Visual light source multi-link collaborative brightness consistency closed-loop control method and system thereof

Abstract

This invention discloses a closed-loop control method and system for multi-stage collaborative brightness consistency of a visual light source, comprising: detecting and classifying LED lamp bead parameters; measuring the difference between the transmittance of the optical diffusion component and a preset value as a compensation factor; assembling the lamp beads and diffusion component into a visual light source and aging it; correcting the target brightness value according to the compensation factor to obtain a calibration value; adjusting the drive current of each aged light source in a closed loop until it meets the standard; generating a digital calibration file containing a unique identifier, calibration current value, illuminance value, and uniformity spectrum, writing it into the light source's memory, and binding a machine-readable identifier for remote access by the client; when the light source is powered on, running a closed-loop control algorithm to adjust the drive current with the standard brightness value as the target and the sampling value of the built-in sensor as feedback, and uploading deviation data; recording the process data with a unique identifier in a data traceability database; and having the analysis and optimization module read the accumulated data to calculate the adjustment strategy for material grading standards and / or pre-aging process parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of visual light source production, manufacturing and intelligent control, and in particular to a multi-link collaborative brightness consistency closed-loop control method for visual light sources and a system thereof. Background Art

[0002] A visual light source is a component specially configured to provide illumination for industrial cameras in machine vision inspection equipment, and is usually composed of a plurality of LED beads, a driving circuit and a diffusion plate. The diffusion plate covers above the LED array, and is configured to diffuse spot-shaped light spots into a uniform planar visual light source. At present, the common practice for producing such visual light sources is as follows: after LED beads enter the factory, they are first subjected to one screening to eliminate beads with parameters obviously out of range, and the remaining beads are directly used for assembly; diffusion plates are uniformly purchased and used according to fixed models, and generally their light transmittance will not be measured separately for each batch of diffusion plates; after the assembly of the visual light source is completed, it directly enters the darkroom calibration step, where an illuminance meter is used to measure the illuminance once facing the light-emitting surface of the visual light source. If the deviation exceeds the preset range, the magnitude of the driving current is adjusted to make the illuminance fall within the allowable range, which is regarded as qualified for delivery, and no special aging process is usually arranged in the middle to allow the beads to go through the initial stage with large brightness fluctuations first. After the visual light source leaves the factory, it is delivered to the customer for use, and the visual light source itself will not perform any calibration actions during the subsequent use process; for the problem of inconsistent brightness caused by unit-to-unit differences or long-term use, the common practice on the client side is to perform unified post-adjustment on the image through image processing algorithms such as histogram equalization and grayscale normalization after image acquisition.

[0003] The above method has the following problems in several aspects:

[0004] First, the diffusion plates are uniformly used according to fixed models, and the actual light transmittance differences of different batches of diffusion plates are not measured separately, nor are they used for subsequent calibration. During factory calibration, the driving current is adjusted uniformly according to the same target illuminance, without corresponding correction for the high or low light transmittance of the batch of diffusion plates. Therefore, the batch-to-batch difference of diffusion plates is directly brought into each finished visual light source, resulting in a generally large difference in central illuminance between different units of the same type of visual light source in actual measurement.

[0005] Second, since no aging process is arranged before the calibration step, the visual light source is often calibrated right after assembly when the brightness is still in a stage of rapid change. The driving current value obtained by calibration is only the result at that moment, and the brightness of the visual light source will continue to drop in a very short time after leaving the factory, making the calibration result determined at the time of delivery difficult to maintain.

[0006] Third, after the visual light source completes its factory calibration, the specific measured values ​​of illuminance, current values ​​of each channel, and uniformity of the light-emitting surface are all stored in the factory and are not delivered to the customer along with the visual light source. The customer only receives a piece of hardware that can emit light normally and cannot know how much this visual light source differs from other sources in terms of illuminance and uniformity. If the customer wants their intelligent vision system (intelligent vision systems include semi-automatic vision inspection systems, automated vision inspection systems, and the increasingly popular AI vision inspection systems) to adapt to this difference in the number of sources, they can only perform uniform preprocessing on the acquired images without distinguishing between individual visual light sources. During processing, it is necessary to cover all possible deviation scenarios, which results in a heavy computational burden.

[0007] Fourth, after the visual light source is delivered to the customer, the LED beads will continue to degrade with the increase of the service life. However, the existing visual light source no longer has any calibration capability after leaving the factory. The driving parameters are fixed and cannot be perceived or compensated for the brightness decay that occurs during use.

[0008] Fifth, the factory has no way of knowing the brightness changes of the visual light source in actual use by the client, so it cannot use this actual usage data to adjust the LED bead selection criteria or aging process parameters. The actual performance of previous batches of products cannot be fed back to the production process, and the consistency level of each batch of products is difficult to continuously improve with the accumulation of production experience. Summary of the Invention

[0009] The purpose of this invention is to overcome the aforementioned deficiencies in the prior art and provide a closed-loop control method and system for multi-stage collaborative brightness consistency of visual light sources. This invention solves the problems of existing technologies by converting batch transmittance differences of diffuser plates into compensation factors and introducing them into the factory calibration process before calibration; adding pre-aging treatment before calibration; generating structured digital files from the calibration data and writing them to local storage along with the visual light source, while binding machine-readable identifiers for remote access by clients; enabling the visual light source to perform power-on self-calibration using built-in sensors; and transmitting self-calibration data and full-process data back to a data traceability database to drive the reverse adjustment of material grading parameter screening standards and / or pre-aging process parameters. In some technologies, the differences in light transmittance between batches of diffuser plates are not measured or used for calibration compensation before being incorporated into the finished visual light source. Furthermore, the lack of an aging process before calibration results in calibration results obtained while the brightness of the visual light source is still rapidly changing, making them difficult to maintain. Data on the illuminance and uniformity of the visual light source are stored internally in the factory and not delivered to customers, forcing customers to perform uniform preprocessing on images without distinguishing individual visual light sources. Additionally, the visual light source lacks calibration capabilities after leaving the factory, making it impossible to perceive and compensate for brightness decay during use. Moreover, the factory has no way of knowing the actual brightness changes of the visual light source during use and cannot adjust the LED selection criteria or aging process parameters accordingly.

[0010] To achieve the above objectives, the present invention is accomplished through the following two aspects:

[0011] Firstly, a closed-loop control method for multi-stage collaborative brightness consistency of visual light sources includes the following steps:

[0012] Material grading steps: The optical and electrical parameters of the LED beads are tested, and they are graded according to the preset parameter screening standards to select LED beads that meet the target grade.

[0013] Optical compensation parameter acquisition steps: Measure the actual transmittance of the optical diffusion components used in this batch, obtain the difference between the actual transmittance and the preset target transmittance, and record the difference as a compensation factor associated with this batch;

[0014] Pre-aging process: Assemble the above-mentioned LED beads and the optical diffusion component into a visual light source, and apply accelerated aging process to the visual light source so that the LED beads can pass through the initial stage of rapid brightness decay and the brightness output tends to be stable.

