An Adaptive Manufacturing Method for Recycled Fiber Corrugated Board Based on Online Identification of Raw Paper Quality
By identifying the quality of the base paper in recycled fiber corrugated board online and adjusting manufacturing parameters using controlled preheating excitation and dynamic response, the problems of insufficient bonding and warping in the production of recycled fiber corrugated board have been solved, resulting in a more stable production process.
Patent Information
- Application Number
- CN202611502416.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-24
- Publication Date
- 2026-11-17
AI Technical Summary
Existing corrugated board production lines struggle to identify dynamic quality changes in recycled fiber base paper in real time, leading to a mismatch between adhesive absorption rate and moisture migration rate, resulting in problems such as insufficient bonding, over-drying, and warping.
By acquiring the moisture content, tension, and longitudinal and transverse propagation characteristics of the base paper online, applying controlled preheating excitation, dynamically identifying the quality vector, and adjusting the gluing, preheating, and hot-pressing parameters according to glue absorption requirements, thermal desiccation, and moisture expansion asymmetry index, a joint quality control process is formed.
It improves the adaptability of the recycled fiber corrugated board manufacturing process to fluctuations in the quality of the base paper, ensures bonding quality, reduces warping and drying problems, and improves the stability and efficiency of the production line.
Smart Images

Figure CN123113385A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of corrugated board manufacturing technology, and in particular to an adaptive manufacturing method for recycled fiber corrugated board based on online identification of raw paper quality. Background Technology
[0002] Continuous production of corrugated cardboard typically includes processes such as unwinding the base paper, preheating, corrugating the core paper, applying starch glue, laminating the face paper and liner paper, and hot pressing and shaping on a double-facer. Existing production lines can adjust the steam volume, preheating wrap angle, glue gap, and line speed according to the basis weight of the base paper, the inlet moisture content, the surface temperature, or the degree of warping of the finished product. These adjustments are mostly based on static detection values or delayed finished product results.
[0003] After recycled fiber base paper undergoes multiple drying, reuse, and re-papermaking processes, the fiber porosity, swelling capacity, and surface bonding state change. Ash content, sizing degree, short fiber ratio, and fiber orientation also vary depending on the source of raw materials and the roll composition. Therefore, even if two rolls or different sections within the same roll have the same or similar basis weight and static moisture content, they may exhibit different sizing rates, moisture migration rates due to heat, and differences in longitudinal and transverse swelling.
[0004] If the amount of adhesive or heat is increased directly based solely on the static moisture content, adhesive accumulation and corrugation may occur in base paper with slow adhesive absorption, while base paper with rapid desiccation may become over-dried and lose its flexibility. Mismatch between the moisture expansion of the face paper and the liner paper can also cause upward curling, downward curling, or lateral waviness. Pursuing any one of the indicators such as adhesive strength, exit moisture content, or flatness in isolation may come at the expense of other indicators. In addition, the incoming base paper passes through multiple stations before reaching the exit inspection point. If the paper segment transit time is not considered, the exit feedback can easily affect subsequent paper segments of different qualities, especially near the roll change or in-roll mixing point.
[0005] Therefore, there is a need in the field for an adaptive manufacturing method that can transform the current manufacturing response of the base paper into an observable state on a continuous production line, allocate manufacturing parameters according to the mechanism of action, and ensure consistency of feedback objects. Summary of the Invention
[0006] Based on the above background, the purpose of this invention is to provide an adaptive manufacturing method for recycled fiber corrugated board based on online identification of base paper quality, so as to solve the problems of insufficient bonding, excessive drying and warping caused by multiple execution quantities being over-adjusted in the same direction and mismatched output feedback, which are difficult to identify the actual glue absorption, desiccation and swelling tendency of recycled base paper based solely on the basis weight and static moisture content of the base paper.
[0007] To achieve the above-mentioned objectives, the present invention provides the following technical solution: An adaptive manufacturing method for recycled fiber corrugated board based on online identification of base paper quality, the method comprising the following steps: Before the face paper and liner paper enter the preheater, the inlet reference signal of the corresponding paper is obtained. The inlet reference signal includes at least the moisture content signal, tension signal, and longitudinal and transverse propagation characteristic signals. The base paper is divided into paper segments along the conveying direction. Within the preset manufacturing quality boundary, the paper segment to be identified is subjected to controlled preheating excitation through the preheater. At the same time, the dynamic relationship between the dynamic response of the moisture content, the dynamic response of the tension, and the longitudinal and transverse propagation characteristics of the paper segment to be identified before and after the controlled preheating excitation is obtained. Based on the dynamic response of moisture content, the dynamic response of tension, and the dynamic relationship, the quality vector of the paper segment to be identified is obtained according to the pre-calibrated quality identification mapping. The quality vector includes the glue absorption demand index, the heat desiccation index, and the moisture expansion asymmetry index. The glue application parameters are determined based on the glue absorption demand index, the preheating parameters of the face paper base paper and the liner paper base paper are determined based on the heat desiccation index, and the hot pressing parameters of the double-sided machine are determined based on the moisture expansion asymmetry index, so that the deviation of each quality index is compensated by the execution amount in the corresponding process action direction. According to the determined gluing parameters, preheating parameters, and double-facer hot-pressing parameters, the base paper is sequentially corrugated, sizing, laminating, and hot-pressed. The glue line formation degree, exit paperboard moisture content, and paperboard warpage index are obtained online. Within the common quality boundary of the glue line formation degree, the exit paperboard moisture content, and the paperboard warpage index, the gluing parameters, the preheating parameters, and the double-facer hot-pressing parameters are corrected.
