Automatic packaging production line optimization method and system based on intelligent control

Abstract

The invention relates to an automatic packaging production line optimization method and system based on intelligent control, and the method comprises the steps: collecting the operation data flow of key equipment of a production line in real time, and generating an equipment state vector; adjusting an abnormal detection threshold according to the current production material characteristics, comparing the equipment state vector with the adjusted abnormal detection threshold, and identifying a pseudo abnormal signal; when false abnormal signals exceeding a preset shutdown threshold value are continuously detected in a preset time window, a shutdown instruction triggered by the false abnormal signals is intercepted, and risk calculation is triggered; and executing a control decision according to a risk grade value output by the risk calculation so as to realize the purposes of error shutdown elimination, accurate fault isolation and energy efficiency collaborative optimization of the packaging production line under the material characteristic fluctuation working condition.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial automation control, and in particular to an optimization method for an automated packaging production line based on intelligent control and a system thereof. Background Art

[0002] In the field of automated paper packaging production lines, the operational stability of equipment has long been severely affected by fluctuations in material properties. Dynamic changes in parameters such as cardboard weight and moisture content lead to abnormal fluctuations in vacuum adsorption pressure and harmonic distortion of the transmission system current, triggering traditional fixed threshold protection mechanisms to frequently cause false shutdowns. Although existing technologies attempt to improve monitoring accuracy through multi-sensor fusion, they fail to establish a dynamic mapping relationship between material properties and equipment status, and lack the ability to accurately identify false abnormal signals. At the same time, the control strategy is rigid and cannot distinguish between real faults and material disturbances, resulting in an average annual unplanned downtime of more than 15% for the production line, causing huge production capacity losses. Especially in high-speed and precision packaging scenarios, the mechanical impact caused by sudden stops exacerbates equipment wear and tear, forming a vicious cycle. Summary of the Invention

[0003] The main purpose of the present invention is to provide an optimization method and system for an automated packaging production line based on intelligent control, which achieves the purpose of eliminating false shutdowns, accurately isolating faults and collaboratively optimizing energy efficiency of the packaging production line under conditions of material property fluctuations through dynamic threshold collaborative judgment, equipment status risk decoupling and life adaptive decision-making mechanism.

[0004] To achieve the above objectives, the present invention provides an optimization method for an automated packaging production line based on intelligent control, comprising the following steps: Collect the operation data stream of key equipment on the production line in real time and generate equipment status vectors; Adjusting an anomaly detection threshold according to the current production material characteristics, comparing the device state vector with the adjusted anomaly detection threshold to identify false anomaly signals; When false abnormality signals exceeding the preset shutdown threshold are continuously detected within a preset time window, the shutdown instruction triggered by the false abnormality signal is intercepted and risk calculation is triggered; The control decision is executed according to the risk level value output by the risk calculation, wherein the risk level value includes a vacuum leakage risk level value and a mechanical transmission risk level value.

[0005] Furthermore, the step of collecting the operation data stream of key equipment of the production line in real time and generating the equipment state vector includes: Deploy pressure sensing units at the main gas path nodes of the vacuum adsorption device on the production line to obtain pressure fluctuation time series data through a preset sampling frequency; A current sensing unit is integrated at the power input end of the servo drive mechanism to extract the harmonic spectrum characteristics of the motor operating current; The pressure fluctuation time series data is time-synchronized and fused with harmonic spectrum features to generate a multi-dimensional device state vector containing time-domain pressure indicators and frequency-domain current indicators.

[0006] Further, the step of adjusting the abnormality detection threshold according to the current production material characteristics comprises: obtaining a reference pressure fluctuation threshold and a reference current harmonic threshold in a device calibration phase; calculating a humidity compensation offset of the pressure fluctuation threshold according to real-time ambient humidity data to generate a compensated pressure fluctuation threshold; calculating a grammage scaling coefficient of the current harmonic threshold according to paperboard grammage data to generate a proportionally adjusted current harmonic threshold; outputting the compensated pressure fluctuation threshold and the proportionally adjusted current harmonic threshold as the abnormality detection threshold.