[0015] Factory calibration steps: Read the compensation factor associated with this batch, correct the preset brightness target value according to the compensation factor, and obtain the calibration target value; in a controlled measurement environment, collect the measured illuminance value of each visual light source after pre-aging treatment and each channel with a standard illuminance meter, use the measured illuminance value as feedback, and adjust the drive current of the corresponding channel in a closed loop until the measured illuminance value reaches the allowable error range of the calibration target value;

[0016] Digital calibration file generation steps: Generate a structured digital calibration file, which at least includes the unique identifier of the visual light source, the calibration drive current value of each channel, the measured value of the center illuminance, and the uniformity spectrum generated by multi-point array illuminance sampling of the light-emitting surface of the visual light source and processing it with an interpolation algorithm.

[0017] Data delivery steps: Write the digital calibration file into the built-in memory of the visual light source body; and bind the machine-readable identifier with the unique identifier so that the client can obtain the digital calibration file from the remote server by reading the machine-readable identifier;

[0018] Power-on self-calibration steps: Each time the visual light source is powered on, it reads the standard brightness value stored in the digital calibration file from the built-in memory as the control target value, uses the real-time sampling value of the built-in brightness sensor of the visual light source as the feedback quantity, runs the closed-loop control algorithm to adjust the drive current until the deviation between the measured brightness and the control target value converges to the preset tolerance range, and uploads the brightness deviation data generated by this calibration to the data traceability database through the data communication interface;

[0019] Data traceability steps: After associating the key process data generated in the above steps with the unique identifier, the data traceability database is recorded.

[0020] Closed-loop optimization step: The analysis and optimization module reads multiple batches of cumulative data from the data traceability database, calculates adjustment strategies according to preset analysis rules, and applies the adjustment strategies to the parameter screening criteria of the material grading step and / or the process parameters of the pre-aging treatment step.

[0021] Preferably, in the material grading step, the preset parameter screening criteria include setting target Bin ranges for the three parameters of LED beads: color temperature, luminous flux, and forward voltage, and performing joint screening of the three parameters on the incoming LED beads, classifying only LED beads whose color temperature, luminous flux, and forward voltage all fall within their respective target Bin ranges as qualified.

[0022] After the material grading step and before the pre-aging step, there is also a layout optimization step: taking the measured luminous flux and forward voltage of the LED beads selected in the material grading step as input, the optimal spatial arrangement scheme of each LED bead on the visual light source substrate is calculated by an optimization algorithm; the optimization algorithm takes maximizing the uniformity of the light spot on the light-emitting surface of the visual light source as the first optimization objective, and making the forward voltage of each LED bead in the same electrical circuit match each other as the second optimization objective.

[0023] The closed-loop optimization step further includes: the analysis and optimization module adjusting the weight parameters of the first optimization objective and the second optimization objective in the optimization algorithm used in the arrangement optimization step based on the light spot uniformity data accumulated in the data traceability database.

[0024] Preferably, in the pre-aging process, the accelerated aging process is carried out at a temperature of 60°C to 80°C for a duration of 48 hours to 100 hours; during the aging process, the brightness output values ​​of each driving channel are collected and recorded at preset time intervals, and the predicted light decay curve of the visual light source is generated by fitting the recorded brightness decay data.

[0025] The predicted light decay curve, which is fitted from the brightness decay data recorded in the pre-aging step, is written into the structured digital calibration file and delivered together with the file.

[0026] In the factory calibration step, the calculation method for obtaining the calibration target value by correcting the preset brightness target value is: calibration target value = preset brightness target value × (1 - compensation factor); the specific processing of the closed-loop adjustment of the drive current is to adjust the PWM duty cycle of the drive circuit or the output value of the programmable current source;

[0027] In the factory calibration step, if it is detected that the brightness output of an individual LED cannot be adjusted to the allowable error range of the calibration target value by adjusting the drive current, the LED is replaced, and the pre-aging step and the factory calibration step are re-executed sequentially on the complete visual light source after the LED is replaced.

[0028] Preferably, in the digital calibration file generation step, the specific generation process of the uniformity map obtained by interpolation processing of the multi-point array sampling is as follows: multi-point matrix illuminance sampling is performed on the light-emitting surface of the visual light source using an area array illuminance meter to obtain discrete illuminance distribution data; two-dimensional interpolation is performed on the discrete illuminance distribution data to generate a continuous illuminance distribution map reflecting the spatial distribution of illuminance on the light surface; the continuous illuminance distribution map is converted into a pseudo-color map and stored in the structured digital calibration file in an encoded image format;

[0029] In the digital calibration file generation step, the structured digital calibration file also includes at least one of the following extended contents: the measured average color temperature of the light-emitting surface of the visual light source; the parameters of the driving circuit board; and the manufacturing date and time.

[0030] Preferably, in the data delivery step, the built-in memory of the visual light source is an electrically erasable programmable read-only memory; the machine-readable identifier is one of a QR code, barcode, RFID tag, or near-field communication tag, and the machine-readable identifier is attached to the visual light source body. The client obtains the digital calibration file corresponding to the unique identifier from the cloud server by reading the machine-readable identifier.

[0031] Preferably, in the power-on self-calibration step, the closed-loop control algorithm is a proportional-integral regulator or a pre-calibration lookup table; the closed-loop control algorithm sets a deviation dead zone, and when the absolute value of the deviation between the measured brightness and the control target value does not exceed the deviation dead zone threshold, the drive current adjustment is not performed; when the deviation exceeds the dead zone threshold, the PWM duty cycle or current value is gradually adjusted by a preset step size ratio of the current drive value until the deviation falls back into the dead zone range or reaches the preset maximum number of adjustments;

[0032] In the power-on self-calibration step, the deviation dead zone threshold is ±2%, the preset step size ratio is 0.5% of the current drive value, and the preset maximum number of adjustments is 20.

[0033] In the power-on self-calibration step and the data traceability step, the brightness deviation data uploaded to the data traceability database is used to update the health status record of the corresponding visual light source; when the brightness deviation data exceeds a preset health threshold, the data traceability database generates maintenance warning information for the visual light source.

[0034] Preferably, the closed-loop optimization step is executed in one of the following two modes: automatic execution mode, in which the analysis and optimization module automatically reads data from the data traceability database at a preset cycle, calculates the adjustment amount according to preset analysis rules, and directly issues the updated parameter screening criteria and / or process parameters to the corresponding production module; or auxiliary decision-making mode, in which the analysis and optimization module generates a process optimization suggestion report containing suggested adjustment parameters and adjustment basis, and after manual review and confirmation, issues the adjustment instruction to the corresponding production module.

[0035] Secondly, the present invention provides a closed-loop control system for multi-stage coordinated brightness consistency of a visual light source, comprising:

[0036] The material grading module is configured to detect the optical and electrical parameters of LED beads, classify them according to preset parameter screening standards, and output qualified LED beads.

[0037] The optical measurement module is configured to measure the actual transmittance of the optical diffusion component, obtain the difference between its transmittance and the preset target transmittance, and use it as a compensation factor to associate with the corresponding batch output.

[0038] The pre-aging chamber is configured to provide a controllable temperature environment to accelerate the aging process of the assembled visual light source until the brightness output tends to stabilize.

[0039] An automated calibration station integrates a standard illuminance meter and a programmable power supply. It is configured to correct the calibration target value according to the compensation factor, and then use the measured illuminance of the standard illuminance meter as feedback to adjust the drive current of each visual light source after pre-aging in a closed loop until the measured illuminance reaches the allowable error range of the calibration target value.

