LED lamp automatic production monitoring control system
Through the multi-sensor collaborative monitoring and control system, multi-source data fusion and dynamic compensation in the LED lamp production process are realized, solving the problems of single detection dimensions and weak data traceability, and improving production quality and efficiency.
Patent Information
- Application Number
- CN202510539211.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-27
- Publication Date
- 2025-08-01
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing LED lamp production monitoring technology has problems such as single detection dimensions, lagging response, weak data traceability and high equipment costs, making it difficult to achieve coordinated monitoring of welding quality, equipment status and process environment, and lacks multi-source data fusion and information security mechanisms.
The multi-sensor collaborative monitoring and control system is adopted, and through the visual-current-temperature three-source data fusion technology, combined with HDR image synthesis and current waveform analysis, dynamic compensation algorithm and blockchain encryption traceability, real-time detection of welding defects, adaptive adjustment of pressure and temperature, and an encrypted traceability log is generated.
It significantly improves the detection rate of welding defects, solid crystal mounting accuracy and production efficiency, reduces the false welding rate and false alarm rate, and realizes rapid and accurate traceability of production data.
Smart Images
Figure CN120406254A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of LED Internet collaborative manufacturing, and particularly to an automatic production monitoring and control system for LED lights. Background Art
[0002] Currently, the production monitoring technology for LED lights generally has defects such as single detection dimension, lagging response, and weak data traceability. Traditional solutions mostly adopt independent visual inspection or current monitoring modules, which are difficult to achieve collaborative monitoring of welding quality, equipment status, and process environment, resulting in a high missed detection rate (about 15% - 20%). In terms of temperature control, the fixed threshold alarm mechanism cannot dynamically compensate for furnace temperature fluctuations, and abnormal handling requires shutdown and adjustment, resulting in loss of production efficiency. In addition, production data is mainly stored locally and dispersedly, lacking encryption protection and global association mechanism, and it takes up to several hours to trace the source of quality problems. More seriously, the existing technology relies on expensive equipment such as high-precision spectrometers and high-speed industrial cameras, and small and medium-sized enterprises are difficult to bear the transformation cost of over 500,000 yuan per production line. The root causes of these problems lie in the lack of multi-source data fusion, rigid control strategies, and imperfect information security architecture. Summary of the Invention
[0003] To achieve the above object, the present invention provides an automatic production monitoring and control system for LED lights. The monitoring and control system is used to execute an automatic monitoring and control method during the production of LED lights. The monitoring and control method includes: Collecting images of the welding area of LED lamp bead pins at a preset frame interval rate, obtaining the current value of the operation of the electromagnet of the die bonder at a preset sampling frequency, and arranging thermocouples at the front, middle, and rear sections in the reflow soldering furnace to obtain temperature data in real time; At a certain moment, extracting the solder joint contour information in the welding area image and performing pixel-level comparison with a pre-stored standard template, and generating a welding defect alarm signal when the image similarity is lower than the first threshold; At the same moment, dynamically calculating the pressure compensation amount of the pressing mechanism of the die bonder according to the real-time fluctuation value of the electromagnet working current; At the same moment, comparing the temperature data of the thermocouples at the front and rear sections in the reflow soldering furnace. When the temperature difference value exceeds the second threshold, automatically extend the residence time of this batch of products in the furnace, and the extended duration is calculated by multiplying the detected temperature difference value by 0.5 seconds / °C and then superimposed on the preset conveyor belt speed parameter; Associating the welding defect alarm signal, pressure compensation amount, and temperature compensation operation record with the current production batch number, and generating an encrypted traceability log including a timestamp and a station code.
[0004] Further, in the monitoring and control method, the step of collecting images of the welding area of LED lamp bead pins at a preset frame interval rate includes: Install a circularly polarized light source at a 45° inclination directly above the wire bonding station. An industrial camera is coaxially arranged at the center axis of the light source, and a narrow-band filter is installed at the front end of the camera lens. The transmission wavelength of the filter matches the wavelength of the laser positioning spot of the wire bonder; At the beginning of each production cycle, trigger the industrial camera to synchronously capture three images of the welding area with different exposure durations, where the exposure times are 5 ms, 10 ms, and 15 ms respectively; Generate a high-dynamic-range composite image from the three images through an image fusion algorithm, and identify the image parameters of the high-dynamic-range composite image; Adjust the industrial camera using the image parameters and then capture an image of the LED lamp bead pin welding area to obtain the welding area image.
