An on-line hot press elimination process method for surface pitting of electronic carrier tape

Abstract

This invention discloses an online hot-pressing process for eliminating crystal points on the surface of electronic carrier tapes, relating to the field of electronic carrier tape manufacturing. The method employs an extruder and a vacuum forming mold. The extruder melts and mixes plastic raw materials, then extrudes them through a die to form a continuous sheet. The vacuum forming mold shapes the sheet into an electronic carrier tape with a receiving cavity. The method further includes the following step: setting a hot-pressing roller device between the die outlet of the extruder and the feed end of the vacuum forming mold. This online hot-pressing process for eliminating crystal points on the surface of electronic carrier tapes, by adding a hot-pressing roller device between the die outlet of the extruder and the feed end of the vacuum forming mold, performs online hot-pressing finishing on the freshly extruded molten sheet. Controllable temperature and pressure are used to physically flatten the protruding crystal points on the sheet surface, while simultaneously making the sheet thickness more uniform and significantly improving the surface finish.

Description

Technical Field

[0001] This invention relates to the field of electronic carrier tape manufacturing technology, and specifically to an online hot-pressing process for eliminating crystal points on the surface of electronic carrier tape. Background Technology

[0002] In current electronic carrier tape production, the process of directly vacuum forming extruded sheets is commonly used. However, due to the presence of incompletely melted polymer gels or impurities in the plastic raw materials, coupled with factors such as die flow channel design and temperature fluctuations, tiny protrusions (i.e., crystal points) are easily generated on the surface of the formed sheet. These crystal points not only affect the optical flatness and appearance quality of the carrier tape surface, but also seriously damage its physical consistency. The presence of crystal points has a serious impact on subsequent processes. During cover tape encapsulation, crystal points cause uneven sealing surfaces, increasing the fluctuation of encapsulation peel force and reducing overall sealing performance. This leads to components being susceptible to moisture, oxidation, or contamination. During high-precision punching, crystal points affect the positioning accuracy and punching quality of the carrier tape, causing cavity size deviations and increased burrs, directly affecting the accurate embedding and fixation of components. Moreover, traditional control methods are often lagging and have poor adaptability, making it difficult to meet the high-efficiency and high-consistency production requirements of modern electronics manufacturing. Summary of the Invention

[0003] The purpose of this invention is to provide an online hot-pressing process for eliminating surface crystal points on electronic carrier tapes, in order to solve the problem that in the existing electronic carrier tape production process, the direct vacuum forming process of the extruded sheet is difficult to effectively control the generation of surface crystal points, resulting in poor sealing performance of the subsequent cover tape packaging, low punching positioning accuracy, and seriously affecting product quality and production efficiency.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an online hot-pressing process for eliminating crystal points on the surface of an electronic carrier tape, wherein the method employs an extruder and a vacuum forming mold, the extruder is used to melt and mix plastic raw materials and then extrude them through a die to form a continuous sheet, and the vacuum forming mold is used to form the sheet into an electronic carrier tape with a receiving cavity, the method further comprising the following steps: A hot press roller device is provided between the die outlet of the extruder and the feed end of the vacuum forming mold, the hot press roller device comprising: At least one pair of cooperating heated rollers, namely an upper pressing roller and a lower pressing roller, with a roller gap between them for the sheet to pass through; A rotating bracket is used to mount the upper and lower pressure rollers and fix them on the production line; A heating mechanism, connected to the upper and / or lower pressure rollers, is used to heat the rollers; The pressure regulating mechanism is used to adjust the roller gap pressure between the upper and lower pressure rollers; Step 1, Online Hot Press Finishing: The continuous sheet extruded by the extruder is directly introduced into the gap between the upper and lower pressure rollers of the hot press roller device. Under the controllable temperature provided by the heating mechanism and the controllable pressure provided by the pressure regulating mechanism, the sheet is hot pressed to achieve secondary finishing of the sheet surface, so that the protruding crystal points on the sheet surface are physically flattened, the sheet thickness tends to be uniform, and the surface smoothness is improved. Step 2, Vacuum forming: The sheet material after hot pressing and finishing is put into the vacuum forming mold and formed into an electronic carrier tape with a receiving cavity.

[0005] Furthermore, the heating mechanism of the hot press roller device adopts one of internal oil heating, electric heating rod heating or external induction heating.

[0006] Furthermore, the hot press roller device also includes a temperature sensor assembly, a pressure sensor, and a control system connected to the temperature sensor assembly and the pressure sensor. The temperature sensor assembly includes a first temperature sensor for detecting the temperature of the heatable roller and a second temperature sensor for collecting temperature distribution data of the sheet before it enters the roll gap. The pressure sensor is used to detect the roll gap pressure between the upper and lower pressing rollers. The control system performs closed-loop automatic control of the heating power of the heating mechanism and the output force of the pressure regulating mechanism based on the detection signals from the temperature sensor assembly and the pressure sensor, so as to achieve stable regulation of the temperature of the heatable roller and the roll gap pressure.

[0007] Furthermore, the control system also includes a feedforward control unit, and a thickness detection device for collecting sheet thickness distribution data is also provided at the inlet of the hot press roller device; the feedforward control unit is connected to the second temperature sensor and the thickness detection device in the temperature sensor assembly, and is used to perform feedforward compensation adjustment on the heating power and roller pressure set value of the hot press roller before the sheet enters the roller gap based on the collected sheet temperature distribution data and thickness distribution data.

