Energy-saving light emitting control method and system for multi-spectrum partition annular light source

Abstract

The application relates to the field of photoelectric technology and discloses an energy-saving light-emitting control method and system of a multispectral partition annular light source, the method comprising the following steps: acquiring partition spectrum data and environment signals to generate an initial transmission flow matrix; clustering and grouping spectrum permeation relations to calculate permeation boundaries, and then evaluating light energy distribution fluctuation and determining a flow path; comparing the interaction intensity obtained based on the path with a standard spectrum template to obtain an adjustment coefficient set; determining an optimized partition light intensity configuration according to the coefficient; and finally, fusing real-time flow data to construct a light energy transmission model, and continuously and dynamically regulating light intensity according to the model output. The application solves the problems of uneven illumination and overflow by precisely controlling the light energy flow between partitions, and improves the dynamic adaptability of the system.

Description

Technical Field

[0001] This invention relates to the field of optoelectronic technology, and in particular to an energy-saving luminescence control method and system for a multispectral zoned ring light source. Background Technology

[0002] In a current technology, multispectral zoned ring light sources typically employ independent driving circuits for each luminous zone, allowing for independent adjustment based on preset parameters or simple feedback mechanisms to meet diverse lighting needs. However, due to the geometric characteristics of the ring structure, which result in highly interwoven light paths, this single adjustment struggles to precisely define the boundary influence range, leading to interpenetration between spectral bands and causing unintended energy spillover and uneven distribution. Particularly in dynamic applications, existing methods lack a quantitative understanding of the flow relationships between zones, hindering precise intensity allocation and real-time response, thus limiting system adaptability and reliability.

[0003] In summary, existing technologies are insufficient for precisely controlling the light energy transmission flow between different zones of a multispectral ring light source, and cannot effectively meet the real-time response requirements of dynamic light intensity distribution, resulting in insufficient system adaptability and reliability. Summary of the Invention

[0004] This invention provides an energy-saving luminescence control method and system for a multispectral zoned ring light source, so as to effectively meet the real-time response requirements of dynamic light intensity distribution and improve the dynamic adaptability of the system.

[0005] In a first aspect, to solve the above-mentioned technical problems, the present invention provides an energy-saving luminous emission control method for a multispectral zoned ring light source, comprising:

[0006] Acquire the partitioned spectral data and environmental signals of the multispectral ring light source, and perform data fusion processing to generate an initial transmission flow matrix;

[0007] Based on the initial transmission flow matrix, the spectral penetration relationships between partitions are clustered and grouped. The boundary features of the spectral penetration relationships are calculated based on the grouping results to obtain a set of penetration boundaries.

[0008] If the permeation parameters in the permeation boundary set exceed the preset permeation threshold, then the light energy distribution characteristics are extracted and the volatility of the light energy distribution characteristics is evaluated to determine the flow path in the light energy distribution area with large volatility.

[0009] Based on the traffic path, the interaction intensity data between adjacent partitions is obtained and compared with a preset standard spectral template to obtain a set of adjustment coefficients;

[0010] Based on the set of adjustment coefficients, the transmission traffic is allocated in real time to determine the optimized partition optical intensity configuration;

[0011] Real-time traffic data is acquired, the environmental signal is fused with the real-time traffic data, and based on the fusion result, the optimized partition light intensity configuration is iteratively adjusted to obtain a stable partition light intensity configuration.

[0012] Secondly, the present invention provides an energy-saving luminous control system for a multispectral zoned ring light source, comprising:

[0013] Initial data processing module: acquires the partitioned spectral data and environmental signals of the multispectral ring light source, performs data fusion processing, and generates the initial transmission flow matrix;

[0014] Spectral penetration analysis module: Based on the initial transmission flow matrix, the spectral penetration relationships between partitions are clustered and grouped, and the boundary features of the spectral penetration relationships are calculated to obtain a set of penetration boundaries;

[0015] Flow path determination module: If the permeation parameters in the permeation boundary set exceed the preset permeation threshold, then extract the light energy distribution characteristics and evaluate the fluctuation of the light energy distribution characteristics to determine the flow path in the light energy distribution area with large fluctuations.

[0016] Light intensity coefficient calculation module: Based on the traffic path, it obtains the interaction intensity data between adjacent partitions and compares it with a preset standard spectral template to obtain a set of adjustment coefficients;

[0017] Partition optical intensity optimization module: Based on the set of adjustment coefficients, it allocates the transmission traffic in real time and determines the optimized partition optical intensity configuration;

[0018] Light intensity iterative adjustment module: acquires real-time traffic data, fuses the environmental signal with the real-time traffic data, and iteratively adjusts the optimized partition light intensity configuration based on the fusion result to obtain a stable partition light intensity configuration.

[0019] 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 energy-saving luminous emission control method of the multispectral partitioned ring light source described in any one of the above.

[0020] 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 perform the energy-saving luminous emission control method of the multispectral partitioned ring light source described in any one of the above-described methods.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] (1) This invention acquires the partitioned spectral data and environmental signals of a multispectral ring light source and performs data fusion processing to generate an initial transmission flux matrix. Then, based on this matrix, it clusters and groups the spectral penetration relationships between partitions and calculates the boundary features of these relationships to obtain a set of penetration boundaries. This process overcomes the problem in existing technologies where the highly intertwined light paths make it difficult to accurately define the boundary influence range. Through multi-source data fusion and cluster analysis, it achieves the quantification of the light energy transmission flux relationship between each luminous partition and the precise identification of spectral penetration boundaries. This process can effectively define and quantify the light energy transmission boundaries between each luminous partition of the multispectral ring light source, solving the technical problem in existing technologies where uneven light energy distribution and unexpected spillover are difficult to precisely control.

