A method for controlling the production of a composite metal sheet

Abstract

The present application belongs to the technical field of composite metal sheet rolling production control, and specifically discloses a composite metal sheet rolling production control method. The method collects the upper and lower surface temperatures of the blank and obtains the current rolling temperature gradient. If the current temperature gradient exceeds the allowable deviation range, the reduction rate compensation is directly triggered. Otherwise, based on the current temperature gradient and material combination, the interlayer deformation ratio deviation is retrieved from the historical rolling log to construct a deviation value sequence. The deviation value sequence is subjected to piecewise linear regression to determine its change trend direction. The temperature gradient deviation direction and the change trend direction are used as double judgment variables. When the directions are consistent, the reduction rate compensation is triggered, and an adjustment command is generated to drive the reduction mechanism to execute the adjustment. The present application can accurately distinguish between temperature deviations that require compensation and those that do not, thereby achieving accurate triggering of the reduction rate compensation and enabling the final setting of the subsequent reduction rate to meet the actual needs.

Description

Technical Field

[0001] This invention belongs to the field of composite metal sheet rolling production control technology, and more specifically, relates to a composite metal sheet rolling production control method. Background Technology

[0002] In the production of composite metal sheets, different metals have different deformation resistance at the same temperature, which leads to uncoordinated deformation when rolled at the same temperature. Therefore, it is usually necessary to perform differential temperature rolling to actively achieve coordinated deformation by utilizing the temperature difference, thus highlighting the importance of differential temperature rolling control.

[0003] Currently, heterothermal rolling typically involves heating the difficult-to-deform metal layer separately using a heating device. However, both separate heating and electromagnetic induction heating commonly suffer from spatial separation between heating and rolling, leading to significant heat loss during the transport of the heated metal. Furthermore, when dissimilar metals come into close contact, heat from the high-temperature side is rapidly conducted to the low-temperature side, causing the expected temperature difference to decrease drastically before entering the rolling mill.

[0004] Existing technologies, such as the Chinese invention patent application with application number 202011119639.3, disclose a composite strip differential temperature rolling method and equipment based on the action of friction rollers. It focuses on the online establishment of temperature difference before rolling, and uses the frictional heat generation effect between high-speed rotating friction rollers and the surface of metal strips to achieve differential temperature rise of dissimilar metals. The rotation speed of the friction rollers is adjusted in real time by temperature detection instrument to maintain the target temperature difference.

[0005] Existing technology, such as the single-roll corrugated and flat rolling equipment and method for high-thickness layered metal composite plates with application number 202410292094.8, focuses on optimizing the deformation path of high-thickness composite plates. It adopts segmented electromagnetic induction heating combined with two-pass continuous rolling of corrugated and flat rolls. Through secondary heating, the billet is restored to the target temperature before entering the second rolling unit, which solves the problems of high energy consumption for overall heating of thick plates and large heat loss during roller conveying.

[0006] The aforementioned existing technologies all limit the solution to the temperature deviation problem to improvements in the heating process. However, after being heated by the heating unit, the billet must pass through an intermediate conveyor roller table before entering the rolling unit. During this conveying process, heat continuously dissipates from the billet surface to the environment, and the different thermal conductivity coefficients of dissimilar metals lead to inconsistent temperature drop rates on both sides of the interface, making it easy for the actual temperature of the billet entering the rolling unit to deviate from the preset target temperature. Current processes typically set the rolling reduction rate according to the target temperature, but when the actual temperature gradient deviates from the target value, the preset reduction rate can no longer match the interlayer deformation coordination at the current actual temperature of the billet.

[0007] Furthermore, in differential temperature rolling, heat loss during the conveying process causes the actual temperature gradient to deviate from the target value. However, there is not a one-to-one correspondence between temperature deviation and deformation coordination failure; at the same degree of deviation, deformation coordination may be normal at times and uncontrolled at others. If real-time feedback adjustment is only based on the current temperature deviation value, it is still impossible to distinguish between temperature deviations that lead to deformation coordination failure and temperature deviations that are still within the acceptable tolerance range. This results in the final set value of the reduction rate deviating from the actual requirement, thus failing to accurately suppress deformation coordination failure. Summary of the Invention

[0008] In view of this, in order to solve the above problems, a production control method for composite metal sheet rolling is proposed.

[0009] The objective of this invention can be achieved through the following technical solution: This invention provides a method for controlling the production of composite metal sheet rolling. The method includes: when the billet enters the rolling unit via the intermediate conveyor roller, the temperature of the upper and lower surfaces of the billet is collected by a temperature detection array arranged on the intermediate conveyor roller. The temperature values ​​of the upper and lower surfaces at corresponding positions along the conveying direction are subtracted point by point, and the maximum temperature difference is taken as the current rolling temperature gradient.

[0010] If the current rolling temperature gradient exceeds the allowable deviation range, trigger reduction rate compensation; otherwise, execute the following sub-steps: based on the current rolling temperature gradient and the material combination of the current billet, retrieve the interlayer deformation ratio deviation of each historical batch corresponding to the temperature gradient from the historical rolling log, and construct a deviation value sequence.

[0011] Piecewise linear regression is performed on the deviation value sequence, and the direction of the trend of the deviation value sequence is determined based on the sign distribution of the regression slope of each sequence segment.

[0012] The deviation direction between the rolling temperature gradient and the target temperature gradient is used as the first judgment variable, and the direction of the change trend is used as the second judgment variable. When the directions of the two judgment variables are consistent, the reduction rate compensation is triggered.

[0013] Generate adjustment commands and drive the pressing mechanism to perform adjustments.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention obtains the actual temperature gradient before the entrance of the rolling unit by setting up a temperature detection array in the intermediate conveying roller table and making point-by-point difference in the temperature of the upper and lower surfaces of the billet. It intuitively displays the temperature distribution state before the entrance of the rolling unit, thereby breaking the limitation of the separation of temperature information between the heating link and the rolling link in the prior art, solving the problem that the actual temperature gradient cannot be known due to heat loss during the billet conveying process, and providing a reliable data basis for subsequent control.