[0008] Preferably, the moisture content signal is acquired by a microwave moisture sensor or a near-infrared moisture sensor; the longitudinal and transverse propagation characteristic signals are acquired by ultrasonic sensors arranged along the longitudinal and transverse directions of the base paper; and the inlet reference signal also includes the basis weight signal of the base paper.
[0009] Preferably, the controlled preheating excitation is a step excitation that increases the preheating wrap angle by 3° to 8° or a heat excitation that has the same thermal effect as the step excitation. The controlled preheating excitation is applied to the base paper for 0.5 to 2.0 s, and the change in the moisture content of the base paper before and after the excitation is 0.15 to 0.60 percentage points.
[0010] Preferably, the quality identification mapping is pre-calibrated through the following steps: The rate of change of dynamic response of moisture content, the rate of change of dynamic response of tension, and the ratio of longitudinal propagation characteristics to transverse propagation characteristics were respectively processed to be dimensionless. A lookup table mapping, piecewise linear mapping, or multiple regression mapping is established using the calibration results of the direction of liquid absorption, thermal desiccation, and swelling of the base paper; and the quality identification mapping is recursively calibrated based on the degree of glue line formation, moisture content of the exported paperboard, and paperboard warpage index.
[0011] Preferably, the adhesive parameters include starch gel content and gel gap, wherein the starch gel content is 22% to 28% and the gel gap is 0.10 to 0.25 mm; the double-sided machine hot pressing parameters include hot pressing temperature and hot pressing residence time, and the glue line temperature is kept within the gelatinization temperature window of 62 to 85 ℃.
[0012] Preferably, the common quality boundary includes: the moisture content of the exported paperboard is 6% to 9%, the glue line formation degree is within the formation window calibrated by the random inspection of the adhesive strength, and the paperboard warpage index does not exceed the preset warpage threshold; when any index approaches the corresponding boundary, the execution amount correction that will cause the index to continue to approach the boundary is limited, and the difference in execution amount between adjacent sections of the banner does not exceed 10% to 20% of the full range of the corresponding execution amount.
[0013] Preferably, a sliding detection window is set along the paper feeding direction. When the quality vector in 3 to 8 consecutive sliding detection windows shifts in the same direction as the previously stable quality vector and the shift exceeds the change point threshold, it is determined that a roll change point or an in-roll quality change point has occurred.
[0014] Preferably, after determining that the roll change point or the quality change point within the roll has occurred, the starch solid content and glue gap are first adjusted according to the glue absorption demand index, and then the preheating parameters are adjusted according to the heat desiccation index. When the adjustment of the glue application parameters and the preheating parameters within the common quality boundary still cannot make the quality vector enter the stable range, the production line speed is adjusted, and the rate of change of the glue application parameters, the preheating parameters and the production line speed between adjacent sliding detection windows is limited respectively.
[0015] As a preferred approach, a paper segment delay queue is established based on the station distances between the inlet detection position, preheating position, gluing position, hot pressing position, and outlet detection position, as well as the real-time production line speed. Using the paper segment identifier as the key, the inlet reference signal, quality vector, preheating history, gluing history, hot pressing history, glue line formation degree, outlet paperboard moisture content, and paperboard warpage index of the same paper segment are bound together. The quality identification mapping and process parameters corresponding to the paper segment to be calibrated are calibrated only with the outlet data that has the same paper segment identifier as the paper segment to be calibrated.
[0016] Preferably, the recycled fiber corrugated board is a double-walled corrugated board. The gluing degree of the first gluing line and the second gluing line are obtained respectively, and the two gluing lines are kept to have independent forming windows. Based on the gluing degree of the two gluing lines, the moisture content of the same export board and the board warping index, the corresponding gluing parameters, preheating parameters and double-facer hot pressing parameters are jointly corrected.
[0017] Compared with the prior art, the present invention has the following advantages: The present invention discloses an adaptive manufacturing method for recycled fiber corrugated board based on online identification of base paper quality. By applying controlled preheating excitation to the paper segment to be identified, the dynamic response of the base paper is obtained online and the quality vector is identified. The identification results are then used to adjust the corresponding parameters of gluing, preheating and hot pressing. At the same time, the glue line formation degree, exit moisture content and board warpage index are used for joint constraints to form a control process from base paper quality identification, manufacturing parameter allocation to exit quality correction, thereby improving the adaptability of the recycled fiber corrugated board manufacturing process to fluctuations in base paper quality. This invention, by collecting the moisture content, tension, and longitudinal and transverse propagation characteristics of the base paper, and combining the dynamic response before and after controlled preheating excitation, can characterize the glue absorption, desiccation, and swelling characteristics of the base paper that are difficult to reflect by static detection values. This invention converts the dynamic response of the base paper into the glue absorption demand index, the heat desiccation index, and the moisture swelling asymmetry index, decomposes the differences in base paper quality into the corresponding manufacturing action direction, and improves the interpretability of the quality identification results. This invention determines the gluing parameters, the preheating parameters of the face paper and the liner paper, and the hot pressing parameters of the double-sided machine based on three types of quality indices, so that different execution quantities can compensate for corresponding quality deviations and reduce the over-adjustment of multiple parameters at the same time. This invention uses glue line formation degree, exit paperboard moisture content and paperboard warpage index to form a common quality boundary, which can suppress excessive drying and warpage while ensuring bonding quality, achieve coordinated correction of manufacturing parameters, and improve the stability of continuous production. Attached Figure Description
[0018] Figure 1 This is a flowchart of an adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of raw paper quality according to the present invention; Figure 2 This is a diagram showing the distribution relationship between the three-component quality vector and the amounts of gluing, preheating, and hot pressing in this invention. Figure 3 This is a schematic diagram of the paper segment delayed binding and quality change point sequence switching in this invention. Detailed Implementation
[0019] The present invention will be further described below with reference to the accompanying drawings and embodiments. It should be understood that the implementation of the present invention is not limited to the embodiments described below, and any modifications and / or alterations made to the present invention will fall within the scope of protection of the present invention. Unless otherwise stated, all percentages are mass percentages, and changes in moisture content are expressed in percentage points.
