Parameter adaptive optimization method of semiconductor automatic film mounting machine based on environmental perception

Abstract

The present invention discloses a method for adaptive optimization of parameters of a semiconductor automatic film pasting machine based on environmental perception, which belongs to the field of automated control technology, wherein the method includes: three-dimensional modeling of structural design information; sensor group layout based on a visual model of a film pasting machine; obtaining a film pasting environment operation data set of the semiconductor automatic film pasting machine, and performing control index analysis and adaptive model training on the film pasting environment operation data set; obtaining a film pasting environment data stream, and performing control analysis on the film pasting environment data stream based on a film pasting machine control adaptive model; adaptively controlling the semiconductor automatic film pasting machine based on the film pasting machine operation control parameters, and performing monitoring feedback optimization control on the film pasting control status. The present application solves the technical problem in the prior art that the semiconductor automatic film pasting machine lacks intelligent perception and adaptive control, resulting in low film pasting accuracy and unstable efficiency, thereby realizing accurate and efficient film pasting operation optimization control.

Description

Technical Field

[0001] The present invention relates to the field of automation control technology, and in particular to a method for adaptively optimizing parameters of a semiconductor automatic film mounting machine based on environment perception. Background Art

[0002] With the rapid development of semiconductor manufacturing technology, the requirements for precision and stability in the production process are increasing, especially in micro-processing processes such as film lamination, where any slight deviation will lead to product defects or even the scrapping of the entire batch. In order to improve production efficiency and reduce costs, the semiconductor manufacturing industry is gradually moving towards automation and intelligence. As one of the important production equipment, the automatic film lamination machine undertakes the important task of accurately attaching thin film materials to the surface of semiconductors.

[0003] Although current semiconductor automatic film laminating machines have a certain level of automation, most of them lack intelligent perception and adaptive adjustment capabilities and cannot effectively respond to changes in environmental factors during the production process. Traditional control methods rely on manually set parameters, lack dynamic adjustment mechanisms, and are easily affected by environmental fluctuations and equipment aging, resulting in reduced product quality. In order to meet the needs of high-precision and high-stability film laminating processes in the semiconductor manufacturing process, how to achieve adaptive optimization of automatic film laminating machine parameters through environmental perception and data analysis has become an urgent problem to be solved in the industry. Summary of the invention

[0004] This application provides a method for adaptively optimizing parameters of a semiconductor automatic film laminating machine based on environmental perception, aiming to solve the technical problem in the prior art that the semiconductor automatic film laminating machine lacks intelligent perception and adaptive control, resulting in low film laminating accuracy and unstable efficiency.

[0005] In view of the above problems, the present application provides a method for adaptively optimizing parameters of a semiconductor automatic film mounting machine based on environmental perception.

[0006] The present application discloses a method for adaptive optimization of parameters of a semiconductor automatic film pasting machine based on environmental perception, the method comprising acquiring structural design information of the semiconductor automatic film pasting machine, performing three-dimensional modeling on the structural design information, and generating a visual model of the film pasting machine; arranging a sensor group based on the visual model of the film pasting machine to obtain an environmental perception module of the film pasting machine, wherein the environmental perception module of the film pasting machine integrates a temperature sensor, a humidity sensor, a pressure sensor, and a positioning sensor; acquiring a film pasting environment operation data set of the semiconductor automatic film pasting machine by using data mining technology, performing control indicator analysis and adaptive model training on the film pasting environment operation data set, and obtaining a film pasting machine control adaptive model; monitoring and acquiring a film pasting environment data stream through the film pasting machine environmental perception module, performing control analysis on the film pasting environment data stream based on the film pasting machine control adaptive model, and obtaining film pasting machine operation control parameters; setting a monitoring point group on the visual model of the film pasting machine, adaptively controlling the semiconductor automatic film pasting machine based on the film pasting machine operation control parameters, and monitoring, feedback, and optimization control of the film pasting control status through the monitoring point group.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] By adopting a semiconductor automatic film laminating machine parameter adaptive optimization method based on environmental perception, including the technical solution of constructing and applying an environmental perception module of the film laminating machine and a control adaptive model, the technical problem in the prior art that the semiconductor automatic film laminating machine is unable to accurately perceive and dynamically adjust according to environmental changes, thereby resulting in unstable film laminating accuracy and efficiency is solved. The technical effects of real-time monitoring of the film laminating environment, optimizing operating parameters, and improving film laminating accuracy and stability are achieved, which significantly improves the intelligence level and operating efficiency of the automatic film laminating machine.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flow chart of a method for adaptively optimizing parameters of a semiconductor automatic film mounting machine based on environmental perception is provided for the embodiment of the present application;

[0011] Figure 2 A flow chart of obtaining an environmental perception module of a semiconductor automatic film mounting machine in a method for adaptively optimizing parameters of the semiconductor automatic film mounting machine based on environmental perception is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0012] The overall idea of ​​the technical solution provided by this application is as follows:

[0013] The embodiment of the present application provides a method for adaptively optimizing parameters of a semiconductor automatic film mounting machine based on environmental perception. By constructing an environmental perception module, the working environment of the semiconductor automatic film mounting machine is monitored in real time, and an adaptive control model is established based on these data. Through the optimization of the model, the operating parameters are automatically adjusted to ensure that the film mounting machine can maintain high-precision and stable operation in different environments, thereby improving overall performance and efficiency.