[0007] Further, the step of comparing the device state vector with the adjusted abnormality detection threshold to identify pseudo-abnormal signals comprises: when the peak value of the sliding window of the vacuum pressure fluctuation sequence exceeds the compensated pressure fluctuation threshold, marking a pressure abnormality; when the harmonic energy integral value of the servo current in the harmonic frequency band exceeds the proportionally adjusted current harmonic threshold, marking a current abnormality; when the pressure abnormality mark and the current abnormality mark are triggered synchronously within a preset time, it is determined as a pseudo-abnormal signal caused by material characteristic fluctuation.

[0008] Further, when the pseudo-abnormal signals exceeding the preset shutdown threshold are continuously detected within a preset time window, the shutdown instruction triggered by the pseudo-abnormal signals is intercepted, and the step of triggering risk calculation comprises: setting a hardware signal interceptor at the shutdown instruction output end of the programmable logic controller; when the pseudo-abnormal signals accumulate to reach a preset trigger frequency threshold within a preset time window, the hardware signal interceptor is activated to block the shutdown instruction; a priority interrupt signal is sent to the coprocessor at the same time to start a real-time risk calculation process based on the device state vector within the sliding time window.

[0009] Further, the step of starting a real-time risk calculation process based on the device state vector within the sliding time window comprises: constructing a probabilistic graphical model containing device state correlation factors, the correlation factors including a vacuum sealing state factor and a transmission stability factor; inputting the device state vector into the probabilistic graphical model for belief propagation calculation; outputting a two-channel probability value representing the vacuum system risk and the mechanical system risk as the risk level value.

[0010] Further, the step of executing control decisions according to the risk level value output by the risk calculation comprises: When the vacuum leakage risk level value is lower than the vacuum risk threshold, a pulse width modulation waveform reconstruction instruction of the vacuum electromagnetic valve is generated, and through the pulse width modulation waveform reconstruction instruction, an anti-phase compensation waveform is generated by the vacuum electromagnetic valve to suppress pressure fluctuation; When the mechanical transmission risk level value is higher than the mechanical risk threshold, a staged deceleration stop instruction sequence of the servo driver is generated, and through the staged deceleration stop instruction sequence, the servo driver is controlled to decelerate to stop at a preset time gradient; Wherein, the vacuum risk threshold and the mechanical risk threshold are dynamically calibrated according to the equipment running time.

[0011] Further, the step of generating the pulse width modulation waveform reconstruction instruction of the vacuum electromagnetic valve comprises: Based on the phase distribution characteristics of the pressure fluctuation sub-vector, an anti-phase compensation waveform of the electromagnetic valve drive is constructed; The compensation waveform is superimposed on the original driving signal through a feedforward control channel, and the pressure fluctuation amplitude is converged to a safe interval.

[0012] The application also provides an intelligent control-based automated packaging production line optimization system, comprising: A data acquisition module for acquiring real-time production line key equipment operation data streams and generating equipment state vectors; A state judgment module for adjusting abnormality detection thresholds according to current production material characteristics, comparing the equipment state vectors with the adjusted abnormality detection thresholds, and identifying pseudo-abnormal signals; A signal interception module for intercepting stop commands triggered by pseudo-abnormal signals when continuously detecting pseudo-abnormal signals exceeding a preset stop threshold within a preset time window, and triggering risk calculation; A risk calculation unit for executing control decisions according to the risk level value output by the risk calculation.

[0013] The application also provides a computer device comprising a memory and a processor, the memory storing a computer program, and the processor executing the computer program to realize the steps of the above-mentioned intelligent control-based automated packaging production line optimization method.

[0014] The automated packaging production line optimization method and system based on intelligent control provided by the present invention have the following beneficial effects: the present invention accurately identifies pseudo-abnormal signals caused by fluctuations in material properties through a dual-threshold collaborative judgment and time window constraint mechanism, reduces the problem of false shutdown caused by traditional fixed threshold protection, and ensures production continuity; decouples the equipment state vector into independent channels of vacuum leakage risk and mechanical failure risk through the constructed probability graph model, outputs quantitative risk level values, provides a theoretical basis for hierarchical control decisions, and improves the accuracy of fault identification; the dynamic threshold calibration mechanism based on the equipment's operating years automatically adjusts the risk response standard to adapt to the aging law of the equipment, and reduces maintenance costs while extending the life of key components; and actively suppresses vacuum pressure fluctuations through inverse compensation waveform reconstruction and feedforward control technology, thereby simultaneously improving system stability and energy utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 1 is a flow chart of an optimization method for an automated packaging production line based on intelligent control in one embodiment of the present invention; Figure 2 This is a structural block diagram of an automated packaging production line optimization system based on intelligent control in one embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0016] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] Reference Figure 1 , which is a flow chart of an optimization method for an automated packaging production line based on intelligent control proposed by the present invention, comprising the following steps: S1, collects the operating data stream of key equipment on the production line in real time and generates equipment state vectors; S2, adjusting the anomaly detection threshold according to the current production material characteristics, comparing the device state vector with the adjusted anomaly detection threshold, and identifying false anomaly signals; S3, when false abnormality signals exceeding a preset shutdown threshold are continuously detected within a preset time window, the shutdown instruction triggered by the false abnormality signal is intercepted and risk calculation is triggered; S4, performing a control decision according to a risk level value output by the risk calculation, the risk level value including a vacuum leakage risk level value and a mechanical transmission risk level value.

