Intelligent Lighting Automatic Adjustment Method and Device Based on Light Intensity and Personnel Detection

Abstract

The embodiment of the present application provides an intelligent lighting automatic adjustment method and device based on light intensity and personnel detection. By synchronously collecting ambient light intensity and personnel activity data, the data is evaluated for reliability and corrected for anomalies. An innovative lighting activity correlation model is constructed, and feature fusion is achieved using the fuzzy logic algorithm. The system establishes a complete evaluation index system including illuminance uniformity, color rendering index, and glare value, and performs constraint optimization based on the lighting scene adaptation model. By introducing a dynamic compensation mechanism of personnel detection probability value and time period priority factor, precise lighting adjustment control is achieved. This method breaks through the limitations of traditional fixed-mode dimming and provides a comprehensive technical solution for intelligent lighting systems.

Description

Technical Field

[0001] The present application relates to the field of data processing, and particularly to an intelligent lighting automatic adjustment method and device based on light intensity and personnel detection. Background Art

[0002] Traditional lighting control systems mainly rely on fixed-time adjustment or simple light-sensing control, and cannot perform intelligent adjustment according to the actual usage scenario. Existing automatic dimming systems often only consider a single factor of ambient light intensity and ignore the impact of personnel activity characteristics on lighting requirements. Especially in complex indoor environments, a control strategy with a single parameter is difficult to meet the lighting requirements in different scenarios.

[0003] At the same time, there are obvious deficiencies in the existing systems in terms of data processing and model construction. Traditional methods lack a reliability evaluation and outlier handling mechanism for multi-source data, and also fail to effectively utilize the correlation between light and personnel activities. The lighting effect evaluation of the system is also relatively simple, and key indicators such as illuminance uniformity, color rendering index, and glare value are not comprehensively considered.

[0004] In addition, the existing technology is also relatively mechanical in terms of light compensation calculation. There is a lack of a dynamic compensation mechanism based on personnel activity probability and time priority, making it difficult to achieve precise lighting adjustment control. The solution to these problems is of great significance for improving the adaptability and user comfort of intelligent lighting systems. Summary of the Invention

[0005] Aiming at the problems in the existing technology, the present application provides an intelligent lighting automatic adjustment method and device based on light intensity and personnel detection, which can achieve precise lighting adjustment control, break through the limitations of traditional fixed-mode dimming, and provide a comprehensive technical solution for intelligent lighting systems.

[0006] To solve at least one of the above problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides an intelligent lighting automatic adjustment method based on light intensity and personnel detection, including:

[0008] Collecting indoor environmental light intensity data through a light intensity sensor, collecting indoor personnel activity data through a personnel detection sensor, performing reliability evaluation and outlier correction on the light intensity data and the personnel activity data, performing smoothing processing on the corrected data using a moving average filtering algorithm, performing time series alignment on the smoothed data to obtain a multi-dimensional feature sequence, constructing a light-activity association model based on the multi-dimensional feature sequence, and performing feature fusion on the output of the light-activity association model using a fuzzy logic algorithm to generate a light scene feature data set;

[0009] Input the illumination scene feature dataset into a preset intelligent illumination adjustment model for training to obtain an illumination scene adaptation model. Construct an illumination effect evaluation index system including illuminance uniformity, color rendering index, and glare value. Based on the illumination effect evaluation index system, constrain and optimize the output result of the illumination scene adaptation model to obtain the target illumination intensity value and illumination device adjustment parameters that meet the illumination effect evaluation index system.

[0010] Input the target illumination intensity value and the illumination device adjustment parameters into an illumination controller. The illumination controller generates an illumination device control instruction based on an illumination compensation calculation formula, where the illumination compensation calculation formula is that the target illumination intensity value is equal to the sum of the current ambient illumination intensity and the illumination compensation value. The illumination compensation value is determined by the functional relationship between the personnel detection probability value and the time period priority factor. Send the illumination device control instruction to the illumination device execution unit through a control bus for illumination adjustment.

[0011] Further, for the reliability evaluation and outlier correction of the illumination intensity data and the personnel activity data, use a moving average filtering algorithm to smooth the corrected data, and perform time series alignment on the smoothed data to obtain a multi-dimensional feature sequence, including:

[0012] Calculate the probability density values of the illumination intensity data and the personnel activity data based on a Gaussian distribution model. Mark the data with probability density values lower than a preset threshold as abnormal data, and use linear interpolation to correct the abnormal data. Evaluate the reliability by calculating the mean, standard deviation, and coefficient of variation of the corrected data, and resample the data with a reliability evaluation result lower than the preset score.

[0013] Use a moving average filtering algorithm to filter the corrected illumination intensity data and personnel activity data respectively. Based on the timestamp information, interpolate and resample the filtered data according to a unified sampling period, and perform time series alignment processing on the resampled data sequence through the least squares method to obtain a multi-dimensional feature sequence.

[0014] Further, based on the multi-dimensional feature sequence, construct an illumination activity association model, and use a fuzzy logic algorithm to perform feature fusion on the output of the illumination activity association model to generate an illumination scene feature dataset, including:

[0015] Use the kernel density estimation method to calculate the joint probability distribution between the illumination intensity and the personnel activity data in the multi-dimensional feature sequence. Based on a Bayesian network, construct the conditional probability relationship of the joint probability distribution, determine the Bayesian network parameters through maximum likelihood estimation, and establish an illumination activity association model.

[0016] Perform fuzzy membership degree calculation on the conditional probability values output by the light activity association model, set the fuzzy rule set for light intensity and personnel activity, use the fuzzy inference mechanism to match and calculate the fuzzy rule set, and perform defuzzification processing on the inference results of the fuzzy inference mechanism through the centroid method to obtain the light scene feature data set.

[0017] Further, input the light scene feature data set into a preset light intelligent adjustment model for training to obtain a light scene adaptation model, and construct an illumination effect evaluation index system including illuminance uniformity, color rendering index, and glare value, including:

[0018] Construct the light intelligent adjustment model using a deep neural network, divide the light scene feature data set into a training set and a validation set according to a ratio of 8:2, use the backpropagation algorithm to iteratively train the light intelligent adjustment model, determine the optimal number of training rounds based on the model performance on the validation set, and obtain the light scene adaptation model;

[0019] Calculate the illuminance uniformity coefficient of the output result of the light scene adaptation model based on the CIE lighting standard, calculate the general color rendering index Ra value of the light source using the color rendering index measurement method, calculate the indoor lighting glare value using the UGR unified glare value evaluation system, and construct an illumination effect evaluation index system with the illuminance uniformity coefficient, the color rendering index Ra value, and the glare value.

[0020] Further, perform constraint optimization on the output result of the light scene adaptation model based on the illumination effect evaluation index system to obtain the target light intensity value and lighting device adjustment parameters that meet the illumination effect evaluation index system, including:

[0021] Set constraint conditions for the output result of the light scene adaptation model, including that the illuminance uniformity coefficient is not less than 0.7, the color rendering index Ra value is not less than 80, and the glare value does not exceed 19, construct a multi-objective optimization function based on the constraint conditions, and use the Lagrange multiplier method to solve the multi-objective optimization function to obtain the target light intensity value that meets the constraint conditions;

[0022] Construct a lighting device parameter optimization model based on the target light intensity value, use the power, color temperature, and irradiation angle of the lighting device as optimization variables, and use the particle swarm algorithm to solve the lighting device parameter optimization model to obtain the lighting device adjustment parameters that meet the illumination effect evaluation index system.

[0023] Further, input the target light intensity value and the lighting device adjustment parameters into the lighting controller. The lighting controller generates a lighting device control instruction based on a light compensation calculation formula, where the light compensation calculation formula is that the target light intensity value is equal to the sum of the current ambient light intensity and the light compensation value, and the light compensation value is determined by the functional relationship between the personnel detection probability value and the time period priority factor, including:

[0024] Input the target light intensity value and the lighting device adjustment parameters into the lighting controller. Divide a day into five time periods: early morning, morning, afternoon, evening, and night based on the time period division rule. Set a priority factor for each time period. Perform a weighted calculation on the personnel detection probability value and the time period priority factor to obtain the light compensation value. Generate a lighting device control instruction according to the calculation formula that the target light intensity value is equal to the sum of the current ambient light intensity and the light compensation value;

[0025] Use a communication protocol conversion module to convert the lighting device control instruction into the DALI communication protocol format, and send the converted control instruction to the lighting device execution unit through the data communication interface of the lighting controller. The lighting device execution unit receives and parses the control instruction and then performs the corresponding lighting adjustment operation.

[0026] Further, the step of sending the lighting device control instruction to the lighting device execution unit through the control bus for lighting adjustment includes:

[0027] Encapsulate the lighting device control instruction according to the DALI protocol specification, perform CRC checksum and address encoding on the encapsulated control instruction, and send the control instruction to the lighting device execution unit through the RS485 bus interface. The lighting device execution unit performs data frame checksum and address matching on the received control instruction;

[0028] Based on the driver chip of the lighting device execution unit, parse the control instruction into a PWM dimming signal and device operating parameters, control the output current of the LED drive circuit through the PWM dimming signal, and adjust the color temperature and irradiation angle of the lighting device according to the device operating parameters to achieve intelligent adjustment of the lighting device.