[0040] The file generation and writing module is configured to generate a structured digital calibration file containing a unique identifier of the visual light source, calibration current values ​​of each channel, measured values ​​of central illuminance, and a uniformity spectrum obtained by interpolation of multi-point array sampling, and write the file into the built-in memory of the visual light source.

[0041] The data cloud platform is configured to store the digital calibration files and provide remote on-demand downloads to clients based on a machine-readable identifier bound to the unique identifier;

[0042] The visual light source terminal includes an LED array, a driving circuit, a built-in brightness sensor, a built-in memory, and a processing unit. The processing unit is configured to run a closed-loop control algorithm to adjust the driving current each time it is powered on, using the standard brightness value stored in the built-in memory as the target and the real-time sampling value of the built-in brightness sensor as feedback, and to upload the brightness deviation data through a data communication interface.

[0043] The data traceability database is configured to receive brightness deviation data uploaded by the visual light source terminal, and associate the brightness deviation data with the key process data of the entire process generated by each module and record them together with the unique identifier;

[0044] The analysis and optimization module is configured to calculate and adjust strategies for multiple batches of cumulative data in the data traceability database according to preset analysis rules, and apply the adjustment strategies to the parameter screening criteria of the material grading module and / or the process parameters of the pre-aging chamber.

[0045] Preferably, the built-in memory of the visual light source terminal is an electrically erasable programmable read-only memory; the built-in brightness sensor is a photodiode; the processing unit is configured to run a proportional-integral regulator or a pre-calibration lookup table, and is equipped with a deviation dead zone threshold and a progressive drive adjustment step size.

[0046] Preferably, the system also includes an intelligent design module, which is configured to receive the measured luminous flux and forward voltage data of the LEDs output by the material grading module, and calculate the optimal spatial arrangement of the LEDs on the visual light source substrate through an optimization algorithm. The optimization algorithm aims to maximize the uniformity of the light spot and match the forward voltage of the LEDs within the same electrical circuit. The analysis and optimization module is further configured to adjust the weight parameters of the optimization algorithm in the intelligent design module based on the light spot uniformity data accumulated in the data traceability database.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0048] 1. The advantages of this invention lie in the following steps:

[0049] In the optical compensation parameter acquisition step, the difference in transmittance between batches of diffuser plates is acquired in advance and converted into a compensation factor.

[0050] In the pre-aging process, a pre-aging process is added before factory calibration;

[0051] In the digital calibration file generation step and the data delivery step, the data obtained from factory calibration is generated into a structured digital calibration file that is delivered synchronously with the visual light source.

[0052] In the power-on self-calibration step, the visual light source is given power-on self-calibration capability based on the built-in sensor;

[0053] In the data traceability step and the closed-loop optimization step, a mechanism is established to transmit self-calibration data and full-process data back to the data traceability database to drive the reverse adjustment of material classification standards and pre-aging process parameters.

[0054] By leveraging the advantages mentioned above, the following can be achieved:

[0055] In the optical compensation parameter acquisition step and the factory calibration step, the batch differences of the diffuser plate are actively corrected during the factory calibration stage, and are no longer directly carried into the finished visual light source as in the prior art.

[0056] In the pre-aging process, the driving parameters used for factory calibration are the results measured after the visual light source has passed the initial stage of rapid brightness change, and are no longer difficult to maintain because the visual light source itself is still in the brightness fluctuation period.

[0057] In the digital calibration file generation step and the data delivery step, the measured values ​​of the center illuminance of the visual light source, the calibration current values ​​of each channel, and the optical characteristic data of the uniformity spectrum are delivered to the customer simultaneously with the product, changing the previous situation where such data was kept inside the factory and could not be obtained by the customer.

[0058] In the power-on self-calibration step, the visual light source can continuously self-calibrate with the factory benchmark as the target throughout the entire usage cycle, and has the ability to perceive and compensate for the brightness decay during use.

[0059] In the data traceability step and the closed-loop optimization step, the brightness changes reflected in actual field use are transmitted back to the production end to adjust the material screening standards and pre-aging process parameters for subsequent batches, filling the gap that the factory previously had no way of knowing the field usage.

[0060] The combined effect of the above factors narrows the difference in center illuminance between different displays of the same type of visual light source from ±15% to ±25% commonly found in existing technologies, and stabilizes it within ±5%.

[0061] 2. In addition to the above effects, the present invention also has the following technical advantages: Because the unique identifier runs through every stage from material grading, optical compensation, pre-aging, factory calibration to customer self-calibration, the full-process data in the data traceability database can be accurately linked to specific material batches, diffuser batches, and process parameter values. Once a quality anomaly occurs, the specific stage can be located, eliminating the need to inspect the entire batch of products as in existing technologies. The closed-loop optimization steps allow material grading standards and pre-aging process parameters to be continuously adjusted based on on-site data, transforming the production process from a one-time setting to a continuous iterative process. The consistency level of visual light sources in subsequent batches is expected to gradually improve with data accumulation. Furthermore, after obtaining the specific optical characteristic data of each visual light source, the customer can perform targeted processing on the images collected by the corresponding visual light source, eliminating the need for a uniform preprocessing method that does not distinguish between individuals, as is the case with existing technologies.

[0062] 3. In summary, the visual light source multi-stage collaborative brightness consistency closed-loop control method and system of the present invention integrates previously independent and data-disconnected stages such as material classification, optical compensation, pre-aging, factory calibration, digital file delivery, client self-calibration, data traceability and closed-loop optimization into a complete technical system through unique identification and unified data flow. This systematically improves the visual light source in many aspects, including factory batch consistency, long-term brightness stability during the service life, transparency of optical characteristic data, and the sustainable improvement capability of the production process itself, providing a more reliable lighting foundation for the application of visual light sources in the field of machine vision inspection. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a schematic diagram of the exploded structure of one type of visual light source provided by existing technology;

[0065] Figure 2 This is a flowchart illustrating a closed-loop control method for multi-stage collaborative brightness consistency of a visual light source, provided by an embodiment of the present invention.

[0066] Figure 3 This is a flowchart illustrating the data delivery steps in a closed-loop control method for multi-stage collaborative brightness consistency of a visual light source, provided by an embodiment of the present invention.

[0067] Figure 4 This is a flowchart illustrating the power-on self-calibration step in a closed-loop control method for multi-stage collaborative brightness consistency of a visual light source, as provided in an embodiment of the present invention.

[0068] Figure 5 This is a flowchart illustrating the closed-loop optimization steps in a closed-loop control method for multi-stage collaborative brightness consistency of a visual light source, provided by an embodiment of the present invention. Detailed Implementation

[0069] The technical solution of this embodiment of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiment is one embodiment of the present invention, and not all embodiments thereof. Based on this embodiment of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0070] Example:

[0071] Please see Figures 1 to 5 The present invention provides a method and system for closed-loop control of brightness consistency across multiple stages of a visual light source, comprising the following two aspects:

[0072] In a first aspect, embodiments of the present invention provide a closed-loop control method for multi-stage coordinated brightness consistency of a visual light source, comprising the following steps:

[0073] Material grading steps: The optical and electrical parameters of the LED beads are tested, and they are graded according to the preset parameter screening standards to select LED beads that meet the target grade.

