A method and system for collaborative control of sponge-flame composite process

By employing a collaborative control method for the flame bonding process, and utilizing multi-source data to invert interfacial viscosity, the airflow tension is predicted and collaboratively compensated. This solves the bonding quality problem caused by uneven heat input during the sponge bonding process, achieving higher production stability and product consistency.

CN123120152APending Publication Date: 2026-11-17QINGDAO RENCHENG SPONGE PROD CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611469567.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-21
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing flame lamination production lines struggle to maintain the bonding quality of the sponge surface when heat input is insufficient or excessive, leading to problems such as inadequate interface wetting, insufficient peel strength, sponge collapse, or reduced thickness. Furthermore, the control methods are unable to identify critical states and make timely adjustments.

Method used

By acquiring multi-source flame data, we can correlate material segments and deconstruct heat, invert interfacial viscosity, predict viscosity changes, and perform coordinated compensation for airflow tension. This ensures data correspondence for sponge material segments during heating, transportation, and compounding stages, reduces control errors, and identifies undermelting, overmelting, and critical instability trends in advance.

Benefits of technology

It improves the consistency of the composite interface and the dimensional stability of the finished product, reduces problems such as sponge collapse, uneven composite peel strength and edge curling of roll material, and enhances the control precision and stability of the continuous production process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123120152A_ABST
    Figure CN123120152A_ABST
Patent Text Reader

Abstract

This invention relates to the field of sponge composite processing technology, and more particularly to a collaborative control method and system for the sponge flame composite process. The method includes the following steps: acquiring multi-source flame data, and correlating material segments based on the multi-source flame data to obtain material segment correlation data; performing thermal deconstruction on the material segment correlation data to obtain interface heating data; performing viscosity inversion based on the interface heating data to obtain interface viscosity data; performing hysteresis prediction based on the interface viscosity data to obtain viscosity prediction data; performing window margin deconstruction on the viscosity prediction data to obtain viscosity deviation data; and performing collaborative compensation based on the viscosity deviation data to obtain airflow tension control data. This invention improves the bonding uniformity and production stability of sponge flame composite through interface viscosity prediction and airflow-tension collaborative compensation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of sponge composite processing technology, and in particular to a method and system for collaborative control of the sponge flame composite process. Background Technology

[0002] Flame lamination is a continuous processing technology that involves instantaneously heating the surface of polyurethane foam with a flame, softening or slightly melting the foam walls, and then bonding it to fabric, TPU film, or other sheet materials under the pressure of a lamination roller. The bonding quality of the lamination interface is closely related to the actual degree of heating of the foam surface. This actual heating state is affected not only by flame temperature, fuel gas flow rate, combustion air flow rate, and spray gun spacing, but also by the foam density, thickness, pore structure, roll conveying speed, and tension. Insufficient heat input results in low melting of the foam surface, easily leading to inadequate wetting and insufficient peel strength; excessive heat input causes cell collapse, reduced foam thickness, surface hardening, and localized melt-through. Existing flame lamination production lines set process parameters based on flame temperature, roll speed, or peel results after lamination, and adjust them using temperature control loops and tension control loops respectively. Existing control methods typically implement corrections only after the temperature or bonding quality exceeds a set threshold. This makes it difficult to distinguish between the critical state where the interfacial viscosity is still within the target range but is gradually approaching the boundary, and the actual unstable state where the interfacial viscosity has exceeded the target range. This leads to control conflicts and causes instability in the flame bonding of sponges. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention proposes a collaborative control method and system for the sponge-flame composite process, thereby resolving at least one of the aforementioned technical problems.

[0004] This application provides a method for the coordinated control of a sponge-flame composite process, including the following steps: Acquire multi-source flame data and perform material segment association based on the multi-source flame data to obtain material segment association data; Thermal deconstruction of the material segment correlation data yields the interface heating data; Viscosity inversion is performed based on interfacial heating data to obtain interfacial viscosity data; Viscosity prediction data is obtained by performing hysteresis prediction based on interfacial viscosity data. The viscosity prediction data is deconstructed using a window margin to obtain viscosity deviation data; Collaborative compensation based on viscosity deviation data yields airflow tension control data.

[0005] This invention correlates multiple data sources, including flame temperature, airflow state, roll material running state, and material properties, across material segments. This ensures accurate data correspondence for the same sponge material segment during heating, transport, and lamination stages, reducing control errors caused by misaligned sampling timing. By deconstructing the effective heat and inverting interfacial viscosity based on the thermal conductivity hysteresis characteristics of the sponge's porous elastic structure, deviations caused by judging melting degree solely based on surface temperature are avoided. By predicting viscosity changes when the sponge reaches the lamination position and quantifying the approach, intrusion, and out-of-bounds states of the target viscosity window, undermelting, overmelting, and critical instability trends can be identified in advance. Coordinated compensation of flame airflow and roll material tension based on viscosity deviations reduces problems such as sponge collapse and hardening, uneven lamination peel strength, and roll material edge curling, improving the consistency of the lamination interface, the dimensional stability of the finished product, and the control precision of continuous production processes.

[0006] Optionally, thermal deconstruction includes: Heat source mapping is performed based on material segment correlation data to obtain material segment heat source data; residence conversion is performed on material segment heat source data to obtain residence heat flux data; loss deconstruction is performed based on residence heat flux data to obtain effective heat input data; tension correction is performed based on effective heat input data to obtain compression heating data; and lateral fusion is performed based on compression heating data to obtain interface heating data.

[0007] This invention maps the associated data of material segments to the corresponding flame action area to ensure consistency in the heat source, action location, and time relationship of each material segment, reducing heat mismatch caused by continuous roll material transport. By combining transport speed and flame action length for dwell calculation, instantaneous flame parameters can be converted into the actual heat flux per unit area. By eliminating ineffective heat caused by environmental heat dissipation through loss deconstruction, the effective heat input is made closer to the actual heating state of the sponge surface foam wall. Tension correction is used to compensate for the influence of roll material stretching and thickness compression on the heat density, and the heating difference between the edge and center of the roll material is characterized by lateral fusion, thereby obtaining continuous and comparable interface heating data. This provides accurate thermal basis for subsequent interface viscosity inversion and airflow-tension coordinated control, helping to reduce problems such as local overmelting, undermelting, and uneven lateral composite strength.

[0008] Optionally, viscosity inversion includes: Temperature rise data is obtained by analyzing the interface heating data; thermal hysteresis data is corrected based on the interface temperature rise data to obtain interface temperature change data; melting degree is characterized based on the interface temperature change data to obtain interface melting data; rheological parameter mapping is performed based on the interface melting data to obtain interface rheological data; and viscosity state inversion is performed based on the interface rheological data to obtain interface viscosity data.

