Gluing monitoring method for capacitive touch sensor
By collecting and analyzing the real-time data and environmental data of the automatic dispenser, and combining with pre-trained models for comprehensive monitoring, the monitoring problem of glue injection speed, position and area in sensor glue coating monitoring is solved, and the accuracy and reliability of monitoring are improved adapt to different environmental conditions, reduce manual intervention and reduce costs.
Patent Information
- Application Number
- CN202411468227.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-21
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing sensor glue coating monitoring methods are difficult to achieve comprehensive monitoring and analysis of glue injection speed, glue coating position and glue coating area, resulting in low accuracy and reliability of monitoring results, especially when environmental factors such as temperature and humidity change. .
By collecting the single glue injection time, substrate weight change information and operating environment data of the automatic dispenser, input it to the pre-trained glue injection speed analysis model and glue coating surface image analysis model to obtain the glue injection speed, glue coating position and coating The abnormality monitoring value of the glue area is measured, and the weight comprehensive analysis is carried out, and the comparison is carried out in combination with the preset abnormality threshold, so as to realize multi-dimensional monitoring of glue coating operations.
It improves the monitoring accuracy and reliability of glue coating operations, reduces the interference of environmental factors on monitoring results, and can promptly detect abnormal situations, reduce the generation of bad products, improve production efficiency, and reduce labor costs.
Smart Images

Figure CN119056698A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of sensor detection, and in particular to a capacitive tactile sensor gluing monitoring method. Background Art
[0002] Capacitive tactile sensors play an increasingly important role in the field of modern industrial manufacturing, especially in precision machinery, electronic equipment and automation systems. During the manufacturing process of capacitive tactile sensors, glue needs to be applied on the substrate in order to stick the capacitor sheet on the substrate. The uniformity and accuracy of the glue coating have a direct impact on the performance of the sensor.
[0003] The existing sensor glue coating monitoring method often only performs qualified monitoring of the glue coating area through visual technology processing after the automatic dispensing machine completes the glue injection. However, changes in environmental factors such as temperature and humidity will affect the fluidity of the glue and thus affect the glue coating quality. The existing monitoring method is difficult to achieve comprehensive monitoring and analysis of the glue injection speed, glue coating position and glue coating area, resulting in low accuracy and reliability of the monitoring results. Summary of the invention
[0004] In order to solve the above technical problems, the present invention provides a capacitive tactile sensor gluing monitoring method which can significantly improve the monitoring level of gluing operation and ensure the quality and reliability of sensor manufacturing.
[0005] In a first aspect, the present invention provides a method for monitoring glue coating of a capacitive tactile sensor, the method comprising:
[0006] Collect the preset single glue injection time of the automatic glue dispensing machine, the weight change information of the substrate during the glue injection process, and the glue coating operation environment data information;
[0007] The single glue injection time preset by the automatic glue dispensing machine, the weight change information of the substrate during the glue injection process, and the glue coating operation environment data information are input into the pre-trained glue injection speed analysis model to obtain the glue injection speed abnormal monitoring quantitative value;
[0008] Collect the image of the substrate coating surface after the automatic dispensing machine completes the glue injection;
[0009] Inputting the substrate glue coating surface image into the pre-trained glue coating surface image analysis model to obtain the glue coating position abnormality monitoring quantitative value and the glue coating area abnormality monitoring quantitative value;
[0010] Perform a weighted comprehensive analysis on the abnormal monitoring quantitative values of the glue injection speed, the abnormal monitoring quantitative values of the glue coating position and the abnormal monitoring quantitative values of the glue coating area to obtain the abnormal monitoring quantitative values of the glue coating operation;
[0011] The abnormal monitoring quantitative value of the gluing operation is compared with the preset abnormal threshold to obtain the gluing monitoring result of the capacitive tactile sensor.
[0012] Furthermore, the method for obtaining the quantitative value of the abnormal monitoring of the injection speed includes:
[0013] Collect glue type data information of the capacitive touch sensor to be produced;
[0014] Divide the single glue injection time preset by the automatic glue dispensing machine into multiple weight collection time points;
[0015] According to the multiple weight collection time points, the glue injection weight on the substrate is collected in sequence, and the glue injection speed between adjacent weight collection time points is calculated to obtain the glue injection speed in multiple time periods;
[0016] Based on the glue coating operation environment data information and glue type data information, the corresponding glue injection speed standard range is extracted from the pre-built glue injection speed database;
[0017] Compare the injection speeds of multiple time periods with the standard range of injection speeds in turn, extract the time periods where the injection speeds are not within the standard range of injection speeds, and mark them as abnormal injection speed periods;
[0018] All periods of abnormal injection speed are summarized and analyzed in proportion to the duration of a single injection to obtain the quantitative value of abnormal injection speed monitoring.
[0019] Furthermore, the method for obtaining the quantitative value of monitoring the abnormality of the glue coating position and the quantitative value of monitoring the abnormality of the glue coating area includes:
[0020] Collect specification data information of capacitive touch sensors;
[0021] Based on the specification data information, the pre-built glue coating plane feature database is traversed to extract the corresponding glue coating position standard range and glue coating area standard range;
[0022] Performing image recognition processing on the glue-coated surface image of the substrate to obtain the glue-coated position and glue-coated area; the glue-coated position is the position vector between the center of the glue-coated portion and the center of the substrate;
[0023] The error amplitudes between the gluing position and the standard range of the gluing position, and between the gluing area and the standard range of the gluing area are calculated respectively to obtain the quantitative value of the abnormal monitoring of the gluing position and the quantitative value of the abnormal monitoring of the gluing area.
