the following: starting temperature difference, deformation difference, deformation speed difference, heat preservation and cooling condition difference, or equipment execution timing difference. Next, classify the above differences according to the interval to which the historical organization offset degree belongs, and determine the average value of the difference, the median value of the difference, or the difference that appears most frequently in the historical qualified batches within the same offset degree interval as the correction range corresponding to that offset degree interval. The constraint adjustment range formation method is composed of the offset degree interval, constraint item name, adjustment direction, and correction range.

[0098] Step 405: Read the basic control boundary of the next pass. Specifically, first, read the basic process parameter range corresponding to the next pass from the multi-pass hot processing process plan, then read the parameter range allowed to be executed by the equipment from the executable range of the hot processing equipment, and take the intersection of the basic process parameter range and the parameter range allowed to be executed by the equipment to obtain the basic control boundary of the next pass. The basic control boundary of the next pass includes at least one of the following: the basic boundary of the starting temperature of the next pass, the basic boundary of the deformation amount distribution, the basic boundary of the deformation rate, the basic boundary of the heat preservation and cooling, and the basic boundary of the equipment execution timing. The basic control boundary of the next pass is used to represent the basic range of control parameters that can be adopted for the next pass without considering the current pass's microstructure offset result.

[0099] Step 406: Search for the corresponding relationship item in the preset organizational relationship. Specifically, first read the offset organizational state type, organizational offset direction, organizational offset degree, and offset calculation partition in the current track's organizational offset result, and then determine the offset calculation partition type corresponding to the offset calculation partition. Subsequently, according to the matching conditions of the same organizational state type, the same organizational offset direction, the organizational offset degree falling into the corresponding organizational offset degree interval, and the same offset calculation partition type, search for a completely corresponding relationship item in the preset organizational relationship. If a completely corresponding relationship item is found, read the next track constraint item, constraint adjustment direction, and constraint adjustment magnitude formation method in the relationship item.

[0100] If no completely corresponding successor relationship item is found, then a superior successor relationship item is searched according to the matching conditions of the same organizational state quantity type and the same organizational offset direction. The superior successor relationship item does not limit the offset calculation partition type or the organizational offset degree range, but still limits the organizational state quantity type and organizational offset direction. The next pass constraint item, constraint adjustment direction and constraint adjustment magnitude formation method in the superior successor relationship item are read, and the correction magnitude is determined according to the current pass organizational offset degree. Then, the corresponding next pass basic control boundary is corrected according to the next pass constraint item, constraint adjustment direction and correction magnitude to determine the next pass constraint boundary.

[0101] Step 407: Correct the next-pass basic control boundary based on the successor relationship item to form the next-pass constraint boundary. Specifically, first, determine the next-pass basic control boundary that needs correction based on the next-pass constraint item in the successor relationship item; then, determine the correction direction of the basic control boundary based on the constraint adjustment direction in the successor relationship item; subsequently, determine the correction magnitude of the basic control boundary based on the constraint adjustment magnitude formation method in the successor relationship item and the current pass organization offset degree. Correct the next-pass basic control boundary according to the correction direction and correction magnitude to obtain the next-pass constraint boundary. The next-pass constraint boundary includes the constraint item name, constraint boundary range, correction direction, correction magnitude, and corresponding organization offset basis.

[0102] Step 408: Process the next pass constraint boundaries corresponding to multiple organizational offset bases; specifically, if the current pass organizational offset result includes a primary offset base and an auxiliary offset base, then find the successor relationship item based on the primary offset base and the auxiliary offset base respectively, and form the corresponding next pass constraint boundary respectively; if multiple organizational offset bases correspond to the same next pass constraint item, then perform intersection processing on multiple next pass constraint boundaries under the same constraint item, and take the range after intersection processing as the final constraint boundary of the constraint item.

[0103] If no executable range is found after intersection processing, the constraint boundary corresponding to the main offset is read first, and the intersection of this constraint boundary with the executable range of the thermal processing execution equipment is performed. If the intersection exists, the intersection range is determined as the final constraint boundary of the constraint item, and the constraint boundary corresponding to the auxiliary offset is transferred to the subsequent pass for further processing. If the intersection still does not exist, the constraint item is determined as an unexecutable constraint item, and the process returns to step 405 to reread the basic control boundary of the next pass. At the same time, the executable range of the thermal processing execution equipment is used as the boundary limitation basis for the constraint item, and steps 406 to 408 are re-executed. If multiple tissue offset bases correspond to different next pass constraint items, the next pass constraint boundary corresponding to each constraint item is determined as the final constraint boundary.

