Pressure adaptive prediction and dynamic trim control system for industrial packaging processes

Abstract

The application relates to the technical field of industrial control systems, and discloses a pressure self-adaptive prediction and dynamic adjustment control system for an industrial packaging process, which comprises the following steps: an initial control parameter mapping model is established based on the physical characteristics of a controlled object; a feedback execution module collects displacement feedback and pressure data in real time; a transient state identification module performs differential operation on the displacement and the pressure; a transient contact stiffness model representing the evolution of the physical characteristics of an interface is constructed; a dynamic adjustment module injects a virtual damping factor according to the stiffness hardening rate; and a posture compensation module performs real-time deviation compensation on a posture vector. The application realizes predictive control by using transient stiffness identification, effectively suppresses stress overshoot, solves the workpiece damage problem caused by sudden changes in nonlinear stiffness, and ensures the reliability of the packaging process.

Description

Technical Field

[0001] This invention relates to a pressure adaptive prediction and dynamic adjustment control system for industrial packaging processes, belonging to the field of industrial control system technology. Background Technology

[0002] Industrial control systems are widely used in automated packaging, particularly in the sealing of aluminum caps for pharmaceutical containers. Conventional techniques employ preset stroke or pressure feedback control strategies, adjusting the actuator's output by collecting sensor signals and combining proportional-integral-derivative (PID) algorithms. However, the aluminum cap sealing process exhibits nonlinear stiffness abrupt changes. At the moment of sealing, the threshold for the aluminum cap to transition from elastic deformation to plastic flow is extremely narrow, and the physical consistency difference between the bottle body and the aluminum cap causes instantaneous fluctuations in contact stiffness. In high-speed response scenarios, there is a mismatch between the closed-loop response period of the control system and the evolution frequency of physical contact stress. Traditional pressure control schemes, limited by signal filtering delays and the physical inertia of the actuator, struggle to capture and suppress stress peaks at the microsecond-level contact instant. Furthermore, hysteresis-based compensation methods result in localized stress damage to the workpiece even when the system-displayed average pressure is at a safe threshold, ultimately leading to container breakage or seal failure. Conventional optimization methods, such as increasing sampling frequency or improving sensor accuracy, cannot eliminate the inherent physical hysteresis effect of the control system.

[0003] Conventional technologies mitigate packaging pressure fluctuations by optimizing mold shapes or introducing auxiliary buffer components and improving hardware. While hardware improvements can alleviate pressure fluctuations, the passive adaptation of the physical structure cannot solve the control lag problem in high-speed dynamic processes. For example, the utility model patent with authorization announcement number CN206834093U discloses a high-efficiency pressure control switch packaging mold, which sets a spring pressure head in the inner cavity of the upper mold and uses the physical elasticity of the spring to pre-fix the loose parts. Although the mold structure is improved to enhance the double-sided packaging efficiency, this pressure compensation based on physical elasticity is a static passive adjustment. The compensation force is limited by the inherent linear stiffness of the spring and cannot perform active real-time torque correction for nonlinear stiffness abrupt changes when the material's elastic deformation transforms into plastic flow. This solution lacks dynamic perception and compensation logic for the terminal posture deviation. When the physical consistency of the workpiece is poor or the conveying posture is offset, it is difficult to achieve predictive pressure closed-loop adjustment before the stress peak is formed, causing structural damage to the precision packaged parts.

[0004] Therefore, the technical problem to be solved by this invention is how to construct a predictive control mechanism to adjust the driving force before the stress peak is formed, while maintaining the packaging cycle time. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process, comprising:

[0006] The data acquisition module is used to acquire digital characterization data that represents the physical characteristics of the controlled object, and to establish a mapping relationship model between the physical characteristics of the controlled object and the initial control parameters based on the digital characterization data, thereby generating the target value of the sealing pressure and the corresponding initial pose command.

[0007] The feedback execution module is used to issue drive current according to the initial pose command and to collect the displacement feedback amount of the execution terminal and the real-time pressure data fed back by the pressure sensor in real time.

[0008] The transient identification module is used to perform differential operations on the displacement feedback and real-time pressure data, obtain the transient contact stiffness that characterizes the pressure evolution with displacement, construct a transient contact stiffness model, and identify the physical state migration trend of the sealing interface from elastic deformation to plastic flow based on the transient contact stiffness model.

[0009] The dynamic adjustment module is used to monitor the rate of change of transient contact stiffness over time to obtain the stiffness hardening rate. When the stiffness hardening rate reaches a preset threshold, a virtual damping factor is actively injected into the driving current to attenuate the execution gain of the driving force before stress overshoot occurs, thereby performing predictive pressure closed-loop dynamic adjustment on the execution terminal.

[0010] The attitude compensation module is used to introduce multi-dimensional real-time compensation logic for the execution attitude based on process trajectory deviation perception under the constraint of predictive pressure closed-loop dynamic adjustment. It performs real-time deviation compensation on the attitude vector of the execution terminal, so that the execution terminal and the sealed end face of the controlled object are in the same frequency.