[0024] Furthermore, the method for constructing the injection speed analysis model includes:
[0025] Collect historical data about the injection process, including data on normal and abnormal conditions;
[0026] Clean the collected data to remove noise, outliers and missing values;
[0027] Selecting a deep learning model as the basic architecture of the injection speed analysis model, wherein the deep learning model includes linear regression, logistic regression, random forest, gradient boosting tree and neural network;
[0028] The cleaned data is divided into training set, validation set and test set; the training set is used to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the final performance of the model;
[0029] Use the training set data to train the selected model and adjust the model parameters through an iterative optimization algorithm to minimize the loss function;
[0030] Use the validation set to evaluate the performance of the model and adjust the model's hyperparameters based on the performance feedback of the validation set;
[0031] Evaluate the performance of the ensemble model on the validation set;
[0032] The trained model is deployed in the actual production environment and integrated with the automatic dispensing machine system to receive input data in real time and output quantitative values for abnormal monitoring of injection speed.
[0033] Furthermore, the method for collecting the image of the substrate coating surface after the automatic glue dispensing machine completes the glue injection includes:
[0034] Wait for the automatic glue dispensing machine to complete the entire glue injection process and receive a completion signal from the automatic glue dispensing machine;
[0035] After the glue injection is completed, the substrate is accurately positioned and fixed;
[0036] Select an industrial camera as the image acquisition device and configure supplementary lighting equipment;
[0037] When the substrate is stable and the light source lighting conditions meet the requirements, the industrial camera is started to collect images; the camera captures the image of the glue-coated surface and saves it as a digital file;
[0038] After the image is saved, the image quality is verified; if the image quality does not meet the requirements, the light source and camera settings are readjusted to acquire the image;
[0039] The collected image data is transferred to the computer.
[0040] Furthermore, the structure of the glue-coated surface image analysis model includes:
[0041] Input layer: used to receive the substrate image after glue coating;
[0042] The first convolutional layer: uses a small-sized convolution kernel with a step size of 1 for feature extraction;
[0043] The second convolution layer: uses a small-sized convolution kernel to further extract complex features;
[0044] Pooling layer: set after each convolutional layer to reduce the spatial dimension and retain the main features;
[0045] Flattening layer: used to flatten the output of the last pooling layer into a one-dimensional vector;
[0046] Output layer: used to output the quantitative value of abnormal monitoring of glue coating position and the quantitative value of abnormal monitoring of glue coating area.
[0047] Furthermore, the calculation formula for the quantitative value of abnormal monitoring of the gluing operation is:
[0048] Q=w s ×V s +w p ×V p +w a ×V a
[0049] Among them, Q represents the quantitative value of abnormal monitoring of glue coating operation, w s represents the weight of the injection speed abnormal monitoring quantization value, V s Indicates the quantitative value of injection speed abnormal monitoring, w p Represents the weight of the quantitative value of the abnormal monitoring of the glue coating position, V p Indicates the quantitative value of abnormal monitoring of glue coating position, w a Represents the weight of the quantitative value of the abnormal monitoring of the glue coating area, V a Indicates the quantitative value of abnormal monitoring of glue coating area.
[0050] Furthermore, factors affecting the setting of the preset abnormal threshold include product specification quality, production process standards, historical data analysis, environmental factors, equipment performance stability, raw material characteristics, failure frequency and cost analysis.
[0051] On the other hand, the present application provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and the computer program implements the steps of any one of the above methods when executed by the processor.
[0052] In a third aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps in any one of the above-mentioned methods when executed by a processor.
[0053] Compared with the prior art, the present invention has the following beneficial effects: the method not only focuses on the monitoring of the results after the glue injection is completed, but also realizes the full-process and multi-dimensional monitoring of the glue injection speed, glue coating position and glue coating area by collecting real-time data and working environment data during the glue injection process, thereby improving the accuracy and reliability of the monitoring;
[0054] Taking into account the impact of environmental factors such as temperature and humidity on glue fluidity and glue coating quality, this method effectively reduces the interference of environmental factors on monitoring results and improves the environmental adaptability of the system by introducing operating environment data information and combining it with a pre-trained model for analysis;
[0055] The pre-trained model is used to quantitatively evaluate the injection speed, glue coating position and glue coating area, and a weighted comprehensive analysis mechanism is introduced to more scientifically evaluate the overall quality of the glue coating operation, making the monitoring results more objective and accurate.
[0056] By comparing the abnormal monitoring quantitative value of the gluing operation with the preset abnormal threshold, the abnormal situation in the gluing operation can be discovered in time and clear monitoring results can be given; it is helpful to quickly respond to and deal with problems, reduce the generation of defective products, and improve production efficiency;
[0057] The entire monitoring process is highly automated and intelligent, reducing the impact of manual intervention and subjective judgment. Data analysis through pre-trained models improves the efficiency and accuracy of monitoring and reduces labor costs. The design of this method is scalable and flexible, and can be adjusted and optimized according to different gluing requirements and operating environments.
