Braid overturning device of integrated circuit test braid machine

Abstract

The invention discloses a braid overturning device of an integrated circuit testing braid machine, and relates to the technical field of integrated circuit manufacturing equipment. A driving control assembly; the intelligent operation and maintenance module comprises a data acquisition and preprocessing unit and a data processing unit. The data acquisition and preprocessing unit comprises a plurality of sensors. The model training and predicting module comprises a deep learning model which learns the association between the operation data of the overturning braiding device main body and the fault so as to predict the occurrence of the fault; an early warning unit; an augmented reality (AR) auxiliary maintenance subunit; the communication interaction module comprises an industrial Ethernet interface unit; a data sharing and synchronizing unit; according to the overturning braiding device of the integrated circuit test braiding machine, innovation is achieved in the aspects of mechanical structure, intelligent operation and maintenance, augmented reality auxiliary maintenance and communication interaction, the braiding efficiency is improved, fault prediction and processing are achieved, and production collaboration and flexibility are enhanced.

Description

Technical Field

[0001] The invention relates to the technical field of integrated circuit manufacturing equipment, in particular to a flipping and taping device of an integrated circuit testing and taping machine. Background Art

[0002] In the current integrated circuit manufacturing industry, the traditional taping machine flipping taping device has many limitations in structure and function. On the one hand, its mechanical transmission structure is relatively complex, and the indirect transmission method adopted has problems of low efficiency and insufficient precision. In addition, the overall weight of the device is heavy, and the traditional materials used limit the flexibility and mobility of the equipment. On the other hand, due to the lack of advanced intelligent operation and maintenance methods, the monitoring and processing of equipment failures are usually in post-processing mode, and it is difficult to predict the occurrence of failures in a timely manner, resulting in frequent shutdowns due to equipment failures during the production process, seriously affecting production efficiency. In addition, in terms of data interaction with upstream and downstream equipment, the communication function of traditional devices is simple, and it is difficult to adapt to the high requirements for data real-time and coordination in modern automated production lines, which restricts the efficient operation of the entire integrated circuit production process.

[0003] In view of this, this application is hereby filed. Summary of the invention

[0004] The object of the present invention is to provide a flipping and taping device for an integrated circuit test taping machine to solve the problems mentioned in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides an integrated circuit test taping machine flipping taping device, comprising: a flipping taping device body: comprising: a mechanical structure component: the mechanical structure component comprises a high-precision flipping arm, a chip-carrying tray and a transmission mechanism, the high-precision flipping arm is driven by the transmission mechanism to realize the flipping action of the carrier tray, and is used to flip the chip from an initial position to a required position for taping; a drive control component: the drive control component comprises a motor, a controller and a sensor, the motor provides power for the transmission mechanism, the controller controls the operation of the motor, and the sensor is distributed at key positions of the flipping taping device body, and is used to monitor the operating parameters of the motor and the state parameters of the mechanical structure component; Intelligent operation and maintenance module: connected to the flipping and taping device body in communication, including: data acquisition and preprocessing unit: including: multiple sensors, arranged at different positions of the flipping and taping device body, the sensors include but are not limited to current sensors, temperature sensors, displacement sensors, angle sensors and environmental sensors, the current sensor is used to measure the running current of the motor and its fluctuation, the temperature sensor is used to monitor the temperature of the motor and key components, the displacement sensor is used to measure the displacement of the flip arm, the angle sensor is used to monitor the flip angle of the flip arm, the environmental sensor is used to detect the ambient temperature and humidity, the multiple sensors collect data at a predetermined sampling frequency, and the predetermined sampling frequency is set according to the characteristics of different sensors; a data cleaning subunit, used to remove noise and abnormal values ​​in the data collected by the sensor, and filter the collected data through an algorithm to ensure the accuracy of the data; a normalization subunit, normalizes the cleaned data, maps the data of different physical quantities to a unified standard range, and forms a historical database; Model training and prediction module: including: deep learning model: using long short-term memory network (LSTM) as the deep learning model, using the data of the historical database to train the deep learning model, the deep learning model learns the relationship between the operation data of the flipping and braiding device body and the fault to predict the occurrence of the fault; early warning unit: inputting real-time operation data into the deep learning model, when the fault prediction probability output by the deep learning model exceeds the preset threshold, the early warning unit issues a fault warning information, the fault warning information includes the components that may fail, the fault type and the estimated time of the fault occurrence; augmented reality AR assisted maintenance subunit: the augmented reality AR assisted maintenance subunit includes: AR identification components : AR identification marks are set on various components of the flip-taping device body, and the AR identification marks are used to identify different components; AR display component: maintenance personnel wear AR glasses, and by scanning the AR identification marks, the AR display component displays the maintenance information corresponding to the scanned components in the field of view of the AR glasses, and the maintenance information includes the disassembly steps of the components, required tools, operation guides and common troubleshooting methods; remote collaboration unit: when maintenance personnel encounter complex faults, the remote collaboration unit transmits the on-site images to remote experts through the network, and the remote experts send guidance information to the AR glasses through the network, and the guidance information is displayed or played in the form of images or voices to assist maintenance personnel in completing maintenance operations; Communication interaction module: connected with the flip taping device body and the intelligent operation and maintenance module, including: Industrial Ethernet interface unit: The industrial Ethernet interface unit includes a communication interface that supports PROFINET and ETHERNET / IP mainstream industrial Ethernet protocols. The communication interface is connected to the upstream and downstream automation equipment for communication, receives chip test information and production task instructions from the upstream equipment, and sends the taping progress and equipment status information to the downstream equipment; Data sharing and synchronization unit: The data sharing and synchronization unit adopts a unified data format to encapsulate the key data of the taping system, including the number of taping, yield rate, and fault information, in a predetermined data structure, and shares data with the production line central control system. The production line central control system adjusts the production scheduling and dispatching strategies according to the shared data. The data sharing and synchronization unit includes a data confirmation mechanism to ensure the accuracy and integrity of data transmission.