[0011] Preferably, the dynamic calibration module includes a virtual momentum assessment unit. The virtual momentum assessment unit is used to monitor the differential slope change of displacement feedback and real-time pressure data in real time after the execution terminal moves to the preset contact point, and to assess the collision momentum between the execution terminal and the controlled object using the differential slope change. When the collision momentum exceeds the preset safety threshold, the virtual momentum assessment unit simulates physical buffering through a logic algorithm and recalculates the output power duty cycle of the execution terminal to suppress the high-frequency stress impact generated by the execution terminal on the controlled object.

[0012] Preferably, the digital characterization data includes the geometric deformation parameters of the end of the controlled object and the thickness tolerance data of the sealing component; the data acquisition module constructs a digital workpiece feature map based on the digital characterization data, and presets a consistency difference compensation amount for a specific workpiece in the initial control parameters according to the digital workpiece feature map.

[0013] Preferably, the transient identification module is also used to monitor the current loop PWM duty cycle of the power drive unit in the feedback execution module; the transient identification module establishes cross-level linkage logic between the result of the differential operation and the current loop PWM duty cycle to achieve a closed-loop response with a response period of less than 1ms.

[0014] Preferably, the dynamic adjustment module sets a dynamic control law self-tuning algorithm based on the physical state migration trend. The dynamic control law self-tuning algorithm is used to offset the contact stress fluctuations generated by the controlled object in the nonlinear stiffness change stage during the sealing operation, so that the sealing stress of the sealing end face is stabilized within the preset target range.

[0015] Preferably, the transient identification module calculates the transient contact stiffness coefficient. To construct a transient contact stiffness model, transient contact stiffness coefficients Follow the formula below: ,in, This is the transient contact stiffness coefficient. This represents the pressure change in real-time pressure data between two adjacent control cycles. This refers to the displacement change of the displacement feedback quantity within two adjacent control cycles; the dynamic adjustment module adjusts the quantity based on the transient contact stiffness coefficient. The time-varying rate is used to determine whether there is a risk of stress overshoot.

[0016] Preferably, the attitude compensation module includes a trajectory deviation sensing unit, which is used to acquire the attitude offset vector of the execution terminal relative to the central axis of the controlled object in real time, and generate multi-dimensional real-time compensation instructions based on the attitude offset vector to drive the execution terminal to perform attitude correction.

[0017] Preferably, the digital characterization data also includes the material modulus data of the sealed interface of the controlled object; the data acquisition module determines the elastic deformation threshold in the physical state migration trend based on the material modulus data, and uses the elastic deformation threshold as the logical judgment basis for triggering the dynamic adjustment module to execute predictive control.

[0018] Preferably, the feedback execution module is also used to perform high-frequency filtering processing on the collected real-time pressure data with a frequency higher than 1000Hz to extract characteristic signals reflecting the mechanical inertia of the execution terminal; the dynamic adjustment module performs compensation correction on the virtual damping factor based on the characteristic signals.

[0019] Preferably, the system also includes a health diagnosis module, which is used to monitor the displacement deviation trend of the execution terminal in multiple operations in real time, and output an early warning signal when the displacement deviation trend continues to exceed the preset aging judgment threshold, and feed it back to the attitude compensation module to increase the compensation weight factor of the attitude vector.

[0020] Compared with the prior art, the beneficial effects of the present invention are:

[0021] 1. In pressure adaptive prediction, a transient contact stiffness model is constructed by real-time monitoring of the differential correlation between the displacement of the actuator and the feedback pressure. The system can identify the physical evolution trend of the interface between the actuator and the controlled object, and actively inject a virtual damping factor when the stiffness hardening rate reaches a preset threshold. This predictive control mechanism, which attenuates the driving force before the actual formation of the pressure peak, avoids the risk of stress overshoot caused by physical hysteresis in traditional pressure feedback regulation, and ensures that the controlled object smoothly transitions from elastic deformation to plastic flow state at the moment of sealing, thereby solving the problem of microcracks that are prone to occur in brittle containers under high-speed conditions.

[0022] 2. The deep coupling of real-time attitude vector deviation compensation and transient impedance identification control enables the system to automatically adapt to workpiece posture deviation and physical consistency fluctuations. This synchronous correction logic of multi-dimensional spatial state and contact stress ensures that the execution terminal always maintains constant effective packaging stress in complex packaging scenarios. This not only eliminates stress damage caused by workpiece size tolerances, but also improves the operational reliability of industrial control systems when facing controlled objects with nonlinear stiffness changes.

[0023] 3. By utilizing the microsecond-level instruction cycle of the high-speed controller, the waveform of the power drive module is fine-tuned before the actuator completes a single step pulse, thereby matching the closed-loop response cycle of the control system with the frequency of physical contact stress evolution. This technical approach, which simulates physical buffering at the underlying drive level through algorithms, offsets the stress impact caused by communication delays without changing the mechanical rigidity structure or reducing the packaging cycle time, thus realizing the transformation of the industrial control system from a lag correction to an in-process suppression control method. Attached Figure Description

[0024] Figure 1 This is a flowchart of the closed-loop control logic of the system that integrates transient identification and attitude compensation in this invention.

[0025] Figure 2 This is a diagram illustrating the hardware hierarchy architecture and data interaction topology between the controller and underlying devices of this invention. Detailed Implementation

[0026] To make the technical objectives, technical solutions, and beneficial effects of the present invention clearer, the present invention will be described below in conjunction with specific embodiments. It should be noted that the following embodiments are intended to explain the present invention and are not intended to limit the scope of protection of the present invention.

