Automatic detection method and system for sealing property of vacuum cavity of full-automatic blister equipment

Abstract

The application discloses a kind of automatic detection method and system of full-automatic blister equipment vacuum cavity tightness, specifically related to the field of tightness test, including flow state adaptive segmentation step: in the pre-evacuation vacuum stage of each forming cycle of blister equipment, the pressure value in vacuum cavity, cavity wall temperature and ambient humidity are synchronously collected with sampling frequency ≥100Hz.The application breaks the limitation of fixed pressure detection, identifies the molecular flow-viscous flow transition time with the second derivative inflection point of pressure, realizes flow state adaptive segmentation, uses attenuation slope ratio R to distinguish three kinds of leakage types, realizes leakage mechanism identification;Establish temperature and humidity compensation model to eliminate thermal interference, build three-level response mechanism of leakage rate and leakage type linkage, create transducer marshalling positioning driven by leakage type, combined with self-calibration compensation aging drift, realize full-automatic blister equipment tightness online intelligent detection.

Description

Technical Field

[0001] This invention relates to the field of sealing test technology, and more specifically, to an automatic method and system for detecting the sealing performance of a vacuum chamber in a fully automatic thermoforming device. Background Technology

[0002] In vacuum forming, the sealing performance of the vacuum chamber directly determines the product molding quality. Leakage will lead to defects such as bubbles and wrinkles in the product, and in severe cases, it will cause batch scrapping. At present, the vacuum chamber sealing performance test mainly adopts the pressure decay method: after the vacuum chamber is evacuated to the target vacuum level, the valve is closed and the pressure holding state is entered. The leakage rate is evaluated by monitoring the pressure change during the pressure holding period.

[0003] However, existing methods typically use a fixed holding time and a single pressure change slope to determine leakage. But during the leakage process, the gas flow state changes with the pressure difference between the inside and outside of the cavity. Initially, it is molecular flow, and later it gradually transitions to viscous flow. The two pressure change patterns are fundamentally different. A single slope cannot accurately characterize the complete leakage process. The detection results heavily depend on the selection of the holding window and cannot distinguish between different leakage mechanisms.

[0004] In thermoforming, thermal radiation from the heating furnace and fluctuations in ambient humidity can significantly affect the accuracy of pressure sensors. Existing technologies lack effective compensation schemes for thermal and humid coupling interference in thermoforming equipment, leading to inaccurate leakage rate determination.

[0005] Existing solutions can only determine "whether there is a leak" but cannot locate the leak point. Maintenance personnel need to check each point one by one, resulting in long downtime. A few ultrasonic testing solutions perform full array scanning of all channels, resulting in a large amount of data and the positioning accuracy is easily affected by the noise of the equipment.

[0006] Existing detection thresholds are mostly fixed values. As the equipment ages and the seals deteriorate, these fixed thresholds become inapplicable, leading to an increased false positive rate. There is a lack of adaptive adjustment methods. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art, the present invention provides an automatic detection method and system for the vacuum cavity sealing of a fully automatic thermoforming equipment, so as to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: an automatic detection method for the vacuum cavity sealing performance of a fully automatic thermoforming equipment, comprising the following steps:

[0009] Adaptive flow segmentation steps: During the pre-vacuuming stage of each forming cycle of the thermoforming equipment, at a sampling frequency... ≥100Hz synchronous acquisition of pressure values ​​inside the vacuum chamber Cavity wall temperature and ambient humidity ; at the end of vacuuming Enter the pressure holding observation window as the starting point Regarding the pressure value Perform second derivative analysis, through Automatic identification of the inflection point transition time between the molecular flow region and the viscous flow region The pressure-holding observation window is divided into the first molecular flow attenuation region. Second viscous flow attenuation region The first attenuation slope was obtained by performing linear fitting on the two segments respectively. Second attenuation slope ;

[0010] Multiphysics Real-Time Compensation Steps: Based on the cavity wall temperature and the ambient humidity The pressure value is calculated using a pre-established sheet material-temperature-humidity coupled compensation model. Real-time corrections are performed to obtain the compensated pressure sequence. The coupling compensation model The compensation coefficient group is automatically selected from the preset material mapping table based on the material of the current module sheet;

[0011] Leakage channel feature identification steps: Calculate the first attenuation slope With the second attenuation slope ratio The ratio The leakage type is identified by comparing it with a preset leakage path feature model library. The leakage types include sealing ring failure, structural cracking, and material penetration.

[0012] Graded response steps: based on the compensated pressure sequence Calculate the equivalent leakage rate ,in For cavity volume, Standard atmospheric pressure; based on the aforementioned equivalent leakage rate Implement differentiated response strategies based on the type of leakage.