[0014] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0015] Examples, such as Figure 1 As shown, the embodiment of the present application provides a method for adaptively optimizing parameters of a semiconductor automatic film mounting machine based on environment perception, the method comprising:

[0016] Step S100: obtaining structural design information of a semiconductor automatic film mounting machine, performing three-dimensional modeling on the structural design information, and generating a visual model of the film mounting machine.

[0017] Specifically, a semiconductor automatic film attacher is a device used to accurately attach thin film materials to the surface of a semiconductor wafer or chip. It is usually used in steps such as packaging, protection or material deposition in the semiconductor manufacturing process. Structural design information refers to the physical and functional design parameters of the device, including its mechanical structure, component size, material properties, layout of functional modules, etc.

[0018] First, you need to obtain the structural design information of the equipment, and then use computer-aided design (CAD) software such as SolidWorks, AutoCAD or Siemens NX to convert these two-dimensional design drawings or data information into a three-dimensional model. This three-dimensional model is a visual model of the laminating machine, which can display the various components and overall layout of the equipment in a virtual environment.

[0019] By building a 3D visualization model of the semiconductor automatic film mounting machine, the operation process of the equipment can be intuitively observed and simulated. Design defects can be quickly discovered during the development phase, equipment layout can be optimized, and rework in design and manufacturing can be reduced. It is convenient to dynamically analyze the operating behavior of the equipment, improving design efficiency and accuracy.

[0020] Step S200: Arrange a sensor group based on the film laminating machine visualization model to obtain a film laminating machine environment perception module, wherein the film laminating machine environment perception module integrates a temperature sensor, a humidity sensor, a pressure sensor, and a positioning sensor.

[0021] Specifically, temperature sensors are used to monitor temperature changes in the working environment. Humidity sensors are used to sense the humidity level in the air. Pressure sensors are used to detect air pressure or mechanical pressure inside the equipment or working area. Positioning sensors are used to accurately locate the position and movement trajectory of equipment or equipment parts.

[0022] Perform spatial analysis on the overall structure of the laminating machine to find the best sensor placement points. Temperature sensors are usually placed in areas susceptible to ambient temperature fluctuations, humidity sensors can be placed in areas prone to moisture accumulation, pressure sensors are usually placed on the robot arm or transmission mechanism, and positioning sensors are mainly used during the movement of the robot arm or laminating head.

[0023] By rationally deploying sensors based on a visualization model, a complete environmental perception module is formed, which can monitor the working environment of the film laminating machine in real time and feedback important environmental data.

[0024] Step S300: using data mining technology to obtain a film-mounting environment operation data set of the semiconductor automatic film-mounting machine, performing control indicator analysis and adaptive model training on the film-mounting environment operation data set, and obtaining a film-mounting machine control adaptive model.

[0025] Specifically, control index analysis refers to extracting key control indicators (such as temperature, humidity, pressure, position, etc.) from the data set and analyzing them to determine the extent of their impact on the film laminating process.

[0026] Use data mining techniques to conduct in-depth analysis of this data. You can use libraries such as pandas in Python or data analysis tools such as MATLAB to clean, organize and analyze the data. In this process, control indicators are extracted from the data set, that is, parameters that have a greater impact on device performance, such as temperature, humidity, film attachment speed and pressure. For example, it was found that when the ambient humidity is higher than 60%, the film adhesion effect is poor, so humidity becomes an important control indicator.

[0027] After the control indicators are clear, the data are trained through machine learning technology to build an adaptive model that can automatically adjust according to environmental changes, that is, the film laminating machine control adaptive model. Specifically, the model will understand how different environmental conditions (such as temperature and humidity) affect the quality of film laminating through historical data. When new environmental data enters, the model will adjust the parameters of the film laminating machine (such as film laminating speed, pressure, etc.) based on these data to achieve the best film laminating effect. That is adaptive control, that is, the film laminating machine control adaptive model can "learn" environmental changes during operation and adapt to different working conditions by automatically adjusting control parameters.