[0027] This invention provides a pressure adaptive prediction and dynamic calibration control system for industrial packaging processes, comprising a data acquisition module, a feedback execution module, a transient identification module, a dynamic calibration module, an attitude compensation module, and a health diagnosis module. The data acquisition module establishes an initial control parameter mapping model based on the physical characteristics of the controlled object. The feedback execution module and the transient identification module work collaboratively to achieve real-time acquisition of displacement feedback and pressure data, as well as the construction of a transient contact stiffness model. The dynamic calibration module injects a virtual damping factor based on stiffness evolution to suppress stress overshoot. The attitude compensation module executes multi-dimensional real-time deviation compensation logic. The health diagnosis module monitors the displacement deviation trend of the execution terminal and outputs early warning signals. All modules are connected via a high-speed industrial communication bus, achieving a closed-loop response cycle of less than [missing information]. The control command flow; in the aluminum cap sealing operation in the pharmaceutical field, the difference in physical consistency between the bottle body and the aluminum cap can cause instantaneous fluctuations in contact stiffness, making it difficult for conventional preset stroke control to ensure sealing quality; the data acquisition module is used to acquire digital characterization data that characterizes the physical characteristics of the controlled object, including the geometric deformation parameters of the controlled object's end, the thickness tolerance data of the sealing component, and the material modulus data of the controlled object's sealing interface; the data acquisition module constructs a digital workpiece feature map based on the digital characterization data, and presets the consistency difference compensation amount for a specific workpiece in the initial control parameters according to the digital workpiece feature map.

[0028] In the process of generating initial pose commands, the digital workpiece feature map executes a linear weighted mapping algorithm based on feature weights. The specific execution path is as follows: the system calculates the axial height tolerance components of the acquired end of the controlled object. radial deformation component and material modulus scalar Normalization is performed to obtain the deviation rate of each physical feature relative to a preset reference value. The deviation rate is then multiplied by a preset command gain weight matrix to calculate the expected initial pose of the execution terminal when it moves to the preset contact point. The axial height tolerance weight coefficient, radial deformation weight coefficient, and material modulus weight coefficient in the instruction gain weight matrix are determined through a discrete calibration procedure during the production line startup phase. The system drives the execution terminal to perform contact pressure testing on 10 sets of standard workpieces at a constant speed of 1 mm per second, recording the displacement increment as the pressure feedback value increases from 0 N to 10 N. The axial height tolerance weight coefficient, radial deformation weight coefficient, and material modulus weight coefficient are calibrated to 0.45, 0.25, and 0.30, respectively, to ensure that the initial pose expectation value is numerically equal to the sum of the axial height tolerance component multiplied by 0.45, the radial deformation component multiplied by 0.25, and the material modulus scalar multiplied by 0.30, plus the system's preset stroke step constant. Follow the formula below: ,in, This is the expected initial pose value, in units of ; This represents the axial height tolerance component, in units of... ; Radial deformation component, unit: ; This is a scalar value for the material modulus, with units of 1. ; , , These are the instruction gain weighting coefficients corresponding to the physical characteristics, and are dimensionless. The preset travel step constant for the system, in units of The system uses this mapping algorithm to transform the discrete physical characteristics of the workpiece into the underlying drive pulse reference of the actuator, realizing the deterministic calculation of the initial control parameters. Based on the digital workpiece feature map, the elastic deformation threshold in the physical state transition trend is determined, and this elastic deformation threshold is used as the logical basis for triggering the dynamic adjustment module to execute predictive control. In this procedure, the system generates the sealing pressure target value and the corresponding initial pose command based on the workpiece end tolerance. If the thickness tolerance of a certain batch of aluminum caps is detected to be 0.05mm larger, the image association model automatically corrects the initial displacement, providing an accurate static reference for subsequent pressure compensation.

[0029] Because the threshold for the aluminum cap to transition from elastic deformation to plastic flow at the moment of sealing is extremely narrow, and the physical hysteresis effect causes sensor feedback to often lag behind stress abrupt changes, the feedback execution module is used to issue drive current based on the initial pose command and to collect the displacement feedback quantity of the execution terminal and the real-time pressure data fed back by the pressure sensor in real time; the transient identification module is used to perform differential operations on the displacement feedback quantity and real-time pressure data to obtain the transient contact stiffness characterizing the pressure evolution characteristics with displacement, and then construct a transient contact stiffness model; in the process of constructing this model, the transient identification module calculates the transient contact stiffness coefficient. To characterize the physical evolution trend; transient contact stiffness coefficient Follow the formula below: ,in, This is the transient contact stiffness coefficient, in units of... ; This represents the pressure change in real-time pressure data between two adjacent control cycles, expressed in units of... ; The displacement change of the displacement feedback quantity within two adjacent control cycles, in units of The transient identification module is also used to monitor the current loop pulse width modulation (PWM) duty cycle of the power drive unit in the feedback execution module. By establishing the linkage logic between the differential operation result and the current loop PWM duty cycle, a high-frequency closed-loop response is achieved.

