LED packaging machine table error correction method based on data fusion

Abstract

The invention relates to the technical field of error correction, in particular to an LED packaging machine table error correction method based on data fusion, which comprises the following steps of: acquiring operation data of an LED packaging machine table in a production cycle, constructing an initial operation data set, slicing and dividing the initial operation data set, and meanwhile, dividing the initial operation data set into an initial operation data set; identifying error features by using a hierarchical iteration mechanism; aiming at the identified error operation feedback parameters, correcting the error operation feedback parameters layer by layer by utilizing a double-layer local correction mechanism, and correcting and verifying the corrected operation feedback parameters by utilizing a stability evaluation function; according to the method, a layered iterative error screening mechanism is adopted, so that the effect of step-by-step screening and sorting of error operation feedback parameters is achieved; by adopting an adjacent slice comparison technology based on spatial similarity, an effect of accurately judging error parameters with propagation trends is realized.

Description

Technical Field

[0001] This invention relates to the technical field of error correction, specifically to an error correction method for LED packaging machines based on data fusion. Background Technology

[0002] The LED industry is currently undergoing rapid iteration towards miniaturization and high integration. Wire bonding technology has entered a critical stage of transitioning from the "micron level" to the "nano level," with new materials such as copper alloy wires and silver wires, as well as new processes such as thermoforming and laser welding, being widely used. However, in the LED packaging production process, packaging machines typically operate under a multi-process, multi-parameter collaborative state, and the correction of parameter errors still faces the following challenges:

[0003] In existing technologies, most error correction methods rely solely on static threshold judgments based on a single time point or a single parameter, lacking analysis of the error's evolution characteristics over time. This makes it difficult to distinguish between systematic errors with a continuous propagation trend and isolated errors without such a trend. Furthermore, existing correction schemes typically lack a unified verification mechanism for the stability of the correction results after parameter adjustment, which can easily lead to "over-correction" or "mis-correction," thereby causing new operational instability issues. Summary of the Invention

[0004] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for correcting errors in LED packaging machines based on data fusion, comprising the following steps:

[0005] Collect and identify the error characteristics of LED packaging machines, specifically:

[0006] The system collects operating data of the LED packaging machine during the production cycle, constructs an initial operating data set, divides the initial operating data set into slices, and uses a hierarchical iteration mechanism to identify error characteristics.

[0007] Furthermore, error correction for LED packaging machines is performed based on the identified error characteristics, specifically as follows:

[0008] For the identified error operation feedback parameters, a two-layer local correction mechanism is used to correct the error operation feedback parameters layer by layer, and a stability evaluation function is used to verify the correction of the operation feedback parameters.

[0009] As a preferred embodiment of the LED packaging machine error correction method based on data fusion described in this invention, the step of identifying error features using a hierarchical iterative mechanism is as follows:

[0010] Collect the operating data of the LED packaging machine under the current batch of orders within the current production cycle, and construct the collected operating data into an initial operating data set. Then, divide the constructed initial operating data set into slices according to the operating time sequence.

[0011] For each data slice, calculate the runtime bias.

[0012] Based on the calculated operating deviation, the parent error slice for error identification of machine tool operating status is determined;

[0013] Based on the determined parent error slices, error operation feedback parameters are selected using error thresholds.

[0014] As a preferred embodiment of the LED packaging machine error correction method based on data fusion described in this invention, the parent error slice for error identification of the fixed machine operating state is specifically as follows:

[0015] Based on the calculated running deviation of each data slice, and arranging the running deviations of each data slice in ascending order, the running data slice with the largest running deviation is selected from all running deviations as the parent error slice.

[0016] As a preferred embodiment of the LED packaging machine error correction method based on data fusion described in this invention, the step of using an error threshold to filter error operation feedback parameters is as follows:

[0017] For a given parent error slice, by setting an error threshold, the error operation feedback parameters in the parent error slice are identified, then we have:

[0018] Set an error parameter judgment threshold, calculate the mean of all running feedback parameters in the parent error slice, and then calculate the absolute difference between each running feedback parameter in the parent error slice and the mean of all running feedback parameters in the parent error slice. Compare the calculated absolute difference with the set error parameter judgment threshold, and identify the error running feedback parameter based on the comparison result. Specifically:

[0019] If the comparison results satisfy the formula This indicates the first error slice in the parent generation. The first running feedback parameter is the error running feedback parameter; otherwise, it represents the first running feedback parameter in the parent error slice. This running feedback parameter is not an error running feedback parameter;

[0020] The selected error feedback parameters are combined in ascending order of their calculated absolute differences to form a second-generation error set.