[0015] (2) This invention retrieves the deformation ratio deviation of each historical batch under the same conditions from the historical records based on the current material type and the measured temperature gradient, and establishes a correlation between the current temperature deviation state and the deformation coordination performance under the same historical working conditions. This allows the risk of deformation coordination failure to be predicted based on the direction of the current temperature gradient approaching the boundary of the allowable range. This overcomes the defect of the prior art that cannot distinguish between temperature deviations that need compensation and those that do not, and ensures the accuracy of the reduction rate adjustment decision, so that the subsequent reduction rate setting can meet the actual needs.

[0016] (3) This invention performs piecewise linear regression on the deviation value sequence and determines the direction of the trend of the deviation value sequence based on the sign distribution of the regression slope of each sequence segment. It retains the local evolution characteristics of the deviation value sequence on the time axis, making the trend judgment result closer to the real process. It provides accurate trend information for subsequent consistency judgment with the temperature gradient deviation direction and avoids triggering incorrect pressure reduction compensation due to trend misjudgment.

[0017] (4) This invention uses the direction of the change trend of the deviation value sequence and the direction of the temperature gradient deviation as dual judgment variables for reduction rate compensation. Compensation is triggered only when the two directions are consistent, avoiding frequent invalid adjustments caused by natural temperature fluctuations or measurement noise. It solves the problem of mismatch between the preset reduction rate and the interlayer deformation coordination under actual temperature conditions, and realizes accurate triggering of reduction rate compensation. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall implementation steps of the method of the present invention;

[0019] Figure 2 This is a schematic diagram of the pressure reduction rate compensation trigger judgment process of the present invention;

[0020] Figure 3 This is a schematic diagram of the piecewise linear regression results of the deviation value sequence of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Currently, the current temperature gradient deviation is typically used as the sole criterion. When this deviation exceeds a preset allowable range, reduction rate compensation is triggered. However, the loss of control over interlayer deformation coordination is influenced by both the direction of temperature gradient deviation and the historical trend of deviation evolution. When the actual temperature deviation caused by heat loss during billet transport is consistent with the historical trend of deformation coordination deviation under the same working conditions, existing technology cannot distinguish whether this temperature deviation is a systemic cause of deformation coordination loss of control. It may misjudge it as a normal temperature fluctuation and directly trigger reduction rate compensation, resulting in ineffective adjustment or misaligned compensation timing.

[0023] Based on this, the present invention targets the heterothermal rolling process of composite metal sheets. By identifying the co-directional coupling relationship between the current rolling temperature gradient deviation direction and the historical interlayer deformation ratio deviation trend direction, when the temperature gradient is close to the allowable boundary but has not yet exceeded the allowable range, it determines that the temperature deviation has had a systematic impact on deformation coordination based on the co-directional coupling relationship and triggers the reduction rate compensation in advance, thereby realizing the deviation trend-driven predictive online control.

[0024] Example 1

[0025] Please refer to the details. Figure 1 As shown, the present invention provides a method for controlling the production of composite metal sheet rolling, the method comprising:

[0026] S1. When the billet enters the rolling unit via the intermediate conveyor rollers, the temperature of the upper and lower surfaces of the billet is collected by the temperature detection array installed on the intermediate conveyor rollers. The temperature values ​​of the upper and lower surfaces at corresponding positions along the conveying direction are subtracted point by point, and the maximum temperature difference is taken as the current rolling temperature gradient.

[0027] The billet is a composite billet formed by stacking two or more dissimilar metals, including a difficult-to-deform metal layer and a easily deformable metal layer. The material combinations of the composite metal sheets include, but are not limited to, titanium / aluminum composite, molybdenum / copper composite, titanium / copper / titanium composite, etc.

[0028] Furthermore, the specific arrangement of the temperature detection array is as follows: the temperature detection array includes an upper surface temperature detection subarray and a lower surface temperature detection subarray. The upper surface temperature detection subarray is arranged above the intermediate conveyor rollers, and the lower surface temperature detection subarray is arranged below the gap between adjacent rollers. Each temperature detection subarray consists of multiple non-contact infrared temperature sensors arranged at equal intervals along the width of the billet. The non-contact infrared temperature sensors are suspended above or below the surface of the billet, maintaining a preset distance (e.g., 50mm to 200mm) to avoid contact with the moving billet. The spacing between adjacent sensors is determined according to the width of the billet, such as ensuring that the adjacent spacing does not exceed one-tenth of the billet width, to guarantee temperature distribution covering the entire width of the billet.

[0029] Considering that in heterogeneous temperature rolling, as the billet is heated and enters the rolling unit via the intermediate conveyor rollers, heat continuously dissipates from the billet surface to the environment. Furthermore, the different thermal conductivity coefficients of the different metal layers lead to inconsistent temperature drop rates on both sides of the interface. Once the interlayer temperature difference deviates from the target value, the ratio of the deformation resistance of the two metal layers changes, resulting in uncontrolled deformation coordination. Therefore, it is necessary to obtain the actual interlayer temperature difference distribution of each cross-section of the billet before entering the rolling unit to provide a basis for subsequent compensation decisions. The cross-section with the largest temperature difference has the highest risk of deformation coordination in the entire billet; using it as the control benchmark ensures that the coordination of the most unfavorable cross-section is guaranteed. Based on this, this embodiment obtains the current rolling temperature gradient in the following way: when the billet is conveyed along the intermediate conveyor rollers towards the rolling unit, each sensor in the upper surface temperature detection subarray synchronously collects the upper surface temperature value of the billet at the corresponding width position, and each sensor in the lower surface temperature detection subarray synchronously collects the lower surface temperature value of the billet at the corresponding width position. The sampling frequency is matched with the billet conveying speed, so that several sampling sections are formed along the billet conveying direction over the entire length of the billet. Each sampling section contains multiple upper surface temperature values ​​and lower surface temperature values ​​distributed along the width direction.