[0023] (2) This invention extracts light energy distribution characteristics and evaluates their volatility when the permeation parameters in the permeation boundary set exceed a preset threshold, thereby determining the flow path in the light energy distribution area with large volatility. This method uses a convolutional neural network to capture the spatial characteristics of light energy distribution and quantify its volatility, which can accurately locate specific areas and flow paths where light energy transmission is unstable and prone to overflow. It makes up for the shortcomings of existing technologies in quantifying the flow relationship between intervals, effectively identifies potential light energy overflow areas and problematic flow paths, and provides a data foundation and clear control targets for subsequent accurate light intensity allocation and real-time response.

[0024] (3) Based on the determined flow path with large fluctuations, this invention obtains the interaction intensity data between adjacent partitions and compares it with a preset standard spectral template to obtain a set of adjustment coefficients. According to this set of adjustment coefficients, the transmission flow is allocated in real time, and the optimized partition light intensity configuration is finally determined. This data-driven feedback adjustment mechanism generates fine adjustment coefficients by accurately quantifying the deviation between the light energy interaction intensity and the standard template. This allows for optimization based on the actual light energy penetration and overflow situation, rather than simply adjusting independently. It achieves real-time and accurate allocation of transmission flow in each luminous partition of the multispectral ring light source, effectively overcoming the technical difficulty of achieving accurate light intensity allocation in existing technologies and ensuring the uniformity of light energy distribution.

[0025] (4) This invention acquires real-time traffic data, fuses the environmental signal with the real-time traffic data, and iteratively adjusts the optimized partitioned light intensity configuration based on the fusion result, thereby constructing a light energy transmission model. Subsequently, the output parameters of the light energy transmission model are continuously monitored, and the optimized partitioned light intensity configuration is dynamically adjusted according to the monitoring results. This method, through closed-loop iterative optimization and dynamic monitoring and control, enables the system to respond quickly and adaptively to traffic fluctuations caused by external and internal factors such as environmental changes and light source aging, significantly improving the system's adaptability. It constructs a light energy transmission model that can respond to real-time dynamic needs, effectively addressing the real-time response requirements of dynamic light intensity distribution, and fundamentally improving the adaptability and reliability of the multispectral ring light source system in dynamic application scenarios. Attached Figure Description

[0026] Figure 1 This is a schematic flowchart of the energy-saving luminescence control method for a multispectral partitioned ring light source provided in the first embodiment of the present invention;

[0027] Figure 2 This is a schematic diagram of the energy-saving light-emitting control system for a multispectral zoned ring light source provided in the second embodiment of the present invention. Detailed Implementation

[0028] 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.

[0029] Reference Figure 1 The first embodiment of the present invention provides an energy-saving luminous emission control method for a multispectral zoned ring light source, comprising the following steps:

[0030] S11: Acquire the partitioned spectral data and environmental signals of the multispectral ring light source, and perform data fusion processing to generate the initial transmission flow matrix;

[0031] S12, based on the initial transmission flow matrix, the spectral penetration relationships between partitions are clustered and grouped, and the boundary features of the spectral penetration relationships are calculated according to the grouping results to obtain a set of penetration boundaries;

[0032] S13, if the permeation parameters in the permeation boundary set exceed the preset permeation threshold, then extract the light energy distribution characteristics and evaluate the volatility of the light energy distribution characteristics to determine the flow path in the light energy distribution area with large volatility.

[0033] S14. Based on the traffic path, obtain the interaction intensity data between adjacent partitions and compare it with the preset standard spectral template to obtain the set of adjustment coefficients;

[0034] S15, Based on the set of adjustment coefficients, allocate the transmission traffic in real time and determine the optimized partition optical intensity configuration;

[0035] S16, acquire real-time traffic data, fuse the environmental signal with the real-time traffic data, and iteratively adjust the optimized partition light intensity configuration based on the fusion result to obtain a stable partition light intensity configuration.

[0036] It is worth noting that in this invention, the term "flow rate" mainly refers to the physical quantity of light energy transmission in different partitions or paths, i.e., luminous energy flux, which can be measured in photons per second or lumens, etc. Therefore, the "transmission flux matrix" used in the following steps specifically refers to the "luminous energy flux matrix".

[0037] In step S11, the partitioned spectral data and environmental signals of the multispectral ring light source are acquired, and data fusion processing is performed to generate an initial transmission flow matrix.

[0038] It should be noted that, in the first embodiment of the present invention, this processing step specifically includes steps S111 to S114:

[0039] S111: Obtain the partitioned spectral data and environmental signals of the multispectral ring light source to obtain the original spectral dataset and environmental signal dataset;

[0040] S112, if the original spectral dataset and environmental signal dataset contain noise, then the original spectral dataset and environmental signal dataset are denoised to obtain denoised spectral dataset and environmental signal dataset.

[0041] S113, Based on the denoised spectral dataset and environmental signal dataset, the partitioned spectral data and environmental signals are integrated to obtain a comprehensive spectral feature dataset;

[0042] S114. Based on the comprehensive spectral feature dataset, calculate the initial transport flow matrix and determine the spectral partition and permeation parameters corresponding to each element in the matrix.

[0043] In step S111, the partitioned spectral data and environmental signals of the multispectral ring light source are acquired to obtain the original spectral dataset and the environmental signal dataset.