[0029] In a second aspect, the present application provides an intelligent lighting automatic adjustment device based on light intensity and personnel detection, including:

[0030] A feature fusion module, which is used to collect indoor environmental light intensity data through a light intensity sensor, collect indoor personnel activity data through a personnel detection sensor, conduct reliability evaluation and outlier correction on the light intensity data and the personnel activity data, perform smoothing processing on the corrected data by using a moving average filtering algorithm, perform time series alignment on the smoothed data to obtain a multi-dimensional feature sequence, construct a lighting activity association model based on the multi-dimensional feature sequence, and perform feature fusion on the output of the lighting activity association model by using a fuzzy logic algorithm to generate a lighting scene feature dataset;

[0031] A model evaluation module, which is used to input the lighting scene feature dataset into a preset lighting intelligent adjustment model for training to obtain a lighting scene adaptation model, construct an illumination effect evaluation index system including illuminance uniformity, color rendering index, and glare value, and perform constraint optimization on the output result of the lighting scene adaptation model based on the illumination effect evaluation index system to obtain a target light intensity value and lighting device adjustment parameters that meet the illumination effect evaluation index system;

[0032] An automatic adjustment module, which is used to input the target light intensity value and the lighting device adjustment parameters into a lighting controller, and the lighting controller generates a lighting device control instruction based on a lighting compensation calculation formula, where the lighting compensation calculation formula is that the target light intensity value is equal to the sum of the current ambient light intensity and the lighting compensation value, and the lighting compensation value is determined by the functional relationship between the personnel detection probability value and the time period priority factor, and the lighting device control instruction is sent to the lighting device execution unit through a control bus for lighting adjustment.

[0033] In a third aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the steps of the intelligent lighting automatic adjustment method based on light intensity and personnel detection are implemented.

[0034] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the intelligent lighting automatic adjustment method based on light intensity and personnel detection are implemented.

[0035] In a fifth aspect, the present application provides a computer program product, including a computer program / instructions, and when the computer program / instructions are executed by a processor, the steps of the intelligent lighting automatic adjustment method based on light intensity and personnel detection are implemented.

[0036] As can be seen from the above technical solutions, the present application provides an intelligent lighting automatic adjustment method and device based on light intensity and personnel detection. By synchronously collecting ambient light intensity and personnel activity data, the data is evaluated for reliability and corrected for anomalies. An innovative lighting-activity correlation model is constructed, and fuzzy logic algorithms are used to achieve feature fusion. The system establishes a complete evaluation index system including illuminance uniformity, color rendering index, and glare value, and performs constraint optimization based on the lighting scene adaptation model. By introducing a dynamic compensation mechanism for personnel detection probability values and time period priority factors, precise lighting adjustment control is achieved. This method breaks through the limitations of traditional fixed-mode dimming and provides a comprehensive technical solution for intelligent lighting systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0038] Figure 1 FIG. 1 is one of the schematic flowcharts of the intelligent lighting automatic adjustment method based on light intensity and personnel detection in the embodiments of the present application;

[0039] Figure 2 FIG. 2 is another schematic flowchart of the intelligent lighting automatic adjustment method based on light intensity and personnel detection in the embodiments of the present application;

[0040] Figure 3 FIG. 3 is yet another schematic flowchart of the intelligent lighting automatic adjustment method based on light intensity and personnel detection in the embodiments of the present application;

[0041] Figure 4 FIG. 4 is still another schematic flowchart of the intelligent lighting automatic adjustment method based on light intensity and personnel detection in the embodiments of the present application;

[0042] Figure 5 FIG. 5 is yet another schematic flowchart of the intelligent lighting automatic adjustment method based on light intensity and personnel detection in the embodiments of the present application;

[0043] Figure 6 FIG. 6 is still another schematic flowchart of the intelligent lighting automatic adjustment method based on light intensity and personnel detection in the embodiments of the present application;

[0044] Figure 7 FIG. 7 is yet another schematic flowchart of the intelligent lighting automatic adjustment method based on light intensity and personnel detection in the embodiments of the present application;

[0045] Figure 8Structural diagram of the intelligent lighting automatic adjustment device based on light intensity and personnel detection in the embodiments of the present application;

[0046] Figure 9 Structural schematic diagram of the electronic device in the embodiments of the present application.

[0047] Reference numerals:

[0048] Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. Detailed implementation manners

[0049] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the scope of protection of the present application.

[0050] In the technical solutions of the present application, the acquisition, storage, use, processing, etc. of data all comply with the relevant provisions of national laws and regulations.

[0051] Considering the problems existing in the prior art, the present application provides an intelligent lighting automatic adjustment method and device based on light intensity and personnel detection. By synchronously collecting ambient light intensity and personnel activity data, the data is evaluated for reliability and corrected for anomalies. An innovative lighting activity association model is constructed, and a fuzzy logic algorithm is used to achieve feature fusion. The system establishes a complete evaluation index system including illuminance uniformity, color rendering index and glare value, and performs constraint optimization based on the lighting scene adaptation model. By introducing a dynamic compensation mechanism of personnel detection probability value and time period priority factor, accurate lighting adjustment control is achieved. This method breaks through the limitations of traditional fixed-mode dimming and provides a comprehensive technical solution for intelligent lighting systems.

[0052] To be able to achieve accurate lighting adjustment control, break through the limitations of traditional fixed-mode dimming, and provide a comprehensive technical solution for intelligent lighting systems, the present application provides an embodiment of an intelligent lighting automatic adjustment method based on light intensity and personnel detection. Refer to Figure 1 The intelligent lighting automatic adjustment method based on light intensity and personnel detection specifically includes the following contents:

[0053] Step S101: Collect indoor environmental light intensity data through a light intensity sensor, collect indoor personnel activity data through a personnel detection sensor, conduct reliability assessment and outlier correction on the light intensity data and the personnel activity data, use a moving average filtering algorithm to smooth the corrected data, perform time series alignment on the smoothed data to obtain a multi-dimensional feature sequence, construct a light-activity association model based on the multi-dimensional feature sequence, and use a fuzzy logic algorithm to perform feature fusion on the output of the light-activity association model to generate a light scene feature data set;

[0054] Optionally, in this embodiment, a high-density sensor network layout scheme is first constructed in the intelligent office scenario. The light intensity sensor adopts a hierarchical and zonal arrangement strategy, with one measurement unit set every 20 square meters in the office area, and each unit contains three light sensor nodes at different heights (0.2m, 0.8m, 1.6m) to achieve three-dimensional monitoring of the light intensity at different working surfaces. The personnel detection sensor adopts a dual-verification mechanism, fusing and deploying a thermal imaging sensor and a millimeter-wave radar sensor, with 4 detection units set in each office area, and improving the recognition accuracy of personnel position and activity status through multi-sensor data fusion.

[0055] In the data collection link, the light intensity sensor samples continuously at a frequency of 1Hz, with a measurement range of 0 - 100000 lux and a resolution of 1 lux, ensuring that it can accurately capture the minute changes in natural light. The personnel detection sensor scans at a frequency of 2Hz, generating an activity data stream containing multi-dimensional features such as position coordinates, movement speed, and thermal maps. Considering the requirements for data transmission reliability in the office environment, a redundant transmission scheme combining a wired RS485 bus and a wireless Zigbee network is adopted to ensure the real-time and reliable data collection.

[0056] The data reliability assessment adopts a multi-level anomaly detection mechanism. First, establish a Gaussian probability distribution model for each sensor node at different time periods based on historical data, and calculate the probability density value of the real-time data. For the light intensity data, the system comprehensively considers factors such as weather conditions and time periods, and establishes a dynamic threshold judgment mechanism. For example, under cloudy weather conditions, the system will correspondingly adjust the tolerance range of light intensity fluctuations. For the personnel activity data, combine the functional division and time characteristics of the office area to establish an activity pattern library for identifying abnormal personnel aggregation or movement behaviors.

[0057] The abnormal data correction process adopts a context-aware intelligent interpolation algorithm. The system first analyzes the data change characteristics within 10 time windows around the abnormal data point, and extracts local trends and periodic characteristics. Then, based on the physical characteristics of light and the patterns of human activities, the optimal interpolation strategy is selected. For example, for anomalies caused by sudden changes in natural light, exponential smoothing interpolation is used; for the missing data of human activity data, it is supplemented based on trajectory prediction. The corrected data is evaluated for reliability through variance analysis and autocorrelation tests to ensure that the data quality meets the modeling requirements.

[0058] The data smoothing process innovatively introduces an adaptive sliding average filtering algorithm. This algorithm dynamically adjusts the size of the filtering window according to the data change rate, and the window range floats between 3 and 15 data points. When a rapid change in light intensity is detected, such as a sudden change in natural light caused by cloud movement, the system automatically reduces the filtering window to 3 - 5 points to maintain the ability to quickly respond to light changes. During periods of relatively stable light, the filtering window is expanded to 10 - 15 points to provide a smoother data output.

[0059] The time series alignment process adopts a multi-step precise alignment scheme. First, based on the timestamp information, the data from different sensors are uniformly interpolated to a standard sampling period of 100 ms. Then, through the cross-correlation analysis of the sliding time window, the system delays between different sensors are identified and compensated. Finally, the least squares method with constraints is used for alignment optimization to ensure the precise correspondence of the light intensity data and the human activity data in the time dimension.

[0060] In the link of light-activity association modeling, this embodiment develops a probability graph model based on kernel density estimation. First, the Gaussian kernel function is used to calculate the joint probability density of the light intensity and human activity data, and the kernel bandwidth is optimally selected through the cross-validation method. Then, a three-layer Bayesian network structure is constructed, including an environment layer, an activity layer, and a scene layer, to describe the conditional dependence relationship between variables. The model can accurately depict how to infer the current scene type based on the changes in the light environment and the state of human activities.

[0061] The feature fusion process adopts an improved fuzzy logic algorithm. An expert knowledge base containing 25 fuzzy rules is designed, covering combinations of scene features such as different light conditions, human density, and activity frequency. The fuzzy inference adopts the Mamdani inference mechanism, and the membership function is optimally designed according to the actual scene features. For example, for the lighting requirements in the work area, fuzzy linguistic variables such as "comfort level" and "concentration level" are designed to achieve an intelligent mapping from quantitative parameters to qualitative evaluations.

[0062] Through the implementation of this technical solution, this embodiment not only achieves accurate perception of the lighting status and personnel activities in the office environment, but more importantly, establishes an intelligent association mechanism between scene features and lighting requirements. This solution demonstrates excellent environmental adaptability and scene recognition ability in practical applications, providing an accurate decision-making basis for subsequent intelligent lighting control. Especially in the dynamic light environment adjustment of office places, this solution can accurately identify the lighting requirement characteristics of different regions and different time periods, providing strong support for achieving the balance between energy conservation and comfort.