[0030] For each sampling section, the temperature values ​​of the upper and lower surfaces at the same width location are subtracted point by point to obtain the temperature difference value at that width location. The average temperature difference value is then calculated for all width locations within the sampling section. This method is repeated for all sampling sections along the entire length of the billet to obtain a temperature difference sequence distributed along the billet conveying direction. The maximum temperature difference value is selected from this sequence as the current rolling temperature gradient. This maximum temperature difference characterizes the maximum interlayer temperature difference between the upper and lower surfaces of the billet in this batch before entering the rolling unit, reflecting the temperature gradient state at the location where heat loss is most severe during billet conveying.

[0031] This step involves deploying a temperature detection array on the intermediate conveyor rollers to obtain the actual temperature gradient before the rolling unit entrance by calculating the temperature difference between the upper and lower surfaces of the billet point by point. This provides a visual representation of the temperature distribution before the rolling unit entrance, thus breaking the limitation of the separation of temperature information between the heating and rolling stages in the existing technology. It also solves the problem that the actual temperature gradient cannot be known due to heat loss during billet conveying, providing a reliable data foundation for subsequent control.

[0032] S2. Determine whether the pressure rate compensation is triggered.

[0033] Specifically, please refer to Figure 2As shown, the specific steps for determining whether to trigger reduction rate compensation are as follows: First, determine if the current rolling temperature gradient exceeds the allowable deviation range. If it does, trigger reduction rate compensation. The allowable deviation range refers to the acceptable fluctuation range of the target temperature gradient during production settings. This range is preset by the operator based on the product quality requirements of the current billet material combination. Different material combinations correspond to different allowable temperature gradient deviation ranges. "Exceeding the allowable deviation range" means exceeding the upper limit of the allowable deviation range or falling below the lower limit of the allowable deviation range.

[0034] If the allowable deviation range is not exceeded, the following sub-steps are performed: A1. Based on the current rolling temperature gradient and the current billet material combination, retrieve the interlayer deformation ratio deviation of each historical batch corresponding to the temperature gradient from the historical rolling log, and construct the deviation value sequence.

[0035] The interlayer deformation ratio refers to the ratio of the relative compressive deformation degrees of each metal layer along the thickness direction after rolling deformation of the composite metal sheet. Specifically, for a two-layer composite metal sheet composed of a difficult-to-deform metal layer and a easily deformable metal layer, the interlayer deformation ratio is defined as the ratio of the actual reduction rate of the difficult-to-deform metal layer to the actual reduction rate of the easily deformable metal layer. In this invention, the difficult-to-deform metal refers to the metal layer with relatively large deformation resistance under the same temperature conditions, and the easily deformable metal refers to the metal layer with relatively small deformation resistance under the same temperature conditions. For example, in a titanium / aluminum composite plate, titanium is a difficult-to-deform metal and aluminum is an easily deformable metal. In a molybdenum / copper composite plate, molybdenum is a difficult-to-deform metal and copper is an easily deformable metal. In a titanium / copper / titanium composite plate, titanium is a difficult-to-deform metal layer and copper is an easily deformable metal layer.

[0036] Understandably, the historical rolling log records the process parameters and rolling result data of each historical batch of composite metal sheets. Among them, the process parameters include at least the rolling temperature gradient of each historical batch, and the rolling result data includes at least the actual value of the interlayer deformation ratio of each historical batch and the deviation from the target value.

[0037] A2. Perform piecewise linear regression on the deviation value sequence, and determine the direction of the trend of the deviation value sequence based on the sign distribution of the regression slope of each sequence segment.

[0038] Understandably, piecewise linear regression is used to divide the deviation value sequence into multiple continuous intervals, i.e. sequence segments, along the time axis, and to perform linear regression on each sequence segment to obtain the regression slope of each sequence segment.

[0039] After piecewise linear regression, the regression slopes corresponding to each sequence segment represent the rate and direction of change of the deviation value within the corresponding sequence segment as the historical batch sequence changes: a positive slope indicates that the deviation value within the segment is increasing, a negative slope indicates that the deviation value within the segment is decreasing, and the larger the absolute value of the slope, the faster the rate of change of the deviation value within the segment.

[0040] A3. The deviation direction between the rolling temperature gradient and the target temperature gradient is used as the first judgment variable, and the direction of the change trend is used as the second judgment variable. When the directions of the two judgment variables are consistent, the reduction rate compensation is triggered.

[0041] The method for determining the direction of the deviation between the rolling temperature gradient and the target temperature gradient is as follows: taking the target temperature gradient as the zero point, calculate the difference between the rolling temperature gradient and the target temperature gradient. When the difference is positive, it is determined to be a positive deviation, that is, the actual temperature gradient is higher than the target value. When the difference is negative, it is determined to be a negative deviation, that is, the actual temperature gradient is lower than the target value.

[0042] When the deviation direction of the first judgment variable is positive and the trend direction of the second judgment variable is positively increasing, or when the deviation direction of the first judgment variable is negative and the trend direction of the second judgment variable is negatively decreasing, then the two judgment variables are determined to be in the same direction.

[0043] It should be noted that if the current temperature gradient deviation direction is the same as the evolution trend of the historical deviation sequence, it indicates that the current temperature deviation is developing in the same direction as the historical deviation, and the temperature deviation has affected the deformation compatibility. There is a correlation between the temperature gradient deviation and the loss of control of deformation compatibility, and intervention is required through compression rate compensation.

[0044] When the directions of the two judgment variables are inconsistent, the current temperature deviation does not develop in the direction of the historical deviation trend, indicating that the temperature deviation is still within the normal fluctuation range of the acceptable tolerance, and the reduction rate compensation is not triggered. The rolling unit maintains the current reduction rate setting value.

