A film transfer imaging apparatus and a control system thereof

Abstract

The present application relates to the field of heat transfer printing control, and specifically to a film transfer printing imaging device and a control system thereof. The device includes a rack, a transfer printing assembly, a process sensing assembly, an imaging assembly, and a controller. The transfer printing assembly includes a transfer printing cavity, a bearing table, a pressing mechanism, an actuator, and a transfer printing execution unit. The transfer printing execution unit is a heat transfer printing execution structure, including a heating pressure head and a counter-pressure table. Under the control of the controller, the heat transfer printing execution structure outputs heat and cooperates with the pressing mechanism to apply heat and pressure to the film or transfer printing medium to complete the transfer printing. The process sensing assembly is arranged in the transfer printing cavity and its adjacent execution area to obtain temperature, pressure, material feeding speed, tension, and deviation correction signal. The imaging assembly includes a light-shielded imaging cavity, an imaging bearing table, an imaging optical module, and an image sensing module. The controller is electrically connected with the above-mentioned units. During the heat transfer printing process, the controller controls the actuator to perform deviation correction actions such as load reduction, suspension, and back-pressing. After the transfer printing is completed, the controller controls the imaging assembly to collect the pattern image after the transfer printing and store it in association.

Description

Technical Field

[0001] This invention relates to the field of thermal transfer control, specifically to a film transfer imaging device and its control system. Background Technology

[0002] Industrial heat transfer printing is widely used for pattern transfer and surface decoration on substrates such as textiles, films, plastic parts, and coated sheets. A typical heat transfer process includes: overlapping the transfer medium with the substrate to be transferred, completing the pattern transfer under heat and pressure, and then visually inspecting or image-checking the transferred pattern to identify defects and achieve quality traceability. In actual engineering, to adapt to different patterns and material systems, heat transfer equipment usually provides adjustable heating settings, pressing pressure, and execution cycle time parameters, and can adjust the material feeding alignment through a correction mechanism; the inspection side often uses a combination of camera and illumination to acquire images, and judges defects such as gaps, breaks, partial missing parts, background dirt, and overexposure by manual visual inspection or software algorithms.

[0003] However, the existing solutions mentioned above still have objective shortcomings in terms of large-scale replication and stable consistency, mainly due to the sensitivity of the thermal transfer process to the coupling of multiple factors and the separation between process data and result data.

[0004] Firstly, the quality of heat transfer printing is highly sensitive to factors such as temperature field distribution, heat accumulation, pressure distribution uniformity, pressure level deviation, feed rate and residence time per unit area, tension fluctuations, and alignment deviations. These factors often change dynamically during the transfer process and directly affect the morphology of defects. For example, uneven temperature distribution of the indenter or localized heat accumulation can easily lead to localized over-transfer, smearing, edge deformation, or increased background; uneven pressure distribution or pressure level deviations can easily lead to insufficient pattern migration, edge gaps, or localized missing parts; deviations in feed rate / residence time, tension fluctuations, and correction deviations can cause pattern stretching, ghosting, breakage, and abnormal localized residence. Because these factors are hidden and coupled within the process, relying solely on fixed parameter formulations or simple threshold protection makes it difficult to consistently obtain consistent results under different batches of materials, different environmental conditions, or different equipment states, leading to process window drift and quality fluctuations.

[0005] Secondly, the process control of existing heat transfer equipment mainly relies on open-loop execution of preset programs. Even when process monitoring exists, it often remains at the level of single-quantity alarms or over-limit protection, lacking a quantitative indicator system that maps multi-source process signals into the "degree of deviation from the effective transfer window." In actual production lines, when defects occur, the cause can often only be traced back after the transfer is completed through visual inspection or imaging detection results, and troubleshooting is carried out by replacing consumables, reloading, and repeatedly adjusting temperature / pressure / speed. This post-event correction not only increases rework and material waste, but also makes anomaly localization dependent on experience, making it difficult to form reusable process strategies across shifts, equipment, and batches.

[0006] Finally, in many production organization methods, the data link between the transfer printing station and the imaging inspection station is not closed: the inspection side usually outputs defect judgments or image results, but lacks a structured correlation with the process curves, segment parameters, and abnormal segments of the transfer stage. This makes it impossible to quickly answer key questions such as which stage of operation caused the defect, whether it was caused by temperature / pressure / material flow, or imaging aperture / exposure. Furthermore, without a cross-stage consistency judgment mechanism, even if an anomaly is detected, it is difficult to effectively write the detection results back to the control side to form executable self-tuning suggestions, thus making it difficult to achieve segment parameter updates based on result feedback and energy-controlled optimized operation. Summary of the Invention

[0007] The purpose of this invention is to provide a film transfer imaging device and its control system to solve the technical problems mentioned in the background art.

[0008] Based on the above ideas, the present invention provides the following technical solution:

[0009] A film transfer imaging device, comprising:

[0010] The assembly includes a frame, transfer unit, process sensing unit, imaging unit, and controller; among which,

[0011] The transfer assembly includes a transfer cavity, a support platform disposed in the transfer cavity, a pressing mechanism disposed opposite to the support platform, an actuator that drives the pressing mechanism to switch between a pressing position and a load reduction and retraction position relative to the support platform, and a transfer execution unit disposed in the transfer cavity.

[0012] The transfer execution unit is a thermal transfer execution structure, which includes a heating head and a pressing table that cooperates with the heating head. Under the control of the controller, the heating head outputs heat and cooperates with the pressing mechanism to apply heat and pressure to the film or transfer medium to complete the transfer.

[0013] The process sensing component is disposed in the transfer cavity and its adjacent execution area, and is used to acquire temperature signal, pressure signal, feed speed signal, tension signal and deviation signal of the heat transfer process;

[0014] The imaging assembly includes a light-shielding imaging cavity, an imaging stage located within the light-shielding imaging cavity, an imaging optical module, and an image sensing module.

