Vacuum hot press mechanism and control method thereof

Abstract

This invention relates to the field of hot-press film application technology, specifically to a vacuum hot-press mechanism and its control method. The mechanism comprises a lower cavity and an upper cavity forming an openable, closed sealed cavity. A fixture mounting plate and a product fixture are disposed within the lower cavity, with the product fixture mounted on the mounting plate and equipped with a fixture heating structure. A linear guide rail and a rodless cylinder are located at the bottom of the lower cavity. A flexible pressure plate and a pressure plate mounting plate are disposed within the upper cavity. The flexible pressure plate is equipped with a pressure plate heating structure, and the pressure plate cylinder drives the flexible pressure plate to press downwards towards the product fixture. An electro-proportional valve is connected to the pressure plate cylinder to adjust its output pressure. A cavity lifting cylinder is connected to the upper cavity via a connecting rod assembly to drive the upper cavity to rise and fall, thereby opening and closing the upper and lower cavities. A linear bearing cooperates with a guide shaft to guide the lifting and lowering movement of the upper cavity. This invention can solve problems such as bubbles, poor adhesion, film rebound, product warping, or batch quality fluctuations in existing hot-press film application methods.

Description

Technical Field

[0001] This invention relates to the field of hot-press film application technology, and in particular to a vacuum hot-pressing mechanism and its control method. Background Technology

[0002] Lamination technology is widely used in the manufacturing of display modules, electronic substrates, packaging protection, and functional film bonding. It places high demands on the cleanliness, flatness, adhesive layer spread, and bonding consistency of the bonding surface. In existing lamination or hot-pressing processes, tiny air particles can easily be trapped between the film and the product surface, potentially leading to defects such as bubbles, bright spots, dark spots, and localized floating during subsequent heating or pressing. Furthermore, products of different sizes and with varying degrees of rigidity and flexibility may warp, exhibit thickness differences, or experience localized stress concentrations under pressure. If pressure, temperature, or decompression processes are not controlled consistently, uneven adhesive flow, insufficient adhesion, or film rebound can easily occur. For continuous production equipment, the operational status of components such as vacuuming, heating, and pressing directly affects bonding quality, and traditional methods relying on manual inspection or post-production maintenance struggle to detect abnormal trends in a timely manner. Summary of the Invention

[0003] In view of the above technical problems, the present invention provides a vacuum hot pressing mechanism and its control method, which aims to solve the problems of air bubbles, poor bonding, film rebound, product warping or batch quality fluctuations caused by air residue, uneven pressure, insufficient temperature and pressure stability and difficulty in timely identification of abnormal equipment operation in the existing film hot pressing process. It is especially suitable for product film application scenarios with high requirements for bubble-free bonding, uniform pressing and process stability.

[0004] Other features and advantages of the invention will become apparent from the following detailed description, or may be learned in part by practice of the invention.

[0005] According to one aspect of the present invention, a vacuum hot pressing mechanism is provided, the mechanism comprising a linear guide rail, a lower cavity, a fixture mounting plate, a product fixture, a rodless cylinder, an upper cavity, a flexible pressure plate, a pressure plate mounting plate, an electric proportional valve, a pressure plate cylinder, a linear bearing, a guide shaft, a connecting rod assembly, and a cavity lifting cylinder; The lower cavity and the upper cavity form an openable and closable sealed cavity. The fixture mounting plate and the product fixture are disposed in the lower cavity. The product fixture is mounted on the fixture mounting plate and is provided with a fixture heating structure for heating the product to be applied. The linear guide rail and the rodless cylinder are disposed at the bottom of the lower cavity. The rodless cylinder is used to drive the lower cavity to move along the linear guide rail so that the lower cavity switches between the waiting position and the hot pressing position. The flexible pressure plate and the pressure plate mounting plate are disposed in the upper cavity. The flexible pressure plate is provided with a pressure plate heating structure. The pressure plate cylinder is used to drive the flexible pressure plate to press down toward the product fixture so as to heat press and bond the film material on the surface of the product. The electro-proportional valve is connected to the pressure plate cylinder and is used to adjust the output pressure of the pressure plate cylinder; The cavity lifting cylinder is connected to the upper cavity through the connecting rod assembly and is used to drive the upper cavity to lift and lower, so that the upper cavity and the lower cavity can open and close. The linear bearing cooperates with the guide shaft to guide the lifting and lowering movement of the upper cavity. After the lower cavity moves below the upper cavity, the upper cavity descends and closes with the lower cavity to form a sealed cavity. After the sealed cavity is evacuated by a vacuum device, the flexible pressure plate presses down on the surface of the film material under heating and performs pressure-holding bonding.

[0006] Furthermore, the flexible pressure plate is an airbag. After the upper cavity descends and closes with the lower cavity to form a sealed cavity and a vacuum is drawn, and the flexible pressure plate is pressed down, the flexible pressure plate is filled with compressed air to flexibly pressurize the product to be laminated.

[0007] Furthermore, the product fixture is equipped with a retractable magnet to position and level the film-to-be-applied product, which has magnetic attraction properties.

[0008] Furthermore, the product fixture is a vacuum suction fixture.

[0009] Furthermore, the flexible pressure plate is a rubber-coated plate.

[0010] According to another aspect of the present invention, a method for controlling a vacuum hot pressing mechanism is provided, the method being performed on a vacuum hot pressing mechanism as described above, the method comprising steps S1 to S8: S1. Obtain the product type, film material type, adhesive layer type, product size, pre-applied film status, and hot-pressing bonding quality requirements of the product to be laminated, and call the corresponding vacuum hot-pressing preset program according to the product type and film material type. The vacuum hot-pressing preset program includes at least the vacuum degree target range, temperature target range, pressure target range, pressing stage sequence, pressure holding conditions, pressure release conditions, sampling cycle, abnormal judgment threshold group, and equipment health status judgment rules. S2. Load the product to be coated with the pre-applied film material into the product fixture, and control the heating structure of the fixture and the heating structure of the pressure plate to preheat according to the vacuum hot pressing preset program; S3. Control the rodless cylinder to drive the lower cavity to move along the linear guide rail to below the upper cavity, and control the cavity lifting cylinder to drive the upper cavity to descend along the guide shaft through the connecting rod assembly, so that the upper cavity and the lower cavity close to form the sealed cavity. Detect the sealing status of the sealed cavity, and stop the subsequent hot-pressing bonding process or perform sealing compensation when the sealing status does not meet the preset sealing conditions. S4. After the sealed cavity is formed, the vacuum device is activated to bring the vacuum level in the sealed cavity into the target vacuum level range, and the vacuum level change, temperature change and pressure preparation status are continuously monitored during the vacuuming process; when the vacuum level is detected to deviate from the target vacuum level range, the vacuum device is controlled to perform air extraction compensation or maintenance control according to the direction of deviation, so that the residual gas between the film material and the product to be laminated is discharged before hot pressing; S5. Collect multiphase current data of at least one monitored device according to the sampling period. The monitored device includes at least one of the fixture heating structure, the pressure plate heating structure, the vacuum device, the heating adjustment component, and the pressure plate cylinder. Perform rolling time window processing on the multiphase current data and obtain a feature set based on the multiphase current data. The feature set includes phase current balance characteristics, phase difference characteristics, change characteristics between the current sampling time and the historical sampling time, equipment start-up characteristics, stable operation characteristics, and shutdown characteristics. S6. Input the feature set into the anomaly monitoring process. First, determine whether the monitored equipment is in the start-up, stable operation, or termination state. Then, determine whether the monitored equipment has a current anomaly exceeding the anomaly judgment threshold group. When the current anomaly belongs to a stored historical anomaly rule and does not affect the current hot-press bonding quality, record the current anomaly as a traceable anomaly and continue to execute the current process. When the current anomaly belongs to an anomaly affecting vacuum, heating, or pressing stability, output a warning, alarm, or shutdown control signal. When the current anomaly does not belong to the historical anomaly rule, input the corresponding operating status data into a machine learning classification model for classification, and update the classification result, manual confirmation result, or unsupervised clustering result to the anomaly rule library. S7. When the vacuum degree, the temperature of the product fixture, the temperature of the flexible pressure plate, and the health status of the monitored equipment all meet the pressing allowable conditions, the pressure plate cylinder is controlled to drive the flexible pressure plate to move toward the product fixture, and the output pressure of the pressure plate cylinder is adjusted according to the preset pressure curve by the electric proportional valve, so that the flexible pressure plate performs pre-pressing, holding pressure and releasing pressure on the surface of the membrane material in sequence under heating state. S8. After hot pressing and bonding are completed, the flexible pressure plate is controlled to retract according to the vacuum hot pressing preset program, and the sealed cavity is subjected to gradient depressurization and vacuum breaking treatment; then the upper cavity is controlled to rise, the lower cavity is controlled to move to the waiting position, and a process record containing process parameter curves, abnormal records, equipment health status, early warning information and bonding batch information is generated. At the same time, the process record is used to update the abnormal rule base and the machine learning classification model.