[0021] As a preferred embodiment of the LED packaging machine error correction method based on data fusion described in this invention, the step of correcting the error operation feedback parameters layer by layer using a two-layer local correction mechanism is as follows:

[0022] Based on the constructed second-generation error set, an error operation feedback parameter is arbitrarily selected from the second-generation error set. At the same time, based on the selected error operation feedback parameter, operation feedback parameters at the same position are selected from adjacent data slices. The spatial similarity between the two operation feedback parameters is calculated, and the error parameter is locally corrected based on the calculated spatial similarity.

[0023] As a preferred embodiment of the LED packaging machine error correction method based on data fusion described in this invention, the local correction of the error parameters is specifically as follows:

[0024] Select the first from the second-generation error set. Error operation feedback parameters And select the running feedback parameters from the same location in adjacent data slices. Calculate the spatial similarity between the two;

[0025] Set spatial similarity threshold Error propagation feedback parameters are determined based on the set spatial similarity threshold.

[0026] For error operation feedback parameters with a propagation trend, calculate the absolute difference between the error operation feedback parameter and the operation feedback parameter at the same position in the adjacent slice, and the error operation feedback parameter with a propagation trend. Then, make a secondary determination of the propagation of the error operation feedback parameter based on the calculated absolute difference.

[0027] As a preferred embodiment of the LED packaging machine error correction method based on data fusion described in this invention, the secondary determination of error operation feedback parameter propagation based on the calculated absolute difference is as follows:

[0028] The calculated absolute difference is compared with the error parameter threshold. If the comparison result satisfies the formula... This indicates that the second-generation error set is selected from the first set. Given a propagation state for each error feedback parameter, and performing local corrections on the current error feedback parameters, we have:

[0029] For the error operation feedback parameters that determine the propagation state, local corrections are made based on the spatial vector direction of the error operation feedback parameters.

[0030] As a preferred embodiment of the LED packaging machine error correction method based on data fusion described in this invention, local self-convergence correction is performed on error operation feedback parameters that do not have a propagation trend, specifically as follows:

[0031] Obtain the parameter value range of the machine parameters corresponding to the error operation feedback parameters during the historical stable operation phase, and determine the stability center value;

[0032] Based on the calculated stable center value, a local self-converging correction is performed.

[0033] As a preferred embodiment of the LED packaging machine error correction method based on data fusion described in this invention, the step of using a stability evaluation function to correct and verify the corrected operating feedback parameters is as follows:

[0034] Collect and correct the operational feedback data of the machine tool;

[0035] Calculate the degree of deviation between the corrected operational feedback data and the baseline operational feedback data, the magnitude of change in machine parameters before and after correction, and the degree of consistency between the corrected operational feedback data and historical stable operational feedback data.

[0036] The calculation results are used as stability evaluation indicators, and a stability evaluation function is constructed. At the same time, the modified running feedback parameters are verified by using the parameter stability threshold.

[0037] The beneficial effects of this invention are:

[0038] This invention achieves refined error identification of the operating status of LED packaging machines by employing multi-source operating data fusion and time slice analysis technology. It can accurately locate the error slice with the largest operating deviation within a continuous operating cycle, thereby solving the problems of coarse error identification granularity and inaccurate positioning in the prior art.

[0039] By adopting a hierarchical iterative error screening mechanism, the step-by-step screening and sorting of error operation feedback parameters is achieved, which effectively avoids the misjudgment problem caused by relying solely on a single threshold judgment and solves the technical problem of insufficient robustness of error parameter identification in the existing technology.

[0040] By employing a spatial similarity-based adjacent slice comparison technique, the system achieves accurate determination of error parameters with propagation trends. It can identify the propagation behavior of errors in the time dimension, thereby performing targeted local corrections for propagating errors and solving the problem that existing technologies cannot identify the propagation of systematic errors.

