Intelligent garage lamp control method and system with scene switching function

Abstract

The invention relates to the technical field of intelligent control, and discloses an intelligent garage lamp control method and system with a scene switching function, and the method comprises the steps: obtaining real-time garage data, and carrying out the grid division, and obtaining scene activity information; analyzing the frequency intensity and the path track of the activity in the garage, and determining the activity density characteristics in the garage; in combination with the activity density characteristics, obtaining activity activeness; obtaining a demand division result of illumination according to the activity activeness; and formulating a preliminary illumination scheme according to the demand division result, and collecting real-time data for correction after executing the preliminary illumination scheme to obtain a final illumination scheme. According to the invention, the problem of insufficient scene adaptability of garage lamp control in the prior art can be effectively solved.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to an intelligent garage light control method and system with scene switching function. Background Technology

[0002] Currently, with the deepening development of smart city and green building concepts, intelligent control of garage lighting systems based on electroluminescence has become a key aspect of improving energy efficiency and user experience. Traditional garage lighting generally adopts a constant-on or simple zone control mode using high-power LED components, which makes it difficult to make fine adjustments based on the real-time activities of vehicles and people in the garage. There is an urgent need to achieve on-demand lighting through intelligent sensing and dynamic control technologies to achieve energy-saving goals while ensuring safety and comfort.

[0003] In existing technologies, intelligent garage lighting control methods typically employ lighting strategies based on timed control or simple sensor triggering. These methods rely on preset time periods combined with the switching control of basic LED components, or simply triggering the on / off state of local electroluminescent units via vehicle detection sensors. This approach is relatively simplistic in its scene perception, failing to comprehensively analyze the dynamic correlation between activity density, pedestrian flow paths, and ambient light intensity in different areas of the garage. For example, the system might be configured with a high-brightness LED array to illuminate the entire area at fixed times at night, without identifying the presence of actual activity in specific areas; or it might only illuminate the corresponding lane lights via electroluminescent units when a vehicle enters, but fail to predict the walking paths of people after they exit the vehicle and the surrounding lighting needs. This lack of multi-dimensional scene perception and predictive dimming mechanisms results in delayed lighting response of electroluminescent units in high-activity areas, negatively impacting user experience and demonstrating insufficient scene adaptability.

[0004] Existing technologies suffer from insufficient adaptability to different scenarios. Summary of the Invention

[0005] This invention provides an intelligent garage light control method and system with scene switching function to solve the problem of insufficient scene adaptability of existing technologies.

[0006] Firstly, in order to solve the above-mentioned technical problems, the present invention provides an intelligent garage light control method with scene switching function, comprising:

[0007] Acquire activity data and light intensity data of vehicles and people in the garage; filter high-activity, low-light areas based on the activity data and light intensity data; acquire movement trajectory data of vehicles and people in the high-activity, low-light areas; and obtain scene activity information.

[0008] The activity frequency and movement trajectory of each region are extracted from the scene activity information and cluster analysis is performed to obtain the activity density characteristics of each region;

[0009] Extract the duration of activity in each region, and combine it with the activity density characteristics to assess the activity level in each region;

[0010] Based on the activity level, high-demand lighting areas are selected, insufficient brightness areas within the high-demand lighting areas are identified, and the lighting needs of the insufficient brightness areas are analyzed to obtain the demand segmentation results.

[0011] Based on the demand segmentation results, combined with the pre-acquired lighting fixture grouping strategy and energy-saving information, a preliminary lighting plan is formulated;

[0012] The preliminary lighting scheme is executed, ambient light change data is collected, and the preliminary lighting scheme is adjusted based on the ambient light change data to obtain the final lighting scheme.

[0013] Preferably, the activity data of vehicles and people in the garage and the light intensity data are acquired. Based on the activity data and the light intensity data, high-activity, low-light areas are filtered, and the movement trajectory data of vehicles and people within the high-activity, low-light areas are acquired to obtain scene activity information, including:

[0014] Collect and integrate activity data of vehicles and people in the garage, as well as light intensity data, to obtain the raw dataset;

[0015] The garage area is divided into grids, and the original dataset is grouped by grid cells to obtain grid activity data;

[0016] Extract the frequency of human and vehicle activity and the light intensity of each grid cell from the grid activity data. If the frequency of human and vehicle activity is higher than a preset frequency threshold and the light intensity is lower than a preset light threshold, then mark the grid cell as a high-activity, low-light area.

[0017] The movement trajectory data of vehicles and people in the high-activity, low-light area are obtained to obtain scene activity information.

[0018] Preferably, the activity frequency and movement trajectory of each region are extracted from the scene activity information and cluster analysis is performed to obtain the activity density characteristics of each region, including:

[0019] Extract the activity frequency and movement trajectory data of each grid cell from the scene activity information;

[0020] Cluster analysis is performed on the activity frequency and movement trajectory data, and the activity density characteristics of each grid region are determined based on the clustering results.

[0021] Preferably, the activity duration of each region is extracted, and the activity intensity of each region is evaluated in combination with the activity density characteristics, including:

[0022] Extract the duration of activity in each region and match it with the activity density feature to obtain the correspondence between activity duration and activity density;

[0023] The triggering conditions for grid activity density are extracted from the correspondence, and the activity density characteristics are combined to determine the activity distribution pattern of each grid region within a specific time period.

[0024] Based on the activity distribution pattern, assess the activity level of each grid area.

[0025] Preferably, high-demand lighting areas are selected based on the activity level, insufficient brightness areas within the high-demand lighting areas are identified, and the lighting needs of the insufficient brightness areas are analyzed to obtain a demand segmentation result, including:

[0026] If the activity level is higher than a preset activity threshold, the grid area exceeding the threshold will be marked as a high-demand lighting area;

[0027] Acquire lighting data within the high-demand lighting area and analyze the matching degree between the activity level and the lighting data to determine areas with insufficient brightness; wherein, the lighting data includes brightness distribution data and ambient light intensity records;

[0028] Extract the spatial location and time period division information of the insufficient brightness area, combine it with the preset adjustment range data to generate the lighting priority of each area, and determine the lighting demand division result.