[0009] This invention analyzes the temperature rise of the sponge surface under flame conditions by analyzing the interfacial heating data, thus avoiding the use of instantaneous temperature values ​​to characterize the melting state. By correcting for thermal hysteresis, it compensates for the slow heat conduction within the high-porosity sponge and the temperature response deviation caused by the continued migration of residual heat after leaving the flame zone, making the interfacial temperature change data closer to the actual thermal state of the material. The degree of melting characterizes the temperature change into the degree of foam wall softening, localized melting, and pore compression. Then, through rheological parameter mapping, a correspondence between the melting state and viscoelastic properties is established, yielding continuously quantifiable interfacial viscosity data. This reduces misjudgments of undermelting and overmelting caused by relying solely on surface temperature measurement, thereby improving the consistency of composite strength and reducing sponge collapse, hardening, and localized delamination.

[0010] Optionally, the lagged forecasts include: Viscosity time delay data is obtained by identifying time delay parameters based on interfacial viscosity data; viscosity evolution data is obtained by recursively extrapolating heat conduction based on viscosity time delay data; and viscosity prediction data is obtained by predicting composite time points based on viscosity evolution data.

[0011] In this invention, by identifying time-delay parameters in the interfacial viscosity data, the time offset between internal heat migration and surface melting response during the process of transporting the sponge from the flame-heated position to the composite pressing position can be represented, avoiding the direct equation of the currently measured interfacial state with the state at subsequent composite moments. Based on the viscosity time-delay data, heat conduction recursion can be performed to continuously characterize the evolution trends of residual heat diffusion, foam wall softening, and viscosity recovery or decline, giving the viscosity change process a clear temporal continuity. By predicting the composite timing, the interfacial viscosity when the material segment reaches the composite roller can be obtained in advance, providing a forward-looking basis for airflow pre-compensation and tension synergistic correction. This reduces problems such as under-melting, over-melting, peel strength fluctuations, and localized sponge collapse caused by control lag, improving the responsiveness, control stability, and finished product consistency of the flame composite process.

[0012] Optionally, the window margin deconstruction includes: Based on the viscosity prediction data and the preset target viscosity range data, interval intrusion analysis and window out-of-bounds deconstruction are performed to obtain the first deviation data and the second deviation data, respectively; based on the first deviation data and the second deviation data, dual-domain margin fusion is performed to obtain the viscosity deviation data.

[0013] This invention compares predicted viscosity data with the target viscosity range and performs range intrusion analysis and window boundary deconstruction respectively. This allows for the differentiation between potential risks that are still within the target range but continuously approaching the boundary and under-melting or over-melting instability states that have already exceeded the target range. This avoids the traditional single threshold judgment that only triggers adjustment after boundary crossing. The first deviation data reflects the direction and degree of intrusion of the predicted viscosity towards the upper and lower boundaries, while the second deviation data characterizes the boundary crossing direction, amplitude, and duration. Through dual-domain margin fusion, the warning domain and instability domain can be uniformly converted into continuous and comparable viscosity deviation quantities. This improves the sensitivity of critical state identification and the hierarchy of control decisions, reducing problems such as under-melting delamination, over-melting collapse, and frequent abrupt changes in control quantities at composite interfaces.

[0014] Optionally, interval intrusion analysis includes: Boundary distance data is obtained by calculating the boundary distance based on the viscosity prediction data and the preset target viscosity range data; invasion direction data is obtained by identifying the invasion direction based on the boundary distance data; invasion depth data is obtained by quantifying the invasion depth based on the invasion direction data and the boundary distance data; and first deviation data is obtained by aggregating the interval margin based on the invasion depth data.

[0015] This invention calculates the distance between the predicted viscosity data and the upper and lower boundaries of the target viscosity range, continuously characterizing the remaining safety margin of the predicted viscosity within the stable window, avoiding the identification of anomalies only after the viscosity actually exceeds the limit. Identifying the intrusion direction based on the boundary distance clarifies whether the predicted viscosity is approaching the high-viscosity undermelting side or the low-viscosity overmelting side. By combining the intrusion direction and boundary distance to quantify the intrusion depth, it can distinguish different risk levels such as slight approach, critical approach, and deep intrusion, and differentiate between instantaneous fluctuations and continuous deterioration trends. By aggregating the interval margins to form the first deviation data, the boundary risks of multiple material segments and different lateral positions can be uniformly converted into continuously comparable early warning quantities. This allows for the early initiation of airflow and tension compensation before actual boundary crossing occurs at the interface, reducing the probability of undermelting delamination, overmelting collapse, and frequent boundary adjustments, thereby improving the proactive control capability and stability of the flame composite process.

[0016] Optionally, window out-of-bounds destructuring includes: Based on viscosity prediction data and preset target viscosity range data, cross-boundary orientation is determined to obtain cross-boundary orientation data; based on cross-boundary orientation data and viscosity prediction data, cross-boundary amplitude is quantized to obtain cross-boundary amplitude data; based on cross-boundary amplitude data, cross-boundary persistence is analyzed to obtain cross-boundary persistence data; based on cross-boundary persistence data and cross-boundary orientation data, instability quantization mapping is performed to obtain second deviation data.

[0017] In this invention, by comparing the predicted viscosity data with the upper and lower boundaries of the target viscosity range, the location of the viscosity deviation (too high or too low) can be identified first, thus distinguishing between two opposite abnormal states: under-melting and over-melting at the interface. Quantifying the amplitude of the deviation accurately reflects the severity of the predicted viscosity deviating from the stable window, avoiding a binary judgment based solely on whether the deviation occurs. Through continuous analysis of the deviation, transient disturbances and continuous instability can be distinguished, reducing misadjustments caused by flame fluctuations, sensor noise, or local material differences. Mapping the deviation direction and duration to a second deviation data point creates a comprehensive characterization of the anomaly type, deviation degree, and instability trend. This helps reduce insufficient peel strength caused by under-melting of the sponge surface, as well as cell collapse, thickness reduction, and surface hardening caused by over-melting, improving the stability of the continuous flame composite process and the consistency of the finished product.

[0018] Optionally, dual-domain margin fusion includes: The first and second deviation data are calibrated in the deviation domain to obtain dual-domain calibrated data; margin normalization is performed on the dual-domain calibrated data to obtain dual-domain normalized data; weight allocation is performed on the dual-domain normalized data to obtain dual-domain weighted data; deviation vector fusion is performed on the dual-domain weighted data to obtain viscosity deviation data.

[0019] In this invention, by calibrating the deviation domains of the first and second deviation data, the state ranges corresponding to the intrusion risk within the interval and the out-of-range instability can be clearly defined, avoiding direct comparison between the two types of deviations due to differences in dimensions, directions, and value ranges. Through margin normalization, the under-melting side, the over-melting side, and different risk levels can be uniformly mapped to a consistent deviation direction and numerical scale. Weight allocation based on dual-domain normalized data allows for adjustment of the contribution ratio between the warning domain and the instability domain according to the intrusion depth, the degree of out-of-range, and the duration of the out-of-range condition, maintaining high sensitivity during the critical approach phase and highlighting severe instability information after actual out-of-range events. By fusing deviation vectors to form continuous and directional viscosity deviation data, a unified control basis can be provided for airflow pre-compensation and tension collaborative correction, reducing control oscillations caused by conflicts between multi-source deviation information and improving the smoothness of the flame recombining process, response accuracy, and consistency of interface bonding quality.