[0041] By adopting a local self-convergence correction technique driven by historical stable intervals, a smooth regression correction effect on non-propagating error parameters is achieved, enabling isolated error parameters to gradually return to the historical stable center value, avoiding interference with the overall operating status of the machine tool, and solving the problem that non-propagating errors are difficult to correct effectively in the existing technology.

[0042] By adopting a unified stability verification mechanism, the comprehensive consistency evaluation of the machine parameters after double correction is achieved. The reliability of the correction results is verified by multi-dimensional indicators such as deviation degree, parameter change range and historical consistency, thereby solving the technical problem of lack of stability verification of correction results and easy to cause secondary anomalies in the existing technology. Attached Figure Description

[0043] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0044] Figure 1 This is a schematic diagram of the overall method steps of the LED packaging machine error correction method based on data fusion according to the present invention. Detailed Implementation

[0045] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0046] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0047] Example 1

[0048] Reference Figure 1 As an embodiment of the present invention, a method for correcting errors in LED packaging machines based on data fusion is provided, comprising the following steps:

[0049] S1: Collect and identify the operating error characteristics of the LED packaging machine.

[0050] Specifically, collecting and identifying the operational error characteristics of the LED packaging machine involves collecting multi-source operational data of the LED packaging machine under target order conditions, and then identifying errors in the machine's operational status based on the collected multi-source operational data. The specific implementation is as follows:

[0051] Through the machine control interface, production management system, and quality inspection system, the operating data of the LED packaging machine for the current batch of orders within the current production cycle is collected. This includes machine parameter configuration data, actual operation feedback data, and corresponding order process constraint data. The collected operating data is then used to construct an initial operating data set.

[0052]

[0053] in, Indicates the first Machine parameter data corresponding to each run. This represents the operational feedback data corresponding to the machine parameter vector. This represents the order process constraint feature data. This indicates the total number of times data was collected within the current period. This represents the initial set of running data that has been constructed.

[0054] Based on the constructed initial operating data set, the machine's operating status is scanned and errors are identified, specifically as follows:

[0055] If the initial set of running data is divided into slices according to the running time sequence, then we have:

[0056]

[0057] in, This indicates the first running data slice. Indicates the first One running data slice, This indicates the total number of slices, which can be set by the implementer based on the actual application scenario.

[0058] For each data slice, the runtime bias is calculated, specifically:

[0059] Before calculating the runtime deviation for each data slice, a baseline deviation threshold is set using runtime feedback data. Then, we have...

[0060] Randomly select all operational feedback data within a historical operational period, and calculate the mean of all operational feedback data within the historical operational period according to the total number of slices. Use the calculated mean of operational feedback data as the deviation benchmark threshold.

[0061] Based on the set deviation baseline threshold, the running deviation for each data slice is calculated, and then...

[0062]

[0063] in, Indicates the first One running data slice, This indicates the set deviation benchmark threshold. This indicates the total number of slices. Indicates the first The runtime deviation of each running data slice is used for error identification.

[0064] Specifically, error identification utilizes a hierarchical iterative mechanism to identify errors in the machine's operating status layer by layer, as detailed below:

[0065] Based on the calculated operational deviation, the parent error analysis for error identification of the machine tool's operating status is determined as follows:

[0066] Based on the calculated running deviation of each data slice, and arranging the running deviations of each data slice in ascending order, the running data slice with the largest running deviation is selected as the parent error slice. Then, we have...

[0067]

[0068] in, Indicates the first One running data slice, Indicates the first The runtime deviation of each running data slice Let represent a defined parent error slice, which is a set formed by combinations of multiple runtime feedback parameters. Then, we have:

[0069]

[0070] in, This represents the first runtime feedback parameter in the parent error slice. Indicates the first error slice in the parent generation One running feedback parameter, This indicates the total number of running feedback parameters in the parent error slice, which can be set by the implementer according to the actual application scenario.

[0071] Based on the determined parent error slice, error operation feedback parameters are selected using an error threshold, specifically:

[0072] For a given parent error slice, by setting an error threshold, the error operation feedback parameters in the parent error slice are identified, then we have:

[0073] Set an error parameter judgment threshold, calculate the mean of all running feedback parameters in the parent error slice, and then calculate the absolute difference between each running feedback parameter in the parent error slice and the mean of all running feedback parameters in the parent error slice. Compare the calculated absolute difference with the set error parameter judgment threshold, and identify the error running feedback parameter based on the comparison result. Specifically:

[0074] If the comparison results satisfy the formula This indicates the first error slice in the parent generation. The first running feedback parameter is the error running feedback parameter; otherwise, it represents the first running feedback parameter in the parent error slice. This running feedback parameter is not an error running feedback parameter;

[0075] The selected error feedback parameters are combined in ascending order of their calculated absolute differences to form a second-generation error set.