[0045] The following supplementary explanations are required during the execution of steps A1 to A3 above:

[0046] Preferably, the specific retrieval method for retrieving from the historical rolling log in step A1 is as follows: A11. Based on the material combination of the current billet, obtain the deformation resistance temperature change rate of the difficult-to-deform metal layer at the target rolling temperature as the first change rate, and at the same time obtain the deformation resistance temperature change rate of the easily-deformable metal layer at the target rolling temperature as the second change rate. Take the absolute value of the difference between the first change rate and the second change rate as the difference in deformation resistance temperature change rate.

[0047] Understandably, the deformation resistance temperature change rate is obtained as follows: At the target rolling temperature and at temperature points ±5℃, ±10℃, ±20℃, and ±30℃, isothermal constant strain rate compression tests are conducted on both the difficult-to-deform and easily-deformable metal layers using a Gleeble thermal simulation testing machine. The tests at each temperature point maintain the same strain rate (e.g., 0.1 sec) and the same strain rate loading history. From the true stress-true strain curves measured at each temperature point, the flow stress values ​​corresponding to the same strain value (e.g., true strains of 0.2, 0.3, 0.4, and 0.5) are extracted. The average value of the linear regression slope of the flow stress value with temperature change at each strain point is taken as the deformation resistance temperature change rate of the metal layer at the target rolling temperature.

[0048] A12. Obtain the critical value of interlaminar flow stress deviation allowed for coordinated deformation under the current billet material combination, and use the ratio of the critical value of interlaminar flow stress deviation to the difference in the temperature change rate of deformation resistance as the upper limit of the target temperature gradient value deviation.

[0049] Understandably, the critical value of interlayer flow stress deviation refers to the maximum allowable difference in interlayer flow stress to ensure coordinated deformation of the two metals. When the actual difference in flow stress between the two metal layers exceeds this critical value, the soft layer metal deforms excessively due to excessively low flow stress, while the hard layer metal deforms insufficiently due to excessively high flow stress, and the distribution of interlayer deformation deviates from the target ratio.

[0050] The critical value was determined as follows: For each material combination, an isothermal constant strain rate (e.g., 0.1 sec) compression test was conducted using a Gleeble thermal simulation testing machine at the target rolling temperature. The specimen was cylindrical, with dimensions of φ8mm × 12mm, and glass lubricant was applied to both ends of the specimen to reduce the influence of end friction. During the test, thermocouples were welded to the center of the specimen surface to monitor the temperature in real time, and the true stress-true strain curves of the corresponding difficult-to-deform metal layer and easy-to-deform metal layer for each specimen were recorded.

[0051] Based on this, a two-dimensional thermo-mechanical coupled finite element model of the composite metal sheet rolling process was established using finite element analysis software (such as ABAQUS / Standard or DEFORM). In the model, the true stress-true strain curves measured by the Gleeble test were used as the deformation resistance of each of the two metal layers as input data. The initial temperatures of the two metal layers were set according to the preset target temperature of heterothermal rolling, so that the difference between the two initial temperatures was equal to the target temperature gradient.

[0052] When calibrating the critical value of interlayer flow stress deviation, the true stress-true strain curve of the difficult-to-deform metal layer is kept unchanged, and an incremental scaling factor is applied to the true stress-true strain curve of the easily deformable metal layer. The initial value of the scaling factor is set to 1, that is, the original flow stress of the easily deformable metal layer is kept unchanged. The increment step size is determined according to the calibration accuracy requirements. Those skilled in the art can select it in the range of 0.005 to 0.02 according to actual requirements. The higher the accuracy requirement, the smaller the step size. When the ratio deviation of the interlayer deformation calculated under a certain scaling factor first reaches the upper limit of the allowable limit, the increment is stopped, and the interlayer flow stress difference corresponding to the scaling factor is determined as the critical value.

[0053] A13. Using the current rolling temperature gradient value as the center value and the upper limit of the target temperature gradient value deviation as the search radius, retrieve all historical batch records from the historical rolling log that have the same material combination as the current billet and whose temperature gradient values ​​fall within the search radius.

[0054] It should be noted that if a search is performed with the current rolling temperature gradient value as the center value and the upper limit of the target temperature gradient value deviation as the search radius, and no historical batch record with the same material combination as the current billet and whose temperature gradient value falls within the search radius is found, then the reduction rate compensation will not be triggered and the record will be marked.

[0055] A14. Extract the deviation between the actual and target values ​​of the interlayer deformation ratio from each retrieved historical batch record, and arrange them in chronological order to form a deviation value sequence. This deviation value sequence reflects the evolution of the deviation in interlayer deformation compatibility among historical batches under the same or similar temperature gradient conditions.

[0056] Preferably, the specific segmentation method for performing piecewise linear regression on the deviation value sequence in step A2 is as follows: First, obtain the difference between adjacent deviation values ​​in the deviation value sequence to generate an adjacent point difference sequence, and extract the points in the adjacent point difference sequence where the sign changes as candidate segmentation boundaries. These candidate segmentation boundaries represent the locations in the deviation value sequence where the trend direction may change.

[0057] Then, the number of candidate segment boundaries to be retained is determined based on the current temperature gradient state. Specifically, based on the target temperature gradient and its allowable deviation range, the upper and lower boundary values ​​of the allowable temperature gradient interval are obtained. The absolute values ​​of the differences between the current rolling temperature gradient and the upper and lower boundary values ​​are calculated respectively, and the smaller value is taken as the safety margin value of the current temperature gradient.

[0058] Subsequently, the difference between the upper and lower boundary values ​​of the allowable temperature gradient interval is calculated to obtain the safety margin interval width. The safety margin value is divided by the safety margin interval width to obtain the safety margin ratio. The difference between 1 and the safety margin ratio is used as the boundary retention ratio. The boundary retention ratio is multiplied by the total number of candidate segment boundaries and rounded up to obtain the target segment boundary number.