[0029] Preferably, based on the demand segmentation results, and combined with the pre-acquired grouping strategy and energy-saving information of lighting fixtures, a preliminary lighting scheme is formulated, including:

[0030] The brightness distribution data of each region is obtained from the lighting demand division results. According to the pre-acquired grouping strategy of lighting fixtures, the regions where the brightness distribution data is lower than the brightness threshold are marked as regions that need adjustment.

[0031] Obtain the ambient light intensity record of the area to be adjusted. If the ambient light intensity record does not match the preset energy-saving information, adjust the brightness switching speed to obtain the brightness switching speed range of each area.

[0032] Based on the speed range, priority information for each region is extracted, and the regions with high priority are dynamically adjusted to obtain a preliminary lighting scheme.

[0033] Preferably, the preliminary lighting scheme is executed, ambient light change data is collected, and the preliminary lighting scheme is adjusted based on the ambient light change data to obtain the final lighting scheme, including:

[0034] The preliminary lighting scheme is executed, and ambient light change data is collected in real time. When the ambient light change data exceeds the preset light change threshold range, the area exceeding the threshold is marked as the intervention area.

[0035] Extract the current brightness state of the lighting equipment in the intervention area, calculate the difference between the brightness state and the preset target value, adjust the brightness according to the difference value, and obtain a dynamic adjustment command;

[0036] Based on the dynamic adjustment command, the preliminary lighting scheme is adjusted, the final brightness control command is determined, and the final lighting scheme is obtained.

[0037] Secondly, the present invention provides an intelligent garage light control system with scene switching function, comprising:

[0038] The data acquisition module is used to acquire activity data and light intensity data of vehicles and people in the garage, filter high-activity low-light areas based on the activity data and light intensity data, acquire the movement trajectory data of vehicles and people in the high-activity low-light areas, and obtain scene activity information.

[0039] The feature extraction module is used to extract the activity frequency and movement trajectory of each region from the scene activity information and perform cluster analysis to obtain the activity density features of each region;

[0040] The activity assessment module is used to extract the duration of activities in each region and, in conjunction with the activity density characteristics, assess the activity level in each region.

[0041] The demand analysis module is used to filter high-demand lighting areas based on the activity level, identify insufficient brightness areas in the high-demand lighting areas, analyze the lighting demand of the insufficient brightness areas, and obtain demand segmentation results.

[0042] The scheme generation module is used to formulate a preliminary lighting scheme based on the required division results, combined with the pre-acquired grouping strategy and energy-saving information of lighting fixtures;

[0043] The feedback adjustment module is used to execute the preliminary lighting scheme, collect ambient light change data, adjust the preliminary lighting scheme according to the ambient light change data, and obtain the final lighting scheme.

[0044] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the intelligent garage light control method with scene switching function described in any one of the above.

[0045] Fourthly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the intelligent garage light control method with scene switching function described above.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] (1) This invention obtains activity density characteristics by cluster analysis of activity frequency and movement trajectory, and evaluates activity level by combining activity duration. It constructs a multi-dimensional scene perception system, which breaks through the limitations of traditional methods that rely only on fixed triggering of time or single sensing. It can dynamically capture the complex correlation and change pattern of human and vehicle activities in the garage, accurately identify the differences in real-time lighting needs in different areas, and make the system response shift from "passive triggering" to "active adaptation", which significantly enhances the system's adaptability to dynamic scenes.

[0048] (2) This invention solves the problem of the inability of traditional technology to adapt to dynamic environmental changes by implementing a closed-loop control of scheme execution, data acquisition and real-time optimization. It can collect ambient light change data in real time, quickly calculate and dynamically adjust the brightness difference of areas exceeding the threshold, and ensure that the lighting brightness always accurately matches the current scene requirements. It avoids insufficient brightness caused by environmental changes, and enables the system to flexibly adapt to the dynamic fluctuations of the light environment in the garage, greatly improving scene adaptability.

[0049] (3) This invention divides the garage area into grids and groups multi-source data into grids to achieve precise binding of activity data with spatial location, solving the problem of poor scene adaptability caused by traditional unified control of the whole area. This solution refines the control unit to the grid and dynamically adjusts the brightness switching speed in combination with lighting priority, so that the brightness adjustment can accurately focus on the specific scene requirements of each grid. At the same time, by optimizing the switching speed, the discomfort caused by sudden changes in light is avoided, and the system control is transformed from coarse and general to fine adaptation, which significantly enhances the adaptability to fragmented scenes in the garage. Attached Figure Description

[0050] Figure 1 This is a schematic flowchart of an intelligent garage light control method with scene switching function provided in the first embodiment of the present invention;

[0051] Figure 2This is a schematic diagram of an intelligent garage light control system with scene switching function provided in the second embodiment of the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Reference Figure 1 The first embodiment of the present invention provides the following steps:

[0054] S11, acquire activity data and light intensity data of vehicles and people in the garage, filter high-activity low-light areas based on the activity data and light intensity data, acquire movement trajectory data of vehicles and people in the high-activity low-light areas, and obtain scene activity information;

[0055] S12, extract the activity frequency and movement trajectory of each region from the scene activity information and perform cluster analysis to obtain the activity density characteristics of each region;

[0056] S13, extract the duration of activity in each region, and evaluate the activity level of each region in combination with the activity density characteristics;

[0057] S14, based on the activity level, filter high-demand lighting areas, identify insufficient brightness areas in the high-demand lighting areas, analyze the lighting needs of the insufficient brightness areas, and obtain the demand segmentation results;

[0058] S15. Based on the demand segmentation results, and combined with the pre-acquired lighting fixture grouping strategy and energy-saving information, formulate a preliminary lighting plan.