[0013] Preferably, in the flow state adaptive segmentation step, the automatic identification method for the transition time τ includes:

[0014] Regarding the pressure value Perform sliding window polynomial smoothing filtering;

[0015] Calculate the second derivative of the smoothed pressure sequence ;

[0016] by The transition time is defined as the last zero-crossing point from positive to negative. ;

[0017] If no zero-crossing point meeting the conditions is detected within the entire pressure holding observation window, the midpoint of the pressure holding observation window is used as the default transition time. .

[0018] Preferably, in the multiphysics field real-time compensation step, the preset material mapping table includes at least the compensation coefficient groups corresponding to four commonly used blister packaging sheets: ABS, PS, PET, and PP.

[0019] Among them, the temperature compensation weights for PET and PP materials are higher than those for ABS and PS materials.

[0020] Preferably, in the leakage channel feature model library, the ratio range corresponding to the sealing ring compression deformation model is: The ratio range corresponding to the microcrack model at the joint is: The ratio range corresponding to the material penetration model is: ,in And the intervals do not overlap;

[0021] When the ratio When the value falls within the ratio range of the seal ring compression deformation model, the leakage type is determined to be seal ring failure;

[0022] When the ratio When the ratio falls within the range of the joint microcrack model, the leakage type is determined to be structural cracking;

[0023] When the ratio When the value falls within the ratio range of the material penetration model, the leakage type is determined to be material penetration.

[0024] Preferably, the differentiated response strategy in the hierarchical response step includes:

[0025] when At that time, it was judged to be grade A qualified, and production was allowed to continue;

[0026] when Furthermore, when the leakage type is material penetration, it is determined to be a Class B warning, the current batch is marked and the next inspection cycle is shortened, while the pre-vacuuming time of the next molding cycle is automatically extended;

[0027] when If the leakage type is a sealing ring failure or structural cracking, it is determined to be a Class C alarm, and the machine is immediately shut down and a maintenance prompt is output.

[0028] Preferably, it further includes an ultrasonic phased array linkage positioning step:

[0029] When the equivalent leakage rate At that time, the ultrasonic phased array sensor array arranged on the outer wall of the vacuum cavity is triggered to start collecting leakage acoustic emission;

[0030] The ultrasonic phased array sensor array includes N ultrasonic transducers evenly distributed circumferentially along the seam of the outer wall of the vacuum cavity, where N≥8. Each transducer is fixed to the non-heated area of ​​the outer wall of the cavity by magnetic attraction or high-temperature adhesive. The operating frequency band is 40kHz-100kHz.

[0031] Acoustic emission signals received by each transducer are collected, and vacuum pump harmonic interference and high-frequency noise from sheet cutting are filtered out by a digital band-stop filter.

[0032] Based on the arrival time difference of the acoustic emission signals from each transducer, a three-dimensional time difference localization algorithm is used to calculate the spatial coordinates of the leakage point;

[0033] When the same spatial coordinate is located in three consecutive molding cycles, the coordinate is marked as a confirmed leak source.

[0034] Preferably, the stopband frequency of the digital bandstop filter is automatically determined based on the vacuum pump model of the current thermoforming equipment and the motor speed of the sheet cutting mechanism. The stopband frequency includes at least the vacuum pump fundamental frequency and its first three harmonics, and the sheet cutting motor fundamental frequency and its first two harmonics.

[0035] A fully automatic vacuum chamber sealing detection system for thermoforming equipment is also provided, comprising:

[0036] The multi-source sensing module includes:

[0037] High-precision pressure transmitter installed on the vacuum chamber pumping line or chamber body, with an accuracy ≤0.1% of full scale and a range covering 0.1Pa-100kPa;

[0038] The NTC temperature probe array attached to the outer wall of the vacuum chamber contains at least four probes arranged around the perimeter of the chamber.

[0039] An ambient humidity sensor is mounted on the equipment rack;

[0040] The 8-16 channel ultrasonic phased array transducers are distributed around the outer wall and top and bottom of the vacuum chamber. Each transducer is fixed by magnetic attraction or high-temperature adhesive.

[0041] The edge computing unit includes an embedded industrial controller and a high-speed data acquisition card. The embedded industrial controller is electrically connected to the multi-source sensing module through the high-speed data acquisition card. The embedded industrial controller has embedded a flow adaptive segmentation module, a multi-physics compensation module, a leakage channel identification module, and a hierarchical response decision module.