[0059] Understandably, the safety margin ratio represents the relative safety margin between the current temperature and the allowable boundary. The closer this ratio is to 1, the closer the current temperature is to the target value, the more ample the safety margin, and the less need there is for fine segmentation of the historical deviation sequence; coarse segmentation is sufficient to meet the trend judgment requirements, thus reducing the number of candidate segment boundaries. Conversely, the closer this ratio is to 0, the closer the current temperature is to the allowable boundary, the tighter the safety margin, and even small trend changes may trigger exceedances. Finer segmentation is needed to capture the local evolution characteristics of the deviation sequence, and the number of candidate segment boundaries should be increased. In this invention, the difference between 1 and the safety margin ratio is used as the boundary retention ratio, causing the segmentation fineness to monotonically increase as the temperature approaches the boundary, achieving a matching relationship where higher risks result in finer observations, and lower risks result in coarser observations.

[0060] After determining the number of target segment boundaries, each candidate segment boundary is screened. For each candidate segment boundary, the absolute value of the difference between the adjacent deviation value to the right and the adjacent deviation value to the left of the boundary position is calculated as the jump amplitude of the candidate segment boundary.

[0061] Finally, all candidate segment boundaries are sorted in descending order of jump magnitude, and the top N candidate segment boundaries are selected as the final segment boundaries, where N equals the number of target segment boundaries. The deviation value sequence is divided into segments based on the position of each segment boundary, and linear regression is performed on each segment to obtain the regression slope.

[0062] In another preferred embodiment, the process of determining the trend direction of the deviation value sequence based on the sign distribution of the regression slope of each sequence segment in step A2 includes: recording the sequence segments with positive slopes as positive sequence segments and the sequence segments with negative slopes as negative sequence segments, and counting the number of positive sequence segments and the number of negative sequence segments.

[0063] The absolute values ​​of the differences between the current rolling temperature gradient and the upper and lower boundary values ​​of the allowable temperature gradient range are respectively denoted as the first difference and the second difference.

[0064] When the first difference is less than the second difference, it indicates that the current temperature gradient is closer to the upper limit boundary value. At this time, the interlayer deformation ratio deviation is more likely to trigger the risk of exceeding the limit if it evolves in the direction of positive increase. If the number of positive sequence segments is greater than or equal to the number of negative sequence segments, the direction of change is determined to be positive increase; otherwise, it is determined to be negative decrease.

[0065] When the first difference is greater than the second difference, it indicates that the current temperature gradient is closer to the lower limit boundary value. At this time, the interlayer deformation ratio deviation is more likely to trigger the risk of exceeding the limit if it evolves in the direction of negative decrease. If the number of negative sequence segments is greater than or equal to the number of positive sequence segments, the direction of change is determined to be negative decrease; otherwise, the direction of change is determined to be positive increase.

[0066] When the first difference equals the second difference, it indicates that the current temperature gradient is at the center of the allowable range, and the risks in the positive and negative directions are balanced. When the number of positive sequence segments is greater than the number of negative sequence segments, it is judged as a positive increase. When the number of positive sequence segments is less than the number of negative sequence segments, it is judged as a negative decrease. When the two are equal, it is judged as an unclear direction.

[0067] When the direction of the change trend is determined to be unclear, it is considered that the directions of the two judgment variables are inconsistent, and the reduction rate compensation is not triggered. The rolling unit maintains the current reduction rate setting value.

[0068] S3. Generate adjustment commands and drive the pressing mechanism to perform adjustment.

[0069] After determining that reduction rate compensation has been triggered, the first step is to determine the direction of compensation. Specifically, the current rolling temperature gradient is compared with the target temperature gradient: if the current value is higher than the target value, it indicates that the surface temperature of the billet is relatively high, and the flow stress of the difficult-to-deform metal layer is lower than expected due to the high temperature. Therefore, the reduction rate needs to be increased to compensate for the excess deformation capacity, and the compensation direction is to increase the reduction rate. If the current value is lower than the target value, it indicates that the billet is too cold, and the flow stress of the difficult-to-deform metal layer increases. Therefore, the reduction rate needs to be decreased to avoid insufficient deformation, and the compensation direction is to decrease the reduction rate.

[0070] Once the direction is determined, the difference between the current rolling temperature gradient and the target temperature gradient is calculated and its absolute value is taken to obtain the temperature gradient adjustment amount. This adjustment amount reflects the degree to which the current temperature deviates from the target value. The greater the temperature gradient deviation, the further the difference in flow stress between the two metal layers in the deformation zone deviates from the expected value, and the greater the required reduction rate correction amount.

[0071] Subsequently, the reduction rate compensation amount under the unit temperature gradient corresponding to the current billet material combination is retrieved and multiplied by the temperature gradient adjustment amount to obtain the reduction rate compensation amount.

[0072] Understandably, the reduction rate compensation under a unit temperature gradient is predetermined as follows: For each material combination, the difference in the temperature change rate of the forming resistance is obtained according to the aforementioned method for obtaining the difference in the temperature change rate of the forming resistance, denoted as... Multi-pass rolling calibration experiments were conducted on the target composite plate using an experimental rolling mill. Different preset reduction rates (e.g., 10%, 15%, 20%, 25%, 30%) were used for each pass. The experiment was repeated at least three times under each preset reduction rate condition, and the average rolling force was taken as the effective rolling force at that reduction rate. Based on the rolling force and the workpiece dimensions, the corresponding deformation resistance value was calculated using the following formula: , This represents the deformation resistance value. For effective rolling force, For the width of the billet, For the contact arc length, the contact arc length is calculated as follows: Calculate, where, Where is the radius of the roll. It is the difference between the inlet thickness of the billet before rolling and the outlet thickness of the billet after rolling.