[0104] Step 409: Generate the next pass's organizational acceptance control constraints. Specifically, the final constraint boundaries, organizational offset basis, correction direction, and correction magnitude corresponding to each constraint item are organized according to the next pass's constraint items to form the next pass's organizational acceptance control constraints. The next pass's organizational acceptance control constraints include at least one of the following: next pass starting temperature constraint, deformation amount allocation constraint, deformation speed constraint, heat preservation and cooling constraint, and equipment execution timing constraint. The next pass's organizational acceptance control constraints are used in step 50 to screen candidate control parameters for the next pass.

[0105] In some implementations, the following example is used to illustrate the process of generating the next organizational control constraint: For example, the primary offset criterion in the current pass's microstructure offset result is that the recrystallization state quantity in the central calculation zone is lower than the microstructure judgment boundary, while the secondary offset criterion is that the grain size characterization quantity in the central calculation zone exceeds the microstructure judgment boundary. In the preset microstructure continuity relationships, the continuity items corresponding to the recrystallization state quantity being lower than the boundary include: microstructure state quantity type is recrystallization state quantity, microstructure offset direction is lower than the boundary, and the next pass constraint items include the next pass starting temperature constraint and the next pass deformation amount allocation constraint. The constraint adjustment direction is to increase the lower boundary of the starting temperature and increase the lower boundary of the deformation amount, and the constraint adjustment magnitude is determined based on the interval to which the microstructure offset degree belongs and the historical qualified batch correction records. The continuity items corresponding to the grain size characterization quantity exceeding the boundary include: microstructure state quantity type is grain size characterization quantity, microstructure offset direction is exceeding the boundary, and the next pass constraint items include the next pass heat preservation and cooling constraint. The constraint adjustment direction is to shorten the allowable heat preservation duration range or enhance the cooling constraint, and the constraint adjustment magnitude is determined based on the interval to which the microstructure offset degree belongs and the historical qualified batch correction records. After reading the next pass's basic control boundary, the starting temperature basic boundary, deformation distribution basic boundary, and insulation and cooling basic boundary are corrected according to the above-mentioned inheritance relationship items to form the corresponding next pass constraint boundaries. If the main offset basis and auxiliary offset basis both act on the same next pass constraint item, the intersection processing is performed on multiple next pass constraint boundaries corresponding to the same constraint item, and the range after the intersection processing is determined as the final constraint boundary. The next pass's organizational inheritance control constraint is formed by each final constraint boundary, the organizational offset basis, the correction direction, and the correction magnitude, and is used for subsequent screening of the next pass candidate control parameter group.

[0106] Step 50: Generate the control parameter results for the next pass based on the control constraints of the next pass organization. Specific implementation steps include: Step 501: Generate candidate control parameters for the next pass. Specifically, first, read the planned ranges for the starting temperature, deformation amount, deformation speed, heat preservation and cooling, and equipment execution timing from the next pass's heat treatment process plan. Then, read the executable ranges for the starting temperature, deformation amount, deformation speed, heat preservation and cooling, and equipment execution timing from the executable range of the heat treatment equipment. Subsequently, take the intersection of each planned range with the corresponding executable range of the equipment to obtain the selectable ranges for each candidate parameter. If the planned range of a parameter item does not intersect with the executable range of the equipment, then the parameter item is determined as an unexecutable parameter item. For unexecutable parameter items, the executable range of the equipment is used as the selectable range of candidate parameters, and the deviation of the planned range corresponding to the unexecutable parameter item is used as the basis for subsequent parameter sorting. Then, continue to determine the selectable ranges of other candidate parameters.

[0107] After obtaining the selectable range of each candidate parameter, multiple candidate control parameter groups are formed based on preset parameter value intervals or historically available process schemes. The preset parameter value intervals are determined based on the control resolution of the thermal processing equipment, the allowable accuracy of the process plan, and the parameter distribution of historically qualified batches. Historically available process schemes are combinations of control parameters from subsequent passes that have been executed in historically qualified batches and whose final organizational test results fall within the organizational determination boundary corresponding to the target product's organizational requirements.