[0030] To proactively suppress stress peaks and address workpiece damage caused by abrupt changes in nonlinear stiffness, a dynamic adjustment module monitors the rate of change of transient contact stiffness over time to obtain the stiffness hardening rate. When the transient identification module detects a transition from elastic deformation to plastic flow at the sealing interface, and the stiffness hardening rate reaches a preset threshold, the dynamic adjustment module actively injects a virtual damping factor into the drive current. This attenuates the drive force's execution gain before stress overshoot occurs, thus performing predictive pressure closed-loop dynamic adjustment on the execution terminal. The virtual damping factor is mapped through the current loop pulse width modulation duty cycle actuator of the underlying drive circuit. Specifically, for every unit increase in the virtual damping factor, the system automatically reduces the current PWM duty cycle register value by 0.5%. If the calculated virtual damping factor reaches 10 units, then... The duty cycle is directly reduced from the initial 60% to 55% in steps, thereby reducing the output current amplitude of the power drive unit to achieve physical-level drive force attenuation. The dynamic calibration module includes a virtual momentum evaluation unit, which monitors the change in differential slope in real time after the execution terminal moves to the preset contact point to evaluate the collision momentum between the execution terminal and the controlled object. When the collision momentum exceeds the preset safety threshold, the system recalculates the output power duty cycle of the execution terminal through an algorithm, for example, reducing the current loop duty cycle from 60% to 45% to suppress the high-frequency stress impact generated by the execution terminal on the controlled object. In addition, the dynamic calibration module sets a dynamic control law self-tuning algorithm according to the physical state migration trend to offset the contact stress fluctuations generated by the controlled object in the nonlinear stiffness change stage during the sealing operation, so that the sealing stress of the sealing end face is stabilized within the target range.

[0031] The dynamic tuning module injects a virtual damping factor into the drive current. Utilizing stiffness hardening rate With preset stiffness hardening rate threshold Difference and preset damping gain coefficient Calculate the virtual damping factor The calculation formula is: ; For virtual damping factor, This is the damping gain coefficient. For stiffness hardening rate, Stiffness hardening rate threshold; damping gain coefficient During the commissioning phase, stepped response testing confirmed that, with the execution terminal in a no-load state, a step current command with an amplitude of 10% of the pressure target value was applied to the power drive unit, and the current was adjusted incrementally. The displacement feedback overshoot is within the preset 2% deviation range, and the stiffness hardening rate threshold is met. The process-driven execution terminal is determined, and 50 sealing operation cycles are continuously performed under standard packaging conditions. The transient identification module records the stiffness hardening rate corresponding to the sampling points of each cycle. Extract the maximum points and calculate the arithmetic mean. Take the arithmetic mean A value of 1.2 times is used as the stiffness hardening rate threshold. Collision momentum safety threshold By calculating energy conversion efficiency Confirmed, energy conversion efficiency To determine the conversion rate of kinetic energy from the actuator to deformation energy at the sealing interface, during the calibration phase, the actuator was driven to press against the controlled object at a stepped velocity from 10 mm / s to 50 mm / s, and the pressure and displacement evolution curves were recorded to identify the energy conversion efficiency. The instantaneous momentum at the moment of a sudden change in slope is set as a safe threshold for collision momentum. .

[0032] Under the constraint of predictive pressure closed-loop dynamic adjustment, the attitude compensation module introduces multi-dimensional real-time compensation logic based on process trajectory deviation perception to perform real-time deviation compensation on the attitude vector of the execution terminal, ensuring that the execution terminal and the sealed end face of the controlled object are in spatial synchronization. The attitude compensation module includes a trajectory deviation perception unit, which is used to acquire the attitude offset vector of the execution terminal relative to the central axis of the controlled object in real time, and generate multi-dimensional real-time compensation commands based on the attitude offset vector to drive the execution terminal to perform attitude correction. In this execution path, the feedback execution module is also used to process the collected real-time pressure data at a frequency higher than 1000. High-frequency filtering is used to extract characteristic signals reflecting the mechanical inertia of the actuator. The dynamic adjustment module compensates and corrects the virtual damping factor based on these characteristic signals, ensuring dynamic stability during the multi-dimensional compensation process. To address displacement drift caused by long-term operation of the actuator, a health diagnosis module monitors the displacement deviation trend of the actuator in real time during multiple operations. When the displacement deviation trend continuously exceeds a preset aging threshold, the health diagnosis module outputs a warning signal and feeds it back to the attitude compensation module to increase the compensation weight factor of the attitude vector. By introducing this closed-loop diagnostic mechanism, the system can automatically adapt to gap changes caused by mechanical aging, improving the operational reliability of the industrial control system without reducing the packaging cycle time.