[0020] Optionally, collaborative compensation includes: Airflow pre-compensation is performed based on viscosity deviation data to obtain airflow compensation data; tension co-correction is performed on airflow compensation data and viscosity deviation data to obtain tension compensation data; coupled constraint solution is performed based on airflow compensation data and tension compensation data to obtain airflow tension control data.

[0021] This invention utilizes pre-compensation of airflow based on viscosity deviation data to feedforward correction of flame heat transfer intensity and uniformity before the interface viscosity deviates from the target range, reducing the continuous accumulation of undermelting or overmelting trends. By combining airflow compensation results with viscosity deviation for tension synergistic correction, it can simultaneously compensate for the secondary effects of decreased elastic modulus, thickness compression, and roll stretching caused by heat softening of the sponge on the interface heating state, avoiding edge shrinkage, wrinkling, or deviation of the roll caused by adjusting only flame parameters. Through coupled constraint solving, the airflow compensation and tension compensation amounts are uniformly coordinated, limiting the variation range, response rate, and conflicting relationships of the two types of execution parameters, thereby obtaining continuous, smooth, and executable airflow and tension control data. This improves the matching degree between flame heat input and the mechanical state of the roll, reducing problems such as uneven peel strength at the composite interface, sponge collapse and hardening, and edge curling, thus improving the control stability and finished product consistency of the continuous composite process.

[0022] Optionally, this application also provides a collaborative control system for the sponge-flame composite process, comprising: At least one memory for storing computer-executable instructions; At least one processor, the processor being configured to execute computer-executable instructions stored in the memory to implement the collaborative control method for the sponge-flame composite process as described above; The data transmission interface, connected to the processor, is used to receive multi-source flame data and send airflow tension control data to the flame airflow adjustment device and the roll tension adjustment device.

[0023] The purpose of this invention is to: firstly, based on multi-source flame data, accurately map flame temperature, airflow state, roll speed, tension changes, and material properties to the same continuous material segment through material segment association, reducing interference caused by misalignment in the timing of multi-source sampling on control judgment; subsequently, perform thermal deconstruction on the flame thermal effects corresponding to the material segment, and obtain heating data that more closely reflects the actual state of the sponge interface based on residence time, heat loss, tension compression, and lateral heating differences. Combining temperature rise response, thermal conduction hysteresis, and melt rheological properties to invert interfacial viscosity, and predicting the viscosity evolution when the material segment reaches the composite position, thereby avoiding hysteresis-based adjustments based solely on the instantaneous temperature at the flame exit. By deconstructing the window margin, the boundary intrusion trend of viscosity within the target range and the degree of undermelting and overmelting after exceeding the boundary can be identified simultaneously, forming viscosity deviation data with directionality and risk level. Based on this, flame airflow and coil tension can be compensated in a coordinated manner, so that heat input adjustment and material mechanical state adjustment are matched with each other, reducing problems such as sponge overmelting and collapse, surface hardening, insufficient composite peel strength and coil edge curling, and improving the response timeliness, lateral consistency and continuous production stability of the flame composite process. Attached Figure Description

[0024] Other features, objects, and advantages of this application will become more apparent from the following detailed description of the non-limiting embodiments, taken with reference to the accompanying drawings: Figure 1 A flowchart illustrating the steps of a collaborative control method for a sponge-flame composite process according to an embodiment is shown. Figure 2 A flowchart illustrating the steps of a thermal deconstruction method according to one embodiment is shown; Figure 3 A flowchart illustrating the steps of a viscosity inversion method according to one embodiment is shown. Figure 4 A flowchart illustrating the steps of a hysteresis prediction method according to one embodiment is shown. Figure 5 A flowchart illustrating the steps of a collaborative compensation method according to an embodiment is shown. Detailed Implementation

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

[0026] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. Functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0027] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0028] A polyurethane foam roll with a thickness of 8mm and a width of 1.6m was selected for flame lamination. The roll running speed was set to 0.25m / s. The material was divided into segments with a length of 0.5m, and the width of the material was divided into three transverse interface units: left, middle, and right. The sampling timestamps of flame images, flame temperature, gas flow rate, oxidizing gas flow rate, roll speed, tension, and encoder pulse data were matched according to a preset synchronization threshold of 50ms to obtain the material segment association data corresponding to each material segment.

[0029] Taking the 25th material segment as an example, the heat source range, gas flow rate, and residence time of the two flame stations were mapped and calculated. After deducting convective and radiative heat loss, and through tension expansion area correction and lateral fusion, the effective heat per unit area of ​​the left, middle, and right interface units were obtained as 3.47 kJ / m², 3.19 kJ / m², and 2.90 kJ / m², respectively. Temperature rise analysis and thermal hysteresis correction were performed based on material density, specific heat capacity, effective interface thickness, and thermal diffusivity, yielding corresponding interface temperatures of 420 K, 410 K, and 400 K. The material softening initiation temperature was set to 393 K, and the full melting temperature to 433 K, resulting in melting degrees of 0.675, 0.425, and 0.175 for the three interface units. Viscosity inversion was performed based on a pre-calibrated rheological parameter table, yielding current interface viscosities of 1913 Pa·s, 2832 Pa·s, and 4162 Pa·s for the three interface units.

[0030] The sampling interval for interface viscosity data was set to 20 ms, the preset maximum hysteresis order was set to 10, and the relevant attenuation threshold was set to 0.2. Based on the time delay parameter identification, the viscosity hysteresis time of the right interface unit was determined to be 0.12 s. Using the current sampling time as the starting point for heat conduction recursion, the predicted viscosities of the right interface unit at the first and second prediction times were found to be 3480 Pa·s and 3560 Pa·s, respectively. The preset target viscosity range was [1800, 3200] Pa·s. Since the right interface unit exceeded the target viscosity limit for two consecutive prediction times, its out-of-bounds duration was 0.12 s. The preset allowable out-of-bounds duration was 0.3 s, so the viscosity deviation at the second prediction time was 0.360. Based on the lateral position of the material segment, this viscosity deviation was mapped to the right flame nozzle partition. With the deviation dead zone set to 0.05, the tension start threshold set to 0.15, the airflow correction coefficient set to 0.6 L / min, and the tension correction coefficient set to 3 N, the airflow pre-compensation amount is: 0.6 × (0.360 - 0.05) = 0.186 L / min.

[0031] The current gas flow rate at the right nozzle is 12.00 L / min. After adjusting for rate of change and upper / lower limit constraints, the target gas flow rate is adjusted to 12.19 L / min. Since the viscosity deviation exceeds the tension activation threshold, the tension compensation is calculated as: -3 × (0.360 - 0.15) = -0.630. The current coil tension is 16.00 N. After adjusting for amplitude and rate of change constraints, the target tension is adjusted to 15.37 N. The system sends the target gas flow rate and target tension to the flame flow rate regulating device and the coil tension regulating device, respectively. After three control cycles, the predicted viscosity of the right interface unit decreases from 3560 Pa·s to 3110 Pa·s, re-entering the target viscosity range. Furthermore, the maximum viscosity difference between the left, middle, and right interface units decreases from 2248 Pa·s to 610 Pa·s.

