Packaging parameter monitoring method based on waste paper packaging

Abstract

The invention discloses a packaging parameter monitoring method based on waste paper packaging, and the method comprises the steps: executing a micro-rollback action after main compression is finished, obtaining the pressure and displacement changes in a rollback process in real time, quantifying the elastic compliance of waste paper, and accurately evaluating the rebound characteristic of a material. In combination with mechanical work, formed bundle length and incoming material moisture in the compression process, the stable density is estimated online by using a semi-physical fusion model, and accurate estimation of the density can be realized without an online weighing system. According to the method, by monitoring the density and the rebound trend in real time and automatically adjusting process parameters such as ribbon tension and pressure maintaining time, precise process optimization is achieved, excessive pressure maintaining and ribbon consumption are reduced, the production efficiency is improved, and energy consumption and auxiliary material consumption are reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of packaging, and in particular relates to a packaging parameter monitoring method based on waste paper packaging. Background Art

[0002] Waste paper baling is a critical process in the recycled paper industry. The compression and baling quality of waste paper directly impact subsequent transportation and processing efficiency. Traditional waste paper balers typically monitor the hydraulic system's pressure, stroke, and time during the compression process to determine whether the desired compaction effect has been achieved. However, existing monitoring methods have limitations, particularly when determining the density and stability of the resulting bale, as they often fail to fully account for material elastic rebound and moisture fluctuations. Existing technologies primarily rely on empirical experience or simple parameter thresholds (such as pressure or dwell time) to control the compression process. However, these methods are subject to significant variability in actual production, particularly for different paper types, whose rebound and compaction behaviors can vary significantly. Furthermore, traditional methods often overlook the impact of moisture on the density of waste paper during baling. Large moisture fluctuations can lead to skewed compression results. While some equipment incorporates online weighing systems to monitor density, these systems are costly and space-consuming, making them unsuitable for all production environments. Some plants have attempted to monitor motor power to reflect compression energy, but since hydraulic system efficiency is significantly affected by oil temperature, friction, and sealing conditions, relying solely on electrical energy statistics to assess compression effectiveness also results in significant inaccuracies.

[0003] Therefore, existing technologies generally lack accurate and real-time online monitoring methods, making it difficult to achieve precise control and optimization of the packaging process. Summary of the Invention

[0004] To solve the above problems, the present invention provides a packaging parameter monitoring method based on waste paper packaging. By real-time monitoring of density and rebound trend, it automatically adjusts process parameters such as tie tension and holding time, thereby achieving precise process optimization, reducing excessive holding pressure and tie consumption, and improving production efficiency.

[0005] The technical solutions provided by the present invention are as follows: A method for monitoring packaging parameters of waste paper packaging, comprising the following steps: S1: Perform multi-channel synchronous data acquisition during the baler compression process, and obtain at least the changes in compression pressure over time, the changes in the pressure head or cylinder position over time, the motor voltage and current or power, the length of the formed bale, and the moisture content of the incoming material; S2: During the main compression phase, the mechanical work done during the compression process is calculated based on the relationship between pressure and displacement. S3: When the compression reaches the set compaction level, the pressure head is controlled to retract with a small displacement, and the pressure change and corresponding displacement change during the retraction process are recorded, thereby obtaining the equivalent elastic compliance or equivalent stiffness of the current batch of materials; S4: Based on the equivalent elastic compliance or equivalent stiffness and in combination with the compressive state at the end of the main compression, estimate the recoverable elastic energy, and deduct the elastic energy from the total compression energy to obtain the irrecoverable plastic energy; S5: inputting the plastic energy, the length of the formed bale and the moisture content of the incoming material into a semi-physical fusion model to obtain an estimated value of the stable density of the formed bale; S6: Compare the estimated stable density value with the target threshold value to generate a monitoring result; when the requirements are not met, output an early warning and adjust at least one of the packaging process parameters, including the holding time, the number of pre-pressing times, the number of tie straps and / or the tie tension.

