An automated control method and system for an LED chip production line

Abstract

This invention relates to the field of automation control technology, specifically to an automation control method and system for an LED bead production line. The method includes the following steps: collecting data to obtain the average thrust value to form a batch status; configuring a sliding window to perform regression calculations to obtain trend characteristics by continuously decreasing frequency; combining background thresholds to determine random fluctuations or actual drift results; determining the control level and generating compensation instructions based on the average value and remaining margin; executing adjustments and setting transition period data collection and verification indicators to update control records. In this invention, by configuring dynamic monitoring intervals to extract evolution features and identify deep drift to block fluctuation interference, the intervention level is pre-determined based on deviation amplitude and remaining margin, and fine-tuning instructions are issued to overcome rigid passive limitations. After adjustment, the control path is continuously corrected through transition verification comparison to form a closed-loop feedforward mechanism, transforming lag alarms into early interventions to suppress process edge deviations and prevent batch welding defects.

Description

Technical Field

[0001] This invention relates to the field of automation control technology, and in particular to an automation control method and system for an LED bead production line. Background Technology

[0002] This invention relates to the field of automation control technology. After the reflow soldering process in the LED chip production line, a thrust tester is usually configured to conduct random checks on the welding strength of the LED chips on the substrate and upload the test data to the central control system. The existing system calculates the average thrust data of the batch and compares it with a preset lower limit value for acceptable thrust. If the average value is greater than the lower limit value, the batch is deemed to be welded and the current temperature or chain speed parameters of the reflow oven are kept unchanged. If it is lower than the lower limit value, an alarm is triggered and parameters are required to be adjusted. In actual continuous production, the heating elements inside the reflow oven may slowly age or the heat capacity of the bearing fixture may change, resulting in a change in the actual heat absorbed by the LED chips on the substrate. This physical state drift is reflected in the thrust test as a continuous and gradual downward trend in the batch average value.

[0003] The existing system performs single-point comparison judgment based on the current batch average and static lower limit. This evaluation mode breaks the continuous changing trajectory, resulting in the loss of the ability to perceive slow heat decay. When the average value approaches the qualified boundary but does not penetrate the lower limit, the system continues to maintain the original process operation, causing the welding heat absorption to show a hidden decline. The single rigid threshold triggering mechanism causes the system to lose the early pre-compensation window. The production line operates in a critical state, causing the cumulative deviation to explode and triggering a delayed shutdown alarm. The passive response logic causes defective products to flow into downstream processes and increase the investigation cost. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and to propose an automated control method and system for LED bead production lines.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an automated control method for an LED chip production line, comprising the following steps:

[0006] S1: Apply lateral shear force to the lamp bead solder joints using a thrust testing machine, collect raw thrust data, remove outliers to obtain valid test data, calculate the arithmetic mean to obtain the thrust mean, and associate it with batch number, test time and reflow oven process parameters to form batch thrust status data;

[0007] S2: Configure a sliding window based on the batch thrust state data, perform linear regression and continuous decrease statistics on the thrust mean within the sliding window, and calculate the cumulative offset and remaining margin according to the preset benchmark steady-state mean, the preset static thrust qualified lower limit and the thrust mean, to obtain the change slope, cumulative offset, remaining margin and the number of continuously decreasing batches, forming thrust trend characteristic data;

[0008] S3: Based on the thrust trend feature data, by comparing the change slope and the number of consecutive decreasing batches with the preset background noise threshold range and the preset continuity judgment threshold respectively, random fluctuation results or real drift results are generated to obtain drift identification results;

[0009] S4: Based on the drift identification results and the thrust trend characteristic data, the reflux furnace control level is determined by comparing the average thrust with the preset static thrust qualification lower limit and comparing the remaining margin with the preset margin threshold.

[0010] The reflux furnace control level is one of the following: alarm shutdown level, parameter hold level, micro-compensation level, or enhanced compensation level.

[0011] S5: Based on the reflow furnace control level and the thrust trend characteristic data, generate corresponding reflow furnace control commands according to the reflow furnace control level;

[0012] The reflux furnace control command is one of the following: parameter hold command, micro-compensation command, enhanced compensation command, or alarm shutdown command.

[0013] S6: When the reflow oven control command is a micro-compensation command or a strong compensation command, adjust the reflow oven process parameters and set a transition period. After the transition period, collect the average thrust of the verification batch to form compensation verification thrust data. Calculate the verification slope and verification margin based on the compensation verification thrust data. Maintain or update the reflow oven control level and the reflow oven control command based on the verification slope and verification margin, and output the LED production line control record.

[0014] As a further aspect of the present invention, the process of S1 is specifically as follows:

[0015] S111: Apply lateral shear force to the lamp bead solder joint using a thrust testing machine, detect the peak thrust when the lamp bead solder joint breaks under the action of lateral shear force, collect the peak thrust value at break, and obtain the raw thrust data;

[0016] S112: Outlier values ​​are removed from the original thrust data to obtain valid test data. The arithmetic mean of the valid test data is calculated to obtain the thrust mean. The thrust mean is correlated with the batch number, test time and reflow furnace process parameters to establish batch thrust status data.

[0017] The reflow furnace process parameters in the batch thrust status data are the adjustment targets for parameter compensation corresponding to the reflow furnace control level.

[0018] As a further aspect of the present invention, the process of S2 is specifically as follows:

[0019] S211: Call the batch thrust status data, determine the order of the average thrust values ​​of each batch according to the batch number and detection time, configure a sliding window, and perform linear regression calculation on the average thrust values ​​within the sliding window to obtain the slope of change;

[0020] S212: Based on the preset benchmark steady-state average value, calculate the difference between the average thrust value and the benchmark steady-state average value to obtain the cumulative offset; based on the preset static thrust qualification lower limit value, calculate the difference between the average thrust value and the static thrust qualification lower limit value to obtain the remaining margin.

[0021] S213: Count the number of consecutive decreases in the average thrust of adjacent batches within the sliding window to obtain the number of consecutive decrease batches. Then, associate the change slope, the cumulative offset, the remaining margin, and the number of consecutive decrease batches to establish thrust trend feature data.

[0022] As a further aspect of the present invention, the process of S3 is specifically as follows:

[0023] S311: Call the thrust trend feature data, compare the change slope with the preset background noise threshold range, and compare the number of consecutive decreasing batches with the preset continuity determination threshold to obtain drift comparison data;

[0024] S312: Based on the drift comparison data, when the change slope is not lower than the lower bound of the background noise threshold interval and the number of consecutive decreasing batches is less than the continuity determination threshold, a random fluctuation result is generated; when the change slope is lower than the lower bound of the background noise threshold interval or the number of consecutive decreasing batches is not less than the continuity determination threshold, a true drift result is generated, and a drift identification result is obtained.