[0076] S2: Based on the constructed second-generation error set, perform error correction on the LED packaging machine.

[0077] Specifically, based on the constructed second-generation error set, a local correction of error parameters is achieved using an adjacent slice comparison mechanism, as follows:

[0078] Based on the constructed second-generation error set, an error operation feedback parameter is arbitrarily selected from the set. Simultaneously, based on this selected parameter, operation feedback parameters at the same location are selected from adjacent data slices. The spatial similarity between the two parameters is calculated. Based on this spatial similarity, local corrections are made to the error parameter. Then, we have…

[0079] Select the first from the second-generation error set. Error operation feedback parameters And select the running feedback parameters from the same location in adjacent data slices. Calculating the spatial similarity between the two, we have:

[0080]

[0081] in, Indicates the selection of the second generation error set. Each error operation feedback parameter This indicates the runtime feedback parameters selected from adjacent data slices at the same location. Let represent the calculated spatial similarity, used for local correction of error parameters. Then, we have:

[0082] Set spatial similarity threshold Based on the set spatial similarity threshold, error propagation feedback parameters are used for judgment, then we have:

[0083] If the calculated spatial similarity, when compared with the set spatial similarity threshold, satisfies the formula... If the error feedback parameter is consistent with the error feedback parameter at the same position in the adjacent slice, then the error feedback parameter has a propagation trend. Conversely, if the error feedback parameter is inconsistent with the error feedback parameter at the same position in the adjacent slice, then the error feedback parameter does not have a propagation trend.

[0084] For error operation feedback parameters with a propagation trend, calculate the absolute difference between the error operation feedback parameter and the operation feedback parameter at the same position in adjacent slices, and the error operation feedback parameter with a propagation trend. Then, based on the calculated absolute difference, perform a secondary determination of the propagation of the error operation feedback parameter. Specifically:

[0085] The calculated absolute difference is compared with the error parameter threshold. If the comparison result satisfies the formula... This indicates that the second-generation error set is selected from the first set. Given a propagation state for each error feedback parameter, and performing local corrections on the current error feedback parameters, we have:

[0086] For the error operation feedback parameters that determine the propagation state, a local correction is made based on the spatial vector direction of the error operation feedback parameters, then we have:

[0087]

[0088] in, This represents the error feedback parameter used to determine the propagation state. This represents the correction factor, which is set by the implementers based on the actual application scenario. This indicates the error operation feedback parameter and the operation feedback parameter at the same position in adjacent slices. This indicates the corrected error feedback parameters.

[0089] For the corrected error operation feedback parameter, re-determine the error operation feedback parameter against the error parameter judgment threshold. If the corrected operation feedback parameter is still an error operation feedback parameter, it means that the current local correction is not complete. Readjust the correction coefficient and perform local correction until the corrected operation feedback parameter passes the error operation feedback parameter verification.

[0090] For error operation feedback parameters that do not exhibit a propagation trend, local self-convergence correction is performed based on the historical stability interval of the corresponding machine parameters. Specifically:

[0091] Obtain the parameter value range of the machine parameters corresponding to the error operation feedback parameters during the historical stable operation phase, and determine the stability center value, specifically:

[0092] Randomly select one error operation feedback parameter from the error operation feedback parameters that have never undergone local correction. And during the historical stable operation phase, extract the stable operation range of the corresponding error operation feedback parameters. And by calculating the stability center value of the error operation feedback parameter, we have:

[0093]

[0094] in, This represents the minimum value within the stable operating range of the error feedback parameter. This represents the maximum value within the stable operating range of the error operation feedback parameter. This represents the center value of the calculated error feedback parameter;

[0095] Based on the calculated stable center value, a local self-convergent correction is performed, specifically as follows:

[0096]

[0097] in, This represents the stability center value of the calculated error feedback parameter. This indicates the error run feedback parameter selected from the error run feedback parameters that have never undergone local correction. This represents the convergence correction coefficient, which is set by the implementer according to the actual application scenario. It represents the error operation feedback parameter after local self-convergence correction.