[0019] In one embodiment, for step S1, comprising: The step of collecting the running data stream of the key equipment of the production line in real time and generating the equipment state vector comprises: Deploying a pressure sensing unit at the main gas path node of the vacuum adsorption device of the production line to obtain pressure fluctuation time series data through a preset sampling frequency; Integrating a current sensing unit at the power input end of the servo drive mechanism to extract the harmonic spectrum characteristics of the motor operating current; Time-synchronously fusing the pressure fluctuation time series data and the harmonic spectrum characteristics to generate a multi-dimensional equipment state vector containing time-domain pressure indicators and frequency-domain current indicators.

[0020] In a specific embodiment, a piezoresistive pressure sensing unit is deployed at the main gas path node (gas source distributor outlet) of the vacuum adsorption device of the production line, and a preset sampling frequency is used to obtain pressure fluctuation time series data, which needs to meet the constraint condition that the maximum separation frequency for high-speed transmission of paperboard is 8-12 Hz, and the typical value is 8-12 Hz. The original pressure signal is denoised by a Kalman filter to calculate the standard deviation of pressure fluctuation as a time-domain pressure indicator: wherein, is the standard deviation of pressure fluctuation, and the average adsorption pressure; a Hall current sensing unit is integrated at the power input end of the servo drive mechanism to collect three-phase current and extract the harmonic spectrum through fast Fourier transform (FFT) to calculate the energy integral value of the 2-5 harmonic frequency band wherein, f is the power grid fundamental frequency; the pressure data and the harmonic data are time-synchronously aligned, and a delay compensation algorithm is used to eliminate time series deviation, wherein, d is the sensor response compensation amount; finally, a two-dimensional equipment state vector is fused and generated wherein, quantifies the stability of the vacuum system, and represents the mechanical transmission load state. This step provides key feature basis for subsequent pseudo-abnormality identification through double-parameter coupled change.

[0021] In one embodiment, for step S2, comprising: The step of adjusting the abnormality detection threshold according to the current production material characteristics comprises:​​​ Obtain the reference pressure fluctuation threshold and reference current harmonic threshold during the equipment calibration phase; Calculate the humidity compensation offset of the pressure fluctuation threshold according to the real-time ambient humidity data to generate the compensated pressure fluctuation threshold; Calculate the weight scaling coefficient of the current harmonic threshold according to the cardboard weight data, and generate the current harmonic threshold after proportional adjustment; The compensated pressure fluctuation threshold and the proportionally adjusted current harmonic threshold are output as abnormality detection thresholds.

[0022] In a specific embodiment, the reference pressure fluctuation threshold value determined during the equipment calibration phase is obtained. (typical value 0.08kPa) and the reference current harmonic threshold (Typical value 0.25); Based on the real-time ambient humidity sensor data H (unit: %RH), the pressure threshold offset is calculated using a linear compensation model: Where, is the humidity compensation coefficient, Generates compensated pressure fluctuation threshold for humidity calibration At the same time, based on the online measured cardboard weight G (unit: g / m²), a nonlinear scaling function is used Calculate the harmonic threshold adjustment coefficient ( is the attenuation factor, As the benchmark weight, 、 is the process range boundary), generates the proportionally adjusted current harmonic threshold ;Finally output the dynamic anomaly detection threshold set The dynamic threshold mechanism of this embodiment reduces the false anomaly misjudgment rate caused by material disturbances compared to the traditional fixed threshold solution.