[0008] Furthermore, the control system also includes a process knowledge base, which is connected to the control system and used to store historical production data; the feedforward control unit adopts an intelligent feedforward control algorithm based on data-driven modeling and rolling time-domain optimization, including the following steps: Step 1: Extract features from the sheet temperature distribution data collected by the second temperature sensor in the temperature sensor assembly set at the inlet of the hot press roller device and the thickness distribution data collected by the thickness detection device to obtain an inlet state feature vector including the average temperature, the maximum temperature gradient, the average thickness, the standard deviation of thickness fluctuation, and the main frequency of thickness fluctuation. Step 2: Based on historical production data, a dynamic response model is established using neural networks or support vector regression to determine the relationship between the crystal point elimination effect on the sheet surface and the inlet state feature vector, hot press roller temperature, roller pressure, and production line speed. The crystal point elimination effect is monitored in real time by a visual inspection unit installed at the outlet of the hot press roller device. Step 3: With the optimization objectives of maximizing crystal point elimination, minimizing energy consumption, and minimizing the thickness fluctuation of the sheet after hot pressing, and with the hot pressing roller temperature, roller pressure, and production line speed as control variables, under the preset process constraints, a multi-objective evolutionary algorithm is used to solve the optimal control parameter sequence within a set time period in the future, and the optimal control parameters at the current moment are sent to the heating mechanism and pressure regulating mechanism for execution before the sheet enters the roller gap. Step 4: Based on the actual crystal point elimination effect monitored in real time by the visual inspection unit at the outlet of the hot press roller device, calculate the deviation from the model prediction value in Step 2. When the deviation exceeds the preset threshold, trigger the online correction of the dynamic response model parameters. The online correction uses the recursive least squares method or Kalman filter algorithm to update the model parameters until the deviation between the model prediction value and the actual detection value stabilizes again within the preset threshold range. Step 5: Store the data from each production run into the process knowledge base. When the plastic raw material produced again has the same grade as the historical raw material, or when the deviation of at least one of its key physical properties parameters, such as melt index, density, or Vicat softening temperature, is within a preset threshold range, automatically retrieve the historical optimal control parameters stored in the process knowledge base as the initial values ​​for the optimization algorithm in Step 3, and use the Bayesian optimization method to accelerate the parameter optimization process.

[0009] Furthermore, the visual inspection unit is a machine vision system, including an industrial camera and an image processor, used to identify and count the number and size of crystal points on the surface of the hot-pressed sheet in real time, and to feed the inspection data back to the control system.

[0010] Furthermore, the surfaces of the upper and lower pressure rollers are provided with a high-gloss polished layer or coated with a wear-resistant and non-stick coating.

[0011] Furthermore, the pressure regulating mechanism is one of a hydraulically driven pressure regulating mechanism, a pneumatically driven pressure regulating mechanism, or an electromechanically driven pressure regulating mechanism.

[0012] Furthermore, the hot press roller device is positioned such that the straight-line distance between the outlet of the extruder die and the feed end of the vacuum forming die is within the range of 0.01 meters to 0.5 meters.

[0013] Compared with the prior art, the present invention provides an online hot-pressing process for eliminating crystal points on the surface of electronic carrier tape. By adding a hot-pressing roller device between the extruder die outlet and the vacuum forming die feed end, the freshly extruded molten sheet is subjected to online hot-pressing finishing treatment. The raised crystal points on the sheet surface are physically flattened by controlling temperature and pressure, while the sheet thickness becomes more uniform and the surface smoothness is significantly improved. This effectively solves the problem of difficulty in controlling crystal points caused by incompletely melted polymer gels and impurities in the raw materials, as well as factors such as die flow channel design or temperature fluctuations in the prior art. The method achieves a high standard and stable compliance in crystal point elimination. By setting up temperature sensor components, pressure sensors, and a control system, closed-loop automatic control of the hot press roller temperature and roller gap pressure is achieved, realizing stable adjustment of process parameters. By setting up a feedforward control unit, the heating power and roller pressure setpoint of the hot press roller are adjusted using temperature and thickness distribution data at the inlet, overcoming the shortcomings of existing control methods that are lagging behind and unable to cope with dynamic fluctuations in the inlet state of the sheet. Furthermore, through intelligent feedforward control algorithms, the effects of inlet state feature extraction, dynamic response model construction, rolling time-domain multi-objective optimization, online adaptive model updating, and batch-to-batch self-learning optimization are achieved, enabling quantitative prediction and dynamic optimization of crystal point elimination effects. This allows the system to adapt to factors such as changes in raw material batches, production line speed adjustments, and environmental temperature and humidity drift, ensuring that the hot pressing process always operates in a dynamically optimal state, significantly improving the automation level and control accuracy of the production process. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0015] Figure 1 A schematic diagram showing the combination of the extruder, vacuum forming mold, and hot press roller device provided in an embodiment of the present invention; Figure 2 This is a partial rear view of the extruder, vacuum forming mold, and hot press roller device provided in an embodiment of the present invention.

