A multi-section tension model adaptive control method and system

Abstract

This invention relates to the field of adaptive control technology, and discloses a multi-segment tension model adaptive control method and system. The method involves acquiring type switching data and transition boundary advance speed to determine the longitudinal stiffness parameter and transition boundary position state quantity after the switch; determining the corrected longitudinal stiffness parameter and longitudinal stiffness residual reduction based on a second tension feedback value and a preset tension prediction model; correcting the corrected boundary position state quantity when the longitudinal stiffness residual reduction is not greater than a preset residual reduction threshold and the corrected longitudinal stiffness parameter is equal to the switched longitudinal stiffness parameter; determining a candidate second downstream speed setpoint based on a second downstream speed feedback value, and determining the second downstream target speed setpoint by combining the corrected longitudinal stiffness parameter, the corrected boundary position state quantity, and the preset tension setpoint. This invention can reduce model state correction errors and decrease tension control deviation during the type switching phase.

Description

Technical Field

[0001] This invention belongs to the field of adaptive control technology, specifically relating to an adaptive control method and system for a multi-segment tension model. Background Technology

[0002] In model-based adaptive control, a predictive model is typically built based on the input data, feedback data, and operating state of the controlled object, and the control quantity is determined based on the prediction results. When the type or operating condition of the controlled object changes, the model parameters and model state may deviate from the actual operating state, thus affecting the prediction results and the determination of subsequent control quantities.

[0003] For example, Chinese patent CN101349893B discloses an adaptive model predictive control device that updates the model online; patent application CN105739310A discloses an adaptive control system for a servo system based on multiple models, selecting model parameters according to a cost function; and Chinese patent CN112506045A discloses an adaptive control method and system for a non-standard discrete-time nonlinear system, updating parameter estimates and state estimates based on estimation errors. While existing technologies can identify and correct changes in model parameters or system states, they still struggle to accurately determine the correction target based on the same prediction deviation when both model parameter deviations and dynamic state deviations simultaneously affect the model output.

[0004] In multi-segment continuous tension control, after a type switch, the longitudinal stiffness parameter after the switch may deviate from the actual longitudinal characteristics. Simultaneously, speed detection errors, drive execution errors, and cumulative errors in boundary position recursion may cause the transition boundary position state variable to deviate from the actual transition boundary position. Both of these deviations will cause a discrepancy between the predicted tension and the second tension feedback value.

[0005] Therefore, when directly correcting the longitudinal stiffness parameter or transition boundary position state quantity after switching based solely on the tension deviation, it is difficult to determine the source of the deviation, which can easily lead to the wrong selection of the model state correction object, causing the corrected preset tension prediction model to be inconsistent with the actual operating state, thereby affecting the accuracy of determining the second downstream target speed setting value. Summary of the Invention

[0006] To address the issue that deviations in longitudinal stiffness parameters and transition boundary position states during type switching can cause tension prediction errors, leading to incorrect selection of model state correction objects and further affecting the accuracy of determining the second downstream target velocity setpoint, this invention provides a multi-segment tension model adaptive control method and system.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides a multi-segment tension model adaptive control method, comprising:

[0009] Acquire type switching data and transition boundary advance speed, determine longitudinal stiffness parameters and transition boundary position state quantities after switching based on the type switching data and transition boundary advance speed, initialize corrected longitudinal stiffness parameters with the longitudinal stiffness parameters after switching, and initialize corrected boundary position state quantities with the transition boundary position state quantities.

[0010] When the transition boundary position state quantity is within the preset boundary correction range, a second tension feedback value is obtained, and the longitudinal stiffness parameter after switching is corrected according to the second tension feedback value and the preset tension prediction model, and the corrected longitudinal stiffness parameter and the reduction amount of longitudinal stiffness residual are determined.

[0011] When the reduction in longitudinal stiffness residual is not greater than a preset residual reduction threshold and the corrected longitudinal stiffness parameter is equal to the switched longitudinal stiffness parameter, the corrected boundary position state quantity is corrected according to the second tension feedback value and the preset tension prediction model.

[0012] Obtain a second downstream velocity feedback value, determine a candidate second downstream velocity setting value based on the second downstream velocity feedback value, calculate the target predicted tension for each candidate second downstream velocity setting value based on the corrected longitudinal stiffness parameter and the corrected boundary position state quantity, determine the second downstream target velocity setting value based on the target predicted tension and the preset tension setting value, and control the second downstream drive mechanism.

[0013] Furthermore, the method also includes:

[0014] The first open tension control section and the second open tension control section are set sequentially along the running direction, and they share the same intermediate boundary;

[0015] Obtain the inlet tension of the first open tension control section and determine it as the first inlet tension feedback value; obtain the actual tension of the second open tension control section and determine it as the second tension feedback value; obtain the intermediate tension feedback value.

[0016] The system acquires a first upstream speed feedback value, an intermediate speed feedback value, and a second downstream speed feedback value; acquires the first upstream boundary position, the intermediate boundary position, and the downstream boundary position of the second open tension control section; the second downstream speed feedback value is the actual driving linear speed of the second downstream drive mechanism; and the preset tension setting value is the tension setting value of the second open tension control section.

[0017] Furthermore, the determination of the longitudinal stiffness parameter after switching, the transition boundary position state quantity, and the preset boundary correction interval includes:

[0018] Obtain the type before switching, the type after switching, and the initial transition boundary position state from the type switching data;

[0019] The longitudinal stiffness parameter before switching, the longitudinal stiffness parameter after switching, the free longitudinal strain parameter before switching, and the free longitudinal strain parameter after switching are determined according to the mapping relationship between the preset type and the longitudinal characteristic parameter.

[0020] The transition boundary position state quantity is initialized with the initial transition boundary position state quantity, and the transition boundary position state quantity and the corrected boundary position state quantity are advanced according to the transition boundary advancement speed.

[0021] The interval after the two ends of the second open tension control section are respectively offset inward by a preset boundary safety distance is determined as the preset boundary correction interval.

[0022] Furthermore, the determination of the segment type occupancy status of the preset tension prediction model includes:

[0023] Obtain the predicted open tension control section length and upstream boundary position, wherein the upstream boundary position is either the first upstream boundary position or the intermediate boundary position. Limit the position difference obtained by subtracting the upstream boundary position from the transition boundary position state quantity or the corrected boundary position state quantity input to the preset tension prediction model to between zero and the predicted open tension control section length, thereby obtaining the type-occupied length after switching.

[0024] When the length occupied by the switched type is greater than zero, the switched type is determined as the entry type; when the length occupied by the switched type is zero, the original type is determined as the entry type.

[0025] When the length occupied by the type after switching is less than the length of the predicted open tension control section, the type before switching is determined as the exit type; when it is equal to the length of the predicted open tension control section, the type after switching is determined as the exit type.

[0026] Furthermore, the calculation of the segment parameters of the preset tension prediction model includes:

[0027] Subtract the type-occupied length after switching from the predicted open tension control section length to obtain the type-occupied length before switching.

[0028] The average free strain of the section is obtained by dividing the sum of the product of the pre-switching type occupancy length and the pre-switching free longitudinal strain parameter and the product of the post-switching type occupancy length and the post-switching free longitudinal strain parameter by the predicted open tension control section length.

[0029] The longitudinal flexibility of the section is obtained by adding the ratio of the length occupied by the type after the switch to the longitudinal stiffness parameter after the switch and the ratio of the length occupied by the type before the switch to the longitudinal stiffness parameter before the switch.

[0030] Furthermore, the determination of the initial parameters of the preset tension prediction model includes:

[0031] The switched longitudinal stiffness parameter used in the preset tension prediction model is used as the input longitudinal stiffness parameter; during normal prediction, the current corrected longitudinal stiffness parameter is taken; during candidate screening, the value of each longitudinal stiffness candidate parameter is taken; during verification, the switched longitudinal stiffness parameter or the longitudinal stiffness parameter to be verified is taken.

[0032] When predicting the first open tension control section, the first inlet tension feedback value, the first upstream velocity feedback value, and the intermediate velocity feedback value are respectively determined as upstream tension, upstream boundary velocity, and downstream boundary velocity; when predicting the second open tension control section, the intermediate tension feedback value, the intermediate velocity feedback value, and the second downstream velocity feedback value are respectively determined as upstream tension, upstream boundary velocity, and downstream boundary velocity.

[0033] The predicted tension is initialized with the upstream tension of the predicted open tension control section, and the average total strain of the section and the inlet total strain are determined based on the predicted tension, the longitudinal flexibility of the section, the average free strain of the section, and the length of the predicted open tension control section.

[0034] Furthermore, the recursive calculation of the predicted tension includes:

[0035] The total strain at the outlet is calculated based on the predicted tension at the previous velocity sampling time, the outlet longitudinal stiffness parameter corresponding to the outlet type, and the outlet free longitudinal strain parameter.

[0036] The outlet longitudinal stiffness parameter and the outlet free longitudinal strain parameter are taken as the switched longitudinal stiffness parameter and the switched free longitudinal strain parameter when the outlet type is the switched type, and as the unswitched longitudinal stiffness parameter and the unswitched free longitudinal strain parameter when the outlet type is the unswitched type.

[0037] The average total strain increment of the section is calculated based on the upstream boundary velocity, the downstream boundary velocity, the total inlet strain, the total outlet strain, and the preset velocity sampling period.

[0038] The average total strain increment of the section is accumulated to the average total strain of the section, and the predicted tension is calculated based on the average total strain of the section, the average free strain of the section, and the longitudinal flexibility of the section.

[0039] Furthermore, the correction of the longitudinal stiffness parameter after the switch includes:

[0040] The longitudinal stiffness parameter after the switch and the longitudinal stiffness values ​​that are greater than zero obtained by adding and subtracting the preset longitudinal stiffness step size respectively are determined as candidate longitudinal stiffness parameter values.

[0041] The intermediate tension feedback value, intermediate velocity feedback value, and second downstream velocity feedback value are respectively determined as the upstream tension, upstream boundary velocity, and downstream boundary velocity of the second open tension control section, and the transition boundary position state quantity is used as the boundary position input value of the preset tension prediction model.

[0042] Within the preset longitudinal stiffness correction data window, a predicted tension sequence is calculated based on each of the longitudinal stiffness candidate parameter values. The predicted tension sequence is the predicted tension sequence of the second open tension control section. The longitudinal stiffness prediction difference is determined based on the difference between each of the predicted tension sequences.

[0043] When the longitudinal stiffness prediction difference is greater than the preset longitudinal stiffness identifiable threshold, longitudinal stiffness candidate residuals are calculated based on each of the predicted tension sequences and the second tension feedback value.

[0044] When the difference in longitudinal stiffness prediction is not greater than the preset longitudinal stiffness identifiable threshold, the corrected longitudinal stiffness parameter is kept as the switched longitudinal stiffness parameter.

[0045] Furthermore, the selection of the candidate longitudinal stiffness parameter values ​​includes:

[0046] The longitudinal stiffness candidate residual obtained from the switched longitudinal stiffness parameters is determined as the current longitudinal stiffness residual, and the minimum longitudinal stiffness candidate residual is determined as the minimum longitudinal stiffness residual.

[0047] The reduction in longitudinal stiffness residual is obtained by subtracting the minimum longitudinal stiffness residual from the current longitudinal stiffness residual.

[0048] When the reduction in longitudinal stiffness residual is greater than the preset residual reduction threshold and the minimum longitudinal stiffness candidate residual is obtained from a unique longitudinal stiffness candidate parameter value, the longitudinal stiffness candidate parameter value is determined as the longitudinal stiffness parameter to be verified.

[0049] Furthermore, the verification of the longitudinal stiffness parameter to be verified includes:

[0050] Within the preset longitudinal stiffness verification window, the predicted tension sequence is calculated using the switched longitudinal stiffness parameters and the longitudinal stiffness parameters to be verified, respectively.

[0051] The reduction in longitudinal stiffness verification residual is determined based on the difference between the residuals of the two predicted tension sequences and the second tension feedback value;

[0052] When the reduction in the longitudinal stiffness verification residual is greater than the preset residual reduction threshold, the longitudinal stiffness parameter to be verified is determined as the corrected longitudinal stiffness parameter.

[0053] When the reduction in the longitudinal stiffness verification residual is not greater than the preset residual reduction threshold, the corrected longitudinal stiffness parameter is kept the same as the switched longitudinal stiffness parameter.

[0054] Furthermore, the correction of the corrected boundary position state quantity includes:

[0055] The candidate values ​​of the boundary position are determined by using the state quantity of the transition boundary position at the beginning of the data window at the preset boundary position correction and the positions obtained by adding and subtracting the preset position step size respectively.

[0056] The intermediate tension feedback value, the intermediate velocity feedback value, and the second downstream velocity feedback value are respectively determined as the upstream tension, upstream boundary velocity, and downstream boundary velocity of the second open tension control section;

[0057] Each of the candidate boundary positions is used as the boundary position input value of the preset tension prediction model. The predicted tension sequence is calculated, and the candidate boundary position residual is calculated based on the predicted tension sequence and the second tension feedback value. Each of the candidate boundary positions is advanced according to the transition boundary advancement speed.