[0023] In one embodiment, the step of comparing the device state vector with the adjusted anomaly detection threshold to identify a false anomaly signal includes: When the sliding window peak value of the vacuum pressure fluctuation sequence exceeds the compensated pressure fluctuation threshold, a pressure anomaly is marked; When the integral value of the servo current harmonic energy in the harmonic frequency band exceeds the current harmonic threshold after proportional adjustment, the current is marked as abnormal; When the pressure anomaly flag and the current anomaly flag are triggered synchronously within the preset time, it is determined to be a false anomaly signal caused by fluctuations in material characteristics.

[0024] In a specific embodiment, the device state vector Perform feature decoupling and extract vacuum pressure fluctuation sequence ; Using a sliding time window (window length , sliding step length ) Real-time calculation of peak pressure within the window ,when Exceeding the post-compensation pressure fluctuation threshold When the pressure abnormality mark is triggered ;Synchronous analysis of current harmonic energy , when its value exceeds the proportionally adjusted harmonic threshold When the current abnormality flag is triggered ; Introduce synchronization time constraint mechanism, only when and exist When overlap occurs within the time window (i.e., ,in are the abnormality mark triggering moments respectively), and are judged as pseudo abnormality signals caused by fluctuations in material characteristics. For example, when the moisture content of the cardboard increases from 8% to 15%, the pressure peak Up to 0.19kPa ( ) At the same time, harmonic energy , and the double anomaly is triggered synchronously within 5 seconds, and the system accurately identifies it as a false anomaly rather than a real fault; this mechanism reduces the misjudgment rate of the traditional single parameter judgment scheme, in which the time constraint parameter Through 200 sets of material switching experiments, the paper packaging production line was optimized and determined to effectively distinguish between instantaneous interference (<2s) and continuous material disturbance (>5s). It reflects the strong coupling relationship between vacuum pressure fluctuation and current harmonic distortion in the paper packaging production line. That is, the decrease in cardboard permeability caused by humidity leads to abnormal adsorption pressure, which is synchronously accompanied by the sudden change in transmission load current distortion caused by the change in grammage. The dual characteristics are The spatiotemporal correlation within the window becomes the key feature for identifying material disturbances.

[0025] In one embodiment, step S3 includes: When false abnormality signals exceeding a preset shutdown threshold are continuously detected within a preset time window, a shutdown instruction triggered by the false abnormality signal is intercepted and a risk calculation step is triggered, including: A hardware signal interceptor is set at the shutdown command output end of the programmable logic controller; When the number of false abnormal signals reaches the preset triggering threshold within the preset time window, the hardware signal interceptor is activated to block the shutdown command; At the same time, a priority interrupt signal is sent to the coprocessor to start the real-time risk calculation process based on the device state vector within the sliding time window.

[0026] In a specific embodiment, a hardware signal interceptor is connected in series to the shutdown command output of the programmable logic controller (PLC) on the packaging production line. The interceptor adopts a high-speed optical coupling isolation circuit design to form a physical isolation barrier to block the voltage. , ensuring signal blocking response time , through the response time constraint to ensure that the interception action is faster than the response of the mechanical actuator (typical solenoid valve action time is 20ms); set the trigger logic based on event accumulation, when the false abnormal signal is within the preset time window (covering the typical duration of material disturbance 15-25 seconds) the cumulative number of times reaches the preset threshold Second time (i.e. satisfying ,in is the pseudo-anomaly indicator function, the counting threshold It can avoid transient interference (such as single cardboard shifting) and immediately activate the hardware signal interceptor to cut off the transmission path of the stop instruction to the actuator to prevent unnecessary shutdown caused by transient material disturbance; it also sends a high-priority interrupt signal to the coprocessor at the same time (interrupt level ≥ 3, which can preempt 90% of non-real-time tasks (according to IEC 61131-3 standard)). The interrupt forces the current background task to be suspended and starts the real-time risk calculation process, which is based on the latest Device state vector sequence within the time window (reserve All state history within) to build a dynamic risk model, process activation delay (Compare to the operating system task switching delay of more than 100ms). For example, when the cardboard weight mutation causes 5 consecutive false exceptions ( The hardware interceptor blocks the shutdown signal output by the PLC within 0.8ms, while the coprocessor responds to the interrupt within 5ms and loads 128 sets of state vectors to start risk calculation.