[0006] Furthermore, the transmission mechanism in the mechanical structure assembly adopts a direct drive mode, so that the motor directly drives the high-precision flip arm, reducing intermediate transmission links and improving transmission efficiency and precision.

[0007] Furthermore, the sensors in the data acquisition and preprocessing unit are redundantly configured, and multiple similar sensors are arranged at the same monitoring position or key component to measure the same parameter simultaneously. The reliability of the data is verified by comparing the measurement results of different sensors. When the measurement result deviation exceeds a preset range, a data abnormality alarm is triggered.

[0008] Furthermore, the training process of the deep learning model includes: dividing the data of the historical database into a training set, a validation set and a test set, adjusting the parameters of the deep learning model according to the training requirements of different stages, the training set is used for the initial training of the model, the validation set is used to optimize the parameters of the model to prevent overfitting, and the test set is used to evaluate the performance of the model. The back propagation algorithm is used during the training process to minimize the error between the predicted fault and the actual fault.

[0009] Furthermore, the AR display component in the AR-assisted maintenance subunit can adjust the position and size of the displayed information in real time according to the perspective of the maintenance personnel, ensuring that the maintenance information is clearly visible at different observation angles. The AR display component can also display virtual operation auxiliary lines to guide the maintenance personnel's operating steps.

[0010] Furthermore, during the data transmission process, the data sharing and synchronization unit in the communication interaction module sets the transmission priority of different types of data, gives high priority to key data that affects the production process, and ensures real-time transmission of key data, wherein the key data includes chip test information and equipment failure information.

[0011] Furthermore, the industrial Ethernet interface unit in the communication interaction module adopts a redundant network connection and sets a primary and backup communication link. When the primary communication link fails, it automatically switches to the backup communication link to ensure that data communication with upstream and downstream devices is not interrupted.

[0012] Furthermore, the early warning unit of the intelligent operation and maintenance module can automatically generate maintenance suggestions while issuing fault warning information. The maintenance suggestions provide corresponding maintenance operation procedures based on the predicted fault type and fault location. The maintenance operation procedures are stored in a local or cloud maintenance database for reference by maintenance personnel.