[0108] Step 502: Determine the composition of the candidate control parameter groups. Specifically, each candidate control parameter group represents a control scheme that can be directly used in the next execution. Each candidate control parameter group includes at least one of the following: candidate starting temperature, candidate deformation amount, candidate deformation rate, candidate heat preservation and cooling conditions, and candidate equipment execution sequence. The candidate starting temperature is derived from the selectable range of starting temperature; the candidate deformation amount is derived from the selectable range of deformation amount; the candidate deformation rate is derived from the selectable range of deformation rate; the candidate heat preservation and cooling conditions are derived from the selectable range of heat preservation and cooling conditions; and the candidate equipment execution sequence is derived from the selectable range of equipment execution sequence. The parameter items maintain a correspondence under the same execution scheme, and parameter items are not extracted from different candidate control parameter groups and reassembled.

[0109] Step 503: Select candidate control parameter groups based on the next pass's organizational acceptance control constraints. Specifically, read the next pass's organizational acceptance control constraints formed in step 40, and read the final constraint boundaries corresponding to each constraint item. Then, read each candidate control parameter group one by one, and compare the candidate starting temperature in the candidate control parameter group with the final constraint boundary corresponding to the next pass's starting temperature constraint, compare the candidate deformation amount with the final constraint boundary corresponding to the deformation amount allocation constraint, compare the candidate deformation speed with the final constraint boundary corresponding to the deformation speed constraint, compare the candidate heat preservation and cooling conditions with the final constraint boundary corresponding to the heat preservation and cooling constraint, and compare the candidate equipment execution timing with the final constraint boundary corresponding to the equipment execution timing constraint. If any parameter item is not within the corresponding final constraint boundary, the candidate control parameter group is removed; if all parameter items are within the corresponding final constraint boundaries, the candidate control parameter group is retained.

[0110] Step 504: Generate candidate control parameter group screening results. Specifically, the retained candidate control parameter groups are organized according to their identifiers to form the candidate control parameter group screening results. The candidate control parameter group screening results include the candidate control parameter group identifier, candidate starting temperature, candidate deformation amount, candidate deformation rate, candidate heat preservation and cooling conditions, candidate equipment execution sequence, and the corresponding organizational acceptance control constraints. If the candidate control parameter group screening results contain only one candidate control parameter group, that candidate control parameter group is determined as the next pass control parameter result.

[0111] Step 505: When the candidate control parameter group screening result contains multiple candidate control parameter groups, determine the control parameter result for the next pass according to the preset parameter selection rules. Specifically, first read the main offset basis in the current pass organization offset result, and then determine whether each candidate control parameter group corresponds to the main offset basis. If the candidate control parameter group satisfies the final constraint boundary generated by the main offset basis, then determine that the candidate control parameter group corresponds to the main offset basis, and sort the candidate control parameter groups corresponding to the main offset basis first.

[0112] For each candidate control parameter group that corresponds to the main offset basis, the equipment execution margin is calculated. The equipment execution margin is determined based on the distance between the parameter values ​​in the candidate control parameter group and the executable boundary of the hot processing equipment. Candidate control parameter groups whose equipment execution margins are within the preset executable range are ranked first. For candidate control parameter groups whose equipment execution margins are all within the preset executable range, the degree of deviation between the candidate control parameter group and the next hot processing process plan is calculated. The degree of deviation is determined based on the difference between the parameter values ​​in the candidate control parameter group and the basic parameter values ​​in the next hot processing process plan. The candidates are ranked from smallest to largest, and the candidate control parameter group ranked first is determined as the control parameter result for the next process.

[0113] Step 506: When no available candidate control parameter group exists after screening, determine the processing direction for regenerating candidate control parameters; specifically, read the parameter items that failed the final constraint boundary. If the next pass starting temperature constraint is not passed, regenerate the candidate starting temperature or candidate heat preservation and cooling conditions; if the deformation amount allocation constraint is not passed, regenerate the candidate deformation amount; if the deformation speed constraint is not passed, regenerate the candidate deformation speed; if the heat preservation and cooling constraint is not passed, regenerate the candidate heat preservation and cooling conditions; if the equipment execution timing constraint is not passed, regenerate the candidate equipment execution timing.

