Wafer automatic pickup and feedback control system based on suction nozzle flow monitoring

Abstract

The invention relates to the technical field of semiconductor manufacturing, in particular to an automatic wafer picking and feedback control system based on suction nozzle flow monitoring. Comprising a suction nozzle flow monitoring module, a data acquisition module, a data processing module, a feedback control module, a communication module, an alarm module, a power management module, a flow sensor precision analysis module and a flow anomaly detection module. Through a real-time flow monitoring and feedback control mechanism, the problem that picking failure in the semiconductor wafer carrying and loading process cannot be detected in time is solved, the safety and the operation efficiency of equipment are remarkably improved, and by combining low-energy-consumption and modular design, the system keeps operation stability and reduces energy consumption and production cost, and meanwhile, the system reliability is improved. And the requirement of efficient and automatic production is met. The system provides reliable guarantee for intelligent and automatic development of modern semiconductor manufacturing, and has wide application prospect and market value.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor manufacturing technology, and in particular to an automatic wafer picking and feedback control system based on nozzle flow monitoring. Background Art

[0002] In the process of semiconductor wafer handling and loading, the suction and placement of wafers are key links. Usually, the suction nozzle generates negative pressure through vacuum to pick up the wafer from the material box and move it to the target position.

[0003] However, during the picking process, the suction sometimes fails due to nozzle clogging, insufficient suction, or foreign matter on the wafer surface. If the wafer is not successfully picked up but the conveyor system continues to operate, it may cause equipment damage, wafer breakage, or even production suspension. Therefore, in semiconductor manufacturing, a method is needed to monitor the flow status of the nozzle in real time to determine whether the wafer is successfully picked up, and feed back the judgment result to the control system to achieve automated operation and error control. Summary of the invention

[0004] The purpose of the present invention is to solve the problem in the prior art that it is impossible to determine whether a wafer is successfully picked up, and to propose an automatic wafer picking and feedback control system based on nozzle flow monitoring.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A wafer automatic picking and feedback control system based on nozzle flow monitoring, comprising a nozzle flow monitoring module, a data acquisition module, a data processing module, a feedback control module, a communication module, an alarm module, a power management module, a flow sensor accuracy analysis module and a flow anomaly detection module, wherein the nozzle flow monitoring module comprises a flow sensor interface, an initial flow calibration unit, a flow signal amplification and filtering, and a flow stability monitoring unit, and the nozzle flow monitoring module is used to monitor the gas flow of the nozzle in real time, including the flow change before and after picking, so as to identify the picking status of the wafer;

[0007] The data acquisition module includes a data acquisition controller, a sampling rate adjustment unit, a data buffer, and a timestamp recording unit. The data acquisition module is used to collect real-time data of the nozzle flow sensor and record the collection time to form a data sequence;

[0008] The data processing module includes a data denoising unit, a filtering algorithm processor, a data standardization unit, and an abnormal flow marker. The data processing module is used to analyze the collected data to determine the wafer picking status, including calculations such as flow mean, flow difference, and fluctuation conditions;

[0009] The judgment basis includes flow difference calculation, standard deviation and variance calculation and Bayesian calculation of the probability of successful pickup. The flow difference calculation is used to calculate the flow difference between the nozzle without wafer and with wafer to judge whether the pickup is successful or not:

[0010] ΔQ=Q noWafer -Q actual

[0011] If ΔQ>ε, it is judged as absorption failure; otherwise, it is successful;

[0012] Standard deviation and variance calculation To eliminate random noise in flow data, the flow fluctuation range is calculated using the standard deviation and variance of the collected data:

[0013]

[0014] Where N is the number of samples, Q i is the i-th sampling value, The average flow rate. If the flow rate fluctuation exceeds the tolerance range, it is considered as an aspiration abnormality;

[0015] The probability of successful absorption P (Success|Q) is calculated by combining the flow difference and standard deviation through the Bayesian formula:

[0016]

[0017] Where P(Success) is the prior probability of successful absorption, P(Success) is the probability density function of traffic in the successful state, and P(Q) is the total probability of traffic;

[0018] The feedback control module includes a PID control unit, a feedback gain adjustment unit, a response delay optimizer, and a pressure compensation control unit. The feedback control module is used to perform self-priming control on the conveying system according to the judgment result, and use the PID control algorithm to adjust the negative pressure of the suction nozzle;

[0019] In feedback control, the PID controller adjusts the negative pressure P(t) according to the error signal e(t), and its control equation is as follows:

[0020]