[0020] like Figure 1 As shown, an embodiment of the present invention discloses an adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of the quality of the original paper. The method includes the following steps: Before the face paper and liner paper enter the preheater, the inlet reference signal of the corresponding paper is obtained. The inlet reference signal includes at least the moisture content signal, tension signal, and longitudinal and transverse propagation characteristic signals. The raw paper is divided into paper segments along the conveying direction. Within the preset manufacturing quality boundary, a controlled preheating excitation is applied to the paper segment to be identified through a preheater. At the same time, the dynamic relationship between the dynamic response of the moisture content, the dynamic response of the tension, and the longitudinal and transverse propagation characteristics of the paper segment to be identified before and after the controlled preheating excitation is obtained. Based on the dynamic response of moisture content, the dynamic response of tension, and the dynamic relationship, the quality vector of the paper segment to be identified is obtained according to the pre-calibrated quality identification mapping. The quality vector includes the glue absorption demand index, the heat desiccation index, and the moisture expansion asymmetry index. The glue application parameters are determined based on the glue absorption demand index, the preheating parameters of the face paper and liner paper are determined based on the heat desiccation index, and the hot pressing parameters of the double-sided machine are determined based on the moisture expansion asymmetry index, so that the deviation of each quality index is compensated by the execution amount in the corresponding process action direction. According to the determined gluing parameters, preheating parameters, and double-facer hot pressing parameters, the base paper is sequentially corrugated, sized, laminated, and hot-pressed. The glue line formation degree, exit paperboard moisture content, and paperboard warpage index are obtained online. Within the common quality boundary of the glue line formation degree, exit paperboard moisture content, and paperboard warpage index, the gluing parameters, preheating parameters, and double-facer hot pressing parameters are corrected.
[0021] This embodiment employs a continuous corrugated board production line sequentially equipped with a base paper rack, an inlet inspection station, a preheater, a single-facer, a gluing unit, a laminating station, a double-facer hot-pressing section, and an outlet inspection station. At least the face paper and liner paper contain recycled fibers. The basis weight of the base paper is 90–250 g / m². 2 The moisture content before entering the preheater is 5% to 11%, and the production line speed is 80 to 250 m / min.
[0022] The entry inspection station simultaneously acquires moisture content, tension, longitudinal propagation characteristics, and transverse propagation characteristics on each sheet of raw paper. Moisture content is obtained using a horizontal scanning microwave moisture sensor, or alternatively, a near-infrared moisture sensor; basis weight is obtained as an auxiliary input using a radiation-type or optical basis weight sensor. Tension is obtained from tension sensors at both ends of the guide rollers. Longitudinal and transverse propagation characteristics are acquired by two sets of non-contact ultrasonic transmitting and receiving units. The measurement directions of the two sets of units are along the longitudinal and transverse directions of the raw paper, respectively, outputting propagation time, attenuation, or the equivalent stiffness calculated from them.
[0023] The controller divides the continuous paper web into multiple segments based on the real-time linear velocity. The duration of each segment is the same as the excitation duration, or it can be an integer multiple of the excitation duration; each segment is assigned a unique segment identifier. All sensors sample using a unified clock, and the sampling records include at least the segment identifier, sampling time, linear velocity, moisture content, tension, longitudinal propagation characteristics, and lateral propagation characteristics. To eliminate the influence of local fluctuations in the paper web, multiple detection zones are divided along the web, with each zone forming a separate signal sequence. The average web width and zone deviation are then calculated.
[0024] A first moisture detection point is set upstream of the preheater, and a second moisture detection point is set downstream of the preheater. Tension and propagation characteristics can be measured either upstream and downstream of the preheater separately, or continuously upstream, and the response segments before and after excitation can be extracted using the paper segment propagation time. The distance from each measurement point to the preheater's operating area is pre-registered, and the controller corrects the sampling window based on the real-time linear velocity to ensure that the same set of before and after signals belongs to the same paper segment.
[0025] like Figure 2 As shown, dynamic moisture content and tension response are mainly included in the adhesive absorption demand index, moisture content change rate and recovery response are mainly included in the thermal desiccation index, and longitudinal and transverse stiffness relationship and interlayer difference are mainly included in the wet expansion asymmetry index. The three indices correspond to the adhesive application, preheating and hot pressing manufacturing actions, respectively.
[0026] For paper segments to be identified that are in a stable conveying state and have not yet approached the production quality boundary, the controller will increase the preheating wrap angle by 3° to 8° based on the current steady-state wrap angle, maintain it for 0.5 to 2.0 s, and then restore it. Alternatively, the controller can briefly increase the steam volume or radiant heat flux while keeping the wrap angle constant, so that its thermal effect is equivalent to the aforementioned wrap angle step. The excitation intensity is characterized by the change in moisture content at the second moisture detection position relative to the first moisture detection position, and this change is controlled within 0.15 to 0.60 percentage points. If the inlet moisture content, surface temperature, or tension of the paper segment is close to the set manufacturing quality boundary, the excitation is reduced or the paper segment is skipped to avoid the identification action itself causing excessive dryness, excessive wetness, or abnormal tension.