[0033] Example 1: Under a pharmaceutical packaging process with a cyclic production rate of 7200 units, the controlled object sealing process faces an aluminum cap thickness tolerance of 0.05. Furthermore, the physical challenge lies in the uneven distribution of the pressure resistance limit of glass bottles. When a batch of aluminum caps exhibits a positive thickness deviation, the data acquisition module constructs a digital workpiece feature map by acquiring the geometric deformation parameters of the controlled object's end. It then determines the compensation amount for the specific workpiece in the initial control parameters, enabling the system to lower the initial pose command by 0.05mm based on this digital workpiece feature map at the beginning of the displacement phase. This corrects the static error caused by thickness fluctuations before physical contact occurs. The feedback execution module acquires the displacement feedback amount with a sampling period of 1ms. With real-time stress data The transient identification module performs differential operations on the displacement and pressure increments of adjacent sampling periods to obtain the transient contact stiffness coefficient, which reflects the evolution of the elastic deformation of the sealing interface into plastic flow. The specific calculation formula is as follows: ,in, This is the transient contact stiffness coefficient, in units of... ; The pressure change within adjacent sampling periods, in units of ; The displacement change within adjacent sampling periods, in units of .

[0034] When the transient contact stiffness coefficient is monitored When the force jumps from 80 N / mm to 150 N / mm in three consecutive cycles, the dynamic adjustment module determines that the physical state migration trend has reached the critical point of plastic flow. The virtual momentum assessment unit triggers the virtual damping injection program, switching the duty cycle of the current loop pulse width modulation of the power drive unit from 60% to 45%, attenuating the output power gain of the execution terminal. The aforementioned predictive control based on stiffness evolution law ensures that the driving force is attenuated before the stress peak is formed, offsetting the stress overshoot caused by the physical inertia of the actuator. At the same time, the attitude compensation module drives the execution terminal to perform spatial angle correction based on the attitude offset vector obtained by the trajectory deviation sensing unit, eliminating the stress concentration caused by the tilt of the aluminum cover. The health diagnosis module monitors the statistical variance of the displacement feedback, which decreases from 0.02 to 0.005, indicating that the running trajectory of the actuator matches the physical characteristics of the controlled object, eliminating container damage caused by contact stress fluctuations, and ensuring that the sealed workpiece is in good condition. No leakage was observed during the pressure test.

[0035] Example 2: This experiment verifies the stability of the pressure adaptive prediction scheme in a pharmaceutical container aluminum cap packaging control simulation platform. A servo pressure actuator and a pressure detection unit with a sampling frequency of no less than 10kHz are used to simulate the sealing impedance characteristics of vials. To simulate an industrial environment, Gaussian white noise with a signal-to-noise ratio of 25dB is actively superimposed in the experimental environment. The experiment focuses on determining the core parameter sampling period T. The value of sampling period T is constrained by the real-time requirements of the controller and the requirements for signal aliasing suppression. If the sampling period T is too large, the transient identification module will be unable to capture the abrupt change in the stiffness of the sealing interface from linear to nonlinear. If the sampling period T is too small, it will lead to redundant consumption of computing resources. Based on the frequency distribution of high-frequency components in the physical state transition trend, the system determines that the range of the sampling period T should tend towards the lower limit of its computing power. In the specific deployment of this experiment, the sampling period T is set to 1ms. The data acquisition module pre-obtains the geometric deformation parameters of the controlled object's end as 0.032mm and constructs a digital workpiece feature map as a correction benchmark for the initial pose command.

[0036] This experimental design includes the present invention's sample group, control group, partially missing control group, and out-of-range control group. The control group adopts a proportional-integral feedback control method based on a preset stroke. The partially missing control group enables transient identification but disables the virtual damping injection logic in the dynamic calibration module. The out-of-range control group adjusts the sampling period. The sampling time was set to 2.5ms to verify the impact of sampling frequency on identification accuracy; during the experiment, the pressure detection unit collected real-time pressure data. The data is then uploaded to the control center. The transient identification module performs differential operations on the data from adjacent sampling intervals to determine the transient contact stiffness coefficient. The calculation formula is as follows: ,in, This is the transient contact stiffness coefficient, in units of... ; This represents the change in real-time pressure data between two adjacent sampling periods, in units of... ; This represents the change in displacement feedback within two adjacent sampling periods, in units of... See Table 1.

[0037] Table 1: Measurement values ​​of encapsulation stress characteristics and container damage probability under different control modes

[0038]

[0039] Based on the data analysis in Table 1, the control group, due to feedback delay, could not adjust its output power in time when physical contact occurred, resulting in a peak contact stress measurement of 245.3 N, which exceeded the stress fatigue limit of the glass material. The sample group of this invention, through the transient identification module, identified the evolution of the transient contact stiffness coefficient K from 92.4 N / mm in the elastic stage to 148.6 N / mm. The dynamic adjustment module then reduced the duty cycle of the current loop pulse width modulation, allowing the driving force to decay before stress overshoot, thereby stabilizing the peak contact stress measurement at 155.6 N. While the partially missing control group could identify stiffness abrupt changes, due to the lack of virtual damping injection measures, its peak pressure control effect was better than traditional feedback, but it still had the risk of local overpressure. The out-of-range control group, due to the lack of virtual damping injection measures... The reduced sampling frequency leads to a lag in the perception of the physical state transition trend, making it impossible to perform predictive dynamic adjustments during the nonlinear stiffness abrupt change stage, resulting in an increased container breakage rate. This gradient comparison data confirms the synergistic effect of 1ms sampling accuracy and damping compensation mechanism in suppressing high-frequency stress impact. The experimental results objectively demonstrate that the pressure adaptive prediction and dynamic adjustment control system for industrial packaging processes, under real-world operating conditions that include interference noise, can utilize the transient contact stiffness model to achieve real-time analysis of the physical characteristics of the packaging interface. Through the cooperation of the virtual momentum evaluation unit and the virtual damping injection logic, the system possesses the ability to actively intervene at the microsecond-level contact instant, eliminating stress damage caused by differences in the physical consistency of the workpiece, and ensuring that the sealed controlled object does not suffer structural failure under a 0.5MPa pressure environment.