[0021] Where K p , K i , K d are the proportional, integral and differential gains of the PID controller, e(t) = Q target -Q actual is the flow error. The PID controller adjusts the suction force in real time according to the error to ensure system stability;

[0022] The communication module includes a data transmission unit, an alarm information transmitter, a communication protocol processor, and a status feedback unit. The communication module is used to feedback the system status and picking results in real time, provide communication functions with external devices, and transmit the picking status data and control information to the monitoring terminal;

[0023] The alarm module includes an abnormal flow alarm unit, a pickup failure alarm, a multi-level alarm system, and a historical alarm recorder, and the alarm module is used to issue an audible and visual signal warning;

[0024] The power management module includes a power regulation unit, a backup power management unit, a power monitoring unit, and a power failure alarm. The power management module is responsible for the power supply of the system;

[0025] The flow sensor accuracy analysis module includes a data storage module, which includes a flow data storage, an alarm record storage, a configuration parameter storage unit, and a data backup unit. The flow sensor accuracy analysis module is used to ensure the accuracy of flow detection. The flow sensor should meet the following uncertainty conditions and calculate the uncertainty U based on the flow root mean square error (RMS):

[0026]

[0027] Where Q i is the flow measurement value, Q target is the ideal flow value, Q target is the sampling times. This formula is used to evaluate the measurement error and flow detection accuracy;

[0028] The flow anomaly detection module includes a flow difference calculator, a Bayesian anomaly detector, a control chart analyzer, and an automatic threshold adjustment unit. The flow anomaly detection module uses a control chart method to detect abnormal flow fluctuations, and selects mean and range control charts, which are represented by mean Q and range R respectively:

[0029]

[0030] R=Q max -Q min

[0031] If the mean Q exceeds the set control limit, it is considered as flow abnormality. This detection method is used to identify sudden failures during the picking process.

[0032] Preferably, the flow sensor interface is used to connect to a nozzle flow sensor and is responsible for collecting flow signals in real time;

[0033] The initial flow calibration unit is used to calibrate the flow reference when the system is started and to set the reference flow in the no-wafer pickup state;

[0034] Flow signal amplification and filtering is used to amplify and filter the collected flow signal to remove environmental noise interference;

[0035] The flow stability monitoring unit is used to monitor the stability of the nozzle flow.

[0036] Preferably, the data acquisition controller is used to control the frequency and time of data acquisition;

[0037] The sampling rate adjustment unit is used to dynamically adjust the sampling rate according to different stages of the nozzle operation;

[0038] The data buffer is used to temporarily store the collected traffic data;

[0039] The time stamp recording unit is used to add a time stamp to each collected data, so as to analyze the time series of traffic changes.

[0040] Preferably, the data denoising unit is used to perform denoising processing on the collected data;

[0041] The filter algorithm processor is used to process the flow data using a specific filter algorithm;

[0042] The data standardization unit is used to standardize the processed data;

[0043] The abnormal traffic marker is used to mark abnormal data that deviates significantly from the baseline traffic.

[0044] Preferably, the PID control unit is used to adjust the negative pressure value of the suction nozzle using a PID control algorithm;

[0045] The feedback gain adjustment unit is used to adjust the gain of PID control according to real-time flow data;

[0046] The response delay optimizer is used to reduce the feedback delay during the picking process;

[0047] The pressure compensation control unit is used to compensate and adjust the nozzle pressure in case of pick-up failure or abnormal flow.

[0048] Preferably, the data transmission unit is used to transmit the collected and processed data to a host computer or a remote monitoring system;

[0049] The alarm information transmitter is used to send an alarm signal to the host computer when an abnormality is detected;

[0050] The communication protocol processor supports multiple communication protocols;

[0051] The state feedback unit is used to feedback the system state and pick up results in real time.

[0052] Preferably, the abnormal flow alarm unit is used to trigger an alarm when the flow data exceeds a normal range;

[0053] The pickup failure alarm is used to send out an alarm reminder when a pickup failure is detected;

[0054] Multi-level alarm system sets multi-level alarms to facilitate operators to handle them in different levels;

[0055] The historical alarm recorder is used to record each alarm information.

[0056] Preferably, the power supply regulating unit is used to dynamically adjust the power supply voltage and current according to the requirements of different modules;

[0057] Backup power management is used to switch to backup power when the main power fails;

[0058] The power supply monitoring unit is used to monitor the power supply voltage and current in real time;

[0059] The power failure alarm is used to send out an alarm signal in time when a power failure is detected.