[0042] The PLC interaction interface includes an industrial Ethernet communication module for real-time data interaction with the main programmable logic controller of the thermoforming equipment;

[0043] And human-computer interaction terminals, including touch screens.

[0044] Preferably, it further includes a self-calibration module, the self-calibration module comprising:

[0045] A standard leak port interface is installed on the vacuum chamber pumping line for connecting a standard leak port with an adjustable leak rate.

[0046] The calibration control unit is configured to close all external valves of the vacuum chamber and connect to the standard leak before the first mold of each shift, and perform a standard pressure decay test to calibrate the reference first decay slope under the current operating conditions. Reference second attenuation slope The calibration results are then updated to the operating parameter library of the edge computing unit, along with the tolerance threshold.

[0047] Preferably, the ultrasonic phased array transducer includes:

[0048] The first set of transducers, consisting of 4-8 units, is arranged along the joint between the upper and lower molds of the vacuum cavity to detect the sealing status of the mold joint surface.

[0049] The second set of transducers, consisting of 4-8 units arranged along the circumferential sidewall of the vacuum cavity and distributed at equal angular intervals, is used to detect the structural integrity of the cavity sidewall.

[0050] The edge computing unit is embedded with a transducer grouping selection module. The transducer grouping selection module is configured to: when the leakage channel identification module determines that the leakage type is a sealing ring failure, selectively enable the first group of transducers for precise positioning; when the leakage type is determined to be a structural crack, selectively enable the second group of transducers for scanning detection.

[0051] The technical effects and advantages of this invention are as follows:

[0052] 1. By performing second derivative analysis on the pressure curve, Automatically identify the transition moment between the molecular flow region and the viscous flow region at the inflection point from positive to negative. The pressure holding window is dynamically divided into two segments, and the attenuation slope is extracted for each segment. and Further utilize the ratio of the attenuation slopes of the two segments. As a characteristic identification index of leakage channels, the difference in attenuation slope between the molecular flow region and the viscous flow region directly reflects the geometric scale characteristics of the leakage channel. Different leakage channel morphologies correspond to different... The value range elevates leak detection from a binary judgment of "whether there is a leak" to the mechanistic level of "what kind of leak channel". It can accurately distinguish three types of leaks with completely different failure modes and degrees of harm: sealing ring compression deformation, joint micro-cracks, and material micropore penetration. This breaks through the limitations of traditional fixed-duration pressure holding detection.

[0053] 2. A coupled compensation model for sheet material, temperature, and humidity was established. The system automatically selects compensation coefficients based on different sheet materials such as ABS, PS, PET, and PP; and constructs a three-level differentiated response mechanism driven by the equivalent leakage rate and leakage type: Level A is qualified and production is maintained; Level B is a warning and micro-leakage is compensated by extending the pre-vacuuming time by 10%-30%; Level C is an alarm and the machine is stopped immediately. This achieves unified optimization of detection sensitivity and equipment uptime. Through the organic linkage of real-time compensation of multi-physics fields and graded differentiated response, a closed loop is achieved from accurate signal acquisition to intelligent matching of execution strategies.

[0054] 3. By using the leakage channel identification results as the basis for group selection, only the first group of transducers at the mold joint is activated when the sealing ring fails, and only the second group of transducers on the circumferential sidewall is activated when the structure cracks. The serial decision-making architecture of identification first and positioning later, combined with the digital band-stop filter for precise notch filtering of the vacuum pump fundamental frequency and its first three harmonics, and the cutting motor fundamental frequency and its first two harmonics, improves the positioning accuracy. At the same time, automatic self-calibration is performed before the first mold of each shift, and real-time calibration is performed through standard leak holes. , The system incorporates tolerance thresholds to compensate for threshold drift caused by aging of seals, ensuring consistent judgment criteria throughout the entire lifecycle of the detection system. This has significant industrial application value and creates a leakage type-driven ultrasonic phased array transducer grouping selection strategy, solving the problems of data redundancy and noise interference in existing full-array scanning solutions. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of the method flow structure of the present invention.

[0056] Figure 2 This is a schematic diagram of the core structure of the system of the present invention. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] As attached Figure 1As shown, the present invention provides an automatic detection method for the vacuum cavity sealing performance of a fully automatic thermoforming equipment, comprising the following steps:

[0059] Adaptive flow segmentation steps: During the pre-vacuuming stage of each forming cycle of the thermoforming equipment, at a sampling frequency... ≥100Hz synchronous acquisition of pressure values ​​inside the vacuum chamber Cavity wall temperature and ambient humidity ; at the end of vacuuming Enter the pressure holding observation window as the starting point ,in The value ranges from 5 seconds to 60 seconds, and the specific value is set according to the vacuum chamber volume and process requirements.