[0006] In some embodiments, in step S1, time reference calibration and band-limited denoising are performed before acquisition; and during the micro-backoff action in step S3, a high sampling mode is enabled to obtain the pressure-displacement tangent relationship in the backoff interval, thereby improving the estimation accuracy of the equivalent elastic compliance or equivalent stiffness.

[0007] In some embodiments, when the oil circuit state changes during the compression process, resulting in a change in the effective pressure area or force state, the main compression stage is divided into multiple sections, and the mechanical work done in each section is calculated and accumulated, and continuity correction is performed at the section boundaries.

[0008] In some embodiments, when stable pressure and displacement data cannot be obtained, the equivalent electrical energy is calculated as a substitute based on the change in motor voltage and current or power over time, and converted into equivalent mechanical work through model mapping and temperature compensation for subsequent calculation of elastic energy and plastic energy.

[0009] In some embodiments, in step S4, the estimation of elastic energy is inferred based on the approximate linear elastic relationship of the main compression end, and a correction coefficient is introduced to consider the effective pressure of the channel section and the influence of boundary friction. The correction coefficient is obtained by fitting the offline weighing data of a small number of calibration bundles.

[0010] In some embodiments, in step S5, the semi-physical fusion model is an interpretable model with plastic energy, formed bale length and incoming material moisture as main independent variables, and its parameters are initially fitted by real mass density data of a small number of calibration bales, and are fine-tuned and updated online during operation based on sparse random sampling.

[0011] In some embodiments, before the main compression, the pressure-displacement curve of the pre-compression stage is used to extract features including peak value, slope and segmented energy to form a fingerprint parameter set for distinguishing paper types and looseness, and the parameter group of the semi-physical fusion model is selected or interpolated based on this.

[0012] In some embodiments, step S5 further includes cable tie tension as an auxiliary variable. When the stable density estimate is lower than a threshold value and the equivalent elastic compliance indicates an increased rebound trend, priority is given to increasing the cable tie tension or increasing the number of cable tie paths. When the stable density meets the threshold value and the equivalent elastic compliance is abnormally high, a diagnostic prompt for slider sticking, hydraulic leakage, or friction abnormality is triggered.

[0013] In some embodiments, a parameter record including a timestamp, a fingerprint of the pre-pressing stage, an estimated value of stable density, incoming material moisture, plastic energy, and length of the formed bale is generated for each formed bale for quality traceability and process optimization.

[0014] In some embodiments, a sparse closed-loop calibration step is further included, in which a number of formed bundles are randomly selected for offline weighing in each shift or preset period, and the obtained true mass density is used to correct the semi-physical fusion model and correction coefficient to suppress long-term drift and improve estimation consistency.

[0015] In summary, the beneficial effects of the present invention are: (1) The present invention performs a micro-retraction action after the main compression process to obtain the elastic compliance during the compression process in real time, thereby quantitatively evaluating the rebound effect. This method overcomes the problem that traditional compression monitoring methods cannot accurately reflect material rebound by relying solely on pressure or stroke, and provides a more accurate basis for density estimation during the packaging process.

[0016] (2) By combining a semi-physical fusion model of plastic energy, bale length, and incoming material moisture, this method can accurately estimate the stable density of each bale of waste paper without relying on online weighing. This not only improves the accuracy of density estimation but also reduces the high cost and complexity of traditional weighing systems, making equipment modification more economical and efficient.

[0017] (3) By monitoring density and rebound trends in real time, the present invention enables precise process adjustments, such as automatically adjusting tie tension or hold time based on density estimates. This precise adjustment can effectively reduce inefficient consumption such as excessive tying and excessive hold time, thereby improving production efficiency and reducing energy consumption. It also reduces the consumption of auxiliary materials in the waste paper packaging process, helping enterprises achieve energy conservation and emission reduction goals. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0019] In order to deepen the understanding of the present invention, the present invention will be further described in detail below with reference to the examples. The following examples are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.