[0060] Preferably, the flow data storage device is used to store the flow data collected each time in the memory, so as to facilitate subsequent analysis and playback;

[0061] The alarm record memory stores detailed information of the alarm event;

[0062] The configuration parameter storage unit is used to store the parameter configuration file of the system;

[0063] The data backup unit is used to back up data regularly.

[0064] Preferably, the flow difference calculator determines the abnormal situation according to the difference between the current flow data and the reference flow;

[0065] The Bayesian Anomaly Detector detects traffic anomalies during the picking process through the Bayesian inference algorithm;

[0066] The control chart analyzer uses the control chart method to perform statistical monitoring of flow to determine whether the flow is stable within the normal range;

[0067] The automatic threshold adjustment unit is used to dynamically adjust the threshold of anomaly detection based on historical data and real-time data.

[0068] Compared with the prior art, the present invention has the following advantages:

[0069] 1. The present invention can judge the wafer picking status in real time according to the nozzle flow data, ensure timely feedback control of the nozzle in the event of a pickup failure, and avoid the risk of damage caused by equipment misoperation. Through real-time monitoring and feedback mechanism, production safety is improved.

[0070] 2. The present invention can accurately identify different flow states of the suction nozzle and trigger an abnormal alarm, providing clear fault information for the operator, helping to quickly troubleshoot the problem, reducing downtime, ensuring that the suction nozzle proceeds to the next step after successful picking up, avoiding damage to the equipment due to misoperation, and reducing manual intervention, thereby improving the service life and production efficiency of the equipment.

[0071] 3. The present invention supports the independent operation of multi-functional modules, and has good compatibility and scalability, is easy to maintain and upgrade, is suitable for a variety of industrial application scenarios, and has a dynamic sampling adjustment function, which can adjust the sampling rate according to different operation stages, effectively saving energy while ensuring the real-time data. At the same time, the low power consumption design effectively reduces the operating cost BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 A wafer picking judgment and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0073] Figure 2 A flow chart of a wafer picking judgment and feedback control system based on nozzle flow monitoring for a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0074] Figure 3 A schematic diagram of a nozzle flow monitoring module of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0075] Figure 4 A schematic diagram of a data acquisition module of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0076] Figure 5 A schematic diagram of a data processing module of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0077] Figure 6 A schematic diagram of a feedback control module of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0078] Figure 7 A schematic diagram of a communication module of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0079] Figure 8 A schematic diagram of an alarm module of an automatic wafer picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0080] Fig. 9A schematic diagram of a power management module of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0081] Fig.10 A schematic diagram of a data storage module of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0082] Fig.11 A schematic diagram of a flow anomaly detection module of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0083] Fig.12 This is a system feedback diagram of scenario 1 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0084] Fig.13 A system feedback diagram of scenario 2 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0085] Fig.14 A system feedback diagram of scenario 2 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0086] Fig.15 This is a system feedback diagram of scenario 3 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0087] Fig.16 This is a system feedback diagram of scenario 3 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0088] Fig.17 A system feedback diagram of scenario 4 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0089] Fig.18 A system feedback diagram of scenario 4 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0090] Fig.19 A system feedback diagram of scenario 5 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0091] Fig. 20 A system feedback diagram of scenario 5 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0092] Fig.21A system feedback diagram of scenario 6 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0093] Fig. 22 A system feedback diagram of scenario 6 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0094] Fig.23 A system feedback diagram of scenario 6 of a wafer automatic picking and feedback control system based on nozzle flow monitoring proposed by the present invention;

[0095] Fig.24 A comparison chart of nozzle flow rates under different conditions of an automatic wafer picking and feedback control system based on nozzle flow monitoring proposed by the present invention. DETAILED DESCRIPTION

[0096] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0097] Reference Figure 1-Figure 23 , a wafer automatic picking and feedback control system based on nozzle flow monitoring, including a nozzle flow monitoring module, a data acquisition module, a data processing module, a feedback control module, a communication module, an alarm module, a power management module, a flow sensor accuracy analysis module and a flow anomaly detection module, the nozzle flow monitoring module includes a flow sensor interface, an initial flow calibration unit, a flow signal amplification and filtering, and a flow stability monitoring unit, the nozzle flow monitoring module is used to monitor the gas flow of the nozzle in real time, including the flow change before and after picking up, so as to identify the picking status of the wafer;

[0098] The data acquisition module includes a data acquisition controller, a sampling rate adjustment unit, a data buffer, and a timestamp recording unit. The data acquisition module is used to collect real-time data from the nozzle flow sensor and record the collection time to form a data sequence;