[0073] Record the deformation resistance value for each pass, and perform linear fitting with the preset reduction rate on the x-axis and the deformation resistance value on the y-axis. The slope of the fitted line is the change in deformation resistance per unit reduction rate of the material combination at the target rolling temperature, denoted as K, with the unit being MPa / .

[0074] Determine the reduction rate compensation amount using the following formula : The reduction rate compensation value of each material combination is pre-calibrated and stored. During online operation, the corresponding reduction rate compensation value is retrieved according to the material combination of the current billet.

[0075] Finally, the compensation direction and reduction rate compensation value are encoded together as adjustment commands, which are output to the reduction mechanism of the rolling unit in the form of electrical or hydraulic signals to drive the hydraulic cylinder or electric reduction device to adjust the roll gap value until the actual reduction rate reaches the target value, thus completing the compensation adjustment of the reduction rate.

[0076] Example 2

[0077] To facilitate understanding of the calculation processes involved in steps S1 to S3 of this invention, the following uses a titanium / aluminum composite plate as an example to perform a complete numerical calculation of each numerical calculation process involved in steps S1 to S3 of this embodiment, so as to facilitate understanding of the specific implementation of each calculation step of this invention.

[0078] The blank specifications for the titanium / aluminum two-layer composite metal sheet are as follows: width 600mm, thickness of titanium layer 2mm and aluminum layer 3mm, target rolling temperature of titanium layer 750℃ and aluminum layer 350℃, target temperature gradient of 400℃, allowable deviation range of ±30℃, i.e. 370℃ to 430℃, and allowable deviation of interlayer deformation ratio upper limit of no more than ±5%.

[0079] When the material enters the rolling unit via the intermediate conveyor rollers, a temperature detection array arranged along the conveying direction collects 120 sampling sections. The temperature of the upper surface (titanium layer side) and the lower surface (aluminum layer side) of each section are measured, and the average temperature difference of each section (average temperature of the upper surface minus the average temperature of the lower surface) is calculated. The temperature data of some sampling sections are shown in Table 1.

[0080] Table 1. Temperature data of some sampling sections

[0081]

[0082] As shown in Table 1, the average temperature difference of the 5th sampling section is 415℃, which is the maximum value among all 120 sections. According to the rule of taking the maximum temperature difference of the section as the current rolling temperature gradient, the current rolling temperature gradient is determined to be 415℃.

[0083] Subsequently, after testing by Gleeble thermal simulation compression test, the deformation resistance temperature change rate of the titanium layer (difficult-to-deform metal) at 750℃, i.e., the first change rate, was 0.48 MPa / ℃, and the deformation resistance temperature change rate of the aluminum layer (easily deformable metal) at 350℃, i.e., the second change rate, was 0.15 MPa / ℃. Therefore, the difference in the deformation resistance temperature change rate between the difficult-to-deformable metal layer and the easily deformable metal layer at the target rolling temperature was 0.33 MPa / ℃.

[0084] Using the upper limit of ±5% for interlaminar deformation ratio deviation specified in the product standard as a constraint, and through calibration using Gleeble thermal simulation test data combined with finite element back calculation, the critical value of the allowable interlaminar flow stress deviation for coordinated deformation of the titanium / aluminum composite plate was determined to be 9.24 MPa. Therefore, the upper limit of the target temperature gradient deviation is 9.24 / 0.33 = 28℃.

[0085] In the current embodiment, with the current rolling temperature gradient of 415°C as the center value and 28°C as the search radius, historical batch records with temperature gradients falling within the range of 387°C to 443°C are retrieved from the historical rolling log.

[0086] Based on the target temperature gradient and the allowable deviation range, the upper and lower boundary values ​​of the allowable temperature gradient interval are 430℃ and 370℃, respectively. The absolute difference between the current rolling temperature gradient of 415℃ and the upper boundary value of 430℃ is 15℃, and the absolute difference between it and the lower boundary value of 370℃ is 45℃. The smaller value of 15℃ is taken as the safety margin value of the current temperature gradient. The width of the safety margin interval is 60℃, the safety margin ratio is 15 / 60 = 0.25, and the boundary retention ratio is 0.75.

[0087] To construct the deviation value sequence, 12 batches of titanium / aluminum composite plates with temperature gradients falling within the range of 368℃ to 424℃ were retrieved from the historical rolling log. The interlayer deformation ratio deviation of each batch was extracted and arranged in chronological order to obtain the deviation value sequence Y, Y=[+1.0, +1.5, +1.2, +1.8, +1.4, +2.0, +1.6, +2.2, +1.8, +2.5, +2.1, +2.8].

[0088] The difference between each adjacent data point in the deviation value sequence was calculated and the sign was determined. The points where the sign changed were extracted as candidate segment boundaries, resulting in a total of 10 candidate segment boundaries. The left and right adjacent deviation values ​​are shown in Table 1.

[0089] Table 2. Candidate Segment Boundary Jump Amplitude Data

[0090]

[0091] It is important to note that the jump magnitudes in Table 2 are the absolute values ​​of the differences between the right and left neighboring deviation values ​​of the candidate boundary. The candidate boundary position number indicates that the boundary lies between adjacent batches in the deviation value sequence. For example, position 4 represents the boundary between batch 4 and batch 5.

[0092] The number of target segment boundaries is obtained based on the total number of candidate segment boundaries and the boundary retention ratio. The target segment boundary number is 8, that is, the 8 with the largest jump amplitude are selected from 10 candidate segment boundaries as the final segment boundaries.

[0093] Sort by jump magnitude in descending order, and select the first 8 as the final segment boundaries, namely positions 9, 11, 3, 5, 7, 4, 6, and 8. Each position corresponds to... Figure 3 The red dashed line indicates that the deviation value sequence is divided into 9 segments, each of which undergoes linear regression. Figure 3 As shown, the regression slopes for each segment are: +0.1, +0.6, -0.4, +0.6, -0.4, +0.6, -0.4, +0.35, and +0.7, with 6 positive sequence segments and 3 negative sequence segments.