[0025] As a further aspect of the present invention, the process of S4 is specifically as follows:

[0026] S411: Invoke the drift identification results and the thrust trend feature data;

[0027] When the average thrust is not greater than the preset lower limit of static thrust, it is determined to be an alarm shutdown level;

[0028] When the average thrust is greater than the preset static thrust qualification lower limit and the drift identification result is the random fluctuation result, it is determined to be a parameter maintenance level;

[0029] When the average thrust is greater than the preset static thrust qualification lower limit, the drift identification result is the actual drift result, and the remaining margin is greater than the preset margin threshold, it is determined to be a micro-compensation level.

[0030] When the average thrust is greater than the preset static thrust qualification lower limit, the drift identification result is the actual drift result, and the remaining margin is not greater than the preset margin threshold, it is determined to be an enhanced compensation level;

[0031] S412: The determined alarm shutdown level, parameter hold level, micro-compensation level, or enhanced compensation level shall be used as the reflow oven control level.

[0032] As a further aspect of the present invention, the process of S5 is specifically as follows:

[0033] S511: Invoke the reflow oven control level and the thrust trend characteristic data, and generate corresponding instructions based on the reflow oven control level, including:

[0034] When the reflow oven control level is the parameter holding level, a parameter holding command is generated;

[0035] When the reflow oven control level is the alarm shutdown level, an alarm shutdown command is generated, and the board feeding stop control is executed according to the alarm shutdown command;

[0036] When the reflow oven control level is the micro-compensation level, the basic compensation amount is calculated based on the difference between the preset target recovery average and the thrust average, and the basic compensation amount is calculated based on the proportional coefficient corresponding to the micro-compensation level. The trend compensation amount is calculated based on the change slope and the differential coefficient corresponding to the micro-compensation level. The basic compensation amount and the trend compensation amount are summed to generate a micro-compensation command.

[0037] When the reflow furnace control level is the enhanced compensation level, the basic compensation amount is calculated based on the difference between the preset target recovery average and the thrust average, and the basic compensation amount is calculated based on the proportional coefficient corresponding to the enhanced compensation level. The trend compensation amount is calculated based on the change slope and the differential coefficient corresponding to the enhanced compensation level. The enhanced compensation command is generated by summing the basic compensation amount and the trend compensation amount.

[0038] S512: The obtained parameter holding instruction, micro-compensation instruction, enhanced compensation instruction, or alarm shutdown instruction shall be used as the reflow oven control instruction.

[0039] As a further aspect of the present invention, the process of obtaining the micro-compensation command is specifically as follows:

[0040] The sum of the basic compensation amount and the trend compensation amount corresponding to the micro-compensation level is used as the micro-process parameter correction amount. According to the preset parameter mapping table, the micro-process parameter correction amount is mapped to the first reflow furnace temperature zone setting temperature adjustment amount or the first conveyor chain speed adjustment amount in the reflow furnace process parameters. The micro-compensation command is generated according to the first reflow furnace temperature zone setting temperature adjustment amount or the first conveyor chain speed adjustment amount.

[0041] The process of obtaining the enhanced compensation command is as follows:

[0042] The sum of the basic compensation amount and the trend compensation amount corresponding to the strengthening compensation level is used as the strengthening process parameter correction amount. According to the preset parameter mapping table, the strengthening process parameter correction amount is mapped to the second reflow furnace temperature zone setting temperature adjustment amount or the second conveyor chain speed adjustment amount in the reflow furnace process parameters. The strengthening compensation command is generated according to the second reflow furnace temperature zone setting temperature adjustment amount or the second conveyor chain speed adjustment amount.

[0043] As a further aspect of the present invention, the process of S6 is specifically as follows:

[0044] S611: When the reflow oven control command is the micro-compensation command or the enhanced compensation command, the reflow oven control command is invoked to adjust the reflow oven process parameters. A transition period is set according to the thermal inertia of the reflow oven. During the transition period, drift identification and parameter compensation are suspended. After the transition period, the average thrust of the verification batch is collected by the thrust testing machine to form the compensation verification thrust data. The verification slope and verification margin are calculated based on the compensation verification thrust data.

[0045] S612: When the verification slope is not lower than the lower bound of the background noise threshold range and the verification margin is greater than the preset margin threshold, maintain the reflow oven process parameters; when the verification slope is lower than the lower bound of the background noise threshold range or the verification margin is not greater than the preset margin threshold, update the reflow oven control level and the reflow oven control command; summarize the reflow oven control level, the reflow oven control command and the compensation verification thrust data, and output the LED production line control record.

[0046] As a further aspect of the present invention, the process of setting the transition period based on the thermal inertia of the reflow oven specifically includes:

[0047] Obtain the preset reflow oven thermal stabilization time, the preset batch cycle time, and the conveyor chain speed in the reflow oven process parameters. Determine the number of transition batches based on the reflow oven thermal stabilization time and the batch cycle time. Multiply the product of the number of transition batches and the batch cycle time by the conveyor chain speed to obtain the transition production interval. Set the transition production interval as the transition period.

[0048] An automated control system for an LED chip production line, the automated control system for executing the above-described automated control method for an LED chip production line, the system comprising:

[0049] The thrust status integration module applies lateral shear force to the lamp bead solder joints through a thrust testing machine, collects raw thrust data, removes outliers to obtain valid test data, calculates the arithmetic mean to obtain the thrust mean, and associates it with batch number, test time and reflow oven process parameters to form batch thrust status data;

[0050] The trend feature extraction module configures a sliding window based on the batch thrust state data, performs linear regression and continuous decrease statistics on the thrust mean within the sliding window, and calculates the cumulative offset and remaining margin based on the preset benchmark steady-state mean, the preset static thrust qualified lower limit and the thrust mean, to obtain the change slope, cumulative offset, remaining margin and number of consecutive decrease batches, forming thrust trend feature data;

[0051] The drift pattern identification module, based on the thrust trend feature data, compares the change slope and the number of consecutive decreasing batches with a preset background noise threshold range and a preset continuity judgment threshold, respectively, to generate random fluctuation results or real drift results, and obtains drift identification results.

[0052] The compensation level decision module determines the reflow furnace control level by comparing the average thrust with the preset static thrust qualification lower limit and the remaining margin with the preset margin threshold, based on the drift identification results and the thrust trend characteristic data.

[0053] The reflux furnace control level is one of the following: alarm shutdown level, parameter hold level, micro-compensation level, or enhanced compensation level.

[0054] The instruction conversion and output module generates corresponding reflow furnace control instructions based on the reflow furnace control level and the thrust trend characteristic data.