[0063] Step S102: Input the lighting scene feature dataset into a preset lighting intelligent adjustment model for training to obtain a lighting scene adaptation model, construct an illumination effect evaluation index system including illuminance uniformity, color rendering index, and glare value, and perform constraint optimization on the output result of the lighting scene adaptation model based on the illumination effect evaluation index system to obtain the target lighting intensity value and lighting device adjustment parameters that meet the illumination effect evaluation index system;

[0064] Optionally, based on the lighting scene feature dataset, this embodiment constructs a lighting intelligent adjustment model with a deep learning architecture. This model adopts an encoder-decoder structure, where the encoder includes three layers of convolutional neural networks for extracting the spatio-temporal dependence relationship of lighting scene features; the decoder uses a bidirectional LSTM network to achieve serialized prediction of lighting adjustment strategies. The input features of the model include multi-dimensional features such as lighting intensity time series data, personnel activity density, and spatial location information, and the output includes the target lighting intensity and the dimming parameters of each lighting device.

[0065] During the model training process, this embodiment adopts a transfer learning strategy. First, it uses large-scale office scene data for pre-training, and then fine-tunes it for specific application scenarios. The training data is divided into a training set, a validation set, and a test set according to a ratio of 8:1:1. To improve the generalization ability of the model, data augmentation techniques are introduced during the training process, including methods such as random cropping of time series features and noise injection, so that the model can adapt to scene changes under different lighting environments.

[0066] For the evaluation of illumination effect, this embodiment establishes a multi-dimensional evaluation index system. The illuminance uniformity index is measured by calculating the ratio of the standard deviation to the average value of the illuminance at multiple measurement points on the working surface. Considering the lighting requirement differences in different functional areas, different uniformity thresholds are set. The color rendering index evaluation adopts the CIERa index standard, combined with the spectral characteristics of LED light sources, to ensure that the color rendering index meets the requirements of the office environment for color reproduction. The glare value evaluation is based on the UGR (Unified Glare Rating) calculation method, considering multiple factors such as the observer's position and background brightness, and quantifies the impact of glare on visual comfort by establishing a visual comfort model.

[0067] In the model output optimization stage, this embodiment innovatively proposes an optimization algorithm based on multi-objective constraints. First, an optimization objective function is established, and three indicators of illuminance uniformity, color rendering index, and glare value are weighted and combined. The weight coefficients are dynamically adjusted through expert experience and user feedback data to balance the lighting requirements in different scenarios. The optimization process uses an improved particle swarm algorithm, where the position of the particle represents the dimming parameters of each lighting device, and the optimal solution that meets the constraint conditions is found through iterative optimization.

[0068] To handle the deviation between the model output and the actual lighting effect, this embodiment designs a closed-loop feedback mechanism. The actual lighting effect indicators are calculated through real-time collected lighting data, compared with the model prediction results, and an error signal is generated. This error signal is used to fine-tune the model parameters online through the backpropagation algorithm to achieve the adaptive optimization of the model.

[0069] In practical applications, when the personnel distribution in the office area is detected to change, the lighting scene adaptation model can respond quickly and generate updated lighting parameters according to the new scene characteristics. For example, when a certain area enters the concentrated office state, the model will automatically adjust the lighting parameters of that area to improve the illuminance uniformity and reduce the glare value, creating a comfortable office environment for employees. At the same time, in areas with sufficient natural light, the model will appropriately reduce the output of artificial lighting to achieve a balance between energy conservation and comfort.

[0070] Through the implementation of this technical solution, this embodiment effectively solves the problems of the traditional lighting control system, such as lagging response to scene changes and unstable lighting effects. This solution can intelligently adjust the lighting parameters according to real-time scene characteristics, ensure that the lighting effect always meets the requirements of the office environment, and at the same time achieve energy conservation and consumption reduction of the lighting system. Especially in the open office areas with a dense population, this solution shows excellent scene adaptation ability and lighting effect stability, significantly improving the comfort and work efficiency of the office environment.

[0071] Step S103: Input the target light intensity value and the lighting device adjustment parameter into the lighting controller. The lighting controller generates a lighting device control instruction based on the light compensation calculation formula, where the light compensation calculation formula is that the target light intensity value is equal to the sum of the current ambient light intensity and the light compensation value, and the light compensation value is determined through the functional relationship between the personnel detection probability value and the time period priority factor. The lighting device control instruction is sent to the lighting device execution unit through the control bus for lighting adjustment.

[0072] Optionally, this embodiment designs a precise lighting compensation control mechanism after obtaining the target light intensity value and lighting device adjustment parameters. The lighting controller first dynamically calculates the required lighting compensation value based on real-time ambient light data, combined with occupant detection status and time characteristics. To accommodate the dynamic changes in natural light in office environments, this embodiment employs a regional, multi-level compensation strategy to ensure optimal lighting in each functional area.

[0073] To calculate illumination compensation, this embodiment innovatively incorporates a combined influence mechanism for human detection probability and time period priority. The human detection probability is calculated by fusing data from multiple sensor nodes and reflects the credibility of human activity in a specific area. The time period priority factor is divided according to the time characteristics of the office space, for example, giving the highest priority to weekdays from 9:00 AM to 11:30 AM and a lower priority to lunch breaks from 12:00 PM to 1:30 PM. These two factors are combined using a nonlinear mapping function to generate the final illumination compensation value.

[0074] This embodiment designs an adaptive lighting compensation calculation formula, the core concept of which is to achieve precise fill light control through dynamic weighting. When human activity is detected in a certain area, the compensation value calculation will prioritize the impact of the human detection probability value to ensure that the lighting effect meets actual needs. For example, as natural light gradually increases in the morning, if stable human activity is detected in a certain area, the lighting controller will calculate the required compensation light based on the difference between the target light intensity value and the current ambient light intensity, and accurately implement it through the dimming parameters of the LED lamps.

[0075] This embodiment employs a hierarchical dimming strategy during control command generation. The calculated illumination compensation value is first converted into dimming parameters for each lighting device. Taking into account the differences in light efficiency characteristics of different LED lamp types, a nonlinear dimming mapping relationship is established. To prevent frequent brightness fluctuations from disrupting office workers, a smooth dimming transition mechanism is implemented, using an interpolation algorithm to achieve a gradual dimming effect.

[0076] During the command issuance process, this embodiment implements a reliable communication mechanism based on the DALI bus protocol. Control commands are queued and processed according to priority. High-priority dimming commands (such as those requiring fill lighting when a person suddenly enters a dark area) can interrupt ongoing lower-priority commands. Furthermore, a command execution confirmation mechanism verifies the execution status of each control command through feedback signals, ensuring the reliability of lighting control.

[0077] To handle the compensation control in complex lighting environments, this embodiment also designs a scene prediction compensation mechanism. By analyzing the lighting change patterns and personnel activity patterns in historical data, a short-term lighting change prediction model is established. When a forthcoming lighting change (such as a cloud blocking the sun) is predicted, the compensation parameters are adjusted in advance to achieve a smoother lighting transition effect.

[0078] Through the implementation of this technical solution, this embodiment effectively solves problems such as the response lag and inaccurate light compensation of traditional lighting control systems. In an office environment, this solution can accurately calculate the required light compensation amount in real time according to the personnel activity status and time characteristics, ensuring that the working area always maintains a suitable lighting environment. Especially in the case of drastic changes in natural light, this solution shows excellent compensation control effects, significantly improving the lighting comfort and energy-saving effects of the office environment.

[0079] As can be seen from the above description, the intelligent lighting automatic adjustment method based on light intensity and personnel detection provided by the embodiments of this application can synchronously collect ambient light intensity and personnel activity data, perform reliability evaluation and anomaly correction on the data. Innovatively construct a light-activity association model and use a fuzzy logic algorithm to achieve feature fusion. The system establishes a complete evaluation index system including illuminance uniformity, color rendering index, and glare value, and performs constraint optimization based on the lighting scene adaptation model. By introducing a dynamic compensation mechanism of personnel detection probability value and time period priority factor, precise lighting adjustment control is achieved. This method breaks through the limitations of traditional fixed-mode dimming and provides a comprehensive technical solution for intelligent lighting systems.

[0080] In an embodiment of the intelligent lighting automatic adjustment method based on light intensity and personnel detection of this application, refer to Figure 2 , it may specifically include the following content:

[0081] Step S201: Calculate the probability density values of the light intensity data and the personnel activity data based on the Gaussian distribution model, mark the data with probability density values lower than the preset threshold as abnormal data, use the linear interpolation method to correct the abnormal data, perform reliability evaluation by calculating the mean, standard deviation, and coefficient of variation of the corrected data, and resample the data with reliability evaluation results lower than the preset score;

[0082] Step S202: Use the sliding average filtering algorithm to filter the corrected light intensity data and personnel activity data respectively, perform interpolation resampling on the filtered data according to a unified sampling period based on the timestamp information, and perform time series alignment processing on the resampled data sequence by the least squares method to obtain a multi-dimensional feature sequence.

[0083] Optionally, this embodiment designs a comprehensive data preprocessing solution to address possible outliers and noise in the raw light intensity data and human activity data. First, based on the statistical characteristics of the data, a Gaussian distribution model is used to detect anomalies. By calculating the probability density value of each data point and comparing it with an empirical threshold, anomalous data points that deviate from the normal distribution are identified.

[0084] This embodiment uses a context-based linear interpolation method to correct abnormal data. For data points marked as abnormal, a reasonable replacement value is calculated through linear interpolation, taking into account the temporal continuity of the data and using the preceding and following normal data points as a benchmark. For example, if the light intensity value at a certain point in time is abnormally high, the correction value that conforms to the scene's regularity is generated by analyzing the lighting trends at nearby time points and combining them with the actual lighting characteristics of the office space.

[0085] To ensure the reliability of data correction, this embodiment has designed a multi-dimensional evaluation mechanism. By calculating the statistical characteristics of the corrected data, including the mean, standard deviation, and coefficient of variation, the central tendency and dispersion of the data are comprehensively evaluated. When the evaluation score falls below the preset standard, the data resampling mechanism is activated. During the resampling process, data collection is prioritized for time periods with higher confidence levels to ensure that the acquired data accurately reflects the characteristics of the scene.