[0060] During the pressure-holding observation phase, if a leak exists in the vacuum chamber, external gas enters the chamber through the leak channel, causing the internal pressure to gradually rise and the vacuum level to decrease. Initially, due to the large pressure difference between the inside and outside of the chamber and the low gas density within the leak channel, the gas flow is in a molecular flow state, resulting in a rapid pressure rise rate. As the internal pressure gradually increases, the gas density increases, and the gas flow transitions to a viscous flow state, slowing the pressure rise rate. Correspondingly, the second derivative of pressure with respect to time... The inflection point when the flow changes from positive to negative is the transition moment between the molecular flow region and the viscous flow region. ;

[0061] For pressure value Perform second derivative analysis, through Automatic identification of the inflection point transition time between the molecular flow region and the viscous flow region The pressure holding observation window is divided into the first molecular flow attenuation region. Second viscous flow attenuation region The first attenuation slope was obtained by performing linear fitting on the two segments respectively. Second attenuation slope ;

[0062] In the flow-adaptive segmentation process, the automatic identification method for the transition time τ includes:

[0063] For pressure value Perform sliding window polynomial smoothing filtering; the sliding window polynomial smoothing filtering adopts the Savitzky-Golay filter, the filtering window width is 11-21 sampling points, and the polynomial order is 2-3.

[0064] Calculate the second derivative of the smoothed pressure sequence ;

[0065] by The last zero-crossing point from positive to negative is taken as the transition moment. ;

[0066] If no zero-crossing point meeting the conditions is detected within the entire pressure holding observation window, the midpoint of the pressure holding observation window is used as the default transition time. ;

[0067] Linear fitting was performed on the pressure data of the two attenuation zones to obtain the first attenuation slope. Molecular flow region and second attenuation slope In the viscous flow region, the least squares method was used for fitting, and the fitted straight line took the form of: ,in This is the attenuation slope, which is taken as a positive value.

[0068] It should be noted that the above-mentioned passage The method for identifying inflection point transitions in molecular / viscous flow is based on the following physical principle: Knudsen number. , The mean free path of gas molecules, The characteristic dimensions of the leakage channel determine the gas flow state; in the initial stage of pressure holding, the pressure inside the cavity is relatively low. Larger The gas is in a molecular flow state; as the pressure inside the cavity increases, Decrease Gradually decrease, when When the gas enters a viscous flow state, experiments show that... inflection point position and The transition points exhibit good consistency along the time axis; therefore, the inflection point of the second derivative can be used as the criterion for determining the transition moment.

[0069] Multiphysics Real-Time Compensation Steps: Based on Cavity Wall Temperature and ambient humidity The pressure value is calculated using a pre-established sheet material-temperature-humidity coupled compensation model. Real-time corrections are performed to obtain the compensated pressure sequence. Coupled compensation model The compensation coefficient group is automatically selected from the preset material mapping table based on the material of the current module sheet;

[0070] Coupled compensation model The construction method is as follows: ,in The temperature compensation factor is fitted using a quadratic polynomial. ; The humidity compensation factor is calculated using linear correction. , For reference humidity, 50% RH is typically used; temperature compensation factor. coefficients in and humidity compensation factor coefficients in Automatically select from the preset material mapping table based on the material of the current module sheet;

[0071] In the multiphysics field real-time compensation step, the preset material mapping table includes at least the compensation coefficient groups corresponding to the four commonly used blister packaging sheets: ABS, PS, PET, and PP.

[0072] Among them, the temperature compensation weights for PET and PP materials are higher than those for ABS and PS materials;

[0073] That is, the temperature compensation factors for PET and PP The absolute value is greater than that of ABS and PS. value;

[0074] The coefficients in the material mapping table were obtained through experimental calibration: on a vacuum calibration stage, the sheet to be calibrated was placed in a standard vacuum chamber, and the temperature of the chamber wall was controlled. The pressure was varied within the range of 25℃-120℃, and the ambient humidity H was controlled within the range of 30%RH-80%RH. A high-precision standard vacuum gauge was used to record the actual pressure value, and the output value of the pressure transmitter to be calibrated was also recorded. The compensation coefficients of each material were obtained by fitting the data through multiple regression analysis.