[0005] Further, in the monitoring and control method, the step of generating a high-dynamic-range composite image from the three images through an image fusion algorithm includes: After eliminating the brightness and darkness differences in the three images through brightness correction, automatically assign fusion weights according to the brightness and darkness changes in each area of the image for image superposition, generate a composite image with enhanced details, and clarify the edges of the solder joints; Convert the processed composite image to a color saturation analysis mode, generate a black-and-white contrast image by automatically dividing the boundary between the solder joints and the background, and then repair the small gaps in the image to form a complete closed solder joint contour to obtain the high-dynamic-range composite image.
[0006] Further, in the monitoring and control method, the step of obtaining the current value of the die bonder electromagnet working at a preset sampling frequency includes: Connect a high-precision Hall current sensor in series in the power supply circuit of the die bonder electromagnet. The output end of the sensor is connected to a 16-bit analog-to-digital converter. Synchronously collect the current waveform data at a reference frequency of 1 kHz, and dynamically adjust the sampling frequency according to the working stage of the die bonder. The way of dynamically adjusting the sampling frequency includes maintaining a sampling frequency of 1 kHz in the wafer picking stage to capture the current rising edge characteristics; switching to a sampling frequency of 2 kHz in the wafer mounting stage to monitor the current ripple change; Extract the parameters of the current waveform data, where the parameters include the current value, calculate the current peak value, rise time, and statistical adjacent cycle peak difference degree within each working cycle.
[0007] Further, in the monitoring and control method, the step of arranging thermocouples at the front, middle, and rear sections in the reflow soldering furnace to obtain temperature data in real time includes: Install K-type thermocouple groups symmetrically along the top and bottom of the reflow soldering furnace cavity at positions 15 cm from the inlet end, the center point of the furnace body, and 15 cm from the outlet end respectively. Each group contains two temperature measurement points, one on the top and one on the bottom; Synchronously collect the thermocouple data of each segment, and set the sampling period to 0.5 seconds to obtain temperature data.
[0008] Further, in the monitoring and control method, the step of extracting the solder joint contour information in the welding area image and performing pixel-level comparison with a pre-stored standard template, and generating a welding defect alarm signal when the image similarity is lower than the first threshold, includes: Perform grayscale processing and edge sharpening operations on the welding area image, and use morphological opening operation to eliminate image noise and then extract the minimum circumscribed rectangle contour of the solder joint; Perform affine transformation registration on the extracted contour and the standard template; Compare the gray value differences of the contour areas pixel by pixel after registration, count the proportion of different pixels and calculate the similarity. When the similarity is lower than the first threshold, generate an encrypted alarm signal including defect coordinates and types.
[0009] Further, in the monitoring and control method, the step of dynamically calculating the pressure compensation amount of the die bonder pressing mechanism according to the real-time fluctuation value of the electromagnet working current includes: Collect the electromagnet working current data in real time and perform sliding window filtering processing; Calculate the percentage deviation of the current fluctuation amplitude from the nominal value; When the absolute value of the deviation exceeds the set threshold, calculate the pressure compensation amount according to the deviation direction and ratio.
[0010] Further, in the monitoring and control method, the step of comparing the temperature data of the front and rear segments of the thermocouple in the reflow soldering furnace, and automatically extending the residence time of this batch of products in the furnace when the temperature difference value exceeds the second threshold, and the extended duration is calculated by multiplying the detected temperature difference value by 0.5 seconds / °C and then superimposed on the preset conveyor belt speed parameter, includes: Obtain the temperature measurement values of the front and rear segments of the thermocouple in real time; Calculate the absolute value of the temperature difference between the two segments and compare it with the second threshold; When the temperature difference exceeds the second threshold, calculate the compensation duration by multiplying the temperature difference value by 0.5 seconds / °C; Convert the compensation duration into the conveyor belt speed adjustment amount; Update the conveyor belt operation parameters in the PLC controller and perform speed adjustment.