[0094] The absolute values ​​of the differences between the current rolling temperature gradient and the upper and lower boundary values ​​are 15℃ and 45℃, respectively, indicating that the current temperature gradient is closer to the upper boundary value. At this point, the number of negative sequence segments is less than the number of positive sequence segments, indicating a positive increasing trend.

[0095] As can be seen from the above calculations, under the condition that the current temperature gradient is 415℃ and closer to the upper limit boundary value, the trend of the historical deviation sequence is determined to be positively increasing, and the deviation direction between this temperature gradient and the target temperature gradient (400℃) is positively deviated. The two directions are consistent, which satisfies the dual judgment conditions for triggering the reduction rate compensation described in step S23. Therefore, the reduction rate compensation is triggered.

[0096] To verify the control effect of the above judgment logic and the effectiveness of the compensation strategy in the actual rolling process, a comparative rolling test was carried out on the same titanium / aluminum composite plate blank in this embodiment.

[0097] The same batch of billets was divided into four sections along its length, designated as test section E1, control section E2, test section E3, and control section E4. Test sections E1 and E3 employed the control method described in this invention, using the consistency between the direction of the temperature gradient deviation and the historical deviation trend as a dual-judgment variable to trigger compensation. Control sections E2 and E4 employed a traditional control method, triggering compensation solely based on whether the current temperature gradient deviated from the target value beyond the allowable deviation range. All four sections of billets were rolled under the same rolling equipment, conveying speed, and target reduction rate.

[0098] The actual measured temperature gradient of test section E1 and control section E2 was 396℃, which is in the middle of the allowable range and belongs to the normal working condition. The actual measured temperature gradient of test section E3 and control section E4 was 422℃. Although it did not exceed the upper limit of 430℃, it was close to the boundary and belonged to the critical working condition. After rolling, the interlayer deformation ratio deviation and thickness tolerance of the finished billet of each section were tested. In this embodiment, the upper limit of the qualified interlayer deformation ratio deviation is set according to the product standard as +5.0%, that is, the deviation value does not exceed +5.0% is considered qualified. The test results are shown in Tables 2 and 3.

[0099] Table 3 Comparison and verification data under normal operating conditions

[0100]

[0101] As shown in Table 3, under normal operating conditions (396℃), the method of this invention determines that the historical deviation trend direction is inconsistent with the current temperature deviation direction, and therefore does not trigger compensation. The traditional method also does not trigger compensation because the temperature does not exceed the limit. Both methods result in acceptable deviations in the interlayer deformation ratio of the finished products, and the thickness tolerances are both within acceptable limits, indicating comparable effectiveness.

[0102] Table 4 Comparison and verification data under critical operating conditions

[0103]

[0104] As shown in Table 4, under critical operating conditions, although the current temperature gradient has not exceeded the upper limit of the allowable range, it is approaching the boundary value. Traditional methods, lacking trend prediction capabilities, determine not to trigger compensation based solely on the temperature not exceeding the limit, resulting in an interlayer deformation ratio deviation of +6.5%, exceeding the acceptable range and leading to unacceptable (too thin) finished product thickness tolerance. In contrast, the method of this invention identifies the consistency between the current temperature gradient deviation direction (positive deviation) and the historical deviation sequence change trend direction (positive increase), determining that the temperature deviation has affected deformation coordination. It then triggers reduction rate compensation in advance (increasing the reduction rate by 0.33%), controlling the interlayer deformation ratio deviation to +3.9%, within the acceptable range, and ensuring the finished product thickness tolerance is acceptable. This verifies the effectiveness and necessity of the method of this invention under critical operating conditions.

[0105] The comparative verification results in Tables 3 and 4 show that, under normal operating conditions, both the method of this invention and the traditional method can function normally, and the present invention will not cause false triggering due to the introduction of trend judgment. Under critical operating conditions, the traditional method fails due to the omission of compensation requirements, while the method of this invention can accurately identify the risk of deformation coordination loss of control and intervene in advance, effectively solving the problem of mismatch between the preset reduction rate and the interlayer deformation coordination under actual temperature conditions.

[0106] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A method for controlling the production of composite metal sheet rolling, characterized in that, The method includes: When the billet enters the rolling unit via the intermediate conveyor rollers, the temperature of the upper and lower surfaces of the billet is collected by the temperature detection array installed on the intermediate conveyor rollers. The temperature values ​​of the upper and lower surfaces at corresponding positions along the conveying direction are subtracted point by point, and the maximum temperature difference is taken as the current rolling temperature gradient. If the current rolling temperature gradient exceeds the allowable deviation range, trigger reduction rate compensation; otherwise, execute the following sub-steps: Based on the current rolling temperature gradient and the current billet material combination, the interlayer deformation ratio deviation of each historical batch corresponding to the temperature gradient is retrieved from the historical rolling log to construct a deviation value sequence. Perform piecewise linear regression on the deviation value sequence, and determine the direction of the trend of the deviation value sequence based on the sign distribution of the regression slope of each sequence segment; The deviation direction between the rolling temperature gradient and the target temperature gradient is used as the first judgment variable, and the direction of the change trend is used as the second judgment variable. When the directions of the two judgment variables are consistent, the reduction rate compensation is triggered. Generate adjustment commands and drive the pressing mechanism to perform adjustments.

2. The composite metal sheet rolling production control method as described in claim 1, characterized in that: The temperature detection array includes an upper surface temperature detection subarray and a lower surface temperature detection subarray. The upper surface temperature detection subarray is arranged above the middle conveyor roller, and the lower surface temperature detection subarray is arranged below the gap between adjacent rollers. Each temperature detection subarray consists of infrared temperature sensors arranged at equal intervals along the width of the billet, and collects upper surface temperature data and lower surface temperature data point by point along the width of the billet.