[0075] Different geometric shapes of leakage channels result in different attenuation characteristics in the molecular flow and viscous flow regions. Through extensive experiments, the inventors discovered that when the leakage channel is a compressed, flat, slit-like channel resembling a sealing ring, the molecular flow dominance time is longer. and The difference is significant, the ratio Falling into a higher range; when the leakage channel is a microcrack at the joint, or a slender capillary channel, the proportion of viscous flow increases rapidly, and the ratio... It falls into the medium range; when the leakage channel is through the material's micropores, and the micropores are evenly distributed, the gas is essentially in a molecular flow state throughout the entire pressure holding process due to the extremely small diameter of the micropores. Approaching ,ratio Approaching 1

[0076] In the leakage path feature model library, the ratio range corresponding to the sealing ring compression deformation model is: The ratio range corresponding to the microcrack model at the joint is: The ratio range corresponding to the material penetration model is: ,in And the intervals do not overlap;

[0077] The specific values ​​at each interval endpoint were obtained through the following calibration methods: simulating the compression deformation of the sealing ring by adjusting the sealing ring clamping force; simulating microcracks at the joint by machining microcracks of different sizes; simulating material penetration by setting up a micropore array with different pore sizes; and simulating three leakage modes by collecting a large amount of pressure decay data under each mode and statistically analyzing the data under each mode. The distribution range of the values ​​is used to determine the endpoints of each interval;

[0078] When the ratio When the value falls within the ratio range of the seal ring compression deformation model, the leakage type is determined to be seal ring failure;

[0079] When the ratio When the ratio falls within the range of the joint microcrack model, the leakage type is determined to be structural cracking;

[0080] When the ratio When the value falls within the ratio range of the material penetration model, the leakage type is determined to be material penetration;

[0081] Leakage channel feature identification steps: Calculate the first attenuation slope With the second attenuation slope ratio , the ratio The leak type is identified by comparing it with a pre-defined leak path feature model library. Leak types include at least seal ring failure, structural cracking, and material penetration.

[0082] Graded response steps: based on the compensated pressure sequence Calculate the equivalent leakage rate ,in For cavity volume, Standard atmospheric pressure; based on equivalent leakage rate Implement differentiated response strategies based on the type of leak;

[0083] Differentiated response strategies in the graded response process include:

[0084] when At that time, it was judged to be grade A qualified, and production was allowed to continue; The acceptable threshold is expressed in Pa·m. 3 / s, the calibration method is as follows: under the condition that the vacuum chamber is well sealed, perform no less than 30 pressure decay tests, and take 1.2 times the maximum leakage rate obtained from the test as the value. For typical volume The vacuum chamber, under the conditions of ABS sheet material and process vacuum degree ≤100Pa, The calibration range is to Pa·m 3 / s; for volume vacuum chamber, The calibration range is to Pa·m 3 / s; The specific value is obtained by standard leak tests based on the vacuum chamber volume, process vacuum requirements, and sheet material.

[0085] when Furthermore, if the leakage type is material penetration, it is classified as a Level B warning.

[0086] Alarm threshold, unit: Pa·m 3 / s can be calibrated in any of the following ways: (1) Usually taken as / s (2) Under the premise that the molding quality is acceptable, the process limit test is carried out by gradually increasing the simulated leakage rate through standard leaks to determine the critical leakage rate when the molding quality begins to show unacceptable defects. ,Pick ; after calibration It should be strictly greater than For typical volume vacuum chamber, The calibration range is to Pa·m 3 / s, the specific value is obtained through standard leak tests based on equipment process requirements; the current batch is marked and the next inspection cycle is shortened, while the pre-vacuuming time of the next molding cycle is automatically extended; the response principle of the Class B warning is: extending the pre-vacuuming time can make the initial vacuum level in the cavity lower, that is, the initial pressure Smaller size provides a greater time margin for the pressure to rise to the upper limit of the process during the pressure holding stage, thereby compensating for the impact of micro-leakage without shutting down the machine;

[0087] when If the leakage type is due to seal failure or structural cracking, it is judged as a Class C alarm, and the machine is immediately shut down and a maintenance prompt is output.

[0088] It also includes the ultrasonic phased array linkage positioning step:

[0089] When the equivalent leakage rate At that time, the ultrasonic phased array sensor array arranged on the outer wall of the vacuum cavity is triggered to start collecting leakage acoustic emission;

[0090] The ultrasonic phased array sensor array includes N ultrasonic transducers evenly distributed along the seam of the outer wall of the vacuum cavity and in the circumferential direction, where N≥8. Each transducer is fixed to the non-heated area of ​​the outer wall of the cavity by magnetic attraction or high-temperature adhesive. The operating frequency band is 40kHz-100kHz.