[0011] Further, in the monitoring and control method, the step of associating the welding defect alarm signal, the pressure compensation amount and the temperature compensation operation record with the current production batch number to generate an encrypted traceability log including the timestamp and the station code includes: Capture the welding defect alarm signal, the pressure compensation amount data and the temperature compensation operation record in real time; Bind the captured data to the current production batch number in key-value pairs; Embed the timestamp generated by the device clock and the station code read by the PLC in the data block; Perform SHA-256 hashing on the complete data packet to generate an irreversible encryption identifier; Write the encrypted log to the distributed database and synchronize it to the blockchain network nodes.
[0012] Furthermore, the system includes: An acquisition unit for collecting images of the LED lamp bead pin soldering area at a preset frame rate interval, obtaining the current value of the die bonder electromagnet working at a preset sampling frequency, and arranging thermocouples at the front, middle, and rear sections in the reflow soldering furnace to obtain temperature data in real time; An image unit for, at a certain moment, extracting the solder joint contour information in the soldering area image and performing pixel-level comparison with a pre-stored standard template, and generating a soldering defect alarm signal when the image similarity is lower than the first threshold; A pressure unit for, at the same moment, dynamically calculating the pressure compensation amount of the die bonder pressing mechanism according to the real-time fluctuation value of the electromagnet working current; A temperature sensing unit for, at the same moment, comparing the temperature data of the thermocouples at the front and rear sections in the reflow soldering furnace, and when the temperature difference value exceeds the second threshold, automatically extending the residence time of this batch of products in the furnace, and the extended duration is calculated by multiplying the detected temperature difference value by 0.5 seconds / °C and then superimposed on the preset conveyor belt speed parameter; A log unit for associating the soldering defect alarm signal, pressure compensation amount, and temperature compensation operation record with the current production batch number to generate an encrypted traceability log containing a timestamp and a station code.
[0013] The automatic production monitoring and control system for LED lamps provided by the present invention has the following beneficial effects: Through three innovations of multi-sensor collaboration, dynamic compensation algorithm, and data trusted management, the production quality and efficiency are significantly improved. By adopting the visual-current-temperature triple-source data fusion technology, combined with HDR image synthesis and current waveform feature analysis, the soldering defect detection rate is increased from 82% to 98.6%, and the die bonding and mounting accuracy is improved to ±8μm. Secondly, the innovative design of the temperature difference linear compensation and pressure adaptive adjustment algorithm controls the temperature uniformity of the reflow soldering furnace within ±1.5°C, reduces the false soldering rate to 0.7%, and at the same time shortens the compensation response speed to 300ms. Thirdly, through the blockchain encryption traceability and dynamic template update mechanism, accurate traceability of production data can be achieved within 10 minutes, and the false alarm rate is reduced by 42%. Description of the Drawings
[0014] Figure 1It is a schematic flowchart of the monitoring and control method of the LED lamp automatic production monitoring and control system in an embodiment of the present invention; Figure 2 It is a system structure block diagram of the LED lamp automatic production monitoring and control system in an embodiment of the present invention.
[0015] The realization of the purpose of the present invention, functional characteristics and advantages will be further described with reference to the embodiments and the accompanying drawings. Specific embodiments
[0016] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0017] Refer to Figure 1 It is a schematic flowchart of the automatic monitoring and control method of an LED lamp automatic production monitoring and control system proposed by the present invention. The monitoring and control system is used to execute the automatic monitoring and control method during the production of LED lamps. The monitoring and control method includes: S1, collect images of the soldering area of the LED lamp bead pins at a preset frame interval rate, obtain the current value of the operation of the electromagnet of the die bonder at a preset sampling frequency, and arrange thermocouples at the front, middle and rear sections in the reflow soldering furnace to obtain temperature data in real time; S21, at a certain moment, extract the solder joint contour information in the welding area image and perform pixel-level comparison with a pre-stored standard template. When the image similarity is lower than the first threshold, generate a welding defect alarm signal; S22, at the same moment, dynamically calculate the pressure compensation amount of the pressing mechanism of the die bonder according to the real-time fluctuation value of the electromagnet working current; S23, at the same moment, compare the temperature data of the thermocouples at the front and rear sections in the reflow soldering furnace. When the temperature difference value exceeds the second threshold, automatically extend the residence time of this batch of products in the furnace. The extended time is calculated by multiplying the detected temperature difference value by 0.5 seconds / °C and then superimposed on the preset conveyor belt speed parameter; S3, associate the welding defect alarm signal, pressure compensation amount and temperature compensation operation record with the current production batch number, and generate an encrypted traceability log including time stamps and station codes.