[0059] S16, execute the preliminary lighting scheme, collect ambient light change data, adjust the preliminary lighting scheme according to the ambient light change data, and obtain the final lighting scheme;

[0060] In step S11, it is necessary to acquire activity data and light intensity data of vehicles and people in the garage. Based on the activity data and light intensity data, high-activity, low-light areas are filtered out, and the movement trajectory data of vehicles and people in the high-activity, low-light areas are acquired to obtain scene activity information, including:

[0061] Collect and integrate activity data of vehicles and people in the garage, as well as light intensity data, to obtain the raw dataset;

[0062] The garage area is divided into grids, and the original dataset is grouped by grid cells to obtain grid activity data;

[0063] Extract the frequency of human and vehicle activity and the light intensity of each grid cell from the grid activity data. If the frequency of human and vehicle activity is higher than a preset frequency threshold and the light intensity is lower than a preset light threshold, then mark the grid cell as a high-activity, low-light area.

[0064] The movement trajectory data of vehicles and people in the high-activity, low-light area are obtained to obtain scene activity information.

[0065] Firstly, a multi-source sensor network is deployed within the garage. This network employs a hierarchical topology, with 16 sensor nodes evenly distributed across the garage ceiling. Each node contains three types of sensing units: a geomagnetic sensor, an infrared thermal imaging camera, and a lux meter. The node spacing is set to 8-10 meters based on the garage's column grid structure. PoE power supply and a wired backup network ensure system stability. Specifically, the geomagnetic sensor detects vehicle entry and exit at a sampling frequency of 10Hz, the infrared thermal imaging camera captures the contours of human activity at a frame rate of 5fps, and the lux meter continuously records light intensity (in lux). All sensor data is transmitted to the edge computing gateway via the ZigBee wireless communication protocol. A timestamp alignment algorithm is used to synchronize the heterogeneous data in time, and the original signal is filtered using a moving average to eliminate impulse noise. Finally, a multi-dimensional raw dataset containing timestamps, spatial coordinates, activity types, and light values ​​is generated.

[0066] The garage area was then divided into grids, and the original monitoring dataset was grouped by grid unit. A regular grid division method was used: first, a two-dimensional rectangular coordinate system was established based on the garage's CAD floor plan, with the main load-bearing columns as reference points. The optimal coordinate reference axis was obtained through least squares fitting. The grid unit size was set to 6 meters × 4 meters, a parameter determined by cluster analysis of the diameter distribution of hotspot areas in historical activity data, ensuring that each grid completely covers the activity range of 2-3 standard parking spaces. A quadtree coding system was used for grid identification, with each grid having a unique Morton code as a spatial index. In the data mapping stage, an improved spatial point-in-polygon algorithm was used. The ray casting method was used to determine the grid to which the sensor data points belonged. For data points located within a ±0.2-meter buffer zone at the grid boundary, an inverse distance weighted interpolation algorithm was used to assign weights to adjacent grids based on the inverse of the square of the distance. Finally, gridded data groups containing time-series activity counts and light intensity sequences were generated, providing a structured data foundation for subsequent analysis.

[0067] Subsequently, the frequency of human and vehicle activities and the light intensity of each grid cell are extracted from the grid activity data. If the frequency of human and vehicle activities is higher than a preset frequency threshold and the light intensity is lower than a preset light intensity threshold, the grid cell is marked as a high-activity, low-light area. The activity frequency is calculated using a sliding time window statistical method, with a 15-minute window to calculate the total number of activity events per unit time. The frequency threshold is dynamically set to 4 times / minute based on the 75th percentile of historical data, and can also be dynamically adjusted according to the garage type and actual data. The light intensity threshold is fixed at 50 lux (based on safety lighting standards). Grid cells that exceed both thresholds are marked as high-activity, low-light areas, and their spatial coordinates and the time period of exceeding the threshold are recorded.

[0068] It should be noted that the preset frequency threshold is dynamically set through statistical learning methods. Based on historical activity data of 30 consecutive working days, the activity frequency distribution of each grid unit in different time periods is calculated, and the 75th percentile value is taken as the benchmark threshold. An adaptive adjustment range of ±15% is introduced to cope with special scenarios such as holidays. The preset illumination threshold is based on the illuminance requirements of garage work surfaces in the national standard "Standard for Lighting Design of Buildings" (GB 50034-2013), combined with ambient light attenuation experimental data, and is finally set to 50 lux. This value takes into account the balance between human eye comfort and energy saving requirements.

[0069] Finally, the movement trajectory data of vehicles and personnel in the high-activity, low-light area are obtained to obtain scene activity information. This is specifically achieved through multi-source sensor data acquisition and trajectory reconstruction. First, a camera array and a geomagnetic sensor network are deployed to continuously acquire raw location point cloud data at a fixed sampling frequency (e.g., 10 frames per second). A multi-target tracking algorithm (e.g., SORT algorithm) is used to associate detection boxes under continuous timestamps. Kalman filtering is used to smooth the trajectory and eliminate noise to generate a continuous movement trajectory. Then, the trajectory feature vector is extracted, including average movement speed, rate of change of direction angle, and distribution of dwell points, to form a standardized trajectory dataset. Based on this dataset, peak detection was performed on the activity density time series of each grid cell. The sliding window Z-score algorithm was used to calculate the real-time deviation, with the window size set to 7 days to capture cyclical patterns. Time periods exceeding twice the standard deviation of the overall average were identified as local peaks, where the average and standard deviation were dynamically updated based on rolling historical data. Subsequently, the Apriori algorithm was used to mine high-frequency co-occurring spatiotemporal patterns. This algorithm generates candidate itemsets and iteratively calculates support, setting the minimum support to 0.05 and the minimum confidence to 0.6. Hash trees were used to optimize itemset storage, and strong association rules such as "18:00-19:00 & grid B12" were mined.

[0070] It should be noted that, based on statistical analysis of 30 days of historical data, a support level ≥ 0.05 can cover more than 90% of high-frequency activity patterns, and a confidence level ≥ 0.6 can guarantee rule reliability > 85%. Therefore, the minimum support level is set to 0.05 and the minimum confidence level is set to 0.6.