[0114] After regeneration, continue with steps 503 to 505. If the regenerated candidate control parameter group still has no usable results, read the parameter items that failed the final constraint boundary and return to step 408 to redetermine the final constraint boundary of the corresponding constraint item. When redetermining the final constraint boundary, use the executable range of the hot processing execution equipment as the boundary limit basis and retain the constraint requirements corresponding to the main offset basis. After redetermining the final constraint boundary, execute steps 501 to 505 again to determine the control parameter results for the next pass.

[0115] Step 507: Output the control parameter results for the next pass; specifically, the control parameter results for the next pass include at least one of the following determined for the next pass execution from the candidate control parameter group: starting temperature, deformation amount, deformation speed, heat preservation and cooling conditions, and equipment execution sequence, and include the corresponding candidate control parameter group identifier and the next pass organizational acceptance control constraints satisfied; the control parameter results for the next pass are used in step 60 to control the hot processing execution equipment to execute the next hot processing.

[0116] In some implementations, the following example is used to illustrate the process of screening candidate control parameter groups and forming the control parameter results for the next pass: For example, the control constraints for the next stage of the process include the final constraint boundary of the next stage's starting temperature, the final constraint boundary of the next stage's deformation amount, and the final constraint boundary of the next stage's deformation rate. Based on the next stage's hot processing process plan and the executable range of the hot processing equipment, three candidate control parameter groups are generated: the first candidate control parameter group includes a candidate starting temperature of 1060℃, a candidate deformation amount of 10mm, and a candidate deformation rate of 0.35mm / s; the second candidate control parameter group includes a candidate starting temperature of 1080℃, a candidate deformation amount of 14mm, and a candidate deformation rate of 0.42mm / s; and the third candidate control parameter group includes a candidate starting temperature of 1100℃, a candidate deformation amount of 16mm, and a candidate deformation rate of 0.35mm / s. 60mm / s; The candidate starting temperature, candidate deformation amount, and candidate deformation rate in each candidate control parameter group are compared with their corresponding final constraint boundaries. If the candidate deformation amount in the first candidate control parameter group is not within the final constraint boundary of the deformation amount in the next pass, the first candidate control parameter group is eliminated. If all parameters in the second candidate control parameter group are within the corresponding final constraint boundary, the second candidate control parameter group is retained. If the candidate deformation rate in the third candidate control parameter group is not within the final constraint boundary of the deformation rate in the next pass, the third candidate control parameter group is eliminated. If only the second candidate control parameter group is retained after screening, it is determined as the control parameter result for the next pass. If multiple candidate control parameter groups are retained after screening, the candidate control parameter group with the highest ranking is determined based on whether it meets the final constraint boundary generated by the main offset criteria, the equipment execution margin, and the degree of deviation from the next thermal processing plan. The candidate control parameter group with the highest ranking is then determined as the control parameter result for the next pass.

[0117] Step 60: Output the control parameter results for the next pass and repeat the multi-pass hot working process. Specific steps include: Step 601: Output the next pass control parameter results to the hot processing execution equipment. Specifically, first read at least one of the following parameters from the next pass control parameter results: initial temperature, deformation amount, deformation speed, heat preservation and cooling conditions, and equipment execution sequence. Then, convert these parameters into execution parameters recognizable by the hot processing execution equipment. The hot processing execution equipment includes at least one of the following: heating furnace, forging equipment, rolling equipment, extrusion equipment, ring rolling equipment, die temperature control equipment, cooling equipment, or transfer equipment. Different hot processing execution equipment reads the next pass control parameter results according to the parameter items corresponding to its own execution object.

[0118] Step 602: The hot processing equipment executes the next hot processing step according to the control parameters of the next step. Specifically, the heating furnace or mold temperature control equipment executes temperature control according to the starting temperature or heat preservation and cooling conditions; the forging equipment, rolling equipment, extrusion equipment, or ring rolling equipment executes deformation control according to the deformation amount, deformation speed, and equipment execution sequence; the cooling equipment or transfer equipment executes cooling or transfer control according to the heat preservation and cooling conditions and equipment execution sequence. Thus, the starting temperature, deformation amount, deformation speed, heat preservation and cooling conditions, or equipment execution sequence of the next step are constrained by the next step's organizational structure inheritance control constraints generated by the current step's organizational structure evolution results and the current step's organizational structure offset results.