[0011] Furthermore, step S5 specifically includes: According to a preset time interval, the multiphase current data of the monitored device is obtained from the current acquisition module or the power metering module, and the multiphase current data within the same rolling time window is divided into a data group to be judged. For the data set to be determined, extract the representative value of each phase current, the mutual difference between each phase current, the comprehensive representative value of the difference between each phase current, the maximum representative value of the difference between each phase current, and the current change amplitude between adjacent sampling windows. The monitored equipment is judged to have entered the start-up state based on the representative value of each phase current, and to have phase imbalance based on the mutual differences and the comprehensive representative value. The monitored equipment is judged to have sudden abnormality based on the current change amplitude, and to have normal shutdown based on the downward trend of each phase current in the end stage. When the mutual difference is at a first preset imbalance level, the corresponding data will be included in the early warning score; when the mutual difference is at a second preset imbalance level higher than the first preset imbalance level, the corresponding data will be included in the alarm score; when the current change exceeds a preset sudden change level, the corresponding data will be directly included in the serious abnormality record. Based on the warning score, the alarm score, and the serious anomaly record, a status identifier corresponding to the monitored equipment is generated. The status identifier includes at least one of normal status, warning status, alarm status, and fault status. The status identifier is associated with the process stage in the vacuum hot pressing bonding process to distinguish the different abnormal meanings corresponding to the starting current, stable operating current, temperature adjustment current, vacuum maintenance current, and shutdown current.

[0012] Furthermore, step S6 also includes: A first monitoring process is established to identify the current operating status of the monitored device. The first monitoring process determines whether the monitored device has entered the start-up state from the standby state based on whether the multiphase current data exceeds the device start-up judgment condition. After the monitored device enters the start-up state, the operating current is continuously monitored for a preset duration to see if it returns to the stable operating range. A second monitoring process is established for finding the current status. When the operating current deviates from the stable operating range, the second monitoring process calls the anomaly rule base to match the current anomaly. When the current anomaly can be matched with a known anomaly rule, a rule matching process is executed. The rule matching process determines whether the known anomaly rule belongs to a traceable anomaly that does not affect the vacuum maintenance of the sealed cavity, the temperature stability of the heating structure, or the pressing stability of the flexible platen. If it belongs to the traceable anomaly, the anomaly category, the time of occurrence, the corresponding monitored equipment, and the current process stage are recorded, and the hot pressing is allowed to continue. If it does not belong to the traceable anomaly, a warning, alarm, pause, or shutdown control signal is output. When the current anomaly cannot be matched with the known anomaly rule, an unruled confirmation process is executed. The unruled confirmation process uploads the anomaly data to the information processing terminal, uses unsupervised learning to cluster the anomaly samples to determine whether the anomaly samples form a new anomaly category, and writes the new anomaly category, the corresponding process stage, the cause of the anomaly, and the handling strategy into the anomaly rule library after manual confirmation. After completing the rule matching process or the ruleless confirmation process, the monitoring state is returned to the continuous monitoring state so that current status judgment, abnormal current diversion and equipment health status update can continue to be performed for subsequent sampling windows.

[0013] Furthermore, the machine learning classification model is established in the following manner: Collect multiphase current data, temperature data, vacuum data, pressure data, equipment operation data, abnormal records and maintenance records from the historical hot-pressing bonding process, and divide the historical hot-pressing bonding process into the feeding and preheating stage, vacuuming stage, degassing stage, pre-pressing stage, pressure holding stage, pressure release stage, vacuum breaking stage and unloading stage according to the corresponding process stages. Training features are extracted from the historical hot-pressing bonding process. These training features include representative values ​​of current in each phase, representative values ​​of phase-to-phase differences, maximum representative values ​​of phase-to-phase differences, current variation trends, temperature variation trends, vacuum variation trends, pressure variation trends, equipment start-up indicators, equipment stable operation indicators, equipment shutdown indicators, and process stage indicators. Ineffective samples formed under long-term standby, no-load, or low-load conditions are filtered, downweighted, or grouped so that the training features can reflect the effective operating status during the hot-press bonding process. Configure normal labels, early warning labels, alarm labels or fault labels for the training features based on historical anomaly records and maintenance records, and train using at least one of support vector machine model and gradient boosting tree model; When using the support vector machine model, the training features are classified into states by constructing a classification boundary that can distinguish between normal samples and abnormal samples with the largest interval. When using the gradient boosting tree model, an initial weak classification tree is first established, and then the next weak classification tree is generated based on the deviation between the previous classification result and the label. Continuous prediction output is obtained by accumulating the results of multiple weak classification trees. Then, the continuous prediction output is mapped to the probability of different state categories to obtain the classification results of normal, early warning, alarm or fault. New samples obtained from online monitoring, manual confirmation results, and actual maintenance results are periodically added to the training data, enabling the machine learning classification model to be updated according to the aging state of the monitored equipment, changes in membrane material, and changes in preset programs.

[0014] Furthermore, step S7 also includes: Select at least one process mode from the following, based on the film material type and adhesive layer type: thermoplastic adhesive layer process, thermosetting adhesive layer process, rigid substrate process, flexible substrate process, or high-density substrate process. In the thermoplastic adhesive layer process, the temperature change trend of the flexible pressure plate and the product fixture, as well as the pressure holding conditions, are preferentially controlled so that the adhesive layer obtains a state suitable for flow and spreading after being heated. In the thermosetting adhesive layer process, the constant temperature holding state, the pressure holding state, and the pressure release rhythm are preferentially controlled to ensure that the adhesive layer forms a stable cured bond between the film material and the product to be laminated. In the rigid substrate process, a pressure target range is configured according to a higher load-bearing capacity to improve the flatness after bonding; In the flexible substrate process, the preset pressure curve is configured with a low pressing impact, and the elastic deformation of the flexible platen is used to compensate for the local warping and thickness difference of the product to be laminated. In the high-density substrate process, priority is given to maintaining the vacuum stability of the sealed cavity and the smoothness of the pressure release process to reduce the risk of pore blockage, bubble residue and membrane rebound. When the vacuum level is detected to deviate from the target vacuum level range, the pumping or holding state of the vacuum device is adjusted; when the temperature is detected to deviate from the target temperature range, the heating power of the fixture heating structure and the pressure plate heating structure is adjusted; when the pressure is detected to deviate from the target pressure range, the output pressure of the pressure plate cylinder is adjusted through the electro-proportional valve; when the multiphase current data indicates that the fixture heating structure, the pressure plate heating structure, the vacuum device, or the pressure plate cylinder has an operational risk, the rate of change of the corresponding action is reduced, the stability confirmation process is extended, or the current process stage is paused.

[0015] The technical solution of the present invention has the following beneficial effects: Compared with existing technologies, this invention uses an openable and closable sealed cavity to expel residual gas between the film material and the product before pressing, ensuring a stable vacuum environment during the bonding process, which helps reduce air bubbles and floating defects. By heating the product side and the pressing side separately, the adhesive layer maintains a suitable flow or curing state before and after pressing, thereby improving the wetting consistency and adhesion reliability of the film material and the product surface. The flexible pressing method compensates for local warping, thickness differences, and pressing plane errors on the product surface, resulting in a more uniform bonding pressure distribution and reducing the risk of local overpressure, underpressure, and film rebound. Through pressure regulation, staged pressing, pressure holding, and gradient pressure release, different film materials, adhesive layers, and product sizes can be matched with corresponding process windows. By collecting and analyzing the operating data of heating, vacuum, and pressing-related equipment, the invention can identify start-up, stable operation, abnormal fluctuations, and shutdown states during the bonding process. Anomalies that do not affect the current quality are tracked and recorded, and anomalies affecting vacuum, temperature, or pressing stability are promptly warned of or require shutdown, thereby improving bonding yield, process traceability, and equipment maintenance efficiency. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the structure of a vacuum hot pressing mechanism in one of the embodiments of this specification; Figure 2 A schematic diagram of a vacuum hot pressing mechanism in one of the embodiments of this specification; Figure 3 This is a flowchart of a vacuum hot pressing mechanism control method in one of the embodiments of this specification.

[0017] Explanation of reference numerals in the attached drawings: 1. Linear guide rail; 2. Lower cavity; 3. Fixture mounting plate; 4. Product fixture; 5. Rodless cylinder; 6. Upper cavity; 7. Flexible pressure plate; 8. Pressure plate mounting plate; 9. Electro-proportional valve; 10. Pressure plate cylinder; 11. Linear bearing; 12. Guide shaft; 13. Connecting rod assembly; 14. Cavity lifting cylinder. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided to make the invention more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of the invention. However, those skilled in the art will recognize that the technical solutions of the invention may be practiced with one or more of these specific details omitted, or other methods, components, systems, steps, etc., may be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of the invention.

[0019] Furthermore, the accompanying drawings are merely illustrative of the invention. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor systems and / or microcontroller systems.

[0020] Reference Figures 1 to 2 As shown, a vacuum hot pressing mechanism is provided in an embodiment of the present invention. The mechanism includes a linear guide rail 1, a lower cavity 2, a fixture mounting plate 3, a product fixture 4, a rodless cylinder 5, an upper cavity 6, a flexible pressure plate 7, a pressure plate mounting plate 8, an electric proportional valve 9, a pressure plate cylinder 10, a linear bearing 11, a guide shaft 12, a connecting rod assembly 13, and a cavity lifting cylinder 14. The lower cavity 2 and the upper cavity 6 form an openable and closable sealed cavity. The fixture mounting plate 3 and the product fixture 4 are disposed in the lower cavity 2. The product fixture 4 is mounted on the fixture mounting plate 3 and is equipped with a fixture heating structure for heating the product to be applied. Linear guide rail 1 and rodless cylinder 5 are located at the bottom of lower cavity 2. Rodless cylinder 5 is used to drive lower cavity 2 to move along linear guide rail 1 so that lower cavity 2 can switch between waiting position and hot pressing position. The flexible pressure plate 7 and the pressure plate mounting plate 8 are disposed in the upper cavity 6. The flexible pressure plate 7 is provided with a pressure plate heating structure. The pressure plate cylinder 10 is used to drive the flexible pressure plate 7 to press down in the direction of the product fixture 4 so as to heat press the film material on the surface of the product. The electric proportional valve 9 is connected to the pressure plate cylinder 10 and is used to adjust the output pressure of the pressure plate cylinder 10; The cavity lifting cylinder 14 is connected to the upper cavity 6 via the connecting rod assembly 13 and is used to drive the upper cavity 6 to lift and lower, so that the upper cavity 6 and the lower cavity 2 can open and close. The linear bearing 11 cooperates with the guide shaft 12 to guide the lifting and lowering movement of the upper cavity 6. After the lower cavity 2 moves below the upper cavity 6, the upper cavity 6 descends and closes with the lower cavity 2 to form a sealed cavity. After the vacuum device completes the vacuuming of the sealed cavity, the flexible pressure plate 7 presses down on the surface of the film material under heating and performs pressure-holding bonding.