[0044] Specifically, data is acquired in real time through a sensor array deployed on a multispectral ring light source. This sensor array, for example, consists of eight spectral partition sensors and environmental sensors for temperature, humidity, etc., with each spectral partition corresponding to a specific wavelength band (e.g., 400nm, 550nm, 700nm, etc.). The sensor array operates continuously at a preset sampling frequency (e.g., 100Hz) to ensure the capture of dynamic changes in light intensity distribution and band penetration parameters. Within one sampling period, the acquired data constitutes the original spectral dataset (e.g., light intensity of 1200 lux for the 400nm partition) and the environmental signal dataset (e.g., temperature of 25°C, humidity of 60%). It is worth noting that the sampling frequency setting must fully consider the frequency characteristics of the measured signal and follow the Nyquist sampling theorem (i.e., the sampling frequency must be greater than twice the highest frequency of the signal) to avoid signal distortion. In this invention, to ensure acquisition quality, a coefficient higher than twice is typically used. For example, in precision manufacturing scenarios, if the main frequency components of the dynamic changes in light intensity are within 15Hz, and the highest frequency of environmental interference (such as machine tool vibration in a workshop) is approximately 25Hz, then the highest total frequency of the signal is 40Hz. In this case, setting the sampling frequency to 100Hz meets the requirement of 2.5 times the highest total frequency, ensuring distortion-free acquisition. In medical lighting scenarios, if the light source response time is approximately 20ms (corresponding to an equivalent bandwidth of approximately 17.5Hz), and the highest frequency of environmental interference (such as electromagnetic radiation from medical equipment) is approximately 20Hz, then the highest total frequency of the signal is 37.5Hz. In this case, the sampling frequency can be set accordingly to 94Hz. A general principle for choosing the sampling frequency is that it should be 2.5 times the sum of the highest frequency of the dynamic changes in the light source and the highest frequency of environmental interference; in practice, setting it to 100Hz can meet the requirements while leaving a margin.

[0045] In step S112, if the original spectral dataset and environmental signal dataset contain noise, then the original spectral dataset and environmental signal dataset are denoised to obtain denoised spectral dataset and environmental signal dataset.

[0046] Specifically, a median filter algorithm is used to smooth noisy datasets. The algorithm works by iterating through the signal sequence using a sliding window (e.g., containing five consecutive sampling points), sorting the sampled values ​​within the window, and replacing the original value at the center of the window with the sorted median. This nonlinear filtering method is particularly effective at filtering out impulse noise or sudden changes in sensor readings (i.e., "jump noise"). For example, if five consecutive light intensity samples in a 400nm region are [1200, 1190, 2000, 1210, 1205] lux, where 2000 lux is an abnormal jump value, after median filtering, the output value at that point will be corrected to the sorted median of 1205 lux, effectively removing noise interference and obtaining a smoother, more representative denoised dataset.

[0047] It should be noted that the choice of this window size (e.g., 5 sampling points) is the result of a trade-off between noise suppression capability and signal detail preservation.

[0048] Specifically, based on the analysis of the noise characteristics of the sensors used in this system, it was found that most of the sudden noise lasts for no more than two sampling periods. Using a window size of five sampling points ensures that noise pulses lasting one to two sampling points are effectively filtered out (because the noise points will not become the median after sorting), while minimizing the smoothing effect on the original rapid changes in the signal, thus preserving the signal's authenticity.

[0049] In step S113, based on the denoised spectral dataset and the environmental signal dataset, the partitioned spectral data and the environmental signal are integrated to obtain a comprehensive spectral feature dataset.

[0050] Specifically, a weighted average fusion algorithm is used to integrate multi-source data. This algorithm assigns different weights to different data sources based on their relative importance or reliability, and then performs a weighted summation to generate a comprehensive feature description. The weights are determined by performing multiple linear regression analysis on a historical dataset. This historical dataset contains a large amount of multidimensional input data (i.e., light intensity in various spectral zones, environmental signals, etc.) and their corresponding known key performance indicators that can reflect lighting quality or detection effectiveness (e.g., the "product surface defect level" score obtained through high-precision offline detection). By training a regression model with multidimensional input data as independent variables and defect level scores as dependent variables, the contribution of each input variable to the final result, i.e., the regression coefficient, can be obtained. Normalizing the absolute values ​​of these regression coefficients (making their sum equal to 1) yields a set of objective, quantitative weights. For example, if the regression analysis shows that 400nm light intensity contributes the most to defect detection, and its normalized coefficient is 0.4, then its weight is set to 0.4. This method transforms the two vague concepts of "sensitivity" and "potential impact" into a data-driven, reproducible mathematical process.

[0051] The fused result is compared with a preset reasonableness threshold range (e.g., the average light intensity should be within the range of 1000-1500 lux) to verify the validity of the data. This threshold range is determined based on long-term statistical analysis of normal operating data under specific application scenarios. Through this step, multi-dimensional discrete information can be integrated into a more robust comprehensive spectral feature dataset.

[0052] In step S114, based on the comprehensive spectral feature dataset, an initial transport flow matrix is ​​calculated, and the spectral partition and permeation parameters corresponding to each element in the matrix are determined.

[0053] Specifically, before constructing the matrix, a comprehensive spectral eigenvalue (F) needs to be calculated, which integrates the light intensity of each spectral region and the influence of key environmental parameters. In an optional embodiment, this comprehensive spectral eigenvalue F can be calculated using the following weighted summation formula:

[0054]

[0055] In this formula, Represents the total number of spectral partitions. It is the ratio of the current light intensity value of the i-th spectral partition to the preset reference light intensity, and That is its corresponding weighting coefficient; It is a calibration constant with a unit of light intensity (such as lux), for example, it can be the average of the reference light intensities for all zones. This spectral weight... The setting is based on the sensitivity of each spectral band to a specific detection task. This sensitivity can be quantified experimentally (e.g., by testing the detection accuracy under different bands through analysis of variance). For example, if the 400nm band is most sensitive to defect detection, its weight... It can be set to a higher value of 0.4. Similarly, This represents the total number of environmental parameters that affect light energy transmission. It is the ratio of the j-th environmental parameter (e.g., temperature, humidity) to its standard operating condition reference value. This is its corresponding weighting coefficient. This environmental weight... The setting is based on the influence coefficient of this environmental parameter on light intensity distribution, which can be measured and determined through controlled variable experiments; for example, if humidity has a significant impact on light intensity, its weight... It can be set to 0.2. The reference light intensity and standard operating condition reference value are determined based on the equipment factory calibration or baseline testing under specific application scenarios. For example, when the light source is first installed and operated stably in a standard working environment (e.g., temperature 25°C, humidity 50%, no external light interference), the light intensity value and environmental parameter value of each spectral zone measured and recorded are set as their respective reference values.