[0032] To verify the control effect, under the same material, speed, and environmental conditions, a conventional independent airflow control group and a collaborative control group according to this embodiment were set up, each continuously processing 20 rolls, each 100m in length. Test results showed that after adopting this method, the proportion of material length with interfacial viscosity within the target range increased from 83.1% to 96.7%; the average composite peel strength increased from 3.08 N / cm to 3.42 N / cm, and the coefficient of variation of transverse peel strength decreased from 12.4% to 4.9%; the rate of undermelting, overmelting, and local delamination defects decreased from 5.6% to 1.2%; and the incidence of edge curling of the roll decreased from 4.1% to 0.9%. Simultaneously, the average adjustment time from viscosity anomaly identification to re-entry into the target range was shortened from 1.18s to 0.62s. Therefore, this method can unify and correlate the material's heating state, interfacial viscosity evolution, and the mechanical state of the roll material, and simultaneously adjust the flame airflow and roll material tension before or in the early stages of viscosity instability, thereby improving the uniformity of interfacial heating, the consistency of composite peel strength, and the stability of the continuous production process.

[0033] Please see Figures 1 to 5 This application provides a method for collaborative control of a sponge-flame composite process, comprising the following steps: 01. Acquire multi-source flame data and perform material segment association based on the multi-source flame data to obtain material segment association data; In one embodiment, flame image sequences, flame temperature data, gas flow rate data, oxidizing gas flow rate data, roll material running speed data, tension data, and encoder pulse data are acquired at each flame treatment station. Simultaneously, material surface temperature data and ambient temperature data are acquired, and the acquisition timestamps are recorded uniformly. Using the material's initial end as the zero point, the cumulative running distance of the material is calculated based on the encoder pulses. ,in, The cumulative number of pulses up to time t. The number of pulses per encoder revolution. To measure the circumference of the roller. According to the preset material segment length. The materials are numbered, and the position range of the k-th material segment is: Let the installation distance of the j-th flame station from the zero point be . Then the material position at this workstation at time t is: When satisfied At that moment, the flame image, flame temperature, gas flow rate, oxidizing gas flow rate, operating speed, and tension data collected at that time are associated with the k-th material segment. Different data sources are matched according to their most recent timestamps, and the matching time difference must not exceed a preset synchronization threshold. =50ms; exceeding Data is marked as missing and not supplemented across segments. Data is archived according to material segment number, flame station number, entry time, exit time, and corresponding multi-source data sequences to obtain material segment associated data.

[0034] 02. Perform thermal deconstruction on the material segment correlation data to obtain interface heating data; In one embodiment, the corresponding gas flow rate, flame temperature, nozzle coverage area, material running speed, tension, and ambient temperature are read according to the material segment number and flame station number, and the material width is divided into several heating units along the transverse direction. For the k-th material segment, the j-th flame station, and the r-th heating unit, it is determined whether the flame acts on the unit based on the nozzle coverage coordinates; when the heating unit is within the nozzle coverage area, the heat source intensity of the material segment is calculated. ,in, This refers to the gas volume flow rate. This refers to the calorific value per unit volume of fuel gas, which is the equivalent effective heat input per unit volume after calibration through tests of flame temperature, air-fuel ratio, and injection distance. The effective coverage area of ​​the flame on the heated unit is used to obtain the heat source data for the material segment. This is based on the length of the flame-affected zone along the material's direction of travel. and material section running speed Determine the dwell range of the material section at this workstation [t] in ,t out ], where: And perform time integration on the heat source intensity within the residence interval: We obtained the resident heat flux data.

[0035] Based on material surface temperature Ambient temperature convective heat transfer coefficient Surface emissivity and Stefan-Boltzmann constant Calculate the heat loss during the stay: , Let be the cumulative heat loss per unit area of ​​the r-th heated unit during the period when the k-th material segment is subjected to the flame at the j-th flame station. Let be the ambient temperature at time t, and then according to Obtain effective heat input data, Let be the residence heat flux of the r-th heated unit in the k-th material segment within the j-th flame station; when When that happens, set it to zero to avoid creating non-physical negative heat input.

[0036] Based on measured tension and the pre-defined material tension-longitudinal elongation relationship Determine the actual unfolded area of ​​the material segment when heated: , Let the initial area of ​​the r-th heated unit in the k-th material segment be the area when it is not under tension, and calculate the heat received per unit actual area: The compression heating data is obtained. The heat generated by each flame station on the same material segment is directly summed according to the transverse heating unit number: Then, the materials are arranged according to their segment numbers and horizontal coordinates to form interface heating data that characterizes the heating location, heating duration, and effective heat input per unit area of ​​each material segment. Alternatively, the heat generated by each flame station on the same material segment is used according to a preset cooling factor, with the output temperature of the previous station as the initial temperature of the next station. After cooling / diffusion based on the running time between stations, the heat input of the next station is superimposed to obtain interface heating data. The cooling factor is obtained through experimental pre-calibration.

[0037] 03. Viscosity inversion is performed based on the interface heating data to obtain the interface viscosity data; In one embodiment, for the r-th interface heating unit of the k-th material segment, the effective heat received per unit area is read. The duration and sequence of heating, and based on the material density. Specific heat capacity Effective thickness of the interface and initial temperature Calculate the theoretical interface temperature: Based on the lag in heat transfer from the surface to the interface, the thermal hysteresis constant is calculated using the material's thermal diffusivity α: π is the constant of pi, according to the sampling interval =20ms recursive correction of interface temperature: ,in, This represents the interface temperature change data at the m-th sampling time. It is based on the pre-calibrated softening initiation temperature of the material. and full melting temperature Determine the degree of interfacial melting : The reference viscosity can be obtained by referring to the pre-calibrated rheological parameter table using the material grade, interface temperature range, and melting degree range. Activation energy of viscous flow And the unmelted correction factor β, and the interfacial viscosity is inverted according to the following formula: ,in, and All temperatures are Kelvin temperatures, where R is the gas constant. This is the reference temperature for rheological testing. Arranged according to material section number, transverse heating unit number, and sampling time. We obtained the interfacial viscosity data.