[0021] In one embodiment, the flexible pressure plate 7 is an airbag. After the upper cavity 6 descends and closes with the lower cavity 2 to form a sealed cavity, and a vacuum is drawn and the flexible pressure plate 7 is pressed down, the flexible pressure plate 7 is filled with compressed air to flexibly pressurize the product to be laminated. Specifically, after the upper and lower cavities 2 are closed and vacuum heat-pressed, compressed air is injected into the airbag at a specific time point, which causes the airbag to deform and bulge, pressing the surface of the film material and achieving full contact and bonding between the film material and the product. The airbag form mainly corresponds to products without primer coating + UV film.

[0022] In one embodiment, the product fixture 4 is provided with a retractable magnet to position and level the product to be coated with film, which has magnetic attraction properties.

[0023] In one embodiment, the product fixture 4 is a vacuum suction fixture. When there is no magnetic strip available for the product to be coated, the vacuum suction fixture needs to be replaced to attract the product.

[0024] In one embodiment, the flexible pressure plate 7 is a coated plate. The coated plate mainly corresponds to a product with a primer coating and a thermosetting film.

[0025] In one embodiment. Refer to Figure 3 The diagram shown is a schematic flow chart of a vacuum hot pressing mechanism provided in an embodiment of the present invention. The method may specifically include the following steps S101~S106: In step S1, the product type, film type, adhesive layer type, product size, pre-applied film status, and hot-pressing bonding quality requirements of the product to be laminated are obtained. The corresponding vacuum hot-pressing preset program is called according to the product type and film type. The vacuum hot-pressing preset program includes at least the vacuum degree target range, temperature target range, pressure target range, pressing stage sequence, pressure holding conditions, pressure release conditions, sampling cycle, abnormal judgment threshold group, and equipment health status judgment rules.

[0026] The control system acquires the product type, film material type, adhesive layer type, product size, pre-application status, and hot-pressing quality requirements of the product to be laminated, and uses this information as input for process selection and parameter configuration. Specifically, the control system can be a controller. The product type distinguishes the substrate rigidity / flexibility, dimensional range, heat and pressure bearing capacity, and surface flatness of the product to be laminated; the film material type distinguishes the film thickness, flexibility, heat deformation characteristics, and bonding surface compatibility requirements; the adhesive layer type distinguishes the flow, wetting, or curing characteristics of the adhesive layer under heat; the product size determines the lamination area, pressure area, and corresponding pressure setting range; the pre-application status determines whether the film material and product surface have achieved initial alignment and whether there are any localized lifts or residual air; the hot-pressing quality requirements determine requirements for bubble control, lamination flatness, adhesion stability, and batch consistency. The control system calls the corresponding vacuum hot-pressing preset program based on the product type and film material type, enabling different products to enter a hot-pressing process that matches their structural and film material characteristics. The vacuum hot pressing preset program includes target vacuum range, target temperature range, target pressure range, pressing stage sequence, pressure holding conditions, pressure release conditions, sampling cycle, anomaly judgment threshold group, and equipment health status judgment rules. The target vacuum range limits the vacuum control range of the sealed cavity before and during pressing to remove residual gas between the membrane material and the product to be bonded. The target temperature range limits the heating control range of the product fixture and flexible pressure plate to ensure the adhesive layer is in a suitable state for flow, wetting, or curing. The target pressure range limits the allowable pressure applied by the flexible pressure plate to the membrane surface and matches it with the size and type of the product to be bonded. The pressing stage sequence determines the execution order of process stages such as preheating, vacuuming, pre-pressing, pressure holding, pressure release, and vacuum breaking. The pressure holding conditions limit the temperature, pressure, vacuum level, and duration of the pressure holding stage to ensure they meet requirements before proceeding to the next stage. The pressure release conditions control the pressure release and vacuum breaking process to prevent membrane rebound, local desorption, or gas re-entry into the bonding interface due to sudden changes in pressure or vacuum. The sampling period determines the frequency of acquiring operational data from the monitored equipment, enabling subsequent assessment of equipment status based on continuous sampling data. Anomaly detection thresholds are used to determine the degree of deviation in vacuum, temperature, pressure, and multiphase current data. Multiphase current data reflects the operational status of heating, vacuum, and pressing-related actuators during startup, stable operation, and termination. Equipment health status determination rules are used to determine whether the monitored equipment is in a normal, warning, alarm, or fault state based on operational data, anomaly records, and process stage information within each sampling period, providing a basis for determining whether subsequent entry into the hot-pressing bonding stage is permitted.

[0027] In step S2, the product to be coated with film pre-applied material is loaded into the product fixture, and the fixture heating structure and the pressure plate heating structure are controlled to preheat according to the vacuum hot pressing preset program.

[0028] The process involves placing the pre-applied film-coated product into the support position of the product fixture, ensuring the film-coating surface faces the flexible pressure plate and maintaining stable contact between the product and the fixture. During loading, the product is positioned according to its shape, size, and pre-applied film status, aligning with the preset vacuum hot-press program to prevent displacement during subsequent lower cavity movement, vacuuming, and pressing. The product fixture supports and limits the product after loading, ensuring a flat bonding area and providing a stable foundation for subsequent heat conduction and flexible pressing.

[0029] After the film-to-be-applied product is loaded, the control system controls the fixture heating structure and the pressure plate heating structure to enter the preheating state according to the preset vacuum hot pressing program. The fixture heating structure is used to transfer heat from the side of the product to be applied to the bonding area, and the pressure plate heating structure is used to provide heat from the surface of the film material, so that the film material, adhesive layer, and product to be applied gradually approach the target temperature range before entering the vacuuming and pressing stage. During the preheating process, the control system continuously acquires the temperature change status of the fixture heating structure and the pressure plate heating structure and compares the actual temperature with the target temperature range; when the temperature has not yet reached the preset requirements, heating continues or the heating power is adjusted; when the temperature enters the target temperature range and meets the stability conditions, the temperature preparation state for the subsequent closed cavity and vacuuming process is formed.

[0030] The preheating control of the fixture heating structure and the pressure plate heating structure is matched with the film material type, adhesive layer type, and product size. For adhesive layers that need to form a flowing and spreading state after heating, the preheating process softens the adhesive layer to a state suitable for subsequent pressing. For adhesive layers that need to form a stable bond under constant temperature conditions, the preheating process establishes a uniform temperature base for the bonding area of ​​the film material and the product to be laminated. Since the current characteristics of the heating equipment are different during the start-up and stable operation phases, the control system can record the operating data of the fixture heating structure and the pressure plate heating structure according to the sampling period during the preheating process. This data is used to identify whether the heating components are in a normal start-up, stable heating, or abnormal fluctuation state, thereby avoiding entering the hot pressing bonding process when the temperature preparation state is abnormal.

[0031] In step S3, the rodless cylinder is controlled to drive the lower cavity to move along the linear guide rail to below the upper cavity, and the cavity lifting cylinder is controlled to drive the upper cavity to descend along the guide shaft through the connecting rod assembly, so that the upper cavity and the lower cavity close to form a sealed cavity. The sealing status of the sealed cavity is detected, and if the sealing status does not meet the preset sealing conditions, the subsequent hot pressing bonding process is stopped or sealing compensation is performed.

[0032] The control system, after confirming that the product to be coated has been loaded and that the fixture heating structure and pressure plate heating structure are in preheated state, outputs a movement control signal to the rodless cylinder. This causes the rodless cylinder to drive the lower cavity along a linear guide rail from the waiting position to below the upper cavity. The linear guide rail restricts the movement direction of the lower cavity and ensures the repeatability of position switching, ensuring that the product to be coated on the fixture corresponds to the flexible pressure plate in the upper cavity. After the lower cavity moves into position, a positioning detection signal, a stroke detection signal, or a position confirmation signal can be used to determine whether the lower cavity has reached the hot-pressing position. If the hot-pressing position has not been reached, the control system does not allow the upper cavity to descend to avoid misalignment and closure of the upper and lower cavities.

[0033] After the lower cavity reaches the hot-pressing position, the control system outputs a descent control signal to the cavity lifting cylinder. The cylinder, via a connecting rod assembly, drives the upper cavity downwards. Guided by linear bearings and a guide shaft, the upper cavity descends, gradually merging with the lower cavity to form a sealed cavity. The connecting rod assembly transmits the driving force of the lifting cylinder and ensures smooth descent of the upper cavity. The guide shaft limits the lifting path of the upper cavity, and the linear bearing reduces sway and friction during lifting, ensuring good parallelism between the upper and lower cavities. During the descent of the upper cavity, the control system can confirm the lifting position, closing pressure, and cavity closure position to avoid initiating vacuuming or pressing actions before the cavity is fully closed.