[0074] Optical compensation parameter acquisition steps: Measure the actual transmittance of the optical diffusion components used in this batch, obtain the difference between the actual transmittance and the preset target transmittance, and record the difference as a compensation factor associated with this batch;

[0075] Pre-aging process: Assemble the above-mentioned LED beads and the optical diffusion component into a visual light source, and apply accelerated aging process to the visual light source so that the LED beads can pass through the initial stage of rapid brightness decay and the brightness output tends to be stable.

[0076] Factory calibration steps: Read the compensation factor associated with this batch, correct the preset brightness target value according to the compensation factor, and obtain the calibration target value; in a controlled measurement environment, collect the measured illuminance value of each visual light source after pre-aging treatment and each channel with a standard illuminance meter, use the measured illuminance value as feedback, and adjust the drive current of the corresponding channel in a closed loop until the measured illuminance value reaches the allowable error range of the calibration target value;

[0077] Digital calibration file generation steps: Generate a structured digital calibration file, which at least includes the unique identifier of the visual light source, the calibration drive current value of each channel, the measured value of the center illuminance, and the uniformity spectrum generated by multi-point array illuminance sampling of the light-emitting surface of the visual light source and processing it with an interpolation algorithm.

[0078] Data delivery steps: Write the digital calibration file into the built-in memory of the visual light source body; and bind the machine-readable identifier with the unique identifier so that the client can obtain the digital calibration file from the remote server by reading the machine-readable identifier;

[0079] Power-on self-calibration steps: Each time the visual light source is powered on, it reads the standard brightness value stored in the digital calibration file from the built-in memory as the control target value, uses the real-time sampling value of the built-in brightness sensor of the visual light source as the feedback quantity, runs the closed-loop control algorithm to adjust the drive current until the deviation between the measured brightness and the control target value converges to the preset tolerance range, and uploads the brightness deviation data generated by this calibration to the data traceability database through the data communication interface;

[0080] Data traceability steps: After associating the key process data generated in the above steps with the unique identifier, the data traceability database is recorded.

[0081] Closed-loop optimization step: The analysis and optimization module reads multiple batches of cumulative data from the data traceability database, calculates adjustment strategies according to preset analysis rules, and applies the adjustment strategies to the parameter screening criteria of the material grading step and / or the process parameters of the pre-aging treatment step.

[0082] In the material grading step, the preset parameter screening criteria include setting target Bin ranges for three parameters of LED beads: color temperature, luminous flux, and forward voltage. Incoming LED beads are screened using a three-parameter joint screening process, with only those beads falling within their respective target Bin ranges being classified as qualified. The advantage of this approach is that, compared to screening based on a single parameter, the three-parameter joint screening simultaneously constrains the optical consistency (color temperature, luminous flux) and electrical consistency (forward voltage) of the beads. This ensures matching in terms of emission color and brightness, and also guarantees similar electrical characteristics within the same driving circuit, avoiding uneven current distribution in the circuit due to excessive differences in forward voltage. This reduces the deviation that needs to be compensated for in subsequent layout optimization and factory calibration processes from the source of the materials.

[0083] Following the material grading step and preceding the pre-aging step, an arrangement optimization step is included: using the measured luminous flux and forward voltage of the LED beads selected in the material grading step as input, an optimization algorithm calculates the optimal spatial arrangement of each LED bead on the visual light source substrate; the optimization algorithm takes maximizing the uniformity of the light spot on the light-emitting surface of the visual light source as the first optimization objective, and ensuring that the forward voltages of each LED bead within the same electrical circuit are matched as the second optimization objective. This arrangement optimization step, after material grading and before the pre-aging step, proactively intervenes in brightness distribution and electrical matching from the spatial structure design level of the visual light source. It further advances subsequent factory calibration (current compensation) and client self-calibration (long-term drift compensation), partially solving the brightness consistency problem at the design stage. At the same time, by matching the voltage within the same circuit, it reduces the uneven current distribution caused by differences in LED bead voltages, which helps to reduce the working stress of devices within the circuit.

[0084] The closed-loop optimization step further includes: the analysis and optimization module adjusting the weight parameters of the first and second optimization objectives in the optimization algorithm used in the layout optimization step based on the spot uniformity data accumulated in the data traceability database. The spot uniformity data originates from the statistical patterns of the uniformity maps generated for each visual light source in the factory calibration step across multiple batches, rather than from the client. This is because the built-in brightness sensors relied upon for client-side power-on self-calibration are typically single-point devices, lacking area array measurement capabilities and unable to generate uniformity maps on-site. This step ensures that the weight parameters used in the layout optimization algorithm are no longer one-time, fixed empirical values, but can be continuously corrected based on the uniformity performance of actual manufactured products. This proactively optimizes uniformity, an indicator that was originally only passively compensated during calibration, moving it forward to the design stage, forming a closed loop of joint improvement at both the design and manufacturing levels.

[0085] In the pre-aging process, the accelerated aging process is carried out at a temperature of 60°C to 80°C for 48 to 100 hours. During the aging process, the brightness output values ​​of each driving channel are collected and recorded at preset time intervals. Based on the recorded brightness decay data, a predicted light decay curve for the visual light source is generated. The accelerated aging process allows the LED beads to pass through the initial rapid brightness change stage before leaving the factory, ensuring the stability and effectiveness of the factory calibration results. At the same time, the decay data collected synchronously during the aging process is used to fit the predicted light decay curve, so that the originally single-function aging process has the dual functions of "stabilizing the current brightness" and "predicting future trends". One set of process data is reused for two purposes without adding any additional testing steps.

[0086] The predicted light decay curve, generated by fitting the brightness decay data recorded in the pre-aging step, is written into the structured digital calibration file and delivered along with the file. This setup ensures that the digital calibration file obtained by the customer not only reflects the static characteristics of the visual light source at the moment of manufacture but also includes a predictive description of the brightness change trend of the visual light source over subsequent use. Customers can use this information to plan maintenance or calibration in advance, rather than reacting passively after significant brightness anomalies appear.

[0087] In the factory calibration step, the calculation method for obtaining the calibration target value by correcting the preset brightness target value is: calibration target value = preset brightness target value × (1 - compensation factor); the specific processing of the closed-loop adjustment of the drive current is to adjust the PWM duty cycle of the drive circuit or the output value of the programmable current source.

[0088] In the factory calibration step, if the brightness output of an individual LED is found to be unable to reach the allowable error range of the calibration target value through drive current adjustment, the LED is replaced, and the pre-aging step and the factory calibration step are re-executed sequentially for the complete visual light source after the LED replacement. Specifically, "brightness output unable to reach the allowable error range of the calibration target value through drive current adjustment" can be manifested as follows: the drive current of the channel has been adjusted to the preset upper or lower current limit, but the measured illuminance still does not fall within the allowable error range of the calibration target value. In this case, the LED corresponding to that channel is determined to be an abnormal LED. This fault-tolerance mechanism avoids the entire visual light source being directly scrapped due to quality fluctuations of individual LEDs, improving product yield while ensuring consistent factory brightness requirements. By re-performing pre-aging and factory calibration on the entire light source after LED replacement, it is ensured that the visual light source after abnormal handling has the same factory quality and data integrity as visual light sources produced through normal processes.