[0020] like Figure 1 As shown, a method for monitoring packaging parameters based on waste paper packaging is characterized by comprising the following steps: S1: Perform multi-channel synchronous data acquisition during the baler compression process, and obtain at least the changes in compression pressure over time, the changes in the pressure head or cylinder position over time, the motor voltage and current or power, the length of the formed bale, and the moisture content of the incoming material; S2: During the main compression phase, the mechanical work done during the compression process is calculated based on the relationship between pressure and displacement. S3: When the compression reaches the set compaction level, the pressure head is controlled to retract with a small displacement, and the pressure change and corresponding displacement change during the retraction process are recorded, thereby obtaining the equivalent elastic compliance or equivalent stiffness of the current batch of materials; S4: Based on the equivalent elastic compliance or equivalent stiffness and in combination with the compressive state at the end of the main compression, estimate the recoverable elastic energy, and deduct the elastic energy from the total compression energy to obtain the irrecoverable plastic energy; S5: inputting the plastic energy, the length of the formed bale and the moisture content of the incoming material into a semi-physical fusion model to obtain an estimated value of the stable density of the formed bale; S6: Compare the estimated stable density value with the target threshold value to generate a monitoring result; when the requirements are not met, output an early warning and adjust at least one of the packaging process parameters, including the holding time, the number of pre-pressing times, the number of tie straps and / or the tie tension.

[0021] Specifically, this embodiment is applied to horizontal or vertical hydraulic waste paper balers. The system includes: a pressure acquisition device connected in parallel to the main cylinder's high-pressure oil circuit; a displacement acquisition device located at the pressure head or cylinder; a voltage, current, or power acquisition device provided by a frequency converter or energy meter; a device for acquiring the length of the formed bales located at the discharge end; and an online moisture acquisition device located at the feed inlet or on the sidewall of the compression channel. These acquisition devices are electrically connected to an edge computing unit, which communicates with the baler controller via Ethernet or fieldbus, enabling simultaneous sampling, data processing, model inference, and process-linked control. To facilitate operation and traceability, the system can be equipped with a touchscreen terminal to display real-time curves, alarms, and reports.

[0022] Upon powering on the equipment and at the start of each shift, the system performs zero-point calibration and time base alignment. The pressure, displacement, electrical parameters, length, and moisture collection channels are timestamped, and a range and zero offset table is established. Band-limited low-pass filtering is then applied to pressure and displacement, sliding window averaging is used for electrical parameters, and a combination of median and short-window averaging is used for moisture. This eliminates packet loss, out-of-order data, and significant transition points, resulting in a clean data stream suitable for subsequent calculations. If sensor saturation or communication anomalies are detected, the system flags the data bundle as an anomaly and halts subsequent estimation, retaining only the original data for fault analysis.

[0023] During the main compression phase, the edge computing unit calculates the mechanical work done during the compression process based on the changing relationship between pressure and displacement. If the oil circuit state switching causes the stress state or effective pressure area to change, the system automatically divides the main compression process into several sections at the event point, calculates and accumulates each section separately, and performs continuity corrections at the section boundaries to ensure the physical consistency of the curve and energy accumulation. When encountering old models or fault conditions that make it difficult to obtain stable pressure or displacement data, the system calculates the equivalent electrical energy by integrating the voltage, current, or power over time, and converts it into equivalent mechanical work through pre-established model mapping and temperature compensation to ensure uninterrupted process.

[0024] When the compression reaches the set compaction level, the controller triggers a small-displacement retraction action. To improve measurement accuracy, the system switches to high-sampling mode during the retraction process, recording the pressure changes and corresponding displacement changes during the retraction process point by point. This interval data is then subjected to linear or local weighted regression to obtain the pressure-displacement tangent relationship in the retraction interval, and the equivalent elastic compliance or equivalent stiffness of the current batch of materials is calculated. If a sudden drop in pressure, excessive vibration, or a safety interlock is triggered during the retraction period, the system automatically abandons the retraction and marks it, attempting to resume with the next bundle to ensure production safety and consistency of action.