[0055] The reflux furnace control command is one of the following: parameter hold command, micro-compensation command, enhanced compensation command, or alarm shutdown command.

[0056] The compensation verification closed-loop module adjusts the reflow oven process parameters and sets a transition period when the reflow oven control command is a micro-compensation command or an enhanced compensation command. After the transition period, it collects the average thrust of the verification batch to form compensation verification thrust data. Based on the compensation verification thrust data, it calculates the verification slope and verification margin, and maintains or updates the reflow oven control level and the reflow oven control command based on the verification slope and verification margin, and outputs the LED production line control record.

[0057] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0058] In this invention, by extracting the average thrust detection values ​​of multiple batches and configuring dynamic monitoring intervals, a trend feature system is reconstructed by obtaining linear features and continuous decreasing frequency indicators. Based on the continuity judgment benchmark and background noise interval, the direction of equipment evolution and drift is screened and identified, thereby blocking false fluctuation interference and compensating for blind spots in single-point comparison. Based on the state deviation amplitude and remaining margin space, the adjustment level is pre-determined and multi-gradient fine-tuning compensation instructions are issued to break the limitations of rigid threshold passive response. After intervention, a transition period is constructed to collect and verify indicators and compare them with the expected control target, thereby continuously updating and correcting instructions to form a closed-loop control path that combines feedforward and feedback. The delayed alarm is transformed into early deviation intervention to reduce the risk of process marginalization and prevent batch welding defects from flowing into downstream processes. Attached Figure Description

[0059] Figure 1 This is a schematic diagram of the automated control method for the LED bead production line of the present invention;

[0060] Figure 2 This is a schematic diagram of the automated control process of the LED bead production line according to the present invention;

[0061] Figure 3 This is a schematic diagram illustrating the thrust drift mode identification principle of the present invention;

[0062] Figure 4 This is a schematic diagram of the compensation verification closed-loop principle of the present invention;

[0063] Figure 5 This is a schematic diagram of the automated control system for the LED bead production line of the present invention. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this invention clearer, the 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 merely illustrative and not intended to limit the invention.

[0065] This embodiment discloses an automated control method for an LED chip production line. After reflow soldering, the solder joint strength of the LED chips gradually changes with the slow shift in the reflow oven's thermal field. If only a fixed pass / fail threshold is used for sampling inspection, it is difficult to detect the trend of deterioration before the solder joint strength falls below the pass line, and it is also difficult to distinguish between occasional fluctuations and continuous degradation of solder joint strength. This embodiment uses batch-by-batch testing of solder joint strength by a thrust testing machine as the data source, sequentially going through six processes: state integration, trend extraction, drift identification, level decision, instruction generation, and compensation verification, forming a closed-loop adjustment of the reflow oven process parameters. Each process is sequentially connected by data; the output of the previous process becomes the input of the next, forming a complete data flow closed loop. The following is a step-by-step explanation.

[0066] Please see Figure 1 and Figure 2 S1: Lateral shear force is applied to the solder joints of the LED beads using a thrust testing machine. Raw thrust data is collected, outliers are removed to obtain valid test data, and the arithmetic mean is calculated to obtain the average thrust. This average thrust is then correlated with the batch number, testing time, and reflow oven process parameters to form batch thrust state data. The technical purpose of this process is to organize the dispersed single-point solder joint strength test results into structured state data, organized by batch, that can be directly used for subsequent trend analysis. This is based on the fact that solder joints within the same batch undergo the same reflow oven thermal history, and their strength can be characterized by the central tendency within the batch. However, single-point testing is affected by individual differences in solder joints and exhibits dispersion; therefore, centralization processing is necessary to reflect the overall level of the batch. This process is further subdivided into two sub-processes, S111 and S112.

[0067] S111: A thrust testing machine applies lateral shear force to the LED chip solder joints, detects the peak thrust force at which the solder joint breaks under lateral shear force, and collects the peak thrust force value to obtain the raw thrust data. Lateral shear force is a thrust force acting on the root of the solder joint along a direction parallel to the substrate. It increases with continuous loading by the pusher of the thrust testing machine. The solder joint breaks at the moment its load-bearing capacity is exceeded. The maximum thrust force recorded at the moment of breakage is the peak thrust force, which directly reflects the actual bonding strength of the solder joint. The peak thrust force values ​​obtained from testing each solder joint are collected in the testing order to form the raw thrust data. The raw thrust data is an unfiltered collection of point-by-point test values, which includes valid readings reflecting the true strength, but may also contain abnormal readings caused by pusher slippage, solder joint position misalignment, or abnormal testing alignment. Therefore, further processing is required in the next sub-process.

[0068] S112: Outlier removal is performed on the raw thrust data to obtain valid test data. The arithmetic mean of the valid test data is calculated to obtain the mean thrust. This mean thrust is then correlated with the batch number, inspection time, and reflow oven process parameters to establish batch thrust status data. The outlier removal logic is as follows: within the raw thrust data of the same batch, the overall central tendency and dispersion range of the batch readings are first determined. Then, individual readings that significantly deviate from the main distribution and exceed the dispersion range are identified as outliers and removed. The remaining readings are the valid test data. This process ensures that individual detection anomalies no longer interfere with the estimation of the overall batch level. The arithmetic mean of the valid test data is calculated to obtain the mean thrust, which is the concentrated representative value of the weld joint bonding strength of the batch. Subsequently, the mean thrust is correlated with the batch number, the inspection time of this test, and the current reflow oven process parameters to establish batch thrust status data. The batch number identifies the batch to which the data belongs and maintains the sequential relationship between batches. The detection time determines the time sequence of the batches. The reflow oven process parameters record the thermal field conditions that generated the solder joint strength of that batch. When the original thrust data has missing formats, or the number of valid test data for a batch is insufficient to stably estimate the concentration level, this process marks that batch as incomplete and temporarily excludes it from subsequent trend statistics. The data will be integrated after the batch data is complete to avoid unreliable batch levels interfering with trend judgment. The reflow oven process parameters in the batch thrust status data are the adjustment targets for parameter compensation corresponding to the reflow oven control level; that is, the process parameters recorded here are the ones that the subsequent compensation process will correct. After the batch thrust status data is generated, it is received by the trend extraction process.