[0058] In summary, this method has significant advantages in solving the problem of gluing monitoring of capacitive tactile sensors, can significantly improve the monitoring level of gluing operations, and ensure the quality and reliability of sensor manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] Figure 1 is a flow chart of the present invention;
[0060] Figure 2 is a flow chart of a method for obtaining a quantitative value for monitoring abnormal injection speed;
[0061] Figure 3 The invention is a flow chart of a method for obtaining quantitative monitoring values of abnormal glue coating position and abnormal glue coating area. DETAILED DESCRIPTION
[0062] In the description of the present application, those skilled in the art should know that the present application can be implemented as a method, an apparatus, an electronic device, and a computer-readable storage medium. Therefore, the present application can be specifically implemented in the following forms: complete hardware, complete software (including firmware, resident software, microcode, etc.), a combination of hardware and software. In addition, in some embodiments, the present application can also be implemented in the form of a computer program product in one or more computer-readable storage media, and the computer-readable storage medium contains computer program code.
[0063] The above-mentioned computer-readable storage medium may adopt any combination of one or more computer-readable storage media. Computer-readable storage media include: electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media include: portable computer disks, hard disks, random access memories, read-only memories, erasable programmable read-only memories, flash memories, optical fibers, optical disc read-only memories, optical storage devices, magnetic storage devices, or any combination of the above. In the present application, computer-readable storage media can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device.
[0064] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws.
[0065] The present application describes the provided methods, devices, and electronic devices through flowcharts and / or block diagrams.
[0066] It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine, and these computer-readable program instructions are executed by a computer or other programmable data processing device to produce a device that implements the functions / operations specified by the boxes in the flowchart and / or block diagram.
[0067] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to work in a specific manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction device product including functions / operations specified in the blocks in the flowchart and / or block diagram.
[0068] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device, so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby enabling the instructions executed on the computer or other programmable data processing apparatus to provide a process for implementing the functions / operations specified in the blocks in the flowchart and / or block diagram.
[0069] The present application is described below in conjunction with the drawings in the present application.
[0070] like Figures 1 to 3 As shown, a capacitive tactile sensor glue coating monitoring method of the present invention specifically includes the following steps:
[0071] S1, collecting the preset single glue injection time of the automatic glue dispensing machine, the weight change information of the substrate during the glue injection process, and the glue coating operation environment data information;
[0072] The duration of a single injection is an important indicator for evaluating the stability of the injection speed. By recording and analyzing the preset injection time, it is possible to preliminarily determine whether the dispensing machine is operating according to the established parameters when performing the task, and whether the speed is too fast or too slow.
[0073] The single glue injection duration directly reads the preset single glue injection duration data from the control system of the automatic glue dispensing machine; these data are usually stored in the control unit of the glue dispensing machine or the data management system connected thereto;
[0074] The change in substrate weight directly reflects the amount of glue injected. By real-time monitoring of the weight change of the substrate during the injection process, the actual amount of glue injected can be indirectly calculated and compared with the theoretical value to evaluate the accuracy and uniformity of the injection.
[0075] The weight change information of the substrate during the glue injection process is measured by using a high-precision electronic scale or balance placed under the glue dispensing workstation, measuring the weight of the substrate before and after glue injection, and calculating the difference;
[0076] Environmental factors such as temperature and humidity have a significant impact on the fluidity and curing speed of glue, which in turn affects the quality of glue coating. Therefore, collecting data on the glue coating environment is an indispensable part of evaluating the quality of glue coating.
[0077] The environmental data information of the gluing operation is obtained by installing temperature and humidity sensors in the gluing operation area to record and transmit environmental parameters such as temperature and humidity in real time. The sensors should have high accuracy and stability to ensure that the collected environmental data is accurate and reliable. At the same time, the installation location and number of sensors must also be considered to ensure that the environmental conditions of the operation area can be fully reflected.
[0078] In this step, by directly reading the preset single glue injection time of the automatic glue dispenser and comparing it with the actual execution situation, it is possible to quickly identify whether the glue injection speed meets the preset standard and whether there is a problem of too fast or too slow speed; this real-time and accurate monitoring helps to adjust the working parameters of the glue dispenser in time, ensure the stability of the glue injection speed, and thus improve the overall efficiency and quality of the glue coating operation; using a high-precision electronic scale or balance to monitor the weight change of the substrate in the glue injection process in real time, the actual amount of glue injected can be accurately calculated and compared with the theoretical value; it can not only intuitively reflect the accuracy of the glue injection amount, but also evaluate the uniformity of the glue coating through the uniformity of the weight distribution; by installing temperature and humidity sensors, the environmental parameters in the glue coating operation area are recorded and transmitted in real time, and the influence of environmental factors on the characteristics of glue fluidity and curing speed can be fully understood; this environmental monitoring helps to timely discover and correct glue coating quality problems caused by environmental factors, and improve the reliability and stability of the glue coating operation; the data collected in step S1 provides comprehensive and accurate data support for subsequent comprehensive evaluation; these data can comprehensively reflect the status and quality of the glue coating operation through comprehensive analysis, and provide a strong basis for optimizing the glue coating process and improving the manufacturing quality of sensors.