[0025] Based on the equivalent elastic compliance or stiffness derived from micro-retraction and the compressive state at the end of the main compression, the system estimates the recoverable elastic energy of the bundle. The irrecoverable plastic energy is then deducted from the total compression energy. To account for the non-idealities of channel compression and the effects of boundary friction, a correction factor is introduced into the elastic energy estimation. This factor is derived from offline weighing and density calculations of a small number of calibration bundles. After a one-time fitting, it is written into the machine parameters and updated in small steps based on subsequent random inspection samples to accommodate long-term drift in oil temperature, friction, and paper grade combinations.

[0026] The system feeds irreversible plastic energy, bale length, and incoming material moisture into a semi-physical fusion model, which outputs an estimate of the bale's stable density. This interpretable model uses initial parameters derived from the true mass density of a small number of calibration bales. During operation, as random inspection samples arrive, the system performs online fine-tuning and updates at a set learning rate to prevent sudden changes in parameters from impacting the estimated stability. The stable density is defined as the density at which the bale reaches mechanical stability after a short period of natural rebound after demolding and tying.

[0027] The system compares the estimated stable density with the target threshold to generate monitoring results. If the estimated value falls below the threshold and the equivalent elastic compliance obtained through micro-retraction indicates an increasing rebound trend, the system prioritizes issuing adjustment instructions to increase the cable tie tension or the number of cable tie passes. If the estimated value is low but the rebound trend is average, the system prioritizes extending the hold time or increasing the number of pre-presses. If necessary, the system recommends reducing the feed rate to improve filling uniformity. If the estimated value meets the threshold but the rebound trend is abnormally high, the system triggers a device status diagnostic prompt, suggesting inspections for leaks in the slider guide, seals, and oil lines.

[0028] Before the main compression phase, the system optionally extracts features such as peak value, slope, and segmented energy from the pre-compression pressure-displacement curve. This creates a fingerprint parameter set for distinguishing paper type and bulk. When a change in paper type or bulk is detected, the system uses this fingerprint to select or interpolate between multiple sets of model parameters, shortening model convergence time and improving estimation consistency across different paper types. This fingerprint serves only as auxiliary information for model self-adaptation and does not alter the basic process of micro-backoff self-calibration and energy separation.

[0029] When the equipment is equipped to collect tie tension, the system uses tie tension as an auxiliary variable in density assessment and process linkage. If the estimated stable density is low and the springback trend is increasing, the system prioritizes increasing tie tension or adding more tie passes before extending the hold pressure. This uses external constraints to suppress springback and improve molding stability. If the stable density meets the target but the springback trend remains high, the system prioritizes equipment status diagnosis over increasing tension to avoid excessive stretching that could cause material waste or equipment wear.

[0030] To enable quality traceability and process optimization, the system automatically generates parameter records for each formed bale, including timestamp, pre-pressing fingerprint, estimated stable density, incoming material moisture, irreversible plastic energy, bale length, whether electrical energy substitution is enabled, whether linkage adjustment is enabled and its specific items, exception labels, and operator confirmation information. These records can be queried by date, shift, paper type, and equipment on touch terminals or in higher-level systems, and can be exported as reports.

[0031] To mitigate long-term drift and improve estimation consistency, the system incorporates a sparse closed-loop calibration procedure. During each shift or a pre-set period, several randomly selected bundles are weighed offline to calculate the true mass density. This result is used to calibrate the semi-physical fusion model parameters, elastic energy correction coefficients, and electrical energy mapping coefficients. The calibration process utilizes small step updates and versioning, retaining historical parameters for easy retrospective review. Zero-point recalibration and sensor maintenance are also prompted when systematic deviations exceeding thresholds are detected.