[0018] In one embodiment, the step of collecting images of the soldering area of the LED lamp bead pins at a preset frame interval rate includes: Install a circularly polarized light source at an inclination of 45° directly above the wire bonding station. An industrial camera is coaxially arranged at the center axis of the light source. A narrow-band filter is installed at the front end of the camera lens, and the transmission wavelength of the filter matches the wavelength of the laser positioning spot of the wire bonder; At the beginning of each production cycle, trigger an industrial camera to synchronously capture three images of the welding area with different exposure durations, where the exposure times are 5 ms, 10 ms, and 15 ms respectively; Generate a high-dynamic-range composite image from the three images through an image fusion algorithm, and identify the image parameters of the high-dynamic-range composite image; Adjust the industrial camera using the image parameters and then capture an image of the LED lamp pin welding area to obtain the welding area image.
[0019] Specifically, in one embodiment, in the monitoring and control method, the step of generating a high-dynamic-range composite image from the three images through an image fusion algorithm includes: After eliminating the brightness and darkness differences in the three images through brightness correction, automatically assign fusion weights according to the brightness and darkness changes in each area of the image for image superposition to generate a composite image with enhanced details, and perform edge sharpening on the solder joints; Convert the processed composite image to a color saturation analysis mode, generate a black-and-white contrast image by automatically dividing the boundary between the solder joints and the background, and then repair the small gaps in the image to form a complete closed solder joint contour to obtain the high-dynamic-range composite image.
[0020] During the specific implementation of this embodiment, the hardware deployment and parameter settings include 1.1 Optical system configuration Ring polarized light source: Installation angle: 45° ± 2° inclination angle directly above; Light source parameters: diameter 120 mm, color temperature 6000 K, polarization degree ≥ 98%; Wavelength matching: the center wavelength of the filter is 635 nm (matched with the wavelength of the laser positioning spot of the wire bonder, 632.8 nm), bandwidth ± 5 nm.
[0021] 1.2 Industrial camera: Model: Basler acA2000-165um (2 million pixels); Lens: 16 mm fixed-focus lens, F2.8 aperture; Trigger delay: < 1 ms (synchronized with the output pulse of the wire bonder PLC); 1.3 Multi-exposure shooting control Exposure time setting: 5 ms (short exposure), 10 ms (reference exposure), 15 ms (long exposure); Trigger timing: Synchronously trigger three consecutive shots 50 ms after the start of the production cycle, with a frame interval of 2 ms; Data stream bandwidth: Transmitted using a GigE interface, and the total transmission time of the three images ≤ 30 ms.
[0022] The implementation of image fusion algorithm fusion includes: Weight distribution function,
[0023] where I(x,y) is the gray value of pixel point (x,y); μ = 128 (the median value of 8-bit gray scale), σ = 50 (empirical parameter).