[0038] 04. Perform hysteresis prediction based on interface viscosity data to obtain viscosity prediction data; In one embodiment, for the k-th material segment and the r-th interface unit, the interfacial viscosity sequence is arranged according to the sampling time. Calculate the viscosity autocorrelation coefficient for different hysteresis orders l: ,in, This is the mean of the viscosity series. Within the preset maximum hysteresis order... Inside, it will be the first time to meet The lag order is determined as And calculate the viscosity time delay parameters: ,in, For viscosity sampling interval, To preset the relevant attenuation threshold, such as 0.2, , To determine the maximum permissible viscosity hysteresis time based on the material's thermal conductivity test, The sampling interval for the interface viscosity data is M, where M is the total number of valid sampling points involved in time delay identification; if in If the condition is not met, then take... The recursive calculation is based on the relationship between the current viscosity and the viscosity at the corresponding time delay: ,in, This represents the viscosity evolution value at the next sampling time. After each iteration, the newly obtained viscosity evolution value is substituted into the calculation for subsequent time points until the predicted time points are reached. and , Let m be the sampling time. Record the viscosity value, material segment number, interface unit number, and prediction time corresponding to the two prediction time points to obtain the viscosity prediction data.

[0039] 05. Perform window margin decomposition on the viscosity prediction data to obtain viscosity deviation data; In one embodiment, for the k-th material segment and the r-th interface unit, the predicted viscosity values ​​at each prediction time point are read. And obtain the target viscosity range preset for the corresponding material. Set the lower boundary warning width within the target range. and upper boundary warning width The upper and lower boundary warning widths are pre-calibrated using historical experimental data, and the interval intrusion deviation is calculated using the following formula: Here, negative values ​​indicate intrusion into the low-viscosity boundary, positive values ​​indicate intrusion into the high-viscosity boundary, and the absolute value represents the intrusion depth, thus obtaining the first deviation data. It is then determined whether the predicted viscosity exceeds the target viscosity range, and the directional boundary exceedance magnitude is calculated: Starting from the first time the boundary was crossed, the number of samples crossing the boundary in the same direction was continuously counted. And calculate the duration of the out-of-bounds error: ,in, This represents the time interval between adjacent prediction points. The out-of-bounds amplitude is normalized according to the target viscosity interval width, and combined with the out-of-bounds duration to obtain the second deviation data: ,in, A preset allowed out-of-bounds duration, such as 0.3 seconds, is used. Dual-domain fusion is performed using a method that prioritizes out-of-bounds access and supplements for intrusive intervals. When When the viscosity exceeds the limit, the second deviation data is used as the current viscosity deviation; when no limit is exceeded, the first deviation data is used as the current viscosity deviation, that is: The viscosity deviation data is obtained by recording the material segment number, interface unit number, prediction time point, deviation direction, deviation amplitude, and out-of-bounds duration.

[0040] 06. Based on viscosity deviation data, collaborative compensation is performed to obtain airflow tension control data.

[0041] In one embodiment, viscosity deviation data is mapped to the corresponding flame nozzle partition based on the material segment number and lateral position. For the j-th nozzle partition, the set of interface elements it covers is... The interface unit corresponding to the viscosity deviation with the largest absolute value is determined in the following way: The viscosity deviation corresponding to this interface unit is used as the partition control deviation. ,in, This indicates that the predicted viscosity is higher than the target viscosity, requiring increased heating. This indicates that the predicted viscosity is lower than the target state, and heating needs to be suppressed. A deviation dead zone is set. The airflow pre-compensation amount is generated according to the following: ,in, The airflow correction factor is calibrated through material testing. The viscosity deviation is determined based on the normal fluctuation range under stable production conditions, for example, by statistically analyzing the maximum absolute value of the viscosity deviation during stable operation. The pre-compensation amount for the gas flow is then added to the current gas flow rate. And apply amplitude and rate of change constraints: ,in, and These are the minimum and maximum allowable gas flow rates at the nozzle, respectively. The maximum rate of change of airflow. To control the cycle. When Exceeding the tension start threshold When the value is 0.15, tension-based collaborative correction is initiated: ,in, This is the tension correction factor. When viscosity is high, tension is reduced to decrease material stretching and enhance interfacial adhesion; when viscosity is low, tension is increased to suppress excessive melting and localized buildup at the interface. The target tension is: ,in, and The allowable tension range of the material, This represents the maximum rate of change of tension. Check the coupling constraint between airflow and tension. When When, the tension is no longer reduced; when If the airflow and tension reverse direction in adjacent control cycles, the target tension of the previous control cycle is maintained, and only the airflow adjustment for the current control cycle is performed to avoid oscillation of the dual actuators. The nozzle zone number, material section number, and target gas flow rate are then specified. Target tension Together with the execution time, they form the airflow tension control data.

[0042] Optionally, thermal deconstruction includes: 21. Based on the material segment association data, perform heat source mapping to obtain the material segment heat source data; In one embodiment, the gas volume flow rate in the associated data of the material segment is read according to the material segment number, flame station number, and transverse heating unit number. 1. Flame coverage coordinates and material movement position. The nozzle coverage area of ​​the j-th flame station is intersected with the coordinates of the r-th heating unit of the k-th material segment. If an intersection exists, the flame station is identified as the effective heat source for that heating unit, and the effective coverage area is determined based on the intersection area. The heat source intensity of the material section is calculated as follows: ,in, Let be the calorific value per unit volume of the gas; when there is no overlap, let: The material segment number, flame station number, heated unit number, action time, and heat source intensity are combined to form the material segment heat source data.

[0043] 22. Perform residence conversion on the heat source data of the material section to obtain residence heat flux data; In one embodiment, the length of the flame-affected zone along the material travel direction is read. and material section running speed The moment when the leading edge of the material segment enters the flame zone is recorded as... The moment that satisfies the following equation will be the departure moment. : Integrate the heat source intensity of the material segment within the dwelling interval: Thus, the heat source intensity that changes over time is converted into the cumulative heat input per unit area during the actual residence period of the material segment, and residence heat flux data is obtained.

[0044] 23. Based on the resident heat flux data, loss deconstruction is performed to obtain the effective heat input data; In one embodiment, the material surface temperature is read within the same residence interval. and ambient temperature Calculate convective heat loss and radiative heat loss separately: ,in, The convective heat transfer coefficient is... For the surface emissivity of the material, Here is the Stefan-Boltzmann constant. The effective heat input is: ,when When that time, set it to zero, that is: This yields effective heat input data.

[0045] 24. Based on the effective heat input data, tension correction is performed to obtain compression heating data; In one embodiment, the measured tension of the k-th material segment is read. The elongation is determined based on the pre-defined material tension-longitudinal elongation relationship: , To obtain the corresponding longitudinal elongation by substituting the measured tension of the k-th material segment at time t into the pre-calibrated tension-longitudinal elongation relationship function, the initial area of ​​the heated unit is used as the starting point. Calculate the actual unfolded area after being subjected to tension: Then, the effective heat input is converted to the actual unfolded area: This eliminates the change in heat per unit area caused by material stretching, thus obtaining compression heat data.

[0046] 25. Horizontal fusion is performed based on compressed heating data to obtain interface heating data.

[0047] In one embodiment, the compression heating data of each flame station acting on the same heating unit are directly accumulated according to the material segment number and the transverse heating unit number: If a flame station does not cover the heated unit, then its corresponding compression heat is set as follows: ,Will The interface heating data are obtained by arranging the material segments according to their numbers, lateral positions, heating start and end times, and effective heat received per unit area.

