Paperboard production process parameter self-adaptive control method based on intelligent cruise system

Abstract

The application provides a paperboard production process parameter self-adaptive control method based on an intelligent cruise system, comprising: collecting order specifications, layered base paper attributes, equipment topological relations, real-time operation parameters and paperboard quality detection results through an industrial control system, calculating a quality stability margin, forming paperboard production cruise state data; selecting a perturbation bearing parameter according to the paperboard production cruise state data, generating a zero-mean process perturbation identification sequence and superimposing it to the corresponding cruise control reference value; identifying cross-machine response according to the perturbation identification execution record, dividing virtual paperboard batch sections, correcting each station transmission position, and forming a cross-machine process history; inputting the cross-machine process history into a shadow production line, respectively deducing the quality risk and unit qualified area energy consumption corresponding to different compensation paths, and forming counterfactual compensation contribution; identifying the compensation state according to the counterfactual compensation contribution, and updating the cruise control parameter.

Description

Technical Field

[0001] This invention relates to the field of paperboard production control technology, and in particular to an adaptive control method for paperboard production process parameters based on an intelligent cruise system. Background Technology

[0002] A corrugated cardboard production line typically includes a paper stack, preheating unit, gluing machine, double-facer, and slitting / cross-cutting stations. During production, cardboard quality is affected by parameters such as paper basis weight, initial moisture content, cardboard running speed, glue viscosity, hot plate temperature, hot plate pressure, and traction tension. Current cardboard production systems collect operating parameters from each station through a control system, and, combined with order specifications, operational experience, and online monitoring results, adjust the machine speed, steam pressure, preheating wrap angle, hot plate temperature, and traction tension.

[0003] This type of control can achieve a certain degree of automation, but it usually still relies on the current station status or existing quality anomalies as the basis for adjustment. Since corrugated cardboard is a continuously operating material, as it passes through the single-facer, bridge, gluing, double-facer, and slitting stations, it is affected by changes in bridge paper storage length, station speed differences, thermal expansion and contraction, and local slippage. When a quality anomaly is detected in a certain section of cardboard at the backend, relying solely on the data collection time and fixed station spacing for backtracking makes it difficult to accurately determine the actual process parameters corresponding to that section at each station, easily leading to incorrect parameter adjustment targets.

[0004] Therefore, this invention proposes an adaptive control method for paperboard production process parameters based on an intelligent cruise system. The information disclosed in the background section is only for enhancing understanding of the background of this disclosure and may therefore contain prior art information that is not common knowledge to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by providing an adaptive control method for paperboard production process parameters based on an intelligent cruise system, thereby solving the technical problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The adaptive control method for paperboard production process parameters based on an intelligent cruise system includes the following steps:

[0008] S1. Collect order specifications, layered paper properties, equipment topology, real-time operating parameters and paperboard quality inspection results through the industrial control system, calculate the quality stability margin, and form paperboard production cruise status data;

[0009] S2. Select the perturbation load parameters based on the paperboard production cruise status data, determine the safety injection boundary, generate a zero-mean process perturbation identifier sequence, and superimpose it onto the corresponding cruise control reference value to form a perturbation identifier execution record.

[0010] S3. Based on the perturbation identifier execution record, identify cross-machine response, divide virtual cardboard batches, correct the transmission position of each station, bind the corresponding station parameters, cardboard quality inspection results and cruise parameter control records, and form a cross-machine process history.

[0011] S4. Input the cross-machine process history into the shadow production line, and respectively deduce the current compensation path, the non-execution compensation path, the delayed execution compensation path, and the alternative compensation path. Compare the quality risk and energy consumption per unit qualified area of ​​each path to form the counterfactual compensation contribution.

[0012] S5. Identify the compensation status based on the counterfactual compensation contribution, update the cruise control parameters and process disturbance identification sequence, and use the updated cruise control parameters as the cruise control reference value for the next control cycle.

[0013] S1 specifically includes: reading order specifications, layered paper properties, equipment topology, process allowable range, and quality allowable boundary to form basic production line configuration data; collecting real-time operating parameters, paperboard quality inspection results, energy metering data, and production line status identifiers according to a preset control cycle, completing time alignment and anomaly verification, including range verification, timestamp verification, and rate of change verification, and marking abnormal data in a single cycle with supplementary values ​​to form cycle operation data; and calculating the normalized residual ratio of each parameter from the corresponding allowable boundary based on the basic production line configuration data and cycle operation data, selecting the minimum value as the quality stability margin, and combining them to form paperboard production cruise status data.

[0014] S2 specifically includes: selecting perturbation bearing parameters from the traction speed reference value, bridge output speed set value, and preheating wrap angle set value based on the paperboard production cruise status data; determining the actual perturbation amplitude by combining the quality stability margin, process allowable range, and actuator rated amplitude upper limit, thus forming perturbation injection constraint data; generating a process perturbation identification sequence composed of multiple perturbation units and maintaining zero mean based on the perturbation injection constraint data; determining the directional arrangement, unit duration, and injection time, thus forming a perturbation identification injection plan; superimposing the process perturbation identification sequence onto the corresponding cruise control reference value to generate cruise control commands; recording the actuator feedback value and execution deviation, thus forming a perturbation identification execution record.

[0015] S3 specifically includes: determining the response candidate window based on the perturbation identifier execution record, extracting response data from the periodic operation data, calculating the response matching degree, and forming a cross-machine response correspondence when at least two response data pass the valid response judgment; dividing virtual paperboard batches based on the cross-machine response correspondence, the cumulative pulse count of the identifier reference encoder, the equipment topology relationship, and the cross-bridge paper storage length, correcting the transmission position of each station, and forming a virtual paperboard batch transmission trajectory; and extracting and binding the time sequence data of each station parameter, the representative value of the station parameter, the paperboard quality inspection result, and the cruise parameter control record based on the virtual paperboard batch transmission trajectory to form a cross-machine process history.

[0016] S4 specifically includes: extracting virtual cardboard batch transfer trajectories, representative values ​​of workstation parameters, and cruise parameter control records based on cross-machine process history; fitting and correcting process state transmission relationships according to workstation sections to form the initial state of the shadow production line; generating current compensation paths, non-executed compensation paths, delayed execution compensation paths, and alternative compensation paths when the shadow production line is effective; deriving the quality risks and energy consumption per unit qualified area for each path; screening target alternative compensation paths to form compensation path deduction results; using the current compensation path as a comparison benchmark, calculating the quality improvement contribution, timing improvement contribution, and energy consumption improvement contribution; and recording the number of delay cycles for the target delayed execution compensation path to form the counterfactual compensation contribution.

[0017] S5 specifically includes: based on the counterfactual compensation contribution, identifying the compensation state in the order of quality correction state, lag compensation state, inefficient compensation state, excessive compensation state, necessary compensation state, steady-state maintenance state, and unidentified state, forming a compensation state identification result; updating the corresponding target parameters based on the compensation state identification result, and after verification by the shadow production line, sending the effective updated cruise control parameters to the corresponding actuator controller, and recording the applicable batch segment identifier and parameter effective time; calculating the predicted quality stability margin based on the updated cruise control parameters, adjusting the process perturbation identifier sequence when the injection conditions are met, and using the updated cruise control parameters as the cruise control reference value for the next control cycle.