[0027] For each paper segment to be identified, a reference window, an excitation response window, and a recovery window are set. The reference window is used to obtain the steady-state moisture content and steady-state tension; the response window is used to calculate the rate of change of moisture content and tension over time; the recovery window is used to confirm that the signal returns to the allowable fluctuation range. The rate of change can be obtained by the ratio of the difference between the first and last values within the window to the time difference, or by fitting a straight line to the sampling points within the window and taking the slope. The longitudinal and transverse propagation relationship is expressed by the ratio of longitudinal propagation velocity to transverse propagation velocity, the ratio of longitudinal equivalent stiffness to transverse equivalent stiffness, or a combination of both.
[0028] The rate of change in moisture content, the rate of change in tension, and the longitudinal and transverse propagation relationships are dimensionless relative to the center value and allowable range of the accepted reference paper, respectively, to obtain the moisture response component, tension response component, and anisotropic component. The controller then obtains the adhesive absorption demand index, the thermal desiccation index, and the moisture expansion asymmetry index through a pre-calibrated quality identification mapping. All three indices can be normalized to values between 0 and 1; an increase in the index indicates a greater degree of compensation required in the corresponding manufacturing action direction.
[0029] The adhesive absorption demand index is mainly determined by the moisture response component, the tension response component, and their recovery time. When the moisture content of the base paper changes slowly under the same heat, the tension recovery is slow, and the adhesive penetration is insufficient as indicated by the adhesive line calibration, the adhesive absorption demand index increases. The heat desiccation index is mainly determined by the rate of change in moisture content and the degree of moisture recovery within the recovery window; when the moisture content of the base paper decreases rapidly after heating, the heat desiccation index increases, which is used to limit subsequent heat. The moisture swelling asymmetry index is jointly determined by the longitudinal and transverse propagation relationship of the face paper and the liner paper, the tension response difference, and the cross-sectional deviation; the greater the difference between the two layers of base paper or different zones, the greater the moisture swelling asymmetry index.
[0030] The quality identification mapping employs a calibration lookup table. Specifically, the three dimensionless components are divided into several intervals. During the initial calibration production, the optimal glue amount, preheating heat, hot pressing amount, and exit quality for each interval are recorded, thereby establishing a three-dimensional lookup table. The quality identification mapping can also use a piecewise linear relationship or a multiple regression relationship. Regardless of the form used, the input to the mapping remains the dynamic response of the current paper segment, and the output remains the quality index in the three manufacturing action directions, all of which are subsequently trimmed through a common quality boundary.
[0031] The adhesive absorption demand index is used to determine the adhesive application parameters. These parameters include starch solids content, adhesive gap, and adhesive application rate per unit width. The starch solids content is set at 22%–28%, and the adhesive gap at 0.10–0.25 mm. When the adhesive absorption demand index increases and the adhesive line formation degree is below the lower limit of the formation window, the adhesive application rate or adhesive solids content can be increased within this range. If the starch absorption characteristics on the adhesive line surface indicate adhesive accumulation, the combination of solids content and adhesive gap should be adjusted first, rather than continuing to increase the adhesive application rate unilaterally.
[0032] The thermal desiccation index is used to determine the preheating parameters for the face paper and liner paper. These parameters include the preheating wrap angle, steam volume, and heat flux. A higher thermal desiccation index indicates faster moisture migration in the base paper under a given heat input; the controller then reduces the steady-state preheating temperature of the corresponding paper layer or shortens the duration of high heat. Conversely, if the thermal desiccation index is low and the glue line temperature is insufficient, the preheating temperature can be increased. The excitation wrap angle is only briefly superimposed on the steady-state wrap angle within the identification window; after identification, the system operates according to the newly obtained steady-state preheating parameters.
[0033] The moisture swelling asymmetry index is used to determine the hot-pressing parameters of the duplexer and the preheating difference between the face paper and the liner paper. The hot-pressing parameters include hot-pressing temperature, pressure, and hot-pressing residence time. The controller adjusts the residence time by regulating the effective length of the hot plate, the ballast contact area, or the linear speed, and keeps the glue line temperature within the gelatinization temperature window of 62–85 °C. When the face paper has a higher tendency to swell relative to the liner paper, the controller limits additional heat dehumidification of the face paper and adjusts the preheating or hot-pressing residence time of the liner paper to make the moisture release after exiting the board more consistent.
[0034] The three quality indices and three types of execution quantities are controlled in a primary correspondence rather than completely independently. The adhesive absorption demand index is mainly based on the glue application channel, the heat desiccation index is mainly based on the preheating channel, and the moisture expansion asymmetry index is mainly based on the interlayer preheating difference and hot pressing channel; the cross-effects generated by other channels are handled by common quality boundaries. In this way, it is avoided to significantly increase the amount of adhesive, preheating heat, and residence time simultaneously when the adhesion is insufficient.
[0035] After adhesive application, at least one of the following parameters is measured near the adhesive line: adhesive line temperature, adhesive line moisture content, and starch absorption characteristics, to determine the adhesive line formation degree. Adhesive line temperature can be obtained by infrared thermometry or thermal imaging; adhesive line moisture content is obtained by near-infrared absorption; starch absorption characteristics are obtained by the absorption difference between the upstream and downstream adhesive films of the adhesive roller or by the reflected absorption signal of the adhesive line. The adhesive line formation degree is calibrated using periodically sampled dry bond strength and strength retention rate after wet heat conditioning to obtain the upper and lower limits of the formation window.