[0034] After the upper and lower cavities are closed, the control system detects the sealing status of the sealed cavities. The sealing status can be judged by the vacuum establishment, vacuum maintenance, seal compression, or cavity closure. During detection, an initial negative pressure is established in the sealed cavity, and the vacuum change is monitored within a preset time. If the vacuum level enters the preset sealing judgment range and the change range meets the requirements, the sealing status is judged to meet the preset sealing conditions. If the vacuum level drops too quickly, fails to reach the preset negative pressure, the cavity closure position is abnormal, or the seal compression is abnormal, the sealing status is judged to not meet the preset sealing conditions. Vacuum change is a fundamental judgment condition for subsequent vacuuming and hot-pressing bonding stability. Only when the sealed cavity has a reliable sealing status can the residual gas between the film material and the product to be laminated be stably discharged.

[0035] When the sealing condition does not meet the preset sealing conditions, the control system stops proceeding to the subsequent hot-press bonding process to avoid performing vacuuming, heating, and pressure holding actions under conditions of leakage or poor closure. The control system can also perform seal compensation based on the type of anomaly. Sealing compensation may include reconfirming the lower cavity's positioning, adjusting the upper cavity's closed position, re-driving the upper cavity to press the sealing surface, extending the seal confirmation time, controlling the vacuum device for short-term evacuation compensation, or issuing a seal anomaly alert to check the sealing surface, sealing components, and product loading status. Only after the sealing condition meets the preset sealing conditions will the control system allow subsequent vacuuming and hot-press bonding processes to proceed, thus ensuring that vacuum degassing and flexible pressing occur in a stable cavity environment.

[0036] In step S4, after the sealed cavity is formed, the vacuum device is activated to bring the vacuum level in the sealed cavity into the target vacuum level range. During the vacuuming process, the changes in vacuum level, temperature, and pressure preparation status are continuously monitored. When the vacuum level is detected to deviate from the target vacuum level range, the vacuum device is controlled to perform gas extraction compensation or maintain control according to the direction of deviation, so that the residual gas between the film material and the product to be laminated is discharged before hot pressing.

[0037] After the sealed cavity is closed and meets the sealing conditions, the control system activates the vacuum device to evacuate the cavity, gradually bringing the vacuum level within the cavity into the target range. The vacuuming process reduces the gas content within the cavity before hot pressing and facilitates the removal of residual gas between the film material and the product to be bonded from the bonding interface. This prevents residual gas from expanding due to heat during subsequent heating and pressing, which could lead to air bubbles, loose bonding, or poor localized adhesion.

[0038] During the vacuuming process, the control system continuously monitors changes in vacuum level, temperature, and pressure readiness. Vacuum level changes are used to determine the rate of negative pressure build-up and maintenance within the sealed cavity, as well as the presence of leaks or insufficient evacuation. Temperature changes are used to determine whether the fixture heating structure and pressure plate heating structure are still within the target temperature range, preventing entry into the pressing stage before the temperature is stable. Pressure readiness status is used to determine whether the pressure plate cylinder, electro-proportional valve, and flexible pressure plate are ready to perform subsequent pre-pressing and pressure-holding actions. All of these statuses serve as process confirmation conditions before entering the hot-press bonding process, ensuring that vacuum degassing, temperature readiness, and pressing readiness are matched.

[0039] When the detected vacuum level is lower than the vacuum requirement corresponding to the target vacuum range, the control system controls the vacuum device to continue pumping or increase the pumping action to compensate for the vacuum in the sealed cavity. When the detected vacuum level has entered the target vacuum range, the control system controls the vacuum device to enter a holding control state, keeping the sealed cavity in a stable vacuum environment. When the detected vacuum level deviates from the target direction, indicating a decrease in vacuum holding capacity, the control system can restart the pumping compensation and determine whether there is a sealing abnormality based on the vacuum level change trend. If the vacuum level cannot be restored to the target vacuum range within a preset time, the control system can restrict the subsequent pressing stage to avoid hot pressing with residual gas not being fully discharged.

[0040] Once the vacuum level is reached and maintained within the target range, air between the film material and the product to be laminated is pre-expelled, allowing the adhesive layer to flow, wet, or cure more evenly during subsequent heating and pressure treatment. By continuously confirming the vacuum, temperature, and pressure readiness status before lamination, the risks of residual air bubbles, discontinuous bonding interfaces, and localized stress concentrations can be reduced, and stable process conditions can be provided for the subsequent pre-pressing, holding, and depressurization of the flexible laminator.

[0041] In step S5, multiphase current data of at least one monitored device is collected according to the sampling period. The monitored device includes at least one of the following: a fixture heating structure, a pressure plate heating structure, a vacuum device, a heating adjustment component, and a pressure plate cylinder. The multiphase current data is processed by rolling time window, and a feature set is obtained based on the multiphase current data. The feature set includes phase current balance characteristics, phase difference characteristics, change characteristics between the current sampling time and the historical sampling time, equipment start-up characteristics, stable operation characteristics, and shutdown characteristics.

[0042] Step S5 specifically includes: acquiring multiphase current data of the monitored equipment from the current acquisition module or the power metering module at preset time intervals, and dividing the multiphase current data within the same rolling time window into a data group to be judged; extracting the representative value of each phase current, the mutual difference between each phase current, the comprehensive representative value of the difference between each phase current, the maximum representative value of the difference between each phase current, and the current change amplitude between adjacent sampling windows for each data group to be judged; determining whether the monitored equipment has entered the start-up state based on the representative value of each phase current, determining whether there is phase imbalance in the monitored equipment based on the mutual difference and the comprehensive representative value, determining whether there is a sudden abnormality in the monitored equipment based on the current change amplitude, and determining the downward trend of each phase current in the final stage. Whether the monitored equipment has stopped normally; when the difference between them is at the first preset imbalance level, the corresponding data will be included in the early warning score; when the difference between them is at the second preset imbalance level, which is higher than the first preset imbalance level, the corresponding data will be included in the alarm score; when the current change exceeds the preset sudden change level, the corresponding data will be directly included in the serious anomaly record; based on the early warning score, alarm score and serious anomaly record, a status identifier corresponding to the monitored equipment will be generated. The status identifier includes at least one of the following: normal status, early warning status, alarm status and fault status. The status identifier will be associated with the process stage in the vacuum hot pressing bonding process to distinguish the different abnormal meanings corresponding to the starting current, stable operating current, temperature adjustment current, vacuum maintenance current and shutdown current.

[0043] The control system collects multiphase current data from the monitored equipment according to a sampling period. The monitored equipment can be one or more of the following: a fixture heating structure, a pressure plate heating structure, a vacuum device, a heating power adjustment component, and a pressure plate cylinder. Multiphase current data can be provided by a current acquisition module or an energy metering module, and the collected data includes at least the R-phase current, S-phase current, and T-phase current of the monitored equipment at the same sampling time. The control system organizes the multiphase current data acquired within the same rolling time window into a set of data to be judged, so as to use continuous data to determine the operating status of the monitored equipment in the current process stage. The rolling time window can cover multiple continuous sampling points, ensuring that fluctuations in a single sampling do not directly determine abnormal results, and enabling continuous identification of process characteristics such as startup, stable operation, temperature adjustment, vacuum maintenance, and shutdown.

[0044] For the data set to be judged, the control system extracts the representative value of each phase current, the mutual difference between phase currents, the comprehensive representative value of the phase current differences, the maximum representative value of the phase current differences, and the current change amplitude between adjacent sampling windows. The representative value of each phase current can be the average value, maximum value, or stable segment representative value of the corresponding phase current within the rolling time window. Let R, S, and T represent the representative values ​​of the three-phase currents in the same data set to be judged, and the average representative value of the three-phase currents can be calculated using the following formula.

[0045] ; in, Used to characterize the overall current level of the monitored equipment within the current rolling time window, it can serve as a benchmark value for determining whether the equipment has started up, is operating stably, and whether the phase-to-phase difference exceeds the allowable proportion. For loads such as heating components and vacuum devices, The lower value corresponds to standby, no-load, or stop states. When the device startup criteria are exceeded, it can be classified as either startup or running.

[0046] The comprehensive representative value of the current difference between each phase can be calculated using the following formula: ; Here, Y represents the degree of imbalance between the three-phase currents. A larger Y indicates a more significant difference in the three-phase currents, suggesting potential issues such as uneven phase-to-phase load, abnormal localized heating, abnormal power regulation, or malfunctioning actuators in the monitored equipment. The control system can correlate Y with... When comparing proportions, when Y is greater than 10 percent and less than When Y is 20% of the target value, the corresponding data group to be judged will be included in the early warning score; when Y is greater than 20%, the corresponding data group to be judged will be included in the early warning score. When the imbalance reaches 20%, the corresponding data group to be judged will be included in the alarm score. This allows for the differentiation and handling of mild phase imbalance and more severe phase imbalance.

[0047] For heating power regulation components, the difference between the maximum values ​​of the three-phase current can be used for judgment. Let... These represent the maximum representative values ​​of the current in phase R, phase S, and phase T within the same rolling time window. The maximum representative values ​​of the three-phase current can be calculated using the following formula: ; The representative value of the maximum current difference between the three phases can be calculated using the following formula: ; Wherein, C represents the overall level of the three-phase current peak values ​​within the current rolling time window, and Z represents the degree of difference between the three-phase peak values. The control system can determine the current balance status of the heating power regulation component based on the relationship between C and Z. When C is greater than 0.9Z, it can be determined as a normal state; when C is greater than 0.8Z and less than 0.9Z, it can be determined as a warning state; when C is less than 0.8Z, it can be determined as an alarm state. If the change in current of any phase between adjacent sampling windows exceeds the preset abrupt change level, the corresponding data will be recorded in the serious anomaly record.

