A Smart Control Method for Lighting Driver Power Supply

Abstract

This application provides an intelligent control method for a lighting driver power supply, relating to the field of lighting control technology. The method includes: building a digital twin model based on regional information; distributing virtual test points according to regional evaluation results, performing lighting fitting based on the digital twin model, and monitoring lighting effects through the virtual test points to establish location point monitoring results; monitoring and evaluating the location point monitoring results using a lighting evaluation network to generate target control results for each lighting device; establishing a control prediction model based on historical control information, whereby the control prediction model is a prediction model for output control results based on power supply parameters; using the target control results as the control objective, performing inverse fitting of power supply parameters through the control prediction model to generate calibrated power supply parameters; and performing drive control of the lighting devices. This application can solve the technical problem of poor lighting effects in existing lighting systems, achieving the technical effect of improving lighting effects.

Description

Technical Field

[0001] This application relates to the field of lighting control technology, and in particular to an intelligent control method for a lighting driver power supply. Background Technology

[0002] With the development of intelligent lighting systems, the automation and intelligence of lighting control have become key to improving energy efficiency and user experience. Existing intelligent control methods for lighting drivers typically involve monitoring the lighting environment and using feedback-based control strategies to adjust parameters such as lighting intensity and color temperature to adapt to different environments and user needs. This includes using traditional sensors for physical measurements and controlling based on the measurement results. However, existing technologies have some shortcomings that affect the overall performance and efficiency of the lighting system.

[0003] Currently, traditional methods are limited by fixed sensor networks, resulting in a limited number and location of monitoring points, which cannot fully cover the entire lighting area, thus affecting the uniformity and adaptability of the lighting effect. In addition, there is often a delay in responding to environmental changes, a lack of rapid processing and feedback capabilities for real-time data, and a lack of personalized and context-aware control strategies, leading to insufficient precision in lighting adjustments and energy waste.

[0004] In summary, existing technologies suffer from limitations in the number and location of monitoring points in traditional lighting systems, resulting in deficiencies in dynamic environmental adaptability, efficiency, and personalized control. This leads to poor performance in energy utilization, user satisfaction, and operational flexibility, ultimately affecting overall lighting quality and economic benefits. Summary of the Invention

[0005] The purpose of this application is to provide an intelligent control method for lighting drive power supplies, in order to solve the technical problems in the prior art where the limited number and location of monitoring points in traditional lighting systems result in deficiencies in dynamic environmental adaptability, efficiency, and personalized control, leading to poor performance of lighting systems in terms of energy utilization, user satisfaction, and operational flexibility, thereby affecting the overall lighting quality and economic benefits.

[0006] In view of the above problems, this application provides an intelligent control method for lighting driver power supply.

[0007] This application provides an intelligent control method for a lighting driver power supply, comprising: acquiring regional information of a target area, the regional information including regional structure information and the layout information of lighting devices within the area; constructing a digital twin model based on the regional information; evaluating the target area; distributing virtual test points according to the regional evaluation results; performing lighting fitting based on the digital twin model; monitoring the lighting effect through the virtual test points; establishing location point monitoring results; configuring a lighting evaluation network; monitoring and evaluating the location point monitoring results using the lighting evaluation network; generating target control results for each lighting device based on the monitoring and evaluation results; acquiring historical control information of the lighting devices; establishing a control prediction model based on the historical control information, the control prediction model being a prediction model for output control results based on power supply parameters; using the target control results as the control target; performing inverse fitting of power supply parameters through the control prediction model to generate calibrated power supply parameters; and driving control of the lighting devices based on the calibrated power supply parameters.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] By acquiring regional information of the target area, including regional structure information and the layout information of lighting devices within the area, a digital twin model is built based on the regional information. The target area is evaluated, and virtual test points are distributed according to the evaluation results. Lighting is fitted based on the digital twin model, and lighting effects are monitored through the virtual test points to establish location point monitoring results. A lighting evaluation network is configured to monitor and evaluate the location point monitoring results, and target control results for each lighting device are generated based on the monitoring and evaluation results. Historical control information of the lighting devices is acquired, and a control prediction model is established based on the historical control information. This control prediction model is a prediction model that outputs control results based on power parameters. Using the target control results as the control target, the power parameters are inversely fitted through the control prediction model to generate calibrated power parameters. The lighting devices are driven and controlled based on the calibrated power parameters. This effectively solves the technical goals of slow response, incomplete monitoring, and inaccurate control in lighting systems, achieving the technical effects of improving the adaptability of lighting systems, reducing energy consumption, and enhancing personalized settings.