[0016] Explanation of reference numerals in the attached figures: 100. Extruder; 200. Vacuum forming mold; 1. Upper pressure roller; 2. Lower pressure roller. Detailed Implementation

[0017] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0018] like Figures 1 to 2 As shown, this invention provides an online hot-pressing process for eliminating crystal points on the surface of an electronic carrier tape. The method employs an extruder 100 and a vacuum forming mold 200. The extruder 100 is used to melt and mix plastic raw materials and then extrude them through a die to form a continuous sheet. The vacuum forming mold 200 is used to form the sheet into an electronic carrier tape with a receiving cavity. The method further includes the following steps: A hot press roller device is installed between the die outlet of the extruder 100 and the feed end of the vacuum forming mold 200. The hot press roller device includes: At least one pair of heatable rollers that cooperate with each other, namely an upper pressing roller 1 and a lower pressing roller 2, with a roller gap between them; A rotating bracket is used to mount the upper calendering roller 1 and the lower calendering roller 2 and fix them on the production line; A heating mechanism, connected to the upper pressure roller 1 and / or the lower pressure roller 2, is used to heat the rollers; The pressure regulating mechanism is used to adjust the roller gap pressure between the upper pressure roller 1 and the lower pressure roller 2; Step 1, Online Hot Press Finishing: The continuous sheet extruded by the extruder 100 is directly introduced into the gap between the upper pressing roller 1 and the lower pressing roller 2 of the hot press roller device. Under the controllable temperature provided by the heating mechanism and the controllable pressure provided by the pressure regulating mechanism, the sheet is hot pressed to achieve secondary finishing of the sheet surface, so that the protruding crystal points on the sheet surface are physically flattened, the sheet thickness tends to be uniform, and the surface smoothness is improved. Step 2, Vacuum forming: The sheet after hot pressing and finishing enters the vacuum forming mold 200 and is formed into an electronic carrier tape with a receiving cavity.

[0019] Specifically, by heating the upper pressure roller 1 and / or the lower pressure roller 2, the surface temperature of the rollers reaches and stabilizes within the hot pressing temperature range of the plastic raw material (usually slightly higher than the glass transition temperature but lower than the melting temperature). As the upper pressure roller 1 and the lower pressure roller 2 rotate, when the sheet enters the roller gap, the heat of the rollers is transferred to the surface of the sheet, softening its surface and giving it a certain fluidity. Thus, under pressure, the crystal points are physically flattened. By using controllable temperature and pressure, the protruding crystal points on the surface of the sheet are physically flattened, while the sheet thickness becomes more uniform and the surface smoothness is significantly improved. This fundamentally solves the problem of difficult crystal point control and achieves a high standard of stable crystal point elimination.

[0020] In one embodiment of the present invention, the heating mechanism of the hot press roller device adopts one of internal oil heating, electric heating rod heating or external induction heating; Specifically, when using electric heating rods, multiple electric heating rods are evenly inserted along the axial direction inside the heatable roller, and the temperature of each heating rod is controlled independently. When using internal oil heating, the heatable roller has a jacket structure with circulating heat transfer oil flowing inside, and the oil temperature is controlled by an external oil temperature controller. When using external induction heating, an electromagnetic induction coil is set on the outer periphery of the heatable roller, and the surface of the heatable roller is heated by electromagnetic induction. Heating methods can be selected according to the characteristics of different plastic raw materials. For example, oil heating is used when processing PC material to ensure temperature uniformity, while electric heating rods are used when processing PS material to reduce equipment costs.

[0021] To achieve real-time response to dynamic fluctuations in the inlet state of the sheet during the hot pressing process of eliminating crystal points using an electron carrier high-speed particle machine, and to quantitatively predict and optimize the crystal point elimination effect, and to overcome the shortcomings of existing control methods such as lag, poor adaptability, and reliance on manual experience for debugging, in one embodiment of the present invention, the control system further includes a feedforward control unit. The second temperature sensor in the temperature sensor assembly is set at the inlet of the hot pressing roller device and is used to collect temperature distribution data of the sheet before it enters the roller gap. A thickness detection device is also set at the inlet of the hot pressing roller device to collect sheet thickness distribution data. The feedforward control unit is connected to the second temperature sensor and the thickness detection device and is used to perform feedforward compensation adjustment on the heating power and roller pressure setpoint of the hot pressing roller before the sheet enters the roller gap based on the collected sheet temperature distribution data and thickness distribution data.