[0091] The transducer group selection module selectively activates transducer groups based on the judgment results of the leakage channel identification module: when the leakage type is sealing ring failure, the first group of transducers arranged along the mold joint is selectively activated for precise positioning; when the leakage type is structural cracking, the second group of transducers arranged along the circumferential sidewall of the vacuum cavity is selectively activated for scanning detection.

[0092] Acoustic emission signals received by each transducer are collected, and vacuum pump harmonic interference and high-frequency noise from sheet cutting are filtered out by a digital band-stop filter.

[0093] Based on the arrival time difference of the acoustic emission signals from each transducer, a three-dimensional time difference localization algorithm is used to calculate the spatial coordinates of the leakage point;

[0094] Specifically, let the time of the leak be... The spatial coordinates of the leak point are , No. The spatial coordinates of the transducers are The sound waves propagate from the leak point to the... The arrival time of each transducer is The speed of sound waves in air is Then it satisfies Using the first transducer as a reference, the coordinates of the leak point are solved by simultaneously solving the arrival time difference data of at least four transducers; speed of sound The temperature near the transducer is collected in real time by an NTC temperature probe and corrected, c = 331.3 + 0.606 × T, where T is the temperature in Celsius;

[0095] When the same spatial coordinate is located in three consecutive molding cycles, the coordinate is marked as a confirmed leak source.

[0096] The stopband frequency of the digital bandstop filter is automatically determined based on the vacuum pump model of the current thermoforming equipment and the motor speed of the sheet cutting mechanism. The stopband frequency includes at least the vacuum pump fundamental frequency and its first three harmonics, and the sheet cutting motor fundamental frequency and its first two harmonics.

[0097] As attached Figure 2 As shown, an automatic detection system for the vacuum cavity sealing performance of a fully automatic thermoforming equipment is also provided, comprising:

[0098] The multi-source sensing module includes:

[0099] High-precision pressure transmitter installed on the vacuum chamber pumping line or chamber body, with an accuracy ≤0.1% of full scale and a range covering 0.1Pa-100kPa;

[0100] The NTC temperature probe array attached to the outer wall of the vacuum chamber contains at least four probes arranged around the perimeter of the chamber.

[0101] An ambient humidity sensor is mounted on the equipment rack;

[0102] The 8-16 channel ultrasonic phased array transducers are distributed around the outer wall and top and bottom of the vacuum chamber. Each transducer is fixed by magnetic attraction or high-temperature adhesive.

[0103] The edge computing unit includes an embedded industrial controller and a high-speed data acquisition card. The embedded industrial controller is electrically connected to the multi-source sensing module through the high-speed data acquisition card. The embedded industrial controller has embedded a flow adaptive segmentation module, a multi-physics compensation module, a leakage path identification module, and a hierarchical response decision module.

[0104] The PLC interaction interface includes an industrial Ethernet communication module for real-time data interaction with the main programmable logic controller of the thermoforming equipment;

[0105] And human-computer interaction terminals, including touch screens.

[0106] It also includes a self-calibration module, which includes:

[0107] A standard leak port interface is installed on the vacuum chamber pumping line for connecting a standard leak port with an adjustable leak rate.

[0108] The calibration control unit is configured to close all external valves of the vacuum chamber and connect a standard leak before the first mold of each shift, and perform a standard pressure decay test to calibrate the baseline first decay slope under the current operating conditions. Reference second attenuation slope And tolerance thresholds, and update the calibration results to the operating parameter library of the edge computing unit;

[0109] The self-calibration process is as follows:

[0110] (1) Close the inlet valve, exhaust valve and pumping isolation valve of the vacuum chamber to isolate the vacuum chamber from the outside atmosphere and the vacuum pump;

[0111] (2) Open the calibration valve at the standard leak port interface and connect the standard leak port with a known leak rate to the vacuum chamber, so that the standard leak port becomes the only communication channel between the vacuum chamber and the external gas source.

[0112] (3) Control the vacuum pump to pump the vacuum chamber to the preset reference vacuum level through the pumping pipeline, and then close the isolation valve between the vacuum pump and the vacuum chamber to make the vacuum chamber enter the pressure holding state;

[0113] (4) Record pressure data under pressure holding conditions, and perform the same flow regime adaptive segmentation and slope fitting process as online detection to obtain the benchmark first attenuation slope under the current working condition. Reference second attenuation slope ;

[0114] (5) Based on the known standard leak rate and the measured... Calculate the tolerance threshold under the current operating conditions;

[0115] (6) Close the calibration valve at the standard leak interface to restore the vacuum chamber to normal working condition;

[0116] (7) The calibration results , The tolerance threshold is automatically updated to the operating parameter library of the edge computing unit, replacing the original reference value, and used as the leakage judgment benchmark for subsequent detection cycles.