[0013] Furthermore, the carrying tray of the flip-taping device body has an adjustable positioning structure, and the position of the positioning structure is adjusted according to chips of different sizes to ensure accurate placement and positioning of chips of different sizes on the carrying tray.

[0014] Furthermore, the mechanical structural components of the flip braiding device body are made of lightweight and high-strength materials to reduce the overall weight of the device while ensuring the structural strength and stability of the device. The lightweight and high-strength materials include carbon fiber composite materials or aluminum alloy materials.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. The direct drive mode and lightweight, high-strength materials are used to reduce the intermediate transmission links, which not only improves the transmission efficiency and accuracy, but also reduces the overall weight of the device, ensures the structural strength and stability, improves the mechanical properties, makes the chip flipping and taping action more accurate and rapid, enhances the reliability and durability of the device, and thus improves the overall efficiency of the taping process; 2. Design of intelligent operation and maintenance module: By deploying multiple sensors for data collection and using deep learning models for fault prediction, warning information is issued and maintenance suggestions are automatically generated when the fault prediction probability exceeds the threshold, which greatly advances the fault discovery time and reduces unplanned downtime caused by equipment failure. At the same time, maintenance personnel can quickly locate and solve problems based on automatically generated maintenance suggestions, which improves the efficiency and reliability of equipment maintenance and reduces the impact of equipment failure on production progress; 3. Application of augmented reality-assisted maintenance subunit: By setting AR identification marks on device components, maintenance personnel wearing AR glasses can clearly see information such as component disassembly steps and tool requirements, and can also achieve remote collaboration, which reduces the difficulty of maintenance personnel in repairing complex equipment, shortens maintenance time, and ensures the accuracy and efficiency of maintenance operations. In particular, the maintenance effect for complex faults is more significant, further improving the operational stability of the entire production system; 4. Design of communication interaction module: Redundant network connection and data priority setting are adopted to ensure the reliability and real-time performance of data transmission. It can receive and send various key information, seamlessly coordinate with upstream and downstream equipment, and adjust the production scheduling and dispatching strategies according to the data of the production line central control system, so that the operation of the entire integrated circuit production line is smoother and more efficient, and the flexibility and adaptability of the production process are enhanced. The production arrangement can be dynamically optimized according to the actual situation, which improves the production coordination. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is the principle block diagram of the flipping tape device of the integrated circuit testing tape machine. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] See also Figure 1 The present invention provides a technical solution: an integrated circuit test taping machine flip taping device, comprising: 1. System installation and initialization: 1. Hardware assembly and connection: Assemble the mechanical components of the flipping and taping device, including connecting the high-precision flipping arm to the motor through the transmission mechanism to ensure a firm connection and smooth transmission. Install the carrying tray on the flipping arm so that it can stably carry the integrated circuit chip and ensure that the tray can complete the precise flipping action driven by the flipping arm.

[0019] Install the controller in the drive control assembly in a suitable position, and make electrical connections between the motor, sensor and controller. Ensure that the sensor is installed in the correct position, for example, the current sensor is closely attached to the motor power line, the temperature sensor is installed in the heat dissipation part of the motor and key components, the displacement sensor and angle sensor are installed in the corresponding position of the flip arm to accurately monitor its displacement and flip angle, and the environmental sensor is installed in the appropriate position inside the braiding machine chassis.

[0020] Install the intelligent operation and maintenance module and the communication interaction module, and connect the intelligent operation and maintenance module with the flip-taping device through a data cable to ensure that data acquisition and control signals can be transmitted normally. For the communication interaction module, connect the industrial Ethernet interface unit with upstream and downstream automation equipment (such as chip testers, placement machines, and packaging machines) through Ethernet cables to ensure reliable connections.

[0021] 2. System initialization settings: Turn on the power, start the flip taping device body, intelligent operation and maintenance module and communication interaction module. Initialize the controller, including setting the initial operating parameters of the motor, such as speed, torque, etc., and adjust the parameters according to different taping tasks and chip specifications.