[0086] This embodiment uses an improved sliding average filtering algorithm for data smoothing. To adapt to the characteristics of different data types, different filter window sizes are set for light intensity data and human activity data. Because light intensity data is affected by natural light fluctuations, a larger filter window is used to preserve slowly changing trends; while a smaller filter window is used for human activity data to preserve sudden changes.

[0087] This embodiment innovatively designs a timestamp-based data resampling mechanism. Because the original sampling periods of light intensity and human activity data may differ, they need to be unified to the same time scale. By analyzing the temporal distribution characteristics of the data, an appropriate resampling period is selected, and piecewise linear interpolation is used to generate an equally spaced data sequence. This processing approach maintains the temporal continuity of the data while ensuring the correspondence between features of different dimensions.

[0088] To achieve precise alignment of multidimensional feature sequences, this embodiment uses a time series alignment algorithm based on the least squares method. First, feature correspondence within a time window is established. Then, the optimal alignment parameters are calculated by minimizing the sum of squares of temporal deviations. This method effectively addresses time delays and drift during data acquisition, ensuring precise alignment of features of different dimensions along the time axis.

[0089] Through the implementation of this technical solution, this embodiment effectively solves problems such as outlier interference, data discontinuity, and sampling inconsistency in the original data. In the intelligent lighting application in office premises, this solution can provide a high-quality data foundation, providing reliable support for subsequent scene feature analysis and lighting control strategy optimization. Especially in office areas with complex environmental conditions and frequent personnel activities, this solution demonstrates excellent data processing capabilities, significantly improving the stability and reliability of the lighting control system.

[0090] In an embodiment of the intelligent lighting automatic adjustment method based on light intensity and personnel detection in this application, refer to Figure 3 , and it may specifically include the following content:

[0091] Step S301: Calculate the joint probability distribution between the light intensity and personnel activity data in the multi-dimensional feature sequence by using the kernel density estimation method, construct the conditional probability relationship of the joint probability distribution based on the Bayesian network, determine the Bayesian network parameters through maximum likelihood estimation, and establish a light activity association model;

[0092] Step S302: Calculate the fuzzy membership degree of the conditional probability value output by the light activity association model, set the fuzzy rule set of light intensity and personnel activity, use the fuzzy inference mechanism to match and calculate the fuzzy rule set, and perform defuzzification processing on the inference result of the fuzzy inference mechanism through the centroid method to obtain the light scene feature data set.

[0093] Optionally, for the analysis of light intensity and personnel activity data in this embodiment, a joint distribution calculation method based on kernel density estimation is first designed. Specifically, a Gaussian kernel function is selected as the kernel function, and the kernel bandwidth is optimized and determined by the cross-validation method. For the light intensity x and personnel activity data y at any time point t, calculate their joint probability density value. In practical applications, for example, in office area A at 9 am, it is observed that the light intensity is 500 lux, the personnel density is 0.3 people per square meter, and the activity frequency is 20 times per hour. The joint probability of this set of data appearing is obtained through kernel density estimation, thereby quantitatively describing the typical degree of this scene feature.

[0094] To accurately describe the conditional probability relationship between data, this embodiment constructs a three-layer Bayesian network structure. The first layer is the environmental factor node, including time period, weather condition, etc.; the second layer is the light intensity node, including natural light intensity, artificial lighting intensity, etc.; the third layer is the personnel activity node, including personnel density, activity frequency, activity duration, etc. The edges between the nodes reflect the conditional dependence relationship between variables. For example, the personnel activity frequency node takes the light intensity node as the parent node, indicating the probability dependence of the activity frequency on the lighting conditions.

[0095] In this embodiment, the parameters of the Bayesian network are determined by the maximum likelihood estimation method. First, the conditional probability table of each node is calculated based on historical data. For example, within the time period from 9 am to 11 am, the probability of observing "high-frequency human activities" given the condition of "sufficient natural light" is calculated. To improve the accuracy of parameter estimation, the expectation-maximization algorithm is used to handle the possible missing values in the data, and the optimal parameter estimation value is obtained through multiple rounds of iterative optimization.

[0096] For the conditional probability values output by the Bayesian network, this embodiment designs an adaptive fuzzy membership function. The light intensity is divided into five fuzzy subsets according to the human visual perception characteristics: {very dark, slightly dark, moderate, slightly bright, very bright}, and the corresponding membership function uses a Gaussian function; the human activity characteristics are divided into three fuzzy subsets: {low-frequency activity, medium-frequency activity, high-frequency activity}, and the membership function uses a trapezoidal function. The function parameters are determined by analyzing the light-activity correspondence relationship in historical data.

[0097] This embodiment establishes a fuzzy rule set containing 25 rules, covering the light-activity association patterns in different scenarios. For example, Rule R1: "If the natural light intensity is slightly dark and the human activity frequency is high, then the scene characteristics tend to require supplementary lighting"; Rule R2: "If the natural light intensity is moderate and the human activity frequency is low, then the scene characteristics tend to maintain the status quo". The weight coefficients of the rules are determined by analyzing the trigger frequency and accuracy rate of the rules in historical data.

[0098] In the fuzzy reasoning process, this embodiment adopts the Mamdani reasoning mechanism. First, the fuzzification process is carried out to convert the input conditional probability values into fuzzy sets; then, the reasoning is carried out based on the fuzzy rules, and the minimum-maximum composition method is used to calculate the activation intensity of the rules; finally, the reasoning results of multiple rules are weighted and synthesized. For example, when the natural light intensity of 450 lux and the human activity frequency of 15 times per hour are detected at a certain moment, the corresponding scene characteristic tendency is obtained through fuzzy reasoning.

[0099] This embodiment innovatively improves the centroid defuzzification algorithm. The traditional centroid method may produce deviations in the case of multiple peaks. The improved algorithm introduces a local weight factor to weight the centroid values in different regions. Specifically, first, the output fuzzy set is divided into several sub-intervals, the local centroid and weight coefficients of each sub-interval are calculated, and then the final defuzzification result is obtained through weighted averaging.

[0100] Through the implementation of this technical solution, this embodiment achieves accurate scene feature recognition in the office environment. Traditional methods often overlook the dynamic correlation between lighting conditions and human activities, resulting in scene recognition results that do not match the actual requirements. By combining probability statistics and fuzzy logic organically, this embodiment can not only accurately capture the statistical laws among data, but also simulate the decision-making thinking of human experts, realizing the intelligent recognition of complex scene features. Especially in open office areas with variable lighting conditions and frequent human activities, this solution exhibits excellent adaptability and robustness.

[0101] Another significant advantage of this embodiment is its scene prediction ability. By analyzing the lighting-activity patterns in historical data and combining the probability distribution characteristics of the current time period, it can predict the trend of scene changes in the future. This prediction ability provides a decision-making basis for active lighting control, effectively improving the response speed and service quality of the lighting system. For example, when detecting an increasing trend in the frequency of human activities, the system can adjust the lighting parameters in advance to avoid the lag of traditional passive control.

[0102] In an embodiment of the intelligent lighting automatic adjustment method based on light intensity and human detection in this application, referring to Figure 4 , it may specifically include the following content:

[0103] Step S401: Construct the intelligent lighting adjustment model using a deep neural network. Divide the lighting scene feature dataset into a training set and a validation set in a ratio of 8:2. Use the backpropagation algorithm to iteratively train the intelligent lighting adjustment model, and determine the optimal number of training epochs based on the model performance on the validation set to obtain the lighting scene adaptation model;

[0104] Step S402: Calculate the illuminance uniformity coefficient of the output result of the lighting scene adaptation model based on the CIE lighting standard, calculate the general color rendering index Ra value of the light source using the color rendering index measurement method, and calculate the indoor lighting glare value using the UGR unified glare value evaluation system. Construct an illumination effect evaluation index system with the illuminance uniformity coefficient, the color rendering index Ra value, and the glare value.

[0105] Optionally, this embodiment designs an intelligent lighting adjustment scheme based on deep learning to achieve adaptive adjustment of lighting parameters for different scenes by constructing a multi-layer neural network. This neural network adopts a five-layer structure, including an input layer, three hidden layers, and an output layer. The input layer receives the feature parameters in the lighting scene feature dataset, including light intensity distribution, human activity characteristics, time characteristics, etc.; the output layer generates corresponding lighting parameter adjustment strategies, including the on / off state of lighting devices, dimming levels, etc.

[0106] The hidden layers use an improved ReLU activation function, introducing a LeakyReLU mechanism to avoid the "neuron death" problem that can occur with traditional ReLU. A residual connection module is also incorporated into the network structure, effectively alleviating the vanishing gradient problem during deep network training. A Dropout layer is added after each hidden layer to prevent overfitting and improve generalization.

[0107] In this example, the collected illumination scene feature dataset was divided into a training set and a validation set in an 8:2 ratio. During data preprocessing, the feature data was normalized to eliminate the influence of dimensional differences between features. Model training was performed using a mini-batch gradient descent method with a batch size of 32 and an adaptive learning rate strategy with an initial value of 0.001.

[0108] During model training, an improved backpropagation algorithm is used. The loss function uses a combination of mean squared error and cross entropy, which considers both the accuracy of continuous value predictions and the effectiveness of classification decisions. To improve training efficiency, a momentum term and a learning rate decay mechanism are introduced. Model performance is evaluated on a validation set, and an early stopping strategy is used to determine the optimal number of training rounds, effectively avoiding overfitting.

[0109] Based on the trained illumination scene adaptation model, this embodiment establishes a complete lighting effect evaluation system. First, the illumination uniformity coefficient is calculated according to the CIE lighting standard. Multiple illumination detection points are set up within the office area, and the illumination uniformity is evaluated by the ratio of the minimum illumination value to the average illumination value. For example, in an open office area, the illumination uniformity coefficient on the work surface must meet a requirement of no less than 0.7.