[0028] By using data mining technology to analyze the film-mounting environment operation data set, extracting key control indicators, and training the film-mounting machine control adaptive model through machine learning, the semiconductor automatic film-mounting machine can dynamically adjust its operating parameters in a changing environment, greatly improving the intelligence of the equipment, reducing manual intervention, and realizing automatic control.

[0029] Step S400: monitoring and acquiring the film pasting environment data stream through the film pasting machine environment perception module, and performing control analysis on the film pasting environment data stream based on the film pasting machine control adaptive model to obtain the film pasting machine operation control parameters.

[0030] Specifically, the film laminating machine environment perception module refers to a system module composed of multiple sensors, which is used to monitor the environmental conditions of the film laminating machine in real time. The sensors include temperature, humidity, pressure, positioning, etc., which can collect key data when the equipment is running. The film laminating environment data stream refers to the environment and equipment operation data that is continuously collected and transmitted in real time through the environment perception module. It reflects the working status of the equipment under different environmental conditions, and is dynamic and continuous.

[0031] During operation, the environment perception module monitors the surrounding environmental conditions in real time, such as temperature, humidity, pressure, etc., through sensors installed in various parts of the equipment. These data are continuously collected to form a film laminating environment data stream. Next, the film laminating machine control adaptive model is used to perform real-time control analysis on these film laminating environment data streams to calculate the optimal operating parameters of the equipment, that is, the film laminating machine operation control parameters.

[0032] Through this step, the semiconductor automatic film laminating machine can achieve intelligent adaptive control, which can not only perceive environmental changes in real time, but also quickly perform control analysis based on the film laminating environment data stream through the trained adaptive model to determine the optimal operation control parameters. This enables the equipment to dynamically adjust its operating parameters to maintain the best film laminating effect.

[0033] Step S500: Setting a monitoring point group on the visual model of the laminating machine, adaptively controlling the semiconductor automatic laminating machine based on the laminating machine operation control parameters, and monitoring and feedback optimizing control of the laminating control status through the monitoring point group.

[0034] Specifically, the monitoring point group is a number of monitoring points set on the device model to monitor the key parameters and status of the device in real time during operation. Each monitoring point corresponds to a specific parameter, such as temperature, humidity, pressure, position, etc.

[0035] First, multiple monitoring point groups are set on the visual model of the semiconductor automatic film laminating machine. The monitoring points are usually located at key positions of the equipment, such as the film laminating head, the robotic arm, the pressure device, etc. These point groups will monitor the status data of the equipment in real time. Then, based on the real-time collected data and combined with the established operation control parameters, the semiconductor automatic film laminating machine is adaptively controlled. As the equipment runs, the system will continuously collect real-time data through the set monitoring point groups and perform monitoring feedback optimization control. The collected monitoring data will be transmitted to the control center of the equipment in real time and analyzed through the film laminating machine control adaptive model. The feedback loop can be implemented through the SCADA system, which can manage the entire monitoring, feedback, and optimization process in real time. For example, if the film is slightly offset in the edge area during the film laminating process, the monitoring point will detect this offset and feedback to the control system. The control system will adjust the precise position of the robotic arm according to this feedback signal to ensure that the film is aligned again.

[0036] This step, based on the adaptive control system of the visualization model and monitoring point group, significantly improved the intelligence level and production efficiency of the film laminating machine. By setting up monitoring point groups at key locations, the equipment can obtain operation data in real time, forming a closed-loop feedback control system, and realizing accurate monitoring and optimization of the equipment status.

[0037] Further, such as Figure 2 As shown, the film laminating machine environment perception module is obtained, including: performing functional structure decomposition and structural attribute marking based on the film laminating machine visualization model to obtain a film laminating machine functional structure set and a functional structure attribute set; performing perception demand analysis on the film laminating machine functional structure set according to the film laminating machine functional structure attribute set to obtain functional structure environment perception demand information; performing sensor selection on the film laminating machine functional structure set based on the functional structure environment perception demand information to obtain a functional structure sensor demand parameter set; performing layout position planning on the functional structure sensor demand parameter set based on the film laminating machine visualization model to determine the functional structure sensor layout position parameters; performing sensor group layout according to the functional structure sensor demand parameter set and the functional structure sensor layout position parameters to obtain the film laminating machine environment perception module.

[0038] Specifically, functional structure splitting is to modularize the equipment and divide the complex mechanical equipment into different functional units, each of which realizes a specific operation function, such as film pasting, positioning, pressure control, etc. Structural attribute marking: It is to identify the specific characteristics of each functional module of the equipment, such as the physical properties (size, material) or functional characteristics (such as pressure, position) of each structure.