[0015] The controller is electrically connected to the transfer execution unit, the actuator, the process sensing component, and the imaging component. It is used to control the actuator to perform load reduction, pause, and retraction and repressing correction actions during the heat transfer process, and to control the imaging component to collect image data of the transferred pattern and perform associated storage after the heat transfer is completed.

[0016] By integrating the transfer cavity and the light-shielding imaging cavity on the same device, and configuring a support stage, pressing mechanism, actuator, and thermal transfer execution structure, the thermal transfer action has controllable pressing position and unloading retraction position. Together with the process sensing component and imaging component, a complete link of transfer, imaging, and associated storage is formed, which can simultaneously meet the closed-loop basic conditions for process execution and result acquisition at the device level, reducing the traceability difficulties caused by station switching and data fragmentation.

[0017] A control system for a film transfer imaging device,

[0018] Used to control the aforementioned film transfer imaging device;

[0019] The controller includes a processor, a memory, and control and acquisition interfaces connected to the film transfer imaging device; the memory stores a computer program, and when the processor executes the computer program, the system is configured to perform the following:

[0020] S1. Drive the film transfer imaging device into the loading and alignment state, load the film or transfer medium and the substrate to be transferred, and read or generate identification information related to this batch of consumables.

[0021] S2. Select the starting heat transfer program based on the substrate type, transfer medium type, and target pattern type, and initialize the sampling and control parameters;

[0022] S3. Start the thermal transfer execution through the thermal transfer execution structure, and enter the process acquisition state through the process sensing component;

[0023] S4. During the heat transfer process, process signals are collected. Temperature-related characteristics are obtained based on temperature non-uniformity and temperature change rate. Pressure uniformity is obtained based on pressure distribution uniformity. Pressure level deviation is obtained based on the deviation between the average pressure and the target pressure. Speed ​​and residence time deviation is obtained based on the deviation between the material feeding speed and the residence time per unit area relative to the target range. Material feeding stability is obtained based on tension fluctuation and correction deviation. Process evaluation indicators are calculated from the above characteristics. When the process evaluation indicators meet the abnormality criteria, the actuator is controlled to perform a reversible progressive repair action of load reduction, pause, and retraction and repressing. At the same time, the segmented heat input setting value and the segmented material feeding speed setting value are reduced to downgrade to a conservative program. The abnormal segment is bound and recorded with the consumable batch identifier.

[0024] S5. After the heat transfer is completed, control the imaging component to acquire image data, extract the image quality parameter set from the image data, and obtain the process prediction quality characterization based on the temperature-related characterization and the pressure uniformity characterization. Then, fuse the image quality parameter set with the process prediction quality characterization to obtain the cross-stage consistency evaluation index, and write the cross-stage consistency evaluation index and its associated process segment index back to the heat transfer control strategy library.

[0025] S6. Based on the process evaluation index and the cross-stage consistency evaluation index, output the optimized program parameter set and protection threshold set for the next heat transfer, and update the starting process selection rule to achieve cross-batch self-tuning.

[0026] By calculating process evaluation indicators that deviate from the effective window during the heat transfer process and calculating cross-stage consistency evaluation indicators during the imaging stage, and then implementing degradation, recording, and self-tuning output based on the two types of evaluation indicators, the system is upgraded from traditional open-loop process execution to closed-loop control with quantifiable processes, loss prevention of anomalies, write-back of results, and self-tuning of parameters. This improves the stability and consistency of different batches of materials, different environments, and different operator conditions, and reduces rework rate and parameter tuning dependence.

[0027] Preferred,

[0028] S4 further includes:

[0029] Within a preset time window, the rate of change of the pressure head temperature field and the average temperature field value are synchronously collected at a preset sampling period. The pressing pressure distribution and the average pressure are collected. The material feeding speed is collected to determine the residence time per unit area. The tension is collected to determine the degree of tension fluctuation. The correction deviation is collected to determine the degree of material feeding alignment.

[0030] Based on the comparison results between the process evaluation index and the preset threshold range, a sequence of corrective actions is triggered in order from mild to severe: when the process evaluation index is in the first abnormal range, the actuator is controlled to perform a short-term unloading and resume pressing; when the process evaluation index is in the second abnormal range, the actuator is controlled to pause and revert to pressing; when the process evaluation index continuously exceeds the upper limit of the window for more than the preset duration, the process is switched to a conservative program and a reload prompt is output.

[0031] By synchronously collecting temperature field, pressure distribution, material feed speed / dwell time, tension fluctuation and correction deviation within a preset time window, and triggering correction action sequences in stages such as slight load reduction, pause and retraction before pressing, and switching to conservative program, the system can repair abnormalities with minimal intervention in the early stages and stop losses in time when abnormalities persist, thus avoiding the continued production of unusable transfer results and balancing efficiency and yield.

[0032] Preferred,

[0033] The process evaluation indicators are determined by the following quantitative representations:

[0034] Temperature-related characterization, which is determined by the normalized result of the difference between the maximum and minimum values ​​of the head temperature distribution relative to the allowable temperature difference, and the normalized result of the rate of change of the average temperature field relative to the allowable heating rate.

[0035] Pressure uniformity characterization, wherein the pressure uniformity characterization is determined by the normalized result of the pressure distribution dispersion relative to the pressure mean;

[0036] Pressure level deviation characterization, which is determined by the degree of deviation of the pressure mean relative to the target pressure and by normalization of the allowable pressure deviation;

[0037] The deviation between speed and residence time is characterized by the degree of deviation of the material feed speed from the target material feed speed and normalization of the allowable speed deviation, and by the degree of deviation of the residence time per unit area from the target residence time and normalization of the allowable residence time deviation. The two are combined to form the deviation between speed and residence time.