[0027] In one embodiment, the step of initiating a real-time risk calculation process based on a device state vector within a sliding time window includes: Constructing a probabilistic graphical model including equipment state correlation factors, wherein the correlation factors include a vacuum sealing state factor and a transmission stability factor; Input the device state vector into the probabilistic graphical model for belief propagation calculation; The dual-channel probability value representing the vacuum system risk and the mechanical system risk is output as the risk level value.

[0028] In a specific embodiment, the key hidden state node in the probabilistic graphical model is defined as the vacuum sealing state factor (0 means normal sealing, 1 means leakage) and transmission stability factor (0 means normal transmission, 1 means failure), and the device state vector is established The observation condition probability relationship with the hidden node is: wherein, is the characteristic mean vector when the vacuum leaks (the pressure fluctuation amplitude is significantly increased), is the covariance matrix; 128 groups of device state vectors in the sliding window are input into the model, and the posterior probability of the hidden node is iteratively calculated by the belief propagation algorithm: The final output is a two-channel risk level value: a vacuum system risk value and a mechanical system risk value . The embodiment maps the device state vector into an interpretable risk quantization value through the probabilistic graphical model, providing a theoretically rigorous mathematical basis for control decisions, and breaking through the limitations of traditional schemes relying on empirical thresholds.

[0029] In one embodiment, for step S4, the following is included: The step of performing control decisions according to the risk level value output by the risk calculation includes: When the vacuum leakage risk level value is lower than the vacuum risk threshold value, a pulse width modulation waveform reconstruction instruction of the vacuum electromagnetic valve is generated, and the vacuum electromagnetic valve generates an anti-phase compensation waveform through the pulse width modulation waveform reconstruction instruction to suppress pressure fluctuations; When the mechanical transmission risk level value is higher than the mechanical risk threshold value, a staged deceleration shutdown instruction sequence of the servo driver is generated, and the servo driver is controlled to decelerate to shutdown at a preset time gradient through the staged deceleration shutdown instruction sequence; Wherein the vacuum risk threshold value and the mechanical risk threshold value are dynamically calibrated according to the equipment operation period.

[0030] In specific embodiments, when the vacuum leakage risk level value is lower than the dynamically calibrated vacuum risk threshold value , that is, when , a pulse width modulation (PWM) waveform reconstruction instruction of the vacuum electromagnetic valve is generated, and the vacuum electromagnetic valve generates an anti-phase compensation waveform through the instruction, which specifically includes: generating a driving signal that is anti-phase with the original disturbance based on the instantaneous phase feature of the pressure fluctuation sub-vector, and injecting the vacuum control system through an independent feedforward control channel (delay <0.1 ms) to make the pressure fluctuation amplitude converge to a safe interval; when the mechanical transmission risk level value is higher than the mechanical risk threshold value (that is, when ​​), generates a graded deceleration stop instruction sequence for the servo drive, which controls the servo drive to execute a three-stage speed reduction stop, including a three-stage speed reduction curve to eliminate mechanical shock: The speed difference between adjacent stages (less than 40% of rated speed), duration of each stage ; Risk threshold and According to the operating life of the equipment (Unit: year) Dynamic adjustment, the formula is: in, is the vacuum risk threshold benchmark value for new machines, is the mechanical risk threshold benchmark value, is the aging sensitivity coefficient of the vacuum system, is the mechanical system aging gain coefficient). This embodiment uses a dynamic calibration dual-threshold decision-making mechanism to extend the service life of the production line while ensuring equipment safety, providing the packaging industry with a life cycle adaptive intelligent control system solution.

[0031] In one embodiment, the step of generating a pulse width modulation waveform reconstruction instruction for a vacuum solenoid valve includes: Based on the phase distribution characteristics of the pressure fluctuation vector, the anti-phase compensation waveform of the solenoid valve drive is constructed; The compensation waveform is superimposed on the original drive signal through the feedforward control channel to converge the pressure fluctuation amplitude to a safe range.