[0069] Please see Figure 1 S2: Based on batch thrust state data, a sliding window is configured. Linear regression and continuous decline statistics are performed on the thrust mean within the sliding window. The cumulative offset and remaining margin are calculated based on the preset benchmark steady-state mean, the preset static thrust qualification lower limit, and the thrust mean, yielding the change slope, cumulative offset, remaining margin, and the number of batches with continuous decline, forming thrust trend characteristic data. The technical purpose of this process is to extract characteristic quantities that can characterize the direction, degree of deviation, and persistence of intensity evolution from the batch-by-batch thrust mean sequence, providing a basis for subsequently distinguishing between occasional fluctuations and continuous drift. This is based on the fact that the thrust mean of a single batch can only reflect the instantaneous level, while intensity degradation is a gradual process spanning multiple batches. Its trend must be examined within an observation interval consisting of consecutive batches to separate the gradual trend from the normal fluctuations between batches. This process is further subdivided into three sub-processes: S211, S212, and S213.

[0070] S211: Retrieve batch thrust status data, determine the order of the average thrust values ​​for each batch according to batch number and detection time, configure a sliding window, and perform linear regression calculation on the average thrust values ​​within the sliding window to obtain the slope of change. The sliding window is a fixed-length observation interval that moves forward along the batch time sequence. It always covers the average thrust values ​​of the most recent consecutive batches. Whenever new batch data enters, the earliest batch data is moved out, thus continuously updating the observation interval as production progresses. The length of the sliding window is preset according to the required batch span, ensuring it covers a sufficient number of batches to reflect the trend without overwhelming recent changes due to an excessively long interval. The order is determined by batch number and detection time to ensure that the average thrust values ​​within the window are arranged according to the actual production sequence. The linear regression calculation fits a straight line within the window, with batch order as the independent variable and average thrust as the dependent variable, that best matches the trend of these data points. The slope of this line is the slope of change. A negative slope indicates that the average thrust within the window is generally decreasing, with the magnitude reflecting the rate of decrease; a slope close to zero indicates that the intensity level is basically stable within the window. The slope thus condenses the overall trend of a batch interval into a single directional characteristic.

[0071] S212: Based on the preset benchmark steady-state mean, calculate the difference between the average thrust and the benchmark steady-state mean to obtain the cumulative offset. Based on the preset static thrust qualification lower limit, calculate the difference between the average thrust and the static thrust qualification lower limit to obtain the remaining margin. The benchmark steady-state mean is the reference level that the average thrust of the weld joint should maintain when the process is in a stable state. It is taken from the statistical calibration of the average thrust of the weld joint during the process stabilization period. That is, during the production period when the thermal field is confirmed to be stable and the strength is qualified, the average thrust of each batch during this period is summarized and its concentration level is used as the benchmark steady-state mean to measure whether the current strength deviates from the normal state. The cumulative offset is taken from the difference between the current average thrust and the benchmark steady-state mean. It represents the extent to which the current strength level has deviated from the normal state. The larger the deviation, the further the strength level is from the steady state. The static thrust qualification lower limit is the minimum level that the weld joint strength must reach to be judged as qualified. It is set according to the quality requirements of the weld joint strength and is the bottom line that cannot be exceeded in terms of quality. The remaining margin is taken as the difference between the current average thrust and the lower limit of the static thrust qualification. It represents how much margin is left between the current strength level and the qualification threshold. The smaller the remaining margin, the closer the strength is to the failure boundary, and the more limited the room for compensation and adjustment. The cumulative offset and the remaining margin together characterize the current strength state from the perspectives of both the degree of deviation and the safety margin.

[0072] S213: The number of consecutive decreases in the thrust mean of adjacent batches within a sliding window is counted to obtain the number of consecutively decreasing batches. This number is then correlated with the slope of change, cumulative offset, remaining margin, and the number of consecutively decreasing batches to establish thrust trend characteristic data. The processing logic for counting the number of consecutive decreases involves comparing the thrust mean of adjacent batches one by one along the batch order within the window. When the later batch is lower than the earlier batch, it is counted as one decrease. These consecutive decreases are accumulated to determine the number of batches currently experiencing a continuous decrease, i.e., the number of consecutively decreasing batches. This feature complements the slope of change from a continuity perspective. The slope of change characterizes the strength of the overall trend, while the number of consecutively decreasing batches characterizes whether the decrease is continuous and uninterrupted. Combining these two features avoids mistaking sporadic fluctuations in opposite directions for a trend. The slope of change, cumulative offset, remaining margin, and number of consecutively decreasing batches are linked to establish thrust trend characteristic data. This thrust trend characteristic data provides a complete characteristic description of the intensity evolution within the current batch interval and is received by both the drift identification process and the grade decision-making process.

[0073] Please see Figure 1 , Figure 2 and Figure 3 S3: Based on thrust trend characteristic data, by comparing the change slope and the number of consecutive decreasing batches with preset background noise threshold ranges and preset continuity judgment thresholds respectively, random fluctuation results or true drift results are generated, resulting in drift identification results. The technical purpose of this process is to solve the problem of difficulty in distinguishing between occasional fluctuations and continuous drift, separating normal fluctuations in intensity levels from true degradation with trends and continuity, avoiding misjudging normal fluctuations as degradation and triggering unnecessary adjustments, and avoiding missing the adjustment opportunity due to missed true degradation. The basis for this is that in normal production, the strength of weld joints is affected by various random factors and fluctuates slightly. These fluctuations are not continuous in direction and fall within the background noise range in amplitude; while true drift is continuously downward in direction and exceeds the background noise range in amplitude. Therefore, it can be distinguished by setting judgment conditions for both the direction amplitude and continuity dimensions simultaneously. This process is further subdivided into two sub-processes, S311 and S312.

[0074] S311: Call the thrust trend characteristic data, compare the slope of change with the preset background noise threshold range, and compare the number of consecutive decreasing batches with the preset continuity judgment threshold to obtain drift comparison data. The background noise threshold range is a preset range used to define the slope range corresponding to normal random fluctuations. It is taken from the calibration of the normal fluctuation range of the slope of change during stable production. That is, the fluctuation range of the slope of change is statistically analyzed in a stable period without trend degradation. Its boundary is used as the basis for distinguishing noise from trend. Its lower boundary marks the boundary of the decreasing slope that can still be regarded as noise. If the slope of change falls within this range, it means that the trend amplitude has not exceeded the normal fluctuation. If it falls below its lower boundary, it means that the trend amplitude has significantly decreased. The continuity judgment threshold is a preset batch number threshold used to define whether the decrease constitutes a continuous trend. It is set according to the allowable number of consecutive decreasing batches. If the number of consecutive decreasing batches reaches this threshold, it means that the decrease has been continuous. This sub-process completes the comparison of the slope of change with the lower boundary of the background noise threshold range and the comparison of the number of consecutive decreasing batches with the continuity judgment threshold, and combines the results of the two comparisons into drift comparison data for comprehensive judgment in the next sub-process.