[0048] The control system determines whether the monitored equipment has entered the startup state based on representative values ​​of the current in each phase. During startup, the current in the monitored equipment typically rises significantly; for example, the starting current of fan-type equipment may be higher than that during stable operation. Subsequently, the current drops and enters the stable operating range. During stable operation, the three-phase current remains relatively stable for a period of time, and the phase differences do not exceed a preset imbalance level. During shutdown, the current in each phase shows a decreasing trend at the end of the shutdown phase and eventually approaches zero or a low load level. The control system uses the current change amplitude between the current sampling window and historical sampling windows to determine if there are any sudden abnormalities in the monitored equipment; and uses the current decreasing trend at the end of the shutdown phase to determine whether the monitored equipment has shut down normally.

[0049] The control system generates status identifiers corresponding to the monitored equipment based on early warning scores, alarm scores, and severe anomaly records. Status identifiers include normal state, early warning state, alarm state, and fault state. Early warning scores record minor phase imbalances or slight deviations from the stable operating range; alarm scores record higher-level phase imbalances or continuous deviations from the stable operating range; and severe anomaly records record current surges exceeding a preset threshold between adjacent sampling windows. The control system associates these status identifiers with different process stages in the vacuum hot-press bonding process, giving the same current change different anomalous meanings at different stages. For example, a momentary current increase during equipment startup can be identified as a startup current characteristic; continuous phase differences during stable operation can be identified as phase imbalance; current fluctuations during temperature adjustment can be identified as changes in heating power adjustment current; abnormal current changes during vacuum maintenance can be identified as a risk to vacuum system operation; and current decreases during the end stage can be identified as a normal shutdown characteristic. Through this process, the operating characteristic set of the monitored equipment can be obtained without deviating from the current hot-press bonding process, providing a data foundation for subsequent anomaly diversion and equipment health status updates.

[0050] In step S6, the feature set is input into the anomaly monitoring process. First, it is determined whether the monitored equipment is in the start-up, stable operation, or termination state. Then, it is determined whether the monitored equipment has a current anomaly exceeding the anomaly judgment threshold group. When the current anomaly belongs to the stored historical anomaly rules and does not affect the current hot-press bonding quality, the current anomaly is recorded as a traceable anomaly and the current process continues. When the current anomaly belongs to the anomaly affecting vacuum, heating, or pressing stability, a warning, alarm, or shutdown control signal is output. When the current anomaly does not belong to the historical anomaly rules, the corresponding operating status data is input into the machine learning classification model for classification, and the classification results, manual confirmation results, or unsupervised clustering results are updated to the anomaly rule library.

[0051] Step S6 further includes: establishing a first monitoring process for identifying the current operating status of the monitored equipment. The first monitoring process determines whether the monitored equipment has entered the start-up state from the standby state based on whether the multiphase current data exceeds the equipment start-up judgment condition, and continuously monitors whether the operating current returns to the stable operating range within a preset duration after the monitored equipment enters the start-up state; establishing a second monitoring process for finding the current status. When the operating current deviates from the stable operating range, the second monitoring process calls the abnormal rule library to match the current abnormality; when the current abnormality can be matched with a known abnormal rule, the rule matching process is executed. The rule matching process determines whether the known abnormal rule belongs to a traceable abnormality that does not affect the vacuum maintenance of the sealed cavity, the temperature stability of the heating structure, or the pressing stability of the flexible pressure plate; if it belongs to... If a traceable anomaly is detected, the anomaly type, occurrence time, corresponding monitored equipment, and current process stage are recorded, and hot pressing is allowed to continue. If the anomaly is not traceable, a warning, alarm, pause, or shutdown control signal is output. When a current anomaly cannot be matched with a known anomaly rule, a ruleless confirmation process is executed. This process uploads the anomaly data to the information processing terminal, uses unsupervised learning to cluster the anomaly samples to determine whether they form a new anomaly category, and writes the new anomaly category, corresponding process stage, anomaly cause, and handling strategy into the anomaly rule base after manual confirmation. After completing the rule matching process or the ruleless confirmation process, the monitoring status returns to continuous monitoring status to continue performing current status judgment, anomaly diversion, and equipment health status updates for subsequent sampling windows.

[0052] The machine learning classification model is established as follows: Multiphase current data, temperature data, vacuum data, pressure data, equipment operation data, anomaly records, and maintenance records from historical hot-pressing processes are collected. The historical hot-pressing processes are then divided into four stages according to the corresponding process phases: material loading and preheating, vacuuming, degassing, pre-pressing, pressure holding, pressure release, vacuum breaking, and material unloading. Training features are extracted from the historical hot-pressing processes. These features include representative values ​​of current for each phase, representative values ​​of inter-phase differences, maximum representative values ​​of inter-phase differences, current change trends, temperature change trends, vacuum change trends, pressure change trends, equipment start-up indicators, equipment stable operation indicators, equipment shutdown indicators, and process stage indicators. Ineffective samples generated during long periods of standby, no-load, or low-load operation are filtered, weighted, or grouped to ensure that the training features reflect the effective operating status during the hot-pressing process. Historical anomaly records and maintenance records are used to assign normal, early warning, alarm, or fault labels to training features, and are trained using at least one of the support vector machine (SVM) and gradient boosting tree (GPRS) models. When using the SVM model, the training features are classified into states by constructing a classification boundary that can distinguish between normal and abnormal samples with the largest possible interval. When using the GPRS model, an initial weak classification tree is first built, and then the next weak classification tree is generated based on the deviation between the previous classification result and the label. Continuous prediction output is obtained by accumulating the results of multiple weak classification trees, and then the continuous prediction output is mapped to the probability of different state categories to obtain the classification results of normal, early warning, alarm, or fault. New samples obtained from online monitoring, manual confirmation results, and actual maintenance results are periodically added to the training data, so that the machine learning classification model can be updated according to the aging state of the monitored equipment, membrane material changes, and preset program changes.

[0053] In this process, the feature set obtained in step S5 is input into the anomaly monitoring process. Based on the multiphase current data, current change trends, and process stage identifiers of the monitored equipment, the process first identifies whether the monitored equipment is in standby, startup, stable operation, or termination state, and then determines whether there is a current anomaly exceeding the anomaly judgment threshold group. When the monitored equipment enters the startup state from the standby state, the multiphase current data will rise from a low load level to above the equipment startup judgment condition. After entering the startup state, if the operating current returns to the stable operation range within a preset duration and the phase difference does not exceed the corresponding threshold, the monitored equipment is determined to have completed startup and entered the stable operation state. If the operating current deviates from the stable operation range during the stable operation phase, or if the phase difference or current change amplitude reaches the corresponding condition in the anomaly judgment threshold group, then the current anomaly judgment is initiated.

[0054] The first monitoring process identifies the current operating status of the monitored equipment. This process determines whether the monitored equipment has transitioned from standby to startup based on whether the multiphase current data exceeds the equipment startup criteria, and continuously monitors whether the operating current returns to the stable operating range after startup. If the operating current falls back and remains stable after startup, this current change is identified as a startup characteristic and not directly treated as a fault. If the operating current continuously deviates from the stable operating range after startup, or if phase imbalance, sudden changes, or abnormal drops occur during the stable operating phase, the corresponding characteristic is transferred to the second monitoring process.

[0055] The second monitoring process is used to locate the current status. When the operating current deviates from the stable operating range, this process calls the anomaly rule library to match the current anomaly with known anomaly rules. If the current anomaly matches a known anomaly rule, the rule matching process is executed. The rule matching process determines whether the known anomaly rule is a traceable anomaly that does not affect the vacuum maintenance of the sealed cavity, the temperature stability of the heating structure, or the pressing stability of the flexible pressure plate. If it is a traceable anomaly, the anomaly category, the time of occurrence, the corresponding monitored equipment, and the current process stage are recorded, and the current hot pressing process is allowed to continue. If it is not a traceable anomaly, or if the anomaly affects any of the following: vacuum device evacuation and maintenance, heating structure temperature stability, or flexible pressure plate pressing stability, a warning, alarm, pause, or shutdown control signal is output, restricting or terminating the current process stage.

[0056] When a current anomaly cannot be matched with a known anomaly rule, a ruleless confirmation process is executed. This process uploads the anomaly data to the information processing terminal and uses unsupervised learning to cluster the anomaly samples to determine if a new anomaly category has formed. After manual confirmation, the new anomaly category, corresponding process stage, cause of the anomaly, and handling strategy are written into the anomaly rule base. After completing either the rule matching or ruleless confirmation process, the monitoring status returns to continuous monitoring, and current status judgment, anomaly diversion, and equipment health status updates continue for subsequent sampling windows. Through this process, known anomalies that do not affect the current bonding quality can be continuously tracked, anomalies affecting vacuum, heating, or pressing stability can be handled promptly, and unknown anomalies can be classified and precipitated into new rules.

[0057] The machine learning classification model is built based on multiphase current data, temperature data, vacuum data, pressure data, equipment operation data, anomaly records, and maintenance records from historical hot-pressing processes. The historical hot-pressing process is divided into four stages: material loading and preheating, vacuuming, degassing, pre-pressing, pressure holding, pressure release, vacuum breaking, and material unloading. Each historical sample is associated with a corresponding process stage to avoid confusing start-up current, temperature adjustment current, vacuum maintenance current, and shutdown current with the same anomaly meaning. Training features include representative values ​​for each phase current, representative values ​​for inter-phase differences, the maximum representative value for inter-phase differences, current change trends, temperature change trends, vacuum change trends, pressure change trends, equipment start-up indicators, equipment stable operation indicators, equipment shutdown indicators, and process stage indicators. Ineffective samples generated during prolonged standby, no-load, or low-load conditions can be filtered, weighted down, or grouped to ensure that the training features primarily reflect the effective operating status during the hot-pressing process. Configure normal labels, early warning labels, alarm labels, or fault labels for training features based on historical anomaly records and maintenance records, and train using at least one of support vector machine models and gradient boosting tree models.