[0036] The export inspection station obtains the moisture content and warpage index of the exported paperboard. The moisture content of the exported paperboard is controlled between 6% and 9%. The warpage index is represented by the maximum height difference, curvature, or edge warping amount measured by the horizontal laser displacement array, and compared with the warpage threshold of the corresponding paperboard specification. The controller combines the glue line formation degree, the moisture content of the exported paperboard, and the warpage index into a combined quality status.
[0037] If the glue line formation degree is below the lower limit while the outlet moisture content is within the allowable range, prioritize adjusting the glue gap or solid content according to the glue absorption demand index. If the glue line formation degree is below the lower limit and the outlet moisture content is close to the lower limit, it is prohibited to further increase the preheating heat and hot pressing residence time. If the glue line formation degree is qualified and the outlet moisture content is close to the lower limit, reduce the preheating or hot pressing amount. If the moisture content is qualified and the warpage index is close to the threshold, adjust the preheating difference between the face paper and the liner paper or the heat of the horizontal section according to the wet expansion asymmetry index. The difference in the execution amount between adjacent sections of the horizontal section is limited to 10% to 20% of the corresponding execution amount's full range to avoid local thermal stress or transverse waviness.
[0038] After each calibration, the controller retains the original quality vector, parameters before calibration, parameters after calibration, and the joint quality state. Only when all three quality indices are within limits and the calibration direction aligns with the primary correspondence of the quality indices is this record used to update the quality identification mapping. This prevents a single export indicator from erroneously reinforcing the mapping.
[0039] like Figure 3 As shown, the paper segment propagates along each station: inlet detection, preheating excitation, gluing and lamination, hot pressing, and outlet detection. After the change point is confirmed, the parameters are switched in the order of gluing, preheating, and line speed.
[0040] The controller sets a sliding detection window for the quality vector along the conveying direction. The three-component quality vector of the current window is compared with the center vector of a previously stable window. When at least one component shifts in the same direction across 3 to 8 consecutive windows and the vector distance exceeds the corresponding change point threshold, a roll change point or an in-roll quality change point is confirmed. If the splicer provides a roll change mark, the mark is used to narrow down the change point search range, but the quality of the new paper segment is ultimately confirmed by the dynamic response.
[0041] After confirming the change point, first adjust the starch solids content and glue gap to bring the glue line formation back to the formation window; then adjust the preheating of the face paper and liner paper to bring the exit moisture content back to 6%–9%; when gluing and preheating have reached a common quality boundary but the glue line or moisture content is still unstable, then adjust the production line speed and the hot pressing residence time of the double-sided machine. Each execution channel has a maximum change rate for adjacent windows, and the subsequent channel starts after the previous channel has reached stability or the boundary. This sequence allows the glue supply, paper moisture, and conveying residence time to transition according to different time scales, reducing parameter oscillations at roll changes.
[0042] The conveying distances between the inlet detection position, preheating position, gluing position, hot pressing position, and outlet detection position are pre-stored in the controller. For any two stations, the paper propagation delay is obtained by dividing the distance between the two stations by the real-time linear velocity of the corresponding time period. When the linear velocity changes, the controller can integrate the encoder pulses and use the moment when the cumulative conveying length reaches the station distance as the paper arrival time.
[0043] Each record in the paper segment delay queue includes a paper segment identifier, inlet reference signal, quality vector, preheating history, gluing history, hot pressing history, glue line formation degree, outlet moisture content, and warpage index. When a paper segment passes through a station, the corresponding station only appends a record to the paper segment identifier. After the outlet quality arrives, the controller retrieves the inlet quality vector and execution history of the same paper segment identifier, calculates the impact of this execution on the quality status, and only updates the identification mapping of the quality interval to which the paper segment belongs.
[0044] For example, if the distance between the inlet and outlet inspection stations is 120 m, and the linear velocity is 120 m / min, the nominal propagation delay is 60 s. If the linear velocity drops to 100 m / min within 20 s after the roll change, the controller does not use the fixed 60 s delay, but instead accumulates the actual conveying length of each sampling cycle to 120 m, and then assigns the outlet data to the corresponding roll change segment. This avoids using the outlet feedback of the old roll for the new roll.
[0045] For single-wall corrugated board, quality vectors are established for the face paper and liner paper respectively, coordinating gluing, preheating, and hot pressing under the same glue line formation window and the same exit quality condition. For double-wall corrugated board, each paper layer entering the first and second laminating stations is identified separately, and independent glue line formation windows are maintained for the first and second glue lines. The first glue line correction preferentially acts on the corresponding facer and the first glue application channel, while the second glue line correction preferentially acts on the second glue application channel; both glue lines are subject to the final exit moisture content and warpage index.
[0046] The effectiveness of the method of the present invention was verified through the following examples and comparative examples. All groups used the same continuous corrugated cardboard production line, selecting a basis weight of 90–250 g / m². 2 The recycled fiber face paper, liner paper, and core paper, with an inlet moisture content of 5%–11%, use the same corrugated type, starch glue raw material, paper clamping method, environmental conditioning conditions, and testing methods. The starch glue solids content is controlled at 22%–28%, the glue gap at 0.10–0.25 mm, the glue line temperature at 62–85 ℃, the production line speed at 80–250 m / min, and the outlet paperboard moisture content at 6%–9%. Except for the control functions or incentive ranges explicitly listed for each group, all other paper specifications, glue batches, equipment, and manufacturing conditions remain consistent.