[0024] Synthetic image calculation,
[0025] The analysis of the advantages of data is as follows in the table:
[0026] The above industrial camera adjusts by using image parameters in the following way: Dynamic adjustment of brightness feedback control, by calculating the average gray value of the HDR synthetic image , target range [90, 160]
[0027] The focus evaluation function of the dynamically adjusted HDR synthetic image is:
[0028] The comparison table of taking pictures of the welding area of the LED lamp bead pins after adjusting by using image parameters is as follows:
[0029] In one embodiment, the steps of obtaining the current value of the die bonder electromagnet working at a preset sampling frequency include: Connect a high-precision Hall current sensor in series in the power supply circuit of the die bonder electromagnet. The output end of the sensor is connected to a 16-bit analog-to-digital converter, synchronously collect the current waveform data at a reference frequency of 1 kHz, and dynamically adjust the sampling frequency according to the working stage of the die bonder. The method of dynamically adjusting the sampling frequency includes maintaining a sampling frequency of 1 kHz in the wafer picking stage to capture the current rising edge feature; switching to a sampling frequency of 2 kHz in the wafer mounting stage to monitor the change of current ripple; Extract the parameters of the current waveform data, and the parameters include the current value, calculate the current peak value and rise time in each working cycle, and statistically analyze the peak difference degree between adjacent cycles.
[0030] In the specific implementation process, The design of the hardware configuration and signal chain includes: Current monitoring unit, Hall sensor: Model: LEM LAH 100-P (accuracy ±0.5%, bandwidth 200 kHz) Installation location: ≤5 cm from the electromagnet coil, shielding layer grounding impedance <0.1 Ω Analog-to-Digital Converter: Model: TI ADS8860 (16-bit, sampling rate 500kSPS) Reference Voltage: ±5V (corresponding to a measurement range of ±10A) Nonlinear Error Compensation: Second-order polynomial correction is adopted (coefficients a = 0.0032, b = 1.0015) Stage Identification Trigger Source: Method 1: Receive the IO signal output by the die bonder PLC (pickup stage = high level, placement stage = low level) Method 2: Autonomously identify the characteristics of the current waveform (rising edge slope > 50A / ms is determined as the pickup stage) The methods for extracting the parameters of the above-mentioned current waveform data include: Obtained by using moving average filtering, peak detection, time calculation, and difference evaluation. Specifically, Moving average filtering algorithm
[0031] Peak detection algorithm
[0032] Time calculation
[0033] Difference evaluation algorithm
[0034] Based on the above algorithms, the parameters of the accurate current waveform data are obtained.
[0035] In one embodiment, in the monitoring and control method, the steps of respectively arranging thermocouples at the front, middle, and rear sections in the reflow soldering furnace to obtain temperature data in real time include: Install K-type thermocouple groups symmetrically along the top and bottom of the furnace cavity at positions 15 cm from the inlet end, the center point of the furnace body, and 15 cm from the outlet end in the reflow soldering furnace cavity, and each group contains two temperature measurement points, one on the top and one on the bottom; Synchronously collect the thermocouple data of each section, and set the sampling period to 0.5 seconds to obtain temperature data.
[0036] In one embodiment, in the monitoring and control method, the steps of extracting the solder joint contour information in the welding area image and performing pixel-level comparison with a pre-stored standard template, and generating a welding defect alarm signal when the image similarity is lower than the first threshold include: Perform grayscale processing and edge sharpening operations on the welding area image, and use morphological opening operation to eliminate image noise and then extract the minimum circumscribed rectangle contour of the solder joint; Perform affine transformation registration on the extracted contour and the standard template; Compare the gray - scale values of the pixels in the contour area after registration pixel - by - pixel, count the proportion of different pixels, calculate the similarity, and generate an encrypted alarm signal containing the defect coordinates and types when the similarity is lower than the first threshold.
[0037] In the specific implementation process, the operations of grayscale processing and edge sharpening on the welding area image are as follows: Gray - scale conversion formula
[0038] Among them, the edge sharpening operator calculation is
[0039] The specific method of using morphological opening operation to eliminate image noise and then extract the minimum circumscribed rectangle contour of the solder joint is as follows: Structuring element: 3×3 cross - shaped convolution kernel
[0040] Among them, the method of contour extraction and registration is Small circumscribed rectangle: Calculate the convex hull of the contour pixel point set, and use the rotating caliper method to obtain the minimum - area bounding rectangle.
[0041] Affine transformation matrix:
[0042] Solve the transformation parameters through SIFT feature point matching (at least 3 pairs of matching points), registration error threshold: RMSE ≤2 pixels.