[0018] The beneficial effects of this invention are as follows:

[0019] This invention establishes virtual paperboard batches for continuously running paperboard by superimposing process perturbation identification sequences onto cruise control reference values, and combines encoder pulses, bridge paper storage lengths, and cross-machine responses to correct the batch transmission position, thereby improving the accuracy of the correspondence between paperboard quality inspection results and front-end process parameters.

[0020] This invention binds the time-series data of workstation parameters, representative values ​​of workstation parameters, paperboard quality inspection results, and cruise parameter control records corresponding to virtual paperboard batches to form a cross-machine process history. This reduces ineffective adjustments caused by mismatches between abnormal paperboard sections and process parameters. By using a shadow production line to deduce the current compensation path, the non-execution compensation path, the delayed execution compensation path, and the alternative compensation path, the impact of different parameter correction methods on paperboard quality and energy consumption per unit qualified area can be determined without interfering with the operation of the physical paperboard production line.

[0021] This invention identifies quality correction states, lag compensation states, inefficient compensation states, overcompensation states, necessary compensation states, and steady-state maintenance states. It updates corresponding target parameters for different production states, avoiding indiscriminate adjustments to multiple process parameters that could lead to new parameter coupling. By updating the traction speed reference value, bridge output speed setpoint, preheating wrap angle setpoint, and other cruise control parameters within the constraints of quality allowable boundaries and safety injection boundaries, it can reduce unnecessary steam consumption and electricity usage while ensuring paperboard quality, thereby improving the stable operating efficiency of the paperboard production line. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the adaptive control method for paperboard production process parameters according to the present invention;

[0023] Figure 2 This is a schematic diagram illustrating the generation of process perturbation identifier sequences and virtual paperboard batch segmentation in this invention.

[0024] Figure 3 This is a schematic diagram illustrating the cross-machine process history construction and shadow production line counterfactual inference of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Example: Figures 1 to 3 As shown, this embodiment provides an adaptive control method for paperboard production process parameters based on an intelligent cruise system, including the following steps:

[0027] S1. Collect order specifications, layered paper properties, equipment topology, real-time operating parameters and paperboard quality inspection results through the industrial control system, calculate the quality stability margin, and form paperboard production cruise status data;

[0028] S2. Select the perturbation load parameters based on the paperboard production cruise status data, determine the safety injection boundary, generate a zero-mean process perturbation identifier sequence, and superimpose it onto the corresponding cruise control reference value to form a perturbation identifier execution record.

[0029] S3. Based on the perturbation identifier execution record, identify cross-machine response, divide virtual cardboard batches, correct the transmission position of each station, bind the corresponding station parameters, cardboard quality inspection results and cruise parameter control records, and form a cross-machine process history.

[0030] S4. Input the cross-machine process history into the shadow production line, and respectively deduce the current compensation path, the non-execution compensation path, the delayed execution compensation path, and the alternative compensation path. Compare the quality risk and energy consumption per unit qualified area of ​​each path to form the counterfactual compensation contribution.

[0031] S5. Identify the compensation status based on the counterfactual compensation contribution, update the cruise control parameters and process disturbance identification sequence, and use the updated cruise control parameters as the cruise control reference value for the next control cycle.

[0032] S1 specifically includes the following sub-steps:

[0033] S110, The industrial control system reads the order specifications, layered paper properties, equipment topology, process allowable range, and quality allowable boundary corresponding to the current production order. The order specifications are read by the production management system, including paperboard width, number of paperboard layers, flute type, target production length, target cross-cutting length, and target paperboard running speed.

[0034] The properties of the layered base paper are read from the base paper roll identification information, the base paper rack station number, and the online moisture content detection device set at the rear of the corresponding base paper rack. These properties include the basis weight of each layer of base paper, the initial moisture content of each layer of base paper, and the roll diameter of each layer of base paper. For 3-layer corrugated board, the properties of the base paper corresponding to the face paper, core paper, and liner paper are recorded separately. For 5-layer corrugated board, the properties of the base paper corresponding to the face paper, the first core paper, the middle divider, the second core paper, and the liner paper are recorded separately. The average value of the properties of each layer of base paper is not used to replace the layered recording results.

[0035] The equipment topology is entered into the industrial control system during the equipment commissioning phase. This includes the arrangement order of single-sided machines, bridges, gluing machines, double-sided machines, and longitudinal and transverse cutting stations, the reference transmission distance between adjacent stations, and the reference bridge paper storage length when the bridge paper storage length is at the reference value.

[0036] The permissible ranges for the process include the permissible range of the traction speed reference value, the permissible range of the bridge output speed setting value, the permissible range of the preheat wrap angle setting value, the permissible range of the steam pressure setting value, the permissible range of the roller gap setting value, the permissible range of the temperature setting value of each hot plate zone of the double-sided machine, the permissible range of the hot plate pressure setting value, and the permissible range of the traction tension setting value.

[0037] The quality allowable boundaries include the allowable range of moisture content of the paperboard exiting the double-facer, the allowable range of surface temperature of the paperboard exiting the double-facer, the upper limit of transverse warping of the paperboard, the upper limit of longitudinal warping of the paperboard, the lower limit of glue line continuity, the allowable state of exposed corrugations, and the upper limit of cross-cutting length deviation.

[0038] The industrial control system prioritizes reading historical qualified production data from the enterprise's process database that has the same number of paperboard layers and flute type as the current order, and whose basis weight deviation of each layer of paperboard is within the preset matching range. It then takes the intersection of the historical valid value range of each cruise control parameter with the rated operating range of the equipment to obtain the process allowable range.

[0039] Historical qualified production data refers to historical production records where the corresponding paperboard quality inspection results all meet the quality allowable boundaries. When there is insufficient matching historical qualified production data, the industrial control system reads the standard process range corresponding to the same number of paperboard layers and flute type. The standard process range is a generally known control range in the corrugated paperboard industry (e.g., the general range of preheating wrap angle is 60° to 120°), and selects the standard process range with the smallest difference in basis weight of each layer of base paper as the initial process allowable range.

[0040] The basic configuration data for the production line is formed by combining order specifications, layered paper properties, equipment topology, process allowable range, and quality allowable boundary.

[0041] S120, the industrial control system operates according to a preset control cycle. The system synchronously collects real-time operating parameters, cardboard quality inspection results, energy metering data, and production line status indicators. The preset control cycle is set according to the refresh frequency of the industrial control system; in this embodiment, it is set to 1 second.