[0022] In this embodiment, the control system also includes a process knowledge base, which is connected to the control system and used to store historical production data; the feedforward control unit adopts an intelligent feedforward control algorithm based on data-driven modeling and rolling time-domain optimization, including the following steps: Step 1, Inlet State Feature Extraction: Extract features from the sheet temperature distribution data collected by the second temperature sensor in the temperature sensor assembly set at the inlet of the hot press roller device and the thickness distribution data collected by the thickness detection device to obtain an inlet state feature vector including the average temperature, the maximum temperature gradient, the average thickness, the standard deviation of thickness fluctuation, and the main frequency of thickness fluctuation. Specifically, after receiving the raw temperature distribution data along the width of the sheet from the second temperature sensor and the thickness distribution data synchronously collected by the thickness detection device, the feedforward control unit executes the entry state feature extraction algorithm to convert the raw data into an entry state feature vector S that can quantitatively characterize the sheet state. in The S in It contains five key components: Average temperature ,in This represents the temperature value at the i-th sampling point, and N represents the total number of sampling points. The average temperature along the width of the sheet reflects the overall heat level of the sheet after it flows out of the extruder die, and directly affects the flowability of the melt in the roll gap. Maximum temperature gradient Where T is the temperature distribution function and x is the coordinate along the width of the plate. This represents the rate of temperature change along the width of the plate. The maximum absolute value of the temperature gradient quantifies the non-uniformity of the transverse temperature distribution of the plate. An excessively large temperature gradient is an important cause of residual stress after the subsequent flattening of crystal points. average thickness ,in This represents the thickness value at the i-th sampling point. The average thickness in the width direction of the sheet ensures that the base dimensions of the final formed carrier tape meet the design requirements; Standard deviation of thickness fluctuation ,in Let be the thickness value at the i-th sampling point. The average thickness The standard deviation is the thickness, which quantifies the uniformity of the sheet thickness. The larger the standard deviation, the more severe the original unevenness of the sheet surface (including the height difference of crystal points). Thickness fluctuation main frequency It is through the thickness distribution function Obtained by performing a Fast Fourier Transform, i.e. ,in Along the width direction of the sheet The thickness distribution function, This represents the Fast Fourier Transform operation. The transformed amplitude spectrum, This indicates the frequency component corresponding to the maximum amplitude. This is the dominant frequency of thickness fluctuation, used to identify periodic thickness fluctuations caused by extruder gear pump pulsation or screw rotation.

[0023] The above five feature components together constitute the entry state feature vector. This serves as the input basis for subsequent dynamic response models and optimization algorithms.

[0024] Step 2, Dynamic Response Model Construction: Based on historical production data, a dynamic response model is established using neural networks or support vector regression methods to establish the relationship between the crystal point elimination effect on the sheet surface and the inlet state feature vector, hot press roller temperature, roller pressure, and production line speed; the crystal point elimination effect is obtained in real time through a visual inspection unit set at the outlet of the hot press roller device. Specifically, in order to achieve accurate prediction of the crystal point elimination effect, the feedforward control unit uses a hybrid modeling method to construct a dynamic response model: First, a simplified physical mechanism model is established based on the principles of heat transfer and plastic rheology. This model takes the inlet state characteristic vector, hot press roller temperature, roller pressure and production line speed as inputs, and outputs a preliminary prediction of the crystal point elimination effect. Then, for the nonlinear part that is difficult to describe precisely by the mechanism model, a three-layer backpropagation neural network is used for correction. The number of input layer nodes of this neural network is matched with the dimension of the input variables, that is, it receives five feature components in the inlet state feature vector, hot press roller temperature, roller pressure and production line speed, a total of eight input variables. The number of hidden layer nodes was determined to be sixteen through experimental optimization, and the number of output layer nodes is one, which is the final predicted value of the crystal point elimination effect. The neural network is trained using historical production data. During each production process, the control system synchronously records the inlet state feature vector, hot press roller temperature, roller pressure, production line speed, and the corresponding actual crystal point elimination effect, forming a set of training samples. After accumulating a certain number of samples, the neural network is trained offline to bring the network weights to the optimal value, thereby obtaining the complete dynamic response model function relationship. ; in, The crystal point elimination effect predicted by the model can be quantified as the number of crystal points per unit area on the surface of the export sheet or the proportion of crystal point area. This is the feature vector of the entry state. Temperature of the hot press roller For roller pressing, For production line speed; Once the dynamic response model is established, the feedforward control unit can quickly predict the crystal point elimination effect under the current operating conditions based on the real-time collected inlet state data and current control parameters, providing a basis for calculating the objective function for subsequent rolling time-domain optimization.

[0025] Step 3, Rolling Time Domain Multi-Objective Optimization: With the optimization objectives of maximizing crystal point elimination effect, minimizing energy consumption, and minimizing sheet thickness fluctuation after hot pressing, and with hot pressing roller temperature, roller pressure, and production line speed as control variables, under the preset process constraints, a multi-objective evolutionary algorithm is used to solve the optimal control parameter sequence within a future set time period. The optimal control parameters at the current moment are sent to the heating mechanism and pressure regulating mechanism for execution before the sheet enters the roller gap. Specifically, based on the completion of the dynamic response model construction and the obtained predicted values ​​of crystal point elimination effect, the feedforward control unit executes a rolling time-domain multi-objective optimization algorithm; this optimization problem uses three interrelated objective functions as optimization directions: The first objective is to maximize the crystal point elimination effect, that is, to minimize the predicted crystal point residue index, mathematically expressed as: ; The second objective is to minimize energy consumption, specifically to minimize the weighted sum of the heating power and pressurization energy consumption of the hot press roller, mathematically expressed as: ,in and These are the weighting coefficients for heating power and pressurization energy consumption, which can be calibrated according to the sensitivity to energy consumption in actual production. The third objective is to minimize the thickness fluctuation of the sheet after hot pressing, i.e., to minimize the standard deviation of the sheet thickness at the exit point, which can be mathematically expressed as: ,in The standard deviation of the thickness fluctuation of the sheet after hot pressing can be obtained by real-time monitoring through a thickness detection device at the exit or by model prediction. The three objective functions described above constitute a multi-objective optimization problem, with the control variables being... , , The constraints include equipment hard limiting constraints. , , And process safety constraints, such as the upper temperature limit to prevent material degradation. And the upper limit of roller pressure to prevent the sheet from being pressed too thin. ; The optimization algorithm employs the Non-Dominated Sorting Genetic Algorithm with Elite Strategy (NSGA-II) to solve the multi-objective optimization problem.