[0032] In a specific embodiment, the pressure fluctuation subvector is extracted from the device state vector (N is the number of sampling points in the sliding window), and its instantaneous phase distribution is calculated by Hilbert transform , where the expression of Hilbert transform is (* denotes convolution operation), which physically means converting the time-domain pressure fluctuation into an analytical signal to extract the instantaneous phase ; The window length is used here , which can cover the main frequency of pressure fluctuation (8-12Hz) to ensure the reduction of phase estimation error. Construct the anti-phase compensation waveform based on the phase characteristics: in is the gain coefficient, , is the original driving signal amplitude, and the gain adopts an adaptive mechanism to meet ( is the safety interval boundary); is the fundamental frequency of the vacuum solenoid valve (typical value 25Hz); phase offset The compensation wave can be precisely aligned with the peak-trough of the pressure disturbance to achieve waveform inversion. Superimposed on the original PWM drive signal , generate reconstructed synthetic drive instructions The synthetic instruction is injected into the vacuum control system through the digital-to-analog converter, so that the pressure fluctuation amplitude converges to a safe range. This embodiment solves the problem of dynamic balance control of vacuum pressure in packaging production lines through precise phase tracking and feedforward direct control technology, which facilitates the production of high-speed and precise packaging.

[0033] Reference Figure 2 , is a structural block diagram of an automated packaging production line optimization system based on intelligent control in one embodiment of the present invention, including: Data acquisition module, used to collect real-time operating data streams of key equipment on the production line and generate equipment state vectors; A state judgment module is used to adjust the abnormality detection threshold according to the characteristics of the current production material, compare the device state vector with the adjusted abnormality detection threshold, and identify false abnormality signals; A signal interception module is used to intercept the shutdown instruction triggered by the false abnormality signal and trigger risk calculation when false abnormality signals exceeding the preset shutdown threshold are continuously detected within a preset time window; The risk calculation unit is used to execute control decisions based on the risk level value output by the risk calculation.

[0034] For the specific implementation of each module in the above device example, please refer to the above method embodiment, which will not be repeated here.

[0035] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, memory, display screen, input device, network interface and database connected via a system bus. The processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the above method is implemented.

[0036] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied.

[0037] To sum up, the present application realizes the real-time collection of the running data stream of the key equipment of the production line and generates the equipment state vector; adjusts the abnormality detection threshold according to the current production material characteristics, compares the equipment state vector with the adjusted abnormality detection threshold, and identifies the pseudo abnormality signal; when the pseudo abnormality signal exceeding the preset shutdown threshold is continuously detected within the preset time window, the shutdown instruction triggered by the pseudo abnormality signal is intercepted, and the risk calculation is triggered; and the control decision is executed according to the risk level value output by the risk calculation, so as to realize the purpose of eliminating the false shutdown, accurately isolating the fault and synergistically optimizing the energy efficiency of the packaging production line under the fluctuating working condition of the material characteristics.

[0038] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. Any reference to memory, storage, database or other medium provided by the present application and used in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM) and memory bus dynamic RAM, etc.

[0039] It should be noted that in this document, the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusions, so that a process, device, article or method including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such a process, device, article or method. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, device, article or method including the element.

[0040] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An optimization method for an automated packaging production line based on intelligent control, characterized in that: The following steps are involved: Collect the operation data stream of key equipment on the production line in real time and generate equipment status vectors; Adjusting an anomaly detection threshold according to the current production material characteristics, comparing the device state vector with the adjusted anomaly detection threshold to identify false anomaly signals; When false abnormality signals exceeding the preset shutdown threshold are continuously detected within a preset time window, the shutdown instruction triggered by the false abnormality signal is intercepted and risk calculation is triggered; The control decision is executed according to the risk level value output by the risk calculation, wherein the risk level value includes a vacuum leakage risk level value and a mechanical transmission risk level value.

2. The method for optimizing an automated packaging production line based on intelligent control according to claim 1, characterized in that: The step of collecting the operation data stream of key equipment of the production line in real time and generating the equipment state vector includes: Deploy pressure sensing units at the main gas path nodes of the vacuum adsorption device on the production line to obtain pressure fluctuation time series data through a preset sampling frequency; A current sensing unit is integrated at the power input end of the servo drive mechanism to extract the harmonic spectrum characteristics of the motor operating current; The pressure fluctuation time series data and harmonic spectrum characteristics are time-synchronously fused to generate a multi-dimensional device state vector containing time-domain pressure indicators and frequency-domain current indicators.