[0042] Real-time operating parameters include the running speed of the single-facer, the running speed of the gluing machine, the running speed of the double-facer inlet, the paper storage length of the bridge, the detected value of the bridge output speed, the feedback value of the preheating wrap angle, the detected value of the steam pressure, the feedback value of the glue roller gap, the viscosity of the paste, the paper surface temperature at the double-facer inlet, the detected values ​​of the temperature of each hot plate zone of the double-facer, the detected values ​​of the hot plate pressure, the detected values ​​of the traction tension, the longitudinal cutting speed, the transverse cutting speed, and the number of encoder pulses at each station.

[0043] The industrial control system also reads the current cruise control reference values ​​from the control command register. These reference values ​​include the traction speed reference value, bridge output speed setting value, preheating wrap angle setting value, steam pressure setting value, glue roller gap setting value, temperature settings for each hot plate zone of the double-facer, hot plate pressure setting value, and traction tension setting value. The paperboard quality inspection results include the moisture content of the paperboard exiting the double-facer, the surface temperature of the paperboard exiting the double-facer, the transverse warpage of the paperboard, the longitudinal warpage of the paperboard, glue line continuity, exposed corrugation identification results, and cross-cut length deviation.

[0044] The transverse warpage and longitudinal warpage of the cardboard are the maximum height difference between the cardboard surface and the reference plane in the width direction and the running direction, respectively, in mm; the glue line continuity is the proportion of the effective glue line length identified by the line scanning vision inspection device set between the gluing machine and the double-facer to the total glue line length to be detected; the exposed corrugation identification result is binary data, recorded as 1 when exposed corrugation peaks are identified, otherwise recorded as 0; the cross-cut length deviation is the absolute value of the difference between the actual cross-cut length and the target cross-cut length, in mm.

[0045] Energy metering data includes the electricity consumption output from the electricity meters of each machine and the steam consumption output from the steam flow meter. Production line status indicators include start-up, shutdown, order change, paper splicing, fault handling, and manual emergency intervention status; when any production line status indicator is triggered, the current cardboard production line is recorded as an unstable production state.

[0046] Industrial control systems use the end time of each control cycle as the alignment time and select the effective sampled value with the smallest time difference from this alignment time.

[0047]

[0048]

[0049] In the formula, This represents the alignment value of the i-th parameter within the k-th control cycle; This indicates the alignment time of the k-th control cycle; This represents the set of all valid sampling times of the i-th parameter within the current control cycle; This represents the effective sampling time with the smallest time difference from the alignment time.

[0050] The industrial control system sequentially performs range verification, timestamp verification, and rate of change verification. When a parameter exceeds the corresponding allowable range, the acquisition timestamp is missing, the time difference between the sampling time and the alignment time is greater than twice the corresponding sensor sampling period, or the equipment is in operation but the parameter maintains the same value for 10 consecutive control cycles, the corresponding data is marked as abnormal data.

[0051] For abnormal data within a single control cycle, the parameter value from the previous valid control cycle is used as the replacement, and a supplementary value marker is added. For parameters that have no valid sampled values ​​for three consecutive control cycles, no supplementary value is performed, the injection of new process perturbation identifier sequences is paused, and the cruise control baseline value of the previous control cycle is maintained. The verified real-time operating parameters, cardboard quality inspection results, energy metering data, production line status identifiers, and supplementary value markers are combined to form the cycle operating data.

[0052] S130: The industrial control system calculates the normalized residual ratio of each real-time operating parameter and the paperboard quality inspection result from the corresponding allowable boundary based on the production line basic configuration data and cycle operation data.

[0053] For interval parameters that simultaneously have both a lower and upper allowable limit, the following formula applies:

[0054]

[0055] In the formula, This represents the normalized residual ratio of the j-th interval parameter within the k-th control cycle; This indicates the actual value of the corresponding parameter; and These represent the lower and upper limits of the allowable limits, respectively. This indicates amplitude limiting; the result is set to 0 if it is less than 0 and 1 if it is greater than 1.

[0056] For indicators that only have an upper limit, the difference between the upper limit and the actual value is divided by the upper limit and a limiting process is performed; for adhesive line continuity that only has a lower limit, the difference between the actual value and the lower limit is divided by the lower limit and a limiting process is performed; when the exposed fiber identification result is 0, the corresponding normalized residual ratio is recorded as 1, and when the exposed fiber identification result is 1, it is recorded as 0.

[0057] The quality stability margin is calculated using the following formula:

[0058]

[0059] In the formula, This represents the quality stability margin corresponding to the k-th control cycle; to This represents the normalized residual proportion of each item involved in the calculation; n represents the total number of parameters involved in the calculation.

[0060] When the quality stability margin is 0, the injection of new process perturbation identifier sequences is suspended. The production line basic configuration data, cycle operation data, and quality stability margin are combined to form paperboard production cruise status data, which is used by S210 to determine perturbation load parameters and safe injection boundaries.

[0061] S2 specifically includes the following sub-steps:

[0062] S210: The industrial control system reads the cardboard production cruise status data generated in S130 and selects one parameter from the traction speed reference value, the bridge output speed setpoint, and the preheating wrap angle setpoint as the perturbation carrying parameter. The perturbation carrying parameter refers to a single cruise control parameter used to superimpose the process perturbation identifier sequence within a perturbation identifier injection window; the perturbation identifier injection window refers to the continuous time interval from the start of execution of the first perturbation unit to the end of execution of the last perturbation unit.

[0063] The industrial control system adopts a single-parameter sequential injection method. It determines the first parameter that meets the injection conditions in the order of traction speed reference value, bridge output speed set value and preheating wrap angle set value, so that the subsequent S310 can attribute the cross-machine response to the determined perturbation bearing parameter.

[0064] The injection conditions include: the candidate parameter is within the corresponding allowable range determined in S110; the candidate parameter has not been marked as abnormal data or supplementary data in the last 10 control cycles, and there is no continuous unidirectional change; the quality stability margin formed in S130 is not less than 0.15; the paperboard production line is not in the unstable production state defined in S120; and the previous perturbation label injection window has ended. When none of the three candidate parameters meet the injection conditions, no new process perturbation label sequence is generated in the current control cycle.

[0065] The maximum permissible perturbation amplitude for industrial control systems is calculated using the following formula:

[0066]

[0067] In the formula, p represents the candidate parameter number; This represents the maximum allowable perturbation amplitude corresponding to the p-th candidate parameter within the k-th control cycle; This represents the injection ratio coefficient corresponding to the p-th candidate parameter; The definition in S130 is adopted; This represents the current value of the p-th candidate parameter within the k-th control cycle; and Let represent the lower and upper limits of the allowed parameters for the p-th candidate parameter, respectively.

[0068] The injection ratio coefficient is read from the enterprise's process database and determined based on records in historical qualified production data where the corresponding candidate parameters, after slight changes, still meet the quality allowable boundaries. In this embodiment, the injection ratio coefficients corresponding to the traction speed reference value, the bridge output speed setting value, and the preheating wrap angle setting value are set to 0.20, 0.15, and 0.10, respectively.