[0079] S2, inputting the single glue injection time preset by the automatic glue dispensing machine, the weight change information of the substrate during the glue injection process, and the glue coating operation environment data information into the pre-trained glue injection speed analysis model to obtain the glue injection speed abnormal monitoring quantitative value;
[0080] The method for constructing the injection speed analysis model comprises:
[0081] Collect a large amount of historical data about the glue injection process, including data on normal and abnormal conditions; this data should include the preset single glue injection time of the automatic glue dispensing machine, the actual glue injection time, the weight change information of the substrate during the glue injection process, the glue coating operation environment data information, and the corresponding glue injection speed abnormality label;
[0082] Clean the collected data to remove noise, outliers, missing values, etc. to ensure data quality;
[0083] Select the appropriate model type according to the characteristics of the problem and the complexity of the data; for monitoring the abnormal injection speed, select the deep learning model as the basic architecture of the injection speed analysis model, which includes linear regression, logistic regression, random forest, gradient boosting tree and neural network;
[0084] The cleaned data is divided into training set, validation set and test set; the training set is used to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the final performance of the model;
[0085] Use the training set data to train the selected model and adjust the model parameters through an iterative optimization algorithm to minimize the loss function;
[0086] Use the validation set to evaluate the performance of the model, and adjust the model's hyperparameters based on the performance feedback of the validation set to further improve the model performance;
[0087] Evaluate the performance of the ensemble model on the validation set and select an appropriate ensemble strategy;
[0088] Deploy the trained model to the actual production environment, integrate it with the automatic dispensing machine system, receive input data in real time and output quantitative values for monitoring abnormal injection speed;
[0089] The method for obtaining the quantitative value of abnormal monitoring of injection speed includes:
[0090] Record in detail the type of glue used for each capacitive tactile sensor, including its physical and chemical properties such as viscosity, fluidity, and curing time; collect and record in real time environmental parameters such as temperature, humidity, and air pressure during the gluing operation; environmental factors directly affect the fluidity and curing speed of the glue, thereby affecting the injection speed and gluing quality;
[0091] Divide the preset single injection time of the automatic dispensing machine into multiple equally or unequally spaced weight collection time points in a more detailed manner; the selection of time points should be based on the glue characteristics and injection process requirements to ensure that key changes in the injection process can be captured; use a high-precision electronic scale or other weighing equipment to accurately measure the injection weight on the substrate at each weight collection time point; ensure that the measurement accuracy meets the monitoring requirements to reduce errors;
[0092] Calculate the injection speed in each time period based on the injection weight difference and time difference between adjacent weight collection time points; smooth the injection speed curve to eliminate noise and abnormal fluctuations; at the same time, analyze the change trend and characteristic points of the curve to provide a basis for subsequent abnormality detection;
[0093] Based on historical data and experimental data, a database containing standard ranges of injection speeds under different glue types and environmental conditions is pre-built; the database should cover the range, fluctuations, and possible abnormal patterns of normal injection speeds; according to the current glue coating operation environment data information and glue type data information, the corresponding standard range of injection speeds is searched and extracted in the database;
[0094] Compare the calculated injection speeds of multiple time periods with the extracted injection speed standard range in turn; if the injection speed of a certain time period exceeds the standard range, mark it as an abnormal injection speed period; for each abnormal injection speed period, its abnormality can be further evaluated; summarize all the abnormal injection speed periods and calculate their proportion to the entire single injection time, which is the quantitative value of injection speed abnormality monitoring;
[0095] The abnormal monitoring quantitative value of injection speed is fed back to the production control system or operator in time so that timely measures can be taken for adjustment and optimization; according to the monitoring results and actual conditions, the injection speed analysis model and the glue coating surface image analysis model are continuously optimized to improve the accuracy and reliability of monitoring.
[0096] In this step, by real-time monitoring of the injection speed and quantifying the abnormal monitoring value, the speed abnormality in the injection process can be discovered and corrected in time, thereby avoiding problems such as uneven, excessive or insufficient glue coating caused by improper speed, and improving product consistency and quality stability; at the same time, reducing rework and scrap caused by product defects and improving production efficiency; applying advanced artificial intelligence technologies such as deep learning to injection speed monitoring realizes the intelligence and automation of the production process; not only reduces the burden of manual monitoring, but also improves the accuracy and real-time nature of monitoring, making the production process more efficient and controllable; by collecting and analyzing a large amount of historical data, an injection speed analysis model is established to provide data support for decision-making in the production process; enterprises can monitor the injection speed abnormality according to the output of the model, and Adjust production parameters or perform equipment maintenance in time to achieve refined management based on data; through real-time monitoring and prediction of injection speed anomalies, potential problems can be discovered and solved in advance to prevent failures; it helps to reduce production interruptions and losses caused by equipment failures and improve the stability and reliability of the production line; the constructed injection speed analysis model can be adjusted and optimized according to different production needs and environmental conditions to meet the requirements of different products and processes; this flexibility and adaptability enables the production line to better respond to market changes and diversified customer needs; this step achieves precise control and abnormal monitoring of injection speed by introducing intelligent monitoring technology and data analysis methods, significantly improves production efficiency and product quality, and promotes the development of the manufacturing industry towards intelligence, automation and refinement.
[0097] S3, collecting the image of the substrate coating surface after the automatic dispensing machine completes the glue injection;
[0098] The method for collecting an image of a substrate coating surface after the automatic glue dispensing machine completes the glue injection comprises:
[0099] Wait for the automatic glue dispensing machine to complete the entire glue injection process and receive a completion signal from the automatic glue dispensing machine;
[0100] After the glue injection is completed, the substrate is accurately positioned and fixed to ensure that the position of the substrate is stable and accurate during image acquisition;
[0101] Select a high-resolution, high-sensitivity industrial camera as the image acquisition device; the camera should be selected based on the characteristics of the coating surface and the required image resolution and accuracy;
[0102] And configure appropriate fill lighting solutions to eliminate shadows, reduce reflections, and enhance the contrast between the glue-coated area and the substrate;
[0103] When the substrate is stable and the light source illumination conditions meet the requirements, the camera is started to acquire images through software or hardware trigger mechanism;
[0104] The camera captures an image of the coated surface and saves it as a digital file; the saved image should contain sufficient details to allow subsequent image analysis and processing;
[0105] After the image is saved, image quality verification is performed to ensure that the image is clear, without blur, shadow or reflection. If the image quality does not meet the requirements, the light source and camera settings need to be readjusted and the image acquisition needs to be repeated.