[0022] In the intelligent operation and maintenance module, the sensors of the data acquisition and preprocessing unit are initialized and calibrated, and the sampling frequency of the sensors is set. For example, the current sensor is set to collect data once every millisecond, the temperature sensor is set to collect data once every 10 milliseconds, the displacement and angle sensors are set to collect data once every 20 milliseconds as needed, and the environmental sensor is set to update once every 30 seconds. At the same time, the corresponding parameters are set for the data cleaning subunit and the normalization subunit, such as the parameters of the data cleaning algorithm (such as the sliding window size, filtering threshold, etc.) and the normalization range (such as mapping different physical quantities to the [0,1] interval).

[0023] For the communication interaction module, configure the communication protocol in the industrial Ethernet interface unit, select mainstream protocols such as PROFINET or ETHERNET / IP according to the protocols supported by the upstream and downstream devices, and set the network parameters such as the IP address and port number of the communication interface. In the data sharing and synchronization unit, set a unified data format (such as XML or JSON), and determine parameters such as the data transmission frequency and the timeout period of the data confirmation mechanism.

[0024] 2. Operation and monitoring during taping operation: 1. Data collection and processing: When the taping operation begins, the sensors of the data acquisition and preprocessing unit start to collect data at a predetermined sampling frequency. The sensors transmit the collected data (motor running time, start and stop times, current fluctuations, temperature, displacement, angle, ambient temperature and humidity, etc.) to the data cleaning subunit in real time.

[0025] The data cleaning subunit uses a pre-set algorithm to clean the data. For example, it uses a sliding window filtering algorithm to remove electromagnetic interference spikes in the current data and remove abnormal jumps in the displacement and angle data caused by mechanical vibration, ensuring that the collected data is accurate and reliable.

[0026] The cleaned data enters the normalization subunit, and the data of different physical quantities are processed according to the set normalization range, mapped to a unified standard range, and form a historical database.

[0027] 2. Fault prediction and early warning: The deep learning model (LSTM) in the model training and prediction module is continuously trained using data from the historical database. At the beginning, the data from the historical database is divided into training set, validation set, and test set in a ratio of 70%:15%:15%. The back propagation algorithm is used to train the model and optimize parameters with the goal of minimizing the error between predicted faults and actual faults.

[0028] During the daily taping operation, the real-time collected and processed operation data is input into the trained LSTM model. The model calculates the probability of fault occurrence based on the input data. When the fault prediction probability output by the model exceeds the preset threshold (such as 0.8), the early warning unit will issue a fault warning information, which includes the name of the component that may fail, the prediction of the fault type, and the estimated time of the fault. The information will be sent to the control system and displayed on the operation interface. At the same time, maintenance suggestions are automatically generated. The maintenance suggestions are extracted from the maintenance database to provide reference for maintenance personnel.

[0029] 3. Use of augmented reality (AR) to assist maintenance: The AR identification marks set on each component of the taping device ensure clear identification. When maintenance personnel need to perform maintenance operations, they wear AR glasses and activate the AR display component of the AR-assisted maintenance subsystem by scanning the AR identification marks on the components.

[0030] The AR display component will retrieve the maintenance information of the corresponding component from the local or cloud database, including the disassembly steps of the component, the required tools, the operation guide and the common troubleshooting methods, and display them in the field of view of the AR glasses in the form of 3D animation, text annotation, etc. At the same time, the position and size of the displayed information will be adjusted in real time according to the maintenance personnel's perspective to ensure that the information is clearly visible, and virtual operation auxiliary lines will be displayed to guide the maintenance personnel's operation steps.

[0031] If maintenance personnel encounter complex faults, the remote collaboration unit transmits the on-site images to remote experts via the network. The remote experts send guidance information to the AR glasses via the network. The guidance information is displayed or played in the form of images or voice to assist maintenance personnel in completing maintenance operations.