[0069] The actual perturbation amplitude of an industrial control system is determined by the following formula:

[0070]

[0071] In the formula, This represents the actual perturbation amplitude corresponding to the p-th candidate parameter within the k-th control cycle; This indicates the upper limit of the rated amplitude that the corresponding actuator is allowed to perform within one perturbation unit.

[0072] The rated amplitude upper limit is read from the equipment controller parameter configuration table. The minimum adjustable resolution of the corresponding actuator is denoted as... The minimum adjustable resolution is read from the equipment controller parameter configuration table or equipment debugging record. (In this embodiment, the rated amplitude upper limit of the traction motor controller is set to 5m / min, and the minimum adjustable resolution is set to 0.1m / min). When Less than If the corresponding candidate parameter does not meet the injection conditions, it is determined that the perturbation bearing parameter, actual perturbation amplitude, minimum adjustable resolution, actuator allowable error, actuator allowable response time, and perturbation identifier injection window length are combined to form perturbation injection constraint data.

[0073] S220: Based on the perturbation injection constraint data formed in S210, the industrial control system generates a process perturbation identification sequence for the current cardboard segment to be identified. The process perturbation identification sequence refers to a sequence of parameter changes composed of multiple perturbation units arranged in a defined direction, maintaining a zero mean within the complete perturbation identification injection window, and used to form an identifiable cross-machine response.

[0074] A perturbation unit refers to the smallest injection segment that maintains the same perturbation direction and actual perturbation amplitude within a continuously preset number of control cycles. In this embodiment, one process perturbation identification sequence includes eight perturbation units, each lasting for two control cycles. Four perturbation units correspond to the increasing direction, and the other four perturbation units correspond to the decreasing direction.

[0075] The perturbation quantity corresponding to the qth perturbation unit is calculated according to the following formula:

[0076]

[0077] In the formula, q represents the perturbation element number; This represents the perturbation amount of the p-th perturbation carrying parameter in the q-th perturbation element; The definition in S210 is adopted; This represents the direction identifier corresponding to the q-th perturbation unit, with 1 for increasing directions and -1 for decreasing directions.

[0078] The directional signs meet the following requirements:

[0079]

[0080] In the formula, Q represents the total number of perturbation units contained in one process perturbation identifier sequence.

[0081] The industrial control system selects a directional arrangement from a pre-stored set of zero-mean directional arrangements that differs from the previous process perturbation identifier sequence. For example, it might use an increasing, decreasing, increasing, increasing, decreasing, decreasing, increasing, decreasing directional arrangement. The industrial control system assigns a unique sequence number to each process perturbation identifier sequence, which includes the order number, perturbation carrying parameter number, injection start time, and sequence number.

[0082] The industrial control system determines the injection start time only if the parameters for three consecutive control cycles are not missing in the periodic operation data, the current order has been running stably for 60 seconds, the last paper receiving action ended more than 60 seconds ago, there are currently no fault alarms or manual emergency interventions, and the previous perturbation label injection window has ended. The perturbation label injection plan is formed by combining the sequence number, perturbation carrying parameters, actual perturbation amplitude, directional arrangement, perturbation unit duration, perturbation label injection window start time, perturbation label injection window end time, and injection start time.

[0083] S230: The industrial control system, according to the perturbation identifier injection plan formed in S220, superimposes the perturbation quantity corresponding to each perturbation unit onto the cruise control reference value of the perturbation carrying parameter, and generates cruise control commands.

[0084]

[0085] In the formula, This represents the cruise control command for the p-th perturbation carrying parameter in the q-th perturbation unit; This indicates the corresponding cruise control reference value; The definition in S220 is adopted.

[0086] When the perturbation load parameter is the traction speed reference value, the cruise control command is sent to the traction motor controller and the traction motor encoder feedback value is read; when the perturbation load parameter is the bridge output speed set value, the cruise control command is sent to the bridge output drive device and the bridge output motor encoder feedback value is read; when the perturbation load parameter is the preheating wrap angle set value, the cruise control command is sent to the preheating wrap angle adjustment mechanism and the preheating wrap angle feedback value is read.

[0087] The industrial control system records the command transmission time, actuator feedback value, and actual arrival time, and calculates the execution deviation according to the following formula:

[0088]

[0089] In the formula, Indicates execution deviation; This indicates the actuator feedback value; Use the definitions from this step.

[0090] When the execution deviation is not greater than the actuator's allowable error, and the time difference between the actual arrival time and the command transmission time is not greater than the actuator's allowable response time, the corresponding perturbation unit is marked as successfully executed. When all perturbation units execute successfully, the sequence number, perturbation carrying parameters, cruise control reference value, cruise control command, perturbation identifier injection window start time, perturbation identifier injection window end time, command transmission time, actuator feedback value, actual arrival time, execution deviation, and execution success mark are combined to form a perturbation identifier execution record for S310 to identify cross-machine response. When any perturbation unit fails to execute, the current process perturbation identifier sequence is marked as an invalid execution sequence and is not sent to S310. The cruise control reference value of the corresponding perturbation carrying parameters is restored, and the reason for the execution failure is recorded.

[0091] S3 specifically includes the following sub-steps:

[0092] S310: The industrial control system reads the perturbation identifier execution record generated in S230 and determines the response candidate window based on the perturbation identifier injection window, minimum response delay, and maximum response delay. The response candidate window refers to the time interval extracted from the actual start time of the process perturbation identifier sequence according to the minimum and maximum response delays of the corresponding response data. The minimum and maximum response delays are determined based on the equipment topology, paperboard transport distance, paperboard running speed at the corresponding workstation, and equipment debugging records generated in S110.

[0093] The industrial control system extracts response data within the response candidate window from the periodic operation data generated by S120 and calculates the response matching degree:

[0094]

[0095] In the formula, the parameter number p, the perturbation element number q, the total number of perturbation elements Q, and the direction identifier are used in S220. Definition of ; h represents the response data number; d represents the number of perturbation units corresponding to the delay of the response to be tested; This represents the average value of the h-th response data after a delay of d perturbation units; This represents the average value of the h-th response data item within the response candidate window; Indicates the response direction coefficient; This indicates the degree of response matching.

[0096] When the perturbation carrying parameter increases, the response direction coefficient is set to 1 when the corresponding response data typically increases; and to -1 when the corresponding response data typically decreases. The response direction coefficient is determined by equipment debugging records or historical qualified production data. The industrial control system sequentially changes d within the preset response delay range and selects the maximum response matching degree. When the maximum response matching degree is not less than 0.70, and the corresponding response amplitude is not less than the sensor's minimum effective resolution (which is read from the equipment controller parameter configuration table), the corresponding response data is recorded as a valid response.

[0097] When the perturbation load parameter is the traction speed reference value, the response data includes the traction motor encoder feedback value, the double-facer inlet cardboard running speed, and the traction tension detection value; when the perturbation load parameter is the bridge output speed setting value, the response data includes the bridge output motor encoder feedback value, the bridge paper storage length, and the double-facer inlet cardboard running speed; when the perturbation load parameter is the preheat wrap angle setting value, the preheat wrap angle feedback value, the double-facer inlet paper surface temperature, the double-facer outlet cardboard surface temperature, and the double-facer outlet cardboard moisture content.