[0047] Each group was prepared according to the following common steps: the face paper, liner paper, and core paper were loaded into the paper rack and conveyed stably; the core paper was preheated and formed into corrugated core paper by corrugating rollers; starch adhesive was applied to the top of the corrugations of the corrugated core paper; the face paper, corrugated core paper, and liner paper were laminated sequentially; the composite paperboard was passed through the hot pressing section of a double-facer to complete gelatinization and shaping; the moisture content and warpage of the paperboard were tested at the exit, and samples were obtained at a uniform sampling location. Each group had 5 independent operating batches, and the dry bond strength, strength retention rate after wet heat conditioning, and warpage or delamination defect rate of each batch were tested.
[0048] Comparative Example 1
[0049] Comparative Example 1 was manufactured using a fixed formula and fixed heat. The inlet inspection station only collected static moisture content and tension for production monitoring; no controlled preheating was applied to the specified paper segments, and no glue absorption demand index, heat desiccation index, or moisture expansion asymmetry index were calculated, nor was joint quality feedback established. Before production began, the starch glue content, glue gap, preheating heat of the face and linerboards, double-facer hot-pressing conditions, and production line speed were set once according to the conventional stable production conditions for this specification of paperboard; during continuous manufacturing, these settings were maintained and did not adaptively change with variations in the quality of the inlet base paper. Corrugation forming, sizing, lamination, hot pressing, and sampling were completed according to the common preparation steps.
[0050] Comparative Example 2
[0051] Comparative Example 2 uses inlet moisture, tension, and longitudinal and transverse propagation characteristic detection, and applies a short-term preheating excitation of 0.5–2.0 s to a specified paper segment. However, instead of calculating the three-component quality vector based on the multi-channel dynamic response before and after excitation, the static moisture content measured before excitation is directly input into a pre-calibrated univariate mapping. The controller adjusts the glue gap or solids content, the preheating heat of the face paper and liner paper, and the hot pressing amount of the duplexer based on this static moisture content mapping, and limits the adjustment range using glue line formation degree, outlet moisture content, and warpage index. The remaining operations are the same as the common preparation steps, and continuous manufacturing and sampling are carried out after the parameters stabilize.
[0052] Comparative Example 3
[0053] Comparative Example 3 applies controlled preheating excitation to a specified paper segment, collecting data on moisture content, tension, and longitudinal and transverse propagation responses before and after excitation. It identifies the glue absorption demand index, heat desiccation index, and moisture expansion asymmetry index, adjusting glue application, preheating, and double-sided machine hot-pressing parameters accordingly. Simultaneously, it detects quality vector offset over 3-8 consecutive windows, adjusting glue application, preheating, and line speed sequentially after confirming quality change points. This group does not implement paper segment time-delay binding; the exit glue line formation degree, exit moisture content, and warpage results are directly fed back based on the controller's current quality status, without tracing back to the same paper segment identifier based on station distance and real-time line speed. The paper is manufactured and sampled according to the above configuration.
[0054] Example 1
[0055] Example 1: Moisture content, tension, and longitudinal and transverse propagation characteristics are simultaneously obtained at the inlet detection station. For designated paper segments that meet safety boundaries, a preheating wrap angle of 3°–8° or an equivalent thermal pulse is applied and maintained for 0.5–2.0 s, ensuring the moisture content change before and after excitation is between 0.15 and 0.60 percentage points. The controller extracts the moisture content change rate, tension change rate, and longitudinal and transverse propagation relationship from the reference window, response window, and recovery window, calculating the glue absorption demand index, heat desiccation index, and moisture expansion asymmetry index. The glue absorption demand index is used to determine the starch glue content and glue gap; the heat desiccation index is used to determine the preheating heat of the face paper and liner paper; and the moisture expansion asymmetry index is used to determine the hot pressing amount of the double-sided machine and the interlayer preheating difference. During manufacturing, the glue line formation degree, outlet moisture content, and warpage index constitute a common quality boundary, limiting and correcting the execution quantity. This group does not use quality change point sequence switching or paper segment time delay queues, completing manufacturing and sampling within the stable quality range of the base paper.
[0056] Example 2
[0057] Example 2 adds a quality change point sequence switching function to Example 1. The controller compares the three-component quality vector of the current sliding window with that of the previously stable window; when at least one component in 3 to 8 consecutive windows shifts in the same direction and the vector distance exceeds a threshold, a roll change point or an in-roll quality change point is confirmed. After confirmation, the starch sizing content and sizing gap are adjusted first, and the preheating of the face paper and liner paper is adjusted after the glue line formation stabilizes or reaches the boundary; the production line speed and hot pressing dwell time are adjusted only when the glue application and preheating reach the boundary but the quality status is still unstable. The maximum change rate of adjacent windows is set for each channel. The paper segment delay queue is not enabled in this group, and manufacturing and sampling are carried out according to the above configuration.
[0058] Example 3
[0059] Example 3 adds paper segment delay binding to Example 2. The controller divides paper segments according to a unified clock and real-time linear velocity and assigns segment identifiers. The inlet reference signal, three-component quality vector, and preheating, gluing, and hot-pressing histories are written into the paper segment record. The paper segment propagation delay is determined by dividing the station distance by the real-time linear velocity; when the linear velocity changes, the arrival time is determined by the cumulative encoder pulse delivery length. Glue line formation, outlet moisture content, and warpage results are only bound to the inlet quality and execution history of the same paper segment identifier, and only the identification mapping of the quality range to which the paper segment belongs is updated. Manufacturing and sampling are performed according to the above configuration.