[0032] 4. Communication interaction and data sharing: A: Upstream data receiving and processing steps: A1: Network initialization and adaptation: During the startup phase of the "IC test taping machine flip taping device", the industrial Ethernet interface unit automatically detects the network environment, identifies and adapts to available mainstream industrial Ethernet protocols such as PROFINET and ETHERNET / IP. It will quickly complete the settings of its own IP address, subnet mask, gateway, etc. according to the preset network configuration parameters to ensure that it is in the same communication network segment with upstream chip testers and other equipment, and seamlessly establish a connection. This feature of adaptive multi-protocol support is different from traditional single-protocol taping devices that require complex manual configuration. It demonstrates a high degree of flexibility and compatibility, and is the basis for ensuring smooth data interaction.

[0033] Start the data monitoring thread, which is responsible for real-time monitoring of data signals from upstream devices and is ready to receive key information at any time.

[0034] A2: Accurate data reception and analysis: Once the upstream equipment sends chip test information and production task instruction data packets, the industrial Ethernet interface unit uses the built-in high-performance protocol parsing engine to quickly disassemble the data packets. Taking the reception of a batch of new microprocessor chip data as an example, it can accurately extract the core performance parameters of the chip, such as the operating frequency reaching 3.5GHz, the operating voltage 1.2V, and the good product identification (the good product rate of this batch is tested to be 92%), the number of tapes is 5,000, and the tapes are required to be taped into 10 batches and other detailed information.

[0035] Different from traditional taping machines that simply receive data, the data analysis of the present invention is in-depth and intelligent, and can automatically adapt the analysis rules according to different chip types to ensure data integrity and accuracy, and avoid taping errors caused by misreading data.

[0036] A3: Real-time parameter adjustment driver: The parsed chip-related information is immediately transmitted to the controller. The controller has a built-in intelligent algorithm to quickly adjust the taping operation parameters based on the chip performance parameters. For example, since high-frequency chips are easily affected by static electricity, the static protection measures of the carrier tray are automatically enhanced; according to the number of taping and batch requirements, the flipping rhythm of the flip arm is accurately planned, and the taping speed is initially set to 300 chips per minute to ensure that the taping process is both efficient and meets quality standards. This ability to adjust dynamically in real time based on upstream data greatly improves the accuracy and adaptability of taping, far exceeding the traditional fixed parameter taping mode.

[0037] B: Downstream data transmission and interaction steps: B1: Local data collection and integration: When the taping system is running, sensors distributed in key locations continuously collect information such as taping progress and equipment status. Every time 100 chips are taped, the system automatically updates the number of taped chips. The speed sensor accurately measures the real-time movement speed of the flip arm and converts the taping speed, such as the current stable speed of 320 chips per minute. At the same time, the fault monitoring module determines in real time whether the equipment is operating normally. Once it detects that the motor current fluctuation exceeds the normal range by 10%, it immediately generates a fault warning signal and counts the yield rate during the taping process, such as the current 93%.

[0038] Integrate these scattered data into a structured data set in preparation for subsequent transmission. This all-round, real-time data collection system can provide more comprehensive and detailed operation feedback for the production line compared to traditional tape machines that only focus on a small amount of basic data.

[0039] B2: Prioritization and Packaging: The integrated data is packaged according to the predetermined data format (such as an optimized XML structure, specially designed for the characteristics of tape data, with clear labels and easy parsing). In this process, the transmission priority is innovatively set according to the key impact of the data on the production process. Chip test information and equipment failure information are given high priority. For example, if it is detected that the key performance parameters of a chip batch do not meet the standards or the equipment has a failure that may affect the chip quality, this information will be packaged and marked first to ensure that it can be delivered to the central control system as soon as possible when the network is congested.

[0040] Relatively minor data such as the progress of taping are listed as medium priority and arranged in order in the data packet. This flexible and intelligent priority strategy ensures the real-time nature of core data and effectively avoids production line decision errors caused by data transmission delays, which is lacking in traditional taping machine communications.

[0041] B3: Reliable data transmission and confirmation: The packaged data is sent to the production line central control system through the industrial Ethernet interface unit. When sending, redundant network connection technology is used to enable both the primary and backup communication links at the same time. If the primary link has packet loss or delays exceeding 10 milliseconds, it will automatically and seamlessly switch to the backup link to ensure uninterrupted data transmission.