[0098] A valid cross-machine response is confirmed only when at least two valid response data exist for each perturbation carrying parameter; if this condition is not met, the corresponding perturbation identifier execution record is marked as a response identification failure record and is not sent to S320. The sequence number, response data number, response start time, response end time, response delay, and maximum response matching degree are combined to form the cross-machine response correspondence.

[0099] S320: The industrial control system defines a continuous section of cardboard actually affected by the same process perturbation identification sequence as a virtual cardboard batch and uses the corresponding sequence number as the batch identifier. The identification reference encoder is an encoder used to record the actual length of cardboard affected by the process perturbation identification sequence. When the perturbation load parameter is the traction speed reference value, the bridge output speed setting value, or the preheating wrap angle setting value, the identification reference encoder is the traction motor encoder, the bridge output motor encoder, or the single-facer exit encoder, respectively.

[0100] The length of a virtual cardboard section is calculated using the following formula:

[0101]

[0102] In the formula, g represents the virtual cardboard batch number; This indicates the length of the g-th virtual cardboard segment, in meters; E represents the identifier reference encoder. This indicates the unit pulse paperboard length corresponding to the reference encoder, in m / pulse, where pulse represents the unit of pulse output by the encoder; and These represent the start times of the perturbation identifier injection window. and the end time of the perturbation identifier injection window The corresponding cumulative pulse count. The unit pulse cardboard length is calculated during the equipment commissioning phase based on the encoder pulse count per revolution, the corresponding effective roller circumference, and the transmission ratio, and then entered into the industrial control system.

[0103] For transmission paths crossing bridges, the industrial control system calculates the effective transmission distance using the following formula:

[0104]

[0105] In the formula, This represents the effective transmission distance when the g-th virtual cardboard segment is transferred from upstream station u to downstream station v; This represents the reference transmission distance between the upstream station u and the downstream station v when the cross-bridge paper storage length is at the reference value. This represents the average bridge storage length during which the g-th virtual cardboard segment passes through the bridge. This indicates the baseline bridge paper storage length entered during the equipment commissioning phase.

[0106] The industrial control system calculates the estimated times for virtual cardboard batches to enter and leave each station based on the cumulative pulse count of the encoders at each station, the effective transmission distance, and the cardboard running speed at the corresponding station. For correction stations where valid response data can be obtained, the estimated times are corrected using the response start time and response end time generated by S310; if the difference between the estimated time and the corresponding response time is no greater than 2 seconds, the corresponding response time is used as the station entry time or station exit time.

[0107] For workstations without response detection devices, the entry and exit times are determined based on the effective transmission distance between adjacent correction workstations and the cardboard running speed. If the difference between the expected time and the corresponding response time is greater than 2 seconds, the corresponding trajectory is marked as an invalid trajectory and is not sent to S330. The batch segment identifier, virtual cardboard batch segment length, entry time of each workstation, exit time of each workstation, average bridge paper storage length, and position correction results are combined to form the virtual cardboard batch segment transmission trajectory.

[0108] Based on the virtual paperboard batch transfer trajectory formed by S320, the industrial control system extracts the initial moisture content of each layer of base paper from the production line basic configuration data formed by S110, and extracts all valid sampled values ​​of each virtual paperboard batch during its passage through the single-facer, bridge, gluing machine, double-facer and longitudinal and transverse cutting stations from the periodic operation data formed by S120, and arranges them according to the collection timestamp to form the station parameter time sequence data.

[0109] For single-facers, the following parameters are considered: board running speed, preheating wrap angle feedback value, steam pressure detection value, and initial moisture content of each layer of base paper. For bridges, the following parameters are considered: bridge paper storage length, bridge output speed detection value, and bridge dwell time obtained by subtracting the entry time from the exit time. For gluing machines, the following parameters are considered: glue roller gap feedback value, glue viscosity, board running speed, and glue line continuity. For double-facers, the following parameters are considered: temperature detection value of each hot plate zone, hot plate pressure detection value, traction speed reference value, and traction tension detection value. For slitting and cross-cutting stations, the following parameters are considered: slitting speed, cross-cutting speed, and cross-cutting length deviation.

[0110] The industrial control system calculates the average value of continuous parameters; records the entry, exit, and maximum values ​​for the cross-bridge paper storage length; records the maximum values ​​for transverse warpage, longitudinal warpage, and cross-cut length deviation of the paperboard; records the minimum value for glue line continuity; and records the corrugation identification result of the corresponding virtual paperboard batch segment as 1 when any exposed corrugation identification result is 1. The above statistical results form the representative values ​​of the workstation parameters.

[0111] The industrial control system further extracts the cruise parameter control records for the corresponding time interval from the control command log according to the virtual cardboard batch transfer trajectory. The cruise parameter control records include the traction speed reference value, bridge output speed setpoint, preheating wrap angle setpoint, steam pressure setpoint, glue roller gap setpoint, temperature setpoints for each hot plate zone of the double-sided machine, hot plate pressure setpoint, and traction tension setpoint corresponding to the cruise control reference value, parameter correction amount, correction command sending time, actuator feedback value, and command source identifier.

[0112] The batch segment identifier, virtual cardboard batch segment transmission trajectory, station parameter timing data, station parameter representative value, cardboard quality inspection results, and cruise parameter control record are combined to form a cross-machine process history, which is used by S410 to establish the initial state of the shadow production line.

[0113] S4 specifically includes the following sub-steps:

[0114] S410, the industrial control system reads the cross-machine process history generated by S330, and extracts the virtual paperboard batch transmission trajectory, station parameter timing data, station parameter representative value, paperboard quality inspection results and cruise parameter control records of the corresponding virtual paperboard batch according to the batch segment identifier.

[0115] A shadow production line refers to a virtual production line established by an industrial control system that does not send control commands to the physical equipment in the paperboard production line, and simulates the segmentation of virtual paperboard batches according to the direction of paperboard movement. Workstation sections include single-facer processing section, bridge conveyor section, gluing machine processing section, double-facer processing section, and slitting and cross-cutting processing section.

[0116] The industrial control system selects historical qualified production data from the enterprise's process database that match the current order's cardboard layer count and flute type, and whose basis weight deviation of each layer is within a preset matching range. It then combines the process state before entering the workstation section, the cruise control parameters within the workstation section, and the process state after leaving the workstation section into a fitting data set. After performing logarithmic mapping on parameters with nonlinear coupling characteristics such as cardboard moisture content and surface temperature, it uses the least squares method to fit the process state transmission coefficient, control parameter action coefficient, and fixed offset of each workstation section.

[0117] The industrial control system uses the most recent 50 valid cross-machine process histories to correct the fitted dataset; if there are fewer than 50 valid cross-machine process histories, all currently available valid cross-machine process histories are used for correction.