[0071] It should be noted that the generated scene activity is a structured data table containing peak time periods, spatial hotspot coordinates, activity intensity index (calculated by peak area integral), and confidence level (based on rule-based lift). This data can guide the lighting system to perform targeted brightness enhancement and inspection path optimization through visual map overlay. For example, during peak activity periods, the light intensity of the corresponding grid area can be adaptively increased to over 100 lux.

[0072] In step S12, it is necessary to extract the activity frequency and movement trajectory of each region from the scene activity information and perform cluster analysis to obtain the activity density characteristics of each region, including:

[0073] Extract the activity frequency and movement trajectory data of each grid cell from the scene activity information;

[0074] Cluster analysis is performed on the activity frequency and movement trajectory data, and the activity density characteristics of each grid region are determined based on the clustering results.

[0075] First, the activity frequency and movement trajectory data of each grid unit are extracted from the scene activity information. This is achieved through time segmentation and feature extraction. The specific process includes two sub-steps: time segmentation and feature extraction. Time segmentation uses a fixed interval method, such as dividing continuous time data into segments with a 30-minute window to ensure that data within each time period is processed independently. In feature extraction, the activity frequency is obtained by counting the number of vehicle or personnel entering and exiting the grid unit within a time period. The movement trajectory features are calculated by analyzing the trajectory point sequence to determine the average speed (i.e., the ratio of total displacement to total time), movement direction (represented by the vector angle of the trajectory start and end points, calculated using the arctangent function), and path length (obtained by accumulating the Euclidean distance between adjacent trajectory points). Specifically, the collected activity data is divided according to a preset time interval, such as 30 minutes. For each grid unit, the activity frequency within that time period, i.e., the number of vehicle or personnel entering and exiting, is calculated. At the same time, features including the average speed, movement direction, and path length of the trajectory point sequence are extracted from the movement trajectory data to form a numerical feature vector.

[0076] The step of clustering the activity frequency and movement trajectory data employs the K-means clustering algorithm. First, the optimal number of clusters K is determined using the elbow rule, i.e., the sum of squares within each cluster corresponding to different K values ​​is calculated. The K value corresponding to the inflection point is selected as the number of clusters. Then, K cluster centers are randomly initialized. A maximum number of iterations and a threshold for center point change are set as convergence conditions. Through the iterative process, the Euclidean distance between each feature vector and the cluster center is calculated (to eliminate the influence of dimensions, each feature is first standardized using Z-scores) and assigned to the nearest cluster. Subsequently, the cluster center is updated to the mean of the data points within the cluster. This iteration is repeated until the maximum number of iterations is reached or the center point movement distance is less than the set threshold. The objective function is to minimize... , where x represents the eigenvector. Let i be the center point of the i-th cluster. For the set of data points within a cluster, The objective function, which determines the number of clusters, ensures that the clustering results maximize the similarity of samples within each cluster, thereby dividing the grid cells into different activity density categories. Finally, based on the clustering results, the activity density characteristics of each grid region are determined.

[0077] It should be noted that the activity density feature is a quantitative indicator that represents the degree of activity concentration in each grid area within a specific time period, such as high density, medium density, or low density. It is obtained based on the category division output by cluster analysis. This feature is used to subsequently assess activity activity to guide the classification of lighting needs. For example, in a parking garage management scenario, assuming that the activity density of grid B05 during the morning period of 7:00-9:00 is determined to be "high density" through cluster analysis, it indicates that the frequency of vehicle entry and exit in this area reaches 8 times per minute. Based on this, the system prioritizes marking this area as a high-demand area in the lighting demand classification and formulates a dimming plan to increase the basic lighting from 50 lux to 70 lux to optimize energy efficiency and safety.

[0078] In step S13, it is necessary to extract the duration of activity in each region and, in conjunction with the activity density characteristics, evaluate the activity level of each region, including:

[0079] Extract the duration of activity in each region and match it with the activity density feature to obtain the correspondence between activity duration and activity density;

[0080] The triggering conditions for grid activity density are extracted from the correspondence, and the activity density characteristics are combined to determine the activity distribution pattern of each grid region within a specific time period.

[0081] Based on the activity distribution pattern, assess the activity level of each grid area.

[0082] First, the duration of activities in each region is extracted and matched with the activity density features to obtain the correspondence between activity duration and activity density. First, the continuous time axis is divided into fixed-length intervals (e.g., 1 hour) using the sliding window technique. The data is then processed using Z-score standardization to eliminate dimensional differences. Then, the linear correlation between activity density and duration is calculated using the Pearson correlation coefficient to generate an association matrix sorted by time period, thus obtaining the correspondence between activity density and duration.

[0083] Next, the triggering conditions for grid activity density are extracted from the correspondence, and combined with the activity density characteristics, the activity distribution pattern of each grid area within a specific time period is determined. Specifically, a multivariate statistical analysis method is adopted. First, the multidimensional variables such as activity triggering conditions, activity density, and timestamps are Z-score standardized to eliminate the influence of dimensions. Then, a covariance matrix is ​​constructed to analyze the correlation between variables. Eigenvalues ​​and eigenvectors are calculated using the principal component analysis algorithm. The principal components with a cumulative contribution rate of more than 85% are selected as the new feature space to achieve data dimensionality reduction. Subsequently, the K-means clustering algorithm is combined, and the elbow rule is used to determine the optimal number of clusters k. The sum of squares within the clusters corresponding to different k values ​​is calculated and a curve is plotted. The inflection point k=3 is selected as the number of categories. After initializing the cluster centers, the intra-cluster distance is minimized through iterative optimization. The activity patterns are divided into high, medium, and low density categories, and a pattern distribution map is generated to visualize the spatial activity distribution pattern.

[0084] Then, based on the activity distribution pattern, the activity level of each grid area is evaluated. If the activity pattern distribution does not match the preset density threshold, the system automatically triggers a spatial-temporal joint query mechanism. The boundary coordinates of the target area are quickly located through the R-tree index structure of the spatial database. The ST_Within spatial operator is used to filter all data points within the polygon range. At the same time, the B-tree temporal index is used to retrieve records that meet the time period requirements. The query statement includes spatial range constraints and time interval overlap judgment conditions. Detailed records of spatial location and temporal distribution of the area are extracted. Subsequently, the activity level is graded using a decision tree classification algorithm. The C4.5 algorithm selects the optimal splitting attribute by calculating the information gain ratio, recursively constructs a classification tree model, and finally generates an evaluation report containing activity level and confidence score. The density threshold is dynamically set through statistical analysis methods and determined based on the 75th percentile of historical activity data.