[0056] After calculating the comprehensive spectral characteristic value F, an initial transport flux matrix can be constructed. The structure of this matrix is ​​predefined; for example, a 3×3 matrix where row vectors correspond to different spectral partitions (e.g., 400nm, 550nm, 700nm), and column vectors correspond to key physical permeation parameters (e.g., transmittance, scattering, reflectance). Each element value in the matrix is ​​calculated by combining the calculated comprehensive spectral characteristic value F with the permeation parameter corresponding to that column. For example, the element value in [row 1, column 1] can be calculated by multiplying the comprehensive light intensity of the 400nm partition by the transmittance of that partition (e.g., 1300 lux × 0.8 = 1040). The resulting initial transport flux matrix structures and quantifies the optical characteristics of each spectral partition, providing clear and well-organized input data for subsequent spectral analysis and decision-making.

[0057] In step S12, based on the initial transmission flow matrix, the spectral penetration relationships between partitions are clustered and grouped, and the boundary features of the spectral penetration relationships are calculated according to the grouping results to obtain a set of penetration boundaries.

[0058] Specifically, this step first employs an unsupervised machine learning algorithm, specifically K-means clustering, to group the spectral partition features contained in the initial transmission flow matrix. The K-means algorithm iteratively divides the feature vectors representing each spectral partition (e.g., composed of parameters such as light intensity, transmittance, and scattering rate) into a predetermined number (K) of clusters. Its goal is to minimize the sum of squared distances from each data point within a cluster to the centroid of that cluster. The predetermined number of clusters, K (e.g., K=2), is determined based on the number of typical feature patterns in a specific industrial inspection scenario, or optimized through contour coefficient analysis of historical data. Through clustering, spectral partitions with similar permeability characteristics can be grouped together; for example, one group could be designated as "high transmittance partitions," and another as "medium transmittance partitions," thus obtaining a structured set of groups.

[0059] Subsequently, based on this grouping set, a boundary feature extraction algorithm is used to quantify the differences between different groups. The core of this algorithm is to identify and calculate the difference in characteristic parameters between spectral partitions that are spectrally adjacent but belong to different clusters. For example, if the 400nm partition (transmittance of 0.8) is classified into the "high transmittance" cluster, while its adjacent 550nm partition (transmittance of 0.7) is classified into the "medium transmittance" cluster, the algorithm will calculate the boundary feature value between these two partitions in the transmittance dimension, i.e., the difference is 0.1. All these calculated boundary feature values ​​together constitute the permeability boundary set, which intuitively quantifies the transition region and the degree of change in the spectral permeability relationship, providing a precise data foundation for subsequent analysis.

[0060] In step S13, if the permeation parameters in the permeation boundary set exceed a preset permeation threshold, then the light energy distribution characteristics are extracted and the volatility of the light energy distribution characteristics is evaluated to determine the flow path in the light energy distribution area with large volatility.

[0061] It should be noted that, in the first embodiment of the present invention, this processing step specifically includes steps S131 to S134:

[0062] S131, Obtain boundary point distribution data from the permeation boundary set, and extract features from the boundary point distribution data to obtain light energy distribution features;

[0063] S132, Based on the light energy distribution characteristics, calculate the fluctuation of the characteristics and determine the light energy distribution area with large fluctuation;

[0064] S133, if the penetration parameter of the highly volatile light energy distribution area exceeds the preset penetration threshold, the flow path in the area is divided to obtain a set of candidate flow paths;

[0065] S134, filter the candidate traffic path set to determine the traffic path with high volatility.

[0066] In step S131, boundary point distribution data is obtained from the set of permeable boundaries, and feature extraction is performed on the boundary point distribution data to obtain light energy distribution features.

[0067] Specifically, this step employs a Convolutional Neural Network (CNN) to automatically extract deep features from the boundary point distribution data. First, to transform the discrete spatial distribution data of boundary points into a format suitable for CNN processing, data preprocessing is required. Using a gridding method, the physical space of the boundary region (e.g., an arc-shaped region containing 5 spectral partitions with a physical size of 10cm x 2cm) is proportionally discretized onto a pre-defined two-dimensional standard matrix (e.g., a 32×32 matrix). In this way, the data of each boundary point becomes a pixel in the matrix, and different physical quantities can serve as different channels for that pixel, forming a multi-channel, image-like input tensor. Each grid cell (pixel) corresponds to a fixed physical size (approximately 3.1mm x 0.6mm in this example), and its value represents the average value of the physical quantity in that small region.

[0068] The CNN model is designed to extract spatial features from the input tensor. The network structure can be designed as follows: the input layer receives the preprocessed multi-channel two-dimensional matrix. Subsequently, the data passes through two convolutional layers (e.g., using 3x3 kernels with a stride of 1 and a ReLU activation function) to effectively extract local spatial features of the boundary regions, such as light intensity gradients and permeability texture patterns. Each convolutional layer can be followed by a max-pooling layer (e.g., 2x2) to reduce the dimensionality of the feature map, decrease computation, and extract the most salient features. After convolution and pooling operations, a flattening layer transforms the multidimensional feature map into a one-dimensional vector, which is then connected to two fully connected layers to output a fixed-dimensional feature vector, representing the light energy distribution features.