[0060] Example 4
[0061] Example 4 was prepared following the entry detection, three-component quality vector identification, manufacturing parameter allocation, and common quality constraint steps of Example 1, without enabling quality change point sequence switching and paper segment delay queues; its controlled preheating excitation caused the moisture content of a specified paper segment to change by 0.3 percentage points before and after excitation. A baseline, response, and recovery window were set for each paper segment, and three quality indices were calculated using the moisture content change rate, tension change rate, and longitudinal and transverse propagation relationship, and the gluing, preheating, and hot-pressing parameters were allocated accordingly. The paper was manufactured and sampled according to this excitation amplitude.
[0062] Comparative Example 4
[0063] Comparative Example 4 was prepared following the same steps as Example 1: inlet detection, three-component quality vector identification, manufacturing parameter allocation, and common quality constraints, without enabling quality change point sequence switching and paper segment delay queues. The only difference was that preheating excitation caused a 0.1 percentage point change in moisture content before and after excitation of a specified paper segment. A baseline, response, and recovery window were still set for each paper segment, and the resulting dynamic response was used to calculate three quality indices. The paper was manufactured and sampled according to this excitation amplitude.
[0064] Comparative Example 5
[0065] Comparative Example 5 was prepared following the same steps as in Example 1: inlet detection, three-component quality vector identification, manufacturing parameter allocation, and common quality constraint, without enabling quality change point sequence switching and paper segment delay queues. The only difference was that the preheating excitation caused a 0.8 percentage point change in moisture content before and after the excitation of a specified paper segment. After the excitation ended, the steady-state preheating setting was restored, and manufacturing parameters were then allocated based on the dynamic response. The paper was manufactured and sampled according to this excitation amplitude.
[0066] Table 1 summarizes the dry bond strength, strength retention rate after wet heat conditioning, and warpage or debonding defect rate for each group. Each group contains the mean and sample standard deviation of 5 independent records.
[0067] Table 1 Summary of main experimental results (mean ± sample standard deviation, n=5)
[0068] As shown in Table 1, the dry adhesive strengths of Comparative Example 1 were 375.44, 377.72, 380.00, 382.28, and 384.56 N / m, with a mean of 380.0 N / m and a sample standard deviation of 3.605 N / m. The dry adhesive strengths of Example 1 were 454.48, 457.24, 460.00, 462.76, and 465.52 N / m, with a mean of 460.0 N / m and a sample standard deviation of 4.364 N / m. The results indicate that, under the same evaluation boundary, using the dynamic response of the current paper segment to determine the adhesive, thermal, and pressing parameters can improve the bonding effect of the adhesive lines.
[0069] Comparative Example 2, lacking a three-component quality vector and directly using static moisture content mapping, had mean values of 410.0 N / m for dry bond strength, 80.0% for strength retention after wet heat conditioning, and 11.0% for warpage or debonding defects. Comparative Example 3, while featuring sequential switching of quality change points, eliminated paper segment delay binding, resulting in mean values of 465.0 N / m, 85.0%, and 8.0%, respectively. These results demonstrate that static moisture content cannot replace dynamic quality decomposition, and incorrectly assigned exit feedback weakens the closed-loop control effect.
[0070] The strength retention rates after wet heat conditioning in Example 1 were 85.226%, 85.742%, 86.000%, 86.258%, and 86.774%, with a mean of 86.0% and a sample standard deviation of 0.577 percentage points. The corresponding results for Example 2 were 89.190%, 89.730%, 90.000%, 90.270%, and 90.810%, with a mean of 90.0% and a sample standard deviation of 0.604 percentage points. The sequential switching in Example 2 reduced the transition deviation caused by simultaneous changes in adhesive amount, heat, and speed during roll changes and changes in roll quality.
[0071] The warpage or delamination defect rates for Example 2 were 4.88%, 4.94%, 5.00%, 5.06%, and 5.12%, respectively, with a mean of 5.0% and a sample standard deviation of 0.095 percentage points. The corresponding results for Example 3 were 2.88%, 2.94%, 3.00%, 3.06%, and 3.12%, respectively, with a mean of 3.0% and a sample standard deviation of 0.095 percentage points. Paper segment delay binding ensures consistency between exit feedback and entry quality and execution history, reducing warpage and glue line failure caused by feedback mismatch.
[0072] The excitation moisture content changes for Comparative Example 4, Example 4, and Comparative Example 5 were 0.1, 0.3, and 0.8 percentage points, respectively. The average dry bond strengths of the three were 440.0, 490.0, and 470.0 N / m, respectively; the average strength retention rates after wet heat conditioning were 84.0%, 91.0%, and 87.0%, respectively; and the average warpage or debonding defect rates were 9.0%, 4.0%, and 7.0%, respectively. The response amplitude of Comparative Example 4 was insufficient to stably distinguish paper segment quality; the excitation amplitude of Comparative Example 5 was too large, compressing the common quality boundary margin; Example 4 achieved better overall results, supporting a preferred moisture content change window of 0.15–0.60 percentage points.
[0073] In summary, controlled excitation, three-component quality vector, common quality constraints, sequential switching of quality change points, and paper segment delay binding collectively improve the continuous manufacturing stability of recycled fiber corrugated board.