[0058] When using a support vector machine (SVM) model, the training features are mapped to a feature space that can distinguish between normal and abnormal samples, and a classification boundary is constructed. Let the training samples be... Category label is ,in Represents a class of states, To represent another type of state, the classification hyperplane can be represented as: ; To maximize the margin between the two classes of samples, two parallel boundaries can be constructed. ; ; The corresponding constraints are: ; ; After merging, it can be written as: ; The training process of a support vector machine model can be represented as solving the problem under the above constraints: ; Where w represents the normal vector of the classification hyperplane, and b represents the bias term. The smaller the value, the larger the interval between the two types of samples. This method allows for the acquisition of classification boundaries that distinguish between normal, warning, alarm, or fault states, and enables the classification of new samples obtained from online monitoring.

[0059] When using the gradient boosting tree model, an initial weak classification tree is first built. Then, the next weak classification tree is generated based on the deviation between the previous classification result and the label. Continuous prediction output is obtained by accumulating the results of multiple weak classification trees. Let the dataset be: ; The corresponding tags are: ; The model in round t tests the samples The output can be represented as: ; And set the initial output The prediction output after the accumulation in round t is: ; The objective function consists of a loss function and a regularization term, and can be expressed as: ; Substituting the cumulative prediction output, we get: ; Where L represents the loss between the prediction result and the label, This represents the regularization term used to suppress overfitting. The regularization term can be expressed as: ; in, and To adjust the parameters, Let be the number of leaf nodes in the tree model at round t. Let be the weight of the j-th leaf node. After performing a second-order Taylor expansion on the loss function, the objective function is approximately: ; in, ; ; This represents the first derivative of the loss function. This represents the second derivative of the loss function. The gradient boosting tree model corrects the prediction bias round by round based on the first and second derivatives mentioned above. After obtaining continuous prediction outputs, the continuous prediction outputs are then mapped to the probabilities of different state categories, thereby obtaining the classification results of normal, warning, alarm, or fault.

[0060] New samples obtained from online monitoring, manual confirmation results, and actual maintenance results can be periodically added to the training data. After the new training data is associated with the corresponding process stage, anomaly category, and handling result, it can be used to update the anomaly rule base and machine learning classification model, enabling the model to be updated according to the aging state of the monitored equipment, changes in membrane material, and changes in the vacuum hot pressing preset program, thereby improving the adaptability of subsequent anomaly identification and equipment health status judgment.

[0061] In step S7, when the vacuum level, the temperature of the product fixture, the temperature of the flexible pressure plate, and the health status of the monitored equipment all meet the pressing allowable conditions, the pressure plate cylinder is controlled to drive the flexible pressure plate to move toward the product fixture, and the output pressure of the pressure plate cylinder is adjusted according to the preset pressure curve through the electric proportional valve, so that the flexible pressure plate performs pre-pressing, holding pressure and releasing pressure on the surface of the film material in the heating state.

[0062] Step S7 further includes: selecting at least one process mode from thermoplastic adhesive layer process, thermosetting adhesive layer process, rigid substrate process, flexible substrate process, or high-density substrate process according to the film material type and adhesive layer type; in the thermoplastic adhesive layer process, prioritizing the control of the temperature change trend of the flexible pressure plate and product fixture as well as the pressure holding conditions, so that the adhesive layer obtains a state suitable for flow and spreading after being heated; in the thermosetting adhesive layer process, prioritizing the control of the constant temperature holding state, the pressure holding state, and the pressure release rhythm, so that the adhesive layer forms a stable and cured bond between the film material and the product to be laminated; in the rigid substrate process, configuring the pressure target range according to the higher load-bearing capacity to improve the flatness after lamination; in the flexible substrate process, configuring the preset pressure curve according to the lower pressure impact, and utilizing the elasticity of the flexible pressure plate. Deformation compensation addresses localized warpage and thickness variations in the product to be laminated; in high-density substrate processes, priority is given to maintaining the vacuum stability of the sealed cavity and the smoothness of the pressure release process to reduce the risks of pore blockage, residual bubbles, and film rebound; when the vacuum level deviates from the target vacuum range, the vacuum device's pumping or holding state is adjusted; when the temperature deviates from the target temperature range, the heating power of the fixture heating structure and the pressure plate heating structure is adjusted; when the pressure deviates from the target pressure range, the output pressure of the pressure plate cylinder is adjusted via an electro-proportional valve; when multiphase current data indicates that there is an operational risk in the fixture heating structure, pressure plate heating structure, vacuum device, or pressure plate cylinder, the rate of change of the corresponding action is reduced, the stability confirmation process is extended, or the current process stage is paused.

[0063] Before executing the pressing action, the control system jointly assesses the vacuum level within the sealed cavity, the temperature of the product fixture, the temperature of the flexible pressure plate, and the health status of the monitored equipment. When the vacuum level is within the target range, it indicates that residual gas between the film material and the product to be bonded has been released. When the temperatures of the product fixture and the flexible pressure plate are within the target range, it indicates that the film material, adhesive layer, and product to be bonded have reached a suitable heating state for hot pressing. When the health status of the monitored equipment meets the allowable pressing conditions, it indicates that the fixture heating structure, pressure plate heating structure, vacuum device, heating adjustment components, and pressure plate cylinder, among other actuators, have not experienced any abnormalities affecting vacuum maintenance, heating stability, or pressing stability. Once all the above conditions are met, the control system allows the pressure plate cylinder to execute the pressing action.

[0064] The control system controls the pressure plate cylinder to drive the flexible pressure plate towards the product fixture, and adjusts the output pressure of the pressure plate cylinder according to a preset pressure curve via an electro-proportional valve. The preset pressure curve is matched with the product type, film type, adhesive layer type, product size, and hot-pressing bonding quality requirements, ensuring that the flexible pressure plate sequentially performs pre-pressing, holding, and depressurizing on the film surface under heating. In the pre-pressing stage, a lower pressure is used to establish initial contact between the film and the surface of the product to be bonded, and the vacuum environment within the sealed cavity continues to expel residual gas from the bonding interface. In the holding stage, the output pressure is adjusted to the target pressure range, allowing the adhesive layer to fully flow, wet, or cure under heat and pressure, and ensuring a stable bond between the film and the product. In the depressurizing stage, the pressing pressure is reduced according to a preset rhythm to avoid sudden pressure changes that could cause film rebound, localized desorption, or re-entrapment of air at the bonding interface.

[0065] The control system selects at least one process mode from thermoplastic adhesive layer process, thermosetting adhesive layer process, rigid substrate process, flexible substrate process, or high-density substrate process based on the film material type and adhesive layer type. In the thermoplastic adhesive layer process, the focus is on controlling the temperature change trend of the flexible pressure plate and product fixture, as well as the pressure holding conditions, so that the adhesive layer achieves a state suitable for flow and spreading after heating, and fills the tiny gaps on the surface of the product to be bonded during the pressure holding process. In the thermosetting adhesive layer process, the focus is on controlling the constant temperature holding state, the pressure holding state, and the pressure release rhythm, so that the adhesive layer forms a stable and cured bond between the film material and the product to be bonded, reducing adhesion instability caused by insufficient pressure holding or excessively rapid pressure release. In the rigid substrate process, a pressure target range can be configured according to a higher load-bearing capacity, so that the film material achieves better bonding flatness under relatively stable support conditions. In the flexible substrate process, a preset pressure curve is configured according to a lower pressure impact, avoiding bending, indentation, or local stress concentration of the flexible substrate due to excessive instantaneous pressure, and utilizing the elastic deformation of the flexible pressure plate to compensate for local warping and thickness differences of the product to be bonded. In high-density substrate processes, priority is given to maintaining the vacuum stability of the sealed cavity and the smoothness of the pressure release process, reducing the risks of gas residue in micropores, adhesive layer blockage, film rebound, and local floating.

[0066] During the lamination process, the control system continuously monitors vacuum level, temperature, pressure, and multiphase current data. When the vacuum level deviates from the target range, the system adjusts the evacuation or holding state of the vacuum device according to the direction of deviation to maintain the sealed cavity in a suitable vacuum environment for lamination. When the product fixture temperature or flexible platen temperature deviates from the target temperature range, the system adjusts the heating power of the fixture heating structure and platen heating structure to maintain the adhesive layer within a suitable temperature range for flow, wetting, or curing. When the pressure deviates from the target pressure range, the system adjusts the output pressure of the platen cylinder via an electro-proportional valve to bring the force exerted by the flexible platen on the film surface back to the range corresponding to the preset pressure curve. When multiphase current data indicates that there is an operational risk in the fixture heating structure, platen heating structure, vacuum device, or platen cylinder, the control system reduces the rate of change of the corresponding action, extends the stabilization confirmation process, or pauses the current process stage to avoid equipment malfunctions directly affecting the hot-press lamination quality.