[0089] In the digital calibration file generation step, the specific generation process of the uniformity map obtained by interpolation processing of the multi-point array sampling is as follows: Multi-point matrix illuminance sampling is performed on the light-emitting surface of the visual light source using an area array illuminance meter to obtain discrete illuminance distribution data; two-dimensional interpolation is performed on the discrete illuminance distribution data to generate a continuous illuminance distribution map reflecting the spatial distribution of illuminance on the light surface; the continuous illuminance distribution map is converted into a pseudo-color map and stored in the structured digital calibration file in an encoded image format. The two-dimensional interpolation operation can employ bilinear interpolation, bicubic interpolation, or other common two-dimensional interpolation algorithms; the encoded image format can be a base64 encoded grayscale image or other image encoding formats such as PNG. By combining multi-point sampling with two-dimensional interpolation to generate a continuous illuminance spatial distribution, compared to recording only the single value of the illuminance at the center point, it can more completely reflect the actual brightness differences of each area of ​​the light-emitting surface of the visual light source. It is converted into a pseudo-color spectrum and stored in a standard image format, which is convenient for parsing and display in client software. It provides the client's AI vision system with a pixel-by-pixel and region-by-region brightness compensation basis, rather than just obtaining a general center illuminance value.

[0090] In the digital calibration file generation step, the structured digital calibration file further includes at least one of the following extended contents: the measured average color temperature of the visual light source's emitting surface; driver circuit board parameters; and manufacturing date and time. These extended contents make the digital calibration file more complete and comprehensive: the manufacturing date and time can serve as a time reference for calculating the actual usage time of the visual light source, and, combined with the predicted light decay curve, can be used to estimate the expected brightness level at the current moment; the measured average color temperature facilitates targeted color compensation for customers in color-sensitive detection scenarios; and the driver circuit board parameters help customers confirm the consistency of model and specifications during subsequent maintenance or replacement of the driver circuit board.

[0091] In the data delivery step, the built-in memory of the visual light source is an electrically erasable programmable read-only memory; the machine-readable identifier is one of a QR code, barcode, RFID tag, or near-field communication tag, and is attached to the visual light source body. The client obtains the digital calibration file corresponding to the unique identifier from the cloud server by reading the machine-readable identifier. The digital calibration file is written to the built-in memory via an I2C communication interface or other common inter-chip communication interfaces such as SPI.

[0092] In the power-on self-calibration step, the closed-loop control algorithm is a proportional-integral regulator or a pre-calibration lookup table. The closed-loop control algorithm sets a deviation dead zone. When the absolute value of the deviation between the measured brightness and the control target value does not exceed the deviation dead zone threshold, no drive current adjustment is performed. When the deviation exceeds the dead zone threshold, the PWM duty cycle or current value is gradually adjusted by a preset step size ratio of the current drive value until the deviation falls back into the dead zone range or reaches the preset maximum number of adjustments. If, after the number of adjustments reaches the preset maximum number of adjustments, the deviation between the measured brightness and the control target value still has not fallen back into the deviation dead zone range, the current adjustment process can be judged as an abnormal state. This abnormal state and the measured brightness value at this time are recorded, and corresponding abnormal prompt information can be selectively generated. The setting of the deviation dead zone avoids frequent and repeated adjustments caused by sensor noise or minor environmental fluctuations, reducing unnecessary actions of the drive circuit and built-in sensor; the progressive step size adjustment method avoids the impact of large current surges on LED beads, making the adjustment process smoother; the preset maximum number of adjustments prevents the adjustment action from being executed infinitely in the case of non-convergence, taking into account both adjustment stability and reasonable consumption of system resources.

[0093] Specifically, the deviation dead zone threshold is ±2%, the preset step size ratio is 0.5% of the current drive value, and the preset maximum number of adjustments is 20.

[0094] In the power-on self-calibration step and the data traceability step, the brightness deviation data uploaded to the data traceability database is used to update the health status record of the corresponding visual light source. When the brightness deviation data exceeds a preset health threshold, the data traceability database generates maintenance warning information for the visual light source. The maintenance warning information can be pushed to the client to remind on-site equipment managers that the visual light source needs attention or maintenance arrangements; it can also be aggregated to the manufacturer's backend to analyze whether there are common abnormal health status trends in visual light sources from specific batches or specific material sources.

[0095] The closed-loop optimization steps are executed in one of two modes: Automatic execution mode, where the analysis and optimization module automatically reads data from the data traceability database at preset intervals, calculates adjustments according to preset analysis rules, and directly issues the updated parameter screening criteria and / or process parameters to the corresponding production modules; and Assisted decision-making mode, where the analysis and optimization module generates a process optimization suggestion report containing suggested adjustment parameters and the basis for adjustment, which is then manually reviewed and confirmed before the adjustment instructions are issued to the corresponding production modules. The automatic execution mode enables rapid process parameter iteration without manual intervention; the Assisted decision-making mode, by introducing a manual review and confirmation step, provides necessary safety redundancy for process parameter adjustments, avoiding irreversible impacts on the entire production line caused by erroneous adjustments due to abnormal data analysis or individual sample deviations. The coexistence of these two modes allows the closed-loop optimization mechanism to achieve a flexible balance between efficiency and safety.

[0096] Secondly, embodiments of the present invention provide a closed-loop control system for multi-stage coordinated brightness consistency of a visual light source, comprising:

[0097] The material grading module is configured to detect the optical and electrical parameters of LED beads, classify them according to preset parameter screening standards, and output qualified LED beads.

[0098] The optical measurement module is configured to measure the actual transmittance of the optical diffusion component, obtain the difference between its transmittance and the preset target transmittance, and use it as a compensation factor to associate with the corresponding batch output.

[0099] The pre-aging chamber is configured to provide a controllable temperature environment to accelerate the aging process of the assembled visual light source until the brightness output tends to stabilize.

[0100] An automated calibration station integrates a standard illuminance meter and a programmable power supply. It is configured to correct the calibration target value according to the compensation factor, and then use the measured illuminance of the standard illuminance meter as feedback to adjust the drive current of each visual light source after pre-aging in a closed loop until the measured illuminance reaches the allowable error range of the calibration target value.

[0101] The file generation and writing module is configured to generate a structured digital calibration file containing a unique identifier of the visual light source, calibration current values ​​of each channel, measured values ​​of central illuminance, and a uniformity spectrum obtained by interpolation of multi-point array sampling, and write the file into the built-in memory of the visual light source.

[0102] The data cloud platform is configured to store the digital calibration files and provide remote on-demand downloads to clients based on a machine-readable identifier bound to the unique identifier;

[0103] The visual light source terminal includes an LED array, a driving circuit, a built-in brightness sensor, a built-in memory, and a processing unit. The processing unit is configured to run a closed-loop control algorithm to adjust the driving current each time it is powered on, using the standard brightness value stored in the built-in memory as the target and the real-time sampling value of the built-in brightness sensor as feedback, and to upload the brightness deviation data through a data communication interface.

[0104] The data traceability database is configured to receive brightness deviation data uploaded by the visual light source terminal, and associate the brightness deviation data with the key process data of the entire process generated by each module and record them together with the unique identifier;

[0105] The analysis and optimization module is configured to calculate and adjust strategies for multiple batches of cumulative data in the data traceability database according to preset analysis rules, and apply the adjustment strategies to the parameter screening criteria of the material grading module and / or the process parameters of the pre-aging chamber.

[0106] The built-in memory of the visual light source terminal is an electrically erasable programmable read-only memory; the built-in brightness sensor is a photodiode; the processing unit is configured to run a proportional-integral regulator or a pre-calibration lookup table, and is set with a deviation dead zone threshold and a progressive drive adjustment step size.