[0119] Step 603: During the next execution cycle, collect the heat treatment process data for the next cycle. Specifically, following the same data collection method as in Step 10, collect the temperature process data, deformation execution data, and load response data for the next cycle, and collect the end-of-cycle status data after the next cycle is completed. The end-of-cycle status data includes at least one of the following: billet size data, billet surface thermal state data, billet shape deviation data, cooling dwell time data, or transfer waiting time data after the next cycle.

[0120] Step 604: Redefine the completed next pass as the new current pass; specifically, use the hot working process data of the next pass as the hot working process data of the new current pass, use the end state data of the next pass as the end state data of the new current pass, use the state of the alloy billet after the completion of the next pass as the billet state before the start of the new current pass, and define the subsequent passes after the next pass as the new next pass; after redefining, continue to execute steps 20 to 60 to generate the new current pass microstructure evolution results, the new current pass microstructure offset results, the new next pass microstructure inheritance control constraints, and the new next pass control parameter results.

[0121] Step 605: Determine whether to stop the loop processing. Specifically, first read the preset number of passes in the multi-pass heat treatment process plan and count the number of completed passes. If the number of completed passes has not reached the preset number of passes, return to steps 603 to 604 to continue processing subsequent passes. If the number of completed passes has reached the preset number of passes, read the dimensional status in the final pass end status data and read the dimensional judgment boundary corresponding to the target product dimensional requirements. Compare the dimensional status with the dimensional judgment boundary to determine the dimensional end judgment result. At the same time, based on the final pass microstructure evolution result, read the microstructure state quantity corresponding to the final pass and read the microstructure judgment boundary in the final target microstructure condition. Compare the microstructure state quantity corresponding to the final pass with the microstructure judgment boundary in the final target microstructure condition to determine the microstructure end judgment result.

[0122] When the dimension completion judgment result indicates that the dimension state is within the dimension judgment boundary, and the structure completion judgment result indicates that the structure state quantity is within the structure judgment boundary of the final target structure condition, the loop processing stops; when the dimension completion judgment result indicates that the dimension state is not within the dimension judgment boundary, or the structure completion judgment result indicates that the structure state quantity is not within the structure judgment boundary of the final target structure condition, the dimension state or structure state quantity that is not within the boundary is read, and its corresponding deviation direction and deviation degree are determined; then, based on the deviation direction and deviation degree, the allowable supplementary pass type is read from the multi-pass hot processing process plan, the supplementary pass type includes at least one of supplementary heating pass, supplementary deformation pass, supplementary heat preservation pass, or supplementary cooling pass; then, the supplementary pass type, the corresponding deviation object, the deviation direction, and the deviation degree are determined as the supplementary pass control input, and the supplementary pass control input is used as the basis for the new next pass processing, and steps 20 to 60 are continued.

[0123] Step 606: Output the multi-pass hot working control results. Specifically, after stopping the loop processing, output the hot working process data corresponding to each pass, the microstructure evolution result of the current pass, the microstructure offset result of the current pass, the microstructure inheritance control constraints for the next pass, and the control parameter results for the next pass. The multi-pass hot working control results are used to record the microstructure inheritance process between each pass and are used for microstructure inheritance control in subsequent hot working processes of similar alloy billets.

[0124] Example 2 See Figure 2As shown, this embodiment provides a digital twin-driven alloy hot working control system. The system is used to execute a digital twin-driven alloy hot working control method as described in Embodiment 1. The system includes a data acquisition module, a microstructure generation module, an offset determination module, a constraint generation module, a parameter generation module, and a loop control module. Each module sequentially performs corresponding data processing in steps 10, 20, 30, 40, 50, and 60. The system includes: The data acquisition module is used to acquire the hot working process data and the end status data of the current pass of the alloy billet. The organization generation module is used to generate the organization evolution result of the current pass based on the heat treatment process data and the current pass end status data, using a digital twin organization evolution model. The offset determination module is used to compare the organization state quantity in the current track organization evolution result with the organization determination boundary in the target organization condition corresponding to the current track item by item to determine the organization offset result of the current track. The constraint generation module is used to generate the next course organization succession control constraint based on the current course organization evolution result and the current course organization offset result. The parameter generation module is used to filter and combine candidate control parameters for the next pass based on the next pass organization acceptance control constraints, and generate the control parameter results for the next pass. The loop control module is used to output the control parameter results of the next pass to the heat processing execution device to execute the next heat processing pass. After the next pass is executed, the next pass is used as the new current pass to continue processing until the multi-pass heat processing process is completed.