[0038] Material feeding stability characterization, wherein the material feeding stability characterization is determined by the normalized result of the tension fluctuation dispersion relative to the tension mean, and the normalized result of the correction deviation relative to the upper limit of the allowable correction deviation;

[0039] The combination rules are reflected in the following: when the temperature-related characteristics increase, the pressure uniformity characteristics increase, the pressure level deviation characteristics increase, the speed and residence time deviation characteristics increase, and the material flow stability characteristics increase, the process evaluation index increases; among them, the allowable temperature difference, allowable heating rate, target pressure, allowable pressure deviation, target material flow speed, allowable speed deviation, target residence time, allowable residence time deviation, and allowable upper limit of correction deviation are obtained from preset calibration experiments and stored in the heat transfer control strategy library.

[0040] By decomposing the key instability mechanisms of heat transfer printing into quantitative characteristics such as temperature non-uniformity and heat accumulation, pressure contact uniformity, pressure level deviation, speed / dwell time deviation, and material feeding stability (tension and alignment), and constraining their normalization scale with calibration thresholds, a comparable and reusable process evaluation system across batches is formed. This system can transform quality risks that are originally difficult to observe directly into calculable process deviations, improve the accuracy and interpretability of anomaly identification, and provide a unified measurement benchmark for subsequent self-tuning.

[0041] Preferred,

[0042] The S5 also includes:

[0043] The imaging stage inside the light-shielding imaging cavity is positioned to a preset imaging position, and the image sensing module is controlled to acquire an image of the transferred pattern in a preset imaging channel.

[0044] The signal-to-noise ratio, background ratio, saturation ratio, edge gap ratio, breakage ratio, local missing ratio, and uniformity index are extracted from the acquired image as a set of image quality parameters. The set of image quality parameters is then associated and stored with the abnormal interval index and consumable batch identifier.

[0045] By acquiring images of the transferred pattern within a light-shielded imaging cavity and extracting image quality parameters such as signal-to-noise ratio, background ratio, saturation ratio, edge gaps, breaks, local defects, and uniformity, and binding them with abnormal segments and consumable batches during the transfer stage, a one-to-one correspondence can be established between the resulting defects and process segments, significantly improving defect location efficiency and traceability capabilities, and providing a data foundation for cross-batch statistics and strategy updates.

[0046] Preferred,

[0047] The cross-stage consistency evaluation index is obtained by fusing the image quality evaluation metric and the process prediction quality evaluation metric:

[0048] The image quality evaluation metric is jointly determined by the signal-to-noise ratio, the background ratio, the saturation ratio, the edge gap ratio, the breakage ratio, the local missing ratio, and the uniformity index. The higher the signal-to-noise ratio, the higher the uniformity index, the lower the background ratio, the lower the saturation ratio, the lower the edge gap ratio, the lower the breakage ratio, and the lower the local missing ratio, the higher the image quality evaluation metric.

[0049] The process prediction quality evaluation quantity is determined by the temperature correlation characterization and the pressure uniformity characterization, wherein the smaller the temperature correlation characterization and the smaller the pressure uniformity characterization, the higher the process prediction quality evaluation quantity.

[0050] Furthermore, the cross-stage consistency evaluation index is obtained by the image quality evaluation quantity and the process prediction quality evaluation quantity according to a preset fusion rule. The fusion rule is as follows: when both are consistent and both are high, the cross-stage consistency evaluation index takes the higher value; when at least one of them decreases, the cross-stage consistency evaluation index decreases accordingly.

[0051] By fusing image quality assessment metrics with process prediction quality assessment metrics, a cross-stage consistency evaluation index is obtained. This enables the system to not only judge image quality but also whether the image result is consistent with the process state, thereby avoiding misjudgments caused by relying solely on imaging results or solely on process signals. When inconsistencies arise between process prediction and image quality, it can serve as a basis for abnormal pattern recognition, improving diagnostic robustness and enhancing the reliability of the self-tuning strategy.

[0052] Preferred,

[0053] S6 further includes:

[0054] A segmented control parameter set is constructed, which includes the heat input setting value, the pressing pressure setting value, the material feeding speed setting value, and the unit area residence time setting value for each segment. Historical records matching the current substrate type and consumable batch are retrieved from the heat transfer control strategy library. The cross-stage consistency evaluation index corresponding to the historical records is used as the self-tuning target. The segmented control parameter set is updated under the condition of satisfying the effective transfer window constraint derived from the process evaluation index, and the starting procedure and protection threshold set for the next heat transfer are output.

[0055] By constructing a segmented set of control parameters (heat input, pressure, speed, residence time) and using cross-stage consistency as the self-tuning target and the process effective window as the constraint, empirical parameter tuning can be transformed into executable automatic updates of segmented parameters. This allows for a more suitable starting procedure and protection threshold when starting up similar substrates / batch of consumables, reducing the number of trials and parameter tuning times, and improving production line cycle time adaptability and batch consistency.

[0056] Preferred,

[0057] The final optimization result output in S6 includes: determining the optimal set of segmented control parameters under the premise of satisfying the following constraints: the process evaluation index does not exceed the upper limit of the preset process window, the cross-stage consistency evaluation index is not lower than the preset target consistency threshold, the highest temperature during the heat transfer process does not exceed the upper limit of the preset safe temperature, and the pressing pressure setting value of each segment does not exceed the upper limit of the preset safe pressure.

[0058] Furthermore, when the above constraints are met, the optimal set of segmented control parameters is determined with the minimum energy consumption of the thermal transfer process as the optimization objective, and the final set of optimization results, including the optimal set of segmented control parameters and the corresponding process evaluation index and cross-stage consistency evaluation index, is output.

[0059] By determining the final set of segmented parameters with the goal of minimizing energy consumption, under the premise of meeting hard constraints such as the upper limit of the process window, the target consistency threshold, the safe temperature and safe pressure, etc., while ensuring quality and safety boundaries, the energy consumption and material damage risks caused by overheating / overpressing can be reduced. This achieves the optimal control result that balances quality, safety and energy consumption, and forms a reusable final optimized parameter set for subsequent batch operations.