[0039] First, the functional structure is split based on the 3D film laminating machine visualization model. This can be done through CAD tools (such as SolidWorks or AutoCAD) to decompose the complex equipment structure into multiple functional modules. For example, the film laminating head, pressure control device, mobile robot arm, etc. can all be split as independent functional units.

[0040] Next, structural attribute marking is performed on each functional unit by assigning physical and functional attributes to each module, such as the pressure range of the film head, the positioning accuracy of the mobile arm, the motion path of the robot arm, etc. For example, the attributes of the pressure control module include the maximum applied pressure and accuracy, while the attribute of the positioning module is the positioning error range. Then, the perception requirements of different functional modules are analyzed based on the set of functional structural attributes to determine which key parameters each functional module needs to monitor and obtain the functional structural environment perception requirement information. According to the perception requirements, suitable sensors are selected to form a set of sensor requirement parameters.

[0041] Next, based on the visual model of the laminating machine, the layout position planning is carried out to determine the installation position of each sensor. For example, the pressure sensor needs to be installed in the center of the laminating head to ensure accurate pressure detection, while the positioning sensor needs to be installed at the key connection part of the robot arm to monitor the movement path in real time.

[0042] Finally, the sensor group is deployed according to the designed deployment plan, and various sensors are installed on the actual equipment to form the film machine environment perception module. It is used to monitor the operating environment and status of the equipment in real time, and the data is transmitted to the central control system through the perception system to support further adaptive control and optimization.

[0043] Through this step, a laminating machine system with environmental perception capabilities can be built. Functional structure splitting and attribute tagging make the equipment modular, which is convenient for monitoring the status of different functional units. Perception demand analysis ensures that the key parameters of each module are effectively monitored, and sensor demand parameter selection and layout location planning ensure that the sensor can accurately monitor the equipment status.

[0044] Furthermore, the film laminating machine control adaptive model is obtained, including: obtaining a film laminating machine control indicator set, classifying and identifying the film laminating machine environment operation data set according to the film laminating machine control indicator set, and obtaining the film laminating machine indicator classification operation data set; performing environmental control analysis training on the film laminating machine indicator classification operation data set respectively, and obtaining a film laminating machine indicator branch controller set; performing multivariate criticality analysis on the film laminating machine indicator branch controller set, and determining the branch controller key coefficient information; performing weighted parallel fusion on the film laminating machine indicator branch controller set based on the branch controller key coefficient information, and obtaining the film laminating machine control adaptive model.

[0045] Specifically, the film laminating machine control index set refers to the indicators set to control various key parameters in the operation process of the film laminating machine, such as temperature, pressure, humidity, position, etc. These indicators are used to monitor and adjust the operation of the equipment. The film laminating environment operation data set refers to a large amount of data generated when the equipment is operating in different environments, including the operating parameters of the equipment under different temperatures, humidity and pressures, film laminating quality and other data. Multivariate criticality analysis is to analyze the importance and influence of different controllers in the entire system to determine the key control parameters. The branch controller key coefficient information refers to the weight and influence of each controller in the entire system obtained through analysis, which is used to characterize the importance of each controller.

[0046] First, obtain the set of control indicators of the film laminating machine, that is, determine which indicators are critical to the film laminating process. These indicators directly affect the quality of the film laminating, so they need to be monitored and adjusted by the system in real time. Next, collect the film laminating environment operation data set, which includes the performance of the equipment when operating under different environmental parameters. After that, classify the data according to different control indicators through classification identification. For example, temperature-related data is classified into one category, and pressure-related data is classified into another category. Each classification-labeled data set corresponds to different operating parameters.

[0047] Next, the film laminating machine indicator classification job data set is trained for environmental control analysis. By using machine learning techniques (such as regression analysis, decision trees, or neural networks), multiple control models are trained to predict how the equipment should adjust control parameters under different environmental conditions. After training, multiple indicator branch controllers are generated, which are responsible for adjusting different control indicators. For example, one controller can be used specifically to adjust pressure, and another to adjust humidity.

[0048] In order to ensure that these controllers work together, a multivariate criticality analysis is required. The importance of each controller in the entire film application process is analyzed and the weight of each controller is determined. Through the branch controller critical coefficient information, the system can assign a weight to each controller, which represents the importance of the controller. Then, using this weight information, the controllers are combined together through weighted parallel fusion to generate a comprehensive adaptive control model. Weighted parallel means that the output of each controller is weighted according to its importance, ensuring that all controllers can optimize the operation of the system.

[0049] Finally, through the control adaptive model, various control parameters can be dynamically adjusted according to environmental data. For example, the system will automatically adjust the operating parameters of the film laminating machine according to the real-time temperature, humidity and pressure data to ensure the best film laminating effect under different environmental conditions.