[0118] Under the m-th compensation path, the first... The predicted process status of each workstation section is updated according to the following formula:

[0119]

[0120] In the formula, m represents the compensation path number; Indicates the workstation section number; and These represent the virtual cardboard batch entering and leaving the first... Predicted process status for each workstation section; This indicates the m-th compensation path. Cruise control parameters corresponding to each workstation section; Represents the process state transfer coefficient matrix; This represents the matrix of control parameter coefficients; This indicates a fixed offset.

[0121] Predicted process conditions include paperboard moisture content, paperboard surface temperature, glue line continuity, traction tension, paperboard transverse warpage, paperboard longitudinal warpage, exposed corrugation risk value, and cross-cut length deviation; cruise control parameters include traction speed reference value, bridge output speed setting value, preheat wrap angle setting value, steam pressure setting value, glue roller gap setting value, temperature setting value of each hot plate zone of the double facer, hot plate pressure setting value, and traction tension setting value.

[0122] The industrial control system uses the layered base paper properties corresponding to the current virtual paperboard batch, the virtual paperboard batch transmission trajectory, and the representative values ​​of the station parameters as the initial inputs to the shadow production line, and uses the current compensation path to re-determine the paperboard quality inspection results.

[0123] When the deviation between the predicted and actual values ​​is not greater than the model allowable error recorded in the enterprise process database (in this embodiment, the model allowable error for the predicted moisture content of the double-sided machine's exit cardboard is set to ±0.5%, and the model allowable error for the predicted surface temperature is set to ±2℃), the shadow production line's effective status identifier is recorded as valid; when any deviation is greater than the corresponding model allowable error, the shadow production line's effective status identifier is recorded as invalid, the cruise control baseline value of the previous control cycle is maintained, and S420 is not entered. The cross-machine process history, the coefficients of each workstation section, and the shadow production line's effective status identifier are combined to form the initial state of the shadow production line.

[0124] S420, the industrial control system generates compensation paths only for virtual cardboard batches whose effective status is identified as valid in the shadow production line. The compensation path refers to the sequence of parameters that, while keeping the initial state of the virtual cardboard batch unchanged, are set according to a preset method for cruise control parameters and input into the shadow production line for segmented extrapolation.

[0125] The current compensation path uses the cruise control parameters that have been actually executed on the physical cardboard production line; if the compensation path is not executed, the parameter correction in the current compensation path will be set to 0, and only the cruise control reference value will be retained; if the compensation path is executed with a delay, the parameter correction will remain unchanged, and the number of delay cycles will be increased gradually starting from a delay of 1 control cycle until the maximum number of delay cycles allowed for the dwell time of the corresponding workstation section is reached.

[0126] The alternative compensation path selects candidate parameters from the cruise control parameters and generates single-parameter paths that increase or decrease the adjustment step size by 1 step size according to the adjustment step size recorded in the equipment controller parameter configuration table or the enterprise process database. In this embodiment, the adjustment step size of the traction speed reference value is set to 2 m / min, and the adjustment step size of the preheating wrap angle setting value is set to 5°.

[0127] If no single-parameter path that meets the quality allowable boundary exists, two candidate parameters are generated, each adjusted by one adjustment step, to form a two-parameter path. All cruise control parameters must not exceed the corresponding allowable range determined in S110.

[0128] For quality indicators with only an allowable upper limit, the corresponding risk is calculated by dividing the predicted value by the allowable upper limit; for adhesive line continuity, the adhesion risk is calculated using the following formula:

[0129]

[0130] In the formula, This represents the adhesion risk corresponding to the m-th compensation path; This indicates the lower limit of allowable continuity of the adhesive line; This indicates the predicted value for the continuity of the adhesive line.

[0131] The industrial control system records the larger of the transverse warping risk and the longitudinal warping risk of the paperboard as the warping risk; if corrugation is predicted to occur, the corrugation risk is recorded as 1, otherwise it is recorded as 0.

[0132] Overall risk is calculated using the following formula:

[0133]

[0134] In the formula, This represents the overall risk corresponding to the m-th compensation path; , and These represent warping risk, exposed rib risk, and cutting deviation risk, respectively.

[0135] When the predicted value of glue line continuity is less than or equal to 0, the predicted value of moisture content of double-sided machine exit paperboard or the predicted value of surface temperature of double-sided machine exit paperboard exceeds the corresponding allowable range, or the risk of exposed corrugations is 1, the overall risk is corrected to 1, and the corresponding path is recorded as a quality non-conforming path.

[0136] Industrial control systems calculate energy consumption per unit qualified area using the following formula:

[0137]

[0138] In the formula, This represents the energy consumption per unit qualified area corresponding to the m-th compensation path, in kWh / m². This indicates electricity consumption, expressed in kWh. This indicates the amount of steam consumed, expressed in kg. This represents the steam conversion factor, with units of kWh / kg; This indicates the area of ​​qualified cardboard, in m².

[0139] The current electricity and steam consumption of the compensation path are obtained from the energy metering data generated by S120, while the corresponding values ​​for other paths are extrapolated from the shadow production line based on historical qualified production data. The industrial control system selects the alternative compensation path with the lowest energy consumption per unit qualified area from the qualified paths as the target alternative compensation path; if no alternative compensation path meets the quality allowable boundary, the target alternative compensation path identifier is recorded as null. The path number, path type, cruise control parameters, parameter correction amount, parameter correction time, various risks, overall risk, energy consumption per unit qualified area, and target alternative compensation path identifier of each compensation path are combined to form the compensation path extrapolation result.

[0140] S430: The industrial control system reads the compensation path deduction result generated by S420, uses the current compensation path as a comparison benchmark, and calculates the counterfactual compensation contribution.

[0141]

[0142]

[0143]

[0144] In the formula, Indicates contribution to quality improvement; Indicates the contribution to timing improvement; Indicates the contribution to energy consumption improvement; , and These represent the overall risks associated with not executing the compensation path, the current compensation path, and the target delayed execution of the compensation path, respectively. and These represent the energy consumption per unit qualified area corresponding to the current compensation path and the target alternative compensation path, respectively.

[0145] The target delayed execution compensation path refers to the delayed execution compensation path that first brings any quality indicator to the quality allowable boundary. If none of the delayed execution compensation paths bring any quality indicator to the quality allowable boundary, the delayed execution compensation path with the highest overall risk is selected as the target delayed execution compensation path. If there is no target alternative compensation path, the energy consumption improvement contribution is recorded as 0.

[0146] The path number, cruise control parameters, various risks, overall risk, energy consumption per unit qualified area, number of delay cycles corresponding to the target delayed execution compensation path, quality improvement contribution, timing improvement contribution, energy consumption improvement contribution, and shadow production line effective status identifier are combined to form the counterfactual compensation contribution amount, which is used by S510 to identify the compensation status and by S520 to update the cruise control parameters.