[0107] This closed-loop control system for multi-stage collaborative brightness consistency of a visual light source also includes an intelligent design module. The intelligent design module is configured to receive measured luminous flux and forward voltage data of the LEDs output by the material grading module, and calculate the optimal spatial arrangement of the LEDs on the visual light source substrate using an optimization algorithm. The optimization algorithm aims to maximize light spot uniformity and match the forward voltage of the LEDs within the same electrical circuit. The analysis and optimization module is further configured to adjust the weight parameters of the optimization algorithm in the intelligent design module based on the light spot uniformity data accumulated in the data traceability database. The optimization algorithm used in the intelligent design module can be one or a combination of common optimization algorithms such as genetic algorithms, simulated annealing algorithms, and linear programming algorithms.

[0108] The present invention provides a closed-loop control method and system for multi-stage collaborative brightness consistency of visual light sources, the specific working principle of which is as follows:

[0109] First, the material grading module individually tests the optical and electrical parameters of incoming LED chips, such as color temperature, luminous flux, and forward voltage. The chips are then categorized according to preset screening criteria, and only those falling within their respective target grading ranges are selected for subsequent assembly. When the system is further equipped with an intelligent design module, it can calculate the optimal spatial arrangement of each LED chip on the visual light source substrate based on the measured luminous flux and forward voltage of the selected chips using an optimization algorithm. This ensures that LED chips with high luminous flux are placed in edge areas prone to optical attenuation, and that LED chips with similar forward voltages are grouped into the same electrical circuit. This approach improves light uniformity and reduces electrical mismatch within the circuit from the initial layout design stage. At the same time, the optical measurement module measures the actual transmittance of the optical diffusion components (diffuser plate, diffuser film or diffuser sheet) used in this batch, obtains the difference value between it and the preset target transmittance, and records the difference value as a compensation factor in relation to this batch for use in the subsequent factory calibration process. This step allows the transmittance difference between batches of diffusion components, which was originally unavoidable, to be quantified in advance before the visual light source is assembled, and is no longer a source of unknown error that is passively borne during the factory calibration stage.

[0110] After the LED beads and diffusion components are assembled into a complete visual light source, they are sent to a pre-aging chamber and aged continuously for 48 to 100 hours in a controlled high-temperature environment of 60°C to 80°C, so that the LED beads can accelerate through the initial rapid brightness decay stage. During the aging process, the brightness decay data of each driving channel is collected and recorded synchronously at preset time intervals to fit the predicted light decay curve of the visual light source. This makes the originally single-function aging process have the dual functions of "stabilizing the current factory state" and "predicting the future long-term trend".

[0111] After aging, the visual light source enters the automated calibration station. The calibration station reads the compensation factor associated with the batch corresponding to the visual light source, and corrects the brightness target value required by the order to the calibration target value that has taken into account the differences of the diffusion components, according to the conversion relationship of calibration target value = preset brightness target value × (1 - compensation factor). Then, the illuminance of each visual light source is measured and collected channel by channel using a standard illuminance meter. The measured value is used as feedback to adjust the PWM duty cycle of the drive circuit or the output value of the programmable current source, and the drive current of each channel is adjusted in a closed loop until the measured illuminance converges to the allowable error range of the calibration target value. If it is found that individual channels still cannot make the measured illuminance fall within the allowable error range even when the current is adjusted to the upper or lower limit, the corresponding LED is determined to be abnormal and replaced. The replacement lamp is then subjected to the pre-aging and factory calibration process again to ensure that the visual light source after abnormal treatment has the same factory data integrity as the visual light source produced by the normal process.

[0112] After successful calibration, a structured digital calibration file is generated by the file generation and writing module. On one hand, multi-point matrix illuminance sampling is performed on the light-emitting surface of the visual light source using an area array illuminance meter. The resulting discrete illuminance data is then converted into a continuous illuminance distribution through two-dimensional interpolation, converted into a pseudo-color spectrum, and stored in a file in an coded image format. On the other hand, the unique identifier of the visual light source, the final calibration current value of each channel, the measured illuminance value at the center point, and the aforementioned uniformity spectrum are written into the file. Optional extensions include the measured average color temperature value, driver board parameters, manufacturing date and time, and the predicted light decay curve obtained from the aforementioned pre-aging stage. This digital calibration file is then written into the built-in memory of the visual light source (such as an electrically erasable programmable read-only memory). Simultaneously, a machine-readable identifier (QR code, barcode, RFID tag, or NFC tag) bound to the unique identifier of the visual light source is attached to the visual light source. The data cloud platform simultaneously saves a cloud copy of this file, allowing customers to remotely access it by reading the machine-readable identifier. This achieves a separate delivery model where the delivery of visual light source hardware and the delivery of optical characteristic data are independent yet interconnected through a unique identifier.

[0113] After the visual light source is delivered to the customer and put into use, the processing unit of the visual light source terminal reads the standard brightness value recorded in the digital calibration file from the built-in memory as the control target value each time it is powered on. It uses the real-time sampling value from the built-in brightness sensor (such as a photodiode) as feedback and runs a proportional-integral regulator or a pre-calibration lookup table algorithm: when the deviation between the measured brightness and the target value falls within a preset dead zone (e.g., ±2%), no adjustment is made to avoid repeated actions caused by sensor noise; when the deviation exceeds the dead zone, the PWM duty cycle or drive current is gradually adjusted according to a preset step size ratio (e.g., 0.5% of the current drive value) until the deviation falls back into the dead zone or reaches the preset maximum number of adjustments (e.g., 20 times). Therefore, the calibration benchmark determined at the factory does not cease to function after the visual light source is delivered, but continues to serve as a reference for on-site self-calibration, extending the brightness maintenance capability of the visual light source throughout its entire service life. After each self-calibration is completed, the processing unit uploads the brightness deviation data generated this time to the data traceability database through the data communication interface. When the uploaded deviation data exceeds the preset health threshold, the data traceability database can also generate corresponding maintenance warning information for the visual light source, so that customers can arrange maintenance in advance, rather than passively responding after the visual light source has obvious failures.

[0114] Meanwhile, the data traceability database uses the unique identifier of the visual light source as an index to link and record key process data generated at each step, such as material grading parameters, batch transmittance of diffusion components, pre-aging process parameters, and factory calibration data, together with the self-calibration deviation data continuously returned by the aforementioned client, forming a complete data archive covering the entire lifecycle of the visual light source. The analysis and optimization module periodically or as needed extracts multiple batches of accumulated data from the data traceability database, calculates adjustment strategies according to preset analysis rules, and applies these adjustment strategies to the parameter screening criteria of the material grading module and / or the process parameters of the pre-aging chamber. This adjustment process can be performed in a fully automatic mode, where the analysis and optimization module directly calculates and issues adjustment instructions; or in an auxiliary decision-making mode, where the analysis and optimization module generates an optimization suggestion report, which is then manually reviewed and confirmed before being issued. The choice between the two modes can be flexibly made based on the maturity of the specific process. If the system is further configured with an intelligent design module, the analysis and optimization module can also adjust the weight parameters of the layout optimization algorithm in the intelligent design module based on the light spot uniformity data accumulated in the data traceability database.