Claims
1. An adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of raw paper quality, characterized in that: The method includes the following steps: Before the face paper and liner paper enter the preheater, the inlet reference signal of the corresponding paper is obtained. The inlet reference signal includes at least the moisture content signal, tension signal, and longitudinal and transverse propagation characteristic signals. The base paper is divided into paper segments along the conveying direction. Within the preset manufacturing quality boundary, the paper segment to be identified is subjected to controlled preheating excitation through the preheater. At the same time, the dynamic relationship between the dynamic response of the moisture content, the dynamic response of the tension, and the longitudinal and transverse propagation characteristics of the paper segment to be identified before and after the controlled preheating excitation is obtained. Based on the dynamic response of moisture content, the dynamic response of tension, and the dynamic relationship, the quality vector of the paper segment to be identified is obtained according to the pre-calibrated quality identification mapping. The quality vector includes the glue absorption demand index, the heat desiccation index, and the moisture expansion asymmetry index. The glue application parameters are determined based on the glue absorption demand index, the preheating parameters of the face paper base paper and the liner paper base paper are determined based on the heat desiccation index, and the hot pressing parameters of the double-sided machine are determined based on the moisture expansion asymmetry index, so that the deviation of each quality index is compensated by the execution amount in the corresponding process action direction. According to the determined gluing parameters, preheating parameters, and double-facer hot-pressing parameters, the base paper is sequentially corrugated, sizing, laminating, and hot-pressed. The glue line formation degree, exit paperboard moisture content, and paperboard warpage index are obtained online. Within the common quality boundary of the glue line formation degree, the exit paperboard moisture content, and the paperboard warpage index, the gluing parameters, the preheating parameters, and the double-facer hot-pressing parameters are corrected.
2. The adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of raw paper quality according to claim 1, characterized in that: The moisture content signal is acquired by a microwave moisture sensor or a near-infrared moisture sensor; the longitudinal and transverse propagation characteristic signals are acquired by ultrasonic sensors arranged along the longitudinal and transverse directions of the base paper; the inlet reference signal also includes the basis weight signal of the base paper.
3. The adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of raw paper quality according to claim 1, characterized in that: The controlled preheating excitation is a step excitation that increases the preheating wrap angle by 3° to 8° or a heat excitation that has the same thermal effect as the step excitation. The controlled preheating excitation is applied to the base paper for 0.5 to 2.0 s, and the change in the moisture content of the base paper before and after the excitation is 0.15 to 0.60 percentage points.
4. The adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of base paper quality according to claim 1, characterized in that: The quality identification mapping is pre-calibrated through the following steps: The rate of change of dynamic response of moisture content, the rate of change of dynamic response of tension, and the ratio of longitudinal propagation characteristics to transverse propagation characteristics were respectively processed to be dimensionless. A lookup table mapping, piecewise linear mapping, or multiple regression mapping is established using the calibration results of the direction of liquid absorption, thermal desiccation, and swelling of the base paper; and the quality identification mapping is recursively calibrated based on the degree of glue line formation, moisture content of the exported paperboard, and paperboard warpage index.
5. The adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of base paper quality according to claim 1, characterized in that: The coating parameters include starch gel content and gel gap, wherein the starch gel content is 22% to 28% and the gel gap is 0.10 to 0.25 mm; the double-sided machine hot pressing parameters include hot pressing temperature and hot pressing residence time, and the glue line temperature is kept within the gelatinization temperature window of 62 to 85 ℃.
6. The adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of base paper quality according to claim 1, characterized in that: The common quality boundaries include: the moisture content of the exported paperboard is 6% to 9%, the glue line formation degree is within the formation window calibrated by the random inspection of the adhesive strength, and the paperboard warpage index does not exceed the preset warpage threshold; when any index approaches the corresponding boundary, the execution amount correction that will cause the index to continue to approach the boundary is limited, and the difference in execution amount between adjacent sections of the banner does not exceed 10% to 20% of the full range of the corresponding execution amount.
7. The adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of base paper quality according to claim 1, characterized in that: A sliding detection window is set along the paper feeding direction. When the quality vector in 3 to 8 consecutive sliding detection windows shifts in the same direction as the previous stable quality vector and the shift exceeds the change point threshold, it is determined that a roll change point or an in-roll quality change point has occurred.
8. The adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of base paper quality according to claim 7, characterized in that: After determining that the roll change point or the quality change point within the roll has occurred, first adjust the starch solid content and glue gap according to the glue absorption demand index, then adjust the preheating parameters according to the heat dehumidification index. When the adjustment of the glue application parameters and preheating parameters within the common quality boundary still cannot make the quality vector enter the stable range, adjust the production line speed and limit the rate of change of the glue application parameters, preheating parameters and production line speed between adjacent sliding detection windows.
9. The adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of base paper quality according to claim 1, characterized in that: Based on the station distances between the inlet detection position, preheating position, gluing position, hot pressing position, and outlet detection position, as well as the real-time production line speed, a paper segment delay queue is established. Using the paper segment identifier as the key, the inlet reference signal, quality vector, preheating history, gluing history, hot pressing history, glue line formation degree, outlet paperboard moisture content, and paperboard warpage index of the same paper segment are bound together. The quality identification mapping and process parameters corresponding to the paper segment to be calibrated are calibrated only with the outlet data that has the same paper segment identifier as the paper segment to be calibrated.
10. The adaptive manufacturing method for recycled fiber corrugated cardboard based on online identification of base paper quality according to claim 1, characterized in that: The recycled fiber corrugated cardboard is a double-walled corrugated cardboard. The gluing degree of the first glue line and the second glue line are obtained respectively, and the two glue lines are kept to have independent forming windows. Based on the gluing degree of the two glue lines, the moisture content of the same export cardboard and the cardboard warpage index, the corresponding gluing parameters, preheating parameters and double-facer hot pressing parameters are jointly corrected.