[0050] Furthermore, the film laminating machine control indicator set includes: extracting structural control indicators from the film laminating environment operation data set to obtain a film laminating machine structural controllable indicator set; performing importance assessment based on the film laminating machine functional structure set to determine the correlation depth of multiple structural control indicators; and performing correlation indicator mining on the film laminating machine structural controllable indicator set according to the multiple structural control indicator correlation depths to obtain a film laminating machine control indicator set.

[0051] Specifically, first, the film laminating environment operation data set is processed and analyzed to extract the control indicators related to the equipment structure, which are called structural controllable indicators. For example, the range of motion of the robot arm of the film laminating machine, the pressure setting of the film laminating head, etc. Next, according to the physical structure of the film laminating machine, it is decomposed into multiple functional structure sets. For example, the film laminating machine can be divided into sub-modules such as the film laminating head, the transmission mechanism, and the control system, and each module has its specific function. By analyzing these functional modules, the importance of each module in the entire system can be obtained.

[0052] Use decision trees or Bayesian networks to rank and analyze different control indicators. Perform importance assessment to determine the influence and priority of each structural control indicator. For example, in some film application scenarios, pressure control is more important than temperature control, which means that the weight of the pressure control indicator will be higher.

[0053] Then, we use correlation analysis technology to calculate the correlation depth between each control indicator. The correlation depth of structural control indicators measures how one indicator affects another indicator. Through correlation analysis, we can better understand the interaction between control indicators. In order to further optimize control, we can identify potential and neglected indicators through correlation indicator mining.

[0054] This step will result in a complete set of control indicators for the laminating machine, each of which has been deeply analyzed and optimized. This control indicator set provides a comprehensive control basis for the system and can be used to monitor and adjust the operating status of the laminating machine in real time.

[0055] Furthermore, obtaining the film laminating machine control indicator set includes: clustering and integrating the film laminating machine structure controllable indicator set according to the film laminating machine function structure set to obtain multiple structure controllable indicator cluster sets; performing correlation analysis on each control indicator in the multiple structure controllable indicator cluster sets in turn based on the film laminating environment operation data set to obtain multiple structure indicator correlation coefficient sets; performing indicator optimization on each control indicator in the multiple structure controllable indicator cluster sets according to the association depth of the multiple structure control indicators and the multiple structure indicator correlation coefficient sets to obtain the film laminating machine control indicator set.

[0056] Specifically, first, the set of structural controllable indicators of the film laminating machine is processed, and integrated according to similarity and functionality through cluster analysis technology to form one or more "structural controllable indicator cluster sets". For example, if in a film laminating process, temperature, humidity and pressure have a direct impact on the quality of the film, they will be classified into the same cluster, while the position accuracy of the film laminating machine and the speed of the conveyor belt will be classified into another cluster. In this way, the control parameters in each cluster can be optimized more targeted.

[0057] Next, for each set of structural controllable indicator clusters, the correlation analysis method is used to analyze the correlation between each control indicator. For example, the increase in temperature will affect the adhesion of the film, and the change in humidity will affect the pressure sensing of the film head. These correlations will be analyzed to obtain the specific structural indicator correlation coefficient. After obtaining the correlation coefficient, the interaction between these control indicators is further analyzed in combination with the correlation depth. The correlation depth measures the degree of influence of a control indicator on other indicators. For example, if the impact of temperature on the entire film laminating process is much greater than that of humidity, then the correlation depth of temperature will be higher. Finally, by analyzing the correlation and correlation depth between different indicators, the most critical indicators for the control of the film laminating machine are screened out.

[0058] Through this step, the film laminating machine can cope with complex working environments more intelligently and optimize various control parameters in the film laminating process, thereby improving efficiency and production quality.

[0059] Furthermore, the obtaining of the film laminating machine control adaptive model includes: performing performance verification and evaluation on the film laminating machine index branch controller set to obtain branch controller performance score information; and constructing a controller fusion coefficient function: ,in, Identifies the fusion coefficient of the Nth group of branch controllers, is the key coefficient information of the Nth group of branch controllers, The Nth group of branch controller performance score information is obtained; based on the controller fusion coefficient function, the branch controller key coefficient information and the branch controller performance score information are calculated to obtain a branch controller fusion coefficient set; based on the branch controller fusion coefficient set, the film laminating machine indicator branch controller set is weighted and parallel fused to obtain the film laminating machine control adaptive model.

[0060] Specifically, first, the performance verification and evaluation of different sets of branch controllers in the film laminating machine are carried out to obtain the performance score information of each branch controller. For example, the controller can be scored by measuring the response speed and film laminating error rate of the controller under temperature changes. Next, a controller fusion coefficient function is constructed, which is calculated based on the performance score information of the controller. If a controller has a higher performance score in the evaluation and its weight is larger, then its contribution to the overall system will be greater.