[0085] For example, in a garage scenario, by analyzing vehicle entry and exit data collected by sensors, the system identifies that the activity level of grid area D05 is "high" during the evening peak hours (17:00-19:00), which is manifested as an average of more than 25 activities per minute. Based on this characteristic, the lighting system automatically increases the brightness of the area from the base value of 40 lux to 60 lux and links the security camera to increase the monitoring frequency, thereby optimizing energy efficiency and safety.

[0086] In step S14, it is necessary to filter high-demand lighting areas based on the activity level, identify insufficient brightness areas within the high-demand lighting areas, analyze the lighting needs of the insufficient brightness areas, and obtain the demand segmentation results, including:

[0087] If the activity level is higher than a preset activity threshold, the grid area exceeding the threshold will be marked as a high-demand lighting area;

[0088] Acquire lighting data within the high-demand lighting area and analyze the matching degree between the activity level and the lighting data to determine areas with insufficient brightness; wherein, the lighting data includes brightness distribution data and ambient light intensity records;

[0089] Extract the spatial location and time period division information of the insufficient brightness area, combine it with the preset adjustment range data to generate the lighting priority of each area, and determine the lighting demand division result.

[0090] First, if the activity level exceeds a preset activity threshold, the grid area exceeding the threshold is marked as a high-demand lighting area. Specifically, a dynamic threshold comparison method is used, employing a sliding window technique to smooth the activity data to reduce noise interference. Then, a weighted accumulation algorithm is used to calculate the activity index, where the number of vehicles is given a higher weight to reflect their mobility impact, and the duration of personnel is given an appropriate weight to capture dwell characteristics. After the index is calculated, the system automatically extracts the corresponding data sequence from the historical 30-day database, uses a percentile statistical algorithm to dynamically set the threshold benchmark value, and introduces an adaptive adjustment mechanism to fine-tune the threshold range according to recent fluctuations. Finally, a real-time comparison module matches the current index with the threshold. When the threshold is exceeded, the area marking logic is immediately triggered to complete the identification of high-demand lighting areas.

[0091] It should be noted that the activity threshold is dynamically set using statistical methods. The initial threshold benchmark is calculated based on the 80th percentile of historical 30-day activity data. An adaptive adjustment mechanism is then introduced, dynamically fine-tuning the upper and lower limits of the threshold based on the standard deviation of the most recent 7 days of data. For example, the threshold range is set to the benchmark value ± 1.5 times the standard deviation to accommodate diurnal fluctuations. For instance, in a garage management scenario, the system calculates the 80th percentile of activity frequency from historical 30-day data to be 20 times per minute, setting 20 times / minute as the benchmark threshold. Subsequently, the standard deviation of the most recent 7 days of data is calculated to be 3 times / minute, resulting in a dynamic threshold range of 20 ± 4.5 times / minute (i.e., 15.5 to 24.5 times / minute). When the real-time activity frequency exceeds 24.5 times, a high activity level is triggered, automatically increasing the lighting level, thus achieving adaptive threshold adjustment based on usage patterns.

[0092] Next, lighting data within the high-demand lighting area is acquired, and the matching degree between the activity level and the lighting data is analyzed to determine areas with insufficient brightness. Specifically, light intensity difference analysis technology is used, and a distributed illuminance sensor network deployed at the top of the grid collects real-time brightness values ​​of 9 measuring points in a high-frequency sampling manner. A continuous iso-illuminance distribution map is generated using a spatial interpolation algorithm to visualize the brightness gradient. Simultaneously, a dynamic ambient light intensity benchmark value is set based on historical data statistical analysis. The Euclidean distance metric algorithm is used to quantify the dispersion of the brightness of each measuring point from the benchmark value. The matching degree index is obtained by calculating the average difference degree and normalizing it. When the matching degree is lower than a preset threshold of 0.7, the area marking logic is automatically triggered, and the area is determined to be a region with insufficient brightness.

[0093] It should be noted that the matching degree threshold of 0.7 was set based on multi-dimensional verification: First, 100 sets of typical scene data were collected through pre-experiment, and the matching degree distribution was analyzed to find that the perceived comfort of the human eye decreased significantly when it was below 0.68; Second, it was combined with the requirement of the industry standard "Standard for Lighting Design of Buildings" (GB50034) that the uniformity of illuminance in garages should not be lower than 0.7; Finally, the analysis of the subject operating characteristic curve determined that 0.7 was the optimal balance point, at which the false alarm rate and the false negative rate were both below 5%.

[0094] Then, the spatial location and time period division information of the insufficient brightness areas are extracted, and the lighting priority of each area is generated by combining the preset adjustment range data to determine the lighting demand division results. First, a three-level hierarchical structure is established, including the target layer (lighting priority), the criterion layer (spatial weight, time weight, adjustment coefficient), and the scheme layer (each area to be evaluated). Then, a judgment matrix is ​​constructed using the 1-9 scaling method to compare the factors pairwise. The spatial weight is calculated inversely proportional to the actual distance from the grid center to the exit, the time weight is calculated based on the peak period coefficient based on historical activity data, and the adjustment coefficient is determined based on the performance parameters of the lighting fixtures. Then, the eigenvector method is used to solve for the weight vector corresponding to the largest eigenvalue, and consistency is verified. The ratio test is used to determine the validity of the matrix (weight allocation is accepted when CR < 0.1). Finally, the weights of each factor are weighted and synthesized to obtain a comprehensive priority score. The lighting demand classification result is generated based on the priority score. For example, in the garage management scenario, for the grid area B08 with insufficient brightness, its spatial weight is calculated to be 0.8 (only 10 meters away from the main exit), its time weight is 0.7 (it is in the peak period of 18:00-19:00), and its adjustment coefficient is 0.5 (the lamp supports stepless dimming). The comprehensive priority score is calculated to be 0.72 by the analytic hierarchy process. It is classified as a high priority area and the dimming scheme is deployed first in the lighting demand classification, increasing the basic brightness from 40 lux to 60 lux.