[0048] Optionally, viscosity inversion includes: 31. Perform temperature rise analysis based on interface heating data to obtain interface temperature rise data; In one embodiment, for the k-th material segment and the r-th interface unit, the effective heat per unit area in the interface heating data is read. And order: According to material density Specific heat capacity Effective thickness of the interface and initial temperature Calculate the theoretical interface temperature: ,in, For the theoretical temperature rise of the interface. This involves assigning the material segment number, interface element number, theoretical temperature rise, and... Record the interface temperature rise data.

[0049] 32. Based on the interface temperature rise data, thermal hysteresis is corrected to obtain interface temperature change data; In one embodiment, based on the effective thickness of the interface Given the material's thermal diffusivity α, calculate the thermal hysteresis constant: ,by As an initial condition, according to the sampling interval Actual temperature at the recursive interface: ,in, Let be the interface temperature at the m-th sampling time. This is used to correct the heat transfer hysteresis between the theoretical temperature rise and the actual interface temperature rise, thus obtaining the interface temperature change data.

[0050] 33. Based on the interface temperature change data, the degree of melting is characterized to obtain interface melting data; In one embodiment, the softening onset temperature of the corresponding material is read. and full melting temperature According to the interface temperature Determine the degree of melting: ,in, This indicates that no effective melting has occurred. This indicates that the material has reached a fully molten state. (The remaining text appears to be incomplete and possibly contains errors.) The corresponding temperature records are interface melting data.

[0051] 34. Rheological parameter mapping is performed based on interfacial melting data to obtain interfacial rheological data; In one embodiment, a corresponding rheological parameter calibration table is determined according to the material grade, and... Temperature range and Use the melting point range as the query condition to read the reference viscosity. Activation energy of viscous flow E a Unmelted correction factor β and reference temperature When the interface temperature or melting degree is at the boundary of an adjacent calibration interval, the rheological parameters of the corresponding boundary interval are used; when the interface temperature or melting degree exceeds the calibration range, the boundary parameters closest to the current state are used, and no calibration interval extrapolation is performed. The above rheological parameters are associated with the material segment number and the interface unit number to obtain the interface rheological data.

[0052] 35. Based on the interface rheological data, the viscosity state is inverted to obtain the interface viscosity data.

[0053] In one embodiment, the interface temperature, degree of melting, and rheological parameters are substituted into the viscosity inversion relationship: ,in Let R be the interfacial viscosity of the r-th interface element in the k-th material segment, and R be the gas constant. and All temperatures were obtained using Kelvin. The inversion was performed according to the material segment number, interface element number, and sampling time. This generates interfacial viscosity data.

[0054] Optionally, the lagged forecasts include: 41. Based on the interface viscosity data, time delay parameters are identified to obtain viscosity time delay data; In one embodiment, for the k-th material segment and the r-th interface unit, the interfacial viscosity sequence is arranged according to the sampling time: And calculate the sequence mean. .exist Within the range, calculate the autocorrelation coefficients corresponding to different lag orders sequentially: ,in, The lag order that first satisfies the following formula is determined as the optimal lag order. : If in If none of the above conditions are met within the range, then let: The viscosity time delay parameter is calculated according to the following formula: ,in, The sampling interval for interface viscosity data. To preset the relevant attenuation threshold. , The corresponding material segment number and interface unit number are recorded as viscosity time delay data.

[0055] 42. Based on viscosity time delay data, heat conduction is recursively extrapolated to obtain viscosity evolution data; In one embodiment, the viscosity at the current sampling time is read. and its forward movement Viscosity at each sampling time The difference between the two values ​​represents the sustained change in viscosity caused by thermal conduction hysteresis, and the viscosity at the next sampling time is recursively calculated using the following formula: ,in, Let be the predicted viscosity at the (m+1)th sampling time. When: When the viscosity increases, the recursive result continues the trend of increasing viscosity; conversely, when it decreases, the recursive result continues the trend of decreasing viscosity. After completing one recursion, The viscosity sequence is added and used as the current viscosity value in the subsequent recursive process to obtain viscosity evolution data containing the predicted time, predicted viscosity and corresponding time delay parameters.

[0056] 43. Based on viscosity evolution data, perform composite time point prediction to obtain viscosity prediction data.

[0057] In one embodiment, the current sampling time is used Starting from this point, we will respectively: and: The first and second prediction time points are set as follows. Calculations are performed periodically according to the aforementioned recursive relationship until the recursive time reaches the corresponding prediction time point. If the prediction time point and the sampling time do not coincide, the first recursive result no earlier than the corresponding prediction time point is selected. The viscosity prediction values ​​corresponding to the two prediction time points are denoted as follows: , The viscosity prediction data is obtained by associating and combining the material segment number, interface unit number, prediction time point and corresponding predicted viscosity.

[0058] Optionally, the window margin deconstruction includes: Based on viscosity prediction data and preset target viscosity range data, interval intrusion analysis and window out-of-bounds deconstruction are performed to obtain the first deviation data and the second deviation data, respectively. In one embodiment, for the k-th material segment and the r-th interface unit, the viscosity prediction value is read. and obtain the preset target viscosity range. Lower boundary warning width and upper boundary warning width When the predicted viscosity falls within the target viscosity range, the range intrusion deviation is calculated based on its distance from the warning boundary: Here, negative values ​​indicate intrusion into the low-viscosity boundary, positive values ​​indicate intrusion into the high-viscosity boundary, and the absolute value represents the intrusion depth, thus obtaining the first deviation data. When the predicted viscosity exceeds the target viscosity range, the out-of-bounds amplitude in the band direction is calculated: Starting from the first time the boundary was crossed, the number of samples that crossed the boundary consecutively in the same direction was counted. If the predicted viscosity returns to the target viscosity range or the direction of the out-of-range change, then reset: The duration of the boundary crossing is: Then, calculate the window out-of-bounds deviation: ,in, The time interval between adjacent prediction time points. The second deviation data is obtained by setting the allowed out-of-bounds duration.

[0059] Viscosity deviation data is obtained by performing dual-domain margin fusion based on the first deviation data and the second deviation data.

[0060] In one embodiment, the first deviation data and the second deviation data are fused according to the method of prioritizing boundary crossings and supplementing with interval intrusions: When the predicted viscosity has exceeded the limit, the second deviation data, which includes the direction, magnitude, and duration of the exceedance, is used; when the predicted viscosity has not yet exceeded the limit but has entered the boundary warning zone, the first deviation data is used; when the predicted viscosity is within the safe range, let: =0, record the material segment number, interface unit number, prediction time point, deviation direction, deviation value and over-limit duration together to obtain viscosity deviation data.