[0061] After obtaining the controller fusion coefficient, the key coefficient information and performance score of each branch controller are analyzed and calculated, and finally a fused controller set is obtained. This set improves the adaptability of the overall system through parallel optimization of multiple controllers. Finally, based on this multi-controller fusion coefficient, the film laminating machine indicators are efficiently fused with each branch controller to realize the adaptive control model of the film laminating machine.

[0062] Through this step, the efficient fusion of multiple controllers is achieved, and the adaptive ability of the film laminating machine is enhanced through weighted and fusion coefficients. The system can automatically optimize the operating parameters in a real-time changing environment, which not only improves the quality of film laminating, but also reduces energy consumption and error rate.

[0063] Furthermore, the monitoring, feedback and optimization control of the film pasting control status through the monitoring point group includes: real-time monitoring and feedback of the film pasting control status through the monitoring point group to obtain film pasting machine control status monitoring data, and effect evaluation of the film pasting machine control status monitoring data to obtain film pasting machine control effect parameters; comparing and analyzing the film pasting machine control effect parameters with preset control effect benchmarks to obtain control effect deviation adjustment parameters; constructing a film pasting machine adjustment strategy library, optimizing and analyzing the control effect deviation adjustment parameters through the film pasting machine adjustment strategy library, and determining control parameter strategy optimization rules; comparing and optimizing the film pasting machine operation control parameters based on the control parameter strategy optimization rules to obtain film pasting machine operation optimization control parameters, and feedback optimization control of the semiconductor automatic film pasting machine through the film pasting machine operation optimization control parameters.

[0064] Specifically, the control effect parameter is an effect indicator of the current control status of the laminating machine calculated based on the monitoring data. It evaluates whether the control performance of the laminating machine in actual operation has achieved the expected effect. The control effect deviation adjustment parameter refers to the parameter used to adjust the system to optimize it to the ideal state when the actual control effect differs from the preset baseline effect. The adjustment strategy library is a database that brings together various adjustment schemes, which is used to select the most appropriate adjustment strategy to optimize the operating parameters of the laminating machine according to the current deviation situation.

[0065] The working status data of the laminating machine is collected in real time through the monitoring point group to form the monitoring data of the control status of the laminating machine. A preliminary effect evaluation is performed on the monitoring data of the control status of the laminating machine to calculate the control effect parameters. For example, if the current film adhesion is not strong enough, the sensor will record abnormal pressure or temperature data, which will be used to generate the control effect parameters.

[0066] Next, these actually measured control effect parameters are compared and analyzed with the ideal parameters preset by the system, that is, the preset control effect benchmark, to generate the control effect deviation adjustment parameter. This deviation parameter can quantify the direction and magnitude of the adjustment.

[0067] Then, the laminating machine adjustment strategy library is called, and the corresponding adjustment rules are selected according to different deviation situations to form control parameter strategy optimization rules, which are analyzed and optimized through the stored historical data. Next, based on the optimization rules, the laminating machine's operation control parameters are adjusted and optimized to obtain new operation optimization control parameters.

[0068] Finally, the optimized control parameters are used to perform feedback optimization control on the laminating machine. This process is a closed-loop control, that is, the system continuously adjusts according to real-time feedback information to keep the laminating machine running in the best state.

[0069] This step enables automatic adaptive feedback control and automatic optimization based on real-time data and preset benchmark parameters. This real-time adjustment not only improves the accuracy and stability of the film application process, but also enables adaptive adjustments when environmental conditions change, ensuring that the quality of the film application is always in the best condition.

[0070] Furthermore, the obtaining of the optimized control parameters for the film laminating machine operation includes: mutating, expanding and updating the film laminating machine operation control parameters based on the control parameter strategy optimization rules to obtain multiple operation control variation parameters; performing effect simulation prediction on the multiple operation control variation parameters to obtain multiple film laminating control effect prediction curves; performing global comparison and optimization on the multiple operation control variation parameters based on the multiple film laminating control effect prediction curves to output the optimized control parameters for the film laminating machine operation.

[0071] Specifically, variation, expansion and update is to diversify and adjust or change the existing operation control parameters, and introduce multiple parameter "variation" versions to explore better control solutions. Effect simulation prediction is to simulate the mutated control parameters through simulation software to predict the effects of these parameters in actual operation.

[0072] First, based on the control parameter strategy optimization rules, the operation control parameters of the laminating machine are mutated, expanded and updated. The mutation and expansion here are similar to the mutation operation in the genetic algorithm, that is, by adjusting the original parameter values ​​within a certain range, multiple different parameter combinations are generated. These combinations are called operation control variation parameters.