[0125] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope defined in the claims.

Claims

1. A digital twin-driven method for controlling alloy hot working, characterized in that, include: Obtain the hot working process data and the end status data of the current pass of the alloy billet; Based on the heat treatment process data and the current pass end status data, the current pass organization evolution result is generated using a digital twin organization evolution model; Compare the organization state quantities in the current track organization evolution result with the organization determination boundaries in the target organization conditions corresponding to the current track item by item to determine the organization offset result of the current track. Based on the current course organization evolution result and the current course organization offset result, the next course organization succession control constraint is generated; Based on the next-stage organizational acceptance control constraints, the candidate control parameters for the next stage are screened and combined to generate the control parameter results for the next stage. The control parameter results for the next pass are output to the heat processing execution device to execute the next heat processing pass. After the next pass is completed, the next pass is used as the new current pass to continue processing until the multi-pass heat processing process is completed.

2. The digital twin-driven alloy hot working control method according to claim 1, characterized in that, Methods for obtaining hot working process data and current pass end status data include: Temperature process data, deformation execution data, and load response data are collected between the start and end times of the current pass, and the settling time is determined based on the time of change of stroke position or the time of change of equipment control command in the deformation execution data. The temperature data between adjacent settling times is determined as the representative temperature value, the load data between adjacent settling times is determined as the representative load value, and the change in stroke position, change in compression, or change in speed between adjacent settling times is determined as the deformation execution amount. After the current pass is completed, the end status data of the current pass is collected, and a correspondence is established between the end status data of the current pass and the hot working process data according to the same alloy billet and the same current pass.

3. The digital twin-driven alloy hot working control method according to claim 1, characterized in that, Methods for generating current track evolution results using digital twin organizational evolution models include: Based on the geometric description of the alloy billet and the deformation contact area of ​​the current pass, the alloy billet is divided into several calculation zones, and the representative temperature value, representative load value, deformation execution amount and deformation speed corresponding to each calculation zone are determined. According to the time period sequence, the temperature representative value, deformation execution amount, deformation speed and load representative value corresponding to each calculation partition are converted into segmented thermal deformation state, and the cumulative thermal deformation state is determined based on the segmented thermal deformation state of each time period. By calculating the organizational state quantities in the digital twin organizational evolution model, the cumulative thermal deformation state is converted into the organizational state quantities corresponding to each calculation partition, thus forming the organizational evolution result for the current pass.

4. The digital twin-driven alloy hot working control method according to claim 1, characterized in that, Methods for determining the organization offset result of the current track number include: Based on the alloy material type, target product microstructure requirements, multi-pass hot working process plan, and current pass sequence position, determine the target microstructure conditions corresponding to the current pass, and determine the microstructure judgment boundary for each microstructure state quantity in the microstructure evolution result of the current pass. When the organization's determination boundary is within the allowable value range, the organization's state quantity is compared with the lower and upper boundaries of the allowable value range to determine the individual comparison results for items that are below the boundary, above the boundary, or have not formed an offset. When the organizational judgment boundary is within the allowable level range, the organizational state quantity is converted into the corresponding level and compared with the allowable level range to determine the individual comparison result of insufficient level, excessive level, or no offset. Based on the comparison results of each item, generate the current track organization offset result, including the organization offset direction and the organization offset degree.

5. The digital twin-driven alloy hot working control method according to claim 1, characterized in that, Methods for generating the next level of organizational control constraints include: Read the organization status quantity in the organization evolution result of the current track, and read the type of offset organization status quantity, offset calculation partition, organization offset direction and organization offset degree in the organization offset result of the current track; Based on the preset organizational succession relationship, find the succession relationship item corresponding to the organizational offset result of the current pass. The succession relationship item includes organizational state quantity type, organizational offset direction, organizational offset degree range, next pass constraint item, constraint adjustment direction, and constraint adjustment magnitude formation method. Read the next basic control boundary, and modify the next basic control boundary according to the next constraint item, constraint adjustment direction and constraint adjustment range formation method in the inheritance relationship item, and generate the next organizational inheritance control constraint.