[0010] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in this application 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 merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating an intelligent control method for a lighting driver power supply according to this application.

[0013] Figure 2 This is a flowchart illustrating the process of establishing a control prediction model based on historical control information in an intelligent control method for a lighting driver power supply according to this application. Detailed Implementation

[0014] This application provides an intelligent control method for lighting driver power supplies, solving the technical problems in existing technologies where the limited number and location of monitoring points in traditional lighting systems lead to deficiencies in dynamic environmental adaptability, efficiency, and personalized control. These deficiencies result in poor energy utilization, user satisfaction, and operational flexibility, ultimately affecting overall lighting quality and economic benefits. The method effectively addresses the technical goals of slow response, incomplete monitoring, and imprecise control in lighting systems, thereby improving system adaptability, reducing energy consumption, and enhancing personalized settings.

[0015] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0016] For examples, please refer to the appendix. Figure 1This application provides an intelligent control method for a lighting driver power supply, which specifically includes the following steps:

[0017] Step 1: Obtain the regional information of the target area, including regional structure information and the layout information of lighting devices within the area, and build a digital twin model based on the regional information.

[0018] Specifically, the target area is the area to be illuminated. Detailed information about the target area is collected, including its layout and the distribution of lighting fixtures. Then, a digital twin model is constructed using this area information. This digital twin is a virtual, digitized copy of the area that simulates the lighting conditions in the real environment, enabling intelligent control of the lighting drive power supply and improving the efficiency and comfort of the lighting system.

[0019] Step 2: Perform regional evaluation on the target area, distribute virtual test points according to the regional evaluation results, perform lighting fitting based on the digital twin model, monitor the lighting effect through the virtual test points, and establish location point monitoring results.

[0020] Specifically, after establishing a digital twin model, a regional evaluation is performed on the target area. This involves arranging virtual test points evenly in a grid, with each virtual test point equally spaced, to understand lighting needs and quality. Based on the regional evaluation results, a series of virtual test points are distributed within the digital twin model to simulate different locations in the actual environment. Next, the digital twin model is used to fit the lighting to the virtual test points, simulating the illumination effect of the lighting fixtures at these points. By monitoring the illumination effect at the virtual test points and collecting illumination data for each point, a comprehensive database of lighting effect monitoring results is established. This helps evaluate the performance of the lighting system and ensures that the lighting effect meets design requirements and standards.

[0021] Step 3: Configure a lighting evaluation network, use the lighting evaluation network to monitor and evaluate the location point monitoring results, and generate target control results for each lighting device based on the monitoring and evaluation results.

[0022] Specifically, to evaluate the performance of a lighting system, a lighting evaluation network is designed. This network comprehensively considers various indicators of lighting effect, monitors virtual test points set in a digital twin model, collects lighting data for each virtual test point, and evaluates the lighting effect to ensure that the lighting system meets predetermined performance standards. The monitoring and evaluation process analyzes various indicators in the lighting evaluation network, such as illuminance uniformity, color consistency, shadow effect, and visual comfort, to obtain comprehensive evaluation results. Based on the monitoring and evaluation results, a target control result is generated for each lighting device, guiding how the device should adjust its output to optimize the lighting effect. For example, if the illuminance uniformity index of a virtual test point is lower than the standard, the corresponding lighting device will be instructed to increase brightness or adjust the angle of the luminaire to improve illuminance uniformity. Similarly, if the color consistency index is poor, it will be suggested to adjust the color temperature to achieve more consistent color performance.

[0023] Step 4: Obtain historical control information of the lighting device, and establish a control prediction model based on the historical control information. The control prediction model is a prediction model that outputs control results based on power parameters.