[0099] The data processing module includes a data denoising unit, a filtering algorithm processor, a data standardization unit, and an abnormal flow marker. The data processing module is used to analyze the collected data to determine the wafer picking status, including calculations such as flow mean, flow difference, and fluctuation;

[0100] The judgment basis includes flow difference calculation, standard deviation and variance calculation and Bayesian calculation of the probability of successful pickup. The flow difference calculation is used to calculate the flow difference between the nozzle without wafer and with wafer to judge whether the pickup is successful or not:

[0101] ΔQ=Q noWafer -Q actual

[0102] If ΔQ>ε, it is judged as absorption failure; otherwise, it is successful;

[0103] Standard deviation and variance calculation To eliminate random noise in flow data, the flow fluctuation range is calculated using the standard deviation and variance of the collected data:

[0104]

[0105] Where N is the number of samples, Q i is the i-th sampling value, and Q is the average flow rate. If the flow rate fluctuation exceeds the tolerance range, it is considered as an aspiration abnormality;

[0106] The probability of successful absorption P (Success|Q) is calculated by combining the flow difference and standard deviation through the Bayesian formula:

[0107]

[0108] Where P(Success) is the prior probability of successful absorption, P(Success) is the probability density function of traffic in the successful state, and P(Q) is the total probability of traffic;

[0109] The feedback control module includes a PID control unit, a feedback gain adjustment unit, a response delay optimizer, and a pressure compensation control unit. The feedback control module is used to perform self-priming control on the conveying system according to the judgment result, and use the PID control algorithm to adjust the suction nozzle negative pressure;

[0110] In feedback control, the PID controller adjusts the negative pressure P(t) according to the error signal e(t), and its control equation is as follows:

[0111]

[0112] Where K p , K i , K d are the proportional, integral and differential gains of the PID controller, e(t) = Q target -Q actual is the flow error. The PID controller adjusts the suction force in real time according to the error to ensure system stability;

[0113] The communication module includes a data transmission unit, an alarm information transmitter, a communication protocol processor, and a status feedback unit. The communication module is used to feedback the system status and picking results in real time, provide communication functions with external devices, and transmit the picking status data and control information to the monitoring terminal;

[0114] The alarm module includes an abnormal flow alarm unit, a pickup failure alarm, a multi-level alarm system, and a historical alarm recorder. The alarm module is used to send out sound and light signal warnings;

[0115] The power management module includes a power regulation unit, backup power management, a power monitoring unit, and a power failure alarm. The power management module is responsible for the power supply of the system;

[0116] The flow sensor accuracy analysis module includes a data storage module, which includes a flow data storage device, an alarm record storage device, a configuration parameter storage unit, and a data backup unit. The flow sensor accuracy analysis module is used to ensure the accuracy of flow detection. The flow sensor should meet the following uncertainty conditions and calculate the uncertainty U based on the flow root mean square error (RMS):

[0117]

[0118] Where Q i is the flow measurement value, Q target is the ideal flow value, Q target is the sampling times. This formula is used to evaluate the measurement error and flow detection accuracy;

[0119] The flow anomaly detection module includes a flow difference calculator, a Bayesian anomaly detector, a control chart analyzer, and an automatic threshold adjustment unit. The flow anomaly detection module uses the control chart method to detect abnormal flow fluctuations, and selects mean and range control charts, respectively. And the range R means:

[0120]

[0121] R=Q max -Q min

[0122] If the mean If the set control limit is exceeded, it is considered as flow abnormality. This detection method is used to identify sudden failures during the picking process.

[0123] The flow sensor interface is used to connect the nozzle flow sensor and is responsible for collecting flow signals in real time;

[0124] The initial flow calibration unit is used to calibrate the flow reference when the system is started and to set the reference flow in the no-wafer pickup state;

[0125] Flow signal amplification and filtering is used to amplify and filter the collected flow signal to remove environmental noise interference;

[0126] The flow stability monitoring unit is used to monitor the stability of the nozzle flow.

[0127] The data acquisition controller is used to control the frequency and time of data acquisition;

[0128] The sampling rate adjustment unit is used to dynamically adjust the sampling rate according to different stages of the nozzle operation;

[0129] The data buffer is used to temporarily store the collected traffic data;

[0130] The time stamp recording unit is used to add a time stamp to each collected data, so as to analyze the time series of traffic changes.