[0060] The technical solution of the present invention may include the following beneficial effects:

[0061] This invention synchronously collects key process states during the heat transfer process using process sensing components. It integrates quantitative characteristics such as temperature non-uniformity and heat accumulation trends, pressure distribution uniformity and pressure level deviations, material feeding speed and residence time per unit area deviations, tension fluctuations and correction deviations into a unified process evaluation system. This forms a process evaluation index that reflects the degree to which the heat transfer process deviates from the effective transfer window. Based on this process evaluation index, the equipment can determine in real time during the transfer process whether risk conditions such as poor contact, localized overheating, uneven pressing, unstable material feeding, or alignment misalignment occur. This avoids the passive situation of traditional solutions that can only deduce the cause from appearance defects after the transfer is completed. Furthermore, this invention designs the correction action as a graded, trigger-based, reversible, and progressive repair sequence: in the mild anomaly stage, a short-term load reduction and restoration of pressing are prioritized to repair local contact / pressure abrupt changes; in the anomalous stage, a pause and reversal of pressing are executed to eliminate local wrinkles, slippage, or insufficient pressing; in the persistent anomaly stage, the segmented heat input setpoint and segmented material feeding speed setpoint are lowered to a conservative level and a reloading prompt is given, thereby achieving an engineered control strategy of repair first, then degrade, and finally stop loss without sacrificing cycle time. Thus, anomalies are identified and addressed at the process stage, significantly reducing the probability of overall defects, minimizing rework and material scrap, and improving adaptability to batch material differences, environmental fluctuations, and operational variations.

[0062] The heat-transfer pattern is imaged and acquired within a light-shielded imaging cavity. This process not only outputs the image itself but also extracts a set of comparable image quality parameters, including signal-to-noise ratio, background ratio, saturation ratio, edge notch ratio, fracture ratio, local missing ratio, and uniformity index, to objectively characterize the appearance quality and defect morphology after transfer. More importantly, this invention associates and stores the aforementioned image quality parameter set with process segment indexes and consumable batch identifiers during the heat transfer process. This ensures that each defect can be mapped to a corresponding time segment and operating condition record, establishing a traceable correspondence between result anomalies and process anomalies. Furthermore, this invention constructs a cross-stage consistency evaluation index: it integrates image quality evaluation quantities with process prediction quality evaluation quantities formed by temperature-related characterization and pressure uniformity characterization. This allows the system to not only determine whether the image meets the standards but also whether the result is consistent with the process state. When both image quality and process prediction quality are poor, the anomaly can be quickly identified as originating from deviations in the thermal transfer process. When image quality is poor but process prediction quality is high, it suggests paying more attention to non-process factors such as imaging aperture, lighting conditions, or sample loading. When image quality is high but process prediction quality is low, it serves as a potential risk indicator, allowing for early adjustment of thresholds or optimization of protection strategies. Therefore, this invention significantly improves the speed and accuracy of anomaly localization, reduces reliance on manual experience, and enhances consistency and traceability across equipment, shifts, and batches.

[0063] This invention introduces a segmented control parameter set at the control level, discretizing the heat transfer process into multiple segments. Each segment is assigned executable setpoints for heat input, pressing pressure, material feed speed, and residence time per unit area, upgrading the controlled object from a single fixed formula to a segmented, adjustable program. During operation, this invention constrains the effective transfer window with process evaluation indicators and uses cross-stage consistency evaluation indicators as self-tuning targets. The process-result correlation data generated in each run is written back to the heat transfer control strategy library for automatic output of more suitable starting programs and protection thresholds under similar substrate and consumable conditions in the next run. This mechanism enables the equipment to adaptively correct for factors such as material batch variations, environmental temperature and humidity changes, pressure head thermal inertia changes, and material feed disturbances, significantly reducing the number of trial-and-error parameter adjustments and improving first-piece yield and stable mass production capabilities. Simultaneously, this invention introduces hard constraints such as upper limits for safe temperature and pressure when outputting the final optimized results, and uses minimum energy consumption as the optimization objective. While meeting quality and safety boundaries, it suppresses overheating and over-pressing, reducing energy consumption and the risk of material thermal damage, thus improving overall economic efficiency. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of the main structure of a film transfer imaging device according to the present invention.

[0065] Figure 2 This is a flowchart illustrating the execution steps of a control system for a film transfer imaging device according to the present invention.

[0066] In the diagram: 1. Frame; 2. Transfer chamber; 3. Support stage; 4. Pressing mechanism; 5. Actuator; 6. Light-shielding imaging wall; 7. Imaging stage; 8. Controller. Detailed Implementation

[0067] Example 1

[0068] like Figure 1 ,

[0069] A film transfer imaging device, comprising:

[0070] Frame 1, transfer assembly, process sensing assembly, imaging assembly, and controller 8; wherein,

[0071] The transfer assembly includes a transfer cavity 2, a support platform 3 disposed in the transfer cavity, a pressing mechanism 4 disposed opposite to the support platform 3, an actuator 5 that drives the pressing mechanism to switch between a pressing position and a load reduction and retraction position relative to the support platform, and a transfer execution unit disposed in the transfer cavity.

[0072] The transfer execution unit is a thermal transfer execution structure, which includes a heating head and a pressing table that cooperates with the heating head. Under the control of the controller, the heating head outputs heat and cooperates with the pressing mechanism to apply heat and pressure to the film or transfer medium to complete the transfer.

[0073] The process sensing component is disposed in the transfer cavity and its adjacent execution area, and is used to acquire temperature signal, pressure signal, feed speed signal, tension signal and deviation signal of the heat transfer process;

[0074] The imaging assembly includes a light-shielding imaging cavity 6, an imaging stage 7 located inside the light-shielding imaging cavity, an imaging optical module, and an image sensing module.

[0075] The controller 8 is electrically connected to the transfer execution unit, the actuator 5, the process sensing component and the imaging component. It is used to control the actuator to perform load reduction, pause and retraction and repressing correction actions during the heat transfer process, and to control the imaging component to collect image data of the transferred pattern and perform associated storage after the heat transfer is completed.