[0110] This example innovatively applies color rendering index measurement methods to lighting effect evaluation. By measuring the color rendering performance of a light source for eight standard color samples, a general color rendering index (Ra) value is calculated. In practical applications, differentiated color rendering index requirements are set for different office scenarios. For example, an Ra value greater than 85 is required for high-precision visual work areas, while an Ra value greater than 80 is required for general office areas.

[0111] For glare control, this embodiment uses the UGR unified glare value evaluation system. It calculates the glare value of indoor lighting by analyzing factors such as the brightness distribution of the light source, background brightness, and observer position. The calculation takes into account the varying viewing angles at multiple viewing positions within the office space, selecting the glare value under the most unfavorable conditions as the evaluation metric.

[0112] Verified through practical applications, the intelligent lighting adjustment solution of this embodiment has significantly improved the lighting quality of the office environment. The deep learning model can accurately identify the lighting demand characteristics of different scenarios and adaptively adjust the lighting parameters to keep the lighting effect in the best state all the time. The complete lighting effect evaluation system provides a reliable feedback mechanism for lighting control to ensure that the adjustment results meet the standard requirements.

[0113] This embodiment is particularly suitable for the lighting control requirements of modern intelligent office buildings. Through the dynamic learning ability of the lighting scene adaptation model, it can continuously optimize the lighting strategy to adapt to various changes in the office environment. The establishment of the evaluation index system provides a scientific basis for the quantitative evaluation of the lighting effect, effectively improving the accuracy and reliability of lighting control.

[0114] In an embodiment of the intelligent lighting automatic adjustment method based on light intensity and personnel detection of this application, refer to Figure 5 , it may specifically include the following content:

[0115] Step S501: Set constraint conditions for the output result of the lighting scene adaptation model, including that the illuminance uniformity coefficient is not less than 0.7, the color rendering index Ra value is not less than 80, and the glare value does not exceed 19. Construct a multi-objective optimization function based on the constraint conditions, and use the Lagrange multiplier method to solve the multi-objective optimization function to obtain the target light intensity value that meets the constraint conditions;

[0116] Step S502: Construct an optimization model for lighting equipment parameters based on the target light intensity value. Take the power, color temperature, and irradiation angle of the lighting equipment as optimization variables, and use the particle swarm algorithm to solve the optimization model for lighting equipment parameters to obtain the lighting equipment adjustment parameters that meet the lighting effect evaluation index system.

[0117] Optionally, this embodiment designs an intelligent lighting control scheme based on multi-objective optimization to achieve precise adjustment of lighting parameters through strict constraint conditions and optimization algorithms. First, set constraint conditions that meet international lighting standards for the output result of the lighting scene adaptation model: the illuminance uniformity coefficient should be not less than 0.7 to ensure the uniformity of lighting distribution in the office area; the color rendering index Ra value should be not less than 80 to ensure the true restoration of object colors in the office environment; the glare value should not exceed 19 to avoid visual discomfort caused by improper lighting.

[0118] Based on the constraint conditions, a multi-objective optimization function is constructed in this embodiment. This function comprehensively considers two main objectives: lighting effect and energy efficiency. On the one hand, it pursues the best visual comfort, and on the other hand, it reduces energy consumption. Weight coefficients are introduced into the optimization function to dynamically adjust the priorities of each objective according to the demand characteristics of different office scenarios. For example, in areas where high-precision visual work is required, the weight of the lighting effect is increased; in ordinary office areas, the weight of the energy-saving objective is appropriately increased.

[0119] This embodiment uses an improved Lagrange multiplier method to solve the multi-objective optimization problem. The traditional Lagrange method may fall into a local optimum when dealing with non-convex optimization problems. The improved algorithm improves the stability and global convergence of the solution by introducing a penalty term and an adaptive step-size adjustment mechanism. Specifically, the original constrained optimization problem is transformed into an unconstrained problem, and the Lagrange multiplier and penalty factor are iteratively updated to gradually approach the optimal solution.

[0120] During the solution process, this embodiment particularly considers the dynamic change characteristics of the lighting requirements in the office environment. By real-time monitoring the change trend of natural light intensity, the parameter weights in the optimization objective function are dynamically adjusted. For example, during periods of sufficient natural light, the target intensity value of artificial lighting is reduced; during periods of insufficient natural light, the compensation intensity of artificial lighting is correspondingly increased. This dynamic optimization mechanism ensures that the lighting effect always remains in the best state.

[0121] Based on the optimized target lighting intensity value, this embodiment further constructs an optimization model for lighting device parameters. This model takes the power, color temperature, irradiation angle, etc. of lighting devices as optimization variables and establishes the mapping relationship between these parameters and the lighting effect. In particular, considering the non-linear dimming characteristics of LED lighting devices, a correction function is introduced into the model to ensure the accuracy of parameter adjustment.

[0122] During the optimization process of lighting device parameters, this embodiment uses an improved particle swarm algorithm for solution. The traditional particle swarm algorithm is prone to premature convergence problems. The improved algorithm improves the search efficiency and the quality of the solution through dynamic inertia weight and adaptive learning factors. The particle positions in the algorithm represent the parameter combinations of lighting devices. By evaluating the lighting effects generated by these parameter combinations, the positions and velocities of the particles are continuously updated, and finally the optimal parameter configuration is found.

[0123] To improve the optimization effect, this embodiment adds a local search strategy to the particle swarm algorithm. When the particle swarm enters a predefined potential optimal region, the local search mechanism is triggered to further optimize the parameter values through refined search. This hybrid optimization strategy not only ensures the global search ability but also improves the local optimization accuracy.

[0124] In practical applications, the optimization solution of this embodiment demonstrates excellent adaptability. For example, in an open office area, when the illuminance uniformity of a local area is detected to be insufficient, the optimization algorithm can quickly adjust the parameters of relevant lighting devices. By changing the power distribution and irradiation angle, the balance of illuminance distribution is achieved. At the same time, through the intelligent adjustment of color temperature, a high color rendering index is maintained, ensuring the visual comfort of the office environment.

[0125] The innovation of this embodiment is also reflected in the collaborative optimization of lighting parameters. By establishing a coupling relationship model between lighting device parameters, the mutual influence of parameter adjustment is considered during the optimization process. For example, the adjustment of power will affect the actual performance of color temperature, and the change of irradiation angle will affect illuminance uniformity. Through collaborative optimization, it is ensured that the adjustment results of various parameters can produce the best comprehensive effect.

[0126] Through the implementation of this technical solution, this embodiment realizes the intelligent and precise control of office environment lighting. The methods of multi-objective optimization and parameter collaborative adjustment not only ensure that the lighting effect meets the standard requirements but also achieve the improvement of energy utilization efficiency. Especially in a complex open office environment, this solution can quickly respond to scene changes and maintain the stability and comfort of the lighting effect.

[0127] In an embodiment of the intelligent lighting automatic adjustment method based on light intensity and personnel detection of this application, refer to Figure 6 , it can also specifically include the following content:

[0128] Step S601: Input the target light intensity value and the lighting device adjustment parameters into the lighting controller. Divide a day into five time periods: early morning, morning, afternoon, evening, and night based on the time period division rule. Set a priority factor for each time period. Perform a weighted calculation on the personnel detection probability value and the time period priority factor to obtain a light compensation value. Generate a lighting device control instruction according to the calculation formula that the target light intensity value is equal to the sum of the current ambient light intensity and the light compensation value;

[0129] Step S602: Use a communication protocol conversion module to convert the lighting device control instruction into the DALI communication protocol format. Send the converted control instruction to the lighting device execution unit through the data communication interface of the lighting controller. After receiving and parsing the control instruction, the lighting device execution unit performs corresponding lighting adjustment operations.

[0130] Optionally, this embodiment designs an intelligent lighting control execution scheme. Through time period division and analysis of personnel activity characteristics, precise adjustment of lighting parameters and reliable execution of control instructions are achieved. First, input the optimized target light intensity value and lighting device adjustment parameters into the lighting controller to establish a lighting compensation mechanism based on time characteristics.

[0131] In this embodiment, an innovative method for dividing time periods based on physiological rhythms is proposed. A day is divided into five key time periods: early morning (5:00 - 8:00), morning (8:00 - 12:00), afternoon (12:00 - 17:00), evening (17:00 - 20:00), and night (20:00 - 5:00). This division fully considers the human physiological rhythm and the law of natural light change, providing differential support for lighting requirements in different time periods.

[0132] In terms of setting time period priorities, this embodiment assigns different priority factors to each time period according to the characteristics of human visual needs and the law of work efficiency. For example, during the morning time period, the concentration of personnel at work is relatively high, and a larger priority factor is set to ensure sufficient lighting support; during the evening time period, considering the gradual weakening of natural light, the priority factor is appropriately increased to compensate for insufficient lighting; during the night time period, the priority factor is reduced to avoid disturbing the human physiological rhythm due to excessive lighting.

[0133] When calculating the light compensation in this embodiment, the personnel detection probability value and the time period priority factor are innovatively weighted and fused. The personnel detection probability value reflects the real-time state of space use. By using a deep learning algorithm to extract the characteristics of personnel activities from video surveillance data, the probability of personnel presence in the current area is calculated. For example, in a meeting room scenario, when multiple people are continuously detected to be present, the personnel detection probability value is increased, and the light compensation intensity is correspondingly increased.

[0134] The calculation of the light compensation value adopts an adaptive weight method. According to the product of the personnel detection probability value and the time period priority factor, combined with a preset reference compensation coefficient, the final compensation value is obtained. This calculation method ensures the flexibility and rationality of lighting adjustment. In the actual application of this embodiment, through the calculation formula that the target light intensity value is equal to the sum of the current ambient light intensity and the light compensation value, the dynamic balance of lighting intensity is achieved.

[0135] In the control instruction generation link, this embodiment establishes a complete instruction mapping system. The optimized lighting device parameters (power, color temperature, irradiation angle, etc.) are converted into a standardized control instruction format, and at the same time, a time stamp and a priority identifier are added to ensure the timing and reliability of instruction execution.