[0095] It should be noted that the lighting demand segmentation result is a structured dataset containing spatial coding, illuminance standards, time strategies, and control parameters. It is obtained through the fusion and optimization calculation of the above multi-source data. When used, it can be directly sent to the new lighting control system to perform dynamic dimming, and can be continuously optimized through model predictive control algorithms based on real-time monitoring data to ensure the balance between lighting effect and energy saving target.

[0096] In step S15, based on the demand segmentation results and combined with the pre-acquired lighting fixture grouping strategy and energy-saving information, a preliminary lighting plan needs to be formulated, including:

[0097] The brightness distribution data of each region is obtained from the lighting demand division results. According to the pre-acquired grouping strategy of lighting fixtures, the regions where the brightness distribution data is lower than the brightness threshold are marked as regions that need adjustment.

[0098] Obtain the ambient light intensity record of the area to be adjusted. If the ambient light intensity record does not match the preset energy-saving information, adjust the brightness switching speed to obtain the brightness switching speed range of each area.

[0099] Based on the speed range, priority information for each region is extracted, and the regions with high priority are dynamically adjusted to obtain a preliminary lighting scheme.

[0100] First, the brightness distribution data of each area is obtained from the lighting demand segmentation results. According to the pre-acquired grouping strategy of lighting fixtures, areas where the brightness distribution data is lower than the brightness threshold are marked as areas that need adjustment. Specifically, this is achieved through data comparison and threshold judgment. Based on the lighting demand segmentation results, the real-time brightness measurement value of each grid unit is extracted. The brightness grouping strategy divides the garage into multiple lighting groups according to the type and installation location of the lighting fixtures. For example, the main road lighting fixtures are divided into a high brightness group and the corner lighting fixtures are divided into a low brightness group. The brightness threshold is set to the minimum standard value for safe lighting, such as 50 lux, through historical data statistics. A point-by-point comparison algorithm is used to compare the brightness data of each grid with the threshold of the corresponding group. If the brightness value is lower than the threshold, the grid is marked as an area that needs adjustment.

[0101] Then, the ambient light intensity record of the area to be adjusted is obtained. If the ambient light intensity record does not match the preset energy-saving information, the brightness switching speed is adjusted to obtain the brightness switching speed range of each area. Specifically, this is accomplished through ambient light intensity analysis and energy-saving mode matching. The least squares method is used to fit ΔE and optimal switching speed data in a large number of actual scenarios to ensure that the speed adjustment matches the actual needs. Since ΔE varies in different areas, a linear mapping function is used. Different switching speeds are calculated for different regions, thus forming a speed range, among which, The difference between ambient light intensity and the upper limit of the energy-saving standard is represented by calibration coefficients k and b, which are obtained through regression analysis of historical data. For example, k = 0.2 (lux / second) and b = 5 lux are determined through regression analysis of 100 sets of historical data, with a coefficient of determination R² = 0.95. First, real-time ambient light intensity data of the area to be adjusted is acquired from the sensor. The energy-saving information is preset to an ideal illumination range based on time periods, such as 30-70 lux during the evening. The difference between the ambient light intensity and the upper limit of the energy-saving standard is calculated using the difference calculation method. ,in To measure the light intensity, For the target light intensity, if ΔE exceeds the tolerance threshold, it is determined to be a mismatch. The brightness switching speed is determined by a linear mapping function based on the magnitude of ΔE. Adjustments were made to obtain the speed range for each region, such as an increase of 10-30 lux per minute.

[0102] Finally, based on the speed range, priority information for each region is extracted, and high-priority regions are dynamically adjusted to obtain a preliminary lighting scheme. This is achieved through priority evaluation and dynamic programming. Priority information is calculated using a weighted scoring algorithm based on factors such as activity level and time period importance. For example, the priority score P = w1A + w2T, where A is activity level, T is time period weight, and w1 and w2 are weight coefficients. Dynamic programming is implemented by defining state variables to represent resource allocation in each region. State transitions are based on priority scores and speed range constraints, and are optimized using a value iteration algorithm such as the Bellman equation V(s) = max[R(s,a) + γV(s')], where s represents the current state, a is the action (e.g., allocating switching speed), R is the immediate reward, γ is the discount factor, and s' is the next state. The state value function is iteratively updated until convergence to obtain the optimal strategy. Higher-priority regions are given faster brightness switching speeds and resources are allocated preferentially. Dynamic adjustment generates a preliminary lighting scheme containing brightness target values, switching sequences, and linkage rules through the above optimization process. The preliminary lighting scheme is a scheme containing brightness... The structured plan for adjusting parameters, switching sequences, and area priorities, obtained through integrated analysis of the above steps, guides subsequent real-time lighting control to optimize energy efficiency and response speed. For example, in a parking garage management scenario, for grid area D08 during the evening peak hours (18:00-19:00), the activity level A = 0.85 (calculated based on sensor data), the time period weight T = 0.9 (higher weight during peak hours), and assuming w1 = 0.7 and w2 = 0.3, the priority score P = 0.7 × 0.85 + 0.3 × 0.9 = 0.865 is calculated. Through dynamic programming optimization, state s is defined as the current brightness (40 lux) and activity intensity of D08. Action a includes increasing the brightness to 50, 60, or 70 lux. The reward R is calculated based on energy efficiency and activity matching degree (such as increased activity smoothness after brightness increase), γ=0.9. After value iteration, the optimal strategy is to prioritize the allocation of fast switching speed (increasing by 20 lux per minute). The generated preliminary plan includes a brightness target of 60 lux for D08 from 18:00 to 19:00, adjusting the switching sequence every 5 minutes, and coordinating with the surrounding area for synchronous optimization.