[0032] System installation and modification follow a process of investigation, deployment, and commissioning. On-site inspections confirm the oil circuit parallel points, the displacement acquisition device installation location, the discharge end length measurement position, and the moisture meter window cleaning plan. After hardware installation, electromagnetic compatibility and protection checks are conducted. Communication and commissioning with the controller are conducted to define the priorities and interlocking conditions for trigger signals, sampling mode switching, and linkage commands. After no-load and loaded calibration, the system is first operated in monitoring mode to verify false alarms and missed alarms over a shift before enabling the automatic linkage strategy.

[0033] If pressure sensor inaccuracy, displacement encoder step loss, electrical parameter noise, or moisture meter contamination occur, the system switches to a conservative strategy based on anomaly detection rules: micro-rollback is suspended, density linkage is disabled, and only a manual inspection prompt is prompted. Once the anomaly is resolved and self-tests pass, the system automatically resumes micro-rollback and normal estimation to ensure production safety and data reliability.

[0034] Example 1 Online monitoring primarily utilizes pressure and displacement. During the main compression phase, the system calculates mechanical work based on pressure and displacement. At the end of the main compression phase, it performs a micro-retraction to obtain equivalent elastic compliance, estimates recoverable elastic energy, and separates the irrecoverable plastic energy. The plastic energy, bundle length, and incoming material moisture are input into a semi-physical fusion model to estimate the stable density. In one production batch, when the estimated value fell below a threshold and rebound trends intensified, the system automatically increased the cable tie tension and added another tie, while also recommending a longer hold time for the next bundle. Subsequent spot checks revealed that the density had returned to the target range and rebound was under control.

[0035] Example 2 Online monitoring using electrical energy substitution. On an older machine model, structural limitations made it difficult to obtain stable displacement data. The system then implemented an electrical parameter substitution scheme, calculating equivalent electrical energy by integrating voltage, current, or power over time. This was then converted to equivalent mechanical work based on machine model mapping and oil temperature compensation. Micro-backoff self-calibration and energy separation were then performed, completing stable density estimation and process linkage similar to Example 1. Operational results showed that after enabling sparse closed-loop calibration, the estimation error was at the same engineering level as Example 1, meeting production requirements.

[0036] Example 3 Model adaptation based on the "paper grade fingerprint" during the pre-pressing phase has been introduced. Under mixed material conditions, the system extracts the peak value, slope, and segmented energy from the pre-pressing curve to form a fingerprint parameter set. When the fingerprint changes, the system automatically interpolates across multiple sets of model parameters to quickly adapt to the new paper grade and bulk. Field tests have shown that when paper grades are frequently switched, adaptive adaptation can shorten model convergence time, reduce inter-batch estimation errors, and minimize ineffective process adjustments caused by misjudged density.

[0037] This implementation method can achieve online monitoring and closed-loop optimization of stable density and key process parameters without changing the main structure of the baler and without the need for online weighing. The sensors and gateways are mature industrial parts and can be transformed in a short time without downtime, making them suitable for upgrading existing equipment and supporting new lines.

[0038] The foregoing description shows and describes preferred embodiments of the present invention. As previously mentioned, it should be understood that the present invention is not limited to the form disclosed herein and should not be construed as excluding other embodiments. Instead, the present invention can be used in various other combinations, modifications, and environments and can be modified within the scope of the inventive concept described herein by the teachings above or by techniques or knowledge in the relevant art. Modifications and variations made by those skilled in the art without departing from the spirit and scope of the present invention are intended to be within the scope of the appended claims.