[0117] The ultrasonic phased array transducer includes:

[0118] The first set of transducers, consisting of 4-8 units, is arranged along the joint between the upper and lower molds of the vacuum cavity to detect the sealing status of the mold joint surface.

[0119] The second set of transducers, consisting of 4-8 units arranged along the circumferential sidewall of the vacuum cavity and distributed at equal angular intervals, is used to detect the structural integrity of the cavity sidewall.

[0120] The edge computing unit has an embedded transducer grouping selection module. The transducer grouping selection module is configured as follows: when the leakage channel identification module determines that the leakage type is sealing ring failure, the first group of transducers is selectively activated for precise positioning; when the leakage type is determined to be structural cracking, the second group of transducers is selectively activated for scanning detection.

[0121] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An automatic detection method for the vacuum cavity sealing performance of a fully automatic thermoforming equipment, characterized in that, Includes the following steps: Adaptive flow segmentation: During the pre-vacuuming stage of each forming cycle in the thermoforming equipment, at a sampling frequency... ≥100Hz synchronous acquisition of pressure values ​​inside the vacuum chamber Cavity wall temperature and ambient humidity ; at the end of vacuuming Enter the pressure holding observation window as the starting point Regarding the pressure value Perform second derivative analysis, through Automatic identification of the inflection point transition time between the molecular flow region and the viscous flow region The pressure-holding observation window is divided into the first molecular flow attenuation region. Second viscous flow attenuation region The first attenuation slope was obtained by performing linear fitting on the two segments respectively. Second attenuation slope ; Real-time multiphysics compensation: based on the cavity wall temperature and the ambient humidity The pressure value is calculated using a pre-established compensation model that couples sheet material, temperature, and humidity. Real-time corrections are performed to obtain the compensated pressure sequence. The coupling compensation model The compensation coefficient group is automatically selected from the preset material mapping table based on the material of the current module sheet; Leakage path feature identification: Calculate the first attenuation slope With the second attenuation slope ratio The ratio The leakage type is identified by comparing it with a preset leakage path feature model library. The leakage types include sealing ring failure, structural cracking, and material penetration. Graded response: based on the compensated pressure sequence Calculate the equivalent leakage rate ,in For cavity volume, Standard atmospheric pressure; based on the aforementioned equivalent leakage rate Implement differentiated response strategies based on the type of leakage.

2. The automatic detection method for vacuum cavity sealing of a fully automatic thermoforming equipment according to claim 1, characterized in that: In the flow regime adaptive segmentation step, the automatic identification method for the transition time τ includes: Regarding the pressure value Perform sliding window polynomial smoothing filtering; Calculate the second derivative of the smoothed pressure sequence ; by The transition time is defined as the last zero-crossing point from positive to negative. ; If no zero-crossing point meeting the conditions is detected within the entire pressure holding observation window, the midpoint of the pressure holding observation window is used as the default transition time. .

3. The automatic detection method for vacuum cavity sealing of a fully automatic thermoforming equipment according to claim 1, characterized in that: In the multiphysics field real-time compensation step, the preset material mapping table includes at least the compensation coefficient groups corresponding to four commonly used blister packaging sheets: ABS, PS, PET, and PP. Among them, the temperature compensation weights for PET and PP materials are higher than those for ABS and PS materials.

4. The automatic detection method for vacuum cavity sealing of a fully automatic thermoforming equipment according to claim 1, characterized in that: In the leakage channel feature model library, the ratio range corresponding to the sealing ring compression deformation model is: The ratio range corresponding to the microcrack model at the joint is: The ratio range corresponding to the material penetration model is: ,in And the intervals do not overlap; When the ratio When the value falls within the ratio range of the seal ring compression deformation model, the leakage type is determined to be seal ring failure; When the ratio When the ratio falls within the range of the joint microcrack model, the leakage type is determined to be structural cracking; When the ratio When the value falls within the ratio range of the material penetration model, the leakage type is determined to be material penetration.

5. The automatic detection method for vacuum cavity sealing of a fully automatic thermoforming equipment according to claim 1, characterized in that: The differentiated response strategies in the hierarchical response steps include: when At that time, it was judged to be grade A qualified, and production was allowed to continue; among them, Equivalent leakage rate, unit: Pa·m 3 / s; The volume of the vacuum cavity is expressed in cubic meters (m³). 3 ; The pressure value after compensation is expressed in Pa. Standard atmospheric pressure, unit: Pa; The acceptable threshold is expressed in Pa·m. 3 / s; when Furthermore, when the leakage type is material penetration, it is determined to be a Level B warning, the current batch is marked and the next inspection cycle is shortened, while the pre-vacuuming time of the next molding cycle is automatically extended; wherein, Alarm threshold, unit: Pa·m 3 / s; when If the leakage type is a sealing ring failure or structural cracking, it is determined to be a Class C alarm, and the machine is immediately shut down and a maintenance prompt is output.