3. The composite metal sheet rolling production control method as described in claim 1, characterized in that: The specific method for retrieving the interlayer deformation ratio deviation corresponding to each historical batch of the temperature gradient from the historical rolling log is as follows: Based on the current material composition of the billet, obtain the difference in the rate of change of deformation resistance temperature between the difficult-to-deform metal layer and the easy-to-deform metal layer at the target rolling temperature, and determine the upper limit of the target temperature gradient value deviation based on the difference in the rate of change of deformation resistance temperature. Using the current rolling temperature gradient value as the center value and the upper limit of the target temperature gradient value deviation as the search radius, retrieve all historical batch records from the historical rolling log that have the same material combination as the current billet and whose temperature gradient values ​​fall within the search radius. Extract the deviation between the actual value and the target value of the interlayer deformation ratio from the retrieved historical batch records, arrange them in chronological order, and form a deviation value sequence.

4. The composite metal sheet rolling production control method as described in claim 3, characterized in that: The upper limit of the target temperature gradient value deviation is determined in the following way: The temperature change rate of the deformation resistance of the difficult-to-deform metal layer at the target rolling temperature is obtained as the first change rate, and the temperature change rate of the deformation resistance of the easily deformable metal layer at the target rolling temperature is obtained as the second change rate. The absolute value of the difference between the first rate of change and the second rate of change is taken as the difference in the rate of change of deformation resistance temperature. Obtain the critical value of interlaminar flow stress deviation allowed for coordinated deformation under the current billet material combination, and use the ratio of the critical value of interlaminar flow stress deviation to the difference in the temperature change rate of deformation resistance as the upper limit of the target temperature gradient value deviation.

5. The composite metal sheet rolling production control method as described in claim 1, characterized in that: Before performing piecewise linear regression on the deviation value sequence, it is also necessary to determine the number of target segment boundaries. The specific determination process includes: Based on the target temperature gradient and its allowable deviation range, the upper and lower boundary values ​​of the allowable temperature gradient interval are obtained. The absolute values ​​of the differences between the current rolling temperature gradient and the upper and lower boundary values ​​are calculated respectively, and the smaller value is taken as the safety margin value of the current temperature gradient. Calculate the difference between the upper and lower boundary values ​​of the allowable temperature gradient range to obtain the width of the safety margin range. Divide the safety margin value by the width of the safety margin range to obtain the safety margin ratio. Subtract the safety margin ratio from 1 to obtain the boundary retention ratio. Obtain the difference between adjacent deviation values ​​in the deviation value sequence, generate an adjacent point difference sequence, extract the points in the adjacent point difference sequence where the sign changes as candidate segment boundaries, and obtain the total number of candidate segment boundaries; Multiply the boundary retention ratio by the total number of candidate segment boundaries, and round the product up to obtain the target number of segment boundaries.

6. The composite metal sheet rolling production control method as described in claim 5, characterized in that: The specific process of performing piecewise linear regression on the deviation value sequence is as follows: For each candidate segment boundary, calculate the absolute value of the difference between the adjacent deviation value to the right and the adjacent deviation value to the left of the boundary position, and use it as the jump amplitude of the candidate segment boundary. Sort all candidate segment boundaries in descending order of the jump magnitude, and select the top N candidate segment boundaries as segment boundaries, where N is equal to the number of target segment boundaries; The deviation value sequence is divided into segments based on the location of each segment boundary. Linear regression is then performed on each segment to obtain the regression slope of each segment.

7. The composite metal sheet rolling production control method as described in claim 5, characterized in that: The direction of the trend of the deviation value sequence is determined as follows: Sequence segments with a positive slope are denoted as positive sequence segments, and sequence segments with a negative slope are denoted as negative sequence segments. The number of positive sequence segments and the number of negative sequence segments are counted. The absolute values ​​of the differences between the current rolling temperature gradient and the upper and lower boundary values ​​are respectively denoted as the first difference and the second difference. When the first difference is less than the second difference, if the number of positive sequence segments is greater than or equal to the number of negative sequence segments, the trend direction is determined to be positively increasing; otherwise, the trend direction is determined to be negatively decreasing. When the first difference is greater than the second difference, if the number of negative sequence segments is greater than or equal to the number of positive sequence segments, the trend direction is determined to be negative decreasing; otherwise, the trend direction is determined to be positive increasing. When the first difference equals the second difference, if the number of positive sequence segments is greater than the number of negative sequence segments, it is determined to be a positive increase; if the number of positive sequence segments is less than the number of negative sequence segments, it is determined to be a negative decrease; if the two are equal, it is determined to be an unknown direction.

8. The composite metal sheet rolling production control method as described in claim 1, characterized in that: The direction of the deviation between the rolling temperature gradient and the target temperature gradient is determined as follows: with the target temperature gradient as the zero point, the difference between the rolling temperature gradient and the target temperature gradient is calculated. A positive difference is considered a positive deviation, and a negative difference is considered a negative deviation.

9. The composite metal sheet rolling production control method as described in claim 8, characterized in that: The conditions under which the first and second judgment variables have the same direction are as follows: The deviation direction of the first judgment variable is positive, and the deviation direction of the second judgment variable is increasing positively; The first judgment variable has a negative deviation direction, and the second judgment variable has a negative decreasing direction.

10. The composite metal sheet rolling production control method as described in claim 4, characterized in that: The specific steps for generating the adjustment command are as follows: If the current rolling temperature gradient is higher than the target temperature gradient, the compensation direction is to increase the reduction rate; if the current rolling temperature gradient is lower than the target temperature gradient, the compensation direction is to decrease the reduction rate. Calculate the difference between the current rolling temperature gradient and the target temperature gradient, and take the absolute value of the difference as the temperature gradient adjustment amount; Retrieve the reduction rate compensation amount under the unit temperature gradient corresponding to the current billet material combination, and multiply it with the temperature gradient adjustment amount to obtain the reduction rate compensation amount; Adjustment commands are generated based on the compensation direction and the amount of reduction rate compensation.

Images