[0136] To achieve precise control of lighting devices, this embodiment uses a communication protocol conversion module to convert the control instruction into the DALI (Digital Addressable Lighting Interface) communication protocol format. As an internationally common intelligent lighting control protocol, the DALI protocol supports two-way communication and multi-level dimming, and can meet the lighting control requirements in complex scenarios.

[0137] During the protocol conversion process, this embodiment particularly considers the integrity and reliability of instructions. By adding a check bit and a response mechanism, the accuracy of instruction transmission is ensured. At the same time, according to the characteristics of the DALI protocol, the control parameters are mapped to the corresponding function codes and address codes to achieve accurate device addressing and parameter setting.

[0138] In the design of the data communication interface of the lighting controller in this embodiment, multiplexing technology is adopted to support the parallel control of multiple lighting circuits. By establishing a priority queue, it is ensured that high-priority instructions can be executed in a timely manner, while maintaining the responsiveness to low-priority instructions. For example, lighting adjustment instructions in case of emergency will obtain the highest priority to ensure a quick response.

[0139] The lighting device execution unit adopts a modular design, including an instruction parsing module, a parameter verification module, and an execution control module. The instruction parsing module is responsible for identifying control instructions in the DALI protocol format and extracting relevant parameters; the parameter verification module ensures that the parsed parameters are within the range supported by the device; the execution control module is responsible for converting the verified parameters into specific hardware control signals.

[0140] Through the implementation of this technical solution, this embodiment realizes the intelligent and precise control of office environment lighting. The dynamic compensation mechanism based on time characteristics and personnel activity characteristics ensures that the lighting effect always meets the usage requirements. The standardized communication protocol and reliable control mechanism guarantee the accurate execution of control instructions. Especially in a complex office environment, this solution can quickly respond to scene changes and maintain the stability and comfort of the lighting effect.

[0141] In an embodiment of the intelligent lighting automatic adjustment method based on light intensity and personnel detection in this application, refer to Figure 7 , and it may specifically include the following content:

[0142] Step S70: Package the lighting device control instructions according to the DALI protocol specification, perform CRC checksum and address encoding on the packaged control instructions, and send the control instructions to the lighting device execution unit through the RS485 bus interface. The lighting device execution unit performs data frame checksum and address matching on the received control instructions;

[0143] Step S702: Based on the driver chip of the lighting device execution unit, parse the control instructions into PWM dimming signals and device working parameters, control the output current of the LED drive circuit through the PWM dimming signals, and adjust the color temperature and irradiation angle of the lighting device according to the device working parameters to achieve the intelligent adjustment of the lighting device.

[0144] Optionally, this embodiment proposes an illumination control instruction execution scheme based on the DALI protocol, which ensures the reliable transmission and precise execution of control instructions through a strict data encapsulation and verification mechanism. In the data encapsulation process, the control instructions are encapsulated into a standard data frame format according to the DALI protocol specifications, including a start bit, an address bit, a command bit, a data bit, and an end bit, realizing the structured organization of control information.

[0145] This embodiment pays special attention to the integrity protection of control instructions during the data encapsulation process. Through the CRC (Cyclic Redundancy Check) algorithm, a checksum is calculated for the encapsulated data frame, and the check code is appended to the end of the data frame. This verification mechanism can effectively detect data errors during transmission, ensuring the accuracy of control instructions. For example, when the lighting equipment in a certain meeting room needs to be adjusted, the control instruction can only be sent after passing the CRC verification, avoiding abnormal lighting effects caused by incorrect instructions.

[0146] In terms of address coding, this embodiment adopts a hierarchical coding strategy, encoding the physical location information and functional attributes of lighting equipment into the address field. Specifically, the address coding includes three levels: area identification, device type, and device serial number, enabling the control system to accurately locate each lighting equipment. For example, in a large open office area, independent control of lighting equipment in different areas and of different types can be achieved through this coding method.

[0147] This embodiment selects the RS485 bus as the data transmission medium, considering its reliability and anti-interference ability in the industrial control field. By adopting a differential signal transmission method, the influence of electromagnetic interference on control signals is effectively reduced. In the design of the bus topology structure, a daisy-chain connection method is adopted, which not only ensures the reliable transmission of signals but also facilitates the expansion and maintenance of the system.

[0148] After receiving the control instruction, the lighting equipment execution unit first performs data frame verification to verify the integrity of the frame format and the correctness of the checksum. Subsequently, address matching is performed. Only when the address in the instruction matches the device's own address will subsequent control operations be executed. This dual verification mechanism effectively prevents the execution of incorrect instructions.

[0149] This embodiment adopts intelligent parsing technology in the design of the driver chip, which can accurately convert control instructions in DALI protocol format into PWM dimming signals and device operating parameters. The driver chip is built-in with a microprocessor, and the mapping from instructions to control signals is realized through a look-up table algorithm, ensuring control accuracy. For example, when adjusting the lighting brightness, the brightness value in the control instruction will be converted into the corresponding PWM duty cycle.

[0150] In terms of LED drive circuit control, this embodiment adopts a high-precision constant current source design to accurately control the output current through the PWM dimming signal. The drive circuit adopts a Buck-Boost topology structure, which can stably adjust the output current within a wide range and achieve linear adjustment of the lighting brightness. At the same time, through the temperature compensation circuit, it ensures stable output characteristics under different temperature conditions.

[0151] In terms of color temperature adjustment, this embodiment adopts a dual-color temperature LED combination design to achieve continuously adjustable color temperature by adjusting the relative brightness of the cold color temperature and warm color temperature LEDs. The control algorithm takes into account the perception characteristics of the human eye to color temperature changes and adopts a non-linear mapping relationship to make the color temperature adjustment more in line with the visual experience. For example, in the morning, a bright and refreshing lighting environment can be created by increasing the color temperature.

[0152] In terms of irradiation angle adjustment, this embodiment designs an electric focusing mechanism to precisely adjust the irradiation angle by controlling the position of the reflector or lens group through a stepper motor. The control algorithm adopts a position feedback design to ensure the accuracy and repeatability of the angle adjustment. For example, in a projection demonstration scenario, the range of the lighting area can be adjusted according to needs.

[0153] Through the implementation of this technical solution, this embodiment realizes high reliability and high precision in the control of lighting equipment. The data transmission mechanism based on the DALI protocol ensures the reliable transmission of control instructions; the design of the intelligent drive chip and the precision drive circuit ensures the accurate execution of control instructions; the realization of multi-parameter coordinated adjustment meets the lighting requirements in different scenarios. Especially in a complex office environment, this solution can achieve precise control and intelligent adjustment of the lighting effect, improving the usage experience and energy efficiency of the lighting system.

[0154] This solution shows excellent adaptability in practical applications. For example, in a conference room scenario where frequent adjustment is required, the reliable transmission and quick response of control instructions ensure timely adjustment of the lighting effect; in an office area where stable lighting needs to be maintained, precise parameter control ensures the continuous comfort of the lighting environment. Through this intelligent control method, the efficient operation and energy saving of the lighting system are realized.

[0155] In order to achieve precise lighting adjustment control, break through the limitations of traditional fixed-mode dimming, and provide a comprehensive technical solution for intelligent lighting systems, this application provides an embodiment of an intelligent lighting automatic adjustment device based on light intensity and personnel detection for implementing all or part of the content of the intelligent lighting automatic adjustment method based on light intensity and personnel detection, see Figure 8 The intelligent lighting automatic adjustment device based on light intensity and personnel detection specifically includes the following content:

[0156] The feature fusion module 10 is used to collect indoor environmental light intensity data through a light intensity sensor, collect indoor personnel activity data through a personnel detection sensor, perform reliability evaluation and outlier correction on the light intensity data and the personnel activity data, smooth the corrected data using a moving average filtering algorithm, perform time series alignment on the smoothed data to obtain a multi-dimensional feature sequence, construct a light-activity association model based on the multi-dimensional feature sequence, and perform feature fusion on the output of the light-activity association model using a fuzzy logic algorithm to generate a light scene feature dataset;

[0157] The model evaluation module 20 is used to input the light scene feature dataset into a preset intelligent lighting adjustment model for training to obtain a light scene adaptation model, construct an illumination effect evaluation index system including illuminance uniformity, color rendering index, and glare value, and perform constraint optimization on the output result of the light scene adaptation model based on the illumination effect evaluation index system to obtain a target light intensity value and lighting device adjustment parameters that meet the illumination effect evaluation index system;

[0158] The automatic adjustment module 30 is used to input the target light intensity value and the lighting device adjustment parameters into a lighting controller. The lighting controller generates a lighting device control instruction based on a light compensation calculation formula, where the light compensation calculation formula is that the target light intensity value is equal to the sum of the current environmental light intensity and the light compensation value, and the light compensation value is determined by the functional relationship between the personnel detection probability value and the time period priority factor. The lighting device control instruction is sent to the lighting device execution unit through a control bus for lighting adjustment.

[0159] As can be seen from the above description, the intelligent lighting automatic adjustment device based on light intensity and personnel detection provided by the embodiments of the present application can synchronously collect environmental light intensity and personnel activity data, perform reliability evaluation and anomaly correction on the data. Innovatively construct a light-activity association model and use a fuzzy logic algorithm to achieve feature fusion. The system establishes a complete evaluation index system including illuminance uniformity, color rendering index, and glare value, and performs constraint optimization based on the light scene adaptation model. By introducing a dynamic compensation mechanism of personnel detection probability value and time period priority factor, precise lighting adjustment control is achieved. This method breaks through the limitations of traditional fixed-mode dimming and provides a comprehensive technical solution for intelligent lighting systems.