[0043] The method of comparing the gray - scale values of the pixels in the contour area after registration pixel - by - pixel, counting the proportion of different pixels, calculating the similarity, and generating an encrypted alarm signal containing the defect coordinates and types when the similarity is lower than the first threshold is as follows: Adopt a differential pixel determination algorithm,
[0044] Then adopt a similarity formula,
[0045] Implement the alarm trigger condition S<90%: Minor defect (record log); S<80%: Severe defect (stop and alarm).
[0046] The generated alarm signal code includes: typedef struct { uint8_t defect type; / / 0x01: cold solder joint, 0x02: offset, 0x03: bridge uint16_t X coordinate; uint16_t Y coordinate; uint32_t timestamp; } AlarmPacket;
[0047] In one embodiment, in the monitoring and control method, the step of dynamically calculating the pressure compensation amount of the die bonder pressing mechanism according to the real-time fluctuation value of the electromagnet working current includes: Real-time collection of electromagnet working current data and sliding window filtering; Calculate the percentage deviation between the current fluctuation amplitude and the nominal value; When the absolute value of the deviation exceeds the set threshold, the pressure compensation amount is calculated based on the deviation direction and ratio.
[0048] In one embodiment, in the monitoring and control method, the steps of comparing the temperature data of the front and rear thermocouples in the reflow oven and automatically extending the residence time of the batch of products in the oven when the temperature difference exceeds a second threshold value, wherein the extension time is calculated by multiplying the detected temperature difference by 0.5 seconds / °C and then added to the preset conveyor belt speed parameter, include: Get the temperature measurement values of the front and rear thermocouples in real time; Calculate the absolute value of the temperature difference between the two sections and compare it with the second threshold; When the temperature difference exceeds the second threshold, the compensation time is calculated by multiplying the temperature difference by 0.5 seconds / °C; Convert the compensation duration into a conveyor belt speed adjustment amount; Update the conveyor belt operating parameters in the PLC controller and perform speed adjustment.
[0049] In one embodiment, in the monitoring and control method, the step of associating the welding defect alarm signal, pressure compensation amount, and temperature compensation operation record with the current production batch number and generating an encrypted traceability log containing a timestamp and a workstation code includes: Real-time capture of welding defect alarm signals, pressure compensation data and temperature compensation operation records; Bind the captured data to the current production batch number as a key-value pair; Embed the timestamp generated by the device clock and the station code read by the PLC in the data block; Perform SHA-256 hash operation on the complete data packet to generate an irreversible encrypted identifier; Write encrypted logs into a distributed database and synchronize them to blockchain network nodes.
[0050] Reference appendix Figure 2 , the system includes: A collection unit, configured to collect images of the soldering area of the LED lamp bead pins at a preset frame rate interval, obtain the current value of the operation of the die bonder electromagnet at a preset sampling frequency, and arrange thermocouples at the front, middle, and rear sections in the reflow soldering furnace to obtain temperature data in real time; An image unit, configured to extract the solder joint contour information in the soldering area image at a certain moment and perform pixel-level comparison with a pre-stored standard template, and generate a soldering defect alarm signal when the image similarity is lower than a first threshold; A pressure unit, configured to dynamically calculate the pressure compensation amount of the die bonder pressing mechanism according to the real-time fluctuation value of the electromagnet working current at the same moment; A temperature sensing unit, configured to compare the temperature data of the thermocouples at the front and rear sections in the reflow soldering furnace at the same moment. When the temperature difference value exceeds a second threshold, automatically extend the residence time of this batch of products in the furnace, and the extended time is calculated by multiplying the detected temperature difference value by 0.5 seconds / °C and then superimposed on the preset conveyor belt speed parameter; A log unit, configured to associate the soldering defect alarm signal, the pressure compensation amount, and the temperature compensation operation record with the current production batch number, and generate an encrypted traceability log including a timestamp and a station code The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied to other related technical fields, shall be included in the patent protection scope of the present invention by the same token.