[0069] The training process of the CNN model is supervised learning. Training data originates from historical boundary point data accumulated over a long period across various industrial inspection scenarios. The training data is also derived from simulated data collection in a standardized experimental environment for specific industrial inspection applications (e.g., surface defect detection of precision optical components). The annotation process is as follows: First, for each collected boundary point distribution data sample, a high-precision goniophotometer is used to perform a fine scan of the boundary region to obtain a high-resolution, ground-truth light energy distribution map. Then, a set of key characteristic parameters of this light energy distribution map (e.g., spatial frequency of light intensity distribution, average gradient, kurtosis coefficient, etc.) are calculated using a physical model, and the vector formed by these parameters is used as the "true" light energy distribution feature label for the data sample. Before model training, the dataset is divided into, for example, 70% (training set), 15% (validation set), and 15% (test set). During training, the boundary point data in the training set is used as input, and the corresponding feature labels are used as the target output. Mean Squared Error (MSE) is used as the loss function, and the Adam optimizer is used for iterative optimization of the model parameters. To ensure stable and efficient training, an initial learning rate of 0.001 is set, and a learning rate decay strategy is employed. Mini-batch gradient descent is used, and the batch size can be set to, for example, 64. Training terminates when the model's loss on the validation set no longer decreases significantly for several consecutive epochs (e.g., 10 epochs) to prevent overfitting. The trained model can automatically and robustly extract deep-level light energy distribution features from new and unknown boundary point distribution data.

[0070] In step S132, based on the light energy distribution characteristics, the fluctuation of the characteristics is calculated, and the light energy distribution area with large fluctuation is determined.

[0071] Specifically, this step employs statistical analysis methods, particularly calculating the standard deviation, to quantify the volatility of the light energy distribution characteristics. Standard deviation is a core indicator for measuring the dispersion of a dataset; a larger value indicates greater data volatility. For example, for light intensity data at two adjacent boundary points extracted from the feature map (e.g., 1300 lux and 1500 lux), the calculated standard deviation is approximately 141 lux. By comparing the calculated standard deviation with a preset volatility threshold (e.g., 100 lux), regions with significant volatility can be objectively identified. The preset volatility threshold can be determined through... The criteria are determined based on statistical analysis of a large amount of historical light energy distribution characteristic data under "steady-state" conditions. The specific process is as follows: First, a dataset of light energy distribution characteristics under ideal, undisturbed conditions when the light source is operating stably is collected, and the mean μ and standard deviation σ of all characteristic standard deviations in this dataset are calculated; then, the volatility threshold is set to... According to the normal distribution theory, this threshold defines a high-confidence interval. Any fluctuation exceeding this threshold (e.g., drastic changes caused by light energy spillover or external interference) can be considered a significant fluctuation requiring attention, rather than normal random noise. Those skilled in the art will understand that... This is a commonly used empirical setting. In scenarios where the sensitivity to volatility detection is more stringent or less stringent, the coefficient "3" can be adjusted within a range of, for example, 2 to 4, depending on the actual needs. If the characteristic standard deviation of a region exceeds this threshold, the region is identified as a region with highly volatile light energy distribution.

[0072] In step S133, if the permeation parameter of the highly volatile light energy distribution area exceeds a preset permeation threshold, the flow paths within the area are divided to obtain a set of candidate flow paths.

[0073] Specifically, the process begins by assessing the identified areas with highly fluctuating light energy distribution. This involves checking whether the key penetration parameters (e.g., average transmittance) of this area exceed a preset penetration threshold (e.g., a transmittance threshold of 0.75). This preset penetration threshold is determined based on the physical signal-to-noise ratio (SNR) requirements of the end application. In applications such as precision inspection, to ensure reliable identification of minute defects, the light signal received by the imaging sensor must achieve a certain SNR, and the light energy penetration parameters (such as transmittance) directly determine the final signal strength. Therefore, by analyzing the minimum SNR required for a specific inspection task (e.g., detecting scratches smaller than 10 micrometers), the minimum effective light energy transmission efficiency that the light source must achieve can be calculated in reverse; this efficiency value is set as the penetration threshold. Areas below this threshold, even if they exhibit internal light intensity fluctuations, will be ignored due to insufficient overall light energy output and inability to form an effective signal, thus avoiding wasting computational resources on areas without practical control value. If this condition is met, a path planning algorithm is initiated to process the area. To apply this algorithm, the light intensity distribution within the region must first be modeled as a weighted graph. This process begins by discretizing the region into an M×N grid, defining the center point of each grid as a graph node, with each node associated with the average light intensity value at its location. Secondly, the edges of the graph are defined as connecting two adjacent nodes in space. For example, each internal node can be connected to its four adjacent nodes in the four directions above, below, left, and right, forming a four-connected graph structure. Finally, the edge weights, i.e., the path cost function, are designed to be inversely proportional to the light intensity gradient. If the light intensity I(u) of node u is greater than that of node v when traveling from node u to its adjacent node v, then the path cost is proportional to the reciprocal of the light intensity difference (I(u) - I(v)), encouraging light energy to propagate along the direction of fastest descent. Conversely, if the light intensity does not decrease, a very large cost value is set to prevent light energy from backtracking to regions with equal or higher light intensity.

[0074] Based on this well-defined weighted graph, algorithms such as Dijkstra's algorithm or A* algorithm, which are based on minimum path cost, can be used to find the path with the highest light intensity in the region as the starting point and the lowest light intensity as the ending point. The minimum cost path found by the algorithm simulates the most likely propagation or flow direction of light energy in that region, thereby generating a set of candidate flow paths that conform to physical laws.

[0075] In step S134, the candidate traffic path set is filtered to determine the traffic path with high volatility.