[0160] From a hardware level, in order to achieve precise lighting adjustment control, break through the limitations of traditional fixed-mode dimming, and provide a comprehensive technical solution for intelligent lighting systems, the present application provides an embodiment of an electronic device for implementing all or part of the content of the intelligent lighting automatic adjustment method based on light intensity and personnel detection. The electronic device specifically includes the following content:

[0161] A processor, a memory, a communications interface, and a bus; wherein, the processor, the memory, and the communications interface complete communication with each other through the bus; the communications interface is used to implement information transmission between the intelligent lighting automatic adjustment device based on light intensity and personnel detection and related devices such as a core business system, a user terminal, and a related database, etc.; this logic controller can be a desktop computer, a tablet computer, a mobile terminal, etc., and this embodiment is not limited thereto. In this embodiment, this logic controller can be implemented with reference to the embodiments of the intelligent lighting automatic adjustment method based on light intensity and personnel detection, and the embodiments of the intelligent lighting automatic adjustment device based on light intensity and personnel detection, the content of which is incorporated herein, and the repeated parts will not be elaborated.

[0162] It can be understood that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0163] In practical applications, part of the intelligent lighting automatic adjustment method based on light intensity and personnel detection can be executed on the electronic device side as described above, or all operations can be completed in the client device. Specifically, it can be selected according to the processing capacity of the client device and the limitations of the user usage scenario, etc. This application does not make any limitations in this regard. If all operations are completed in the client device, the client device may further include a processor.

[0164] The above-mentioned client device may have a communication module (i.e., a communication unit), and can be communicatively connected to a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and in other implementation scenarios, it may also include a server of an intermediate platform, such as a server of a third-party server platform communicatively linked to the task scheduling center server. The server may include a single computer device, or may include a server cluster composed of multiple servers, or a server structure of a distributed device.

[0165] Figure 9 It is a schematic block diagram of the system composition of the electronic device 9600 according to an embodiment of the present application. As Figure 9 shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It should be noted that this Figure 9is exemplary; other types of structures can also be used to supplement or replace this structure to achieve telecommunication functions or other functions.

[0166] In one embodiment, the function of the intelligent lighting automatic adjustment method based on light intensity and human detection can be integrated into the central processor 9100. Among them, the central processor 9100 can be configured to perform the following controls:

[0167] Step S101: Collect indoor environmental light intensity data through a light intensity sensor, collect indoor human activity data through a human detection sensor, perform reliability evaluation and outlier correction on the light intensity data and the human activity data, use a moving average filtering algorithm to smooth the corrected data, align the smoothed data in time series to obtain a multi-dimensional feature sequence, construct a light activity association model based on the multi-dimensional feature sequence, and use a fuzzy logic algorithm to perform feature fusion on the output of the light activity association model to generate a light scene feature data set;

[0168] Step S102: Input the light scene feature data set into a preset light intelligent adjustment model for training to obtain a light scene adaptation model, construct an illumination effect evaluation index system including illuminance uniformity, color rendering index, and glare value, and perform constraint optimization on the output result of the light scene adaptation model based on the illumination effect evaluation index system to obtain a target light intensity value and lighting device adjustment parameters that meet the illumination effect evaluation index system;

[0169] Step S103: Input the target light intensity value and the lighting device adjustment parameters into a lighting controller, and the lighting controller generates a lighting device control instruction based on a light compensation calculation formula, where the light compensation calculation formula is that the target light intensity value is equal to the sum of the current ambient light intensity and the light compensation value, and the light compensation value is determined by the functional relationship between the human detection probability value and the time period priority factor. The lighting device control instruction is sent to the lighting device execution unit through a control bus for lighting adjustment.

[0170] As can be seen from the above description, the electronic device provided in the embodiment of the present application synchronously collects ambient light intensity and human activity data, performs reliability evaluation and anomaly correction on the data. Innovatively constructs a light activity association model and uses a fuzzy logic algorithm to achieve feature fusion. The system establishes a complete evaluation index system including illuminance uniformity, color rendering index, and glare value, and performs constraint optimization based on the light scene adaptation model. By introducing a dynamic compensation mechanism of human detection probability value and time period priority factor, precise lighting adjustment control is achieved. This method breaks through the limitations of traditional fixed-mode dimming and provides a comprehensive technical solution for intelligent lighting systems.

[0171] In another embodiment, the intelligent lighting automatic adjustment device based on light intensity and personnel detection can be separately configured from the central processor 9100. For example, the intelligent lighting automatic adjustment device based on light intensity and personnel detection can be configured as a chip connected to the central processor 9100, and the functions of the intelligent lighting automatic adjustment method based on light intensity and personnel detection can be realized through the control of the central processor.

[0172] As Figure 9 shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It should be noted that the electronic device 9600 does not necessarily have to include Figure 9 all the components shown in Figure 9 ; in addition, the electronic device 9600 may further include

[0173] As Figure 9 shown, the central processor 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor devices and / or logic devices. The central processor 9100 receives inputs and controls the operations of the various components of the electronic device 9600.

[0174] Among them, the memory 9140 can be, for example, one or more of a buffer, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory, or other suitable devices. The above information related to failures can be stored, and in addition, programs for executing relevant information can also be stored. And the central processor 9100 can execute the programs stored in the memory 9140 to implement information storage or processing, etc.

[0175] The input unit 9120 provides inputs to the central processor 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to supply power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display can be, for example, an LCD display, but is not limited thereto.

[0176] The memory 9140 can be a solid-state memory, for example, a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It can also be a memory that stores information even when power is off, can be selectively erased and has more data. Examples of such a memory are sometimes referred to as EPROMs, etc. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 can include an application / function storage unit 9142, which is used to store application programs and function programs or the processes for operating the electronic device 9600 by the central processor 9100.

[0177] The memory 9140 can also include a data storage unit 9143, which is used to store data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 can include various drivers of the electronic device for communication functions and / or for performing other functions of the electronic device (such as a messaging application, an address book application, etc.).

[0178] The communication module 9110 is a transmitter / receiver that transmits and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which can be the same as in the case of a conventional mobile communication terminal.

[0179] Based on different communication technologies, multiple communication modules 9110 can be provided in the same electronic device, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, etc. The communication module 9110 (transmitter / receiver) is also coupled to the speaker 9131 and the microphone 9132 via the audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby implementing the usual telecommunication functions. The audio processor 9130 can include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that it is possible to record on the local machine through the microphone 9132 and play the sounds stored on the local machine through the speaker 9131.

[0180] An embodiment of the present application also provides a computer-readable storage medium capable of implementing all steps of the intelligent lighting automatic adjustment method based on light intensity and personnel detection, where the execution subject in the above embodiment is a server or a client. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, all steps of the intelligent lighting automatic adjustment method based on light intensity and personnel detection, where the execution subject in the above embodiment is a server or a client, are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0181] Step S101: Collect indoor environmental light intensity data through a light intensity sensor, collect indoor personnel activity data through a personnel detection sensor, perform reliability evaluation and outlier correction on the light intensity data and the personnel activity data, perform smoothing processing on the corrected data using a moving average filtering algorithm, perform time series alignment on the smoothed data to obtain a multi-dimensional feature sequence, construct a light activity association model based on the multi-dimensional feature sequence, and perform feature fusion on the output of the light activity association model using a fuzzy logic algorithm to generate a light scene feature data set;

[0182] Step S102: Input the light scene feature data set into a preset light intelligent adjustment model for training to obtain a light scene adaptation model, construct an illumination effect evaluation index system including illuminance uniformity, color rendering index, and glare value, and perform constraint optimization on the output result of the light scene adaptation model based on the illumination effect evaluation index system to obtain a target light intensity value and lighting device adjustment parameters that meet the illumination effect evaluation index system;

[0183] Step S103: Input the target light intensity value and the lighting device adjustment parameters into a lighting controller, and the lighting controller generates a lighting device control instruction based on a light compensation calculation formula, where the light compensation calculation formula is that the target light intensity value is equal to the sum of the current ambient light intensity and the light compensation value, and the light compensation value is determined by the functional relationship between the personnel detection probability value and the time period priority factor. The lighting device control instruction is sent to the lighting device execution unit through a control bus for lighting adjustment.

[0184] As can be seen from the above description, the computer-readable storage medium provided by the embodiments of the present application synchronously collects ambient light intensity and personnel activity data, evaluates the reliability of the data, and corrects anomalies. An illumination-activity correlation model is innovatively constructed, and fuzzy logic algorithms are used to achieve feature fusion. The system establishes a complete evaluation index system including illuminance uniformity, color rendering index, and glare value, and performs constraint optimization based on the illumination scenario adaptation model. By introducing a dynamic compensation mechanism of personnel detection probability value and time period priority factor, precise lighting adjustment control is achieved. This method breaks through the limitations of traditional fixed-mode dimming and provides a comprehensive technical solution for intelligent lighting systems.

[0185] Embodiments of the present application also provide a computer program product capable of implementing all steps of the intelligent lighting automatic adjustment method based on light intensity and personnel detection, where the execution subject in the above embodiments is a server or a client. When the computer program / instructions are executed by a processor, the steps of the intelligent lighting automatic adjustment method based on light intensity and personnel detection are implemented. For example, the computer program / instructions implement the following steps:

[0186] Step S101: Collect indoor ambient light intensity data through a light intensity sensor, collect indoor personnel activity data through a personnel detection sensor, evaluate the reliability of the light intensity data and the personnel activity data, and correct outliers. Use a moving average filtering algorithm to smooth the corrected data, perform time series alignment on the smoothed data to obtain a multi-dimensional feature sequence, construct an illumination-activity correlation model based on the multi-dimensional feature sequence, and use a fuzzy logic algorithm to perform feature fusion on the output of the illumination-activity correlation model to generate an illumination scenario feature data set;

[0187] Step S102: Input the illumination scenario feature data set into a preset illumination intelligent adjustment model for training to obtain an illumination scenario adaptation model, construct an illumination effect evaluation index system including illuminance uniformity, color rendering index, and glare value, and perform constraint optimization on the output result of the illumination scenario adaptation model based on the illumination effect evaluation index system to obtain a target light intensity value and lighting device adjustment parameters that meet the illumination effect evaluation index system;

[0188] Step S103: Input the target light intensity value and the lighting device adjustment parameters into a lighting controller. The lighting controller generates a lighting device control instruction based on an illumination compensation calculation formula, where the illumination compensation calculation formula is that the target light intensity value is equal to the sum of the current ambient light intensity and the illumination compensation value. The illumination compensation value is determined by the functional relationship between the personnel detection probability value and the time period priority factor. The lighting device control instruction is sent to the lighting device execution unit through a control bus for lighting adjustment.