Claims
1. An automatic production monitoring and control system for LED lights, characterized in that, The monitoring and control system is used to execute an automated monitoring and control method during the production of LED lights. The monitoring and control method includes: Collecting images of the welding area of LED lamp bead pins at a preset frame interval rate, obtaining the current value of the die bonder electromagnet working at a preset sampling frequency, and arranging thermocouples at the front, middle, and rear sections in the reflow soldering furnace to obtain temperature data in real time; At a certain moment, extracting the solder joint contour information in the welding area image and performing pixel-level comparison with a pre-stored standard template, and generating a welding defect alarm signal when the image similarity is lower than the first threshold; At the same moment, dynamically calculating the pressure compensation amount of the die bonder pressing mechanism according to the real-time fluctuation value of the electromagnet working current; At the same moment, comparing the temperature data of the thermocouples at the front and rear sections in the reflow soldering furnace. When the temperature difference value exceeds the second threshold, automatically extend the residence time of this batch of products in the furnace, and the extended duration is calculated by multiplying the detected temperature difference value by 0.5 seconds / °C and then superimposed on the preset conveyor belt speed parameter; Associating the welding defect alarm signal, pressure compensation amount, and temperature compensation operation record with the current production batch number to generate an encrypted traceability log including a timestamp and a station code.
2. The automatic production monitoring and control system for LED lights according to claim 1, characterized in that, In the monitoring and control method, the step of collecting images of the welding area of LED lamp bead pins at a preset frame interval rate includes: Installing a circularly polarized light source at a 45° inclination directly above the wire bonding station, coaxially setting an industrial camera at the central axis of the light source, and installing a narrowband filter at the front end of the camera lens. The transmission wavelength of the filter matches the wavelength of the laser positioning spot of the wire bonder; At the beginning of each production beat, triggering the industrial camera to synchronously capture three images of the welding area with different exposure durations, where the exposure times are 5 ms, 10 ms, and 15 ms respectively; Generating a high dynamic range composite image from the three images through an image fusion algorithm and identifying the image parameters of the high dynamic range composite image; Adjusting the industrial camera using the image parameters and then shooting the welding area of the LED lamp bead pins to obtain the welding area image.
3. The automatic production monitoring and control system for LED lights according to claim 2, characterized in that, In the monitoring and control method, the step of generating a high dynamic range composite image from the three images through an image fusion algorithm includes: After eliminating the brightness and darkness differences in the three images through brightness correction, automatically assigning fusion weights according to the brightness and darkness changes in each area of the image for image superposition to generate a composite image with enhanced details, and performing edge sharpening on the solder joints; Converting the processed composite image into a color saturation analysis mode, generating a black and white contrast image by automatically dividing the boundary between the solder joints and the background, and then repairing the small gaps in the image to form a complete closed solder joint contour to obtain the high dynamic range composite image.
4. The automatic production monitoring and control system for LED lights according to claim 1, characterized in that, In the monitoring and control method, the step of obtaining the current value of the die bonder electromagnet working at a preset sampling frequency includes: A high-precision Hall current sensor is connected in series in the power supply circuit of the die bonder electromagnet. The output end of the sensor is connected to a 16-bit analog-to-digital converter to synchronously collect current waveform data at a reference frequency of 1 kHz, and the sampling frequency is dynamically adjusted according to the working stage of the die bonder. The method for dynamically adjusting the sampling frequency includes maintaining a sampling frequency of 1 kHz in the wafer picking stage to capture the current rising edge characteristics; switching to a sampling frequency of 2 kHz in the wafer mounting stage to monitor the current ripple changes; Extract the parameters of the current waveform data, where the parameters include the current value, calculate the current peak value, rise time within each working cycle, and statistically analyze the peak difference degree between adjacent cycles.
5. The automatic production monitoring and control system of the LED lamp according to claim 1, characterized in that, In the monitoring and control method, the steps of arranging thermocouples at the front, middle, and rear sections in the reflow soldering furnace to obtain temperature data in real time include: Install a K-type thermocouple group symmetrically along the top and bottom of the furnace cavity at positions 15 cm from the inlet end, the center point of the furnace body, and 15 cm from the outlet end in the reflow soldering furnace cavity, and each group contains two temperature measurement points, one on the top and one on the bottom; Synchronously collect the thermocouple data of each section, and set the sampling period to 0.5 seconds to obtain temperature data.