[0076] In this embodiment, the process sensing component is used to acquire temperature signals, pressure signals, material feeding speed signals, tension signals, and deviation correction signals during the heat transfer process, and aligns them with a unified timestamp. The temperature signal is preferably the pressure head temperature field T(x,t), with N spatial sampling points. T The sampling points range from 16 to 256, with units of ℃, and the sampling frequency ranges from 10 to 50 Hz. The pressure signals are preferably the pressure distribution p(x,t) and the average pressure value, with N being the number of spatial sampling points. p The sampling frequency is 50–200 Hz, with 16–256 points, in kPa or MPa. The material feeding speed v(t) is in mm / s, and the sampling frequency is 50–200 Hz. The dwell time per unit area s(t) is in seconds, calculated from the speed and effective contact length of the pressure head for feed-type heat transfer printing, and equal to the online cumulative value of the holding pressure duration for fixed-point heat pressing. The tension F(t) is in N, and the sampling frequency is 50–200 Hz. The correction deviation e(t) is in mm, and the sampling frequency is 50–200 Hz. To ensure calculation stability, the controller performs anomaly and missing signal handling for the above signals: based on the main frequency f... s (Preferred 100Hz) Resample each signal and align the timestamps; if no more than 3 sampling points are missing in a single instance, use linear interpolation to fill in the missing points; if more than 3 are missing, mark the time slice as an unusable segment and normalize it according to the number of valid samples during integration / statistics; use sliding window median filtering (window 5 to 11 points) to denoise scalar signals, and replace outliers that deviate more than 6 times the median absolute deviation with the median of the window; use point-by-point spatial median filtering for the temperature / pressure matrix and set an engineering upper limit on the rate of change to remove outliers.

[0077] like Figure 2 A control system for a film transfer imaging device.

[0078] Used to control the aforementioned film transfer imaging device;

[0079] The controller includes a processor, a memory, and control and acquisition interfaces connected to the film transfer imaging device; the memory stores a computer program, and when the processor executes the computer program, the system is configured to perform the following:

[0080] S1. Drive the film transfer imaging device into the loading and alignment state, load the film or transfer medium and the substrate to be transferred, and read or generate identification information related to this batch of consumables.

[0081] S2. Select the starting heat transfer program based on the substrate type, transfer medium type, and target pattern type, and initialize the sampling and control parameters;

[0082] S3. Start the thermal transfer execution through the thermal transfer execution structure, and enter the process acquisition state through the process sensing component;

[0083] S4. During the heat transfer process, process signals are collected. Temperature-related characteristics are obtained based on temperature non-uniformity and temperature change rate. Pressure uniformity is obtained based on pressure distribution uniformity. Pressure level deviation is obtained based on the deviation between the average pressure and the target pressure. Speed ​​and residence time deviation is obtained based on the deviation between the material feeding speed and the residence time per unit area relative to the target range. Material feeding stability is obtained based on tension fluctuation and correction deviation. Process evaluation indicators are calculated from the above characteristics. When the process evaluation indicators meet the abnormality criteria, the actuator is controlled to perform a reversible progressive repair action of load reduction, pause, and retraction and repressing. At the same time, the segmented heat input setting value and the segmented material feeding speed setting value are reduced to downgrade to a conservative program. The abnormal segment is bound and recorded with the consumable batch identifier.

[0084] S5. After the heat transfer is completed, control the imaging component to acquire image data, extract the image quality parameter set from the image data, and obtain the process prediction quality characterization based on the temperature-related characterization and the pressure uniformity characterization. Then, fuse the image quality parameter set with the process prediction quality characterization to obtain the cross-stage consistency evaluation index, and write the cross-stage consistency evaluation index and its associated process segment index back to the heat transfer control strategy library.

[0085] S6. Based on the process evaluation index and the cross-stage consistency evaluation index, output the optimized program parameter set and protection threshold set for the next heat transfer, and update the starting process selection rule to achieve cross-batch self-tuning.

[0086] In this embodiment, the controller executes a closed-loop thermal transfer process. First, it enters a loading and alignment state, loading the transfer medium and the substrate to be transferred, and reading or generating identification information related to this batch of consumables. Simultaneously, it records the substrate type, transfer medium type, and target pattern type for this task, which will serve as search keys for subsequent strategy libraries. Then, it selects the starting thermal transfer program and initializes the sampling and control parameters. This embodiment uses a segmented control parameter vector, where u... k The setpoint for the heat input of the k-th segment (preferably the pressure head setpoint temperature, in °C), p k v is the setpoint for the pressing pressure of the kth segment (in MPa). k The setpoint for the feed speed of segment k (in mm / s), s k This is the set value (in seconds) for the dwell time per unit area of ​​the k-th segment. The number of segments K can be 2 to 5, with 3 segments being preferred to adapt to industrial cycle time. When the strategy store contains historical records of the same key value, the historical best Z is directly read as the starting program. When it does not exist, the default starting range is used for initialization: pressure head setting temperature 140–210℃, pressing pressure 0.2–2.5MPa, material feeding speed 10–200mm / s, and dwell time per unit area 0.2–5.0s. Subsequently, the heat transfer execution is started through the heat transfer execution structure and the process acquisition state is entered. At the same time, the segment number kkk is bound to the time segment to form the segment basis for the abnormal segment index.