[0040] Example 3: This example combines Figures 1 to 2 Description of the pressure adaptive prediction and dynamic adjustment control system for industrial packaging processes, such as... Figure 1 As shown, the data acquisition module establishes an initial control parameter image model and generates a sealing pressure target value and an initial pose command, which are then transmitted to the feedback execution module. The feedback execution module issues a drive current based on the initial command and collects the displacement feedback and pressure data of the execution terminal in real time. The displacement feedback and pressure data are input to the transient identification module, which identifies the physical state migration trend by constructing a transient contact stiffness model and transmits this trend to the dynamic adjustment module. While monitoring the stiffness hardening rate, a virtual damping factor is injected to suppress stress overshoot, and then a predictive pressure closed-loop adjustment signal is output to the feedback execution module. At the same time, the adjustment constraint signal is sent to the attitude compensation module. After performing real-time attitude vector deviation compensation and making the sealing end face spatially synchronized, a multi-dimensional real-time compensation command is generated and fed back to the feedback execution module to form a closed-loop control loop.

[0041] like Figure 2As shown, the hardware architecture mainly consists of a core pressure adaptive controller, a high-speed industrial real-time bus in the intermediate transmission layer, and underlying execution and sensing devices. The pressure adaptive controller integrates a transient identification module, a dynamic adjustment module, an attitude compensation module, and a predictive control flow logic unit. The controller establishes a connection with the lower-level devices through a high-speed industrial real-time bus with a response time of less than 1ms. The lower-level devices include a high-frequency sensing group composed of pressure sensors and displacement sensors, and a feedback execution terminal containing a power drive unit. During operation, the high-frequency sensing group detects the physical characteristics of the controlled workpiece and converts them into collected data, which is then transmitted back to the controller. After processing, the controller sends dynamic adjustment commands to the feedback execution terminal through the bus. The feedback execution terminal then generates driving force and acts on the controlled workpiece, thereby constructing a complete physical system that includes sensing, transmission, decision-making, and execution.

[0042] Example 4: In the aluminum cap packaging production line for pharmaceutical containers, when the cumulative number of executions of the terminal exceeds... At this time, the controlled object's sealing process faces displacement drift caused by the increased mechanical transmission clearance of the servo actuator. When the mechanical transmission clearance changes from the initial 0.005mm to 0.025mm, the initial pose command issued by the feedback execution module deviates from the contact point of the sealing interface, causing a time lag in the physical state migration trend captured by the transient identification module. To determine the aging judgment threshold, the system uses the health diagnosis module to perform a standardized calibration procedure during production line downtime. The health diagnosis module drives the execution terminal to run to the zero reference position under no-load conditions, and the position displacement sensor collects the no-load return error of the execution terminal in 500 reciprocating motions. The system calculates the arithmetic mean of the no-load return error as the basic wear amount, and determines the wear amount based on three times the standard deviation of the basic wear amount. An aging judgment threshold of 0.015 mm is determined. When the average deviation between the real-time monitored displacement feedback and the commanded displacement continuously exceeds this aging judgment threshold, the health diagnosis module determines that the actuator enters the aging compensation mode. Under this condition, the data acquisition module constructs a digital workpiece feature map. The data acquisition module uses a high-precision laser displacement gauge to acquire a three-dimensional point cloud array of the sealed end face of the controlled object, extracts the axial geometric feature values ​​from the point cloud array, and encapsulates these axial geometric feature values ​​into structured data to generate a digital workpiece feature map. The digital workpiece feature map includes a workpiece identifier field, a geometric deviation vector field, and a material modulus scalar field. The geometric deviation vector consists of an axial height tolerance component and a radial deformation component.

[0043] The initial control parameter mapping model reads the digitized workpiece feature map and calculates the initial pose command correction amount by combining the basic wear amount output by the health diagnosis module. Initial pose command correction amount The calculation formula is as follows: ,in, This is the initial pose command correction amount, in units of ; These are the geometric deviation components of the workpiece determined by the digitized workpiece feature map, in units of... ; The basic wear amount is measured by the health diagnostic module, in units of... The initial control parameter mapping model will calculate the initial pose command correction amount. The initial pose command is superimposed on the actual position command, enabling the execution terminal to pre-compute a composite compensation for transmission backlash and workpiece tolerance before contacting the sealing interface; by executing the above procedure, the system reduces the displacement control deviation from 0.025... Reduced to 0.005 Within this range, it is ensured that the sampling starting point of the transient identification module for the transient contact stiffness coefficient coincides with the physical contact point of the sealing interface, thereby enabling the dynamic adjustment module to accurately inject the virtual damping factor when the stiffness hardening rate reaches the preset threshold. This embodiment, through feedback adjustment of the health diagnosis module, enables the system to maintain the sealing pressure control accuracy under mechanical aging conditions, solving the problem of accuracy degradation caused by long-term operation of the actuator.