[0115] Therefore, this invention connects the material classification, (optional step) layout optimization, optical compensation parameter acquisition, pre-aging, factory calibration, digital calibration file generation and delivery, client power-on self-calibration, data traceability and closed-loop optimization, etc., into a complete closed-loop system with bidirectional data flow and interconnected steps through three data threads: the pre-transmission of compensation factors, the connection between digital calibration files and factory benchmarks and on-site self-calibration, and the continuous back-transmission of self-calibration data to the data traceability database. This makes each step no longer an isolated and independent process, but jointly serve the overall goal of improving the brightness consistency of the visual light source at the factory and ensuring its long-term stability.

[0116] Referring to the aforementioned specific working principle, the advantages of the visual light source multi-stage collaborative brightness consistency closed-loop control method and system of the present invention are as follows:

[0117] 1. Significantly improved consistency across product batches. Material grading limits the dispersion range of parameters such as color temperature, luminous flux, and forward voltage from the source of the LED chips; batch-to-batch transmittance differences in diffusion components are measured and converted into compensation factors before the visual light source is assembled, which are then actively used in the factory calibration process; pre-aging treatment ensures that the visual light source enters the calibration process only after it has passed the initial rapid brightness change stage. The combined effect of these multiple technical measures, none of which can be omitted, has stabilized the center illuminance difference between different batches of the same type of visual light source from the commonly existing ±15% to ±25% to within ±5%.

[0118] 2. The optical characteristics of visual light sources are delivered transparently, enabling clients to perform targeted compensation. A structured digital calibration file generated at the factory for each visual light source delivers its unique identifier, calibration current values ​​for each channel, measured center illuminance, and uniformity spectrum, among other characteristic data, to the customer via both built-in storage and cloud download. This changes the previous technology where visual light sources were delivered as a "black box," leaving customers with no way of knowing their specific optical characteristics. The customer's intelligent vision system (which includes semi-automatic vision inspection systems, automated vision inspection systems, and the increasingly prevalent AI vision inspection systems) can then apply targeted image processing or brightness compensation methods to each visual light source, eliminating the need for uniform preprocessing applied to all visual light sources without differentiation.

[0119] 3. Brightness maintenance capability extends from the factory setting to the entire service life. The brightness benchmark established by the factory calibration does not cease to function after the visual light source is delivered. Instead, through the built-in sensor and power-on self-calibration mechanism of the visual light source, it is continuously used as a calibration target during customer use, enabling the brightness of the visual light source to be maintained near the factory benchmark for a long time, reducing the frequency of on-site calibration and maintenance costs caused by long-term light decay.

[0120] 4. The pre-aging process achieves the dual purpose of "stabilizing the current state" and "predicting the future". The aging process is used to accelerate the LED beads through the initial unstable brightness stage and ensure that the factory calibration results are based on a stable state. At the same time, the brightness decay data during the aging period is collected to fit the predicted light decay curve. One process serves two purposes at the same time without adding additional testing links or process costs.

[0121] 5. The production process has the capability for continuous improvement based on actual usage data. Key data from each stage of material, production, calibration, delivery, and client use are all recorded in the data traceability database with a unique identifier for the visual light source. On-site data generated by client self-calibration can be transmitted back, allowing the analysis and optimization module to make reverse adjustments to material grading standards and / or pre-aging process parameters. This ensures that the production process is no longer a one-time, fixed setting, but can be continuously iterated with data accumulation, and the consistency level of visual light sources in subsequent batches is expected to gradually improve.

[0122] 6. It can further balance yield assurance with proactive optimization during the design phase. If individual LED beads are found to have substandard performance during factory calibration, this invention supports replacing the LED bead and then performing a complete pre-aging and calibration process again, avoiding the direct scrapping of the entire visual light source due to the abnormality of a single LED bead. In the implementation of the configuration and layout optimization design, the light emission uniformity and electrical matching within the same circuit can also be proactively improved during the visual light source structure design phase based on the measured parameters of the LED beads, providing another layer of protection for brightness consistency beyond the source.

[0123] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A closed-loop control method for multi-stage coordinated brightness consistency of a visual light source, characterized in that, Includes the following steps: Material grading steps: The optical and electrical parameters of the LED beads are tested, and they are graded according to the preset parameter screening standards to select LED beads that meet the target grade. Optical compensation parameter acquisition steps: Measure the actual transmittance of the optical diffusion components used in this batch, obtain the difference between the actual transmittance and the preset target transmittance, and record the difference as a compensation factor associated with this batch; Pre-aging process: Assemble the above-mentioned LED beads and the optical diffusion component into a visual light source, and apply accelerated aging process to the visual light source so that the LED beads can pass through the initial stage of rapid brightness decay and the brightness output tends to be stable. Factory calibration steps: Read the compensation factor associated with this batch, correct the preset brightness target value according to the compensation factor, and obtain the calibration target value; in a controlled measurement environment, collect the measured illuminance value of each visual light source after pre-aging treatment and each channel with a standard illuminance meter, use the measured illuminance value as feedback, and adjust the drive current of the corresponding channel in a closed loop until the measured illuminance value reaches the allowable error range of the calibration target value; Digital calibration file generation steps: Generate a structured digital calibration file, which at least includes a unique identifier of the visual light source, calibration drive current values ​​for each channel, measured values ​​of central illuminance, and a uniformity spectrum generated by multi-point array illuminance sampling of the light-emitting surface of the visual light source and processing it with an interpolation algorithm. Data delivery steps: Write the digital calibration file into the built-in memory of the visual light source body; The machine-readable identifier is bound to the unique identifier, so that the client can obtain the digital calibration file from the remote server by reading the machine-readable identifier; Power-on self-calibration steps: Each time the visual light source is powered on, it reads the standard brightness value stored in the digital calibration file from the built-in memory as the control target value, uses the real-time sampling value of the built-in brightness sensor of the visual light source as the feedback quantity, runs the closed-loop control algorithm to adjust the drive current until the deviation between the measured brightness and the control target value converges to the preset tolerance range, and uploads the brightness deviation data generated by this calibration to the data traceability database through the data communication interface; Data traceability steps: After associating the key process data generated in the above steps with the unique identifier, the data traceability database is recorded. Closed-loop optimization step: The analysis and optimization module reads multiple batches of cumulative data from the data traceability database, calculates adjustment strategies according to preset analysis rules, and applies the adjustment strategies to the parameter screening criteria of the material grading step and / or the process parameters of the pre-aging treatment step.

2. The method for closed-loop control of multi-stage coordinated brightness consistency of a visual light source according to claim 1, characterized in that, In the material grading step, the preset parameter screening criteria include setting target Bin ranges for the three parameters of LED beads: color temperature, luminous flux, and forward voltage. The incoming LED beads are jointly screened using the three parameters, and only LED beads whose color temperature, luminous flux, and forward voltage all fall within their respective target Bin ranges are classified as qualified. After the material grading step and before the pre-aging step, there is also a layout optimization step: taking the measured luminous flux and forward voltage of the LED beads selected in the material grading step as input, the optimal spatial arrangement scheme of each LED bead on the visual light source substrate is calculated by an optimization algorithm; the optimization algorithm takes maximizing the uniformity of the light spot on the light-emitting surface of the visual light source as the first optimization objective, and making the forward voltage of each LED bead in the same electrical circuit match each other as the second optimization objective. The closed-loop optimization step further includes: the analysis and optimization module adjusting the weight parameters of the first optimization objective and the second optimization objective in the optimization algorithm used in the arrangement optimization step based on the light spot uniformity data accumulated in the data traceability database.