[0026] Within each control cycle (e.g., every 1 second), the algorithm uses the current entry state feature vector. As input, a set of Pareto optimal solutions satisfying the constraints is generated within a predetermined time period in the future (e.g., 10 seconds in the future, i.e., control time domain length M=10). Each solution corresponds to a sequence of control parameters for the next M time moments. ; The control system selects a final execution solution from the Pareto optimal solution set according to a preset fuzzy preference rule. This rule, for example, ensures the crystal point elimination effect first. If the crystal point is below the preset upper limit, then among the solutions that meet this condition, the energy consumption target is selected first. The lowest possible solution; the system executes only the first control action in the selected solution sequence, which is the optimal control parameter at the current moment. , , The data is sent to the heating mechanism and the pressure regulating mechanism respectively to achieve feedforward compensation adjustment of the hot pressing process. When the next control cycle arrives, the system re-executes the above optimization process based on the updated inlet status data, forming a rolling optimization control mode to ensure that the hot pressing process always operates in the dynamic optimal state.

[0027] Step 4, Online Adaptive Update of Model: Based on the actual crystal point elimination effect monitored in real time by the visual inspection unit at the outlet of the hot press roller device, calculate the deviation from the model prediction value in Step 2. When the deviation exceeds the preset threshold, trigger the online correction of the dynamic response model parameters. The online correction uses the recursive least squares method or Kalman filter algorithm to update the model parameters until the deviation between the model prediction value and the actual detection value stabilizes again within the preset threshold range. Specifically, the control system receives real-time monitoring data from the visual inspection unit at the outlet of the hot press roller device, which shows the actual crystal point elimination effect. And compare it with the predicted value of the dynamic response model at the current time in step 2. Compare and calculate prediction bias The system presets a deviation threshold ε (e.g., the deviation in the number of crystal points does not exceed 2 per meter, or the relative deviation does not exceed 5%), and sets a counter and a time window. When the deviation e continuously exceeds the threshold ε for a preset number of times (e.g., 3 consecutive times), it is determined that the model has deviated from the actual working condition, triggering an online correction program. The online correction uses the recursive least squares method or the Kalman filter algorithm to update the model parameters. If the recursive least squares method is used, it utilizes the N sets of data accumulated in the most recent time window (including the inlet state feature vector S). in Control parameters , , and the corresponding actual crystal point elimination effect The model weight vector is updated using a recursive formula. The covariance matrix P is used to minimize the sum of squared errors between the model output and the measured values. If the Kalman filter algorithm is used, the model parameters are treated as state variables, and state equations and observation equations are established, utilizing real-time observations. The system performs an optimal estimate of the state and updates the model parameters. The online correction process continues, and after each update, the deviation between the model's predicted value and the actual value is recalculated until the deviation stabilizes within a preset threshold range for multiple consecutive times (e.g., 5 consecutive times). At this point, it is determined that the model has reapproached the actual process, exits the correction mode, and resumes normal operation. Through the above online adaptive update mechanism, the system can effectively cope with the impact of time-varying factors such as changes in raw material batches and drift in environmental temperature and humidity on the process model, ensuring that the crystal point elimination effect remains stable at the optimal level for a long time.

[0028] Step 5, Batch-to-Batch Self-Learning Optimization: Data from each production run is stored in the process knowledge base. When the plastic raw material produced again has the same grade as the historical raw material, or when the deviation of at least one of its key physical properties parameters, such as melt index, density, or Vicat softening temperature, is within a preset threshold range, the historical optimal control parameters stored in the process knowledge base are automatically retrieved as the initial values ​​for the optimization algorithm in Step 3, and the Bayesian optimization method is used to accelerate the parameter optimization process.