[0103] In step S16, the preliminary lighting scheme needs to be executed, ambient light change data needs to be collected, and the preliminary lighting scheme needs to be adjusted according to the ambient light change data to obtain the final lighting scheme, including:

[0104] The preliminary lighting scheme is executed, and ambient light change data is collected in real time. When the ambient light change data exceeds the preset light change threshold range, the area exceeding the threshold is marked as the intervention area.

[0105] Extract the current brightness state of the lighting equipment in the intervention area, calculate the difference between the brightness state and the preset target value, adjust the brightness according to the difference value, and obtain a dynamic adjustment command;

[0106] Based on the dynamic adjustment command, the preliminary lighting scheme is adjusted, the final brightness control command is determined, and the final lighting scheme is obtained.

[0107] First, the preliminary lighting scheme is executed, and ambient light change data is collected in real time. When the ambient light change data exceeds the preset light change threshold range, the area exceeding the threshold is marked as an intervention area. This is specifically achieved through real-time data acquisition and threshold comparison. This process includes four key steps: data acquisition, real-time comparison, trigger judgment, and area marking. In the data acquisition step, illuminance values ​​are continuously acquired using an ambient light sensor at a fixed sampling frequency. In the real-time comparison step, a point-by-point comparison algorithm is used to traverse each data point to check the threshold. In the trigger judgment step, the marking condition is activated when the data point is below the lower limit or above the upper limit. In the area marking step, spatial indexing technology is used to map the grid cells exceeding the standard as intervention areas. The light change threshold range is set according to safety standards as a lower limit of 40 lux and an upper limit of 80 lux. The real-time data is matched with the threshold using a point-by-point comparison algorithm. If the data exceeds the range, the area marking mechanism is triggered, and spatial indexing technology is used to mark the corresponding grid cell as an intervention area.

[0108] Then, the current brightness state of the lighting equipment in the intervention area is extracted, and the difference between the brightness state and the preset target value is calculated. The brightness is adjusted according to the difference value to obtain a dynamic adjustment command. Specifically, this is achieved through state monitoring and feedback control. This process includes four key steps: state monitoring, error calculation, control law application, and command generation. The state monitoring step reads the current brightness value from the device controller in real time by setting a fixed sampling frequency (e.g., twice per second). The error calculation step uses an arithmetic subtraction formula. The difference between the target value and the actual value is calculated, where The preset value is 60 lux based on the activity requirements; the control law application is based on a proportional control algorithm, using the formula... Calculate the adjustment amount. The proportional coefficient is identified and determined to be 0.5 by the system; the instruction generation stage converts the adjustment amount into a specific control signal and adds a limiting mechanism to prevent over-adjustment and ensure smooth and stable adjustment.

[0109] Finally, based on the dynamic adjustment instructions, the preliminary lighting scheme is adjusted to determine the final brightness control instructions, resulting in the final lighting scheme. This is achieved through instruction integration and scheme optimization, which includes three core stages: instruction fusion, parameter optimization, and strategy generation. The instruction fusion stage employs a priority scheduling algorithm to integrate the dynamic adjustment instructions into the preliminary scheme according to regional importance and timestamps, resolving instruction conflicts. The parameter optimization stage applies a dynamic programming method, defining the state space as the brightness configuration of each grid region and the decision variable as the brightness adjustment action. The Bellman equation is solved using a value iteration algorithm. Where s represents the system state, a is the adjustment action, and R is the immediate reward function. As a discount factor, For the next state, iterative updates are performed until convergence to obtain the optimal brightness target value and switching sequence. In the strategy generation stage, a structured instruction set containing brightness values, priorities, and linkage rules is constructed based on the optimization results. A dynamic programming method is used to map the adjustment instructions to the parameters of the initial scheme, such as modifying the brightness target value and switching sequence. The final lighting scheme is a structured instruction set containing optimized brightness values, execution priorities, and linkage rules. It is obtained through an iterative adjustment process and is used to directly drive lighting equipment to achieve on-demand lighting, thereby improving energy efficiency and user experience.

[0110] Reference Figure 2 The second embodiment of the present invention provides an intelligent garage light control system with scene switching function, characterized in that it includes:

[0111] The data acquisition module is used to acquire activity data and light intensity data of vehicles and people in the garage, filter high-activity low-light areas based on the activity data and light intensity data, acquire the movement trajectory data of vehicles and people in the high-activity low-light areas, and obtain scene activity information.

[0112] The feature extraction module is used to extract the activity frequency and movement trajectory of each region from the scene activity information and perform cluster analysis to obtain the activity density features of each region;

[0113] The activity assessment module is used to extract the duration of activities in each region and, in conjunction with the activity density characteristics, assess the activity level in each region.

[0114] The demand analysis module is used to filter high-demand lighting areas based on the activity level, identify insufficient brightness areas in the high-demand lighting areas, analyze the lighting demand of the insufficient brightness areas, and obtain demand segmentation results.

[0115] The scheme generation module is used to formulate a preliminary lighting scheme based on the required division results, combined with the pre-acquired grouping strategy and energy-saving information of lighting fixtures;

[0116] The feedback adjustment module is used to execute the preliminary lighting scheme, collect ambient light change data, adjust the preliminary lighting scheme according to the ambient light change data, and obtain the final lighting scheme.

[0117] It should be noted that the intelligent garage light control system with scene switching function provided in this embodiment of the invention is used to execute all the process steps of the intelligent garage light control method with scene switching function in the above embodiment. The working principle and beneficial effect of the two are one-to-one, so they will not be described again.

[0118] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as an instruction deployment program. When the processor executes the computer program, it implements the steps in the above-described embodiments of the intelligent garage light control method with scene switching functionality, for example... Figure 1 The step S11 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above system embodiments, such as the precision matching module.