[0087] Specifically,

[0088] In this embodiment, during the heat transfer process, process evaluation indicators are calculated and the effective transfer window is estimated in real time within a preset time window; the preset time window is 3-20 seconds, and the sampling period is preferably 0.01 seconds. For each sampling moment, the maximum temperature, minimum temperature, mean temperature, and its rate of change are calculated from the temperature field (using a difference approximation); the mean pressure and standard deviation are calculated from the pressure distribution; the residence time per unit area is calculated from the material feeding speed; the mean tension and standard deviation are recursively calculated from the tension within the window; the correction deviation is e(t). iThis embodiment uses statistical calibration of qualified samples to obtain allowable and target quantities, which are then stored in the heat transfer control strategy library: the allowable temperature difference is taken as the 95th quantile of qualified samples; the allowable heating rate is taken as the 95th quantile of qualified samples; the target pressure is taken as the mean of qualified samples; the allowable pressure deviation is taken as the 95th quantile of qualified samples; the target feed speed and allowable speed deviation are taken as the mean speed and 95th quantile deviation of qualified samples, respectively; the target residence time and allowable residence time deviation are taken as the mean residence time and 95th quantile deviation of qualified samples, respectively; the upper limit of allowable correction deviation is taken as the 95th quantile of qualified samples. The above allowable / deviation quantities are entered into the normalized denominator to form equivalent weights: the smaller the allowable quantity, the greater the contribution of the corresponding characterization to X; the larger the allowable quantity, the smaller the contribution of the corresponding characterization, thereby achieving engineering reproducibility of the weight coefficient source without changing the formula structure given in the claims.

[0089] Specifically,

[0090] The process evaluation indicators are determined by the following quantitative representations:

[0091] Define the temperature-dependent characterization a(t) as:

[0092]

[0093] Where T max (t) and T min (t) represents the maximum and minimum values ​​of the temperature field T(x,t), respectively, ΔT allow To allow for temperature differences, T allow To allow the heating rate.

[0094] Define the pressure uniformity characterization b(t) as:

[0095]

[0096] Where σ p (t) represents the standard deviation of the pressure distribution p(x,t). This represents the average pressure distribution.

[0097] Define the deviation of pressure level as d(t):

[0098]

[0099] Where p tar For target pressure, p allow To allow for pressure deviation.

[0100] Define the deviation of velocity from dwell time as g(t):

[0101]

[0102] Where v tar For the target material feeding speed, v allow To allow for speed deviation, s tar The target dwell time per unit area, s allow This is the allowable deviation in dwell time.

[0103] Define the material flow stability characterization c(t) as:

[0104]

[0105] Where the standard deviation of σF(t) is... e is the average tension value allow The upper limit of the allowable correction deviation.

[0106] The process evaluation index is defined as follows:

[0107]

[0108] Where τ is the preset time window length; the ΔT allow T allow p tar p allow v tar v allow s tar s allow e allow The results are obtained from preset calibration experiments and stored in the thermal transfer control strategy library.

[0109] in A difference approximation is used, and the integral is discretized using the trapezoidal rule. The effective transfer window constraint is specified as the upper limit X of the process window. lim The result is obtained by taking the 95th quantile of the qualified sample X and multiplying it by a safety factor of 1.1 to 1.3. Further anomaly criteria are defined: the first anomaly interval is 0.9X. lim <X≤X lim The criterion for sustained overlimit is X>X. lim The duration is 0.5–2.0 seconds. These thresholds are determined by calibration data and written into the strategy library, facilitating reproducible deployment on different devices and with different materials.

[0110] In this embodiment, when the online calculated process evaluation index meets the anomaly criteria, the controller drives the actuator to output a correction action and downgrade the program parameters according to the graded strategy. In the first anomaly interval, the actuator performs a short-term unloading and resumes pressing, with an unloading displacement Δz of 0.2–1.0 mm lasting 0.1–0.5 s, followed by resumption of pressing. Simultaneously, the heat input setting value for the next segment is lowered by 1–5°C, and the material feeding speed setting value for the next segment is lowered by 5%–15%. In the second anomaly interval, the actuator performs a pause and retraction re-pressing, with a retraction displacement Δz of 0.5–2.0 mm lasting 0.2–1.0 s, followed by re-pressing. Simultaneously, the heat input setting value for the current segment is lowered by 5–15°C, and the material feeding speed setting value for the current segment is lowered by 10%–30%. When a continuous over-limit condition is met, the controller switches to a conservative program. The conservative program adjusts the remaining segment's u... k With v k The system is uniformly adjusted to the conservative setting preset in the strategy library, and a reload prompt is output. The controller binds and records abnormal segments with consumable batch identifiers. The abnormal segment index includes at least: time interval, segment number, segment parameter vector Z at that time, and the X value of the window, for subsequent write-back and self-tuning.

[0111] In this embodiment, after thermal transfer is completed, imaging and cross-stage consistency calculations are performed. The controller positions the imaging stage to a preset imaging position, controls the imaging optics module to illuminate the transferred pattern using a fixed illumination mode, and the image sensing module acquires a grayscale image (III). The image resolution can be 2048×2048 or 4096×4096, with a bit depth of 12–16 bits, and the exposure time is set to avoid saturation. To improve reproducibility, image preprocessing includes dark field correction (I). ’ =II dark And replace bad pixels with the median of a 3×3 matrix. Then, for I... ’ Construct a pattern region mask M, with the threshold either using the Otsu method or a fixed threshold (normalized gray level 0.15–0.30). Calculate the image quality parameter set from the mask M and the background region: Signal-to-noise ratio (SNR) = μ s / (σ b +ε), where μ s The mean of the pattern area, σ b The standard deviation of the background region is given, and ε is set to 10. −6 Background ratio r bg =μ b / (μ s +ε), where μ b The average value of the background area; saturation ratio r sat The proportion of pixels within the pattern area whose grayscale reaches 98% or more of full scale; the edge notch ratio r gap The fracture ratio r is obtained by calculating the proportion of missing segments along the M boundary length;brk The proportion of extra connected component area to the total pattern area is obtained through connected component analysis; the local missing component ratio r is also shown. miss The uniformity index is obtained by measuring the percentage of window areas within a sliding window (e.g., 32×32) where the local mean is less than 50% of the pattern mean; J = 1−σ s / (μ s +ε), where σ s The standard deviation of the pattern area is given.

[0112] Specifically,

[0113] The S5 also includes:

[0114] The imaging stage inside the light-shielding imaging cavity is positioned to a preset imaging position, and the image sensing module is controlled to acquire an image of the transferred pattern in a preset imaging channel.