[0131] The data denoising unit is used to perform denoising on the collected data;

[0132] The filter algorithm processor is used to process the flow data using a specific filter algorithm;

[0133] The data standardization unit is used to standardize the processed data;

[0134] The abnormal traffic marker is used to mark abnormal data that deviates significantly from the baseline traffic.

[0135] The PID control unit is used to adjust the negative pressure value of the suction nozzle using a PID control algorithm;

[0136] The feedback gain adjustment unit is used to adjust the gain of PID control according to real-time flow data;

[0137] The response delay optimizer is used to reduce the feedback delay during the picking process;

[0138] The pressure compensation control unit is used to compensate and adjust the nozzle pressure in case of pick-up failure or abnormal flow.

[0139] The data transmission unit is used to transmit the collected and processed data to the host computer or remote monitoring system;

[0140] The alarm information transmitter is used to send an alarm signal to the host computer when an abnormality is detected;

[0141] The communication protocol processor supports multiple communication protocols;

[0142] The state feedback unit is used to feedback the system state and pick up results in real time.

[0143] The abnormal flow alarm unit is used to trigger an alarm when the flow data exceeds the normal range;

[0144] The pickup failure alarm is used to send out an alarm reminder when a pickup failure is detected;

[0145] Multi-level alarm system sets multi-level alarms to facilitate operators to handle them in different levels;

[0146] The historical alarm recorder is used to record each alarm information.

[0147] The power supply regulation unit is used to dynamically adjust the power supply voltage and current according to the requirements of different modules;

[0148] Backup power management is used to switch to backup power when the main power fails;

[0149] The power supply monitoring unit is used to monitor the power supply voltage and current in real time;

[0150] The power failure alarm is used to send out an alarm signal in time when a power failure is detected.

[0151] The flow data storage device is used to store the flow data collected each time in the memory for subsequent analysis and playback;

[0152] The alarm record memory stores detailed information of the alarm event;

[0153] The configuration parameter storage unit is used to store the parameter configuration file of the system;

[0154] The data backup unit is used to back up data regularly.

[0155] The flow difference calculator determines abnormal conditions based on the difference between the current flow data and the reference flow;

[0156] The Bayesian Anomaly Detector detects traffic anomalies during the picking process through the Bayesian inference algorithm;

[0157] The control chart analyzer uses the control chart method to perform statistical monitoring of flow to determine whether the flow is stable within the normal range;

[0158] The automatic threshold adjustment unit is used to dynamically adjust the threshold of anomaly detection based on historical data and real-time data.

[0159] It should be noted that this system is used for automatic packaging in the pharmaceutical industry, fine flow control in the chemical industry, automatic filling in the food industry, and pollution monitoring in the environmental protection industry.

[0160] In the process of automated filling and packaging in the pharmaceutical industry, high requirements are placed on the flow control of sensitive materials such as drug powders and liquid medicines. This system can be used to monitor and control the flow distribution of medicines, ensure the accuracy of each filling amount, prevent the loss, contamination or over-dispensing of medicines due to abnormal flow, and improve the stability and safety of drug production.

[0161] In chemical production, fine flow control of some gases or liquids is very important. This system can be applied to flow monitoring and regulation in chemical reaction equipment. Combined with the flow anomaly detection module, it can detect flow anomalies in time and automatically adjust them to ensure the stability of the reaction process, improve output quality, and reduce the risk of reaction out of control.

[0162] In the automatic filling production line of the food industry, precise flow control is required to ensure that the filling amount of each container meets the standard. The multi-level alarm and feedback control function of the system can help food filling equipment monitor the flow in real time and respond promptly when abnormalities occur during the filling process, thereby reducing overflow and waste and improving product quality.

[0163] In the monitoring of pollution source emissions, this system can be used to monitor the flow of gaseous or liquid pollutants. For example, real-time monitoring and abnormal detection of industrial waste gas and wastewater flow can help timely discover and deal with excessive emissions and improve the efficiency of environmental monitoring. In conjunction with the data acquisition and alarm modules, real-time uploading of monitoring data and intelligent alarms can also be achieved.

[0164] In conjunction with the instruction manual Figure 12-Figure 23 , the following conclusions can be drawn from the nozzle flow data analysis results:

[0165] 1. Flow rate rules of the nozzle suction and release process:

[0166] Under vacuum suction, the nozzle flow rate is stable in the range of 3.20-3.25 (scenario 2), indicating that the suction force of the nozzle is in normal working state and the material is not successfully picked up.