[0075] S312: Based on drift comparison data, when the slope of change is not lower than the lower bound of the background noise threshold interval and the number of consecutive decreasing batches is less than the continuity judgment threshold, a random fluctuation result is generated. When the slope of change is lower than the lower bound of the background noise threshold interval or the number of consecutive decreasing batches is not less than the continuity judgment threshold, a true drift result is generated, thus obtaining the drift identification result. The judgment logic of this sub-process considers both the amplitude of the trend and the duration of the decline. When the slope of change is still above the lower bound of the background noise threshold interval and the number of consecutive decreasing batches has not yet reached the continuity judgment threshold, it indicates that there is neither a significant downward trend nor a continuous decline. This situation is judged as a random fluctuation result, indicating that the current change is a normal fluctuation. When the slope of change falls below the lower bound of the background noise threshold interval, or the number of consecutive decreasing batches reaches the continuity judgment threshold, either of these conditions is met, indicating that the intensity level has either significantly decreased in amplitude or has become continuous. This situation is judged as a true drift result, indicating that the current change is a trend of degradation. The final output random fluctuation result or true drift result is the drift identification result. When thrust trend characteristic data is missing, or when a batch is marked as incomplete and a complete comparison cannot be formed, this process does not force a drift identification result. Instead, it uses the previous valid judgment and waits for the data to be supplemented, avoiding misjudgments based on incomplete features. The drift identification result is then received by the grading decision process.

[0076] Please see Figure 1 and Figure 2S4: Based on the drift identification results and thrust trend characteristic data, the reflow oven control level is determined by comparing the average thrust with the preset static thrust acceptable lower limit and comparing the remaining margin with the preset margin threshold. The technical purpose of this process is to transform the intensity state judgment conclusion into a graded handling decision, so that different severity states correspond to different response intensities. This is based on whether the weld strength has fallen below the acceptable threshold, whether the degradation is a genuine trend, and whether the margin from the acceptable threshold is sufficient. These three factors jointly determine whether to shut down, maintain, make minor adjustments, or make major adjustments. Therefore, it is necessary to judge step by step according to these three conditions. The reflow oven control level is one of the following: alarm shutdown level, parameter maintenance level, micro-compensation level, or enhanced compensation level. This process is further subdivided into two sub-processes, S411 and S412.

[0077] S411: Using drift identification results and thrust trend characteristic data, determine the level according to the following conditions: When the average thrust is not greater than the preset static thrust lower limit, it is determined to be an alarm shutdown level. At this time, the weld strength has fallen below the acceptable threshold, a quality risk has occurred, and shutdown is required instead of further adjustment. When the average thrust is greater than the preset static thrust lower limit and the drift identification result is a random fluctuation result, it is determined to be a parameter maintenance level. At this time, the strength is still above the acceptable threshold and the changes are normal fluctuations, requiring no adjustment of process parameters. When the average thrust is greater than the preset static thrust lower limit, the drift identification result is a true drift result, and the remaining margin is greater than the preset margin threshold, it is determined to be a micro-compensation level. At this time, although the strength has actually degraded, there is still sufficient margin from the acceptable threshold. The preset margin threshold is the threshold for judging whether the margin is sufficient. It is set according to the safety margin required for compensation adjustment. Sufficient margin indicates that the degradation is in its early stages and can be corrected with slight adjustment. When the average thrust exceeds the preset lower limit of acceptable static thrust, the drift identification result is a true drift result, and the remaining margin is not greater than the preset margin threshold, the enhancement compensation level is determined. At this point, the strength has truly degraded and the margin from the acceptable threshold is insufficient, requiring stronger adjustment to restore it as quickly as possible. The above four conditions are mutually exclusive and cover various combinations of strength states, ensuring that any state can be classified into a unique level.

[0078] S412: The determined alarm shutdown level, parameter hold level, micro-compensation level, or enhanced compensation level is used as the reflow oven control level. This sub-process formally establishes the level determined in S411 as the reflow oven control level, which serves as the basis for subsequent instruction generation and is received by the instruction generation process.

[0079] Please see Figure 1S5: Based on the reflow oven control level and thrust trend characteristic data, generate corresponding reflow oven control commands according to the reflow oven control level. The technical purpose of this process is to translate the graded decision into action commands that can be directly executed by the reflow oven, thus connecting decision-making and execution. This is based on the fact that different levels correspond to different actions: the hold level only needs to maintain the status quo, the shutdown level needs to interrupt the board feed, and the compensation level needs to calculate specific adjustment amounts based on the magnitude and trend of strength degradation. The reflow oven control command is one of the following: parameter hold command, micro-compensation command, enhanced compensation command, or alarm shutdown command. This process is further subdivided into two sub-processes, S511 and S512.

[0080] S511: Retrieves reflow oven control level and thrust trend characteristic data, and generates corresponding instructions based on the reflow oven control level. When the reflow oven control level is parameter hold level, a parameter hold instruction is generated, and the reflow oven maintains the current process parameters unchanged. When the reflow oven control level is alarm shutdown level, an alarm shutdown instruction is generated, and a stop-board-feed control is executed according to the alarm shutdown instruction, preventing subsequent boards from entering the reflow oven, thereby stopping the continued production of defective batches. When the reflow oven control level is at the micro-compensation level, the required strength recovery range is first determined based on the difference between the preset target recovery average and the thrust average. The target recovery average is the target level to which the expected weld point thrust average is expected to recover after compensation, and is set as the expected strength recovery level. The larger the difference, the more strength needs to be compensated. Then, the difference is converted into a basic compensation amount according to the proportional coefficient corresponding to the micro-compensation level. The proportional coefficient determines the proportion of compensation given according to the current deviation. The basic compensation amount is the main compensation for the current deviation. At the same time, the trend compensation amount is calculated according to the slope of change and the differential coefficient corresponding to the micro-compensation level. The differential coefficient determines how much additional compensation is added according to the current downward trend. The faster the decline reflected by the slope of change, the larger the trend compensation amount. The trend compensation amount is the pre-compensation for the trend. The basic compensation amount and the trend compensation amount are summed to obtain the total compensation amount for this level, and a micro-compensation instruction is generated accordingly. When the reflow oven control level is at the enhanced compensation level, the same processing logic applies. First, the difference between the preset target recovery mean and the thrust mean is calculated. Then, the basic compensation amount is calculated based on the proportional coefficient corresponding to the enhanced compensation level. Next, the trend compensation amount is calculated based on the slope of change and the differential coefficient corresponding to the enhanced compensation level. The basic compensation amount and the trend compensation amount are summed to generate an enhanced compensation command. The proportional coefficient and differential coefficient corresponding to each level are determined through process calibration. During calibration, the process parameter adjustment amount corresponding to the unit deviation and unit direction are examined, and the coefficient of the enhanced compensation level is made greater than that of the micro-compensation level. Thus, under the same deviation and direction, the total compensation amount given by enhanced compensation is greater than that of micro-compensation, corresponding to a stronger adjustment force.