[0029] Specifically, after each production task is completed, the control system encapsulates the key information of that batch into a case data entry and stores it in the process knowledge base. This case data includes raw material information (such as raw material grade, melt index, density, and Vicat softening temperature), production environment parameters (such as ambient temperature and humidity), and a complete inlet state feature vector sequence. The actual trajectory of the control parameters executed The actual crystal point elimination effect reported by the visual inspection unit And the final batch product quality indicators; The process knowledge base is stored and managed using a database management system. Each case data entry is accompanied by a timestamp and a unique identifier for easy retrieval and recall. When a new batch of production starts, the control system first obtains information about the currently used plastic raw materials, including the raw material grade and key physical property parameters such as melt index, density, and Vicat softening temperature, either measured or input. The system compares the current raw material information with historical raw material information stored in the process knowledge base. If the current raw material grade is exactly the same as a historical raw material grade, it is directly determined to be the same raw material. If the grades are different, the system calculates the deviation of the physical property parameters between the current raw material and each historical raw material, such as the melt index deviation. Density deviation Vicat softening temperature deviation When the above deviation values ​​are all less than their respective preset thresholds (e.g., melt index deviation ≤ 5%, density deviation ≤ 1%, Vicat softening temperature deviation ≤ 3℃), they are judged as similar raw materials; once the same or similar historical raw materials are matched, the system automatically retrieves the raw material from the process knowledge base the batch with the best crystal point elimination effect in the historical production batch (i.e., The smallest set of control parameters is used as the initial value for the current batch step 3 rolling time-domain optimization algorithm. Based on this, the system employs a Bayesian optimization method to accelerate the parameter optimization process. Specifically, it uses the historically optimal control parameters as the initial mean function of the Gaussian process surrogate model. A small amount of trial run data from the initial phase of the current batch (e.g., the inlet state feature vectors of the first five control cycles, actual control parameters, and corresponding crystal point elimination effects) is used as observation points to construct the posterior distribution of the Gaussian process. The next most promising combination of control parameters is selected for experimentation by maximizing the acquisition function (e.g., the expected improvement function, EI). The mathematical expression of the acquisition function is: ,in For the Gaussian process model, the control parameters The predicted value, The current optimal observation value is used. After several iterations (usually 3-5 times), the Bayesian optimization algorithm can converge to the vicinity of the optimal process parameters of the current new batch, avoiding large-scale trial and error debugging from scratch, significantly shortening the process adjustment time after batch switching, and realizing cross-batch reuse and self-learning optimization of process knowledge.

[0030] In one embodiment of the present invention, the visual inspection unit is a machine vision system, including an industrial camera and an image processor, used to identify and count the number and size of crystal points on the surface of the hot-pressed sheet in real time, and to feed the inspection data back to the control system. Specifically, the industrial camera is installed directly above or diagonally above the outlet of the hot press roller device, with its optical axis perpendicular to the surface of the sheet, and is equipped with a trigger acquisition mode synchronized with the production line speed to continuously acquire surface images of the hot-pressed sheet at a rate of 50 to 200 frames per second. The LED light source is arranged in a low-angle ring lighting or high-brightness backlighting manner to ensure that crystal point defects present clear contrast in the image. After receiving the raw image data from the industrial camera, the image processor first performs preprocessing operations, including image filtering and denoising, illumination unevenness correction, and geometric distortion correction. Then, it uses a deep learning-based convolutional neural network (CNN) object detection model (such as YOLO or Faster R-CNN) or a traditional image segmentation algorithm (such as threshold segmentation combined with morphological processing) to identify and locate crystal points in the image in real time, and counts the number of crystal points per unit area, as well as quantitative indicators such as the maximum size, average size, or area ratio of each crystal point. The above statistical results are the actual crystal point elimination effect. The image processor feeds back the detection data to the control system in real time through industrial Ethernet or fieldbus. The control system then uses this data to perform subsequent model deviation calculations and parameter corrections, thereby achieving closed-loop monitoring of crystal point elimination effect and process optimization.

[0031] In one embodiment of the present invention, the hot press roller device further includes a temperature sensor assembly, a pressure sensor, and a control system connected to the temperature sensor assembly and the pressure sensor. The temperature sensor assembly includes a first temperature sensor and a second temperature sensor. The first temperature sensor is used to detect the temperature of the heatable roller, and the second temperature sensor is set at the inlet of the hot press roller device to collect temperature distribution data of the sheet before it enters the roller gap. The pressure sensor is used to detect the roller gap pressure between the upper pressing roller 1 and the lower pressing roller 2. The control system performs closed-loop automatic control on the heating power of the heating mechanism and the output force of the pressure regulating mechanism according to the detection signals of the temperature sensor assembly and the pressure sensor, so as to achieve stable regulation of the temperature of the heatable roller and the roller gap pressure.

[0032] Specifically, the first temperature sensor can be a patch thermocouple or a platinum resistance temperature sensor, which is installed near the roller surface or inside the roller body of the upper pressing roller 1 and the lower pressing roller 2, respectively, to collect temperature data of the working surface of the heatable roller in real time; the second temperature sensor can be an infrared temperature array sensor, which is installed at the inlet of the hot pressing roller device to collect two-dimensional temperature distribution data in the width direction of the sheet in a non-contact manner. The pressure sensor can be a strain gauge or piezoelectric pressure sensor, installed at the connection between the pressure regulating mechanism and the bearing seat of the upper pressure roller 1, or integrated into the oil or air circuit of the hydraulic cylinder or air cylinder, for real-time monitoring of the actual roller gap pressure between the upper pressure roller 1 and the lower pressure roller 2. The first temperature sensor, the second temperature sensor, and the pressure sensor are electrically connected to the control system, and transmit the collected real-time signals to the control system. The control system adopts a programmable logic controller (PLC) or an industrial control computer. It has preset target roller temperature and target roller gap pressure values. After receiving feedback signals from temperature and pressure sensors, the control system compares the measured values ​​with the target values ​​through a PID control algorithm. Based on the deviation, it outputs control commands to the heating mechanism and pressure regulating mechanism in real time. The heating power of the heating mechanism is adjusted to stabilize the roller temperature within the target range, while the output force of the pressure regulating mechanism is adjusted to keep the roller gap pressure constant. This achieves closed-loop automatic control and stable regulation of the temperature and pressure of the heatable roller.