[0115] The signal-to-noise ratio, background ratio, saturation ratio, edge gap ratio, breakage ratio, local missing ratio, and uniformity index are extracted from the acquired image as a set of image quality parameters. The set of image quality parameters is then associated and stored with the abnormal interval index and consumable batch identifier.

[0116] Specifically,

[0117] The cross-stage consistency evaluation index is obtained by fusing the image quality evaluation metric and the process prediction quality evaluation metric:

[0118] Define the image quality evaluation metric Q img for:

[0119]

[0120] Wherein, SNR is the signal-to-noise ratio extracted from the acquired image, r bg For background ratio, r sat For saturation ratio, r gap The proportion of edge notches and the proportion of fractures, r brk The proportion of local missing parts r miss .

[0121] Define the process prediction quality evaluation quantity Q pred for:

[0122]

[0123] The cross-stage consistency evaluation index is:

[0124]

[0125] Specifically,

[0126] S6 further includes:

[0127] Construct a piecewise control parameter vector:

[0128]

[0129] Where u k Input the set value for the heat input of segment k, p k The set value for the pressing pressure of segment kkk is v. k The set value for the material feeding speed of the kth segment, s k Set the dwell time per unit area for the kth segment;

[0130] Retrieve historical records matching the current substrate type and consumable batch from the thermal transfer control strategy library, use the cross-stage consistency evaluation index YYY corresponding to the historical records as the self-tuning target, and update Z under the condition of satisfying the effective transfer window constraint derived from the process evaluation index X, and output the starting procedure and protection threshold set for the next thermal transfer.

[0131] Specifically,

[0132] The final optimization result output in S6 includes:

[0133] Define the feasible set:

[0134]

[0135] Where X lim Y is the upper limit of the process window. tar T is the target consistency threshold. safe For the upper limit of safe temperature, p safe This is the upper limit of safety pressure.

[0136] Furthermore, under the premise of satisfying the aforementioned feasible set constraints, the optimal segmented control parameter vector is determined with the minimum energy consumption of the thermal transfer process as the optimization objective:

[0137]

[0138] Wherein, PZ(t) corresponds to the heat input setpoint u of the corresponding segment within each segment. k Correspondingly, the equivalent power is calculated from the equipment's thermal response model;

[0139] Where PZ(t) corresponds to the heat input setpoint u in each segment. k Correspondingly, the equivalent power is calculated from the equipment thermal response model; in implementation, a piecewise constant approximation can be used: PZ(t) = P in the k-th segment. k P kThe energy consumption integral is approximately ∑kPk·τk, calculated from the pressure head set temperature and heating power-temperature calibration curve. Final output

[0140] Output the final set of optimization results:

[0141]

[0142] And Write it into the policy library as a candidate for the next startup procedure, and at the same time, write the corresponding protection threshold (including X). lim Updated along with the anomaly classification threshold.

Claims

1. A film transfer imaging device, characterized in that, include: The assembly includes a frame, transfer unit, process sensing unit, imaging unit, and controller; among which, The transfer assembly includes a transfer cavity, a support platform disposed in the transfer cavity, a pressing mechanism disposed opposite to the support platform, an actuator that drives the pressing mechanism to switch between a pressing position and a load reduction and retraction position relative to the support platform, and a transfer execution unit disposed in the transfer cavity. The transfer execution unit is a thermal transfer execution structure, which includes a heating head and a pressing table that cooperates with the heating head. Under the control of the controller, the heating head outputs heat and cooperates with the pressing mechanism to apply heat and pressure to the film or transfer medium to complete the transfer. The process sensing component is disposed in the transfer cavity and its adjacent execution area, and is used to acquire temperature signal, pressure signal, feed speed signal, tension signal and deviation signal of the heat transfer process; The imaging assembly includes a light-shielding imaging cavity, an imaging stage located within the light-shielding imaging cavity, an imaging optical module, and an image sensing module. The controller is electrically connected to the transfer execution unit, the actuator, the process sensing component, and the imaging component. It is used to control the actuator to perform load reduction, pause, and retraction and repressing correction actions during the heat transfer process, and to control the imaging component to collect image data of the transferred pattern and perform associated storage after the heat transfer is completed.

2. A control system for a film transfer imaging device, characterized in that, Used to control the film transfer imaging device as described in claim 1; The controller includes a processor, a memory, and control and acquisition interfaces connected to the film transfer imaging device; the memory stores a computer program, and when the processor executes the computer program, the system is configured to perform the following: S1. Drive the film transfer imaging device into the loading and alignment state, load the film or transfer medium and the substrate to be transferred, and read or generate identification information related to this batch of consumables. S2. Select the starting heat transfer program based on the substrate type, transfer medium type, and target pattern type, and initialize the sampling and control parameters; S3. Start the thermal transfer execution through the thermal transfer execution structure, and enter the process acquisition state through the process sensing component; S4. During the heat transfer process, process signals are collected. Temperature-related characteristics are obtained based on temperature non-uniformity and temperature change rate. Pressure uniformity is obtained based on pressure distribution uniformity. Pressure level deviation is obtained based on the deviation between the average pressure and the target pressure. Speed ​​and residence time deviation is obtained based on the deviation between the material feeding speed and the residence time per unit area relative to the target range. Material feeding stability is obtained based on tension fluctuation and correction deviation. The process evaluation index is calculated from the above characteristics. When the process evaluation index meets the abnormality criteria, the actuator is controlled to perform a reversible progressive repair action of load reduction, pause and retraction and repressing. At the same time, the segmented heat input setting value and the segmented material feeding speed setting value are reduced to downgrade to a conservative program, and the abnormal segment is bound and recorded with the consumable batch identifier. S5. After the heat transfer is completed, control the imaging component to acquire image data, extract the image quality parameter set from the image data, and obtain the process prediction quality characterization based on the temperature-related characterization and the pressure uniformity characterization. Then, fuse the image quality parameter set with the process prediction quality characterization to obtain the cross-stage consistency evaluation index, and write the cross-stage consistency evaluation index and its associated process segment index back to the heat transfer control strategy library. S6. Based on the process evaluation index and the cross-stage consistency evaluation index, output the optimized program parameter set and protection threshold set for the next heat transfer, and update the starting process selection rule to achieve cross-batch self-tuning.