[0067] During pre-pressing, holding, and depressurization, the operating status of the monitored equipment is continuously correlated with the current process stage. The starting current, stable operating current, temperature adjustment current, vacuum maintenance current, and pressing action current are interpreted according to their respective stages to avoid misjudging normal process actions as faults. If the anomaly is traceable and does not affect vacuum maintenance, temperature stability, or pressing stability, the anomaly information is recorded and the current pressing process continues; if the anomaly affects vacuum, heating, or pressure stability, a warning, alarm, pause, or shutdown control signal is output. Through the above control, the flexible pressure plate can complete the hot-pressing bonding of the membrane surface under stable vacuum, stable temperature, and stable pressure conditions.

[0068] In step S8, after hot pressing is completed, the flexible pressure plate is controlled to retract according to the vacuum hot pressing preset program, and the sealed cavity is subjected to gradient depressurization and vacuum breaking treatment. Then, the upper cavity is controlled to rise, the lower cavity is controlled to move to the waiting position, and a process record containing process parameter curves, abnormal records, equipment health status, early warning information and bonding batch information is generated. At the same time, the process record is used to update the abnormal rule base and machine learning classification model.

[0069] After the hot-pressing bonding reaches the pressure holding condition, the control system controls the flexible pressure plate to retract according to the preset vacuum hot-pressing program, causing the flexible pressure plate to detach from the film surface and return to its initial position. Before the flexible pressure plate retracts, it can be confirmed that the pressing pressure has been released according to the preset pressure curve, avoiding the flexible pressure plate being lifted directly before the adhesive layer has stabilized or the pressure has been released. During the retraction of the flexible pressure plate, the control system can continue to monitor the action status of the pressure plate cylinder and the corresponding current changes. If the retraction action is abnormal or the current change exceeds the abnormal judgment threshold group, the corresponding information will be recorded as an abnormality in the pressing end stage.

[0070] After the flexible pressure plate retracts, the control system performs gradient depressurization and vacuum breaking in the sealed cavity. Gradient depressurization allows the pressure inside the sealed cavity to gradually recover according to a set rhythm, preventing sudden vacuum release from causing film rebound, adhesive layer interface disturbance, or re-entry of gas into the bonding area. After vacuum breaking, the pressure inside the sealed cavity returns to the allowable opening state, and the control system then controls the upper cavity to rise. The upper cavity moves upward with the cooperation of the cavity lifting cylinder, connecting rod assembly, guide shaft, and linear bearing until it reaches the opening position. After the cavity is in place, the control system controls the rodless cylinder to drive the lower cavity to move along the linear guide rail to the waiting position, so as to remove the product that has been laminated and load the next product to be laminated.

[0071] During the processes of flexible pressure plate retraction, gradient pressure release, vacuum breaking, upper cavity rise, and lower cavity return, the control system continues to record multiphase current data of the monitored equipment according to the sampling cycle, and judges whether each actuator has completed its normal operation based on the process stage. For monitored equipment such as fans, vacuum devices, heating structures, and pressure plate cylinders, the current in the final stage should show a downward trend and return to standby or low load status; if the current does not decrease as expected, shows a sudden change, or continuously deviates from the stable operating range, it is recorded as an abnormality in the final stage. The shutdown current characteristics in this stage are treated differently from the aforementioned starting current and stable operating current to avoid misjudging normal shutdown changes as faults.

[0072] After the hot-pressing bonding process is completed, the control system generates a process log. The process log includes process parameter curves, anomaly records, equipment health status, early warning information, and bonding batch information. Process parameter curves include curves showing changes in vacuum degree over time, product fixture temperature over time, flexible pressure plate temperature over time, pressing pressure over time, and multiphase current of the monitored equipment over time. Anomaly records include the anomaly type, time of occurrence, corresponding monitored equipment, corresponding process stage, anomaly duration, and handling result. Equipment health status includes normal status, early warning status, alarm status, or fault status, and can be obtained by combining anomaly records from each stage and health status judgment rules. Early warning information includes traceable anomalies that were output during the hot-pressing bonding process but did not cause a shutdown, anomaly trends requiring maintenance confirmation, and alarm information affecting the stability of the current process. Bonding batch information includes product type, film type, adhesive layer type, product size, vacuum hot-pressing preset program number, and bonding completion time.

[0073] After the process log is generated, the control system uses it to update the anomaly rule base and machine learning classification model. For known anomalies, if they are verified in this batch to not affect the vacuum maintenance of the sealed cavity, the temperature stability of the heating structure, or the pressing stability of the flexible platen, they can be retained as traceable anomalies, with their frequency of occurrence, process stage, and operating trend added. For anomalies that have triggered warnings, alarms, pauses, or shutdowns, the causes and handling strategies are updated based on actual handling and maintenance results. For new anomaly samples that do not match historical anomaly rules, their corresponding multiphase current data, temperature data, vacuum data, pressure data, equipment action data, and manual confirmation results can be added to the training data, increasing the anomaly rule base with new anomaly categories and enabling the machine learning classification model to continuously correct the classification boundaries of normal, warning, alarm, and fault states based on the new samples. Through the above processing, the hot pressing bonding results of the current batch not only form a traceable record but can also be used for anomaly identification and equipment health status assessment in subsequent batches.

[0074] Compared with existing technologies, this invention uses an openable and closable sealed cavity to expel residual gas between the film material and the product before pressing, ensuring a stable vacuum environment during the bonding process, which helps reduce air bubbles and floating defects. By heating the product side and the pressing side separately, the adhesive layer maintains a suitable flow or curing state before and after pressing, thereby improving the wetting consistency and adhesion reliability of the film material and the product surface. The flexible pressing method compensates for local warping, thickness differences, and pressing plane errors on the product surface, resulting in a more uniform bonding pressure distribution and reducing the risk of local overpressure, underpressure, and film rebound. Through pressure regulation, staged pressing, pressure holding, and gradient pressure release, different film materials, adhesive layers, and product sizes can be matched with corresponding process windows. By collecting and analyzing the operating data of heating, vacuum, and pressing-related equipment, the invention can identify start-up, stable operation, abnormal fluctuations, and shutdown states during the bonding process. Anomalies that do not affect the current quality are tracked and recorded, and anomalies affecting vacuum, temperature, or pressing stability are promptly warned of or require shutdown, thereby improving bonding yield, process traceability, and equipment maintenance efficiency.

[0075] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the claims.

[0076] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A vacuum hot press mechanism, characterized by, The mechanism includes a linear guide rail, a lower cavity, a fixture mounting plate, a product fixture, a rodless cylinder, an upper cavity, a flexible pressure plate, a pressure plate mounting plate, an electric proportional valve, a pressure plate cylinder, a linear bearing, a guide shaft, a connecting rod assembly, and a cavity lifting cylinder. The lower cavity and the upper cavity form an openable and closable sealed cavity. The fixture mounting plate and the product fixture are disposed in the lower cavity. The product fixture is mounted on the fixture mounting plate and is provided with a fixture heating structure for heating the product to be applied. The linear guide rail and the rodless cylinder are disposed at the bottom of the lower cavity. The rodless cylinder is used to drive the lower cavity to move along the linear guide rail so that the lower cavity switches between the waiting position and the hot pressing position. The flexible pressure plate and the pressure plate mounting plate are disposed in the upper cavity. The flexible pressure plate is provided with a pressure plate heating structure. The pressure plate cylinder is used to drive the flexible pressure plate to press down toward the product fixture so as to heat press and bond the film material on the surface of the product. The electro-proportional valve is connected to the pressure plate cylinder and is used to adjust the output pressure of the pressure plate cylinder; The cavity lifting cylinder is connected to the upper cavity through the connecting rod assembly and is used to drive the upper cavity to lift and lower, so that the upper cavity and the lower cavity can open and close. The linear bearing cooperates with the guide shaft to guide the lifting and lowering movement of the upper cavity. After the lower cavity moves below the upper cavity, the upper cavity descends and closes with the lower cavity to form a sealed cavity. After the sealed cavity is evacuated by a vacuum device, the flexible pressure plate presses down on the surface of the film material under heating and performs pressure-holding bonding.

2. The vacuum hot press mechanism of claim 1, wherein, The flexible pressure plate is an airbag. After the upper cavity descends and closes with the lower cavity to form a sealed cavity and a vacuum is drawn, and the flexible pressure plate is pressed down, the flexible pressure plate is filled with compressed air to flexibly pressurize the product to be laminated.

3. The vacuum hot press mechanism of claim 1, wherein, The product fixture is equipped with a retractable magnet to position and level the film-to-be-applied product, which has magnetic attraction properties.