[0098] For the corrected error operation feedback parameter, re-determine the error operation feedback parameter against the error parameter judgment threshold. If the corrected operation feedback parameter is still an error operation feedback parameter, it means that the current local correction is not complete. Readjust the correction coefficient and perform local correction until the corrected operation feedback parameter passes the error operation feedback parameter verification.

[0099] It should be noted that for the two locally corrected operational feedback parameters, the corrected parameters were uniformly verified for stability to ensure that the parameters after both corrections were normal.

[0100] Stability verification involves collecting operational feedback data from the corrected machine and calculating the deviation between the corrected operational feedback data and the baseline operational feedback data, the magnitude of changes in machine parameters before and after correction, and the degree of consistency between the corrected operational feedback data and historical stable operational feedback data. Specifically:

[0101] The deviation between the corrected operational feedback data and the baseline operational feedback data is calculated by recombining the operational feedback parameters after the two local corrections, and then comparing the combined set of operational feedback parameters with the baseline set of operational feedback parameters, which consists of the stability center value corresponding to each corrected operational feedback data. The calculated spatial similarity is the deviation between the corrected operational feedback data and the baseline operational feedback data.

[0102] The magnitude of the change in machine parameters before and after correction is calculated as the absolute difference between the set of operational feedback parameters and the set of baseline operational feedback parameters, and the calculated absolute difference is used as the magnitude of the change in machine parameters before and after correction.

[0103] The degree of consistency between the corrected operational feedback data and the historical stable operational feedback data is determined by calculating the cosine similarity between the set of operational feedback parameters and the baseline set of operational feedback parameters, and then using the calculated cosine similarity as the degree of consistency between the corrected operational feedback data and the historical stable operational feedback data.

[0104] Using the deviation between the calculated corrected operational feedback data and the baseline operational feedback data, the magnitude of changes in machine parameters before and after correction, and the degree of consistency between the corrected operational feedback data and historical stable operational feedback data as stability indicators, a stability evaluation function is constructed, resulting in:

[0105]

[0106] in, , , These represent weighting coefficients, which are set by the implementers based on the actual application scenario. This indicates the degree of deviation between the corrected operational feedback data and the baseline operational feedback data. This indicates the magnitude of change in machine parameters before and after the correction. This indicates the degree of consistency between the corrected operational feedback data and the historical stable operational feedback data. This represents the constructed stability evaluation function used to verify the consistency of the corrected runtime feedback parameters, specifically:

[0107] Set parameter stability threshold Based on the set parameter stability threshold, the consistency of the corrected runtime feedback parameters is verified, and then we have:

[0108] If the constructed stability evaluation function and the parameter stability threshold satisfy the formula... This indicates that the machine parameters after the two local corrections are normal parameters, and the error operation feedback parameter correction of the LED packaging machine is complete;

[0109] If the constructed stability evaluation function and the parameter stability threshold satisfy the formula... If the error is not corrected, it means that the machine parameters after the two partial corrections are not normal parameters, and the error operation feedback parameter correction of the LED packaging machine is not complete. The parameter correction will be re-executed until the corrected parameters pass the consistency verification.

[0110] Furthermore, if the aforementioned function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0111] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0112] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0113] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for correcting errors in LED packaging machines based on data fusion, characterized in that: Includes the following steps, Collect and identify the error characteristics of LED packaging machines, specifically: The system collects operating data of the LED packaging machine during the production cycle, constructs an initial operating data set, divides the initial operating data set into slices, and uses a hierarchical iteration mechanism to identify error characteristics. Furthermore, error correction for LED packaging machines is performed based on the identified error characteristics, specifically as follows: For the identified error operation feedback parameters, a two-layer local correction mechanism is used to correct the error operation feedback parameters layer by layer, and a stability evaluation function is used to verify the correction of the operation feedback parameters.