[0058] Furthermore, the screening of the candidate values ​​for the boundary positions includes:

[0059] The candidate boundary position residuals obtained from the transition boundary position state quantities are determined as the current candidate boundary position residuals, and the minimum candidate boundary position residuals are determined as the minimum candidate boundary position residuals.

[0060] Subtract the minimum boundary position candidate residual from the current boundary position candidate residual to obtain the reduction in boundary position residual;

[0061] When the reduction in the boundary position residual is greater than the preset residual reduction threshold and the minimum boundary position candidate residual is obtained from the unique boundary position candidate value, the boundary position candidate value is determined as the boundary position to be verified.

[0062] Furthermore, the verification of the boundary location to be verified includes:

[0063] Within the preset boundary position verification window, the transition boundary position state quantity and the boundary position to be verified are advanced according to the transition boundary advancement speed, and the predicted tension sequence is calculated respectively.

[0064] The boundary position is determined based on the difference between the residuals of the two predicted tension sequences and the second tension feedback value to verify the amount of residual reduction;

[0065] When the reduction in the boundary position verification residual is greater than the preset residual reduction threshold, the boundary position to be verified is determined as the corrected boundary position state quantity.

[0066] When the reduction in the residual of the boundary position verification is not greater than the preset residual reduction threshold, the boundary position to be verified is not used, and the corrected boundary position state quantity continues to advance according to the transition boundary advancement speed.

[0067] Furthermore, the determination of the candidate second downstream speed setpoint and the second downstream target speed setpoint includes:

[0068] The second downstream speed feedback value is located between the preset minimum drive speed and the preset maximum drive speed;

[0069] The candidate second downstream speed setting value range is determined based on the second downstream speed feedback value, the preset allowable speed change in a single control cycle, the preset minimum drive speed, and the preset maximum drive speed. The candidate second downstream speed setting value is determined according to the preset speed search step size.

[0070] The intermediate tension feedback value, intermediate velocity feedback value, and each of the candidate second downstream velocity setting values ​​are respectively determined as the upstream tension, upstream boundary velocity, and downstream boundary velocity of the second open tension control section. The predicted tension sequence of the second open tension control section is calculated based on the corrected longitudinal stiffness parameter and the corrected boundary position state quantity. The predicted tension at the last velocity sampling moment in the predicted tension sequence is determined as the target predicted tension.

[0071] The candidate second downstream speed setting value with the smallest absolute value of the difference from the preset tension setting value is determined as the second downstream target speed setting value.

[0072] Secondly, the present invention provides a multi-segment tension model adaptive control system for implementing the aforementioned multi-segment tension model adaptive control method, comprising a parameter determination module, a longitudinal stiffness correction module, a boundary position correction module, and a drive control module;

[0073] The parameter determination module is used to acquire type switching data and transition boundary advance speed, determine the longitudinal stiffness parameter and transition boundary position state quantity after switching, and initialize the corrected longitudinal stiffness parameter and corrected boundary position state quantity.

[0074] The longitudinal stiffness correction module is used to obtain a second tension feedback value and determine the corrected longitudinal stiffness parameter and the reduction in longitudinal stiffness residual based on the second tension feedback value and a preset tension prediction model.

[0075] The boundary position correction module is used to obtain a second tension feedback value and correct the corrected boundary position state quantity when the reduction in the longitudinal stiffness residual is not greater than a preset residual reduction threshold and the corrected longitudinal stiffness parameter is equal to the switched longitudinal stiffness parameter.

[0076] The drive control module is used to acquire the second downstream speed feedback value, determine the candidate second downstream speed setpoint and the target predicted tension, determine the second downstream target speed setpoint based on the target predicted tension and the preset tension setpoint, and control the second downstream drive mechanism.

[0077] Compared with the prior art, the present invention has the following beneficial effects:

[0078] This invention addresses the issue that deviations in longitudinal stiffness parameters and transition boundary position states can both cause tension prediction errors during type switching. By distinguishing between these two types of model state deviations, the impact of changes in longitudinal stiffness parameters and boundary position states on tension prediction results can be identified separately. This reduces model state correction errors caused by confusion of deviation sources and improves the accuracy of the tension prediction model in representing the current operating state. Based on the corrected model state, the matching degree between the second downstream target velocity setpoint and the preset tension setpoint can be improved, reducing tension deviations caused by model state mismatch during type switching and enhancing the accuracy and stability of multi-segment tension control. Attached Figure Description

[0079] Figure 1 This is a schematic diagram of the multi-segment tension model adaptive control method of the present invention;

[0080] Figure 2 This is a schematic diagram of the multi-segment tension model adaptive control system structure of the present invention;

[0081] Figure 3 This is a schematic diagram of the longitudinal stiffness parameter correction process of the present invention;

[0082] Figure 4 This is a schematic diagram of the boundary position state quantity correction process of the present invention;

[0083] Figure 5This is a schematic diagram of the process for determining the second downstream target speed setpoint of the present invention. Detailed Implementation

[0084] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. The described embodiments are used to illustrate the technical solution of the present invention and do not constitute a limitation on the scope of protection of the present invention. Other implementation methods obtained by those skilled in the art based on the embodiments of the present invention without creative effort should all fall within the scope of protection of the present invention.

[0085] Example 1

[0086] Please see Figure 1 As shown, this embodiment provides a multi-segment tension model adaptive control method. The flexible strip is continuously transmitted and maintained in a longitudinally tensile state between the first upstream drive mechanism, the intermediate drive mechanism, and the second downstream drive mechanism. The material composition of the flexible strip is not a limiting condition in this embodiment; when the flexible strip can obtain type switching data, transition boundary advance speed, tension feedback value, and speed feedback value, and can obtain the longitudinal stiffness parameters before switching, the longitudinal stiffness parameters after switching, the free longitudinal strain parameters before switching, and the free longitudinal strain parameters after switching through pre-calibration, the tension model adaptive control is executed according to the processing procedure described in this embodiment.

[0087] A first open tension control section and a second open tension control section are sequentially arranged along the running direction. Specifically, in this embodiment, the upstream boundary of the first open tension control section is formed by a first upstream drive mechanism, the downstream boundary of the first open tension control section and the upstream boundary of the second open tension control section are formed by the same intermediate drive mechanism, and the downstream boundary of the second open tension control section is formed by a second downstream drive mechanism. The drive boundary where the intermediate drive mechanism is located is defined as the intermediate boundary. An open tension control section refers to a section where the strip is continuously tensioned between adjacent drive boundaries; when guide rollers or tension measuring rollers are set within the open tension control section, the guide rollers or tension measuring rollers do not form independent traction boundaries.

[0088] During equipment installation or commissioning, a one-dimensional path coordinate system is established along the strip's running path, and the strip's running direction is determined as the direction of coordinate value increase. The first upstream boundary position, the intermediate boundary position, and the downstream boundary position of the second open tension control section are calibrated according to the equipment installation dimensions and stored in the controller. The length of the first open tension control section is the length obtained by subtracting the first upstream boundary position from the intermediate boundary position, and the length of the second open tension control section is the length obtained by subtracting the intermediate boundary position from the downstream boundary position of the second open tension control section. In this embodiment, the predicted open tension control section length is either the first open tension control section length or the second open tension control section length used when the preset tension prediction model is currently invoked.

[0089] The first inlet tension feedback value represents the tension of the strip when it enters the first open tension control section. In this embodiment, the first inlet tension feedback value is collected by a tension measuring device located downstream of the first upstream boundary. The intermediate tension feedback value represents the upstream tension of the strip when it enters the second open tension control section from the intermediate boundary. In this embodiment, the intermediate tension feedback value is collected by a force measuring roller located downstream of the intermediate boundary. The second tension feedback value represents the measured tension of the second open tension control section at the current tension calculation moment. In this embodiment, the second tension feedback value is collected by a tension measuring roller located within the second open tension control section. The intermediate tension feedback value serves as the upstream tension input of the second open tension control section, and the second tension feedback value serves as a verification value for the predicted result of the second open tension control section. The intermediate tension feedback value is not used as the second tension feedback value, and vice versa.

[0090] The first upstream speed feedback value is the surface linear velocity at the contact position between the first upstream drive mechanism and the strip; the intermediate speed feedback value is the surface linear velocity at the contact position between the intermediate drive mechanism and the strip; and the second downstream speed feedback value is the surface linear velocity at the contact position between the second downstream drive mechanism and the strip. In this embodiment, encoders installed on the first upstream drive mechanism, intermediate drive mechanism, and second downstream drive mechanism are used to collect rotational speeds. The first upstream speed feedback value, intermediate speed feedback value, and second downstream speed feedback value are calculated based on the encoder speed, transmission ratio, and effective diameter of the drive roller. When a servo driver capable of directly outputting surface linear velocity is used, the controller reads the surface linear velocity output by the servo driver. The second downstream speed feedback value serves as the current downstream boundary velocity of the second open tension control section and as a speed reference for forming candidate second downstream speed setpoints.

[0091] The transition boundary advance speed represents the actual downstream speed at which the material boundary between the strip before and after the switch moves along the strip's running path from its current position. When the equipment is equipped with strip speed measuring devices in different open tension control sections, the controller selects the strip speed value of the section corresponding to the transition boundary position state quantity as the transition boundary advance speed; when a reference speed measuring device capable of characterizing the overall material conveying speed is used, the reference speed value is directly adopted. The transition boundary advance speed is preferably obtained by a contact speed measuring wheel or a non-contact laser speed measuring device; when only a drive roller encoder can be used, the estimated linear velocity is obtained based on the encoder speed and the effective diameter of the drive roller, allowing subsequent boundary position state quantity correction to absorb the deviation of the transition boundary position state quantity caused by drive roller slippage, encoder error, and cumulative integral error. Therefore, this embodiment does not assume absolute slippage between the drive roller and the strip as an implementation condition.

[0092] Before the equipment is put into operation, each tension and velocity measuring device undergoes zero-point, range, and time synchronization calibration. The controller reads velocity data according to a preset velocity sampling period and aligns the tension data to the same velocity sampling time using timestamps. When the sampling period of the tension sensor differs from the preset velocity sampling period, the most recent sampled value can be selected using timestamps or linear interpolation can be performed. The second tension feedback value is a measured tension value at a single velocity sampling time, and forms a sequence of second tension feedback values ​​arranged in chronological order within the preset longitudinal stiffness correction data window, preset longitudinal stiffness verification window, preset boundary position correction data window, and preset boundary position verification window.

[0093] In this embodiment, the velocity sampling time is the discrete moment when the controller executes a recursive calculation of the preset tension prediction model once according to the preset velocity sampling period; the tension calculation time is the moment when the controller generates an actual tension calculation result and performs candidate residual processing. Both the preset longitudinal stiffness correction data window and the preset longitudinal stiffness verification window are composed of temporally continuous velocity sampling times. The preset longitudinal stiffness verification window is set after the preset longitudinal stiffness correction data window, and they do not share the same velocity sampling time. Similarly, both the preset boundary position correction data window and the preset boundary position verification window are composed of temporally continuous velocity sampling times. The preset boundary position verification window is set after the preset boundary position correction data window, and they do not share the same velocity sampling time. The predicted tension sequence is the set of predicted tension values ​​obtained by the preset tension prediction model within a data window according to the velocity sampling time sequence; the longitudinal stiffness candidate residual and the boundary position candidate residual are calculated using the predicted tension and the second tension feedback value from the same data window and the same velocity sampling time.

[0094] S101. Obtain type switching data and transition boundary advance speed, determine the longitudinal stiffness parameter and transition boundary position state quantity after switching, and initialize the corrected longitudinal stiffness parameter and corrected boundary position state quantity.

[0095] The type switching data includes the type before switching, the type after switching, and the initial transition boundary position status quantity. Specifically, the type before switching and the type after switching are read from the production recipe; when the equipment control system uses strip specification identification to manage products, the type before switching and the type after switching are read from the strip specification identification. The initial transition boundary position status quantity indicates the position of the material boundary between the strip after switching and the strip before switching in a one-dimensional path coordinate system when the strip type switching actually begins. The initial transition boundary position status quantity is determined based on the fixed strip formation position or the calibrated switching execution position, and serves as the starting value for recursive transition boundary position status quantity calculation.

[0096] During the equipment commissioning phase, a mapping relationship between preset types and longitudinal characteristic parameters is established. Longitudinal tensile calibration is performed on the flexible strips corresponding to each type. Within the normal operating tension range of the strip, the longitudinal stiffness parameter is determined based on the relationship between the change in longitudinal tensile force and the change in longitudinal relative elongation. The longitudinal stiffness parameter characterizes the strip's ability to resist longitudinal elastic deformation. When total tension is used as the tension representation, the longitudinal stiffness parameter is the one matched to the total tension of the entire strip. When unit width tension is used, the longitudinal stiffness parameter is the one used per unit width, ensuring that the tension and longitudinal stiffness parameters have a consistent width reference.