4. The vacuum hot press mechanism of claim 1, wherein, The product fixture is a vacuum suction fixture.

5. The vacuum hot press mechanism of claim 1, wherein, The flexible pressure plate is a rubber-coated plate.

6. A vacuum hot press mechanism control method characterized by, The method is performed on the vacuum hot pressing mechanism as described in any one of claims 1 to 5, and the method includes steps S1 to S8: S1. Obtain the product type, film material type, adhesive layer type, product size, pre-applied film status, and hot-pressing bonding quality requirements of the product to be laminated, and call the corresponding vacuum hot-pressing preset program according to the product type and film material type. The vacuum hot-pressing preset program includes at least the vacuum degree target range, temperature target range, pressure target range, pressing stage sequence, pressure holding conditions, pressure release conditions, sampling cycle, abnormal judgment threshold group, and equipment health status judgment rules. S2. Load the product to be coated with the pre-applied film material into the product fixture, and control the heating structure of the fixture and the heating structure of the pressure plate to preheat according to the vacuum hot pressing preset program; S3. Control the rodless cylinder to drive the lower cavity to move along the linear guide rail to below the upper cavity, and control the cavity lifting cylinder to drive the upper cavity to descend along the guide shaft through the connecting rod assembly, so that the upper cavity and the lower cavity close to form the sealed cavity. Detect the sealing status of the sealed cavity, and stop the subsequent hot-pressing bonding process or perform sealing compensation when the sealing status does not meet the preset sealing conditions. S4. After the sealed cavity is formed, the vacuum device is activated to bring the vacuum level in the sealed cavity into the target vacuum level range, and the vacuum level change, temperature change and pressure preparation status are continuously monitored during the vacuuming process; when the vacuum level is detected to deviate from the target vacuum level range, the vacuum device is controlled to perform air extraction compensation or maintenance control according to the direction of deviation, so that the residual gas between the film material and the product to be laminated is discharged before hot pressing; S5. Collect multiphase current data of at least one monitored device according to the sampling period. The monitored device includes at least one of the fixture heating structure, the pressure plate heating structure, the vacuum device, the heating adjustment component, and the pressure plate cylinder. Perform rolling time window processing on the multiphase current data and obtain a feature set based on the multiphase current data. The feature set includes phase current balance characteristics, phase difference characteristics, change characteristics between the current sampling time and the historical sampling time, equipment start-up characteristics, stable operation characteristics, and shutdown characteristics. S6. Input the feature set into the anomaly monitoring process, first determine whether the monitored device is in the start-up, stable operation or termination state, and then determine whether the monitored device has a current anomaly exceeding the anomaly judgment threshold group. When the current anomaly belongs to a stored historical anomaly rule and does not affect the current hot-press bonding quality, the current anomaly is recorded as a traceable anomaly and the current process continues; when the current anomaly affects vacuum, heating, or pressing stability, a warning, alarm, or shutdown control signal is output; when the current anomaly does not belong to the historical anomaly rule, the corresponding operating status data is input into a machine learning classification model for classification, and the classification result, manual confirmation result, or unsupervised clustering result is updated to the anomaly rule library. S7. When the vacuum degree, the temperature of the product fixture, the temperature of the flexible pressure plate, and the health status of the monitored equipment all meet the pressing allowable conditions, the pressure plate cylinder is controlled to drive the flexible pressure plate to move toward the product fixture, and the output pressure of the pressure plate cylinder is adjusted according to the preset pressure curve by the electric proportional valve, so that the flexible pressure plate performs pre-pressing, holding pressure and releasing pressure on the surface of the membrane material in sequence under heating state. S8. After hot pressing and bonding are completed, the flexible pressure plate is controlled to retract according to the vacuum hot pressing preset program, and the sealed cavity is subjected to gradient depressurization and vacuum breaking treatment; then the upper cavity is controlled to rise, the lower cavity is controlled to move to the waiting position, and a process record containing process parameter curves, abnormal records, equipment health status, early warning information and bonding batch information is generated. At the same time, the process record is used to update the abnormal rule base and the machine learning classification model.

7. The vacuum hot press mechanism control method according to claim 6, wherein Step S5 specifically includes: According to a preset time interval, the multiphase current data of the monitored device is obtained from the current acquisition module or the power metering module, and the multiphase current data within the same rolling time window is divided into a data group to be judged. For the data set to be determined, extract the representative value of each phase current, the mutual difference between each phase current, the comprehensive representative value of the difference between each phase current, the maximum representative value of the difference between each phase current, and the current change amplitude between adjacent sampling windows. The monitored equipment is judged to have entered the start-up state based on the representative value of each phase current, and to have phase imbalance based on the mutual differences and the comprehensive representative value. The monitored equipment is judged to have sudden abnormality based on the current change amplitude, and to have normal shutdown based on the downward trend of each phase current in the end stage. When the mutual difference is at a first preset imbalance level, the corresponding data will be included in the early warning score; when the mutual difference is at a second preset imbalance level higher than the first preset imbalance level, the corresponding data will be included in the alarm score; when the current change exceeds a preset sudden change level, the corresponding data will be directly included in the serious abnormality record. Based on the warning score, the alarm score, and the serious anomaly record, a status identifier corresponding to the monitored equipment is generated. The status identifier includes at least one of normal status, warning status, alarm status, and fault status. The status identifier is associated with the process stage in the vacuum hot pressing bonding process to distinguish the different abnormal meanings corresponding to the starting current, stable operating current, temperature adjustment current, vacuum maintenance current, and shutdown current.

8. The vacuum hot press mechanism control method according to claim 6, wherein Step S6 also includes: A first monitoring process is established to identify the current operating status of the monitored device. The first monitoring process determines whether the monitored device has entered the start-up state from the standby state based on whether the multiphase current data exceeds the device start-up judgment condition. After the monitored device enters the start-up state, the operating current is continuously monitored for a preset duration to see if it returns to the stable operating range. A second monitoring process is established for finding the current status. When the operating current deviates from the stable operating range, the second monitoring process calls the anomaly rule base to match the current anomaly. When the current anomaly can be matched with a known anomaly rule, a rule matching process is executed. The rule matching process determines whether the known anomaly rule belongs to a traceable anomaly that does not affect the vacuum maintenance of the sealed cavity, the temperature stability of the heating structure, or the pressing stability of the flexible platen. If it belongs to the traceable anomaly, the anomaly category, the time of occurrence, the corresponding monitored equipment, and the current process stage are recorded, and the hot pressing is allowed to continue. If it does not belong to the traceable anomaly, a warning, alarm, pause, or shutdown control signal is output. When the current anomaly cannot be matched with the known anomaly rule, an unruled confirmation process is executed. The unruled confirmation process uploads the anomaly data to the information processing terminal, uses unsupervised learning to cluster the anomaly samples to determine whether the anomaly samples form a new anomaly category, and writes the new anomaly category, the corresponding process stage, the cause of the anomaly, and the handling strategy into the anomaly rule library after manual confirmation. After completing the rule matching process or the ruleless confirmation process, the monitoring state is returned to the continuous monitoring state so that current status judgment, abnormal current diversion and equipment health status update can continue to be performed for subsequent sampling windows.

9. The vacuum hot press mechanism control method according to claim 6, wherein The machine learning classification model is established in the following way: Collect multiphase current data, temperature data, vacuum data, pressure data, equipment operation data, abnormal records and maintenance records from the historical hot-pressing bonding process, and divide the historical hot-pressing bonding process into the feeding and preheating stage, vacuuming stage, degassing stage, pre-pressing stage, pressure holding stage, pressure release stage, vacuum breaking stage and unloading stage according to the corresponding process stages. Training features are extracted from the historical hot-pressing bonding process. These training features include representative values ​​of current in each phase, representative values ​​of phase-to-phase differences, maximum representative values ​​of phase-to-phase differences, current variation trends, temperature variation trends, vacuum variation trends, pressure variation trends, equipment start-up indicators, equipment stable operation indicators, equipment shutdown indicators, and process stage indicators. Ineffective samples formed under long-term standby, no-load, or low-load conditions are filtered, downweighted, or grouped so that the training features can reflect the effective operating status during the hot-press bonding process. Configure normal labels, early warning labels, alarm labels or fault labels for the training features based on historical anomaly records and maintenance records, and train using at least one of support vector machine model and gradient boosting tree model; When using the support vector machine model, the training features are classified into states by constructing a classification boundary that can distinguish between normal samples and abnormal samples with the largest interval. When using the gradient boosting tree model, an initial weak classification tree is first established, and then the next weak classification tree is generated based on the deviation between the previous classification result and the label. Continuous prediction output is obtained by accumulating the results of multiple weak classification trees. Then, the continuous prediction output is mapped to the probability of different state categories to obtain the classification results of normal, early warning, alarm or fault. New samples obtained from online monitoring, manual confirmation results, and actual maintenance results are periodically added to the training data, enabling the machine learning classification model to be updated according to the aging state of the monitored equipment, changes in membrane material, and changes in preset programs.

10. The vacuum hot press mechanism control method according to claim 6, wherein Step S7 also includes: Select at least one process mode from the following, based on the film material type and adhesive layer type: thermoplastic adhesive layer process, thermosetting adhesive layer process, rigid substrate process, flexible substrate process, or high-density substrate process. In the thermoplastic adhesive layer process, the temperature change trend of the flexible pressure plate and the product fixture, as well as the pressure holding conditions, are preferentially controlled so that the adhesive layer obtains a state suitable for flow and spreading after being heated. In the thermosetting adhesive layer process, the constant temperature holding state, the pressure holding state, and the pressure release rhythm are preferentially controlled to ensure that the adhesive layer forms a stable cured bond between the film material and the product to be laminated. In the rigid substrate process, a pressure target range is configured according to a higher load-bearing capacity to improve the flatness after bonding; In the flexible substrate process, the preset pressure curve is configured with a low pressing impact, and the elastic deformation of the flexible platen is used to compensate for the local warping and thickness difference of the product to be laminated. In the high-density substrate process, priority is given to maintaining the vacuum stability of the sealed cavity and the smoothness of the pressure release process to reduce the risk of pore blockage, bubble residue and membrane rebound. When the vacuum level is detected to deviate from the target vacuum level range, the pumping or holding state of the vacuum device is adjusted; when the temperature is detected to deviate from the target temperature range, the heating power of the fixture heating structure and the pressure plate heating structure is adjusted; when the pressure is detected to deviate from the target pressure range, the output pressure of the pressure plate cylinder is adjusted through the electro-proportional valve; when the multiphase current data indicates that the fixture heating structure, the pressure plate heating structure, the vacuum device, or the pressure plate cylinder has an operational risk, the rate of change of the corresponding action is reduced, the stability confirmation process is extended, or the current process stage is paused.

Images