Claims

1. A method for monitoring packaging parameters based on waste paper packaging, characterized in that: The following steps are involved: S1: Perform multi-channel synchronous data acquisition during the baler compression process, and obtain at least the changes in compression pressure over time, the changes in the pressure head or cylinder position over time, the motor voltage and current or power, the length of the formed bale, and the moisture content of the incoming material; S2: During the main compression phase, the mechanical work done during the compression process is calculated based on the relationship between pressure and displacement. S3: When the compression reaches the set compaction level, the pressure head is controlled to retract with a small displacement, and the pressure change and corresponding displacement change during the retraction process are recorded, thereby obtaining the equivalent elastic compliance or equivalent stiffness of the current batch of materials; S4: Based on the equivalent elastic compliance or equivalent stiffness and in combination with the compressive state at the end of the main compression, estimate the recoverable elastic energy, and deduct the elastic energy from the total compression energy to obtain the irrecoverable plastic energy; S5: inputting the plastic energy, the length of the formed bale and the moisture content of the incoming material into a semi-physical fusion model to obtain an estimated value of the stable density of the formed bale; S6: Compare the estimated stable density value with the target threshold value to generate a monitoring result; when the requirements are not met, output an early warning and adjust at least one of the packaging process parameters, including the holding time, the number of pre-pressing times, the number of tie straps and / or the tie tension.

2. The method for monitoring packaging parameters based on waste paper packaging according to claim 1, characterized in that: In step S1, time reference calibration and band-limited denoising are performed before acquisition; during the micro-backoff action in step S3, a high sampling mode is enabled to obtain the pressure-displacement tangent relationship in the backoff interval, thereby improving the estimation accuracy of the equivalent elastic compliance or equivalent stiffness.

3. The method for monitoring packaging parameters based on waste paper packaging according to claim 1, characterized in that: When the oil circuit state changes during the compression process, resulting in a change in the effective pressure area or force state, the main compression stage is divided into multiple sections. The mechanical work done in each section is calculated and accumulated, and continuity correction is performed at the section boundaries.

4. The method for monitoring packaging parameters based on waste paper packaging according to claim 1, characterized in that: When stable pressure and displacement data cannot be obtained, the equivalent electrical energy is calculated as a substitute based on the change in motor voltage and current or power over time. After model mapping and temperature compensation, it is converted into equivalent mechanical work for subsequent calculation of elastic energy and plastic energy.

5. The method for monitoring packaging parameters based on waste paper packaging according to claim 1, characterized in that: In step S4, the elastic energy is estimated based on the approximate linear elastic relationship at the end of the main compression, and a correction coefficient is introduced to consider the effective pressure of the channel section and the influence of boundary friction. The correction coefficient is obtained by fitting the offline weighing data of a small number of calibration bundles.

6. The method for monitoring packaging parameters based on waste paper packaging according to claim 1, characterized in that: In step S5, the semi-physical fusion model is an interpretable model with plastic energy, formed bale length and incoming material moisture as the main independent variables. Its parameters are initially fitted by the real mass density data of a small number of calibration bales, and are fine-tuned and updated online during operation based on sparse random sampling.

7. The method for monitoring packaging parameters based on waste paper packaging according to claim 1, characterized in that: Before the main compression, the pressure-displacement curve of the pre-compression stage is used to extract features including peak value, slope and segmented energy to form a fingerprint parameter set for distinguishing paper types and looseness, and the parameter group of the semi-physical fusion model is selected or interpolated based on this.

8. The method for monitoring packaging parameters based on waste paper packaging according to claim 1, characterized in that: In step S5, the cable tie tension is also included as an auxiliary variable. When the stable density estimate is lower than the threshold and the equivalent elastic compliance indicates an increasing rebound trend, priority is given to increasing the cable tie tension or increasing the number of cable tie paths. When the stable density meets the threshold and the equivalent elastic compliance is abnormally high, a diagnostic prompt for slider sticking, hydraulic leakage or friction abnormality is triggered.

9. The method for monitoring packaging parameters based on waste paper packaging according to claim 1, characterized in that: It also includes the generation of parameter records for each formed bale, including timestamp, pre-pressing stage fingerprint, stable density estimation, incoming material moisture, plastic energy and formed bale length, for quality traceability and process optimization.

10. The method for monitoring packaging parameters based on waste paper packaging according to claim 1, characterized in that: It also includes setting up a sparse closed-loop calibration step, randomly selecting a number of formed bundles for offline weighing within each shift or preset period, and using the obtained true mass density to correct the semi-physical fusion model and correction coefficient to suppress long-term drift and improve estimation consistency.

Images