[0081] When generating compensation instructions, the compensation amount needs to be further converted into process parameter adjustments that can be executed by the reflow oven. The process of obtaining the micro-compensation instruction is as follows: the sum of the basic compensation amount and the trend compensation amount corresponding to the micro-compensation level is used as the micro-process parameter correction amount. According to the preset parameter mapping table, the micro-process parameter correction amount is mapped to the first reflow oven temperature zone setting temperature adjustment amount or the first conveyor chain speed adjustment amount in the reflow oven process parameters. The micro-compensation instruction is then generated based on the first reflow oven temperature zone setting temperature adjustment amount or the first conveyor chain speed adjustment amount. The parameter mapping table is a pre-established correspondence that maps process parameter corrections to specific adjustable parameters in the reflow oven. It is established through prior process experiments, recording the corresponding changes in the reflow oven temperature zone setting temperature and conveyor speed under different process parameter corrections, thus forming a correspondence between the correction and adjustable parameter changes. Using this table, compensation requirements expressed in terms of thrust can be converted into adjustments to the temperature zone setting temperature or conveyor speed. The temperature zone setting temperature affects the input of welding heat, and the conveyor speed affects the duration of heat exposure for the weld joint in the oven; both directly affect the bond strength of the weld joint. The process of obtaining the enhanced compensation command is as follows: the sum of the basic compensation and trend compensation corresponding to the enhanced compensation level is used as the enhanced process parameter correction. According to the pre-established parameter mapping table, the enhanced process parameter correction is mapped to the second reflow oven temperature zone setting temperature adjustment or the second conveyor speed adjustment in the reflow oven process parameters. The enhanced compensation command is then generated based on the second reflow oven temperature zone setting temperature adjustment or the second conveyor speed adjustment. The two acquisition processes use the same mapping mechanism but correspond to their respective correction and adjustment amounts, so that the difference in compensation strength is ultimately reflected in the difference in the adjustment range of the reflux furnace parameters.

[0082] S512: The acquired parameter hold command, micro-compensation command, enhanced compensation command, or alarm shutdown command is used as the reflow oven control command. This sub-process formally establishes the command generated in S511 as the reflow oven control command, which is executed by the reflow oven and received by the compensation verification process. When the parameter mapping table lacks a corresponding relationship matching a certain correction amount, this process does not output an unfounded adjustment amount, but instead reverts to parameter hold and issues a prompt for manual confirmation, avoiding the application of unreliable process adjustments.

[0083] Please see Figure 1 , Figure 2 and Figure 4S6: When the reflow oven control command is a micro-compensation command or a strong compensation command, adjust the reflow oven process parameters and set a transition period. After the transition period, collect the average thrust of the verification batch to form compensation verification thrust data. Calculate the verification slope and verification margin based on the compensation verification thrust data, and maintain or update the reflow oven control level and reflow oven control command according to the verification slope and verification margin, outputting the LED production line control record. The technical purpose of this process is to confirm whether the compensation has achieved the expected effect and close the control loop accordingly, avoiding the assumption that the compensation is effective after it is applied and allowing deviations to occur. The basis for this is that the reflow oven has thermal inertia; the thermal field inside the furnace will not stabilize immediately after the process parameters are adjusted. The intensity of the batch immediately following the adjustment has not yet fully reflected the adjustment effect. It is necessary to wait for the thermal field to stabilize before testing to obtain a reliable verification conclusion. This process is further subdivided into two sub-processes, S611 and S612.

[0084] S611: When the reflow oven control command is a micro-compensation command or a strong compensation command, the reflow oven control command is invoked to adjust the reflow oven process parameters. A transition period is set based on the reflow oven's thermal inertia. During the transition period, drift identification and parameter compensation are suspended. After the transition period, the average thrust of the verification batch is collected using a thrust testing machine to form compensation verification thrust data. The verification slope and verification margin are calculated based on this data. Reflow oven thermal inertia is the hysteresis characteristic exhibited by the reflow oven's thermal field as it gradually approaches a new stable state after process parameter changes. The transition period is the waiting interval during which the thermal field is allowed to stabilize again. Batch strength collected during this period is still in an unstable adjustment stage and cannot be used as a basis for evaluation. Therefore, drift identification and parameter compensation are suspended during the transition period to prevent new misjudgments or superimposed compensation based on unstable data. After the transition period, the average thrust of the verification batch is collected again using a thrust testing machine. The verification batch is the batch used to verify the compensation effect after the transition period ends, and its average thrust is aggregated to form compensation verification thrust data. On the compensation verification thrust data, the verification slope and verification margin are calculated according to the same processing logic as the change slope and remaining margin in S2. The verification slope reflects the trend of the strength level after compensation, and the verification margin reflects the remaining strength level after compensation from the qualified bottom line. The two are used to verify whether the compensation has made the strength trend stabilize and the margin restored.

[0085] The process of setting the transition period based on the reflow oven's thermal inertia is as follows: Obtain the preset reflow oven thermal stabilization time, preset batch cycle time, and conveyor chain speed from the reflow oven process parameters. Determine the number of transition batches based on the reflow oven thermal stabilization time and batch cycle time. Multiply the product of the number of transition batches and the batch cycle time by the conveyor chain speed to obtain the transition production interval, and set this transition production interval as the transition period. The reflow oven thermal stabilization time is the time required for the furnace's internal thermal field to stabilize again after process parameter adjustments, taken from the calibration of the reflow oven's thermal field response. The batch cycle time is the time interval between two adjacent batches entering the production line, taken from the batch feeding rhythm of the production line. The conveyor chain speed is the speed at which the sheet material is conveyed within the reflow oven. The processing logic is as follows: First, compare the reflow oven thermal stabilization time with the batch cycle time to determine how many batches will pass during the thermal field stabilization period, obtaining the number of transition batches. Then, multiply the number of transition batches by the batch cycle time to reconstruct the corresponding duration, and multiply this duration by the conveyor chain speed to convert it into the production length the sheet material has traveled during this period, obtaining the transition production interval. Finally, establish the transition production interval as the transition period. The transition period set in this way matches the time required for the thermal field to stabilize and the actual production line operating speed, so that the sampling of the verification batch occurs after the thermal field has stabilized.