[0061] Optionally, interval intrusion analysis includes: Boundary distance data is obtained by calculating the boundary distance based on the viscosity prediction data and the preset target viscosity range data. In one embodiment, viscosity prediction values ​​are read for the k-th material segment, the r-th interface unit, and the m-th prediction time point. and target viscosity range Calculate the distances to the lower and upper boundaries respectively: , ,when and When the predicted viscosity is within the target viscosity range, it indicates that the predicted viscosity is within the target viscosity range. If any boundary distance is less than zero, it is considered a window out-of-bounds situation, and the interval intrusion amount is not calculated in this step. The two boundary distances and the corresponding prediction time points are recorded as boundary distance data.

[0062] Intrusion direction identification is performed based on boundary distance data to obtain intrusion direction data; In one embodiment, the lower boundary warning width is read. and upper boundary warning width And pre-defined: Determine the direction of invasion as follows : ,in, This indicates an invasion into the low-viscosity boundary. This indicates an invasion into the high-viscosity boundary. This indicates that the area is within a safe zone, from which the intrusion direction data is obtained.

[0063] Intrusion depth data is obtained by quantizing the intrusion direction data and boundary distance data. In one embodiment, the normalized invasion depth is calculated based on the invasion direction and the corresponding boundary distance. : in: The closer the predicted viscosity is to the target viscosity boundary, The closer the value is to 1, the better. If the predicted viscosity is within the safe range or has exceeded the target viscosity range, then let: This yields the intrusion depth data.

[0064] Interval margin aggregation is performed based on intrusion depth data to obtain the first deviation data.

[0065] In one embodiment, the invasion direction and invasion depth are combined to form a directional interval invasion deviation: For the same material segment and the same interface unit, the results generated at the first and second prediction time points Arranged sequentially according to the prediction time; if duplicate records exist for the same prediction time point, retain the record with the larger absolute value, without averaging or weighting. The material segment number, interface unit number, prediction time point, intrusion direction, and intrusion depth are included. Together they form the first deviation data.

[0066] Optionally, window out-of-bounds destructuring includes: Based on viscosity prediction data and preset target viscosity range data, cross-boundary orientation is determined to obtain cross-boundary orientation data; In one embodiment, viscosity prediction values ​​are read for the k-th material segment, the r-th interface unit, and the m-th prediction time point. and target viscosity range Determine the direction of the boundary crossing according to the following formula. : ,in, This indicates that the high viscosity has exceeded the limit. This indicates that the low viscosity has exceeded the limit. This indicates that the boundary has not been exceeded. The boundary deviation location, material segment number, interface unit number, and prediction time point are recorded as boundary deviation location data.

[0067] The out-of-bounds amplitude is quantified based on the out-of-bounds orientation data and viscosity prediction data to obtain the out-of-bounds amplitude data. In one embodiment, the corresponding target viscosity interval boundary is selected based on the boundary crossing direction, and the boundary crossing amplitude in the direction is calculated: ,in, This indicates the amplitude exceeding the upper boundary. This indicates the amplitude below the lower boundary. The corresponding out-of-bounds location record is the out-of-bounds amplitude data.

[0068] Based on the out-of-bounds amplitude data, out-of-bounds persistence analysis is performed to obtain out-of-bounds persistence data; In one embodiment, it is determined whether adjacent predicted values ​​cross the boundary in the same direction according to the order of prediction time points. If the current time point satisfies: The number of consecutive out-of-bounds samples is then updated as follows: If the boundary is crossed for the first time at the current time or the direction of the boundary crossing changes, then let: If the predicted viscosity returns to the target viscosity range, then let: The duration of the boundary crossing is calculated as follows: ,in, This represents the time interval between adjacent prediction time points. The out-of-bounds amplitude, the number of consecutive out-of-bounds samples, and the out-of-bounds duration are correlated and recorded to obtain the out-of-bounds duration data.

[0069] Instability quantization mapping is performed based on out-of-bounds continuous data and out-of-bounds orientation data to obtain the second deviation data.

[0070] In one embodiment, the out-of-bounds amplitude is normalized according to the width of the target viscosity range, and the degree of instability is amplified based on the duration of the out-of-bounds movement. ,in, This is the preset allowed duration for exceeding the limit. When... season: High viscosity boundary crossings correspond to positive deviations, while low viscosity boundary crossings correspond to negative deviations. The absolute value of the deviation increases with the increase of the boundary crossing amplitude and duration. The parameters include material segment number, interface unit number, prediction time point, boundary crossing direction, boundary crossing amplitude, and boundary crossing duration. Together they form the second deviation data.

[0071] Optionally, dual-domain margin fusion includes: The first and second deviation data are calibrated in the deviation domain to obtain dual-domain calibration data. In one embodiment, for the k-th material segment, the r-th interface unit, and the m-th prediction time point, the first deviation data is read. Second deviation data And set interval intrusion domain markers respectively. and window out of bounds marker : , The two deviation values, corresponding domain labels, material segment numbers, interface unit numbers, and prediction time points are associated to obtain dual-domain calibration data.

[0072] Based on the dual-domain calibration data, margin-oriented normalization is performed to obtain dual-domain normalized data; In one embodiment, the first deviation data has been calculated according to the lower boundary warning width. and upper boundary warning width After normalization, the second deviation data has been adjusted according to the target viscosity range width. and the permitted duration of boundary crossing Quantization is performed, therefore repeated scaling calculations are no longer necessary; only the deviation direction is standardized. Based on the aforementioned intrusion direction... and cross-boundary direction ,get: , In this system, positive values ​​uniformly represent the high viscosity direction, and negative values ​​uniformly represent the low viscosity direction, thus obtaining dual-domain normalized data.

[0073] Weight allocation is performed based on the dual-domain normalized data to obtain dual-domain weighted data; In one embodiment, a mutually exclusive selection rule is used for weight allocation. When a window out-of-bounds error occurs, the second bias data is used preferentially; when no window out-of-bounds error occurs, the first bias data is used. , These are respectively formed into interval intrusion weighted components and window out-of-bounds weighted components: , This yields dual-domain weighted data.

[0074] Viscosity deviation data is obtained by fusing deviation vectors based on dual-domain weighted data.

[0075] In one embodiment, the interval intrusion weighted component and the window out-of-bounds weighted component are combined to form a deviation vector: The final viscosity deviation is obtained by summing the components in the same direction: Since the interval intrusion weight and window out-of-bounds weight are mutually exclusive, the actual fusion result is equivalent to: The material segment number, interface unit number, prediction time point, deviation direction, and By recording the data together, we obtain viscosity deviation data.

[0076] Optionally, collaborative compensation includes: 61. Perform airflow pre-compensation based on viscosity deviation data to obtain airflow compensation data; In one embodiment, viscosity deviation data is processed according to material segment number and lateral position. Mapped to the corresponding flame nozzle partition. For the j-th nozzle partition, in the set of interface elements it covers. Select the interface unit with the largest absolute value of deviation: The viscosity deviation corresponding to this interface unit is determined as the zone control deviation. ,in, This indicates a high viscosity deviation, requiring increased airflow to enhance heating; This indicates a low viscosity deviation, requiring reduced airflow to suppress heating. A deviation dead zone should be established. Calculate the airflow pre-compensation amount: Then, the current gas flow rate Apply rate of change constraints and upper and lower bound constraints: ,in, This is the airflow correction factor. and These are the minimum and maximum allowable gas flow rates at the nozzle, respectively. The maximum rate of change of airflow. To control the cycle, the nozzle zones are numbered. , and target gas flow rate Recorded as airflow compensation data.