6. The automatic detection method for vacuum cavity sealing performance of a fully automatic thermoforming equipment according to claim 1, characterized in that: It also includes the ultrasonic phased array linkage positioning step: When the equivalent leakage rate At that time, the ultrasonic phased array sensor array arranged on the outer wall of the vacuum cavity is triggered to start collecting leakage acoustic emission; The ultrasonic phased array sensor array includes N ultrasonic transducers evenly distributed circumferentially along the seam of the outer wall of the vacuum cavity, where N≥8. Each transducer is fixed to the non-heated area of ​​the outer wall of the cavity by magnetic attraction or high-temperature adhesive. The operating frequency band is 40kHz-100kHz. Acoustic emission signals received by each transducer are collected, and vacuum pump harmonic interference and high-frequency noise from sheet cutting are filtered out by a digital band-stop filter. Based on the arrival time difference of the acoustic emission signals from each transducer, a three-dimensional time difference localization algorithm is used to calculate the spatial coordinates of the leakage point; When the same spatial coordinate is located in three consecutive molding cycles, the coordinate is marked as a confirmed leak source.

7. The automatic detection method for vacuum cavity sealing performance of a fully automatic thermoforming equipment according to claim 6, characterized in that: The stopband frequency of the digital band-stop filter is automatically determined based on the vacuum pump model of the current thermoforming equipment and the motor speed of the sheet cutting mechanism. The stopband frequency includes at least the vacuum pump fundamental frequency and its first three harmonics, and the sheet cutting motor fundamental frequency and its first two harmonics.

8. An automatic detection system for the vacuum cavity sealing performance of a fully automatic thermoforming equipment, characterized in that, include: The multi-source sensing module includes: High-precision pressure transmitter installed on the vacuum chamber pumping line or chamber body, with an accuracy ≤0.1% of full scale and a range covering 0.1Pa-100kPa; The NTC temperature probe array attached to the outer wall of the vacuum chamber contains at least four probes arranged around the perimeter of the chamber. An ambient humidity sensor is mounted on the equipment rack; The 8-16 channel ultrasonic phased array transducers are distributed around the outer wall and top and bottom of the vacuum chamber. Each transducer is fixed by magnetic attraction or high-temperature adhesive. The edge computing unit includes an embedded industrial controller and a high-speed data acquisition card. The embedded industrial controller is electrically connected to the multi-source sensing module through the high-speed data acquisition card. The embedded industrial controller has embedded a flow adaptive segmentation module, a multi-physics compensation module, a leakage channel identification module, and a hierarchical response decision module. The PLC interaction interface includes an industrial Ethernet communication module for real-time data interaction with the main programmable logic controller of the thermoforming equipment; And human-computer interaction terminals, including touch screens.

9. The automatic detection system for vacuum cavity sealing of a fully automatic thermoforming equipment according to claim 8, characterized in that: It also includes a self-calibration module, which comprises: A standard leak port interface is installed on the vacuum chamber pumping line for connecting a standard leak port with an adjustable leak rate. The calibration control unit is configured to close all external valves of the vacuum chamber and connect to the standard leak before the first mold of each shift, and perform a standard pressure decay test to calibrate the reference first decay slope under the current operating conditions. Reference second attenuation slope The calibration results are then updated to the operating parameter library of the edge computing unit, along with the tolerance threshold.

10. The automatic detection system for vacuum cavity sealing of a fully automatic thermoforming equipment according to claim 8, characterized in that: The ultrasonic phased array transducer includes: The first set of transducers, consisting of 4-8 units, is arranged along the joint between the upper and lower molds of the vacuum cavity to detect the sealing status of the mold joint surface. The second set of transducers, consisting of 4-8 units arranged along the circumferential sidewall of the vacuum cavity and distributed at equal angular intervals, is used to detect the structural integrity of the cavity sidewall. The edge computing unit is embedded with a transducer grouping selection module. The transducer grouping selection module is configured to: when the leakage channel identification module determines that the leakage type is a sealing ring failure, selectively enable the first group of transducers for precise positioning; when the leakage type is determined to be a structural crack, selectively enable the second group of transducers for scanning detection.

Images