[0097] The free longitudinal strain parameter represents the proportion of natural longitudinal length change of the strip relative to a preset reference process state when it is not subjected to external longitudinal tension. Specifically, for flexible strips of the same type, the reference natural length without external longitudinal tension is measured under the preset reference process state, and the working natural length without external longitudinal tension is measured under the corresponding working process state. The difference between the working natural length and the reference natural length is divided by the reference natural length to obtain the free longitudinal strain parameter. A positive free longitudinal strain parameter indicates that the natural length increases relative to the preset reference process state, while a negative free longitudinal strain parameter indicates that the natural length decreases relative to the preset reference process state. For flexible strips where moisture content or temperature has a significant impact on longitudinal deformation, the longitudinal stiffness parameter and the free longitudinal strain parameter are stored in the mapping relationship between preset type and longitudinal characteristic parameters according to preset moisture content levels or preset temperature levels, respectively.

[0098] The controller queries the mapping relationship between the preset type and the longitudinal characteristic parameters based on the pre-switching type and the post-switching type to obtain the pre-switching longitudinal stiffness parameter, the post-switching longitudinal stiffness parameter, the pre-switching free longitudinal strain parameter, and the post-switching free longitudinal strain parameter. The pre-switching longitudinal stiffness parameter and the pre-switching free longitudinal strain parameter are used to describe the flexible strip portion corresponding to the pre-switching type that has not yet left the predicted open tension control zone; the post-switching longitudinal stiffness parameter and the post-switching free longitudinal strain parameter are used to describe the flexible strip portion corresponding to the post-switching type that has entered the predicted open tension control zone.

[0099] For example, to illustrate the subsequent calculation process, the first upstream boundary position is calibrated as 10.0 meters, the middle boundary position as 12.0 meters, and the downstream boundary position of the second open tension control section as calibrated as 14.0 meters; the longitudinal stiffness parameter before switching is calibrated as 100,000 Newtons, and the longitudinal stiffness parameter after switching is calibrated as 120,000 Newtons; the free longitudinal strain parameter before switching is 0.001, and the free longitudinal strain parameter after switching is 0.002. The above values ​​are example values ​​set to illustrate the calculation process of the preset tension prediction model, do not represent actual equipment test results, and are not intended to limit the scope of protection of this invention.

[0100] The transition boundary position state quantity is initialized using the initial transition boundary position state quantity. Between two adjacent tension calculation moments, the transition boundary advance velocity is integrated over time, and the resulting boundary advance distance is added to the transition boundary position state quantity at the previous tension calculation moment to obtain the transition boundary position state quantity at the current tension calculation moment. Specifically, during the discrete calculation process, the average transition boundary advance velocity at two adjacent velocity sampling moments is multiplied by the time interval between the two velocity sampling moments to obtain the boundary advance distance within that time interval, and the boundary advance distances within each time interval are accumulated sequentially. This calculation method is used to characterize the process of the material boundary between the flexible strip sections corresponding to the type after switching and the type before switching moving downstream as the strip is conveyed.

[0101] The corrected longitudinal stiffness parameter is initialized with the switched longitudinal stiffness parameter, and the corrected boundary position state variable is initialized with the transition boundary position state variable. For a single type switch, the corrected longitudinal stiffness parameter remains the same as the switched longitudinal stiffness parameter until the longitudinal stiffness parameter correction is complete. After the longitudinal stiffness parameter to be verified passes verification, it is written into the corrected longitudinal stiffness parameter and maintained for the remaining control cycle of this type switch, without being re-initialized with the switched longitudinal stiffness parameter. When no new boundary position state variable correction value is received, the corrected boundary position state variable continues to advance from the currently stored value according to the actual acquired transition boundary advancement speed. When the next type switch is detected, the corrected longitudinal stiffness parameter and the corrected boundary position state variable are re-initialized based on the data from the next type switch.

[0102] The preset boundary correction interval is located within the second open tension control section. The upstream boundary of the second open tension control section is offset downstream by a preset boundary safety distance, and the downstream boundary is offset upstream by the same preset boundary safety distance. The interval between these two offset positions is defined as the preset boundary correction interval. The preset boundary safety distance ensures that the candidate boundary position value remains within the second open tension control section as it advances within the preset boundary position correction data window and the preset boundary position verification window. The maximum permissible transition boundary advancement speed is the upper limit of the transition boundary advancement speed allowed by both the equipment's driving capability and the production process. During the equipment commissioning phase, it is determined and stored in the controller based on the maximum permissible driving speed of the first upstream drive mechanism, the intermediate drive mechanism, the second downstream drive mechanism, and the maximum continuous operating speed specified by the production process. The preset boundary safety distance is determined based on the preset position step size, the upper limit of position measurement error, and the maximum boundary advancement distance calculated according to the maximum permissible transition boundary advancement speed.

[0103] S102. Determine the section type occupancy status and section parameters of the preset tension prediction model.

[0104] The preset tension prediction model uses a one-dimensional longitudinal force transmission relationship to describe the predicted open tension control section. Model inputs include the length of the predicted open tension control section, the upstream boundary position, the boundary position input value, the longitudinal stiffness parameter used for prediction, the upstream tension, the upstream boundary velocity, and the downstream boundary velocity. The boundary position input value is the transition boundary position state quantity, the corrected boundary position state quantity, or a candidate boundary position value. The model output is the predicted tension at each velocity sampling time.

[0105] For a predicted open tension control segment, the relative boundary position is obtained by subtracting the upstream boundary position of the predicted open tension control segment from the boundary position input value. When the relative boundary position is less than zero, the occupancy length of the switched type is determined to be zero; when the relative boundary position is between zero and the length of the predicted open tension control segment, the relative boundary position is determined as the occupancy length of the switched type; when the relative boundary position is greater than the length of the predicted open tension control segment, the length of the predicted open tension control segment is determined as the occupancy length of the switched type. Using this method, the occupancy length of the switched type always lies between zero and the length of the predicted open tension control segment.

[0106] The pre-switching type's occupied length is obtained by subtracting the post-switching type's occupied length from the predicted open tension control section length. If the post-switching type's occupied length is greater than zero, it indicates that the flexible strip portion corresponding to the post-switching type has entered the upstream end of the predicted open tension control section, and the post-switching type is determined as the inlet type; if the post-switching type's occupied length is equal to zero, the pre-switching type is determined as the inlet type. If the post-switching type's occupied length is less than the predicted open tension control section length, it indicates that the downstream end of the predicted open tension control section is still the flexible strip portion corresponding to the pre-switching type, and the pre-switching type is determined as the outlet type; if the post-switching type's occupied length is equal to the predicted open tension control section length, the post-switching type is determined as the outlet type.

[0107] The segment-average free strain represents the average unloaded longitudinal deformation state of the flexible strip portions corresponding to the pre-switching and post-switching types within the currently predicted open tension control segment, according to their respective occupied lengths. The segment-average free strain is obtained by adding the product of the pre-switching type's occupied length and the pre-switching free longitudinal strain parameter, and the product of the post-switching type's occupied length and the post-switching free longitudinal strain parameter, and then dividing by the predicted open tension control segment length. The length-weighted approach is used because the flexible strip portions corresponding to the pre-switching and post-switching types occupy different path lengths in series within the same open tension control segment, and their unloaded natural length variation contributes proportionally to the total segment length.

[0108] The longitudinal flexibility of a section is used to represent the equivalent longitudinal length change caused by unit tension in the predicted open tension control section under the current two strip occupancy states. The section longitudinal flexibility is obtained by dividing the occupancy length of the post-switching type by the predicted post-switching longitudinal stiffness parameter, dividing the occupancy length of the pre-switching type by the pre-switching longitudinal stiffness parameter, and then adding the longitudinal flexibility formed by the post-switching strip to the longitudinal flexibility formed by the pre-switching strip. Since the flexible strip portions corresponding to the pre-switching and post-switching types are subjected to series forces along the transmission path, the longitudinal flexibility is added according to the flexibility corresponding to each series length, conforming to the deformation relationship of a one-dimensional axially elastic series system.

[0109] For example, when the first upstream boundary is 10.0 meters, the middle boundary is 12.0 meters, and the downstream boundary of the second open tension control section is 14.0 meters, the length of the second open tension control section is 2.0 meters. When the boundary position of the preset tension prediction model is 12.4 meters, the length occupied by the type after switching is 0.4 meters, and the length occupied by the type before switching is 1.6 meters. The average free strain of the section is (1.6×0.001+0.4×0.002)÷2.0=0.0012; the longitudinal flexibility of the section is 0.4÷120000+1.6÷100000, and the calculated result is approximately 0.00001933 m / Newton. This calculation result shows that within the same open tension control section, the flexible strip parts corresponding to the type before switching and the type after switching participate in the calculation of free longitudinal strain parameters and longitudinal flexibility according to the actual occupied length.

[0110] The input longitudinal stiffness parameters of the preset tension prediction model represent the switched longitudinal stiffness parameters used in this prediction. Normal prediction uses the currently corrected longitudinal stiffness parameters; candidate longitudinal stiffness parameter values ​​are selected using each candidate value; and the longitudinal stiffness parameters to be verified are verified using both the switched longitudinal stiffness parameters and the parameters to be verified. The longitudinal stiffness parameters before the switch are used as known longitudinal characteristic parameters during a type switch process and do not participate in the longitudinal stiffness candidate search.

[0111] S103. Predict the tension recursively based on the preset tension prediction model.

[0112] When predicting the first open tension control section, the first inlet tension feedback value, the first upstream velocity feedback value, and the intermediate velocity feedback value are respectively defined as upstream tension, upstream boundary velocity, and downstream boundary velocity. When predicting the second open tension control section, the intermediate tension feedback value, the intermediate velocity feedback value, and the second downstream velocity feedback value are respectively defined as upstream tension, upstream boundary velocity, and downstream boundary velocity. In the preset tension prediction model, upstream tension, upstream boundary velocity, and downstream boundary velocity are general input names; the first inlet tension feedback value, the first upstream velocity feedback value, and the intermediate velocity feedback value are used when predicting the first open tension control section, and the intermediate tension feedback value, the intermediate velocity feedback value, and the second downstream velocity feedback value are used when predicting the second open tension control section.

[0113] At the initial velocity sampling time of each prediction data window, the predicted tension is initialized with the upstream tension of the predicted open tension control segment. The initialized predicted tension is multiplied by the current segment's longitudinal compliance, divided by the length of the predicted open tension control segment, and added to the current segment's mean free strain to obtain the segment's mean total strain at the initial velocity sampling time. The segment's mean total strain represents the current segment's mean total longitudinal strain state, which includes both the elastic strain caused by tension and the material's free longitudinal strain.

[0114] The inlet longitudinal stiffness parameter and the inlet free longitudinal strain parameter are determined based on the inlet type. When the inlet type is the switched type, the inlet longitudinal stiffness parameter and the inlet free longitudinal strain parameter are the same as those used in this prediction. When the inlet type is the original type, both the inlet longitudinal stiffness parameter and the inlet free longitudinal strain parameter are the same. The upstream tension at the current velocity sampling moment is divided by the inlet longitudinal stiffness parameter, and then the inlet free longitudinal strain parameter is added to obtain the total inlet strain at the current velocity sampling moment. Within the preset longitudinal stiffness correction data window, preset longitudinal stiffness verification window, preset boundary position correction data window, and preset boundary position verification window, the upstream tension is the measured upstream tension aligned with the current velocity sampling moment. In the future prediction of the preset control period in step S106, the upstream tension is the intermediate tension feedback value specified in step S106.

[0115] The longitudinal stiffness parameter and free longitudinal strain parameter of the outlet are determined based on the outlet type. When the outlet type is the post-switching type, the longitudinal stiffness parameter of the outlet is taken from the post-switching longitudinal stiffness parameter used in this prediction, and the free longitudinal strain parameter of the outlet is also taken from the post-switching free longitudinal strain parameter. When the outlet type is the pre-switching type, both the pre-switching longitudinal stiffness parameter and the pre-switching free longitudinal strain parameter are taken. The predicted tension at the previous velocity sampling time is divided by the longitudinal stiffness parameter of the outlet, and then the free longitudinal strain parameter of the outlet is added to obtain the total outlet strain.

[0116] Between two adjacent velocity sampling moments, the difference between the downstream boundary velocity and the upstream boundary velocity is first calculated. Then, the product of the upstream boundary velocity and the total inlet strain is added to the velocity difference, and the product of the downstream boundary velocity and the total outlet strain is subtracted. The result is multiplied by a preset velocity sampling period and then divided by the predicted open tension control segment length to obtain the segment's average total strain increment. This segment's average total strain increment is accumulated and added to the segment's average total strain at the previous velocity sampling moment to obtain the segment's average total strain at the current velocity sampling moment.

[0117] It should be noted that the recursive relationship of the average total strain of the section simultaneously takes into account the change in the geometric length of the section caused by the difference between the downstream boundary velocity and the upstream boundary velocity, as well as the total strain transport carried by the flexible strip portion corresponding to the inlet and outlet types. When the inlet total strain, outlet total strain, and upstream boundary velocity are used as the current calculation inputs, and the downstream boundary velocity is higher than the upstream boundary velocity, the difference between the downstream and upstream boundary velocities increases the average total strain of the section, which is consistent with the actual transmission state of the flexible strip undergoing longitudinal stretching when the downstream traction speed increases.