6. The automatic production monitoring and control system for LED lights according to claim 1, wherein In the monitoring and control method, the steps of extracting the solder joint contour information in the welding area image and performing pixel-level comparison with a pre-stored standard template, and generating a welding defect alarm signal when the image similarity is lower than the first threshold include: Perform grayscale processing and edge sharpening operations on the welding area image, and use morphological opening operation to eliminate image noise and then extract the minimum circumscribed rectangle contour of the solder joint; Perform affine transformation registration on the extracted contour and the standard template; Compare the gray value differences of the registered contour areas pixel by pixel, statistically analyze the proportion of different pixels and calculate the similarity. When the similarity is lower than the first threshold, generate an encrypted alarm signal including the defect coordinates and types.
7. The automatic production monitoring and control system for LED lights according to claim 1, wherein, In the monitoring and control method, the steps of dynamically calculating the pressure compensation amount of the die bonder pressing mechanism according to the real-time fluctuation value of the electromagnet working current include: Real-time collect the electromagnet working current data and perform sliding window filtering processing; Calculate the percentage deviation of the current fluctuation amplitude from the nominal value; When the absolute value of the deviation exceeds the set threshold, calculate the pressure compensation amount according to the deviation direction and ratio.
8. The automatic production monitoring and control system for LED lights according to claim 1, characterized in that, In the monitoring and control method, the steps of comparing the temperature data of the thermocouples at the front and rear sections in the reflow soldering furnace, and automatically extending the residence time of this batch of products in the furnace when the temperature difference value exceeds the second threshold, and the extended duration is calculated by multiplying the detected temperature difference value by 0.5 seconds / °C and then added to the preset conveyor belt speed parameter include: Real-time obtain the temperature measurement values of the thermocouples at the front and rear sections; Calculate the absolute value of the temperature difference between the two sections and compare it with the second threshold; When the temperature difference exceeds the second threshold, calculate the compensation duration by multiplying the temperature difference value by 0.5 seconds / °C; Convert the compensation duration into a conveyor belt speed adjustment amount; Update the conveyor belt operation parameters in the PLC controller and perform speed adjustment.
9. The automatic production monitoring and control system for LED lights according to claim 1, characterized in that, In the monitoring and control method, the steps of associating the welding defect alarm signal, pressure compensation amount, and temperature compensation operation record with the current production batch number to generate an encrypted traceability log including the timestamp and station code include: Real-time capture of welding defect alarm signals, pressure compensation amount data, and temperature compensation operation records; Bind the captured data to the current production batch number in key-value pairs; Embed the timestamp generated by the device clock and the station code read by the PLC in the data block; Perform SHA-256 hashing operation on the complete data packet to generate an irreversible encryption identifier; Write the encrypted log to the distributed database and synchronize it to the blockchain network nodes.
10. The automatic production monitoring and control system for LED lights according to claim 1, characterized in that The system includes: An acquisition unit for collecting images of the welding area of the LED lamp bead pins at a preset frame rate interval, obtaining the current value of the die bonder electromagnet working at a preset sampling frequency, and arranging thermocouples at the front, middle, and rear sections in the reflow soldering furnace to obtain temperature data in real time; An image unit for extracting the solder joint contour information in the welding area image at a certain moment and performing pixel-level comparison with a pre-stored standard template, and generating a welding defect alarm signal when the image similarity is lower than the first threshold; A pressure unit for dynamically calculating the pressure compensation amount of the die bonder pressing mechanism according to the real-time fluctuation value of the electromagnet working current at the same moment; A temperature sensing unit for comparing the temperature data of the thermocouples at the front and rear sections in the reflow soldering furnace at the same moment. When the temperature difference value exceeds the second threshold, automatically extend the residence time of this batch of products in the furnace, and the extended duration is calculated by multiplying the detected temperature difference value by 0.5 seconds / °C and then superimposed on the preset conveyor belt speed parameter; A log unit for associating the welding defect alarm signal, pressure compensation amount, and temperature compensation operation records with the current production batch number to generate an encrypted traceability log containing a timestamp and a station code.