[0044] Example 5: In a pharmaceutical container sealing operation scenario where material modulus data fluctuations exceed 15%, to calibrate the collision momentum safety threshold during the transient contact between the execution terminal and the controlled object, the data acquisition module drives the execution terminal at 10... Up to 50 The system performs a reciprocating test at a stepped speed, and records the corresponding displacement feedback. Real-time pressure data under changing conditions The transient stiffness envelope at different velocities is calculated, and the collision momentum safety threshold is determined based on the energy conversion characteristics of the linear and nonlinear segments in the transient stiffness envelope. ,in, The collision momentum safety threshold, in units of In this procedure, the system calculates the conversion efficiency of the kinetic energy of the execution terminal into the deformation energy of the sealing interface. ,set up The collision momentum at the point of change is the collision momentum safety threshold. The obtained value serves as the logical trigger for the dynamic adjustment module to inject a virtual damping factor into the drive current.

[0045] When the production line faces situations requiring terminal replacement or material property switching of sealing components, the dynamic calibration module determines the stiffness hardening rate threshold by executing a parameter self-calibration program. The transient identification module operates at a speed of 0.5 during the runtime phase. The sampling period is used to obtain the contact stress curve of the aluminum cap as it transitions from elastic deformation to plastic flow, and the transient contact stiffness coefficient is calculated. The rate of change over time, i.e., the stiffness hardening rate Stiffness hardening rate The calculation formula is as follows: ,in, Stiffness hardening rate, in units of ; This is the transient contact stiffness coefficient, in units of... ; Sampling time, in units of The system statistically analyzes the distribution of the maximum stiffness hardening rate during 50 consecutive sealing cycles, and takes 1.2 times the central value of this distribution as the stiffness hardening rate threshold. This enables the control system to identify nonlinear inflection points in the physical state transition trend during the adjustment of the target value of the sealing pressure.

[0046] Example 6: In a scenario where a new batch of sealing components is deployed on a pharmaceutical packaging production line, the controlled object sealing process faces challenges related to the material modulus changing from 70... Fluctuation to 82 To address the physical challenges, the data acquisition module performs a standardized material modulus calibration procedure before operation. The system drives the execution terminal to press against the sealing component at a uniform speed of 1 mm / s and collects real-time pressure data. With displacement feedback The stress-to-strain ratio at the sealing interface during the elastic phase is calculated to obtain the material modulus scalar. Specifically, the formula for calculating the material modulus scalar is as follows: ,in, This is a scalar value for the material modulus, with units of 1. ; Contact stress at the sealed interface, in units of ; The compressive strain of the sealing component is dimensionless. The system records the average value of 10 tests and fills the obtained value into the material modulus scalar field of the digital workpiece feature map, thus correcting the target sealing pressure value generated by the initial control parameter mapping model from 120N to 135N, thereby achieving the matching of the initial control parameters with the material properties. To address the mechanical inertia frequency shift problem caused by the replacement of the execution terminal, the execution module is fed back to execute a spectrum feature identification procedure to determine the filter cutoff frequency. The system injects a sweep excitation current with a frequency range of 10Hz to 2000Hz into the power drive unit. Simultaneously, a pressure sensor monitors the residual current response during the sealing process. The system identifies a mechanical resonance peak in the actuator at 850Hz and sets the filter cutoff frequency based on the resonance isolation principle. It is 1.2 times the resonant frequency, that is, set... It is 1020Hz, where, This is the filter cutoff frequency, in units of By performing this on-site pre-deployment calibration method, the system eliminates interference signals generated by mechanical inertia, enabling the transient identification module to accurately determine the transient contact stiffness coefficient. The differential calculation results reflect the deformation characteristics of the sealing assembly. When the stiffness hardening rate exceeds [a certain value], [the result is observed]. At that time, the dynamic adjustment module issues a command to reduce the duty cycle of the current loop pulse width modulation, suppressing the generated 220N impact load to within 160N.

[0047] The vibration of the conveyor track caused a deviation of 1.5 cm between the center axis of the controlled object and the center axis of the execution terminal. ∘ In pharmaceutical packaging operations, the attitude compensation module executes a spatial alignment procedure by parsing the attitude offset vector, and the trajectory deviation sensing unit obtains the three-dimensional Euler angle deviation of the execution terminal relative to the sealed end face of the controlled object in real time. , , The compensation control law calculates the target deflection angle correction of the servo adjustment mechanism based on the coordinate transformation matrix. This calculation logic is implemented through a three-axis spatial pulse distributor: the number of horizontal and vertical offset pulses of the execution terminal relative to the central axis of the controlled object is obtained, and the horizontal deflection angle, pitch deflection angle, and rotation deflection angle in the three-dimensional Euler angle deviation are mapped to the step pulse increment of the servo driver, where each 0.1 degree deviation of the horizontal deflection angle corresponds to 50 compensation pulses, and the step frequency is set to 2000Hz; by performing a division judgment between the deflection angle deviation and the preset step resolution of 0.002 degrees per pulse, a multi-axis linkage spatial attitude dynamic correction command is generated. The system drives the execution terminal to complete the dynamic correction of the spatial attitude within a 5ms response period, so that the angle between the axial pressure vector and the normal vector of the sealing end face converges from 1.5° to within 0.1°. By injecting the spatial geometric deviation into the attitude correction loop in real time, the system eliminates the single-sided contact stress concentration caused by the inaccuracy of the position and makes the pressure distribution uniformity on the sealing ring surface reach 96.5%.