[0024] Specifically, in the process of intelligent control of lighting systems, to improve control efficiency and accuracy, historical control information of the lighting devices is collected, including past brightness settings, operating times, and energy consumption data. A control prediction model is then established based on this historical control information to predict future output control results based on power supply parameters. Power supply parameters, including voltage, current, and power factor, directly affect the performance and energy consumption of the lighting devices. By analyzing the relationship between historical data and power supply parameters, the model predicts how the lighting devices should adjust their output under different power supply parameters to achieve the desired lighting effect. For example, if historical data shows that lighting demand is low during a specific time period, the control prediction model will suggest reducing the brightness of the lighting devices during that period to save energy. Similarly, if power supply parameters indicate that voltage fluctuations cause unstable lighting effects, corresponding compensatory controls are predicted and implemented to maintain a stable lighting level.

[0025] By establishing control prediction models, lighting systems can more intelligently predict and adapt to environmental changes, achieving efficient and precise lighting control.

[0026] Step 5: Using the target control result as the control target, perform inverse fitting of the power supply parameters through the control prediction model to generate calibrated power supply parameters.

[0027] Specifically, the target control result is used as the control objective. That is, the lighting installation generated by the lighting evaluation network is taken as the ideal state, and the power supply parameters are backfitted using a control prediction model. The power supply parameters of the lighting installation to achieve the target control result are predicted by analyzing the model. Backfitting is a reverse engineering process aimed at identifying which power supply parameter settings will make the actual performance of the lighting installation closest to the target control result. For example, if the target control result specifies a particular illuminance level, the voltage and current settings required to achieve that illuminance are predicted by the control prediction model. By generating calibrated power supply parameters, the lighting installation is precisely controlled. These calibrated power supply parameters can be used to adjust the output of the lighting installation to ensure that the actual lighting effect is consistent with the target control result, thereby optimizing lighting performance and meeting user needs.

[0028] Step 6: Drive and control the lighting device according to the calibrated power parameters.

[0029] Specifically, precise drive control of lighting fixtures is achieved using calibrated power supply parameters generated through a control prediction model. The power input to the lighting fixtures is adjusted based on these calibrated parameters to achieve precise management of brightness, color temperature, power consumption, and other key lighting characteristics. For example, if the calibrated power supply parameters indicate a need to increase brightness to achieve the target control result, the current supplied to the lighting fixtures is increased. Similarly, if a need to reduce power consumption is required, the voltage or current supply is reduced. Drive control of the lighting fixtures ensures that the lighting effect matches the design intent, while optimizing energy use and extending the lifespan of the lighting equipment.

[0030] The aforementioned intelligent control method for lighting drive power supply can effectively solve the technical goals of slow response, incomplete monitoring, and inaccurate control in lighting systems, thereby improving the adaptability of lighting systems, reducing energy consumption, and enhancing personalized settings.

[0031] Furthermore, this application also includes:

[0032] The configured lighting evaluation network is as follows: CEI = w1·U composite +w2·CCI+w3·S+W4·VCD; where CEI characterizes the lighting evaluation network, U composite The comprehensive illuminance uniformity index, E min E represents the minimum brightness value of the location point monitoring results. max The maximum brightness value of the location point monitoring results is represented by N, where N is the total number of virtual test points, i represents any virtual test point, and E i E represents the brightness value of any virtual test point. avg The average value of the location point monitoring results, ω a and ω bThe weights for illuminance uniformity index and standard deviation uniformity index are given by [variable name], and CCI is the color consistency index. CT i Let CT be the color temperature of the i-th virtual test point. avg denoted as the average color temperature of the location monitoring results, S as the shadow evaluation value, VCD as the visual comfort value, and w1, w2, w3, and w4 as the weighting coefficients of the comprehensive illuminance uniformity index, color consistency index, shadow evaluation value, and visual comfort value, respectively.

[0033] Specifically, the minimum brightness value E of the location point monitoring results min The maximum brightness value E of the location point monitoring results max The closer the ratio is to 1, the higher the overall illuminance uniformity index U. composite The larger the value, the more uniform the illumination; conversely, the smaller the value, the worse the uniformity of illumination.

[0034] Next, the brightness value E of any virtual test point i The average value E of the location monitoring results avg The smaller the difference, the better. The closer a value is to 1, the more uniform the illumination; conversely, the smaller the value, the worse the illuminance uniformity. The color temperature CT of the i-th virtual test point. i Average color temperature CT of location point monitoring results avg The smaller the difference, the better. The closer the value is to 0, the stronger the color temperature consistency, the more uniform the color temperature distribution, and the larger the color consistency index (CCI). Conversely, the closer the value is to 0, the worse the uniformity of the color temperature distribution.