[0167] When the nozzle successfully picks up the material, the flow rate is maintained at 3.15-3.20 (Scenarios 4 and 5). Whether the material is located in the center of the nozzle has no significant effect on the flow rate change, indicating that the system can detect whether the material is successfully picked up without judging by the specific location of the material.

[0168] In the vacuum-breaking blowing state, the flow rate is reduced to the range of 2.69-2.98 (scenario 3), which is used to identify the material release or nozzle recovery state.

[0169] 2. Automatic feedback logic for nozzle status monitoring:

[0170] In scenario 6, when the vacuum state changes from on to off, the nozzle flow rate gradually decreases to confirm whether the material is successfully placed, and the blowing process is accompanied to quickly release the material. This process realizes accurate monitoring of the picking and placing process through flow data and feedback control, effectively ensuring the stability of the material absorption and release process.

[0171] 3. The corresponding relationship between nozzle flow and operating status:

[0172] The flow characteristic values ​​of the nozzle under different working conditions are significant, and different flow values ​​can be used to determine whether the nozzle is in a normal working state and whether the material is successfully picked up or placed. This feature provides a good data basis for flow monitoring and feedback control systems, significantly improving the accuracy and intelligence of the system.

[0173] The present invention can be explained by the following operation mode:

[0174] When using this system to work, the workflow is as follows:

[0175] 1. Equipment initialization: Turn on the power of the nozzle flow monitoring system and initialize all modules of the system, including the nozzle flow monitoring module, data acquisition module, data processing module, feedback control module, communication module, alarm module and power management module;

[0176] Ensure that the communication module between the system and the control end is connected normally to ensure smooth data transmission;

[0177] After the system self-test is completed, it enters the standby state and is ready to start the detection process.

[0178] 2. Flow sensor calibration: calibrate the flow sensor to the standard state to ensure the accuracy of the measurement. The calibration reference value can be obtained by taking the average value of multiple tests;

[0179] After calibration, set the initial flow rate of the nozzle flow sensor as the baseline flow rate in the no-wafer pickup state, and store the value for subsequent comparison.

[0180] 3. Data acquisition: The nozzle flow sensor monitors the gas flow Q(t) of the nozzle in real time, and records the flow value at each time point through the data acquisition module;

[0181] At the moment of picking up, data collection begins, recording the sampling time t and flow data Q(t) to form a time-flow sequence {(t i ,Q i )};

[0182] The data sampling frequency should be set to a rate appropriate for the picking process to ensure the completeness and accuracy of the collected data.

[0183] 4. Data preprocessing: preprocess the collected traffic data, including noise removal, filtering and smoothing to reduce data errors;

[0184] Calculate the average value of flow data and variance s 2 , to obtain the fluctuation range of flow:

[0185]

[0186] Where N is the number of sampling points, variance s 2 Used to determine the stability and fluctuation range of flow.

[0187] 5. Determine the wafer picking status:

[0188] Flow difference calculation: Calculate the actual flow Q collected actual With the initial flow Q noWafer The difference ΔQ is:

[0189] ΔQ=Q noWafer -Q actual

[0190] If ΔQ is less than the set threshold ε, it is considered that the wafer has been picked up successfully, otherwise it is judged as a pick-up failure.

[0191] Bayesian probability judgment: Calculate the probability of successful pickup P (Success|Q) based on the Bayesian formula:

[0192]

[0193] The reliability of the picking state is determined by the probability value. If P(Success|Q)>η(the set reliability threshold), the picking is judged to be successful; otherwise, it is a failure.

[0194] Control chart method to detect abnormalities: Calculate the mean and range of flow and draw a control chart. If the flow value exceeds the control limit, it is judged that the flow is abnormal, which may be due to pickup failure or equipment failure.

[0195] 6. Feedback control and PID adjustment:

[0196] Feedback control signal: If it is judged that the picking fails, the feedback control module immediately issues a command to stop the conveyor belt or the conveying action of the robot arm to prevent the wafer from continuing to be conveyed without being picked up;

[0197] PID controller adjusts the suction nozzle negative pressure: If it is judged that the picking is unstable, the PID controller adjusts the suction nozzle negative pressure P(t) to restore the flow rate to the target value Q target The PID control equation is:

[0198]

[0199] Where, error e(t) = Q target -Q actual , K p , K i , K d are proportional, integral and derivative gains respectively;

[0200] The PID controller eliminates the error e(t) by adjusting the negative pressure in the feedback loop to ensure a stable picking state.