3. The control system for a film transfer imaging device according to claim 2, characterized in that, S4 further includes: Within a preset time window, the rate of change of the pressure head temperature field and the average temperature field value are synchronously collected at a preset sampling period. The pressing pressure distribution and the average pressure are collected. The material feeding speed is collected to determine the residence time per unit area. The tension is collected to determine the degree of tension fluctuation. The correction deviation is collected to determine the degree of material feeding alignment. Based on the comparison results between the process evaluation index and the preset threshold range, a sequence of corrective actions is triggered in order from mild to severe: when the process evaluation index is in the first abnormal range, the actuator is controlled to perform a short-term unloading and resume pressing; when the process evaluation index is in the second abnormal range, the actuator is controlled to pause and revert to pressing; when the process evaluation index continuously exceeds the upper limit of the window for more than the preset duration, the process is switched to a conservative program and a reload prompt is output.

4. The film transfer imaging device and its control system according to claim 3, characterized in that, The process evaluation indicators are determined by the following quantitative representations: Temperature-related characterization, which is determined by the normalized result of the difference between the maximum and minimum values ​​of the head temperature distribution relative to the allowable temperature difference, and the normalized result of the rate of change of the average temperature field relative to the allowable heating rate. Pressure uniformity characterization, wherein the pressure uniformity characterization is determined by the normalized result of the pressure distribution dispersion relative to the pressure mean; Pressure level deviation characterization, which is determined by the degree of deviation of the pressure mean relative to the target pressure and by normalization of the allowable pressure deviation; The deviation between speed and residence time is characterized by the degree of deviation of the material feed speed from the target material feed speed and normalization of the allowable speed deviation, and by the degree of deviation of the residence time per unit area from the target residence time and normalization of the allowable residence time deviation. The two are combined to form the deviation between speed and residence time. Material feeding stability characterization, wherein the material feeding stability characterization is determined by the normalized result of the tension fluctuation dispersion relative to the tension mean, and the normalized result of the correction deviation relative to the upper limit of the allowable correction deviation; The combination rules are reflected in the following: when the temperature-related characteristics increase, the pressure uniformity characteristics increase, the pressure level deviation characteristics increase, the speed and residence time deviation characteristics increase, and the material flow stability characteristics increase, the process evaluation index increases; among them, the allowable temperature difference, allowable heating rate, target pressure, allowable pressure deviation, target material flow speed, allowable speed deviation, target residence time, allowable residence time deviation, and allowable upper limit of correction deviation are obtained from preset calibration experiments and stored in the heat transfer control strategy library.

5. The film transfer imaging device and its control system according to claim 4, characterized in that, The S5 also includes: The imaging stage inside the light-shielding imaging cavity is positioned to a preset imaging position, and the image sensing module is controlled to acquire an image of the transferred pattern in a preset imaging channel. The signal-to-noise ratio, background ratio, saturation ratio, edge gap ratio, breakage ratio, local missing ratio, and uniformity index are extracted from the acquired image as a set of image quality parameters. The set of image quality parameters is then associated and stored with the abnormal interval index and consumable batch identifier.

6. The film transfer imaging device and its control system according to claim 5, characterized in that, The cross-stage consistency evaluation index is obtained by fusing the image quality evaluation metric and the process prediction quality evaluation metric: The image quality evaluation metric is jointly determined by the signal-to-noise ratio, the background ratio, the saturation ratio, the edge gap ratio, the breakage ratio, the local missing ratio, and the uniformity index. The higher the signal-to-noise ratio, the higher the uniformity index, the lower the background ratio, the lower the saturation ratio, the lower the edge gap ratio, the lower the breakage ratio, and the lower the local missing ratio, the higher the image quality evaluation metric. The process prediction quality evaluation quantity is determined by the temperature correlation characterization and the pressure uniformity characterization, wherein the smaller the temperature correlation characterization and the smaller the pressure uniformity characterization, the higher the process prediction quality evaluation quantity. Furthermore, the cross-stage consistency evaluation index is obtained by the image quality evaluation quantity and the process prediction quality evaluation quantity according to a preset fusion rule. The fusion rule is as follows: when both are consistent and both are high, the cross-stage consistency evaluation index takes the higher value; when at least one of them decreases, the cross-stage consistency evaluation index decreases accordingly.

7. The film transfer imaging device and its control system according to claim 6, characterized in that, S6 further includes: A segmented control parameter set is constructed, which includes the heat input setting value, the pressing pressure setting value, the material feeding speed setting value, and the unit area residence time setting value for each segment. Historical records matching the current substrate type and consumable batch are retrieved from the heat transfer control strategy library. The cross-stage consistency evaluation index corresponding to the historical records is used as the self-tuning target. The segmented control parameter set is updated under the condition of satisfying the effective transfer window constraint derived from the process evaluation index, and the starting procedure and protection threshold set for the next heat transfer are output.

8. The film transfer imaging device and its control system according to claim 7, characterized in that, The final optimization result output in S6 includes: determining the optimal set of segmented control parameters under the premise of satisfying the following constraints: the process evaluation index does not exceed the upper limit of the preset process window, the cross-stage consistency evaluation index is not lower than the preset target consistency threshold, the highest temperature during the heat transfer process does not exceed the upper limit of the preset safe temperature, and the pressing pressure setting value of each segment does not exceed the upper limit of the preset safe pressure. Furthermore, when the above constraints are met, the optimal set of segmented control parameters is determined with the minimum energy consumption of the thermal transfer process as the optimization objective, and the final set of optimization results, including the optimal set of segmented control parameters and the corresponding process evaluation index and cross-stage consistency evaluation index, is output.

Images