[0035] Then, the shadow rating value S is obtained by visual observation or by calculating the ratio of the shadowed area to the entire area. Shadowed Area refers to the shaded area, while Total Area refers to the entire area.

[0036] Next, the visual comfort VCD was obtained by calculating the average flicker index and glare index for each virtual test point. Among these, Flicker Index avg The flicker index is the average flicker index of the location monitoring results. i Let be the flicker index of the i-th virtual test point. Glare Rating avg The average glare index (Glare Rating) is the result of location point monitoring. i Let VCD be the glare index of the i-th virtual test point. VCD = 0.5 * (1 - FlickerIndex) avg )+0.5·(1-Glare Rating avg).

[0037] Next, the average flicker index is used. avg and average glare rating avg Achieve visual comfort with VCD.

[0038] Next, the comprehensive illuminance uniformity index U... composite The Lighting Evaluation Network (CEI) is obtained by weighting the Color Uniformity Index (CCI), Shadow Evaluation Value (S), and Visual Comfort (VCD). The weighting coefficients for the Comprehensive Illumination Uniformity Index, Color Uniformity Index, Shadow Evaluation Value, and Visual Comfort are customized by those skilled in the art based on specific circumstances. For example, if a higher priority is desired for the Comprehensive Illumination Uniformity Index and Visual Comfort, they can be set to w1 = 0.3, w2 = 0.2, w3 = 0.2, and w4 = 0.3.

[0039] By systematically evaluating and comparing the overall performance of different lighting design schemes, quantitative basis is provided for the optimization of lighting systems.

[0040] Furthermore, this application also includes:

[0041] The control prediction model is as follows: L predicted，j (t+1)=f(V j (t), C j (t), F j (t)); L predicted，j (t+1) represents the control prediction model for the j-th lighting device, where t represents the time node, and V j (t) represents the input voltage of the j-th lighting device at time node t, C j (t) represents the input current of the j-th lighting device at time node t, F j (t) represents the input frequency parameter of the j-th lighting device at time node t; configure the loss function, and perform inverse fitting based on the prediction model through the loss function, as follows: Where J is the loss function, k is the summation index, ranging from 0 to M-1, and L... target，j (t+k) represents the target brightness control result of the j-th lighting device at a future time node t+k, L predicted，j (t+k) represents the predicted brightness of the j-th lighting device at the future time node t+k, where λ is the regularization parameter, (u j (t+k)) 2 The control input power parameters of the j-th lighting device at a future time node t+k are represented.

[0042] Specifically, the control prediction model L for the j-th lighting devicepredicted，j (t+1), where the value of the control prediction model for the j-th lighting device represents the predicted lighting intensity or brightness. This is determined by the input voltage V of the j-th lighting device at time node t. j (t), the input current Cj(t) of the j-th lighting device at time node t, and the input frequency parameter F of the j-th lighting device at time node t. j The predicted lighting intensity is obtained by calculating parameters such as (t).

[0043] Next, (L) target，j (t+k)-L predicted，j (t+k)) 2 The squared error term represents the square of the difference between the target and the prediction, used to quantify the prediction accuracy at each step. A smaller value indicates a smaller prediction error and control cost, i.e., a smaller loss; conversely, a larger value indicates a greater loss. By adjusting the regularization parameter λ, the influence of prediction error and control input is balanced, and the influence of the control input is adjusted to prevent over-control.

[0044] Next, the control input power parameters (u) of the j-th lighting device at future time node t+k. j (t+k)) 2 This is used to evaluate the magnitude of the control input to ensure that the system is not over-regulated, thereby avoiding potential instability or excessive energy consumption. j (t+k)) 2 The smaller, The smaller the value, the smaller the total amount of control activities during the entire prediction period, and the smaller the degree of loss; conversely, the larger the value, the greater the loss.

[0045] By dynamically adjusting the system to achieve optimal performance, and by minimizing the cost function, a set of optimized control strategies can be found.

[0046] Furthermore, this application also includes:

[0047] The power data of the lighting driver is read to establish a steady-state power identification; the control compensation optimization is performed based on the steady-state power identification and the calibrated power parameters; a dynamic response strategy is generated based on the compensation optimization result; and the control compensation of the identified power parameters is performed based on the dynamic response strategy.