[0201] 7. Alarm and communication feedback:

[0202] Alarm module activation: If the system detects a pickup failure or abnormal flow, the alarm module will emit an audible and visual signal to remind the operator to perform inspection and maintenance;

[0203] Data feedback to the control terminal: Through the communication module, the real-time detection data, judgment results and feedback control information are transmitted to the control terminal for the operator to conduct remote monitoring;

[0204] Alarm information record: The system records the alarm time and alarm reason and saves them to the database for subsequent data analysis and troubleshooting.

[0205] 8. Complete the wafer picking process:

[0206] If the system determines that the pick-up is successful, the conveyor belt or robotic arm continues to transport the wafer to the next process station;

[0207] The system completes a picking process and returns to standby mode, waiting for the next picking command.

[0208] 9. Data storage and history:

[0209] Save the flow data, judgment results and feedback control operation records of each pickup to the system database for subsequent analysis and equipment maintenance;

[0210] The data analysis module regularly analyzes historical data and generates system status reports and trend charts to evaluate the long-term stability and maintenance needs of the equipment.

[0211] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A wafer automatic picking and feedback control system based on nozzle flow monitoring, comprising a nozzle flow monitoring module, a data acquisition module, a data processing module, a feedback control module, a communication module, an alarm module, a power management module, a flow sensor accuracy analysis module and a flow anomaly detection module, characterized in that: The nozzle flow monitoring module includes a flow sensor interface, an initial flow calibration unit, a flow signal amplification and filtering unit, and a flow stability monitoring unit. The nozzle flow monitoring module is used to monitor the gas flow of the nozzle in real time, including the flow change before and after picking up, so as to identify the picking status of the wafer; The data acquisition module includes a data acquisition controller, a sampling rate adjustment unit, a data buffer, and a timestamp recording unit. The data acquisition module is used to collect real-time data of the nozzle flow sensor and record the collection time to form a data sequence; The data processing module includes a data denoising unit, a filtering algorithm processor, a data standardization unit, and an abnormal flow marker. The data processing module is used to analyze the collected data to determine the wafer picking status, including calculations such as flow mean, flow difference, and fluctuation conditions; The judgment basis includes flow difference calculation, standard deviation and variance calculation and Bayesian calculation of the probability of successful pickup. The flow difference calculation is used to calculate the flow difference between the nozzle without wafer and with wafer to judge whether the pickup is successful or not: ΔQ=Q no Wafer -Q actual If ΔQ>ε, it is judged as absorption failure; otherwise, it is successful; Standard deviation and variance calculation To eliminate random noise in flow data, the flow fluctuation range is calculated using the standard deviation and variance of the collected data: Where N is the number of samples, Q i is the i-th sampling value, The average flow rate. If the flow rate fluctuation exceeds the tolerance range, it is considered as an aspiration abnormality; The probability of successful absorption P (Success|Q) is calculated by combining the flow difference and standard deviation through the Bayesian formula: Where P(Success) is the prior probability of successful absorption, P(Success) is the probability density function of traffic in the successful state, and P(Q) is the total probability of traffic; The feedback control module includes a PID control unit, a feedback gain adjustment unit, a response delay optimizer, and a pressure compensation control unit. The feedback control module is used to perform self-priming control on the conveying system according to the judgment result, and use the PID control algorithm to adjust the negative pressure of the suction nozzle; In feedback control, the PID controller adjusts the negative pressure P(t) according to the error signal e(t), and its control equation is as follows: Where K p , K i , K d are the proportional, integral and differential gains of the PID controller, e(t) = Q target -Q actual is the flow error. The PID controller adjusts the suction force in real time according to the error to ensure system stability; The communication module includes a data transmission unit, an alarm information transmitter, a communication protocol processor, and a status feedback unit. The communication module is used to feedback the system status and picking results in real time, provide communication functions with external devices, and transmit the picking status data and control information to the monitoring terminal; The alarm module includes an abnormal flow alarm unit, a pickup failure alarm, a multi-level alarm system, and a historical alarm recorder, and the alarm module is used to issue an audible and visual signal warning; The power management module includes a power regulation unit, a backup power management unit, a power monitoring unit, and a power failure alarm. The power management module is responsible for the power supply of the system; The flow sensor accuracy analysis module includes a data storage module, which includes a flow data storage, an alarm record storage, a configuration parameter storage unit, and a data backup unit. The flow sensor accuracy analysis module is used to ensure the accuracy of flow detection. The flow sensor should meet the following uncertainty conditions and calculate the uncertainty U based on the flow root mean square error (RMS): Where Q i is the flow measurement value, Q target is the ideal flow value, Q target is the sampling times. This formula is used to evaluate the measurement error and flow detection accuracy; The flow anomaly detection module includes a flow difference calculator, a Bayesian anomaly detector, a control chart analyzer, and an automatic threshold adjustment unit. The flow anomaly detection module uses a control chart method to detect abnormal flow fluctuations, and selects mean and range control charts, which are represented by mean Q and range R respectively: R=Q max -Q min If the mean If the set control limit is exceeded, it is considered as flow abnormality. This detection method is used to identify sudden failures during the picking process.