[0073] Next, the effects of the generated multiple operation control variation parameters are simulated and predicted. Use software such as MATLAB, Simulink or ANSYS to simulate the actual film-sticking effect of different parameter combinations. The simulation results will generate multiple film-sticking control effect prediction curves, which show the performance of different parameter combinations under simulation conditions, such as the effect of pressure adjustment on film-sticking uniformity, the effect of speed adjustment on diaphragm positioning, etc.

[0074] Then, these simulation results are globally compared and optimized. This can be done through global optimization algorithms, such as genetic algorithms, particle swarm optimization algorithms, or simulated annealing. The performance of all prediction curves is compared, and the parameter combination with the best performance under different environments and conditions is selected.

[0075] Finally, the system outputs the optimal control parameters after optimization, which is the optimal parameter set obtained after variation expansion, simulation prediction, and global comparison. These parameters will be used to actually control the operation of the film laminating machine to ensure that the system can automatically adjust and maintain the best film laminating effect under different working environments.

[0076] In summary, the method for adaptively optimizing parameters of a semiconductor automatic film mounting machine based on environment perception provided in the embodiment of the present application has the following technical effects:

[0077] 1. Through the establishment of the environment perception module and 3D visual modeling, the laminating machine can accurately perceive changes in the working environment and provide real-time data input. This intelligent perception and visual modeling enhances the adaptability of the laminating machine, enabling it to adjust its working state in real time according to different environmental parameters, thereby improving its working efficiency and product quality. At the same time, it reduces the frequency of human intervention and dependence on the working environment, ensuring the stability of automated production.

[0078] 2. Through the decomposition of functional structure and attribute labeling, the functions of the film laminating machine are carefully analyzed, and appropriate sensors are selected in combination with environmental perception requirements. This precise sensor layout planning can significantly improve the machine's environmental perception ability, effectively capture key data that affects the film laminating operation, provide an accurate data basis for subsequent adaptive optimization, and improve the intelligence and flexibility of machine operation.

[0079] 3. The robustness and accuracy of the control model are improved by verifying and evaluating the controller performance and calculating the fusion coefficient. This method ensures that the system can be adaptively adjusted under different working conditions and realizes efficient collaboration of multiple controllers. By integrating the key parameters of the controller, the overall response speed and accuracy of the system are improved, which greatly improves the stability of the production process and the operation effect.

[0080] 4. Through variation expansion and simulation prediction, the system can automatically select the optimal parameter combination from multiple operation control parameters. The simulation prediction curve provides a comparison of the effects of different operating parameters to ensure that the final output optimized control parameters are the most in line with actual production needs. This optimization method improves the system's adaptive ability, ensures efficient and accurate film laminating operations, and reduces faults and errors in production.

[0081] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0082] Furthermore, the first or second mentioned above not only represents an order relationship, but also represents a specific concept, and / or refers to the selection of multiple elements individually or in whole. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application intends to include these modifications and variations.

Claims

1. Parameter adaptive optimization method of semiconductor automatic film mounting machine based on environmental perception, It is characterized in that The method comprises: Acquire structural design information of a semiconductor automatic film mounting machine, perform three-dimensional modeling on the structural design information, and generate a visual model of the film mounting machine; Based on the visualization model of the film laminating machine, a sensor group is arranged to obtain an environment perception module of the film laminating machine, wherein the environment perception module of the film laminating machine integrates a temperature sensor, a humidity sensor, a pressure sensor and a positioning sensor; Using data mining technology to obtain a film-mounting environment operation data set of the semiconductor automatic film-mounting machine, performing control indicator analysis and adaptive model training on the film-mounting environment operation data set, and obtaining a film-mounting machine control adaptive model; The film laminating machine environment perception module monitors and acquires the film laminating environment data stream, and controls and analyzes the film laminating environment data stream based on the film laminating machine control adaptive model to obtain the film laminating machine operation control parameters; A monitoring point group is set on the visual model of the film laminating machine, the semiconductor automatic film laminating machine is adaptively controlled based on the operation control parameters of the film laminating machine, and the film laminating control status is monitored and feedback optimized through the monitoring point group; Wherein, the step of obtaining the film laminating machine environment perception module includes: Based on the film laminating machine visualization model, functional structure splitting and structural attribute marking are performed to obtain a film laminating machine functional structure set and a functional structure attribute set; Performing a perception demand analysis on the functional structure set of the film laminating machine according to the functional structure attribute set of the film laminating machine to obtain the functional structure environment perception demand information; Based on the functional structure environment perception requirement information, a sensor is selected for the film laminating machine functional structure set to obtain a functional structure sensor requirement parameter set; Based on the film laminating machine visualization model, the layout position of the functional structure sensor requirement parameter set is planned to determine the layout position parameters of the functional structure sensor; The sensor group is arranged according to the functional structure sensor requirement parameter set and the functional structure sensor arrangement position parameter to obtain the film laminating machine environment perception module.