[0106] The collected image data is transferred to a computer for subsequent image analysis of the glue-coated surface.
[0107] In this step, by accurately collecting the image of the glue-coated surface of the substrate after the glue injection is completed, key parameters such as the uniformity, position and area of the glue coating can be intuitively observed, so as to more accurately evaluate the glue coating quality; it overcomes the limitations of traditional monitoring methods that only rely on visual inspection or single parameter monitoring, and improves the comprehensiveness and accuracy of monitoring; by configuring a suitable fill light scheme, it can effectively eliminate shadows, reduce reflections, and enhance the contrast between the glue-coated area and the substrate, so that the image acquisition process is less sensitive to environmental factors; it helps to maintain stable image acquisition quality in different working environments and improve the environmental adaptability and reliability of the monitoring system; this step realizes the automation of image acquisition, and reduces human intervention and errors by receiving the completion signal of the automatic dispensing machine and triggering the camera to acquire the image; this not only improves production efficiency, but also reduces labor costs, making the entire glue coating monitoring process more efficient and reliable; acquisition The collected high-quality image data provides a solid foundation for subsequent image analysis and processing; through the pre-trained glue coating surface image analysis model, it can quickly and accurately identify problems such as abnormal glue coating position, insufficient or excessive glue coating area, and give quantitative evaluation results; it helps to discover problems in time and take corresponding corrective measures to ensure the manufacturing quality of capacitive tactile sensors; since this step uses high-resolution, high-sensitivity industrial cameras as image acquisition devices, and supports flexible configuration of light sources and camera settings, it can be adjusted and optimized according to different glue coating requirements and substrate characteristics; it helps to improve the flexibility and adaptability of the production line and meet the customized needs of different customers and products; this step has significant beneficial effects in the glue coating monitoring process of capacitive tactile sensors, which not only improves the accuracy and reliability of monitoring, but also improves production efficiency and flexibility, providing a strong guarantee for the manufacturing quality of sensors.
[0108] S4, inputting the substrate coating surface image into a pre-trained coating surface image analysis model to obtain a coating position abnormality monitoring quantitative value and a coating area abnormality monitoring quantitative value;
[0109] The structure of the glue-coated surface image analysis model includes:
[0110] Input layer: used to receive the substrate image after glue coating;
[0111] The first convolutional layer: uses a small-sized convolution kernel with a step size of 1 for feature extraction;
[0112] The second convolution layer: uses a small-sized convolution kernel to further extract more complex features;
[0113] Pooling layer: After each convolutional layer, there is a max pooling layer to reduce the spatial dimension and retain the main features;
[0114] Flattening layer: used to flatten the output of the last pooling layer into a one-dimensional vector;
[0115] Output layer: used to output the quantitative value of abnormal monitoring of glue coating position and the quantitative value of abnormal monitoring of glue coating area; the two nodes correspond to the quantitative value of abnormal monitoring of glue coating position and the quantitative value of abnormal monitoring of glue coating area respectively;
[0116] The method for obtaining the quantitative value of the abnormal monitoring of the gluing position and the quantitative value of the abnormal monitoring of the gluing area includes:
[0117] Record the specific position requirements of the capacitor plate, the precise shape of the glue coating, and the expected glue coating thickness; considering the changes in glue properties under different environmental conditions, it is necessary to collect standard glue coating data under different environmental conditions for more accurate comparison and analysis;
[0118] In the glue coating plane feature database, not only the standard range of glue coating position and glue coating area should be stored, but also the multi-dimensional information such as the outline of glue coating shape and the distribution of glue coating thickness should be included; a dynamic update mechanism for the database should be established to continuously adjust and optimize the standard range of glue coating position and glue coating area according to the actual data and feedback in the production process to adapt to changes in production conditions;
[0119] Before image recognition, the substrate glue surface image is preprocessed, including denoising, contrast enhancement, edge detection, etc., to improve image quality and facilitate subsequent processing; image processing technology is used to extract the features of the glue coating area, including the precise coordinates of the glue coating position, the outline and area of the glue coating area, and possible defects; the extracted glue coating shape is matched with the preset standard shape to determine the precise vector of the glue coating position; at the same time, image segmentation technology is used to accurately measure the glue coating area;
[0120] The error definitions of the gluing position and the gluing area are clarified. According to the error definitions, the error amplitudes between the gluing position and the standard range of the gluing position, and between the gluing area and the standard range of the gluing area are calculated respectively. For the position error, the components in multiple directions need to be considered. For the area error, the numerical difference or percentage is directly calculated. The calculated error amplitude is compared with the preset threshold. If the error exceeds the threshold, it is determined that the gluing position or the gluing area is abnormal, and the corresponding monitoring quantitative value is given.
[0121] The quantitative monitoring values of the abnormal glue coating position and the abnormal glue coating area are fed back to the production control system or operators in real time so that timely measures can be taken for adjustment and optimization; based on the monitoring results and actual production conditions, the glue coating plane feature database, image recognition algorithm and error calculation model are continuously optimized to improve the accuracy and reliability of monitoring.