[0048] Specifically, to ensure the lighting driver power supply can operate stably and effectively, power data, including parameters such as voltage, current, and power, is read. A steady-state indicator is then established based on this data to determine whether the power supply is in a stable state and whether it can meet the control requirements of the lighting device.

[0049] Next, based on the power supply steady-state indicator and the calculated calibrated power supply parameters, compensation optimization is performed to identify any deviations between the power supply parameters and the calibrated values, and to determine how to adjust these deviations to optimize lighting control. Compensation optimization is carried out based on factors such as the real-time state of the power supply, environmental changes, and the response characteristics of the lighting fixtures.

[0050] Based on the results of compensation optimization, a dynamic response strategy is generated. This strategy guides the lighting system on how to adjust power parameters to achieve optimal control based on current conditions and needs. The dynamic response strategy ensures that the lighting system maintains stable lighting performance even when faced with power fluctuations or external disturbances.

[0051] Finally, based on the dynamic response strategy, the power parameters of the indicator are controlled and compensated, which means that the power input of the lighting device is adjusted in real time to counteract any detected deviations and ensure that the lighting effect meets the preset target.

[0052] Through real-time control compensation, the lighting system can not only adapt to dynamic changes in power supply, but also provide a consistent and high-quality lighting experience.

[0053] Furthermore, this application also includes:

[0054] Set a first steady-state threshold and a second steady-state threshold; determine whether the power supply steady-state indicator meets the second steady-state threshold; if the power supply steady-state indicator fails to meet the second steady-state threshold, directly report an abnormality in the lighting driver power supply; if all the power supply steady-state indicators meet the second steady-state threshold, trigger discrimination of the power supply steady-state indicator is performed through the first steady-state threshold, and trigger discrimination result is generated; perform an abnormality evaluation of the lighting driver power supply based on the trigger discrimination result.

[0055] Specifically, to more accurately monitor and control the stability of the lighting driver power supply, a first steady-state threshold and a second steady-state threshold are set. The first and second steady-state thresholds are predefined standards used to determine whether the power supply is in a normal operating state.

[0056] Next, it is determined whether the power supply steady-state indicator meets the second steady-state threshold. The second steady-state threshold is usually set relatively leniently to detect any significant power supply anomalies. If the power supply steady-state indicator shows that the power supply status does not meet the second steady-state threshold, it indicates a serious power supply problem, and a lighting driver power supply anomaly will be reported directly. In this case, emergency measures should be taken, such as shutting off the lighting fixtures to prevent further damage or safety risks.

[0057] Next, if the power supply steady-state indicator meets the second steady-state threshold, a more refined power supply steady-state indicator trigger judgment is performed using the first steady-state threshold. The first steady-state threshold is typically set more stringently to identify minor power supply fluctuations or potential problems. Based on the trigger judgment results, an anomaly evaluation is performed on the lighting driver power supply to determine if any stability issues requiring attention exist.

[0058] By using a graded discrimination method, not only can serious power supply anomalies be identified and dealt with in a timely manner, but also potential stability problems can be warned in advance, thereby ensuring the reliability and safety of the lighting system.

[0059] Furthermore, this application also includes:

[0060] Obtain the trigger frequency, and generate a first abnormal effect based on the trigger frequency; obtain the trigger value, and generate a second abnormal effect based on the trigger value; establish a trigger discrimination result based on the first abnormal effect and the second abnormal effect.

[0061] Specifically, when monitoring the lighting driver power supply, the trigger frequency is recorded, which is the frequency at which the power supply steady-state indicator meets the first steady-state threshold, reflecting the frequency of power supply state fluctuations. Based on the trigger frequency, a first abnormal effect is generated, representing the potential impact of power supply state fluctuations on the stability of the lighting system.

[0062] Simultaneously, a trigger value is acquired, indicating the degree to which the power supply steady-state indicator exceeds the first steady-state threshold. The magnitude of the trigger value reflects the severity of power supply state fluctuations. Based on the trigger value, a second anomalous effect is generated, representing the direct impact of power supply state fluctuations on the lighting system performance.