[0076] Specifically, this step uses a comparative analysis screening method to determine the final flow path for regulation from a set of candidate flow paths. The core criterion for screening is maximizing volatility. By recalculating the volatility of light intensity data on each candidate path (e.g., calculating the standard deviation of light intensity at all points on the path), a quantified volatility score can be assigned to each path. For example, path A has a volatility score of 150 lux, while path B has a score of 100 lux. Furthermore, to enhance the accuracy of the decision, environmental signals can be introduced for correlation analysis, prioritizing paths whose volatility shows a strong correlation with key environmental factors (e.g., humidity reaching 65%). Finally, the path with the highest volatility score (path A in this example) is selected as the final flow path because it best represents the instability of light energy distribution and its sensitivity to external disturbances, which is a key objective for energy-saving regulation.

[0077] In step S14, based on the traffic path, the interaction intensity data between adjacent partitions is obtained and compared with a preset standard spectral template to obtain a set of adjustment coefficients. It should be noted that, in the first embodiment of the present invention, this processing step specifically includes steps S141 to S143:

[0078] S141, Obtain interaction intensity data between adjacent partitions from the traffic path, and perform spectral analysis on the interaction intensity data to obtain spectral feature distribution;

[0079] S142, if the deviation between the spectral feature distribution and the preset standard spectral template exceeds the preset chromaticity threshold, then color distortion is determined, and the spectral features of the distorted region are identified.

[0080] S143, Based on the spectral characteristics of the distorted region, match it with the preset standard spectral template to obtain a set of adjustment coefficients.

[0081] In step S141, interaction intensity data between adjacent partitions is obtained from the traffic path, and spectral analysis is performed on the interaction intensity data to obtain the spectral feature distribution.

[0082] Specifically, firstly, using a high-precision light intensity sensor or spectrometer, the light energy exchange at the boundary of adjacent spectral partitions along the flow path determined in step S13 is measured to obtain interaction intensity data. This interaction intensity can be quantified as the light intensity difference (e.g., 1200 lux - 1100 lux = 100 lux) at the boundary point between adjacent partitions (e.g., 400 nm and 550 nm partitions). Subsequently, the acquired interaction intensity data (which may be a time series) is processed using the Fast Fourier Transform (FFT) algorithm. The principle of the FFT algorithm is to decompose complex time-domain or spatial-domain signals into their basic frequency components, thereby converting the dynamic changes in interaction intensity into a frequency-domain representation, i.e., the spectral feature distribution. This distribution can reveal the hidden periodic features and main frequency components in the light energy interaction.

[0083] In step S142, if the deviation between the spectral feature distribution and the preset standard spectral template exceeds a preset chromaticity threshold, then color distortion is determined, and the spectral features of the distorted region are identified.

[0084] Specifically, this step uses an objective colorimetric comparison method to determine whether color distortion exists. First, the corresponding chromaticity coordinates (e.g., CIE 1931 x,y chromaticity coordinates) are calculated based on the spectral characteristic distribution obtained in step S141. Then, the calculated chromaticity coordinates are compared with the standard chromaticity coordinates defined by a preset standard spectral template, and the Euclidean distance or difference between the two is calculated. For example, if the chromaticity coordinates of the standard template are (x=0.31, y=0.32), while the actually measured coordinates are (x=0.33, y=0.35), the deviation will be quantified. This deviation will be compared with a preset chromaticity threshold (e.g., 0.02), which is set based on the minimum perceptible difference of color difference for the human eye or the accuracy requirements for color consistency in specific industrial applications. If the deviation exceeds this threshold, color distortion is determined to exist. At this point, the system will further analyze the spectral components that cause the color deviation, for example, determine that it is caused by an abnormal increase in light intensity in the 550nm band, and record this information as the spectral characteristics of the distorted region.

[0085] In step S143, the spectral characteristics of the distorted region are matched with the preset standard spectral template to obtain a set of adjustment coefficients.

[0086] Specifically, this step employs a Proportional-Integral-Derivative (PID) control algorithm derived from classical control theory to dynamically calculate the adjustment coefficients. This algorithm can generate correction commands quickly, stably, and accurately. In terms of its working principle, the algorithm first calculates the error by comparing the current spectral characteristics of the distorted region with the target characteristics in a standard spectral template, thus calculating the deviation between the two. This error represents the degree of deviation between the current state and the ideal state. Next, as the core correction step, the algorithm calculates the proportional (P) term. It calculates a proportional adjustment amount based on the magnitude of the current error. The adjustment magnitude is controlled by a preset proportional gain coefficient Kp, and the proportional term's role is to quickly respond and eliminate most of the error. To further improve control accuracy, the algorithm can also introduce integral (I) and derivative (D) terms as optional optimizations. The integral term is used to eliminate small steady-state errors caused by the inherent characteristics of the system; the derivative term is used to predict the trend of error changes, intervening in advance to prevent overshoot and oscillations during the adjustment process, making the adjustment process smoother. The effects of these three factors are ultimately weighted and summed using the classic PID control formula to obtain the total adjustment. This formula is expressed as:

[0087]

[0088] in, It is an error signal that varies over time, and among which, It is the proportional gain, a dimensionless number used to set the overall response strength of the control action; It is the integral time, measured in "time," used to adjust the speed at which the system eliminates steady-state errors. Smaller... A value implies a stronger integral effect; while It is the differential time, also measured in "time", used to predict the trend of error changes, suppress overshoot and oscillation of the system, and thus improve stability.

[0089] It should be noted that the gain coefficient The optimal gain coefficients were determined through system model simulation optimization. Specifically, a mathematical model was first established to accurately describe the dynamic response characteristics of the light source system. Then, in simulation software (e.g., MATLAB / Simulink), multiple different combinations of gain coefficients were tested to evaluate the system's performance metrics (e.g., response speed, overshoot, steady-state error) under these combinations. Finally, a set of gain coefficients was selected that, while ensuring system stability, achieved the fastest response speed and an overshoot below a preset value (e.g., below 5%). The value is used as a fixed gain coefficient.