[0119] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0120] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0121] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0122] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0123] If the modules / units integrated into the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or system capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0124] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0125] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for controlling intelligent garage lights with scene switching function, characterized in that, include: Acquire activity data and light intensity data of vehicles and people in the garage; filter high-activity, low-light areas based on the activity data and light intensity data; acquire movement trajectory data of vehicles and people in the high-activity, low-light areas; and obtain scene activity information. The activity frequency and movement trajectory of each region are extracted from the scene activity information and cluster analysis is performed to obtain the activity density characteristics of each region; Extract the duration of activity in each region, and combine it with the activity density characteristics to assess the activity level in each region; Based on the activity level, high-demand lighting areas are selected, insufficient brightness areas within the high-demand lighting areas are identified, and the lighting needs of the insufficient brightness areas are analyzed to obtain the demand segmentation results. Based on the demand segmentation results, combined with the pre-acquired lighting fixture grouping strategy and energy-saving information, a preliminary lighting plan is formulated; The preliminary lighting scheme is executed, ambient light change data is collected, and the preliminary lighting scheme is adjusted based on the ambient light change data to obtain the final lighting scheme.

2. The intelligent garage light control method with scene switching function according to claim 1, characterized in that, The process involves acquiring activity data of vehicles and people within the garage, as well as light intensity data. Based on this data, high-activity, low-light areas are filtered, and movement trajectory data of vehicles and people within these areas is obtained to acquire scene activity information, including: Collect and integrate activity data of vehicles and people in the garage, as well as light intensity data, to obtain the raw dataset; The garage area is divided into grids, and the original dataset is grouped by grid cells to obtain grid activity data; Extract the frequency of human and vehicle activity and the light intensity of each grid cell from the grid activity data. If the frequency of human and vehicle activity is higher than a preset frequency threshold and the light intensity is lower than a preset light threshold, then mark the grid cell as a high-activity, low-light area. The movement trajectory data of vehicles and people in the high-activity, low-light area are obtained to obtain scene activity information.

3. The intelligent garage light control method with scene switching function according to claim 1, characterized in that, The process of extracting activity frequency and movement trajectory of each region from the scene activity information and performing cluster analysis to obtain activity density characteristics of each region includes: Extract the activity frequency and movement trajectory data of each grid cell from the scene activity information; Cluster analysis is performed on the activity frequency and movement trajectory data, and the activity density characteristics of each grid region are determined based on the clustering results.

4. The intelligent garage light control method with scene switching function according to claim 1, characterized in that, The extraction of activity duration in each region, combined with the activity density characteristics, to assess the activity level in each region includes: Extract the duration of activity in each region and match it with the activity density feature to obtain the correspondence between activity duration and activity density; The triggering conditions for grid activity density are extracted from the correspondence, and the activity density characteristics are combined to determine the activity distribution pattern of each grid region within a specific time period. Based on the activity distribution pattern, assess the activity level of each grid area.

5. The intelligent garage light control method with scene switching function according to claim 1, characterized in that, The process of filtering high-demand lighting areas based on activity levels, identifying insufficiently lit areas within these high-demand areas, analyzing the lighting needs of these insufficiently lit areas, and obtaining demand segmentation results includes: If the activity level is higher than a preset activity threshold, the grid area exceeding the threshold will be marked as a high-demand lighting area; Acquire lighting data within the high-demand lighting area and analyze the matching degree between the activity level and the lighting data to determine areas with insufficient brightness; wherein, the lighting data includes brightness distribution data and ambient light intensity records; Extract the spatial location and time period division information of the insufficient brightness area, combine it with the preset adjustment range data to generate the lighting priority of each area, and determine the lighting demand division result.

6. The intelligent garage light control method with scene switching function according to claim 5, characterized in that, Based on the demand segmentation results, and combined with the pre-acquired lighting fixture grouping strategy and energy-saving information, a preliminary lighting plan is formulated, including: The brightness distribution data of each region is obtained from the lighting demand division results. According to the pre-acquired grouping strategy of lighting fixtures, the regions where the brightness distribution data is lower than the brightness threshold are marked as regions that need adjustment. Obtain the ambient light intensity record of the area to be adjusted. If the ambient light intensity record does not match the preset energy-saving information, adjust the brightness switching speed to obtain the brightness switching speed range of each area. Based on the speed range, priority information for each region is extracted, and the regions with high priority are dynamically adjusted to obtain a preliminary lighting scheme.

7. The intelligent garage light control method with scene switching function according to claim 1, characterized in that, The process of executing the preliminary lighting plan, collecting ambient light change data, adjusting the preliminary lighting plan based on the ambient light change data, and obtaining the final lighting plan includes: The preliminary lighting scheme is executed, and ambient light change data is collected in real time. When the ambient light change data exceeds the preset light change threshold range, the area exceeding the threshold is marked as the intervention area. Extract the current brightness state of the lighting equipment in the intervention area, calculate the difference between the brightness state and the preset target value, adjust the brightness according to the difference value, and obtain a dynamic adjustment command; Based on the dynamic adjustment command, the preliminary lighting scheme is adjusted, the final brightness control command is determined, and the final lighting scheme is obtained.

8. A smart garage light control system with scene switching function, characterized in that, include: The data acquisition module is used to acquire activity data and light intensity data of vehicles and people in the garage, filter high-activity low-light areas based on the activity data and light intensity data, acquire the movement trajectory data of vehicles and people in the high-activity low-light areas, and obtain scene activity information. The feature extraction module is used to extract the activity frequency and movement trajectory of each region from the scene activity information and perform cluster analysis to obtain the activity density features of each region; The activity assessment module is used to extract the duration of activities in each region and, in conjunction with the activity density characteristics, assess the activity level in each region. The demand analysis module is used to filter high-demand lighting areas based on the activity level, identify insufficient brightness areas in the high-demand lighting areas, analyze the lighting demand of the insufficient brightness areas, and obtain demand segmentation results. The scheme generation module is used to formulate a preliminary lighting scheme based on the required division results, combined with the pre-acquired grouping strategy and energy-saving information of lighting fixtures; The feedback adjustment module is used to execute the preliminary lighting scheme, collect ambient light change data, adjust the preliminary lighting scheme according to the ambient light change data, and obtain the final lighting scheme.

Images