[0122] In this step, the glue coating surface image analysis model can automatically extract multi-level and multi-dimensional features from the glue coating surface image of the substrate through multi-layer convolution and pooling operations. The automated feature extraction method is more accurate and efficient than the traditional method. The model can process and analyze the substrate image after receiving the glue coating, and quickly output the quantitative monitoring value of the abnormal glue coating position and the quantitative monitoring value of the abnormal glue coating area. This enables problems in the production process to be discovered and handled in a timely manner, avoiding the generation of defective products and waste of resources. At the same time, since the model has been trained and optimized with a large amount of data, the monitoring results it outputs are highly accurate. Taking into account the changes in glue properties under different environmental conditions, this step collects standard glue coating data under different environmental conditions and establishes a dynamically updated glue coating plane feature database. This enables the model to adapt to changes in production conditions more accurately and improves the adaptability and stability of monitoring. The quantitative values of position abnormality monitoring and glue coating area abnormality monitoring can be fed back to the production control system or operators in real time; this intelligent feedback mechanism enables problems in the production process to be responded to and handled quickly, thereby improving production efficiency and product quality; at the same time, according to the monitoring results and actual production conditions, the glue coating plane feature database, image recognition algorithm and error calculation model can be continuously optimized to further improve the accuracy and reliability of monitoring; this step not only focuses on the accuracy of the glue coating position and glue coating area, but also uses image segmentation and other technologies to comprehensively monitor and analyze the contour, area and possible defects of the glue coating area; it can more comprehensively reflect the glue coating quality status and provide strong support for quality control in the production process; this step realizes real-time, accurate and comprehensive monitoring and analysis of the glue coating quality of the capacitive tactile sensor by introducing the glue coating surface image analysis model, providing a strong guarantee for improving product quality and production efficiency.
[0123] S5. Perform a weighted comprehensive analysis on the abnormal monitoring quantization value of the glue injection speed, the abnormal monitoring quantization value of the glue coating position, and the abnormal monitoring quantization value of the glue coating area to obtain the abnormal monitoring quantization value of the glue coating operation;
[0124] The weights are assigned according to the degree of influence of each factor on the final product quality. The weight factor needs to be determined based on the relative importance of each factor in the coating process on the final sensor performance.
[0125] The factors affecting the weight distribution of the abnormal monitoring quantization value of the glue injection speed, the abnormal monitoring quantization value of the glue coating position, and the abnormal monitoring quantization value of the glue coating area include:
[0126] Product specifications: Capacitive touch sensors of different specifications and models have different requirements for glue coating quality, including glue coating speed, position and area accuracy; therefore, process standards and technical requirements will directly affect the distribution of weights;
[0127] Production standards: The production standards within the industry or enterprise also stipulate the key indicators of glue coating quality and their importance. These standards can be used as the basis for weight allocation;
[0128] Production process control capability: The stronger the control capability of the gluing process on the production line, that is, the more stable the process, the lower the sensitivity to certain indicators may be, and the corresponding weights will also be adjusted;
[0129] Variability analysis: Analyze the variability of parameters such as gluing speed, position and area through historical data. Parameters with greater variability require higher weights because they have a greater impact on the quality of the final product.
[0130] Environmental factors: Changes in temperature and humidity will directly affect the fluidity of the glue, thereby affecting the quality of glue coating; for parameters that are easily affected by the environment, a higher weight may need to be assigned;
[0131] Fault severity: evaluate the potential impact of abnormalities in different coating parameters on product performance or reliability. Parameters with greater impact should be given higher weights.
[0132] Fault frequency: Based on historical data, evaluate the frequency of abnormal occurrence of each parameter. Frequently occurring abnormalities also require higher weights;
[0133] Cost considerations: Taking into account the cost of monitoring and control, higher weights will be assigned to monitoring parameters with higher cost effectiveness;
[0134] Production efficiency: The weight distribution should also take into account the impact on production efficiency to avoid excessive focus on certain details that may lead to a decrease in overall production efficiency;
[0135] The calculation formula for the quantitative value of abnormal monitoring of gluing operation is:
[0136] Q=w s ×V s +w p ×V p +w a ×V a
[0137] Among them, Q represents the quantitative value of abnormal monitoring of glue coating operation, w s represents the weight of the injection speed abnormal monitoring quantization value, V s Indicates the quantitative value of injection speed abnormal monitoring, w p Represents the weight of the quantitative value of the abnormal monitoring of the glue coating position, V p Indicates the quantitative value of abnormal monitoring of glue coating position, w a Represents the weight of the quantitative value of the abnormal monitoring of the glue coating area, V a Indicates the quantitative value of abnormal monitoring of glue coating area.
[0138] In this step, by comprehensively considering the abnormal monitoring quantitative values of multiple key parameters such as glue injection speed, glue coating position and glue coating area, and allocating weights based on the relative importance of each parameter on the quality of the final product, the overall quality of the glue coating operation can be evaluated more comprehensively, and the actual situation of the glue coating operation can be accurately reflected, thereby improving the accuracy of the monitoring results; the weight allocation takes into account multiple aspects, making the weight allocation more scientific and reasonable, thereby enhancing the reliability of the monitoring results; even in a complex and changeable production environment, it can maintain a high level of monitoring accuracy and stability; the impact of production efficiency is taken into account when allocating weights, avoiding the problem of excessive attention to certain details that leads to a decrease in overall production efficiency; by reasonably allocating weights, the production efficiency can be increased as much as possible while ensuring the quality of glue coating, Reduce production costs; weight allocation is based on different product specifications, production standards and process requirements, so it has high flexibility; when production conditions or product requirements change, the weight factor can be adjusted to adapt to new needs to ensure the accuracy and reliability of monitoring results; through comprehensive analysis of the quantitative values of abnormal monitoring of gluing operations, problems and bottlenecks in the gluing process can be discovered in time, providing strong support for the continuous improvement of the production process; at the same time, it can also provide data support and reference basis for subsequent process optimization and equipment upgrades; the comprehensive weight analysis in step S5 has significant beneficial effects in the gluing monitoring process of capacitive tactile sensors, which can improve monitoring accuracy, enhance reliability, optimize production efficiency, flexibly adapt to different needs and promote continuous improvement.