[0090] The total adjustment This is then converted into specific adjustment coefficients. For example, if distortion analysis determines that the light intensity in the 550nm band is 10% higher than the standard value, the PID controller will calculate the corresponding... This is then converted into a multiplicative light intensity attenuation factor, for example, 0.9. Similarly, if the analysis indicates that the transmittance needs to be increased by 0.05, the algorithm will calculate the corresponding transmittance gain factor, for example, 1.05. All these coefficients calculated for different parameters together constitute the aforementioned set of adjustment coefficients.

[0091] In step S15, the transmission traffic is allocated in real time according to the set of adjustment coefficients to determine the optimized partition optical intensity configuration. It should be noted that, in the first embodiment of the present invention, this processing step specifically includes steps S151 to S153:

[0092] S151, Obtain traffic data from the traffic path, process the traffic data, and analyze the traffic distribution characteristics between partitions;

[0093] S152, Based on the set of adjustment coefficients and the traffic distribution characteristics, a feedback loop mechanism is used to allocate the transmission traffic in real time to obtain a preliminary optical intensity allocation scheme;

[0094] S153, if the deviation between the preliminary light intensity allocation scheme and the preset light intensity standard template exceeds the preset light intensity threshold, then the allocation ratio of the transmission traffic is adjusted to determine the optimized partition light intensity configuration.

[0095] In step S151, traffic data is obtained from the traffic path and processed to analyze the traffic distribution characteristics between partitions.

[0096] Specifically, high-sampling-rate light intensity sensors deployed along highly volatile flow paths are used to collect real-time light energy transfer flow data (e.g., measured in photons per second) at the boundaries of partitions. Subsequently, time-series analysis is employed, specifically through a sliding window technique, to process the continuously acquired flow data. This technique divides the data stream into continuous, overlapping time segments (windows) and calculates statistical characteristics for the data within each window. For example, by analyzing the data within a window, the peak flow rate of the 400nm partition can be determined to be 1200 photons, while the valley flow rate of the adjacent 550nm partition is 700 photons. This analytical method reveals the dynamic imbalances in light energy transfer over short periods, thus providing a precise description of the flow distribution characteristics between partitions.

[0097] In step S152, based on the set of adjustment coefficients and the traffic distribution characteristics, a feedback loop mechanism is used to allocate the transmission traffic in real time to obtain a preliminary optical intensity allocation scheme.

[0098] Specifically, this step establishes a closed-loop feedback control mechanism. This mechanism uses the set of adjustment coefficients generated in step S143 as the target setpoint of the control system and the real-time flow distribution characteristics obtained in step S151 as process feedback variables. The control system iterates at a fixed time period (e.g., every 0.5 seconds), continuously comparing the feedback variables with the setpoint. If a deviation is detected (e.g., the actual flow rate of the 550nm zone is lower than the target value indicated by the adjustment coefficients), the feedback algorithm immediately generates a compensation command to dynamically adjust the light source driving power of that zone, for example, increasing its light intensity to 1100 lux. This instantly adjusted light source state constitutes the preliminary light intensity allocation scheme.

[0099] In step S153, if the deviation between the preliminary light intensity allocation scheme and the preset light intensity standard template exceeds the preset light intensity threshold, the allocation ratio of the transmission traffic is adjusted to determine the optimized partition light intensity configuration.

[0100] Specifically, this step is a verification and fine-tuning of the initial scheme. The initial light intensity allocation scheme obtained in step S152 (e.g., 1100 lux for the 550nm partition) is compared with a more stringent, preset light intensity standard template (e.g., requiring 1000 lux), and their relative deviation is calculated. This deviation is compared with a preset light intensity tolerance threshold (e.g., 8%), which is set according to the process accuracy requirements of the end application. If the deviation exceeds this threshold (10% > 8% in this example), a quadratic programming (QP) optimization algorithm is initiated to calculate the final optimized configuration. Quadratic programming is a mathematical method for finding the optimal solution of a multivariable function under a set of linear constraints. In this step, the optimization objective is defined as minimizing the squared Euclidean distance between the adjusted partition light intensity configuration and the light intensity standard template (i.e., the sum of the squares of the light intensity differences of all partitions). This objective ensures that the final configuration is as close as possible to the standard template overall. Secondly, the constraints are formalized: for example, by setting the sum of the light intensities of all zones to equal a constant value (i.e., the total luminous flux before adjustment), it is transformed into a linear equality constraint to maintain the stability of the overall energy. Simultaneously, the requirement that the light intensity of each zone must be within its effective hardware operating range (e.g., minimum and maximum brightness) is transformed into a linear inequality constraint. Based on this, standard convex optimization algorithms, such as the Interior Point Method or the Active Set Method, can be used to efficiently solve this quadratic programming problem. The solver will calculate a unique set of zone light intensity values ​​that minimize the objective function while satisfying all the above linear constraints. For example, the algorithm's calculation result might determine to adjust the power of the 550nm zone to 1005 lux, and make corresponding fine-tuning of the power of other adjacent zones such as 400nm, to minimize the overall deviation while maintaining the constraint of constant total luminous flux. This globally optimized scheme solved by the QP algorithm is the optimized zone light intensity configuration.

[0101] In step S16, real-time traffic data is acquired, the environmental signal and the real-time traffic data are fused, and based on the fusion result, the optimized partitioned light intensity configuration is iteratively adjusted to obtain a stable partitioned light intensity configuration. It should be noted that, in one embodiment of the present invention, this processing step may specifically include steps S161 to S164:

[0102] S161, the environmental signal is denoised and normalized to obtain the first signal dataset;

[0103] S162, merge the first signal dataset with the real-time traffic data to obtain the fusion result;

[0104] S163, Based on the fusion result, calculate the deviation between the current partition light intensity configuration and the preset standard template. If the deviation exceeds the preset stability threshold, adjust the partition light intensity configuration.