[0063] Finally, by combining the first and second abnormal effects, a trigger discrimination result was established, which comprehensively considered the frequency and severity of power supply state fluctuations, providing a more comprehensive basis for the abnormal evaluation of lighting driver power supplies.

[0064] By using comprehensive evaluation methods, potential problems with lighting driver power supplies can be identified and addressed more accurately, ensuring the stability and reliability of lighting systems.

[0065] Furthermore, such as Figure 2 As shown, this application also includes:

[0066] The historical control information is subjected to control trust authentication to establish a control trust authentication identifier; the required data volume is configured, and the historical control information is eliminated according to the required data volume and the control trust authentication identifier to establish a retained dataset; features are extracted from the retained dataset, and a control prediction model is established based on the feature extraction results.

[0067] Specifically, to ensure the reliability and accuracy of historical control information for lighting installations, control trust certification is performed on this information. The control trust certification process verifies the source, integrity, and accuracy of the data, and establishes a control trust certification identifier for the verified data, ensuring that the data used for modeling is trustworthy.

[0068] Next, configure the required data volume to determine how much data is needed to build the control prediction model. Based on the configured required data volume and the control trust certification identifier, perform data eviction on historical control information, removing data that has not passed certification or is no longer needed, thereby creating a reserved dataset. The reserved dataset contains high-quality, reliable historical data for training the control prediction model.

[0069] Finally, feature extraction is performed on the retained dataset to identify and select the most important attributes or variables that best predict the control outcomes of the lighting devices. Based on the feature extraction results, a control prediction model is established, which can utilize key features from historical data to predict future lighting control needs, thereby improving the intelligent control level and efficiency of the lighting system.

[0070] Furthermore, this application also includes:

[0071] Obtain structural information of the target region and generate an initial evaluation result for the region based on the structural information; obtain regional importance information of the target region and generate an additional evaluation result based on the regional importance information; generate a regional evaluation result based on the initial evaluation result and the additional evaluation result.

[0072] Specifically, structural information about the target area is collected, including the layout of buildings, the size and shape of rooms, etc., and then the structural information is used to generate an initial evaluation result of the target area, reflecting the lighting needs and potential problems based on the area's structure.

[0073] Secondly, information on the importance of the target area is obtained, including pedestrian traffic, activity frequency, and functional requirements, to determine the area's criticality. Based on this information, supplementary evaluation results are generated, providing a further understanding and refinement of the area's lighting needs.

[0074] Finally, by combining the initial evaluation results and the supplementary evaluation results, a comprehensive regional evaluation result is generated. The supplementary evaluation results take into account both the structural characteristics of the region and its importance and functional requirements, providing comprehensive analysis and guidance for lighting design and control.

[0075] In summary, the intelligent control method for a lighting driver power supply provided in this application has the following technical effects:

[0076] By acquiring regional information of the target area, including regional structure information and the layout information of lighting devices within the area, a digital twin model is built based on the regional information. The target area is evaluated, and virtual test points are distributed according to the evaluation results. Lighting is fitted based on the digital twin model, and lighting effects are monitored through the virtual test points to establish location point monitoring results. A lighting evaluation network is configured to monitor and evaluate the location point monitoring results, and target control results for each lighting device are generated based on the monitoring and evaluation results. Historical control information of the lighting devices is acquired, and a control prediction model is established based on the historical control information. This control prediction model is a prediction model that outputs control results based on power parameters. Using the target control results as the control target, the power parameters are inversely fitted through the control prediction model to generate calibrated power parameters. The lighting devices are driven and controlled based on the calibrated power parameters. This effectively solves the technical goals of slow response, incomplete monitoring, and inaccurate control in lighting systems, achieving the technical effects of improving the adaptability of lighting systems, reducing energy consumption, and enhancing personalized settings.