[0118] The predicted tension at the current velocity sampling time is obtained by subtracting the average free strain of the current segment from the average total strain of the current segment, multiplying the difference by the predicted open tension control segment length, and then dividing by the longitudinal flexibility of the current segment. Since the subtraction of the average free strain of the segment from the average total strain retains the equivalent elastic strain formed by longitudinal force, this calculation will not directly mistake the natural expansion and contraction of the material without load for elastic deformation caused by tension.

[0119] The preset velocity sampling period is jointly determined by the effective sampling period of the velocity measuring device and the controller's operation period, and is set to a preset value that ensures stable execution of velocity sampling, tension data time alignment, and the recursive execution of the preset tension prediction model. The preset velocity sampling period remains consistent within the same calibration data window, verification window, and control cycle, ensuring that the predicted tension sequences of different candidate states have the same sampling time.

[0120] The preset tension prediction model is used for control-scale tension prediction of flexible strips under continuous tension and normal traction conditions. The average total strain of the section represents the equivalent average longitudinal strain of an open tension control section at the preset velocity sampling period scale, and is not used to describe the propagation of high-frequency longitudinal stress waves in the strip. When the equipment experiences strip breakage, significant slack, or the tension measuring device exceeds its range, the equipment's basic protection logic suspends the model correction of this embodiment and the search for the second downstream target velocity setpoint. The process of this embodiment will resume after the continuous tension state is restored. This implementation condition avoids the use of a one-dimensional longitudinal tension model under conditions where a stable longitudinal force relationship does not exist.

[0121] S104. When the state quantity at the transition boundary position is within the preset boundary correction range, correct the longitudinal stiffness parameter after switching.

[0122] like Figure 3 As shown, after the transition boundary position state variable enters the preset boundary correction interval, the controller reads the second tension feedback value, the intermediate tension feedback value, the intermediate velocity feedback value, and the second downstream velocity feedback value, and uses the second open tension control section as the predicted open tension control section for longitudinal stiffness correction. The longitudinal stiffness parameter correction uses the second tension feedback value as the measured benchmark for the model residual, and the intermediate tension feedback value as the upstream tension input of the model. The measured tension of the first open tension control section is not used to replace the second tension feedback value.

[0123] The preset longitudinal stiffness step size is determined during the equipment commissioning phase based on the degree of dispersion in the repeated calibration of the longitudinal stiffness parameter after switching, the fluctuation range of the longitudinal stiffness parameter during normal production, and the desired correction resolution. The preset longitudinal stiffness step size must ensure that the longitudinal stiffness value obtained by subtracting this step size from the switched longitudinal stiffness parameter is greater than zero; if the value obtained by subtracting the preset longitudinal stiffness step size is not greater than zero, this subtraction result is not used as a candidate longitudinal stiffness parameter value. The candidate longitudinal stiffness parameter values ​​are determined by the switched longitudinal stiffness parameter itself, the value obtained by adding the preset longitudinal stiffness step size to the switched longitudinal stiffness parameter, and the subtraction result that satisfies the positive value condition.

[0124] The preset longitudinal stiffness correction data window consists of a continuous preset number of velocity sampling times. Its duration is determined during the equipment commissioning phase based on the tension sensor response time and the available dwell time of the transition boundary within the second open tension control section. Before starting a preset longitudinal stiffness correction data window, the controller determines, based on the current transition boundary advance speed, that the transition boundary will still be within the preset boundary correction range before the end of the window, thus avoiding crossing the boundary of the second open tension control section within a candidate residual calculation window.

[0125] The duration of the preset longitudinal stiffness verification window is the same as the duration of the preset longitudinal stiffness correction data window, and both windows contain the same number of continuous velocity sampling moments. The preset longitudinal stiffness verification window is set after the preset longitudinal stiffness correction data window and does not share velocity sampling moments with it. Before starting the preset longitudinal stiffness verification window, the controller confirms, based on the current transition boundary advance speed, that the transition boundary is still within the preset boundary correction interval before the preset longitudinal stiffness verification window ends; if this condition is not met, the longitudinal stiffness parameter to be verified this time is not written into the corrected longitudinal stiffness parameter, and longitudinal stiffness correction is re-executed during the next type switch.

[0126] Within the preset longitudinal stiffness correction data window, a preset tension prediction model is invoked for each candidate longitudinal stiffness parameter value. The boundary position input values ​​use the same transition boundary position state quantity sequence, the upstream tension uses the same intermediate tension feedback value sequence, the upstream boundary velocity uses the same intermediate velocity feedback value sequence, and the downstream boundary velocity uses the same second downstream velocity feedback value sequence. Only the switched longitudinal stiffness parameters used in the prediction are changed, thereby obtaining the predicted tension sequence corresponding to each candidate longitudinal stiffness parameter value.

[0127] For each candidate longitudinal stiffness parameter value, the predicted tension sequence is paired up in a non-repeating pairing manner. For each pair, the two predicted tension sequences are subtracted point-by-point at the same velocity sampling time. The square root of the average of the squared differences is then taken to obtain the root mean square difference for that pair. The maximum value among all the root mean square differences is determined as the longitudinal stiffness predicted difference. The longitudinal stiffness predicted difference is used to determine whether changing the longitudinal stiffness within the current preset longitudinal stiffness correction data window can generate a predicted tension change exceeding the tension measurement fluctuation level.

[0128] A preset longitudinal stiffness identifiable threshold is used to determine whether the difference in predicted tension caused by different longitudinal stiffness candidate parameter values ​​exceeds the resolvable level of current measurement and operational fluctuations. During equipment commissioning, a preset number of non-overlapping tension data windows are obtained under the condition that the type and drive state remain stable. For each tension data window, the average value of the second tension feedback value of that tension data window is subtracted from the second tension feedback value at each speed sampling time, and the root mean square is calculated to obtain the root mean square of tension fluctuation. At the same time, the allowable measurement error of the tension measuring device used to obtain the second tension feedback value after zero-point calibration and range calibration under the current range is read. The maximum value among the preset number of root mean squares of tension fluctuations and the larger value among the allowable measurement errors of the tension measuring device used to obtain the second tension feedback value are multiplied by a preset safety factor to obtain the preset longitudinal stiffness identifiable threshold. The preset safety factor is set and stored in the controller during the equipment commissioning phase according to the allowable misjudgment level. When the longitudinal stiffness prediction difference caused by the longitudinal stiffness candidate parameter values ​​is greater than the preset longitudinal stiffness identifiable threshold, the longitudinal stiffness candidate residual is calculated again.

[0129] When the difference in longitudinal stiffness prediction exceeds the preset identifiable threshold for longitudinal stiffness, for each candidate longitudinal stiffness parameter value, its predicted tension sequence is subtracted point by point from the second tension feedback value at the same velocity sampling time. The square root of the average of the squared differences is then used to obtain the candidate longitudinal stiffness residual for that candidate parameter value. The candidate longitudinal stiffness residual is used to characterize the prediction error level when using that candidate longitudinal stiffness parameter value under the same measured tension data. When the difference in longitudinal stiffness prediction is not greater than the preset identifiable threshold for longitudinal stiffness, no candidate longitudinal stiffness residual is formed. The corrected longitudinal stiffness parameter remains the same as the switched longitudinal stiffness parameter, and the current data window does not enter the boundary position correction judgment dependent on the reduction of the longitudinal stiffness residual.

[0130] The candidate longitudinal stiffness residuals corresponding to the switched longitudinal stiffness parameters are determined as the current longitudinal stiffness residuals. The minimum value among all candidate longitudinal stiffness residuals is determined as the minimum longitudinal stiffness residual. The reduction in longitudinal stiffness residuals is obtained by subtracting the minimum longitudinal stiffness residual from the current longitudinal stiffness residuals. The reduction in longitudinal stiffness residuals represents the reduction in tension prediction residuals within the current data window after switching from the switched longitudinal stiffness parameters to the optimal candidate longitudinal stiffness parameter values.

[0131] A preset residual reduction threshold is used to determine whether the reduction in residual caused by the candidate state exceeds the natural change of the residual itself under stable operating conditions. During equipment commissioning, a preset number of non-overlapping data windows are continuously acquired under the condition that the type and operating state remain stable. The tension prediction residual of each data window is calculated using the same preset tension prediction model, and the absolute value of the difference between the tension prediction residuals of adjacent data windows is calculated. At the same time, the allowable measurement error of the tension measuring device used to obtain the second tension feedback value is obtained. The larger of the maximum value of the absolute value of the residual difference and the allowable measurement error is determined as the preset residual reduction threshold. The reduction in longitudinal stiffness residual, the reduction in longitudinal stiffness verification residual, the reduction in boundary position residual, and the reduction in boundary position verification residual all use the same tension dimension, so they can be compared using the same preset residual reduction threshold.

[0132] When the reduction in longitudinal stiffness residual exceeds a preset residual reduction threshold and the minimum longitudinal stiffness residual is obtained from a unique longitudinal stiffness candidate parameter value, the longitudinal stiffness candidate parameter value that yielded the minimum longitudinal stiffness residual is determined as the longitudinal stiffness parameter to be verified. The longitudinal stiffness parameter to be verified only represents candidate results within the calibration data window and is not replaced by the calibrated longitudinal stiffness parameter before verification is completed. If the minimum longitudinal stiffness residual is not obtained from a unique longitudinal stiffness candidate parameter value, a longitudinal stiffness parameter to be verified is not generated to avoid arbitrarily modifying the longitudinal stiffness parameter when candidate values ​​cannot be uniquely determined.

[0133] Using the switched longitudinal stiffness parameter and the longitudinal stiffness parameter to be verified as input longitudinal stiffness parameters, respectively, two predicted tension sequences are calculated under the same conditions of state variables at the transition boundary position, intermediate tension feedback value, intermediate velocity feedback value, and second downstream velocity feedback value. The root mean square residuals between the two predicted tension sequences and the second tension feedback value at the same velocity sampling time are calculated. The former is determined as the residual before longitudinal stiffness verification, and the latter is determined as the residual after longitudinal stiffness verification.

[0134] Subtracting the residual after longitudinal stiffness verification from the residual before verification yields the reduction in longitudinal stiffness verification residual. If the reduction in longitudinal stiffness verification residual is greater than the preset residual reduction threshold, it indicates that the longitudinal stiffness parameter to be verified can still reduce the prediction residual in subsequent data independent of the correction data window, and the longitudinal stiffness parameter to be verified is determined as the corrected longitudinal stiffness parameter. If the reduction in longitudinal stiffness verification residual is not greater than the preset residual reduction threshold, the corrected longitudinal stiffness parameter is maintained as the switched longitudinal stiffness parameter.