6. The digital twin-driven alloy hot working control method according to claim 5, characterized in that, Methods for generating the control parameter results for the next pass include: Based on the next heat treatment process plan and the executable range of the heat treatment equipment, multiple candidate control parameter groups for the next process are generated. Each candidate control parameter group for the next process includes at least one of the following: candidate starting temperature, candidate deformation amount, candidate deformation rate, candidate heat preservation and cooling conditions, and candidate equipment execution sequence. Compare each parameter item in the next candidate control parameter group with the corresponding final constraint boundary in the next organizational acceptance control constraint, and eliminate any next candidate control parameter group whose parameter item is not within the corresponding final constraint boundary. Among the retained candidate control parameters for the next pass, the control parameter results for the next pass are determined based on whether they meet the final constraint boundary generated by the current pass's organization offset results, the equipment execution margin, and the degree of deviation from the next pass's thermal processing plan.

7. The digital twin-driven alloy hot working control method according to claim 1, characterized in that, The method for continuing processing the next pass as the new current pass includes: During the next execution, collect the heat treatment process data of the next step, and after the next execution is completed, collect the end status data of the next step. The hot working process data of the next pass is used as the hot working process data of the new current pass, the end state data of the next pass is used as the end state data of the new current pass, the state of the alloy billet after the completion of the next pass is used as the billet state before the start of the new current pass, and the subsequent passes after the next pass are defined as the new next pass. Continue generating new current pass organization evolution results, new current pass organization offset results, new next pass organization acceptance control constraints, and new next pass control parameter results, until the number of completed passes reaches the preset number of passes, and the dimensional state of the final pass is within the dimensional judgment boundary corresponding to the target product dimensional requirements, and the organization state quantity of the final pass is within the organization judgment boundary in the final target organization conditions.

8. The digital twin-driven alloy hot working control method according to claim 3, characterized in that, The method for establishing the digital twin organizational evolution model includes: Acquire alloy material type, initial geometric data of alloy billet, multi-pass hot working process plan, range of hot working equipment, historical hot working process data, historical pass end status data and historical microstructure detection data; A geometric description of the billet is established based on the initial geometric data of the alloy billet and the multi-pass hot working process plan; a material thermophysical property relationship is established based on the alloy material type and the historical hot working process data; a hot deformation calculation relationship is established based on the historical hot working process data; and a microstructure state quantity calculation relationship is established based on the historical hot working process data, the historical pass end state data, and the historical microstructure detection data. The input-output relationship between the billet geometry description, the material thermophysical property relationship, the thermal deformation calculation relationship, and the microstructure state quantity calculation relationship is established to form the digital twin microstructure evolution model.

9. A digital twin-driven alloy hot working control method according to claim 6, characterized in that, Methods for determining the final constraint boundaries also include: When the current course organization offset result includes multiple organization offset bases, the corresponding successor relationship item is found according to each organization offset base, and the corresponding next course constraint boundary is formed respectively. When multiple next-level constraint boundaries correspond to the same next-level constraint term, the intersection of the multiple next-level constraint boundaries is processed, and the range after the intersection process is determined as the final constraint boundary of the corresponding next-level constraint term. When there is no executable range for the hot processing equipment after the intersection processing, the next pass constraint boundary formed by the organization offset corresponding to the target product organization requirements in the current pass organization offset result is intersected with the executable range of the hot processing equipment, and the range after the intersection processing is determined as the final constraint boundary of the corresponding next pass constraint item.

10. A digital twin-driven alloy hot working control system, used to implement the digital twin-driven alloy hot working control method according to any one of claims 1-9, characterized in that, The system includes: The data acquisition module is used to acquire the hot working process data and the end status data of the current pass of the alloy billet; The organization generation module is used to generate the organization evolution result of the current pass based on the heat treatment process data and the current pass end status data, using a digital twin organization evolution model. The offset determination module is used to compare the organization state quantity in the current track organization evolution result with the organization determination boundary in the target organization condition corresponding to the current track item by item to determine the organization offset result of the current track. The constraint generation module is used to generate the next course organization succession control constraint based on the current course organization evolution result and the current course organization offset result. The parameter generation module is used to filter and combine candidate control parameters for the next pass based on the next pass organization acceptance control constraints, and generate the control parameter results for the next pass. The loop control module is used to output the control parameter results of the next pass to the heat processing execution device to execute the next heat processing pass. After the next pass is executed, the next pass is used as the new current pass to continue processing until the multi-pass heat processing process is completed.

Images