2. The method for correcting LED packaging machine errors based on data fusion as described in claim 1, characterized in that: The identification of error features using a hierarchical iterative mechanism is as follows: Collect the operating data of the LED packaging machine under the current batch of orders within the current production cycle, and construct the collected operating data into an initial operating data set. Then, divide the constructed initial operating data set into slices according to the operating time sequence. For each data slice, calculate the runtime bias. Based on the calculated operating deviation, the parent error slice for error identification of machine tool operating status is determined; Based on the determined parent error slices, error operation feedback parameters are selected using error thresholds.

3. The LED packaging machine error correction method based on data fusion as described in claim 2, characterized in that: The parent error slice for error identification of the fixed machine's operating status is as follows: Based on the calculated running deviation of each data slice, and arranging the running deviations of each data slice in ascending order, the running data slice with the largest running deviation is selected from all running deviations as the parent error slice.

4. The LED packaging machine error correction method based on data fusion as described in claim 3, characterized in that: The error operation feedback parameters are filtered using an error threshold, as detailed below: For a given parent error slice, by setting an error threshold, the error operation feedback parameters in the parent error slice are identified, then we have: Set an error parameter judgment threshold, calculate the mean of all running feedback parameters in the parent error slice, and then calculate the absolute difference between each running feedback parameter in the parent error slice and the mean of all running feedback parameters in the parent error slice. Compare the calculated absolute difference with the set error parameter judgment threshold, and identify the error running feedback parameter based on the comparison result. Specifically: If the comparison results satisfy the formula This indicates the first error slice in the parent generation. The first running feedback parameter is the error running feedback parameter; otherwise, it represents the first running feedback parameter in the parent error slice. This running feedback parameter is not an error running feedback parameter; The selected error feedback parameters are combined in ascending order of their calculated absolute differences to form a second-generation error set.

5. The LED packaging machine error correction method based on data fusion as described in claim 4, characterized in that: The error operation feedback parameters are corrected layer by layer using a two-layer local correction mechanism, as detailed below: Based on the constructed second-generation error set, an error operation feedback parameter is arbitrarily selected from the second-generation error set. At the same time, based on the selected error operation feedback parameter, operation feedback parameters at the same position are selected from adjacent data slices. The spatial similarity between the two operation feedback parameters is calculated, and the error parameter is locally corrected based on the calculated spatial similarity.

6. The LED packaging machine error correction method based on data fusion as described in claim 5, characterized in that: The local correction of the error parameters is as follows: Select the first from the second-generation error set. Error operation feedback parameters And select the running feedback parameters from the same location in adjacent data slices. Calculate the spatial similarity between the two; Set spatial similarity threshold Error propagation feedback parameters are determined based on the set spatial similarity threshold. For error operation feedback parameters with a propagation trend, calculate the absolute difference between the error operation feedback parameter and the operation feedback parameter at the same position in the adjacent slice, and the error operation feedback parameter with a propagation trend. Then, make a secondary determination of the propagation of the error operation feedback parameter based on the calculated absolute difference.

7. The LED packaging machine error correction method based on data fusion as described in claim 6, characterized in that: The secondary determination of error operation feedback parameter propagation based on the calculated absolute difference is as follows: The calculated absolute difference is compared with the error parameter threshold. If the comparison result satisfies the formula... This indicates that the second-generation error set is selected from the first set. Given a propagation state for each error feedback parameter, and performing local corrections on the current error feedback parameters, we have: For the error operation feedback parameters that determine the propagation state, local corrections are made based on the spatial vector direction of the error operation feedback parameters.

8. The method for correcting LED packaging machine errors based on data fusion as described in claim 7, characterized in that: For errors that do not have a propagation trend, local self-convergence correction is performed on the operation feedback parameters, as follows: Obtain the parameter value range of the machine parameters corresponding to the error operation feedback parameters during the historical stable operation phase, and determine the stability center value; Based on the calculated stable center value, a local self-converging correction is performed.

9. The LED packaging machine error correction method based on data fusion as described in claim 8, characterized in that: The process of using a stability evaluation function to correct and verify the modified operational feedback parameters is as follows: Collect and correct the operational feedback data of the machine tool; The deviation between the corrected operational feedback data and the baseline operational feedback data, the change range of machine parameters before and after correction, and the consistency between the corrected operational feedback data and historical stable operational feedback data are calculated respectively. The calculation results are used as stability evaluation indicators, and a stability evaluation function is constructed. At the same time, the modified running feedback parameters are verified by using the parameter stability threshold.

Images