[0135] To illustrate the screening and verification process of candidate longitudinal stiffness parameters, a preset velocity sampling period of 0.02 seconds is established. The preset longitudinal stiffness correction data window includes six velocity sampling times: 0.00 seconds, 0.02 seconds, 0.04 seconds, 0.06 seconds, 0.08 seconds, and 0.10 seconds. The upstream boundary position of the second open tension control section is 12.00 meters, the length of the second open tension control section is 2.00 meters, the transition boundary position state quantity at the start of the preset longitudinal stiffness correction data window is 12.40 meters, and the transition boundary advance speed is 1.00 meters / second. Therefore, the transition boundary position state quantities corresponding to the six velocity sampling times are 12.40 meters, 12.4... The sampling times were 2 meters, 12.44 meters, 12.46 meters, 12.48 meters, and 12.50 meters. The intermediate tension feedback value at each of the six velocity sampling times was 500.00 Newtons, the intermediate velocity feedback value was 3.00 meters per second, and the second downstream velocity feedback value was 3.05 meters per second. The longitudinal stiffness parameter before switching was 100,000 Newtons, and the longitudinal stiffness parameter after switching was 120,000 Newtons. The preset longitudinal stiffness step size was 5,000 Newtons. The free longitudinal strain parameter before switching was 0.001, and the free longitudinal strain parameter after switching was 0.002. This resulted in three candidate longitudinal stiffness parameter values ​​of 115,000 Newtons, 120,000 Newtons, and 125,000 Newtons. Following step S102, the pre-switching type occupancy length, post-switching type occupancy length, average free strain of the segment, and longitudinal flexibility of the segment are determined at each velocity sampling time. Then, following step S103, the tension is recursively predicted. The predicted tension sequence obtained using 115000 Newtons is 500.00 Newtons, 551.81 Newtons, 602.14 Newtons, 651.02 Newtons, 698.49 Newtons, and 744.60 Newtons. The predicted tension sequence obtained using 120000 Newtons is 500.00 Newtons, 5... The predicted tension sequences obtained using 125,000 Newtons are 51.85 Newtons, 602.23 Newtons, 651.20 Newtons, 698.77 Newtons, and 745.00 Newtons, respectively. The maximum value of the root mean square difference among the three predicted tension sequences calculated using a non-repeating pairing method is approximately 0.418 Newtons, which is determined as the longitudinal stiffness predicted difference.Assuming a preset longitudinal stiffness identification threshold of 0.15 Newtons, and since 0.418 Newtons is greater than 0.15 Newtons, candidate longitudinal stiffness residuals are calculated. Assuming the second tension feedback values ​​at the six velocity sampling times are 500.00 Newtons, 552.28 Newtons, 602.02 Newtons, 651.56 Newtons, 698.83 Newtons, and 745.48 Newtons respectively, each predicted tension sequence is subtracted from the second tension feedback value sequence point by point at the same velocity sampling time. The square root of the average of the squared differences is then taken to obtain the candidate longitudinal stiffness residuals corresponding to 115000 Newtons, 120000 Newtons, and 125000 Newtons, which are approximately 0.485 Newtons, 0.315 Newtons, and 0.238 Newtons respectively. The 0.315 Newtons corresponding to the longitudinal stiffness parameter of 120,000 Newtons after the switch is determined as the current longitudinal stiffness residual, and the 0.238 Newtons corresponding to 125,000 Newtons is determined as the minimum longitudinal stiffness residual, resulting in a reduction of approximately 0.077 Newtons in the longitudinal stiffness residual. Assuming a preset residual reduction threshold of 0.05 Newtons, since 0.077 Newtons is greater than 0.05 Newtons, and the minimum longitudinal stiffness candidate residual is only obtained from 125,000 Newtons, 125,000 Newtons is determined as the longitudinal stiffness parameter to be verified. In the preset longitudinal stiffness verification window, which is located after the preset longitudinal stiffness correction data window and does not share the velocity sampling time with it, the transition boundary position state quantity at the beginning of the window is set to 12.52 meters, the intermediate tension feedback value is 510.00 Newtons, the intermediate velocity feedback value is 3.02 meters / second, the second downstream velocity feedback value is 3.06 meters / second, and the transition boundary advance velocity remains at 1.00 meters / second; the predicted tension sequence obtained by using the switched longitudinal stiffness parameter of 120,000 Newtons is 510.00 Newtons, 551.95 Newtons, and 592 Newtons respectively. The predicted tension sequences obtained using the longitudinal stiffness parameter to be verified, 125,000 N, are 69 N, 632.27 N, 670.72 N, and 708.05 N. The corresponding second tension feedback values ​​at the velocity sampling time are 510.00 N, 551.78 N, 593.08 N, 632.33 N, 671.16 N, and 708.19 N. Using the same root mean square residual calculation method, the residual before longitudinal stiffness verification was approximately 0.256 Newtons, and the residual after longitudinal stiffness verification was approximately 0.191 Newtons, with a reduction of approximately 0.064 Newtons in the longitudinal stiffness verification residual. Since 0.064 Newtons is greater than the preset residual reduction threshold of 0.05 Newtons, 125,000 Newtons was determined as the corrected longitudinal stiffness parameter.The above values ​​are only used to illustrate the process by which the controller determines the corrected longitudinal stiffness parameters based on the candidate longitudinal stiffness parameter values, the predicted tension sequence, and the second tension feedback value. They do not represent actual equipment test results, nor are they intended to limit the scope of protection of this invention.

[0136] It should be noted that after the longitudinal stiffness parameter correction is completed, boundary position correction is only initiated when a reduction in longitudinal stiffness residuals has been formed, the reduction in longitudinal stiffness residuals is not greater than the preset residual reduction threshold, and the corrected longitudinal stiffness parameter is equal to the switched longitudinal stiffness parameter. This judgment indicates that changing the switched longitudinal stiffness parameter within the current preset longitudinal stiffness correction data window has not resulted in a residual improvement exceeding the preset residual reduction threshold, and the corrected longitudinal stiffness parameter remains the same as the switched longitudinal stiffness parameter. When the reduction in longitudinal stiffness residuals is greater than the preset residual reduction threshold but no unique longitudinal stiffness parameter to be verified has been formed, the corrected boundary position state quantity continues to advance according to the transition boundary advancement speed; when the longitudinal stiffness parameter to be verified fails verification within the preset longitudinal stiffness verification window, the corrected boundary position state quantity continues to advance according to the transition boundary advancement speed; when the longitudinal stiffness prediction difference is not greater than the preset longitudinal stiffness identifiable threshold, no reduction in longitudinal stiffness residuals is formed, and boundary position correction based on the reduction in longitudinal stiffness residuals is not initiated.

[0137] The condition for initiating boundary position correction, "the corrected longitudinal stiffness parameter equals the switched longitudinal stiffness parameter," indicates that no new longitudinal stiffness parameter was accepted during the longitudinal stiffness correction process corresponding to this type switch. For each type switch, only one longitudinal stiffness parameter correction process, consisting of a preset longitudinal stiffness correction data window and a preset longitudinal stiffness verification window, is executed. When the longitudinal stiffness parameter to be verified passes verification, it is determined as the corrected longitudinal stiffness parameter, and boundary position correction is not performed again during this type switch. Boundary position correction is initiated when the reduction in longitudinal stiffness residual is not greater than the preset residual reduction threshold and the corrected longitudinal stiffness parameter is still equal to the switched longitudinal stiffness parameter. If the longitudinal stiffness prediction difference is not greater than the preset longitudinal stiffness identifiable threshold, the minimum longitudinal stiffness candidate residual cannot be obtained from the unique longitudinal stiffness candidate parameter value, the longitudinal stiffness parameter to be verified fails verification, or the verification window does not meet the position conditions, the corrected longitudinal stiffness parameter is not updated, and boundary position correction is not initiated. The corrected boundary position state quantity continues to advance according to the transition boundary advancement speed. After detecting the next type switch, the longitudinal stiffness parameter correction process is re-executed.

[0138] S105. Correct the corrected boundary position state quantity based on the second tension feedback value and the preset tension prediction model.

[0139] like Figure 4As shown, the duration of the preset boundary position correction data window is determined during the equipment debugging phase based on the response time of the tension measuring device used to acquire the second tension feedback value and the available dwell time of the transition boundary within the second open tension control section, and consists of continuous speed sampling moments. Specifically, the controller first determines the shortest data duration that can reflect tension changes based on the response time of the tension measuring device used to acquire the second tension feedback value, and then calculates the available dwell time of the transition boundary within the preset boundary correction interval based on the current transition boundary advance speed and the length of the preset boundary correction interval; the duration of the preset boundary position correction data window does not exceed the available dwell time. Before starting the preset boundary position correction data window, the controller confirms, based on the current transition boundary advance speed, that the current boundary position candidate value, negative boundary position candidate value, and positive boundary position candidate value all remain within the second open tension control section before the window ends.

[0140] The duration of the preset boundary position verification window is the same as the duration of the preset boundary position correction data window, and both windows contain the same number of continuous velocity sampling moments. The preset boundary position verification window is set after the preset boundary position correction data window and does not share velocity sampling moments with it. Before starting the preset boundary position verification window, the controller confirms, based on the current transition boundary advancement speed, that the transition boundary and the boundary position to be verified are still within the valid position range before the verification window ends; if this condition is not met, the boundary position to be verified is not used to update the corrected boundary position state quantity.

[0141] When entering boundary position correction, the transition boundary position state quantity at the start of the preset boundary position correction data window is determined as the current boundary position candidate value; the position obtained by subtracting the preset position step size from the transition boundary position state quantity is determined as the negative boundary position candidate value; the position obtained by adding the preset position step size to the transition boundary position state quantity is determined as the positive boundary position candidate value. The preset position step size is determined during the equipment debugging phase based on the cumulative error of the transition boundary advance speed integral, the speed detection error, the position error caused by the slippage between the drive roller and the strip, and the position correction resolution required by the controller.

[0142] The preset boundary safety distance setting ensures that the current boundary position candidate value, negative boundary position candidate value, and positive boundary position candidate value remain within the effective position range of the second open tension control section within the preset boundary position correction data window. Before activating the preset boundary position correction data window, the controller predicts the positions of the current boundary position candidate value, negative boundary position candidate value, and positive boundary position candidate value at the end of the window according to the current transition boundary advancement speed. If any predicted position exceeds the second open tension control section, the boundary position candidate screening is not activated this time, and the corrected boundary position state quantity continues to advance according to the transition boundary advancement speed.

[0143] Within the preset boundary position correction data window, the intermediate tension feedback value, intermediate velocity feedback value, and second downstream velocity feedback value are used as the upstream tension, upstream boundary velocity, and downstream boundary velocity of the second open tension control section, respectively. The corrected longitudinal stiffness parameter is used as the switched longitudinal stiffness parameter for prediction. The corresponding predicted tension sequence is calculated using each boundary position candidate value as the boundary position input value of the preset tension prediction model.

[0144] The candidate boundary positions are not constant throughout the entire data window. For each candidate boundary position, starting from the initial candidate value at the beginning of the preset boundary position correction data window, the position is synchronously advanced according to the time integral of the transition boundary advancement speed between adjacent velocity sampling times. Since each candidate value advances using the same transition boundary advancement speed, the initial position offset between the three candidate boundaries remains constant within the window. What is being compared are different initial boundary position assumptions under the same actual boundary advancement trajectory, rather than static material boundaries.

[0145] For each candidate boundary position value, the predicted tension sequence corresponding to that candidate boundary position value is subtracted point by point from the second tension feedback value at the same velocity sampling time. The square root of the average of the squared differences is then used to obtain the candidate boundary position residual. The candidate boundary position residual calculated from the candidate sequence formed by the transition boundary position state variables at the start of the preset boundary position correction data window is determined as the current candidate boundary position residual. The minimum value among all candidate boundary position residuals is determined as the minimum candidate boundary position residual.

[0146] Subtracting the minimum boundary position candidate residual from the current boundary position candidate residual yields the reduction in boundary position residual. If the reduction in boundary position residual exceeds a preset residual reduction threshold and the minimum boundary position candidate residual is obtained from a unique boundary position candidate value, the boundary position candidate value that yielded the minimum boundary position candidate residual is determined as the boundary position to be verified. If the reduction in boundary position residual does not exceed the preset residual reduction threshold or the minimum boundary position candidate residual cannot be obtained from a unique candidate value, no boundary position to be verified is formed, and the corrected boundary position state variable continues to advance according to the transition boundary advancement speed.

[0147] Within the preset boundary position verification window, both the transition boundary position state quantity and the boundary position to be verified start from their current values ​​at the beginning of the verification window and continue to advance according to the transition boundary advancement speed. Using the advanced transition boundary position state quantity sequence and the boundary position sequence to be verified as the boundary position input values ​​of the preset tension prediction model, two predicted tension sequences are calculated.

[0148] Calculate the root mean square residuals between the two predicted tension sequences and the second tension feedback value at the same velocity sampling time. The residual obtained using the transition boundary position state quantity is determined as the residual before boundary position verification, and the residual obtained using the boundary position to be verified is determined as the residual after boundary position verification. Subtract the residual after boundary position verification from the residual before boundary position verification to obtain the reduction in boundary position verification residual.

[0149] When the reduction in the boundary position verification residual exceeds a preset residual reduction threshold, the boundary position to be verified, which has progressed to the current tension calculation time, is determined as the corrected boundary position state quantity. When the reduction in the boundary position verification residual does not exceed the preset residual reduction threshold, the boundary position to be verified is not used, and the corrected boundary position state quantity continues to advance from the original corrected boundary position state quantity according to the transition boundary advancement speed. It should be noted that the boundary position state quantity correction only changes the spatial position of the material boundary at the correction time, and does not change the motion relationship of the material boundary continuing to advance downstream according to the transition boundary advancement speed in subsequent times.

[0150] To illustrate the screening and verification process for candidate boundary position values, in another illustrative implementation scenario, both the switched and corrected longitudinal stiffness parameters are 120,000 Newtons. In the pre-defined longitudinal stiffness correction data window, the current longitudinal stiffness residual is 0.20 Newtons, the minimum longitudinal stiffness residual is 0.18 Newtons, and the reduction in longitudinal stiffness residual is 0.02 Newtons. Assuming a preset residual reduction threshold of 0.05 Newtons, since the reduction in longitudinal stiffness residual is not greater than the preset residual reduction threshold, and the corrected longitudinal stiffness parameter is equal to the switched longitudinal stiffness parameter, boundary position correction is initiated.

[0151] The preset velocity sampling period is set to 0.02 seconds, and the preset boundary position correction data window includes six velocity sampling times: 0.00 seconds, 0.02 seconds, 0.04 seconds, 0.06 seconds, 0.08 seconds, and 0.10 seconds. The upstream boundary position of the second open tension control section is 12.00 meters, the length of the second open tension control section is 2.00 meters, the transition boundary position state quantity at the start of the preset boundary position correction data window is 12.40 meters, and the preset position step size is 0.05 meters. This forms a candidate value for the current boundary position of 12.40 meters, a candidate value for the negative boundary position of 12.35 meters, and a candidate value for the positive boundary position of 12.45 meters.

[0152] The transition boundary advance speed is set to 1.00 m / s, the intermediate tension feedback value to 500.00 N, the intermediate velocity feedback value to 3.00 m / s, and the second downstream velocity feedback value to 3.05 m / s. The longitudinal stiffness parameter before switching is 100,000 N, and the longitudinal stiffness parameter after switching is 120,000 N. The free longitudinal strain parameter before switching is 0.001, and the free longitudinal strain parameter after switching is 0.002. Within the preset boundary position correction data window, the three boundary position candidate values ​​are synchronously advanced from their respective initial candidate values ​​according to the transition boundary advance speed, and the corresponding predicted tension sequences are calculated according to steps S102 and S103.