[0086] S612: Maintain the reflow oven process parameters when the verification slope is not lower than the lower bound of the background noise threshold range and the verification margin is greater than the preset margin threshold. Update the reflow oven control level and reflow oven control commands when the verification slope is lower than the lower bound of the background noise threshold range or the verification margin is not greater than the preset margin threshold. Summarize the reflow oven control level, reflow oven control commands, and compensation verification thrust data, and output the LED production line control record. The thresholds used in this sub-process are consistent with those used in S312 and S411. When the verification slope is still above the lower bound of the background noise threshold range and the verification margin is greater than the preset margin threshold, it indicates that the intensity trend after compensation has stabilized, the margin has been restored sufficiently, the compensation is effective, and the current reflow oven process parameters are maintained without further change. When the verification slope falls below the lower bound of the background noise threshold range, or the verification margin is not greater than the preset margin threshold, either of these conditions indicates that the compensated intensity is still declining or the margin is still insufficient, and the compensation is inadequate to correct the degradation. In this case, the process returns to the level decision to redetermine the reflow oven control level and update the reflow oven control command accordingly, applying adjustment again to form a closed loop of continuous correction. Regardless of whether it is maintained or updated, the reflow oven control level, reflow oven control command, and compensation verification thrust data are summarized and output as the LED production line control record. The LED production line control record is a complete record of the level adopted, the command issued, and the verification results of this round of control, for traceability and subsequent analysis. When the verification batch cannot be collected due to shutdown or testing interruption, or the compensation verification thrust data is incomplete, this process does not determine the success or failure of compensation based on incomplete data. Instead, it extends the transition period to wait for the verification batch to be completed, and then makes a judgment after the data is complete, ensuring the reliability of the closed-loop judgment.

[0087] Please see Figure 5 In this embodiment, the method is executed by the automated control system of the LED chip production line. The thrust state integration module performs detection and batch state integration in S1, the trend feature extraction module performs sliding window configuration and trend feature extraction in S2, the drift pattern identification module performs drift identification in S3, the compensation level decision module performs reflow oven control level determination in S4, the instruction conversion and output module performs reflow oven control instruction generation in S5, and the compensation verification closed-loop module performs compensation adjustment, transition period setting, verification, and control record output in S6. Each module receives data in sequence according to the above steps and together completes the automated closed-loop control of the solder joint strength of the LED chip production line.

[0088] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An automated control method for an LED chip production line, characterized in that, Includes the following steps: S1: Apply lateral shear force to the lamp bead solder joints using a thrust testing machine, collect raw thrust data, remove outliers to obtain valid test data, calculate the arithmetic mean to obtain the thrust mean, and associate it with batch number, test time and reflow oven process parameters to form batch thrust status data; S2: Configure a sliding window based on the batch thrust state data, perform linear regression and continuous decrease statistics on the thrust mean within the sliding window, and calculate the cumulative offset and remaining margin according to the preset benchmark steady-state mean, the preset static thrust qualified lower limit and the thrust mean, to obtain the change slope, cumulative offset, remaining margin and the number of continuously decreasing batches, forming thrust trend characteristic data; S3: Based on the thrust trend feature data, by comparing the change slope and the number of consecutive decreasing batches with the preset background noise threshold range and the preset continuity judgment threshold respectively, random fluctuation results or real drift results are generated to obtain drift identification results; S4: Based on the drift identification results and the thrust trend characteristic data, the reflux furnace control level is determined by comparing the average thrust with the preset static thrust qualification lower limit and comparing the remaining margin with the preset margin threshold. The reflux furnace control level is one of the following: alarm shutdown level, parameter hold level, micro-compensation level, or enhanced compensation level. S5: Based on the reflow furnace control level and the thrust trend characteristic data, generate corresponding reflow furnace control commands according to the reflow furnace control level; The reflux furnace control command is one of the following: parameter hold command, micro-compensation command, enhanced compensation command, or alarm shutdown command. S6: When the reflow oven control command is a micro-compensation command or a strong compensation command, adjust the reflow oven process parameters and set a transition period. After the transition period, collect the average thrust of the verification batch to form compensation verification thrust data. Calculate the verification slope and verification margin based on the compensation verification thrust data. Maintain or update the reflow oven control level and the reflow oven control command based on the verification slope and verification margin, and output the LED production line control record.

2. The automated control method for an LED bead production line according to claim 1, characterized in that, The process of S1 is as follows: S111: Apply lateral shear force to the lamp bead solder joint using a thrust testing machine, detect the peak thrust when the lamp bead solder joint breaks under the action of lateral shear force, collect the peak thrust value at break, and obtain the raw thrust data; S112: Outlier values ​​are removed from the original thrust data to obtain valid test data. The arithmetic mean of the valid test data is calculated to obtain the thrust mean. The thrust mean is correlated with the batch number, test time and reflow furnace process parameters to establish batch thrust status data. The reflow furnace process parameters in the batch thrust status data are the adjustment targets for parameter compensation corresponding to the reflow furnace control level.

3. The automated control method for an LED bead production line according to claim 1, characterized in that, The process of S2 is as follows: S211: Call the batch thrust status data, determine the order of the average thrust values ​​of each batch according to the batch number and detection time, configure a sliding window, and perform linear regression calculation on the average thrust values ​​within the sliding window to obtain the slope of change; S212: Based on the preset benchmark steady-state average value, calculate the difference between the average thrust value and the benchmark steady-state average value to obtain the cumulative offset; based on the preset static thrust qualification lower limit value, calculate the difference between the average thrust value and the static thrust qualification lower limit value to obtain the remaining margin. S213: Count the number of consecutive decreases in the average thrust of adjacent batches within the sliding window to obtain the number of consecutive decrease batches. Then, associate the change slope, the cumulative offset, the remaining margin, and the number of consecutive decrease batches to establish thrust trend feature data.

4. The automated control method for an LED bead production line according to claim 1, characterized in that, The process of S3 is as follows: S311: Call the thrust trend feature data, compare the change slope with the preset background noise threshold range, and compare the number of consecutive decreasing batches with the preset continuity determination threshold to obtain drift comparison data; S312: Based on the drift comparison data, when the change slope is not lower than the lower bound of the background noise threshold interval and the number of consecutive decreasing batches is less than the continuity determination threshold, a random fluctuation result is generated; when the change slope is lower than the lower bound of the background noise threshold interval or the number of consecutive decreasing batches is not less than the continuity determination threshold, a true drift result is generated, and a drift identification result is obtained.