[0048] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0049] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process, characterized in that, include: The data acquisition module is used to acquire digital characterization data that represents the physical characteristics of the controlled object, and to establish a mapping relationship model between the physical characteristics of the controlled object and the initial control parameters based on the digital characterization data, thereby generating the target value of the sealing pressure and the corresponding initial pose command. The feedback execution module is used to issue drive current according to the initial pose command and to collect the displacement feedback amount of the execution terminal and the real-time pressure data fed back by the pressure sensor in real time. The transient identification module is used to perform differential operations on the displacement feedback and real-time pressure data, obtain the transient contact stiffness that characterizes the pressure evolution with displacement, construct a transient contact stiffness model, and identify the physical state migration trend of the sealing interface from elastic deformation to plastic flow based on the transient contact stiffness model. The dynamic adjustment module is used to monitor the rate of change of transient contact stiffness over time to obtain the stiffness hardening rate. When the stiffness hardening rate reaches a preset threshold, a virtual damping factor is actively injected into the driving current to attenuate the execution gain of the driving force before stress overshoot occurs, thereby performing predictive pressure closed-loop dynamic adjustment on the execution terminal. The attitude compensation module is used to introduce multi-dimensional real-time compensation logic for the execution attitude based on process trajectory deviation perception under the constraint of predictive pressure closed-loop dynamic adjustment. It performs real-time deviation compensation on the attitude vector of the execution terminal, so that the execution terminal and the sealed end face of the controlled object are in the same frequency.

2. The pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process according to claim 1, characterized in that, The dynamic calibration module includes a virtual momentum evaluation unit. After the execution terminal moves to the preset contact point, the virtual momentum evaluation unit monitors the change of the differential slope of the displacement feedback and real-time pressure data in real time, and uses the change of differential slope to evaluate the collision momentum between the execution terminal and the controlled object. When the collision momentum exceeds the preset safety threshold, the virtual momentum evaluation unit simulates the physical buffer through a logic algorithm and recalculates the output power duty cycle of the execution terminal.

3. The pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process according to claim 1, characterized in that, Digital characterization data includes geometric deformation parameters of the controlled object's end and thickness tolerance data of the sealing assembly; The data acquisition module constructs a digital workpiece feature map based on the digital characterization data, and presets the consistency difference compensation amount for a specific workpiece in the initial control parameters according to the digital workpiece feature map.

4. The pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process according to claim 1, characterized in that, The transient identification module is also used to monitor the current loop PWM duty cycle of the power drive unit in the feedback execution module; The transient identification module establishes cross-level linkage logic between the result of differential operation and the PWM duty cycle of the current loop.

5. The pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process according to claim 1, characterized in that, The dynamic adjustment module sets a dynamic control law self-tuning algorithm based on the physical state migration trend. The dynamic control law self-tuning algorithm is used to counteract the contact stress fluctuations caused by the nonlinear stiffness change stage of the controlled object during the sealing operation, so that the sealing stress of the sealing end face is stabilized within the preset target range.

6. The pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process according to claim 1, characterized in that, The transient identification module calculates the transient contact stiffness coefficient. To construct a transient contact stiffness model, transient contact stiffness coefficients Follow the formula below: ,in, This is the transient contact stiffness coefficient. This represents the pressure change in real-time pressure data between two adjacent control cycles. This refers to the displacement change of the displacement feedback quantity within two adjacent control cycles; the dynamic adjustment module adjusts the quantity based on the transient contact stiffness coefficient. The time-varying rate is used to determine whether there is a risk of stress overshoot.

7. The pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process according to claim 1, characterized in that, The attitude compensation module includes a trajectory deviation sensing unit, which is used to acquire the attitude offset vector of the execution terminal relative to the central axis of the controlled object in real time, and generate multi-dimensional real-time compensation instructions based on the attitude offset vector to drive the execution terminal to perform attitude correction.

8. The pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process according to claim 1, characterized in that, Digital characterization data also includes material modulus data of the sealed interface of the controlled object; The data acquisition module determines the elastic deformation threshold in the physical state migration trend based on the material modulus data, and uses the elastic deformation threshold as the logical basis for triggering the dynamic adjustment module to execute predictive control.

9. The pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process according to claim 1, characterized in that, The feedback execution module is also used to perform high-frequency filtering processing (frequency higher than 1000Hz) on the collected real-time pressure data to extract characteristic signals reflecting the mechanical inertia of the execution terminal; the dynamic adjustment module performs compensation correction on the virtual damping factor based on the characteristic signals.

10. The pressure adaptive prediction and dynamic adjustment control system for an industrial packaging process according to claim 1, characterized in that, The system also includes a health diagnosis module, which is used to monitor the displacement deviation trend of the execution terminal in multiple operations in real time. When the displacement deviation trend continues to exceed the preset aging judgment threshold, an early warning signal is output and fed back to the attitude compensation module to increase the compensation weight factor of the attitude vector.

Images