2. According to claim 1, a wafer automatic picking and feedback control system based on nozzle flow monitoring is characterized in that: The flow sensor interface is used to connect the nozzle flow sensor and is responsible for collecting flow signals in real time; The initial flow calibration unit is used to calibrate the flow reference when the system is started and to set the reference flow in the no-wafer pickup state; Flow signal amplification and filtering is used to amplify and filter the collected flow signal to remove environmental noise interference; The flow stability monitoring unit is used to monitor the stability of the nozzle flow.

3. The wafer automatic picking and feedback control system based on nozzle flow monitoring according to claim 1 is characterized in that: The data acquisition controller is used to control the frequency and time of data acquisition; The sampling rate adjustment unit is used to dynamically adjust the sampling rate according to different stages of the nozzle operation; The data buffer is used to temporarily store the collected traffic data; The time stamp recording unit is used to add a time stamp to each collected data, so as to analyze the time series of traffic changes.

4. The wafer automatic picking and feedback control system based on nozzle flow monitoring according to claim 1 is characterized in that: The data denoising unit is used to perform denoising processing on the collected data; The filter algorithm processor is used to process the flow data using a specific filter algorithm; The data standardization unit is used to standardize the processed data; The abnormal traffic marker is used to mark abnormal data that deviates significantly from the baseline traffic.

5. The wafer automatic picking and feedback control system based on nozzle flow monitoring according to claim 1, characterized in that: The PID control unit is used to adjust the negative pressure value of the suction nozzle using a PID control algorithm; The feedback gain adjustment unit is used to adjust the gain of PID control according to real-time flow data; The response delay optimizer is used to reduce the feedback delay during the picking process; The pressure compensation control unit is used to compensate and adjust the nozzle pressure in case of pick-up failure or abnormal flow.

6. The wafer automatic picking and feedback control system based on nozzle flow monitoring according to claim 1, characterized in that: The data transmission unit is used to transmit the collected and processed data to a host computer or a remote monitoring system; The alarm information transmitter is used to send an alarm signal to the host computer when an abnormality is detected; The communication protocol processor supports multiple communication protocols; The state feedback unit is used to feedback the system state and pick up results in real time.

7. The wafer automatic picking and feedback control system based on nozzle flow monitoring according to claim 1 is characterized in that: The abnormal flow alarm unit is used to trigger an alarm when the flow data exceeds the normal range; The pickup failure alarm is used to send out an alarm reminder when a pickup failure is detected; Multi-level alarm system sets multi-level alarms to facilitate operators to handle them in different levels; The historical alarm recorder is used to record each alarm information.

8. The wafer automatic picking and feedback control system based on nozzle flow monitoring according to claim 1, characterized in that: The power supply regulation unit is used to dynamically adjust the power supply voltage and current according to the requirements of different modules; Backup power management is used to switch to backup power when the main power fails; The power supply monitoring unit is used to monitor the power supply voltage and current in real time; The power failure alarm is used to send out an alarm signal in time when a power failure is detected.

9. The wafer automatic picking and feedback control system based on nozzle flow monitoring according to claim 1, characterized in that: The flow data storage device is used to store the flow data collected each time in the memory, so as to facilitate subsequent analysis and playback; The alarm record memory stores detailed information of the alarm event; The configuration parameter storage unit is used to store the parameter configuration file of the system; The data backup unit is used to back up data regularly.

10. The wafer automatic picking and feedback control system based on nozzle flow monitoring according to claim 1, characterized in that: The flow difference calculator determines abnormal conditions based on the difference between the current flow data and the reference flow; The Bayesian Anomaly Detector detects traffic anomalies during the picking process through the Bayesian inference algorithm; The control chart analyzer uses the control chart method to perform statistical monitoring of flow to determine whether the flow is stable within the normal range; The automatic threshold adjustment unit is used to dynamically adjust the threshold of anomaly detection based on historical data and real-time data.

Images