5. The automated control method for an LED bead production line according to claim 1, characterized in that, The process of S4 is as follows: S411: Invoke the drift identification results and the thrust trend feature data; When the average thrust is not greater than the preset lower limit of static thrust, it is determined to be an alarm shutdown level; When the average thrust is greater than the preset static thrust qualification lower limit and the drift identification result is the random fluctuation result, it is determined to be a parameter maintenance level; When the average thrust is greater than the preset static thrust qualification lower limit, the drift identification result is the actual drift result, and the remaining margin is greater than the preset margin threshold, it is determined to be a micro-compensation level. When the average thrust is greater than the preset static thrust qualification lower limit, the drift identification result is the actual drift result, and the remaining margin is not greater than the preset margin threshold, it is determined to be an enhanced compensation level; S412: The determined alarm shutdown level, parameter hold level, micro-compensation level, or enhanced compensation level shall be used as the reflow oven control level.

6. The automated control method for an LED bead production line according to claim 1, characterized in that, The process of S5 is as follows: S511: Invoke the reflow oven control level and the thrust trend characteristic data, and generate corresponding instructions based on the reflow oven control level, including: When the reflow oven control level is the parameter holding level, a parameter holding command is generated; When the reflow oven control level is the alarm shutdown level, an alarm shutdown command is generated, and the board feeding stop control is executed according to the alarm shutdown command; When the reflow oven control level is the micro-compensation level, the basic compensation amount is calculated based on the difference between the preset target recovery average and the thrust average, and the basic compensation amount is calculated based on the proportional coefficient corresponding to the micro-compensation level. The trend compensation amount is calculated based on the change slope and the differential coefficient corresponding to the micro-compensation level. The basic compensation amount and the trend compensation amount are summed to generate a micro-compensation command. When the reflow furnace control level is the enhanced compensation level, the basic compensation amount is calculated based on the difference between the preset target recovery average and the thrust average, and the basic compensation amount is calculated based on the proportional coefficient corresponding to the enhanced compensation level. The trend compensation amount is calculated based on the change slope and the differential coefficient corresponding to the enhanced compensation level. The enhanced compensation command is generated by summing the basic compensation amount and the trend compensation amount. S512: The obtained parameter holding instruction, micro-compensation instruction, enhanced compensation instruction, or alarm shutdown instruction shall be used as the reflow oven control instruction.

7. The automated control method for an LED bead production line according to claim 6, characterized in that, The process of obtaining the micro-compensation command is as follows: The sum of the basic compensation amount and the trend compensation amount corresponding to the micro-compensation level is used as the micro-process parameter correction amount. According to the preset parameter mapping table, the micro-process parameter correction amount is mapped to the first reflow furnace temperature zone setting temperature adjustment amount or the first conveyor chain speed adjustment amount in the reflow furnace process parameters. The micro-compensation command is generated according to the first reflow furnace temperature zone setting temperature adjustment amount or the first conveyor chain speed adjustment amount. The process of obtaining the enhanced compensation command is as follows: The sum of the basic compensation amount and the trend compensation amount corresponding to the strengthening compensation level is used as the strengthening process parameter correction amount. According to the preset parameter mapping table, the strengthening process parameter correction amount is mapped to the second reflow furnace temperature zone setting temperature adjustment amount or the second conveyor chain speed adjustment amount in the reflow furnace process parameters. The strengthening compensation command is generated according to the second reflow furnace temperature zone setting temperature adjustment amount or the second conveyor chain speed adjustment amount.

8. The automated control method for an LED bead production line according to claim 1, characterized in that, The process of S6 is as follows: S611: When the reflow oven control command is the micro-compensation command or the enhanced compensation command, the reflow oven control command is invoked to adjust the reflow oven process parameters. A transition period is set according to the thermal inertia of the reflow oven. During the transition period, drift identification and parameter compensation are suspended. After the transition period, the average thrust of the verification batch is collected by the thrust testing machine to form the compensation verification thrust data. The verification slope and verification margin are calculated based on the compensation verification thrust data. S612: When the verification slope is not lower than the lower bound of the background noise threshold range and the verification margin is greater than the preset margin threshold, maintain the reflow oven process parameters; when the verification slope is lower than the lower bound of the background noise threshold range or the verification margin is not greater than the preset margin threshold, update the reflow oven control level and the reflow oven control command; summarize the reflow oven control level, the reflow oven control command and the compensation verification thrust data, and output the LED production line control record.

9. The automated control method for an LED bead production line according to claim 8, characterized in that, The process of setting the transition period based on the thermal inertia of the reflux furnace is as follows: Obtain the preset reflow oven thermal stabilization time, the preset batch cycle time, and the conveyor chain speed in the reflow oven process parameters. Determine the number of transition batches based on the reflow oven thermal stabilization time and the batch cycle time. Multiply the product of the number of transition batches and the batch cycle time by the conveyor chain speed to obtain the transition production interval. Set the transition production interval as the transition period.

10. An automated control system for an LED bead production line, characterized in that, The system is used to implement the automated control method for the LED bead production line according to any one of claims 1-9, and the system includes: The thrust status integration module applies lateral shear force to the lamp bead solder joints through a thrust testing machine, collects raw thrust data, removes outliers to obtain valid test data, calculates the arithmetic mean to obtain the thrust mean, and associates it with batch number, test time and reflow oven process parameters to form batch thrust status data; The trend feature extraction module configures a sliding window based on the batch thrust state data, performs linear regression and continuous decrease statistics on the thrust mean within the sliding window, and calculates the cumulative offset and remaining margin based on the preset benchmark steady-state mean, the preset static thrust qualified lower limit and the thrust mean, to obtain the change slope, cumulative offset, remaining margin and number of consecutive decrease batches, forming thrust trend feature data; The drift pattern identification module, based on the thrust trend feature data, compares the change slope and the number of consecutive decreasing batches with a preset background noise threshold range and a preset continuity judgment threshold, respectively, to generate random fluctuation results or real drift results, and obtains drift identification results. The compensation level decision module determines the reflow furnace control level by comparing the average thrust with the preset static thrust qualification lower limit and the remaining margin with the preset margin threshold, based on the drift identification results and the thrust trend characteristic data. The reflux furnace control level is one of the following: alarm shutdown level, parameter hold level, micro-compensation level, or enhanced compensation level. The instruction conversion and output module generates corresponding reflow furnace control instructions based on the reflow furnace control level and the thrust trend characteristic data. The reflux furnace control command is one of the following: parameter hold command, micro-compensation command, enhanced compensation command, or alarm shutdown command. The compensation verification closed-loop module adjusts the reflow oven process parameters and sets a transition period when the reflow oven control command is a micro-compensation command or an enhanced compensation command. After the transition period, it collects the average thrust of the verification batch to form compensation verification thrust data. Based on the compensation verification thrust data, it calculates the verification slope and verification margin, and maintains or updates the reflow oven control level and the reflow oven control command based on the verification slope and verification margin, and outputs the LED production line control record.

Images