[0042] After the data is sent, the confirmation waiting timer is started, and the preset timeout is 500 milliseconds. If a confirmation reply is received from the central control system within this time, it means that the data has been successfully received and the transmission process ends; if not, the data is resent immediately and the number of retransmissions is recorded. When the number of retransmissions reaches 3 and still fails, the manual intervention alarm is triggered to notify the technician to troubleshoot the network failure. This rigorous confirmation and retransmission mechanism is in sharp contrast to the unstable and feedback-free data transmission of traditional tape machines, and greatly improves the reliability of data transmission.

[0043] C: Central control system data processing guarantee steps: C1: Dual backup storage architecture: After receiving the data from the tape system, the production line control system first parses it according to the established format to extract key information. Then, the data is stored in two different types of storage media at the same time, such as a high-speed solid-state drive (SSD) as the main storage area for real-time data query and fast analysis; and a large-capacity mechanical hard disk (HDD) as the backup storage area for regular data archiving.

[0044] This dual backup strategy ensures that when a hardware failure occurs on any storage medium (such as SSD power failure or HDD track failure), data can be completely restored from the other storage medium, providing solid data protection for the long-term stable operation of the production line. Compared with traditional single-storage central control systems, reliability is greatly improved.

[0045] C2: Regular consistency check: The central control system automatically starts the data consistency check program every hour. By comparing the data verification, record quantity, key data values ​​and other indicators of the main storage area and the backup storage area, the data consistency is fully checked. For example, the number of taped chips in the two storage areas is compared to see if they are the same. If a difference is found, the data source is traced back based on the data update timestamp or log record, and the backup data is used for repair to ensure the accuracy and integrity of the data, effectively preventing production line scheduling chaos caused by data errors.

[0046] In the communication interaction and data sharing links, through a series of innovative designs such as multi-protocol adaptive network connection, intelligent in-depth data analysis, dynamic parameter adjustment, priority transmission, redundant and reliable transmission links, rigorous confirmation and retransmission mechanism, and dual backup verification of the central control system, the collaborative efficiency, data reliability and production adaptability of the integrated circuit test taping machine's flipping taping device and the upstream and downstream equipment of the production line are comprehensively improved. Compared with traditional technologies, it has significant advantages and provides strong support for the efficient production of integrated circuits.

[0047] 3. System maintenance and optimization: 1. Regular system maintenance: According to the fault warning information and maintenance suggestions of the intelligent operation and maintenance module, maintenance personnel perform regular maintenance on the system. For parts that may fail, according to the information provided by the AR-assisted maintenance subsystem, preventive maintenance is performed using corresponding tools and operating procedures, such as replacing motor brushes in advance based on the warning information of motor brush wear.

[0048] Regularly check each part of the mechanical structure assembly, such as whether the connection between the high-precision flip arm and the transmission mechanism is loose, whether the carrying tray is worn or deformed, tighten the loose connections, and replace or repair the worn or deformed parts.

[0049] Check the communication interaction module to ensure that the communication link of the industrial Ethernet interface unit is normal and the primary and backup communication links are working properly. If there is a primary link failure, check the cause of the failure and repair it, and automatically switch to the backup link to ensure uninterrupted communication.

[0050] 2. System optimization: As the system operation time increases, new operation data is continuously added to the historical database, and the deep learning model is regularly updated and trained using this new data to improve the model's fault prediction performance. The model can be set to be updated and trained once a week or month.

[0051] According to the feedback and data analysis of the central control system, the operating parameters of the taping system are optimized and adjusted. For example, according to the taping conditions of different batches of chips, the motor speed, the flipping speed and angle of the flip arm are adjusted to improve the taping efficiency and yield rate. At the same time, according to the information fed back by the data sharing and synchronization unit, the production scheduling and dispatching strategies of the production line are optimized. For example, when the yield rate of a batch of chips is low, the taping priority of the batch of chips is adjusted, and the batch is arranged for random inspection or rework.

[0052] In summary: through innovative mechanical structure component design, such as direct drive and the use of new materials, integrated intelligent operation and maintenance modules to achieve fault prediction and maintenance suggestions based on deep learning, combined with the optimization of augmented reality-assisted maintenance and communication interaction modules, the integrated circuit test taping machine's flipping and taping device can achieve efficient and accurate operation when flipping and taping chips, early warning and processing of faults, and highly coordinated optimization with the production line.