[0139] S6. Compare the abnormal monitoring quantization value of the gluing operation with the preset abnormal threshold value to obtain the gluing monitoring result of the capacitive tactile sensor;
[0140] The factors affecting the setting of the preset abnormal threshold include:
[0141] Product specification quality: Capacitive touch sensors of different models and specifications have different requirements for glue coating quality; therefore, the preset abnormal threshold needs to be set according to the technical specifications and performance requirements of the specific product;
[0142] Production process standards: The production process standards within the industry or the company stipulate the acceptable range of glue coating quality. These standards can serve as an important basis for setting abnormal thresholds;
[0143] Historical data analysis: By analyzing the data of historical gluing operations, we can understand the distribution and variability of parameters such as gluing speed, position and area; based on these data, we can set reasonable thresholds to distinguish between normal and abnormal gluing operations;
[0144] Environmental factors: Environmental factors such as temperature and humidity have a significant impact on the fluidity of glue, which in turn affects the quality of glue coating. Therefore, the preset abnormal threshold needs to take into account the fluctuation range of environmental factors and their impact on the glue coating operation;
[0145] Equipment performance stability: The performance stability and accuracy of the automatic dispensing machine directly affect the quality of glue coating. The aging, wear or calibration status of the equipment may affect the accuracy of the glue coating parameters. Therefore, when setting the abnormal threshold, the actual performance and stability of the equipment need to be considered.
[0146] Raw material characteristics: The type, viscosity, curing time and other characteristics of the glue will also affect the gluing effect; different types of glue require different gluing parameters and abnormal thresholds;
[0147] Fault frequency: evaluate the frequency of various anomalies in the gluing operation and their potential impact on the final product quality or production process; for anomalies that occur frequently and have serious consequences, a lower threshold should be set to facilitate timely detection and handling;
[0148] Cost analysis: The setting of preset abnormal thresholds also needs to consider the cost of monitoring and control as well as production efficiency; too high a threshold will lead to missed detections, while too low a threshold will increase the false alarm rate, affecting production efficiency and costs.
[0149] In this step, by setting a scientific and reasonable abnormal threshold and comparing it with the abnormal monitoring quantitative value obtained in the actual gluing operation, it is possible to accurately judge whether the gluing operation meets the quality requirements. Compared with the traditional experience judgment or single parameter monitoring, it has higher accuracy and reliability; the setting of the preset abnormal threshold takes into account multiple factors, making quality control more systematic and scientific, and helps to promptly discover and correct problems in the gluing operation, thereby improving the overall quality control capability; reasonable abnormal threshold setting can reduce unnecessary downtime inspections and rework while ensuring the quality of gluing, thereby improving production efficiency; by promptly identifying and handling abnormal gluing operations, it is possible to avoid problems The problem products flow into the next process, reducing the subsequent processing costs and time; cost analysis plays an important role in the setting of preset abnormal thresholds; by balancing the relationship between monitoring and control costs and production efficiency, an abnormal threshold that is both economical and effective can be set; it helps to reduce the additional costs caused by excessive monitoring or missed detection, thereby reducing the overall production cost; by analyzing the comparison results of the quantitative value of the abnormal monitoring of the gluing operation with the preset abnormal threshold, the problems and bottlenecks in the gluing process can be identified; it provides an important basis for the continuous improvement of the production process, helps to optimize the production process, improve equipment performance, improve raw material selection, etc., thereby further improving the quality and efficiency of the gluing operation.
[0150] In addition, the present application also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of the above-mentioned method for controlling output data is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.
[0151] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A capacitive tactile sensor glue coating monitoring method, characterized in that: The method comprises: Collect the preset single glue injection time of the automatic glue dispensing machine, the weight change information of the substrate during the glue injection process, and the glue coating operation environment data information; The single glue injection time preset by the automatic glue dispensing machine, the weight change information of the substrate during the glue injection process, and the glue coating operation environment data information are input into the pre-trained glue injection speed analysis model to obtain the glue injection speed abnormal monitoring quantitative value; Collect the image of the substrate coating surface after the automatic dispensing machine completes the glue injection; Inputting the substrate glue coating surface image into the pre-trained glue coating surface image analysis model to obtain the glue coating position abnormality monitoring quantitative value and the glue coating area abnormality monitoring quantitative value; Perform a weighted comprehensive analysis on the abnormal monitoring quantitative values of the glue injection speed, the abnormal monitoring quantitative values of the glue coating position and the abnormal monitoring quantitative values of the glue coating area to obtain the abnormal monitoring quantitative values of the glue coating operation; The abnormal monitoring quantitative value of the gluing operation is compared with the preset abnormal threshold to obtain the gluing monitoring result of the capacitive tactile sensor.