[0033] In one embodiment of the present invention, the rotating bracket 3 may adopt a split structure, consisting of a fixed base and a movable guide frame; the fixed base is firmly locked to the production line base by anchor bolts, the movable guide frame is provided with a vertical slide groove, the bearing seat of the upper pressure roller 1 is slidably fitted into the slide groove, and the bearing seat of the lower pressure roller 2 is fixedly installed on the lower part of the movable guide frame.

[0034] In one embodiment of the present invention, the rotating bracket 3 may adopt an integral frame structure, with the bottom of the frame fixedly connected to the base. Vertically penetrating elongated guide holes are opened on the two side uprights of the frame. The bearing seat of the upper pressure roller 1 passes through the guide hole and can move up and down along the guide hole. The bearing seat of the lower pressure roller 2 is fixedly installed on the lower part of the frame.

[0035] In one embodiment of the present invention, the pressure regulating mechanism is one of a hydraulically driven pressure regulating mechanism, a pneumatically driven pressure regulating mechanism, or an electromechanically driven pressure regulating mechanism. Specifically, when the pressure regulating mechanism adopts a hydraulic drive, it includes two hydraulic cylinders arranged symmetrically on the left and right. The cylinder body of the hydraulic cylinder is fixed to the top beam or the top of the frame of the rotating bracket 3. The piston rod is vertically downward and directly connected to the bearing seat of the upper pressure roller 1. Pressure oil is supplied through the hydraulic station and the roller pressure is steplessly adjusted in conjunction with the proportional servo valve. When the hydraulic cylinder is activated, it drives the bearing seat of the upper pressure roller 1 to move up and down along the vertical slide groove or guide hole of the bracket.

[0036] When the pressure regulating mechanism is pneumatically driven, it includes a cylinder and a precision pressure regulating valve. The cylinder body is fixed to the top of the rotating bracket 3, and the piston rod is connected to the bearing seat of the upper pressure roller 1. The roller pressure is controlled by adjusting the compressed air pressure.

[0037] When the pressure regulating mechanism adopts an electromechanical drive, it includes a servo motor, a reducer, a ball screw, and a compression spring. The servo motor is fixed to the top of the rotating bracket 3 and drives the spring seat to compress the spring through the reducer and the ball screw. The spring force acts on the bearing seat of the upper pressure roller 1, driving the upper pressure roller 1 to move along the guide structure to adjust the roller gap pressure. The precise position control of the servo motor realizes the digital setting of the roller pressure, which does not require a hydraulic or air source and is suitable for clean production environments. A pressure sensor can be set at the spring seat and connected to the control system to realize closed-loop feedback control of the roller pressure.

[0038] In one embodiment of the present invention, the surfaces of the upper pressure roller 1 and the lower pressure roller 2 are provided with a high-gloss polished layer or coated with a wear-resistant and anti-stick coating to avoid new defects on the sheet surface caused by sticking to the rollers and to reduce wear on the roller surface during long-term production.

[0039] In one embodiment of the present invention, the hot press roller device is positioned within a straight-line distance of 0.01 meters to 0.5 meters between the outlet of the extruder 100 and the feed end of the vacuum forming mold 200. Controlling the straight-line distance within a range of 0.01 meters to 0.5 meters can provide a sufficient stable operating range for the sheet while maintaining a suitable hot pressing temperature, ensuring that the hot press roller device can effectively flatten the crystal points on the surface of the sheet, while also taking into account the compactness and maintainability of the equipment layout. In actual production, the specific installation position of the hot press roller device can be fine-tuned within this range according to factors such as the type of plastic raw material being processed (such as PC, PS, ABS, etc.), sheet thickness, and production line speed. For example, when processing raw materials with poor thermal conductivity or high melting temperature, the distance can be appropriately reduced to reduce temperature drop, while when processing thinner sheets or sheets with higher running speed, the distance can be appropriately increased to ensure stability.

[0040] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An on-line thermal press-out elimination process method of surface crystal points of electronic carrier tape, the method adopts an extruder (100) and a vacuum suction forming die (200), the extruder (100) is used for continuously sheeting after melting and mixing plastic raw materials through a die, and the vacuum suction forming die (200) is used for forming the sheet into an electronic carrier tape with a containing cavity, characterized in that, The method further includes the following steps: A hot press roller device is provided between the die outlet of the extruder (100) and the feed end of the vacuum forming mold (200), the hot press roller device comprising: At least one pair of mutually cooperating heatable rollers, namely an upper pressing roller (1) and a lower pressing roller (2), forming a roller gap between them for the sheet to pass through; A rotating bracket is used to mount the upper pressure roller (1) and the lower pressure roller (2) and fix them on the production line; A heating mechanism, connected to the upper pressing roller (1) and / or the lower pressing roller (2), is used to heat the rollers; The pressure regulating mechanism is used to regulate the roller gap pressure between the upper pressure roller (1) and the lower pressure roller (2); Step 1, Online hot pressing finishing process: The continuous sheet extruded by the extruder (100) is directly introduced into the gap between the upper pressing roller (1) and the lower pressing roller (2) of the hot pressing roller device. Under the controllable temperature provided by the heating mechanism and the controllable pressure provided by the pressure regulating mechanism, the sheet is hot pressed to achieve secondary finishing of the sheet surface, so that the protruding crystal points on the sheet surface are physically flattened, the sheet thickness tends to be uniform, and the surface smoothness is improved. Step 2, Vacuum forming: The sheet after hot pressing and finishing enters the vacuum forming mold (200) and is formed into an electronic carrier tape with a receiving cavity.