[0077] 62. Tension-based co-correction is performed on the airflow compensation data and viscosity deviation data to obtain tension compensation data; In one embodiment, for the k-th material segment, the zone control deviation with the largest absolute value in its corresponding nozzle zone is selected as the basis for tension correction. When At this time, tension compensation is activated: ,in, This is the tension correction factor; high viscosity deviation corresponds to negative tension compensation, and low viscosity deviation corresponds to positive tension compensation. For the current tension... Apply rate of change constraints and upper and lower bound constraints: ,in, and The minimum and maximum allowable tension of the material. This represents the maximum rate of change of tension. The material segments are numbered... , and target tension Recorded as tension compensation data.

[0078] 63. Based on the airflow compensation data and tension compensation data, coupled constraint solution is performed to obtain airflow tension control data.

[0079] In one embodiment, it is checked whether there is excessive compensation in the same direction between the airflow and tension. When and At that time, the tension is no longer reduced, so that: ,when and At that point, the tension is no longer increased, so that: Read the airflow change from the previous control cycle. and tension change If both conditions are met: If both the airflow change and tension change are reversed relative to the previous control cycle, it is determined that the direction has reversed. To avoid control oscillations caused by the dual actuators and to maintain the target tension of the previous control cycle, only airflow compensation for the current control cycle is executed; otherwise, the current target gas flow rate is executed. and target tension Number the nozzle zones, material sections, and target gas flow rates. Target tension The airflow tension control data is obtained by combining the control timing with the airflow tension control data.

[0080] Optionally, this application also provides a collaborative control system for the sponge-flame composite process, comprising: At least one memory for storing computer-executable instructions; At least one processor, the processor being configured to execute computer-executable instructions stored in the memory to implement the collaborative control method for the sponge-flame composite process as described above; The data transmission interface, connected to the processor, is used to receive multi-source flame data and send airflow tension control data to the flame airflow adjustment device and the roll tension adjustment device.

[0081] Therefore, the embodiments should be regarded as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended application documents rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of the equivalents of the application documents be incorporated into the invention.

[0082] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for coordinated control of a sponge-flame composite process, characterized in that, Includes the following steps: Acquire multi-source flame data and perform material segment association based on the multi-source flame data to obtain material segment association data; Thermal deconstruction of the material segment correlation data yields the interface heating data; Viscosity inversion is performed based on interfacial heating data to obtain interfacial viscosity data; Viscosity prediction data is obtained by performing hysteresis prediction based on interfacial viscosity data. The viscosity prediction data is deconstructed using a window margin to obtain viscosity deviation data; Collaborative compensation based on viscosity deviation data yields airflow tension control data.

2. The method for coordinated control of the sponge-flame composite process according to claim 1, characterized in that, Thermal deconstruction includes: Heat source mapping is performed based on material segment correlation data to obtain material segment heat source data; residence conversion is performed on material segment heat source data to obtain residence heat flux data; loss deconstruction is performed based on residence heat flux data to obtain effective heat input data; tension correction is performed based on effective heat input data to obtain compression heating data; and lateral fusion is performed based on compression heating data to obtain interface heating data.

3. The method for coordinated control of the sponge-flame composite process according to claim 1, characterized in that, Viscosity inversion includes: Temperature rise data is obtained by analyzing the interface heating data; thermal hysteresis data is corrected based on the interface temperature rise data to obtain interface temperature change data; melting degree is characterized based on the interface temperature change data to obtain interface melting data; rheological parameter mapping is performed based on the interface melting data to obtain interface rheological data; and viscosity state inversion is performed based on the interface rheological data to obtain interface viscosity data.

4. The collaborative control method for the sponge-flame composite process according to claim 1, characterized in that, Lag forecasts include: Viscosity time delay data is obtained by identifying time delay parameters based on interfacial viscosity data; viscosity evolution data is obtained by recursively extrapolating heat conduction based on viscosity time delay data; and viscosity prediction data is obtained by predicting composite time points based on viscosity evolution data.

5. The method for coordinated control of the sponge-flame composite process according to claim 1, characterized in that, Window margin deconstruction includes: Based on the viscosity prediction data and the preset target viscosity range data, interval intrusion analysis and window out-of-bounds deconstruction are performed to obtain the first deviation data and the second deviation data, respectively; based on the first deviation data and the second deviation data, dual-domain margin fusion is performed to obtain the viscosity deviation data.

6. The method for coordinated control of the sponge-flame composite process according to claim 5, characterized in that, Interval intrusion analysis includes: Boundary distance data is obtained by calculating the boundary distance based on the viscosity prediction data and the preset target viscosity range data; invasion direction data is obtained by identifying the invasion direction based on the boundary distance data; invasion depth data is obtained by quantifying the invasion depth based on the invasion direction data and the boundary distance data; and first deviation data is obtained by aggregating the interval margin based on the invasion depth data.

7. The method for coordinated control of the sponge-flame composite process according to claim 5, characterized in that, Window out-of-bounds destructuring includes: Based on viscosity prediction data and preset target viscosity range data, cross-boundary orientation is determined to obtain cross-boundary orientation data; based on cross-boundary orientation data and viscosity prediction data, cross-boundary amplitude is quantized to obtain cross-boundary amplitude data; based on cross-boundary amplitude data, cross-boundary persistence is analyzed to obtain cross-boundary persistence data; based on cross-boundary persistence data and cross-boundary orientation data, instability quantization mapping is performed to obtain second deviation data.

8. The method for coordinated control of the sponge-flame composite process according to claim 5, characterized in that, Dual-domain margin fusion includes: The first and second deviation data are calibrated in the deviation domain to obtain dual-domain calibrated data; margin normalization is performed on the dual-domain calibrated data to obtain dual-domain normalized data; weight allocation is performed on the dual-domain normalized data to obtain dual-domain weighted data; deviation vector fusion is performed on the dual-domain weighted data to obtain viscosity deviation data.

9. The collaborative control method for the sponge-flame composite process according to claim 1, characterized in that, Collaborative compensation includes: Airflow pre-compensation is performed based on viscosity deviation data to obtain airflow compensation data; tension co-correction is performed on airflow compensation data and viscosity deviation data to obtain tension compensation data; coupled constraint solution is performed based on airflow compensation data and tension compensation data to obtain airflow tension control data.

10. A collaborative control system for a sponge-flame composite process, characterized in that, include: At least one memory for storing computer-executable instructions; At least one processor, the processor being configured to execute computer-executable instructions stored in the memory to implement the collaborative control method for the sponge-flame composite process as described in any one of claims 1 to 9; The data transmission interface, connected to the processor, is used to receive multi-source flame data and send airflow tension control data to the flame airflow adjustment device and the roll tension adjustment device.