2. The method for adaptively optimizing parameters of a semiconductor automatic film mounting machine based on environmental perception as claimed in claim 1, It is characterized in that The step of obtaining the film laminating machine control adaptive model comprises: Acquire a film laminating machine control index set, classify and identify the film laminating environment operation data set according to the film laminating machine control index set, and obtain a film laminating machine index classification operation data set; Performing environmental control analysis training on the film laminating machine index classification operation data set to obtain a film laminating machine index branch controller set; Performing a multivariate criticality analysis on the film laminating machine indicator branch controller set to determine the branch controller critical coefficient information; Based on the key coefficient information of the branch controller, the film laminating machine index branch controller set is weighted and parallel fused to obtain the film laminating machine control adaptive model.

3. The method for adaptively optimizing parameters of a semiconductor automatic film mounting machine based on environmental perception as claimed in claim 2, It is characterized in that The film laminating machine control indicator set includes: Extracting structural control indicators from the film laminating environment operation data set to obtain a set of controllable indicators of the film laminating machine structure; Performing importance assessment based on the functional structure set of the film laminating machine to determine the correlation depth of multiple structural control indicators; According to the correlation depths of the multiple structural control indicators, the controllable indicator set of the film laminating machine structure is mined for related indicators to obtain a film laminating machine control indicator set.

4. The method for adaptively optimizing parameters of a semiconductor automatic film mounting machine based on environmental perception as claimed in claim 3, It is characterized in that The obtained film laminating machine control index set includes: Clustering and integrating the set of controllable indicators of the film laminating machine structure according to the set of functional structures of the film laminating machine to obtain a plurality of cluster sets of controllable indicators of the structure; Based on the film-sticking environment operation data set, correlation analysis is performed on each control indicator in the plurality of structure controllable indicator cluster sets in turn to obtain a plurality of structure indicator correlation coefficient sets; According to the plurality of structural control indicator association depths and the plurality of structural indicator correlation coefficient sets, indicator optimization is performed on each control indicator in the plurality of structural controllable indicator cluster sets to obtain the film laminating machine control indicator set.

5. The method for adaptively optimizing parameters of a semiconductor automatic film mounting machine based on environmental perception as claimed in claim 2, It is characterized in that The step of obtaining the film laminating machine control adaptive model comprises: Performing performance verification and evaluation on the film laminating machine indicator branch controller set to obtain branch controller performance score information; Construct the controller fusion coefficient function: ,in, Identifies the fusion coefficient of the Nth group of branch controllers, is the key coefficient information of the Nth group of branch controllers, The performance score information of the Nth group of branch controllers; Calculating the branch controller key coefficient information and the branch controller performance score information based on the controller fusion coefficient function to obtain a branch controller fusion coefficient set; The film laminating machine index branch controller set is weighted and fused in parallel based on the branch controller fusion coefficient set to obtain the film laminating machine control adaptive model.

6. The method for self-adaptive optimization of parameters of a semiconductor automatic film mounting machine based on environment perception as claimed in claim 1, It is characterized in that The monitoring and feedback optimization control of the film sticking control status through the monitoring point group includes: The film laminating machine control status is monitored and fed back in real time through the monitoring point group to obtain the film laminating machine control status monitoring data, and the film laminating machine control status monitoring data is evaluated to obtain the film laminating machine control effect parameters; Compare and analyze the film laminating machine control effect parameter with a preset control effect benchmark to obtain a control effect deviation adjustment parameter; Constructing a film laminating machine adjustment strategy library, optimizing and analyzing the control effect deviation adjustment parameters through the film laminating machine adjustment strategy library, and determining control parameter strategy optimization rules; Based on the control parameter strategy optimization rules, the operation control parameters of the laminating machine are compared and optimized to obtain the optimized control parameters of the laminating machine operation, and the semiconductor automatic laminating machine is feedback optimized and controlled through the optimized control parameters of the laminating machine operation.

7. The method for self-adaptive optimization of parameters of a semiconductor automatic film mounting machine based on environment perception as claimed in claim 6, It is characterized in that The step of obtaining the optimal control parameters of the laminating machine operation includes: Based on the control parameter strategy optimization rule, the operation control parameters of the film laminating machine are mutated, expanded and updated to obtain a plurality of operation control variation parameters; Performing effect simulation prediction on the multiple operation control variation parameters to obtain multiple film sticking control effect prediction curves; Based on the multiple film laminating control effect prediction curves, a global comparison and optimization is performed on the multiple operation control variation parameters, and the optimized control parameters of the film laminating machine operation are output.

Images