3. The method for optimizing an automated packaging production line based on intelligent control according to claim 1, characterized in that: The step of adjusting the anomaly detection threshold according to the current production material characteristics includes: Obtain the reference pressure fluctuation threshold and reference current harmonic threshold during the equipment calibration phase; Calculate the humidity compensation offset of the pressure fluctuation threshold according to the real-time ambient humidity data to generate the compensated pressure fluctuation threshold; Calculate the weight scaling coefficient of the current harmonic threshold according to the cardboard weight data, and generate the current harmonic threshold after proportional adjustment; The compensated pressure fluctuation threshold and the proportionally adjusted current harmonic threshold are output as abnormality detection thresholds.

4. The method for optimizing an automated packaging production line based on intelligent control according to claim 1, characterized in that: The step of comparing the device state vector with the adjusted anomaly detection threshold to identify a false anomaly signal includes: When the sliding window peak value of the vacuum pressure fluctuation sequence exceeds the compensated pressure fluctuation threshold, a pressure anomaly is marked; When the integral value of the servo current harmonic energy in the harmonic frequency band exceeds the current harmonic threshold after proportional adjustment, the current is marked as abnormal; When the pressure anomaly flag and the current anomaly flag are triggered synchronously within the preset time, it is determined to be a false anomaly signal caused by fluctuations in material characteristics.

5. The method for optimizing an automated packaging production line based on intelligent control according to claim 1, characterized in that: The step of intercepting a shutdown instruction triggered by a false abnormality signal and triggering risk calculation when a false abnormality signal exceeding a preset shutdown threshold is continuously detected within a preset time window includes: A hardware signal interceptor is set at the shutdown command output end of the programmable logic controller; When the number of false abnormal signals reaches the preset triggering threshold within the preset time window, the hardware signal interceptor is activated to block the shutdown command; At the same time, a priority interrupt signal is sent to the coprocessor to start the real-time risk calculation process based on the device state vector within the sliding time window.

6. The method for optimizing an automated packaging production line based on intelligent control according to claim 5, characterized in that: The step of starting a real-time risk calculation process based on the device state vector within the sliding time window includes: Constructing a probabilistic graphical model including equipment state correlation factors, wherein the correlation factors include a vacuum sealing state factor and a transmission stability factor; Input the device state vector into the probabilistic graphical model for belief propagation calculation; The dual-channel probability value representing the vacuum system risk and the mechanical system risk is output as the risk level value.

7. The method for optimizing an automated packaging production line based on intelligent control according to claim 1, characterized in that: The step of executing a control decision based on the risk level value output by the risk calculation includes: When the vacuum leakage risk level value is lower than the vacuum risk threshold, a pulse width modulation waveform reconstruction instruction of the vacuum solenoid valve is generated, and the pulse width modulation waveform reconstruction instruction is used to drive the vacuum solenoid valve to generate an anti-phase compensation waveform to suppress pressure fluctuations; When the mechanical transmission risk level value is higher than the mechanical risk threshold, a graded deceleration and shutdown instruction sequence of the servo drive is generated, and the servo drive is controlled by the graded deceleration and shutdown instruction sequence to decelerate in stages according to a preset time gradient until it stops; The vacuum risk threshold and the mechanical risk threshold are dynamically calibrated according to the operating years of the equipment.

8. The method for optimizing an automated packaging production line based on intelligent control according to claim 7, characterized in that: The step of generating a pulse width modulation waveform reconstruction instruction for the vacuum solenoid valve includes: Based on the phase distribution characteristics of the pressure fluctuation vector, the anti-phase compensation waveform of the solenoid valve drive is constructed; The compensation waveform is superimposed on the original drive signal through the feedforward control channel to converge the pressure fluctuation amplitude to a safe range.

9. An automated packaging production line optimization system based on intelligent control, characterized in that: include: Data acquisition module, used to collect real-time operating data streams of key equipment on the production line and generate equipment state vectors; A state judgment module is used to adjust the abnormality detection threshold according to the characteristics of the current production material, compare the device state vector with the adjusted abnormality detection threshold, and identify false abnormality signals; A signal interception module is used to intercept the shutdown instruction triggered by the false abnormality signal and trigger risk calculation when false abnormality signals exceeding the preset shutdown threshold are continuously detected within a preset time window; The risk calculation unit is used to execute control decisions based on the risk level value output by the risk calculation.

10. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method for optimizing an automated packaging production line based on intelligent control according to any one of claims 1 to 8 are implemented.

Images