[0147] S5 specifically includes the following sub-steps:

[0148] S510: The industrial control system reads the counterfactual compensation contribution generated in S430 and extracts the cruise control parameters, various risks, overall risk, energy consumption per unit qualified area, number of delay cycles, and shadow production line effective status identifier for each compensation path according to the batch segment identifier. An overall risk less than 1 indicates that the quality indicators of the corresponding path are within the quality allowable boundary; an overall risk greater than or equal to 1 indicates that at least one quality indicator has reached or exceeded the quality allowable boundary. When the shadow production line effective status identifier is invalid, the cruise control baseline value of the previous control cycle is maintained, and S520 is not entered.

[0149] When the shadow production line is marked as valid, it is uniquely identified in the following order: quality correction state, delayed compensation state, inefficient compensation state, excessive compensation state, necessary compensation state, steady-state maintenance state, and unidentified state. If the overall risk of the current compensation path is greater than or equal to 1, and no compensation path is executed or there is a path with an overall risk less than 1 among the target alternative compensation paths, it is identified as a quality correction state, and the path with the lowest energy consumption per unit qualified area is recorded as the correction target path.

[0150] When the overall risk of the target delayed execution compensation path is greater than or equal to 1, and the overall risk of the current compensation path is less than 1, it is identified as a delayed compensation state. When the overall risk of the target alternative compensation path is less than 1, the overall risk of the current compensation path is less than 1, and the energy consumption improvement ratio is not less than the preset energy consumption improvement threshold, it is identified as an inefficient compensation state.

[0151] The energy consumption improvement rate is calculated according to the following formula:

[0152]

[0153] In the formula, Indicates the percentage of energy consumption improvement; and The definition in S430 is adopted.

[0154] The preset energy consumption improvement threshold is read from the enterprise's process database; in this embodiment, it is set to 0.03. If the overall risk of not executing a compensation path is less than 1, and its energy consumption per unit qualified area is lower than the energy consumption per unit qualified area corresponding to the current compensation path, it is identified as an over-compensation state. If the overall risk of not executing a compensation path is greater than or equal to 1, the overall risk of the current compensation path is less than 1, and there is no inefficient compensation state, it is identified as a necessary compensation state.

[0155] If the overall risk of both the current compensation path and not executing the compensation path is less than 1, and the conditions for inefficient compensation or overcompensation are not met, the state is identified as steady-state maintenance. If the overall risk of both the current compensation path and not executing the compensation path is greater than or equal to 1, and there is no target alternative compensation path with an overall risk less than 1, the state is identified as unidentified. If the target alternative compensation path is null, the inefficient compensation state is not identified; if the overall risk of not executing the compensation path is less than 1, the quality correction state can still be identified.

[0156] The batch segment identifier, compensation status, target parameter number, correction target path number, target alternative compensation path number, number of delay cycles of target delayed execution compensation path, and energy consumption improvement ratio are combined to form the compensation status identification result.

[0157] S520: The industrial control system reads the compensation state identification result generated by S510, updates only the target parameters corresponding to the compensation state, and keeps the other cruise control parameters unchanged from the cruise control reference values ​​of the previous control cycle. In the quality correction state, it reads the correction target path and uses parameters whose values ​​differ from the current compensation path as candidate update parameters.

[0158] Under the delayed compensation state, the target parameter correction amount remains unchanged, and the command transmission time is adjusted in advance according to the following formula:

[0159]

[0160] In the formula, 'a' represents the target parameter number; Indicates the time when the updated correction command was sent; Indicates the time when the correction command was sent before the update; Indicates the number of delay cycles in the target delayed execution compensation path; The definition of the preset control cycle in S120 is retained.

[0161] In inefficient compensation mode, the target alternative compensation path is read, and parameters whose values ​​differ from those in the current compensation path are used as candidate update parameters. In overcompensation mode, the adjustment step size is rolled back by one step each time, in the direction opposite to the parameter correction direction.

[0162] Candidate update parameters are subject to limiting processing according to the following formula:

[0163]

[0164] In the formula, and These represent the values ​​of the target parameter after the update and before the update, respectively. Indicates the adjustment step size whose sign is opposite to the original parameter correction direction; and These represent the lower and upper limits of the target parameters determined in S110, respectively. Following the definition in S130, the adjustment step size is read from the equipment controller parameter configuration table or the enterprise process database. In this embodiment, the adjustment step size of the bridge output speed setpoint is... Set to 3 m / min, steam pressure setpoint adjustment step Set it to 0.05MPa.

[0165] After each update, the industrial control system inputs the candidate update parameters into the shadow production line for re-analysis. Under quality correction mode, the review is passed when the overall risk is less than 1; under inefficiency compensation mode, the review is passed when the overall risk is less than 1 and the energy consumption improvement ratio is not less than the preset energy consumption improvement threshold; under lag compensation mode, the review is passed when the overall risk decreases and no new quality indicators exceed the limits; under overcompensation mode, if the next rollback will make the overall risk greater than or equal to 1, the parameter values ​​from the previous successful review are used.

[0166] If the review fails, the parameter update validity status is marked as invalid, and the cruise control reference value of the previous control cycle is retained. In necessary compensation and steady-state states, the current compensation path is retained. In unidentified states, the injection of new process disturbance identifier sequences is paused, the safety process parameter group in the enterprise process database is read, and the traction speed reference value is reduced according to the preset deceleration step size (in this embodiment, the preset deceleration step size is set to 5 m / min).

[0167] After verification, the industrial control system sends the updated cruise control parameters to the corresponding actuator controller, recording the applicable batch segment identifier, parameter effective time, command recipient, and actuator feedback value. The updated cruise control parameters are formed by combining the batch segment identifier, compensation status, target parameter number, updated value, correction command sending time, overall risk after verification, energy consumption per unit qualified area after verification, applicable batch segment identifier, parameter effective time, command recipient, actuator feedback value, and parameter update validity status identifier.

[0168] In S530, the industrial control system inputs the updated cruise control parameters generated in S520 into the shadow production line and obtains the predicted quality stability margin according to the calculation rules in S130. The predicted quality stability margin refers to the quality stability margin calculated based on the predicted quality index values ​​output by the shadow production line.

[0169] The industrial control system prioritizes adjusting the actual perturbation amplitude for the next control cycle. The duration of the perturbation unit is only extended when the actual perturbation amplitude reaches the safe injection boundary formed by S210 and the predicted quality stability margins for three consecutive virtual paperboard batches have all increased. The injection time is only adjusted when the injection time overlaps with the response time interval of the actuator corresponding to the order change, paper splicing, or previous parameter correction command. Only one of the following—the actual perturbation amplitude, the perturbation unit duration, and the injection time—is adjusted within the same control cycle.

[0170] The actual perturbation amplitude for the next control cycle is calculated using the following formula:

[0171]

[0172] In the formula, and This indicates the actual perturbation magnitude between the next control cycle and the current control cycle; This represents the perturbation amplitude adjustment coefficient, which is read from the enterprise process database (in this embodiment). (Value is 0.1). This indicates a stability margin in the predicted quality. The definition in S130 is adopted; and The definitions in S210 are all adopted.