[0153] The predicted tension sequences obtained with 12.35 meters as the candidate initial value are 500.00 N, 551.63 N, 601.80 N, 650.56 N, 697.96 N, and 744.01 N, respectively; the predicted tension sequences obtained with 12.40 meters as the candidate initial value are 500.00 N, 551.85 N, 602.23 N, 651.20 N, 698.77 N, and 745.00 N, respectively; and the predicted tension sequences obtained with 12.45 meters as the candidate initial value are 500.00 N, 552.07 N, 602.67 N, 651.83 N, 699.60 N, and 746.00 N, respectively. Let the second tension feedback values ​​at the sampling time at the same speed be 500.00 N, 552.27 N, 602.47 N, 651.93 N, 699.50 N and 746.15 N respectively.

[0154] The three predicted tension sequences were subtracted point-by-point from the second tension feedback value sequence at the same sampling speed. The square root of the average of the squared differences was then taken to obtain the candidate residuals for the three boundary positions at 12.35 meters, 12.40 meters, and 12.45 meters, which were approximately 1.273 Newtons, 0.662 Newtons, and 0.143 Newtons, respectively. The 0.662 Newtons corresponding to 12.40 meters was determined as the current candidate residual for the boundary position, and the 0.143 Newtons corresponding to 12.45 meters was determined as the minimum candidate residual for the boundary position, resulting in a reduction of approximately 0.519 Newtons in the boundary position residual. Since 0.519 Newtons is greater than the preset residual reduction threshold of 0.05 Newtons, and the minimum candidate residual for the boundary position is only obtained from 12.45 meters, 12.45 meters was determined as the boundary position to be verified.

[0155] In the preset boundary position verification window, which is located after the preset boundary position correction data window and does not share the same velocity sampling time, the transition boundary position state quantity and the boundary position to be verified continue to advance according to the transition boundary advancement speed. Let the transition boundary position state quantity and the boundary position to be verified at the beginning of the verification window be 12.50 meters and 12.55 meters, respectively, the intermediate tension feedback value be 505.00 Newtons, the intermediate velocity feedback value be 3.00 meters / second, and the second downstream velocity feedback value be 3.04 meters / second. The predicted tension sequences obtained with a validation window starting at 12.50 meters are 505.00 N, 546.89 N, 587.59 N, 627.14 N, 665.56 N, and 702.89 N, respectively; the predicted tension sequences obtained with a validation window starting at 12.55 meters are 505.00 N, 547.07 N, 587.95 N, 627.66 N, 666.23 N, and 703.71 N, respectively. Assuming the second tension feedback values ​​at the same sampling speed are 505.00 N, 546.97 N, 588.10 N, 627.61 N, 666.33 N, and 703.61 N respectively, the calculated residual before boundary position verification is approximately 0.515 N, and the residual after boundary position verification is approximately 0.096 N, representing a reduction of approximately 0.419 N in the residual. Since 0.419 N is greater than the preset residual reduction threshold of 0.05 N, the boundary position to be verified at the current tension calculation time is determined as the corrected boundary position state quantity.

[0156] The above values ​​are only used to illustrate the process by which the controller corrects the boundary position state quantity based on the boundary position candidate value, the predicted tension sequence, and the second tension feedback value. They do not represent actual equipment test results, nor are they intended to limit the scope of protection of this invention.

[0157] S106. Determine the candidate second downstream speed setpoint based on the second downstream speed feedback value, and determine the second downstream target speed setpoint.

[0158] See Figure 5 The preset minimum drive speed and preset maximum drive speed are determined by the second downstream drive mechanism, transmission mechanism, and the continuous operating speed range allowed by the production process, and are written into the controller during equipment commissioning. When performing the second downstream target speed setpoint search, the second downstream speed feedback value should be between the preset minimum drive speed and the preset maximum drive speed; if the current second downstream speed feedback value has exceeded this normal operating range, the equipment's basic speed protection logic will process it first, and the tension prediction speed search of this embodiment will not be performed.

[0159] The preset control cycle represents the predicted control duration for this speed search, determined based on the speed response time and tension control update cycle of the second downstream drive mechanism. The allowable speed change per preset control cycle is determined based on the allowable acceleration and deceleration of the second downstream drive mechanism and the preset control cycle. When using a symmetrical speed change, the smaller of the absolute values ​​of the allowable acceleration and deceleration can be multiplied by the preset control cycle, ensuring that both positive and negative speed adjustments do not exceed the drive mechanism's capacity.

[0160] The speed value obtained by subtracting the preset allowable speed change per single control cycle from the second downstream speed feedback value is used as the candidate lower speed limit value, and the speed value obtained by adding the preset allowable speed change per single control cycle to the second downstream speed feedback value is used as the candidate upper speed value. The intersection of the interval formed by the candidate lower speed limit value and the candidate upper speed value with the interval formed by the preset minimum drive speed and the preset maximum drive speed is used to obtain the candidate second downstream speed setpoint range. The candidate second downstream speed setpoint range simultaneously satisfies the absolute speed limit and the allowable speed change limit per single control cycle of the second downstream drive mechanism.

[0161] The preset speed search step size is determined based on the minimum executable speed command resolution and the required tension control resolution of the second downstream drive mechanism, and is set to be no less than the minimum speed command change that the drive controller can stably resolve. Starting from the lower limit of the candidate second downstream speed setpoint range, candidate second downstream speed setpoints are sequentially formed according to the preset speed search step size until they do not exceed the upper limit of the range.

[0162] For each candidate second downstream velocity setpoint, the intermediate tension feedback value, intermediate velocity feedback value, and transition boundary advance velocity are read at the current tension calculation time. At each future velocity sampling time within the preset control cycle, the intermediate tension feedback value, intermediate velocity feedback value, and transition boundary advance velocity are all taken from the measured values ​​at the current tension calculation time and remain unchanged. Each candidate second downstream velocity setpoint is used as the predicted execution velocity under that candidate condition throughout its corresponding entire preset control cycle, and is input as a constant downstream boundary velocity into the preset tension prediction model. The currently corrected longitudinal stiffness parameter is used as the switched longitudinal stiffness parameter in the prediction. The corrected boundary position state quantity stored at the current tension calculation time is read, and the corrected boundary position state quantity is used as the boundary position input value input into the preset tension prediction model at the start velocity sampling time of the preset control cycle.

[0163] Within the preset control period, predictions are made according to the temporal order of future velocity sampling times. After each preset velocity sampling period, the measured value of the transition boundary advance velocity at the current tension calculation time is multiplied by the preset velocity sampling period to obtain the boundary advance distance within that preset velocity sampling period. The boundary advance distances from the start of the preset control period to the current future velocity sampling time are then accumulated to obtain the cumulative boundary advance distance at the current future velocity sampling time. The sum of the corrected boundary position state quantity stored at the current tension calculation time and the cumulative boundary advance distance is used as the boundary position input value for the preset tension prediction model at the current future velocity sampling time. The cumulative boundary advance distance and the boundary position input values ​​determined accordingly for each future velocity sampling time are only used for prediction calculations within the current preset control period and are not written into the corrected boundary position state quantity.

[0164] Based on the boundary position input value of the preset tension prediction model at the current future velocity sampling time, the type occupancy length after switching, the type occupancy length before switching, the inlet type, the outlet type, the average free strain of the section, and the longitudinal flexibility of the section are redefined. Then, the predicted tension of the second open tension control section at the current future velocity sampling time is calculated using the recursive relationship of the average total strain of the section recorded in step S103. The predicted tension sequence of the current candidate second downstream velocity setpoint is obtained according to the velocity sampling time sequence, and the predicted tension at the last velocity sampling time in the predicted tension sequence is determined as the target predicted tension. The intermediate tension feedback value, intermediate velocity feedback value, and transition boundary advance speed maintain the value rules of the measured values ​​at the current tension calculation time within the preset control cycle and are only used for future prediction in the current preset control cycle; when entering the next preset control cycle, the intermediate tension feedback value, intermediate velocity feedback value, and transition boundary advance speed are reacquired.

[0165] The preset tension setting value is the tension setting value of the second open tension control section under the current switching type and current production process, which is given in advance by the production formula or process database. For each candidate second downstream speed setting value, the absolute value of the difference between its target predicted tension and the preset tension setting value is calculated, and the candidate second downstream speed setting value with the smallest absolute value is determined as the second downstream target speed setting value. If two candidate second downstream speed setting values ​​obtain the same minimum absolute value in the discrete search, the candidate second downstream speed setting value that first obtains the minimum value is selected according to the search order of candidate second downstream speed setting values ​​from smallest to largest, ensuring that the second downstream target speed setting value is uniquely determined. When the candidate second downstream speed setting value range is empty, it means that the current second downstream speed feedback value or the equipment's allowable speed constraint does not meet the current prediction search conditions. No new second downstream target speed setting value is output, and the current safe speed is maintained by the equipment's original speed closed loop, and a new judgment is made in the next control cycle.

[0166] To illustrate the formation and selection process of the candidate second downstream speed setpoint, the current second downstream speed feedback value is set to 3.00 m / s, the preset minimum drive speed is 2.80 m / s, the preset maximum drive speed is 3.20 m / s, the preset allowable speed change per control cycle is 0.03 m / s, and the preset speed search step size is 0.01 m / s.

[0167] The result of subtracting 0.03 m / s from 3.00 m / s, resulting in 2.97 m / s, is determined as a candidate lower speed limit. The result of adding 0.03 m / s to 3.00 m / s, resulting in 3.03 m / s, is determined as a candidate upper speed limit. The intersection of the interval formed by 2.97 m / s to 3.03 m / s and the interval formed by 2.80 m / s to 3.20 m / s yields a candidate second downstream speed setting interval of 2.97 m / s to 3.03 m / s. Seven candidate second downstream speed setting values ​​of 2.97 m / s, 2.98 m / s, 2.99 m / s, 3.00 m / s, 3.01 m / s, 3.02 m / s, and 3.03 m / s are then sequentially formed according to a preset speed search step size of 0.01 m / s.

[0168] The preset control period includes six future velocity sampling times: 0.00 seconds, 0.02 seconds, 0.04 seconds, 0.06 seconds, 0.08 seconds, and 0.10 seconds. The intermediate tension feedback value at the current tension calculation time is 500.00 Newtons, the intermediate velocity feedback value is 3.00 m / s, the transition boundary advance velocity is 1.00 m / s, the corrected longitudinal stiffness parameter is 125000 Newtons, the corrected boundary position state variable is 12.60 m, the pre-switching longitudinal stiffness parameter is 100000 Newtons, the pre-switching free longitudinal strain parameter is 0.001, and the post-switching free longitudinal strain parameter is 0.002. Under each candidate condition, the corresponding candidate second downstream velocity setpoint is used as the downstream boundary velocity that remains constant within the preset control period. The intermediate tension feedback value and the intermediate velocity feedback value are used as the upstream tension and upstream boundary velocity that remain constant, respectively. The boundary position input values ​​for the six future velocity sampling times are determined sequentially based on the transition boundary advance velocity.

[0169] The predicted tension sequence corresponding to each candidate second downstream velocity setpoint is calculated according to steps S102 and S103 respectively. When 2.97 m / s, 2.98 m / s, 2.99 m / s, 3.00 m / s, 3.01 m / s, 3.02 m / s and 3.03 m / s are used as candidate second downstream velocity setpoints, the target predicted tension at the last velocity sampling time in the predicted tension sequence is approximately 349.55 N, 399.72 N, 449.87 N, 500.00 N, 550.11 N, 600.19 N and 650.25 N respectively.

[0170] Assuming a preset tension setting of 600.00 Newtons, the absolute values ​​of the differences between the predicted tensions of the aforementioned targets and the preset tension setting are approximately 250.45 Newtons, 200.28 Newtons, 150.13 Newtons, 100.00 Newtons, 49.89 Newtons, 0.19 Newtons, and 50.25 Newtons, respectively. Since the absolute value of the difference corresponding to 3.02 m / s is the smallest, 3.02 m / s is determined as the second downstream target speed setting and sent to the drive controller of the second downstream drive mechanism.

[0171] After completing the search for the candidate second downstream velocity setpoint, the controller does not retain the boundary position input values ​​at each future velocity sampling moment formed during the prediction process. Upon entering the next control cycle, the intermediate tension feedback value, intermediate velocity feedback value, transition boundary advance velocity, and second downstream velocity feedback value are reacquired, and the corrected boundary position state is updated based on the actually acquired transition boundary advance velocity. The above values ​​are only used to illustrate the process by which the controller determines the candidate second downstream velocity setpoint range, calculates the target predicted tension, and filters the second downstream target velocity setpoint; they do not represent actual equipment test results, nor are they intended to limit the scope of protection of this invention.

[0172] The controller sends the second downstream target speed setpoint to the drive controller of the second downstream drive mechanism. The drive controller executes the second downstream target speed setpoint according to its own speed closed loop and continues to be protected by the preset minimum drive speed, preset maximum drive speed, and allowed acceleration / deceleration.