[0077] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0078] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for intelligent control of a lighting driver power supply, characterized by, include: Obtain the regional information of the target area, including regional structure information and the layout information of lighting devices within the area, and build a digital twin model based on the regional information; A regional evaluation is performed on the target area, virtual test points are distributed according to the regional evaluation results, lighting is fitted based on the digital twin model, and lighting effect is monitored through the virtual test points to establish location point monitoring results. Configure a lighting evaluation network, use the lighting evaluation network to monitor and evaluate the location point monitoring results, and generate target control results for each lighting device based on the monitoring and evaluation results; The historical control information of the lighting device is obtained, and a control prediction model is established based on the historical control information. The control prediction model is a prediction model that outputs control results based on power parameters. Using the target control result as the control target, the power supply parameters are inversely fitted through the control prediction model to generate calibrated power supply parameters; Drive and control the lighting device according to the calibrated power parameters; The configured lighting evaluation network is as follows: ； in, Characterizing lighting evaluation network, The comprehensive illuminance uniformity index, , The minimum brightness value representing the location point monitoring results. The maximum brightness value representing the location point monitoring results. This represents the total number of virtual test points. Representing any virtual test point, Characterizes the brightness value of any virtual test point. The average value of the monitoring results at the location points. and The weights for the illuminance uniformity index and the standard deviation uniformity index are... The color consistency index. , For the first Color temperature of a virtual test point The average color temperature of the location monitoring results. This is the shadow rating value. For visual comfort, , , , These are the weighting coefficients for the comprehensive illuminance uniformity index, color consistency index, shadow evaluation value, and visual comfort, respectively.

2. The intelligent control method for a lighting driver power supply as described in claim 1, characterized in that, The step of using the target control result as the control target, and performing inverse fitting of the power supply parameters through the control prediction model to generate calibrated power supply parameters, further includes: The control prediction model is as follows: ； Characterizing the first A control prediction model for a lighting device Representing time nodes, Characterizing the first Each lighting device at the time point Input voltage, Characterizing the first Each lighting device at the time point The input current, Characterizing the first Each lighting device at the time point Input frequency parameters; Configure the loss function, and then perform inverse fitting based on the prediction model using the loss function, as follows: ； in, For loss function, For the summation index, the range is from 0 to... , Representing future time nodes The The target brightness control results for each lighting device Characterizing the first Future timeline of a lighting installation Predicted brightness, For regularization parameters, Representing future time nodes The The control input power parameters for each lighting device.

3. The intelligent control method for a lighting driver power supply as described in claim 1, characterized in that, Also includes: Read power data from the lighting driver power supply and establish a power steady-state indicator; The control compensation optimization is performed based on the power supply steady-state identifier and calibrated power supply parameters. A dynamic response strategy is generated based on the compensation optimization result, and control compensation is performed on the identified power supply parameters based on the dynamic response strategy.

4. The intelligent control method for a lighting driver power supply as described in claim 3, characterized in that, Also includes: Set a first steady-state threshold and a second steady-state threshold; Determine whether the power supply steady-state indicator meets the second steady-state threshold. If the power supply steady-state indicator fails to meet the second steady-state threshold, report an abnormality in the lighting driver power supply directly. If all the power supply steady-state indicators can meet the second steady-state threshold, then the triggering judgment of the power supply steady-state indicators is performed through the first steady-state threshold, and a triggering judgment result is generated; Anomalies in the lighting driver power supply are evaluated based on the trigger detection results.

5. The intelligent control method for a lighting driver power supply as described in claim 4, characterized in that, The step of triggering and determining the power supply steady-state identifier based on the first steady-state threshold and generating a triggering result further includes: Obtain the trigger frequency, and generate a first abnormal effect based on the trigger frequency; Obtain the trigger value, generate a second abnormal effect based on the trigger value, and establish a trigger discrimination result based on the first abnormal effect and the second abnormal effect.

6. The intelligent control method for a lighting driver power supply as described in claim 1, characterized in that, The step of acquiring historical control information of the lighting device and establishing a control prediction model based on the historical control information further includes: Perform control trust authentication on the historical control information and establish a control trust authentication identifier; Configure the required data volume, and based on the required data volume and the control trusted authentication identifier, perform data elimination of historical control information to establish a retained dataset; Feature extraction is performed on the retained dataset, and a control prediction model is established based on the feature extraction results.

7. The intelligent control method for a lighting driving power supply as described in claim 1, characterized in that, The step of performing a regional evaluation on the target area and distributing virtual test points according to the regional evaluation results also includes: Obtain structural information of the target region, and generate an initial evaluation result of the region based on the structural information; Obtain the regional importance information of the target area, and generate additional evaluation results based on the regional importance information; Regional evaluation results are generated based on the initial evaluation results and the additional evaluation results.

Images