Claims

1. An integrated circuit test taping machine flip taping device, characterized in that: include: The main body of the flipping and taping device includes: a mechanical structure component: the mechanical structure component includes a high-precision flipping arm, a chip-carrying tray and a transmission mechanism, which flips the chip from the initial position to the required position for taping; a drive control component: the drive control component includes a motor, a controller and a sensor, and the sensor is distributed at the key parts of the main body of the flipping and taping device, and is used to monitor the operating parameters of the motor and the state parameters of the mechanical structure component; Intelligent operation and maintenance module: connected to the flipping and braiding device body in communication, including: data acquisition and preprocessing unit: including: multiple sensors, arranged at different positions of the flipping and braiding device body, the sensors include but are not limited to current sensors, temperature sensors, displacement sensors, angle sensors and environmental sensors; Model training and prediction module: including: deep learning model: using long short-term memory network as the deep learning model, the deep learning model learns the relationship between the operation data of the flipping and braiding device body and the fault to predict the occurrence of the fault; early warning unit: inputting real-time operation data into the deep learning model, when the fault prediction probability output by the deep learning model exceeds the preset threshold, the early warning unit issues fault early warning information, the fault early warning information includes the components that may fail, the fault type and the estimated time of the fault occurrence; Augmented reality AR assisted maintenance subunit: The augmented reality AR assisted maintenance subunit includes: AR identification component: AR identification marks are set on various components of the flip tape device body; AR display component: maintenance personnel wear AR glasses; remote collaboration unit; Communication interaction module: connected to the flip taping device body and the intelligent operation and maintenance module, including: Industrial Ethernet interface unit: the industrial Ethernet interface unit includes a communication interface that supports PROFINET and ETHERNET / IP mainstream industrial Ethernet protocols; Data sharing and synchronization unit: shares the key data of the taping system with the production line central control system.

2. The flipping and taping device of the integrated circuit test taping machine as claimed in claim 1, characterized in that: The transmission mechanism in the mechanical structure assembly adopts a direct drive mode, so that the motor directly drives the high-precision flip arm.

3. The flipping and taping device of the integrated circuit test taping machine as claimed in claim 1, characterized in that: The sensors in the data acquisition and preprocessing unit are redundantly configured, with multiple sensors of the same type arranged at the same monitoring position or key component to measure the same parameter simultaneously. The reliability of the data is verified by comparing the measurement results of different sensors. When the measurement result deviation exceeds a preset range, a data abnormality alarm is triggered.

4. The flipping and taping device of the integrated circuit test taping machine as claimed in claim 1, characterized in that: The training process of the deep learning model includes: dividing the data in the historical database into a training set, a validation set and a test set, and adjusting the parameters of the deep learning model according to the training requirements of different stages.

5. The flipping and taping device of the integrated circuit test taping machine as claimed in claim 1, characterized in that: The AR display component in the AR-assisted maintenance subunit can adjust the position and size of the displayed information in real time according to the viewing angle of the maintenance personnel.

6. The flipping and taping device of the integrated circuit test taping machine as claimed in claim 1, characterized in that: The data sharing and synchronization unit in the communication interaction module sets the transmission priority of different types of data during the data transmission process.

7. The flipping and taping device of the integrated circuit test taping machine as claimed in claim 1, characterized in that: The industrial Ethernet interface unit in the communication interaction module adopts a redundant network connection and sets a primary and backup communication link to ensure that data communication with upstream and downstream devices is not interrupted.

8. The flipping and taping device of the integrated circuit test taping machine as claimed in claim 1, characterized in that: The early warning unit of the intelligent operation and maintenance module can automatically generate maintenance suggestions while issuing fault early warning information.

9. The flipping and taping device of the integrated circuit test taping machine as claimed in claim 1, characterized in that: The carrying tray of the flipping and braiding device body has an adjustable positioning structure, and the position of the positioning structure is adjusted according to chips of different sizes.

Images