[0189] As can be seen from the above description, the computer program product provided by the embodiments of the present application synchronously collects ambient light intensity and personnel activity data, evaluates the reliability of the data, and corrects anomalies. An illumination-activity correlation model is innovatively constructed, and feature fusion is realized using the fuzzy logic algorithm. The system establishes a complete evaluation index system including illuminance uniformity, color rendering index, and glare value, and performs constraint optimization based on the illumination scene adaptation model. By introducing a dynamic compensation mechanism for the personnel detection probability value and the time period priority factor, precise lighting adjustment control is achieved. This method breaks through the limitations of traditional fixed-mode dimming and provides a comprehensive technical solution for intelligent lighting systems.

[0190] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, apparatus, or computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0191] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one or more of the processes Figure 1 or multiple processes and / or blocks

[0192] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that realizes the functions specified in Figure 1 one or more of the processes Figure 1 or multiple processes and / or blocks

[0193] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for realizing the functions in Figure 1One process or multiple processes and / or boxes Figure 1 Steps of the functions specified in one box or multiple boxes.

[0194] In the present invention, specific embodiments are used to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. An intelligent lighting automatic adjustment method based on light intensity and personnel detection, characterized in that The method includes: Collecting indoor environmental light intensity data through a light intensity sensor, collecting indoor personnel activity data through a personnel detection sensor, performing reliability evaluation and outlier correction on the light intensity data and the personnel activity data, smoothing the corrected data using a moving average filtering algorithm, performing time series alignment on the smoothed data to obtain a multi-dimensional feature sequence, calculating the joint probability distribution between the light intensity and the personnel activity data in the multi-dimensional feature sequence using a kernel density estimation method, constructing the conditional probability relationship of the joint probability distribution based on a Bayesian network, determining the Bayesian network parameters through maximum likelihood estimation, and establishing a light activity association model; calculating the fuzzy membership degree of the conditional probability value output by the light activity association model, setting a fuzzy rule set for the light intensity and the personnel activity, using a fuzzy inference mechanism to match and calculate the fuzzy rule set, and performing defuzzification processing on the inference result of the fuzzy inference mechanism through the centroid method to obtain a light scene feature data set; Constructing the intelligent light adjustment model using a deep neural network, dividing the light scene feature data set into a training set and a validation set according to a ratio of 8:2, iteratively training the intelligent light adjustment model using a backpropagation algorithm, determining the optimal number of training rounds based on the model performance on the validation set, and obtaining a light scene adaptation model; calculating the illuminance uniformity coefficient of the output result of the light scene adaptation model based on the CIE lighting standard, calculating the general color rendering index Ra value of the light source using a color rendering index measurement method, calculating the indoor lighting glare value using the UGR unified glare value evaluation system, constructing an illumination effect evaluation index system with the illuminance uniformity coefficient, the color rendering index Ra value, and the glare value, setting constraint conditions for the output result of the light scene adaptation model, including that the illuminance uniformity coefficient is not less than 0.7, the color rendering index Ra value is not less than 80, and the glare value does not exceed 19, constructing a multi-objective optimization function based on the constraint conditions, and solving the multi-objective optimization function using the Lagrange multiplier method to obtain the target light intensity value that meets the constraint conditions; constructing an illumination device parameter optimization model based on the target light intensity value, taking the power, color temperature, and irradiation angle of the illumination device as optimization variables, and solving the illumination device parameter optimization model using a particle swarm algorithm to obtain the illumination device adjustment parameters that meet the illumination effect evaluation index system; Input the target light intensity value and the lighting device adjustment parameters into the lighting controller. Divide a day into five time periods: early morning, morning, afternoon, evening, and night based on the time period division rule. Set priority factors for each time period. Calculate the light compensation value by weighted calculation of the personnel detection probability value and the time period priority factor. The personnel detection probability value is obtained by calculating the probability of the presence of personnel in the current area after extracting the personnel activity characteristics from the video surveillance data through a deep learning algorithm. The personnel detection probability value reflects the real-time state of space usage. Generate a lighting device control instruction according to the calculation formula that the target light intensity value is equal to the sum of the current ambient light intensity and the light compensation value; use a communication protocol conversion module to convert the lighting device control instruction into the DALI communication protocol format, and send the converted control instruction to the lighting device execution unit through the data communication interface of the lighting controller. After receiving and parsing the control instruction, the lighting device execution unit performs the corresponding lighting adjustment operation, and sends the lighting device control instruction to the lighting device execution unit through the control bus for lighting adjustment.

2. The intelligent lighting automatic adjustment method based on light intensity and personnel detection according to claim 1, wherein Perform reliability evaluation and outlier correction on the light intensity data and the personnel activity data, use a moving average filtering algorithm to smooth the corrected data, and perform time series alignment on the smoothed data to obtain a multi-dimensional feature sequence, including: Calculate the probability density values of the light intensity data and the personnel activity data based on the Gaussian distribution model, mark the data with probability density values lower than the preset threshold as abnormal data, correct the abnormal data using the linear interpolation method, perform reliability evaluation by calculating the mean, standard deviation, and coefficient of variation of the corrected data, and resample the data with a reliability evaluation result lower than the preset score; Use a moving average filtering algorithm to perform filtering processing on the corrected light intensity data and personnel activity data respectively, perform interpolation resampling on the filtered data according to a unified sampling period based on the timestamp information, and perform time series alignment processing on the resampled data sequence through the least squares method to obtain a multi-dimensional feature sequence.

3. The intelligent lighting automatic adjustment method based on light intensity and personnel detection according to claim 1, characterized in that The step of sending the lighting device control instruction to the lighting device execution unit through the control bus for lighting adjustment includes: Encapsulate the lighting device control instruction according to the DALI protocol specification, perform CRC checksum and address encoding on the encapsulated control instruction, and send the control instruction to the lighting device execution unit through the RS485 bus interface. The lighting device execution unit performs data frame checksum and address matching on the received control instruction; Based on the driver chip of the lighting device execution unit, parse the control instruction into a PWM dimming signal and device operating parameters, control the output current of the LED drive circuit through the PWM dimming signal, and adjust the color temperature and irradiation angle of the lighting device according to the device operating parameters to achieve intelligent adjustment of the lighting device.

4. An intelligent lighting automatic adjustment device based on light intensity and personnel detection, characterized in that, The device includes: A feature fusion module is used to collect indoor environmental light intensity data through a light intensity sensor and collect indoor personnel activity data through a personnel detection sensor. It conducts reliability assessment and outlier correction on the light intensity data and the personnel activity data, uses a moving average filtering algorithm to smooth the corrected data, aligns the smoothed data in time series to obtain a multi-dimensional feature sequence, uses a kernel density estimation method to calculate the joint probability distribution between the light intensity and the personnel activity data in the multi-dimensional feature sequence, constructs the conditional probability relationship of the joint probability distribution based on a Bayesian network, determines the Bayesian network parameters through maximum likelihood estimation, and establishes a light-activity association model; calculates the fuzzy membership degree of the conditional probability value output by the light-activity association model, sets a fuzzy rule set for the light intensity and personnel activity, uses a fuzzy inference mechanism to match and calculate the fuzzy rule set, and defuzzifies the inference result of the fuzzy inference mechanism through the centroid method to obtain a light scene feature data set; Builds the intelligent light adjustment model using a deep neural network, divides the light scene feature data set into a training set and a validation set in a ratio of 8:2, uses the backpropagation algorithm to iteratively train the intelligent light adjustment model, determines the optimal number of training epochs based on the model performance on the validation set, and obtains a light scene adaptation model; calculates the illuminance uniformity coefficient of the output result of the light scene adaptation model based on the CIE lighting standard, calculates the general color rendering index Ra value of the light source using a color rendering index measurement method, calculates the indoor lighting glare value using the UGR unified glare value evaluation system, constructs an illumination effect evaluation index system with the illuminance uniformity coefficient, the color rendering index Ra value, and the glare value, sets constraint conditions for the output result of the light scene adaptation model, including that the illuminance uniformity coefficient is not less than 0.7, the color rendering index Ra value is not less than 80, and the glare value does not exceed 19, constructs a multi-objective optimization function based on the constraint conditions, and uses the Lagrange multiplier method to solve the multi-objective optimization function to obtain the target light intensity value that meets the constraint conditions; constructs an illumination device parameter optimization model based on the target light intensity value, takes the power, color temperature, and irradiation angle of the illumination device as optimization variables, and uses a particle swarm algorithm to solve the illumination device parameter optimization model to obtain the illumination device adjustment parameters that meet the illumination effect evaluation index system; An automatic adjustment module is used to input the target light intensity value and the lighting device adjustment parameters into the lighting controller. Based on the time period division rule, a day is divided into five time periods: early morning, morning, afternoon, evening, and night. Priority factors are set for each time period, and the personnel detection probability value is weighted with the time period priority factor to obtain the light compensation value. The personnel detection probability value is calculated by extracting the personnel activity characteristics from the video surveillance data through a deep learning algorithm and then calculating the probability of the presence of personnel in the current area. The personnel detection probability value reflects the real-time state of space usage. According to the calculation formula that the target light intensity value is equal to the sum of the current ambient light intensity and the light compensation value, a lighting device control instruction is generated; the communication protocol conversion module is used to convert the lighting device control instruction into the DALI communication protocol format, and the converted control instruction is sent to the lighting device execution unit through the data communication interface of the lighting controller. After receiving and parsing the control instruction, the lighting device execution unit performs the corresponding lighting adjustment operation, and the lighting device control instruction is sent to the lighting device execution unit through the control bus for lighting adjustment.

5. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the intelligent lighting automatic adjustment method based on light intensity and personnel detection according to any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the intelligent lighting automatic adjustment method based on light intensity and personnel detection according to any one of claims 1 to 3.

Images