3. The method for closed-loop control of brightness consistency across multiple stages of a visual light source according to claim 1, characterized in that, In the pre-aging process, the accelerated aging process is carried out at a temperature of 60°C to 80°C for a duration of 48 to 100 hours. During the aging process, the brightness output values ​​of each driving channel are collected and recorded at preset time intervals, and the predicted light decay curve of the visual light source is generated by fitting the recorded brightness decay data. The predicted light decay curve, which is fitted from the brightness decay data recorded in the pre-aging step, is written into the structured digital calibration file and delivered together with the file. In the factory calibration step, the calculation method for obtaining the calibration target value by correcting the preset brightness target value is: calibration target value = preset brightness target value × (1 - compensation factor); the specific processing of the closed-loop adjustment of the drive current is to adjust the PWM duty cycle of the drive circuit or the output value of the programmable current source; In the factory calibration step, if it is detected that the brightness output of an individual LED cannot be adjusted to the allowable error range of the calibration target value by adjusting the drive current, the LED is replaced, and the pre-aging step and the factory calibration step are re-executed sequentially on the complete visual light source after the LED is replaced.

4. The method for closed-loop control of multi-stage coordinated brightness consistency of a visual light source according to claim 1, characterized in that, In the digital calibration file generation step, the specific generation process of the uniformity map obtained by interpolation of the multi-point array sampling is as follows: multi-point matrix illuminance sampling is performed on the light-emitting surface of the visual light source using an area array illuminance meter to obtain discrete illuminance distribution data; two-dimensional interpolation is performed on the discrete illuminance distribution data to generate a continuous illuminance distribution map reflecting the spatial distribution of illuminance on the light surface; the continuous illuminance distribution map is converted into a pseudo-color map and stored in the structured digital calibration file in an encoded image format; In the digital calibration file generation step, the structured digital calibration file also includes at least one of the following extended contents: the measured average color temperature of the light-emitting surface of the visual light source; the parameters of the driving circuit board; and the manufacturing date and time.

5. The method for closed-loop control of brightness consistency across multiple stages of a visual light source according to claim 1, characterized in that, In the data delivery step, the built-in memory of the visual light source is an electrically erasable programmable read-only memory; the machine-readable identifier is one of a QR code, barcode, RFID tag, or near-field communication tag. The machine-readable identifier is attached to the visual light source body, and the client obtains the digital calibration file corresponding to the unique identifier from the cloud server by reading the machine-readable identifier.

6. The method for closed-loop control of brightness consistency across multiple stages of a visual light source according to claim 1, characterized in that, In the power-on self-calibration step, the closed-loop control algorithm is a proportional-integral regulator or a pre-calibration lookup table; the closed-loop control algorithm sets a deviation dead zone, and when the absolute value of the deviation between the measured brightness and the control target value does not exceed the deviation dead zone threshold, the drive current adjustment is not performed; When the deviation exceeds the dead zone threshold, the PWM duty cycle or current value is gradually adjusted by a preset step size ratio of the current drive value until the deviation falls back into the dead zone range or reaches the preset maximum number of adjustments. In the power-on self-calibration step, the deviation dead zone threshold is ±2%, the preset step size ratio is 0.5% of the current drive value, and the preset maximum number of adjustments is 20. In the power-on self-calibration step and the data traceability step, the brightness deviation data uploaded to the data traceability database is used to update the health status record of the corresponding visual light source; When the brightness deviation data exceeds a preset health threshold, the data traceability database generates maintenance warning information for the visual light source.

7. The method for closed-loop control of brightness consistency across multiple stages of a visual light source according to claim 1, characterized in that, The closed-loop optimization step is executed in one of the following two modes: automatic execution mode, in which the analysis and optimization module automatically reads data from the data traceability database according to a preset cycle, calculates the adjustment amount according to the preset analysis rules, and directly issues the updated parameter screening criteria and / or process parameters to the corresponding production module; In the decision support mode, the analysis and optimization module generates a process optimization suggestion report containing suggested adjustment parameters and the basis for adjustment. After manual review and confirmation, the adjustment instruction is issued to the corresponding production module.

8. A closed-loop control system for multi-stage coordinated brightness consistency of a visual light source, characterized in that, include: The material grading module is configured to detect the optical and electrical parameters of LED beads, classify them according to preset parameter screening standards, and output qualified LED beads. The optical measurement module is configured to measure the actual transmittance of the optical diffusion component, obtain the difference between its transmittance and the preset target transmittance, and use it as a compensation factor to associate with the corresponding batch output. The pre-aging chamber is configured to provide a controllable temperature environment to accelerate the aging process of the assembled visual light source until the brightness output tends to stabilize. An automated calibration station integrates a standard illuminance meter and a programmable power supply. It is configured to correct the calibration target value according to the compensation factor, and then use the measured illuminance of the standard illuminance meter as feedback to adjust the drive current of each visual light source after pre-aging in a closed loop until the measured illuminance reaches the allowable error range of the calibration target value. The file generation and writing module is configured to generate a structured digital calibration file containing a unique identifier of the visual light source, calibration current values ​​of each channel, measured values ​​of central illuminance, and a uniformity spectrum obtained by interpolation of multi-point array sampling, and write the file into the built-in memory of the visual light source. The data cloud platform is configured to store the digital calibration files and provide remote on-demand downloads to clients based on a machine-readable identifier bound to the unique identifier; The visual light source terminal includes an LED array, a driving circuit, a built-in brightness sensor, a built-in memory, and a processing unit. The processing unit is configured to run a closed-loop control algorithm to adjust the driving current each time it is powered on, using the standard brightness value stored in the built-in memory as the target and the real-time sampling value of the built-in brightness sensor as feedback, and to upload the brightness deviation data through a data communication interface. The data traceability database is configured to receive brightness deviation data uploaded by the visual light source terminal, and associate the brightness deviation data with the key process data of the entire process generated by each module and record them together with the unique identifier; The analysis and optimization module is configured to calculate and adjust strategies for multiple batches of cumulative data in the data traceability database according to preset analysis rules, and apply the adjustment strategies to the parameter screening criteria of the material grading module and / or the process parameters of the pre-aging chamber.

9. A closed-loop control system for multi-stage coordinated brightness consistency of a visual light source according to claim 8, characterized in that, The built-in memory of the visual light source terminal is an electrically erasable programmable read-only memory. The built-in brightness sensor is a photodiode; the processing unit is configured to run a proportional-integral regulator or a pre-calibration lookup table, and is set with a deviation dead zone threshold and a progressive drive adjustment step size.

10. A closed-loop control system for multi-stage coordinated brightness consistency of a visual light source according to claim 9, characterized in that, It also includes an intelligent design module, which is configured to receive the measured luminous flux and forward voltage data of the LED beads output by the material classification module, and calculate the optimal spatial arrangement of the LED beads on the visual light source substrate through an optimization algorithm; the optimization algorithm takes maximizing the uniformity of the light spot and matching the forward voltage of the LED beads in the same electrical circuit as the optimization objectives; the analysis and optimization module is further configured to adjust the weight parameters of the optimization algorithm in the intelligent design module based on the light spot uniformity data accumulated in the data traceability database.

Images