[0173] The boundary position input values ​​at each future velocity sampling time calculated based on the transition boundary advance speed within this preset control cycle are only used for the predicted tension calculation corresponding to the candidate second downstream velocity setpoint. After the search for the candidate second downstream velocity setpoint ends, they are not retained as the model state and do not change the currently stored corrected boundary position state quantity.

[0174] Upon entering the next control cycle, the second tension feedback value, transition boundary advance speed, intermediate tension feedback value, intermediate speed feedback value, and second downstream speed feedback value are reacquired. Starting from the most recently stored corrected boundary position state quantity, the transition boundary advance speed, actually collected from the actual tension calculation time corresponding to the corrected boundary position state quantity to the current tension calculation time of the next control cycle, is integrated over time. The resulting actual boundary advance distance is added to the most recently stored corrected boundary position state quantity to obtain the corrected boundary position state quantity at the current tension calculation time of the next control cycle. The controller continues tension prediction and second downstream target speed setpoint determination using the corrected boundary position state quantity at the current tension calculation time and the latest corrected longitudinal stiffness parameter. Within the remaining control cycles of the same type switch, the longitudinal stiffness candidate parameter value screening and verification are not repeated; the controller directly uses the finally determined corrected longitudinal stiffness parameter in this type switch for tension prediction and second downstream target speed setpoint determination. When new type switch data is detected, the initialization and model state correction processes specified in steps S101 to S105 are re-executed.

[0175] It should be noted that the data used for longitudinal stiffness correction, boundary position correction, and determination of the second downstream target velocity setting value in this embodiment all have clear acquisition paths. The longitudinal stiffness parameters after switching and before switching are obtained through longitudinal tension calibration; the free longitudinal strain parameters after switching and before switching are obtained through natural length measurement under conditions of no external longitudinal tension; the transition boundary position state quantity is obtained by the time integration of the initial transition boundary position state quantity and the transition boundary advance speed; the intermediate tension feedback value and the second tension feedback value are collected by tension measuring devices at their respective set positions; the first upstream velocity feedback value, the intermediate velocity feedback value, and the second downstream velocity feedback value are obtained by the drive mechanism encoder or servo driver.

[0176] The transition boundary represents the material boundary between the flexible strip sections corresponding to the type after switching and the type before switching, which moves during continuous operation. The transition boundary position state quantity, candidate boundary position values, and the boundary position to be verified are all calculated according to the transition boundary advancement speed. The first and second open tension control sections are treated as having a one-dimensional longitudinal force relationship. The average free strain of the section is length-weighted based on the length occupied by the type before and after switching. The longitudinal flexibility of the section is calculated based on the length occupied by the type before and after switching, the longitudinal stiffness parameter before switching, and the longitudinal stiffness parameter after switching. When slippage occurs between the drive roller and the strip, or when there is a cumulative error in the speed measurement, the resulting material boundary position error is corrected through the boundary position state quantity correction process.

[0177] Example 2

[0178] Please see Figure 2 As shown, this embodiment provides a multi-segment tension model adaptive control system, including a parameter determination module, a longitudinal stiffness correction module, a boundary position correction module, and a drive control module. Specifically, in this embodiment, the parameter determination module, longitudinal stiffness correction module, boundary position correction module, and drive control module are deployed as software function blocks in the same industrial control computer. The industrial control computer reads data from the tension measuring device, speed measuring device, and drive controller via a fieldbus, and sends the second downstream target speed setpoint to the drive controller of the second downstream drive mechanism.

[0179] The preset tension prediction model is stored in the memory of the industrial control computer as a calculation program, without a separate model module. The longitudinal stiffness correction module, boundary position correction module, and drive control module call the same preset tension prediction model, so that the selection of longitudinal stiffness candidate parameter values, boundary position candidate values, and the search for the second downstream target velocity setpoint adopt the same method for determining the occupancy status of the section type, the calculation method for the average free strain of the section, the calculation method for the longitudinal flexibility of the section, and the recursive relationship for the predicted tension.

[0180] The parameter determination module acquires type switching data and transition boundary advance speed. The type switching data includes the type before switching, the type after switching, and the initial transition boundary position state quantity. The parameter determination module obtains the longitudinal stiffness parameter before switching, the longitudinal stiffness parameter after switching, the free longitudinal strain parameter before switching, and the free longitudinal strain parameter after switching based on the preset mapping relationship between the type and longitudinal characteristic parameters. It then determines the transition boundary position state quantity based on the time integral of the initial transition boundary position state quantity and the transition boundary advance speed. The parameter determination module initializes the corrected longitudinal stiffness parameter with the longitudinal stiffness parameter after switching, initializes the corrected boundary position state quantity with the transition boundary position state quantity, and advances the transition boundary position state quantity and the corrected boundary position state quantity not replaced by the new correction value according to the transition boundary advance speed.

[0181] The longitudinal stiffness correction module acquires the second tension feedback value, intermediate tension feedback value, intermediate velocity feedback value, and second downstream velocity feedback value when the state quantity at the transition boundary position is within the preset boundary correction range. For each type switch, the longitudinal stiffness correction module initiates only one complete correction process consisting of a preset longitudinal stiffness correction data window and a preset longitudinal stiffness verification window, and saves a status flag indicating whether the longitudinal stiffness correction for this type switch has been completed. The longitudinal stiffness correction module forms candidate longitudinal stiffness parameter values ​​based on the longitudinal stiffness parameters after the switch and the preset longitudinal stiffness step size. Within the preset longitudinal stiffness correction data window, it calculates the predicted tension sequence, the longitudinal stiffness prediction difference, and the candidate longitudinal stiffness residual. Based on the candidate longitudinal stiffness residual, it obtains the current longitudinal stiffness residual, the minimum longitudinal stiffness residual, and the reduction in longitudinal stiffness residual. After the longitudinal stiffness parameters to be verified are generated, the longitudinal stiffness correction module calculates the residual before longitudinal stiffness verification, the residual after longitudinal stiffness verification, and the reduction in the longitudinal stiffness verification residual within the preset longitudinal stiffness verification window. Based on the reduction in the longitudinal stiffness verification residual, the corrected longitudinal stiffness parameters are determined. After the longitudinal stiffness parameters are corrected, the longitudinal stiffness correction module sets the status flag to "completed". The longitudinal stiffness parameter correction will not be restarted again until the next type switching data is detected.

[0182] The boundary position correction module is activated when a reduction in longitudinal stiffness residual has been achieved, the reduction in longitudinal stiffness residual is not greater than a preset residual reduction threshold, and the corrected longitudinal stiffness parameter is equal to the switched longitudinal stiffness parameter. The boundary position correction module acquires the second tension feedback value, intermediate tension feedback value, intermediate velocity feedback value, second downstream velocity feedback value, and transition boundary advance velocity. Based on the transition boundary position state quantity and a preset position step size, it generates candidate values ​​for the current boundary position, negative boundary position, and positive boundary position. Within the preset boundary position correction data window, it advances each candidate boundary position value and calculates the candidate boundary position residual and the reduction in boundary position residual. After forming the boundary position to be verified, within the preset boundary position verification window, it calculates the residual before boundary position verification, the residual after boundary position verification, and the reduction in boundary position verification residual. Based on the reduction in boundary position verification residual, it determines the corrected boundary position state quantity.

[0183] The drive control module acquires the second downstream speed feedback value. Based on the second downstream speed feedback value, the preset minimum drive speed, the preset maximum drive speed, and the preset allowable speed change per control cycle, it determines the candidate second downstream speed setpoint range and forms candidate second downstream speed setpoints according to the preset speed search step size. At the beginning of each preset control cycle, the drive control module reads the intermediate tension feedback value, intermediate speed feedback value, transition boundary advance speed, and stored corrected boundary position state value at the current tension calculation moment. Within the preset control cycle, the intermediate tension feedback value, intermediate speed feedback value, and transition boundary advance speed are used as constant prediction inputs.

[0184] For each candidate second downstream speed setpoint, the drive control module uses the candidate second downstream speed setpoint as the predicted execution speed under that candidate condition, and inputs it into the preset tension prediction model as a constant downstream boundary speed throughout the entire preset control cycle. The corrected boundary position state quantity stored at the current tension calculation time is used as the boundary position input value at the start speed sampling time of the preset control cycle. At each subsequent future speed sampling time, the cumulative boundary advance distance is determined based on the measured value of the transition boundary advance speed at the current tension calculation time and the cumulative duration from the start time of the preset control cycle to the current future speed sampling time. The sum of the corrected boundary position state quantity and the cumulative boundary advance distance is used as the boundary position input value input into the preset tension prediction model at the current future speed sampling time.

[0185] The boundary position input values ​​generated at each future velocity sampling moment within the preset control period are only used for calculating the predicted tension corresponding to the current candidate second downstream velocity setpoint, and are not written into the corrected boundary position state quantity stored in the parameter determination module. The drive control module calculates the predicted tension sequence of the second open tension control section based on the corrected longitudinal stiffness parameters and the boundary position input values ​​of the preset tension prediction model at each future velocity sampling moment. The predicted tension at the last velocity sampling moment is determined as the target predicted tension, and the second downstream target velocity setpoint is determined based on the absolute value of the difference between the target predicted tension and the preset tension setpoint.

[0186] The drive control module sends the second downstream target speed setpoint to the drive controller of the second downstream drive mechanism. The second downstream drive mechanism executes the second downstream target speed setpoint according to its own speed closed loop, while retaining the original minimum drive speed, maximum drive speed, and allowable acceleration / deceleration protection of the equipment. After the candidate speed search of the current preset control cycle is completed, the drive control module does not retain the boundary position input value of the future speed sampling time formed during the prediction process; when entering the next control cycle, the parameter determination module obtains the current corrected boundary position state value based on the most recently stored corrected boundary position state value and the subsequent actual acquired transition boundary advance speed, and then provides the current corrected boundary position state value to the drive control module. The actual drive linear velocity of the second downstream drive mechanism is again used as the second downstream speed feedback value input in the next control cycle, enabling the drive control module to form continuous closed-loop control.

[0187] The parameter determination module, longitudinal stiffness correction module, boundary position correction module, and drive control module all use the same device clock or a time-synchronized communication clock. The second tension feedback value, intermediate tension feedback value, transition boundary advance speed, intermediate speed feedback value, and second downstream speed feedback value all carry sampling timestamps. The initial pre-set tension prediction model recursion, candidate residual calculation, and verification residual calculation only combine data from the same speed sampling time to avoid false residuals caused by time mismatches between tension and speed data.

[0188] In Examples 1 and 2, the preset objects are calibrated or set before the equipment executes tension model adaptive control and stored in the controller or process database. The mapping relationship between the preset type and the longitudinal characteristic parameters is established by the longitudinal tensile calibration of the flexible strip corresponding to each type and the measurement of the natural length without external longitudinal tension; the preset boundary safety distance and preset position step size are determined based on the position measurement error, the integral error of the transition boundary advance speed, and the boundary advance distance within the correction window; the preset longitudinal stiffness step size is determined based on the longitudinal stiffness calibration dispersion and the longitudinal stiffness correction resolution; the preset longitudinal stiffness identifiable threshold and preset residual reduction threshold are determined based on the tension fluctuation under stable operating conditions, the natural change of the model residual, and the allowable error of tension measurement; the preset minimum drive speed, preset maximum drive speed, preset allowable speed change per control cycle, preset control cycle, and preset speed search step size are determined based on the capability of the second downstream drive mechanism, the control update cycle, and the speed command resolution; the preset tension setpoint is determined based on the production process requirements of the switched type.

[0189] Specifically, the second tension feedback value is obtained through a tension measuring roller located within the second open tension control section; the intermediate tension feedback value is obtained through a force measuring roller located downstream of the intermediate boundary; and the first inlet tension feedback value is obtained through a tension measuring device located downstream of the first upstream boundary. The first upstream speed feedback value, the intermediate speed feedback value, and the second downstream speed feedback value are obtained through an encoder or servo driver of the drive mechanism. The transition boundary advance speed is obtained through a strip speed measuring device. The longitudinal stiffness parameters before and after switching are obtained through longitudinal tensile calibration. The free longitudinal strain parameters before and after switching are obtained through natural length measurement under conditions of no applied longitudinal tension. The transition boundary position state quantity is calculated by time integration of the initial transition boundary position state quantity and the transition boundary advance speed. The second tension feedback value, the intermediate tension feedback value, the first inlet tension feedback value, the first upstream speed feedback value, the intermediate speed feedback value, the second downstream speed feedback value, the transition boundary advance speed, the longitudinal stiffness parameters before and after switching, the free longitudinal strain parameters before and after switching, and the transition boundary position state quantity all have clearly defined data acquisition paths.

[0190] The embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to the specific embodiments described above. The specific embodiments described above are used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Any changes, substitutions or modifications made by those skilled in the art to the above embodiments without departing from the concept and scope of protection of the present invention shall fall within the scope of protection of the present invention.