2. A capacitive tactile sensor glue coating monitoring method as claimed in claim 1, characterized in that: The method for obtaining the quantitative value of abnormal monitoring of injection speed includes: Collect glue type data information of the capacitive touch sensor to be produced; Divide the single glue injection time preset by the automatic glue dispensing machine into multiple weight collection time points; According to the multiple weight collection time points, the glue injection weight on the substrate is collected in sequence, and the glue injection speed between adjacent weight collection time points is calculated to obtain the glue injection speed in multiple time periods; Based on the glue coating operation environment data information and glue type data information, the corresponding glue injection speed standard range is extracted from the pre-built glue injection speed database; Compare the injection speeds of multiple time periods with the standard range of injection speeds in turn, extract the time periods where the injection speeds are not within the standard range of injection speeds, and mark them as abnormal injection speed periods; All periods of abnormal injection speed are summarized and analyzed in proportion to the duration of a single injection to obtain the quantitative value of abnormal injection speed monitoring.
3. A capacitive tactile sensor glue coating monitoring method as claimed in claim 2, characterized in that: The method for obtaining the quantitative value of the abnormal monitoring of the gluing position and the quantitative value of the abnormal monitoring of the gluing area includes: Collect specification data information of capacitive touch sensors; Based on the specification data information, the pre-built glue coating plane feature database is traversed to extract the corresponding glue coating position standard range and glue coating area standard range; Performing image recognition processing on the glue-coated surface image of the substrate to obtain the glue-coated position and glue-coated area; the glue-coated position is the position vector between the center of the glue-coated portion and the center of the substrate; The error amplitudes between the gluing position and the standard range of the gluing position, and between the gluing area and the standard range of the gluing area are calculated respectively to obtain the quantitative value of the abnormal monitoring of the gluing position and the quantitative value of the abnormal monitoring of the gluing area.
4. A capacitive tactile sensor glue coating monitoring method as claimed in claim 1, characterized in that: The method for constructing the injection speed analysis model comprises: Collect historical data about the injection process, including data on normal and abnormal conditions; Clean the collected data to remove noise, outliers and missing values; Selecting a deep learning model as the basic architecture of the injection speed analysis model, wherein the deep learning model includes linear regression, logistic regression, random forest, gradient boosting tree and neural network; The cleaned data is divided into training set, validation set and test set; the training set is used to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the final performance of the model; Use the training set data to train the selected model and adjust the model parameters through an iterative optimization algorithm to minimize the loss function; Use the validation set to evaluate the performance of the model and adjust the model's hyperparameters based on the performance feedback of the validation set; Evaluate the performance of the ensemble model on the validation set; The trained model is deployed in the actual production environment and integrated with the automatic dispensing machine system to receive input data in real time and output quantitative values for abnormal monitoring of injection speed.
5. The method for monitoring glue coating of a capacitive tactile sensor according to claim 1, characterized in that: The method for collecting an image of a substrate coating surface after the automatic glue dispensing machine completes the glue injection comprises: Wait for the automatic glue dispensing machine to complete the entire glue injection process and receive a completion signal from the automatic glue dispensing machine; After the glue injection is completed, the substrate is accurately positioned and fixed; Select an industrial camera as the image acquisition device and configure supplementary lighting equipment; When the substrate is stable and the light source lighting conditions meet the requirements, the industrial camera is started to collect images; the camera captures the image of the glue-coated surface and saves it as a digital file; After the image is saved, the image quality is verified; if the image quality does not meet the requirements, the light source and camera settings are readjusted to acquire the image; The collected image data is transferred to the computer.
6. The method for monitoring glue coating of a capacitive tactile sensor according to claim 1, characterized in that: The structure of the glue-coated surface image analysis model includes: Input layer: used to receive the substrate image after glue coating; The first convolutional layer: uses a small-sized convolution kernel with a step size of 1 for feature extraction; The second convolution layer: uses a small-sized convolution kernel to further extract complex features; Pooling layer: set after each convolutional layer to reduce the spatial dimension and retain the main features; Flattening layer: used to flatten the output of the last pooling layer into a one-dimensional vector; Output layer: used to output the quantitative value of abnormal monitoring of glue coating position and the quantitative value of abnormal monitoring of glue coating area.
7. The method for monitoring glue coating of a capacitive tactile sensor according to claim 1, characterized in that: The calculation formula for the quantitative value of abnormal monitoring of gluing operation is: Q=w s ×V s +w p ×V p +w a ×V a Among them, Q represents the quantitative value of abnormal monitoring of glue coating operation, w s represents the weight of the injection speed abnormal monitoring quantization value, V s Indicates the quantitative value of abnormal injection speed monitoring, w p Represents the weight of the quantitative value of the abnormal monitoring of the glue coating position, V p Indicates the quantitative value of abnormal monitoring of glue coating position, w a Represents the weight of the quantitative value of the abnormal monitoring of the glue coating area, V a Indicates the quantitative value of abnormal monitoring of glue coating area.
8. The method for monitoring glue coating of a capacitive tactile sensor according to claim 1, characterized in that: The factors affecting the setting of the preset abnormal threshold include product specification quality, production process standards, historical data analysis, environmental factors, equipment performance stability, raw material characteristics, failure frequency and cost analysis.
9. An electronic device for monitoring glue coating of a capacitive tactile sensor, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, characterized in that: When the computer program is executed by the processor, the steps in the method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 7 are implemented.
Citation Information
Cited By
Anti-oxidation silver-filled glass millet bead UV glue applying intelligent monitoring control system
CN119689947A