2. The process of claim 1, wherein the process is characterized by, The heating mechanism of the hot press roller device adopts one of the following: internal oil heating, electric heating rod heating, or external induction heating.

3. The online hot-pressing elimination process for surface crystal points of an electron carrier according to claim 1, characterized in that, The hot press roller device also includes a temperature sensor assembly, a pressure sensor, and a control system connected to the temperature sensor assembly and the pressure sensor. The temperature sensor assembly includes a first temperature sensor for detecting the temperature of the heatable roller and a second temperature sensor for collecting temperature distribution data of the sheet before it enters the roller gap. The pressure sensor is used to detect the roller gap pressure between the upper pressing roller (1) and the lower pressing roller (2). The control system performs closed-loop automatic control of the heating power of the heating mechanism and the output force of the pressure regulating mechanism based on the detection signals of the temperature sensor assembly and the pressure sensor, so as to achieve stable regulation of the temperature of the heatable roller and the roller gap pressure.

4. The online hot-pressing elimination process for surface crystal points of an electron carrier according to claim 3, characterized in that, The control system also includes a feedforward control unit, and a thickness detection device for collecting sheet thickness distribution data is also provided at the inlet of the hot press roller device; the feedforward control unit is connected to the second temperature sensor and the thickness detection device in the temperature sensor assembly, and is used to perform feedforward compensation adjustment on the heating power and roller pressure setting value of the hot press roller before the sheet enters the roller gap based on the collected sheet temperature distribution data and thickness distribution data.

5. The online hot-pressing elimination process for surface crystal points of an electron carrier according to claim 4, characterized in that, The control system also includes a process knowledge base, which is connected to the control system and used to store historical production data; the feedforward control unit adopts an intelligent feedforward control algorithm based on data-driven modeling and rolling time-domain optimization, including the following steps: Step 1: Extract features from the sheet temperature distribution data collected by the second temperature sensor in the temperature sensor assembly set at the inlet of the hot press roller device and the thickness distribution data collected by the thickness detection device to obtain an inlet state feature vector including the average temperature, the maximum temperature gradient, the average thickness, the standard deviation of thickness fluctuation, and the main frequency of thickness fluctuation. Step 2: Based on historical production data, a dynamic response model is established using neural networks or support vector regression to determine the relationship between the crystal point elimination effect on the sheet surface and the inlet state feature vector, hot press roller temperature, roller pressure, and production line speed. The crystal point elimination effect is monitored in real time by a visual inspection unit installed at the outlet of the hot press roller device. Step 3: With the optimization objectives of maximizing crystal point elimination, minimizing energy consumption, and minimizing the thickness fluctuation of the sheet after hot pressing, and with the hot pressing roller temperature, roller pressure, and production line speed as control variables, under the preset process constraints, a multi-objective evolutionary algorithm is used to solve the optimal control parameter sequence within a set time period in the future, and the optimal control parameters at the current moment are sent to the heating mechanism and pressure regulating mechanism for execution before the sheet enters the roller gap. Step 4: Based on the actual crystal point elimination effect monitored in real time by the visual inspection unit at the outlet of the hot press roller device, calculate the deviation from the model prediction value in Step 2. When the deviation exceeds the preset threshold, trigger the online correction of the dynamic response model parameters. The online correction uses the recursive least squares method or Kalman filter algorithm to update the model parameters until the deviation between the model prediction value and the actual detection value stabilizes again within the preset threshold range. Step 5: Store the data from each production run into the process knowledge base. When the plastic raw material produced again has the same grade as the historical raw material, or when the deviation of at least one of its key physical properties parameters, such as melt index, density, or Vicat softening temperature, is within a preset threshold range, automatically retrieve the historical optimal control parameters stored in the process knowledge base as the initial values ​​for the optimization algorithm in Step 3, and use the Bayesian optimization method to accelerate the parameter optimization process.

6. The online hot-pressing elimination process for surface crystal points of an electron carrier according to claim 5, characterized in that, The visual inspection unit is a machine vision system, including an industrial camera and an image processor, used to identify and count the number and size of crystal points on the surface of the hot-pressed sheet in real time, and to feed the inspection data back to the control system.

7. The online hot-pressing elimination process for surface crystal points of an electron carrier according to claim 5, characterized in that, The surfaces of the upper pressure roller (1) and the lower pressure roller (2) are provided with a high-gloss polished layer or coated with a wear-resistant and non-stick coating.

8. The online hot-pressing elimination process for surface crystal points of an electron carrier according to claim 3, characterized in that, The pressure regulating mechanism is one of a hydraulically driven pressure regulating mechanism, a pneumatically driven pressure regulating mechanism, or an electromechanically driven pressure regulating mechanism.

9. The online hot-pressing elimination process for surface crystal points of an electron carrier according to claim 3, characterized in that, The hot press roller device is positioned within a straight distance of 0.01 meters to 0.5 meters between the outlet of the extruder (100) and the feed end of the vacuum forming mold (200).

Images