[0173] When the parameter update valid status flag is invalid, the predicted quality stability margin is less than 0.15, the shadow production line valid status flag is invalid, the cycle operation data is missing for three consecutive control cycles, the cardboard production line is in an unstable production state, or the corresponding actuator has an execution failure record, the injection of new process perturbation flag sequence will be suspended, and the updated cruise control parameters that have been reviewed will be maintained.

[0174] When the injection conditions are met, the cruise control parameters are updated as the cruise control reference values ​​for the next control cycle, and the adjusted process perturbation identifier sequence is used as the process perturbation identifier sequence for the next control cycle. The system then returns to S120 to continue collecting cycle operation data. After the subsequent virtual cardboard batch completes actual production, the industrial control system calculates the actual quality stability margin through S130 and writes the difference between the actual quality stability margin and the predicted quality stability margin into the corresponding cross-machine process history for S410 to correct the shadow production line.

[0175] All the above formulas are performed using dimensionless numerical calculations; the relevant formulas are based on empirical models that approximate the real situation, obtained through extensive data collection and software simulation fitting. The preset parameters and thresholds involved in the formulas can be conventionally set and adjusted by those skilled in the art according to the physical constraints of the actual application scenario.

[0176] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0177] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0178] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive control method for paperboard production process parameters based on an intelligent cruise system, characterized in that, Includes the following steps: S1. Collect order specifications, layered paper properties, equipment topology, real-time operating parameters and paperboard quality inspection results through the industrial control system, calculate the quality stability margin, and form paperboard production cruise status data; S2. Select the perturbation load parameters based on the paperboard production cruise status data, determine the safety injection boundary, generate a zero-mean process perturbation identifier sequence, and superimpose it onto the corresponding cruise control reference value to form a perturbation identifier execution record. S3. Based on the perturbation identifier execution record, identify cross-machine response, divide virtual cardboard batches, correct the transmission position of each station, bind the corresponding station parameters, cardboard quality inspection results and cruise parameter control records, and form a cross-machine process history. S4. Input the cross-machine process history into the shadow production line, deduce the current compensation path, the non-execution compensation path, the delayed execution compensation path, and the alternative compensation path respectively, compare the quality risk and energy consumption per unit qualified area of ​​each path, and form the counterfactual compensation contribution.

2. The adaptive control method for paperboard production process parameters based on an intelligent cruise system according to claim 1, characterized in that, Also includes: S5. Identify the compensation status based on the counterfactual compensation contribution, update the cruise control parameters and process disturbance identification sequence, and use the updated cruise control parameters as the cruise control reference value for the next control cycle.

3. The adaptive control method for paperboard production process parameters based on an intelligent cruise system according to claim 1, characterized in that, S1 specifically includes: Read order specifications, layered paper properties, equipment topology, process allowable range, and quality allowable boundary to form basic production line configuration data; According to the preset control cycle, real-time operating parameters, cardboard quality inspection results, energy metering data and production line status indicators are collected to complete time alignment and anomaly verification. Anomaly verification includes range verification, timestamp verification and rate of change verification. Abnormal data in a single cycle is marked with supplementary values ​​to form cycle operating data. Based on the basic configuration data and cycle operation data of the production line, the normalized residual ratio of each parameter from the corresponding allowable boundary is calculated, the minimum value is selected as the quality stability margin, and the data are combined to form the paperboard production cruise status data.

4. The adaptive control method for paperboard production process parameters based on an intelligent cruise system according to claim 1, characterized in that, S2 specifically includes: Based on the paperboard production cruise status data, the perturbation bearing parameters are selected from the traction speed reference value, the bridge output speed setting value and the preheat wrap angle setting value. The actual perturbation amplitude is determined by combining the quality stability margin, the process allowable range and the upper limit of the actuator rated amplitude, thus forming perturbation injection constraint data. Based on the perturbation injection constraint data, a process perturbation identifier sequence consisting of multiple perturbation units with zero mean is generated. The directional arrangement, unit duration, and injection time are determined to form a perturbation identifier injection plan.

5. The adaptive control method for paperboard production process parameters based on an intelligent cruise system according to claim 4, characterized in that, Also includes: The process perturbation identifier sequence is superimposed onto the corresponding cruise control reference value to generate cruise control commands, and the actuator feedback value and execution deviation are recorded to form a perturbation identifier execution record.

6. The adaptive control method for paperboard production process parameters based on an intelligent cruise system according to claim 1, characterized in that, S3 specifically includes: Based on the perturbation identifier execution record, determine the response candidate window, extract the response data from the periodic operation data, calculate the response matching degree, and form a cross-machine response correspondence when at least two response data pass the valid response judgment. Based on the cross-machine response correspondence, the cumulative pulse count of the reference encoder, the equipment topology, and the cross-bridge paper storage length, virtual paperboard batches are divided, the transmission positions of each station are corrected, and a virtual paperboard batch transmission trajectory is formed. Based on the virtual cardboard batch transmission trajectory, extract and bind the time sequence data of each workstation parameter, the representative value of the workstation parameter, the cardboard quality inspection results, and the cruise parameter control records to form a cross-machine process history.

7. The adaptive control method for paperboard production process parameters based on an intelligent cruise system according to claim 1, characterized in that, S4 specifically includes: Based on the cross-machine process history, the virtual cardboard batch segment transfer trajectory, representative values ​​of workstation parameters, and cruise parameter control records are extracted. The process state transfer relationship is fitted and corrected according to the workstation segment to form the initial state of the shadow production line. When the shadow production line is effective, generate the current compensation path, the non-execution compensation path, the delayed execution compensation path, and the alternative compensation path, deduce the quality risk and energy consumption per unit qualified area of ​​each path, screen the target alternative compensation path, and form the compensation path deduction result.

8. The adaptive control method for paperboard production process parameters based on an intelligent cruise system according to claim 7, characterized in that, Also includes: Using the current compensation path as a benchmark, calculate the contribution of quality improvement, timing improvement, and energy consumption improvement, and record the number of delay cycles of the target delayed execution compensation path to form the counterfactual compensation contribution.

9. The adaptive control method for paperboard production process parameters based on an intelligent cruise system according to claim 2, characterized in that, S5 specifically includes: Based on the counterfactual compensation contribution, the compensation state is identified in the following order: quality correction state, delayed compensation state, inefficient compensation state, excessive compensation state, necessary compensation state, steady state maintenance state, and unidentified state, thus forming the compensation state identification result. Based on the compensation status identification results, the corresponding target parameters are updated. After verification by the shadow production line, the effective updated cruise control parameters are sent to the corresponding actuator controller, and the applicable batch segment identifier and parameter effective time are recorded.

10. The adaptive control method for paperboard production process parameters based on an intelligent cruise system according to claim 9, characterized in that, Also includes: The predicted quality stability margin is calculated based on the updated cruise control parameters. When the injection conditions are met, the process perturbation identification sequence is adjusted, and the updated cruise control parameters are used as the cruise control reference value for the next control cycle.

Images