Claims

1. A multi-segment tension model adaptive control method, characterized in that, include: Acquire type switching data and transition boundary advance speed, determine longitudinal stiffness parameters and transition boundary position state quantities after switching based on the type switching data and transition boundary advance speed, initialize corrected longitudinal stiffness parameters with the longitudinal stiffness parameters after switching, and initialize corrected boundary position state quantities with the transition boundary position state quantities. When the transition boundary position state quantity is within the preset boundary correction range, a second tension feedback value is obtained, and the longitudinal stiffness parameter after switching is corrected according to the second tension feedback value and the preset tension prediction model, and the corrected longitudinal stiffness parameter and the reduction amount of longitudinal stiffness residual are determined. When the reduction in longitudinal stiffness residual is not greater than a preset residual reduction threshold and the corrected longitudinal stiffness parameter is equal to the switched longitudinal stiffness parameter, the corrected boundary position state quantity is corrected according to the second tension feedback value and the preset tension prediction model. Obtain a second downstream velocity feedback value, determine a candidate second downstream velocity setting value based on the second downstream velocity feedback value, calculate the target predicted tension for each candidate second downstream velocity setting value based on the corrected longitudinal stiffness parameter and the corrected boundary position state quantity, determine the second downstream target velocity setting value based on the target predicted tension and the preset tension setting value, and control the second downstream drive mechanism.

2. The method according to claim 1, characterized in that, The method further includes: The first open tension control section and the second open tension control section are set sequentially along the running direction, and they share the same intermediate boundary; Obtain the inlet tension of the first open tension control section and determine it as the first inlet tension feedback value; obtain the actual tension of the second open tension control section and determine it as the second tension feedback value; obtain the intermediate tension feedback value. The system acquires a first upstream speed feedback value, an intermediate speed feedback value, and a second downstream speed feedback value; acquires the first upstream boundary position, the intermediate boundary position, and the downstream boundary position of the second open tension control section; the second downstream speed feedback value is the actual driving linear speed of the second downstream drive mechanism; and the preset tension setting value is the tension setting value of the second open tension control section.

3. The method according to claim 2, characterized in that, The determination of the longitudinal stiffness parameter after switching, the transition boundary position state quantity, and the preset boundary correction interval includes: Obtain the type before switching, the type after switching, and the initial transition boundary position state from the type switching data; The longitudinal stiffness parameter before switching, the longitudinal stiffness parameter after switching, the free longitudinal strain parameter before switching, and the free longitudinal strain parameter after switching are determined according to the mapping relationship between the preset type and the longitudinal characteristic parameter. The transition boundary position state quantity is initialized with the initial transition boundary position state quantity, and the transition boundary position state quantity and the corrected boundary position state quantity are advanced according to the transition boundary advancement speed. The interval after the two ends of the second open tension control section are respectively offset inward by a preset boundary safety distance is determined as the preset boundary correction interval.

4. The method according to claim 3, characterized in that, The determination of the section type occupancy status of the preset tension prediction model includes: Obtain the predicted open tension control section length and upstream boundary position, wherein the upstream boundary position is either the first upstream boundary position or the intermediate boundary position. Limit the position difference obtained by subtracting the upstream boundary position from the transition boundary position state quantity or the corrected boundary position state quantity input to the preset tension prediction model to between zero and the predicted open tension control section length, thereby obtaining the type-occupied length after switching. When the length occupied by the switched type is greater than zero, the switched type is determined as the entry type; when the length occupied by the switched type is zero, the original type is determined as the entry type. When the length occupied by the type after switching is less than the length of the predicted open tension control section, the type before switching is determined as the exit type; when it is equal to the length of the predicted open tension control section, the type after switching is determined as the exit type.

5. The method according to claim 4, characterized in that, The calculation of the section parameters of the preset tension prediction model includes: Subtract the type-occupied length after switching from the predicted open tension control section length to obtain the type-occupied length before switching. The average free strain of the section is obtained by dividing the sum of the product of the pre-switching type occupancy length and the pre-switching free longitudinal strain parameter and the product of the post-switching type occupancy length and the post-switching free longitudinal strain parameter by the predicted open tension control section length. The longitudinal flexibility of the section is obtained by adding the ratio of the length occupied by the type after the switch to the longitudinal stiffness parameter after the switch and the ratio of the length occupied by the type before the switch to the longitudinal stiffness parameter before the switch.

6. The method according to claim 5, characterized in that, The determination of the initial parameters of the preset tension prediction model includes: The switched longitudinal stiffness parameter used in the preset tension prediction model is used as the input longitudinal stiffness parameter; during normal prediction, the current corrected longitudinal stiffness parameter is taken; during candidate screening, the value of each longitudinal stiffness candidate parameter is taken; during verification, the switched longitudinal stiffness parameter or the longitudinal stiffness parameter to be verified is taken. When predicting the first open tension control section, the first inlet tension feedback value, the first upstream velocity feedback value, and the intermediate velocity feedback value are respectively determined as upstream tension, upstream boundary velocity, and downstream boundary velocity; when predicting the second open tension control section, the intermediate tension feedback value, the intermediate velocity feedback value, and the second downstream velocity feedback value are respectively determined as upstream tension, upstream boundary velocity, and downstream boundary velocity. The predicted tension is initialized with the upstream tension of the predicted open tension control section, and the average total strain of the section and the inlet total strain are determined based on the predicted tension, the longitudinal flexibility of the section, the average free strain of the section, and the length of the predicted open tension control section.

7. The method according to claim 6, characterized in that, The recursive calculation of the predicted tension includes: The total strain at the outlet is calculated based on the predicted tension at the previous velocity sampling time, the outlet longitudinal stiffness parameter corresponding to the outlet type, and the outlet free longitudinal strain parameter. The outlet longitudinal stiffness parameter and the outlet free longitudinal strain parameter are taken as the switched longitudinal stiffness parameter and the switched free longitudinal strain parameter when the outlet type is the switched type, and as the unswitched longitudinal stiffness parameter and the unswitched free longitudinal strain parameter when the outlet type is the unswitched type. The average total strain increment of the section is calculated based on the upstream boundary velocity, the downstream boundary velocity, the total inlet strain, the total outlet strain, and the preset velocity sampling period. The average total strain increment of the section is accumulated to the average total strain of the section, and the predicted tension is calculated based on the average total strain of the section, the average free strain of the section, and the longitudinal flexibility of the section.

8. The method according to claim 7, characterized in that, The correction of the longitudinal stiffness parameter after switching includes: The longitudinal stiffness parameter after the switch and the longitudinal stiffness values ​​that are greater than zero obtained by adding and subtracting the preset longitudinal stiffness step size respectively are determined as candidate longitudinal stiffness parameter values. The intermediate tension feedback value, intermediate velocity feedback value, and second downstream velocity feedback value are respectively determined as the upstream tension, upstream boundary velocity, and downstream boundary velocity of the second open tension control section, and the transition boundary position state quantity is used as the boundary position input value of the preset tension prediction model. Within the preset longitudinal stiffness correction data window, a predicted tension sequence is calculated based on each of the longitudinal stiffness candidate parameter values. The predicted tension sequence is the predicted tension sequence of the second open tension control section. The longitudinal stiffness prediction difference is determined based on the difference between each of the predicted tension sequences. When the longitudinal stiffness prediction difference is greater than the preset longitudinal stiffness identifiable threshold, longitudinal stiffness candidate residuals are calculated based on each of the predicted tension sequences and the second tension feedback value. When the difference in longitudinal stiffness prediction is not greater than the preset longitudinal stiffness identifiable threshold, the corrected longitudinal stiffness parameter is kept as the switched longitudinal stiffness parameter.

9. The method according to claim 8, characterized in that, The selection of candidate longitudinal stiffness parameter values ​​includes: The longitudinal stiffness candidate residual obtained from the switched longitudinal stiffness parameters is determined as the current longitudinal stiffness residual, and the minimum longitudinal stiffness candidate residual is determined as the minimum longitudinal stiffness residual. The reduction in longitudinal stiffness residual is obtained by subtracting the minimum longitudinal stiffness residual from the current longitudinal stiffness residual. When the reduction in longitudinal stiffness residual is greater than the preset residual reduction threshold and the minimum longitudinal stiffness candidate residual is obtained from a unique longitudinal stiffness candidate parameter value, the longitudinal stiffness candidate parameter value is determined as the longitudinal stiffness parameter to be verified.

10. The method according to claim 9, characterized in that, The verification of the longitudinal stiffness parameter to be verified includes: Within the preset longitudinal stiffness verification window, the predicted tension sequence is calculated using the switched longitudinal stiffness parameters and the longitudinal stiffness parameters to be verified, respectively. The reduction in longitudinal stiffness verification residual is determined based on the difference between the residuals of the two predicted tension sequences and the second tension feedback value; When the reduction in the longitudinal stiffness verification residual is greater than the preset residual reduction threshold, the longitudinal stiffness parameter to be verified is determined as the corrected longitudinal stiffness parameter. When the reduction in the longitudinal stiffness verification residual is not greater than the preset residual reduction threshold, the corrected longitudinal stiffness parameter is kept the same as the switched longitudinal stiffness parameter.

11. The method according to claim 7, characterized in that, The correction of the corrected boundary position state quantity includes: The candidate values ​​of the boundary position are determined by using the state quantity of the transition boundary position at the beginning of the data window at the preset boundary position correction and the positions obtained by adding and subtracting the preset position step size respectively. The intermediate tension feedback value, intermediate velocity feedback value, and second downstream velocity feedback value are respectively determined as the upstream tension, upstream boundary velocity, and downstream boundary velocity of the second open tension control section; Each of the candidate boundary positions is used as the boundary position input value of the preset tension prediction model. The predicted tension sequence is calculated, and the candidate boundary position residual is calculated based on the predicted tension sequence and the second tension feedback value. Each of the candidate boundary positions is advanced according to the transition boundary advancement speed.

12. The method according to claim 11, characterized in that, The filtering of candidate values ​​for the boundary positions includes: The candidate boundary position residuals obtained from the transition boundary position state quantities are determined as the current candidate boundary position residuals, and the minimum candidate boundary position residuals are determined as the minimum candidate boundary position residuals. Subtract the minimum boundary position candidate residual from the current boundary position candidate residual to obtain the reduction in boundary position residual; When the reduction in the boundary position residual is greater than the preset residual reduction threshold and the minimum boundary position candidate residual is obtained from the unique boundary position candidate value, the boundary position candidate value is determined as the boundary position to be verified.

13. The method according to claim 12, characterized in that, The verification of the boundary location to be verified includes: Within the preset boundary position verification window, the transition boundary position state quantity and the boundary position to be verified are advanced according to the transition boundary advancement speed, and the predicted tension sequence is calculated respectively. The boundary position is determined based on the difference between the residuals of the two predicted tension sequences and the second tension feedback value to verify the amount of residual reduction; When the reduction in the boundary position verification residual is greater than the preset residual reduction threshold, the boundary position to be verified is determined as the corrected boundary position state quantity. When the reduction in the residual of the boundary position verification is not greater than the preset residual reduction threshold, the boundary position to be verified is not used, and the corrected boundary position state quantity continues to advance according to the transition boundary advancement speed.

14. The method according to claim 7, characterized in that, The determination of the candidate second downstream speed setpoint and the second downstream target speed setpoint includes: The second downstream speed feedback value is located between the preset minimum drive speed and the preset maximum drive speed; The candidate second downstream speed setting value range is determined based on the second downstream speed feedback value, the preset allowable speed change in a single control cycle, the preset minimum drive speed, and the preset maximum drive speed. The candidate second downstream speed setting value is determined according to the preset speed search step size. The intermediate tension feedback value, intermediate velocity feedback value, and each of the candidate second downstream velocity setting values ​​are respectively determined as the upstream tension, upstream boundary velocity, and downstream boundary velocity of the second open tension control section. The predicted tension sequence of the second open tension control section is calculated based on the corrected longitudinal stiffness parameter and the corrected boundary position state quantity. The predicted tension at the last velocity sampling moment in the predicted tension sequence is determined as the target predicted tension. The candidate second downstream speed setting value with the smallest absolute value of the difference from the preset tension setting value is determined as the second downstream target speed setting value.

15. A multi-segment tension model adaptive control system, characterized in that, The method for implementing a multi-segment tension model adaptive control method as described in any one of claims 1 to 14 includes a parameter determination module, a longitudinal stiffness correction module, a boundary position correction module, and a drive control module. The parameter determination module is used to acquire type switching data and transition boundary advance speed, determine the longitudinal stiffness parameter and transition boundary position state quantity after switching, and initialize the corrected longitudinal stiffness parameter and corrected boundary position state quantity. The longitudinal stiffness correction module is used to obtain a second tension feedback value and determine the corrected longitudinal stiffness parameter and the reduction in longitudinal stiffness residual based on the second tension feedback value and a preset tension prediction model. The boundary position correction module is used to obtain a second tension feedback value and correct the corrected boundary position state quantity when the reduction in the longitudinal stiffness residual is not greater than a preset residual reduction threshold and the corrected longitudinal stiffness parameter is equal to the switched longitudinal stiffness parameter. The drive control module is used to acquire the second downstream speed feedback value, determine the candidate second downstream speed setpoint and the target predicted tension, determine the second downstream target speed setpoint based on the target predicted tension and the preset tension setpoint, and control the second downstream drive mechanism.

Images