[0105] S164, repeat the previous adjustment step until the deviation is not higher than the preset stable threshold, and finally obtain the stable partition light intensity configuration.

[0106] It is worth noting that the 'flow data' here refers to measurement data characterizing the flow of light energy.

[0107] In step S161, the environmental signal is denoised and normalized to obtain a first signal dataset.

[0108] Specifically, this step first preprocesses the high-frequency sampled environmental signal (e.g., background light interference caused by external light sources). Wavelet transform is used to denoise the original signal. Wavelet transform enables multi-scale analysis of the signal, effectively distinguishing the true characteristics of the signal from high-frequency random noise. After denoising, the signal is then subjected to min-max normalization, linearly mapping its numerical range to the [0,1] interval. For example, an ambient light signal with an original fluctuation range of 50 lux to 70 lux will have a fluctuation range of 0.6 to 0.8 after normalization. This processing unifies the data scale, facilitating subsequent fusion of multi-source data to ultimately obtain the first signal dataset.

[0109] In step S162, the first signal dataset and the real-time traffic data are fused to obtain a fusion result.

[0110] Specifically, this step employs the Kalman filter algorithm to dynamically fuse the first signal dataset (processed environmental signals) and real-time optical energy flow data. The Kalman filter is an efficient recursive state estimation algorithm whose core idea is to use noisy measurements (real-time flow data and environmental signals) to correct predictions of the optical energy transmission process state. In this step, the filter uses environmental signal data to correct and optimize the estimation of the real-time flow data, thereby obtaining a fusion result that is statistically closer to the actual physical process and has lower noise.

[0111] In step S163, based on the fusion result, the deviation between the current partition light intensity configuration and the preset standard template is calculated. If the deviation exceeds the preset stability threshold, the partition light intensity configuration is adjusted.

[0112] Specifically, the output of the Kalman filter in step S162 (e.g., estimating the light intensity of the 550nm zone to be 900 lux) is compared with a preset standard template (e.g., requiring 1000 lux), and the relative deviation is calculated. If this deviation (10% in this example) exceeds a preset stability threshold (e.g., 8%) set for stable system operation, an adjustment is initiated. The adjustment amount can be calculated using a linear interpolation method to generate a smooth correction value proportionally to the magnitude of the deviation. For example, the algorithm might calculate that the light source power of the 550nm zone needs to be reduced by 4%, while the power of the 400nm zone needs to be increased by 2% to maintain spectral balance, thus completing this adjustment.

[0113] In step S164, the previous adjustment step is repeated until the deviation is not higher than the preset stable threshold, and finally the stable partition light intensity configuration is obtained.

[0114] Specifically, this step establishes a closed-loop iterative adjustment mechanism. After an adjustment is completed in step S163, the system does not terminate immediately but enters the next control cycle, re-executing the data fusion in step S162 and the deviation calculation in step S163. For example, after the first adjustment, the light intensity of the 550nm partition may become 995 lux. In the new cycle, the system calculates a new deviation of 0.5%. Since 0.5% is not higher than the stability threshold of 8%, the iterative adjustment loop terminates. At this point, the system determines that the current light intensity configuration (995 lux for the 550nm partition) has reached a stable state and outputs it as the stable partition light intensity configuration for continuous emission control of the light source. This process, through a continuous "fusion-comparison-adjustment" loop, ensures that the light intensity configuration can quickly respond to environmental changes and ultimately converge to a stable target state.

[0115] Reference Figure 2 The second embodiment of the present invention provides an energy-saving luminous control system for a multispectral zoned ring light source, comprising:

[0116] Initial data processing module: acquires the partitioned spectral data and environmental signals of the multispectral ring light source, performs data fusion processing, and generates the initial transmission flow matrix;

[0117] Spectral penetration analysis module: Based on the initial transmission flow matrix, the spectral penetration relationships between partitions are clustered and grouped, and the boundary features of the spectral penetration relationships are calculated to obtain a set of penetration boundaries;

[0118] Flow path determination module: If the permeation parameters in the permeation boundary set exceed the preset permeation threshold, then extract the light energy distribution characteristics and evaluate the fluctuation of the light energy distribution characteristics to determine the flow path in the light energy distribution area with large fluctuations.

[0119] Light intensity coefficient calculation module: Based on the traffic path, it obtains the interaction intensity data between adjacent partitions and compares it with a preset standard spectral template to obtain a set of adjustment coefficients;

[0120] Partition optical intensity optimization module: Based on the set of adjustment coefficients, it allocates the transmission traffic in real time and determines the optimized partition optical intensity configuration;

[0121] Light intensity iterative adjustment module: acquires real-time traffic data, fuses the environmental signal with the real-time traffic data, and iteratively adjusts the optimized partition light intensity configuration based on the fusion result to obtain a stable partition light intensity configuration.

[0122] It should be noted that the energy-saving luminous emission control system for the multispectral zoned ring light source provided in this embodiment of the invention is used to execute all the process steps of the energy-saving luminous emission control method for the multispectral zoned ring light source in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0123] 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 a flow path determination program. When the processor executes the computer program, it implements the steps in the above-described embodiments of the energy-saving luminous emission control methods for various multispectral zone ring light sources, 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 spectral penetration analysis module.

[0124] 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.

[0125] 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.

[0126] 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.

[0127] 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.

[0128] Wherein, if the modules / units integrated in 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 device 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.

[0129] 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.

[0130] 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 energy-saving luminous emission control of a multispectral zoned ring light source, characterized in that, include: Acquire the partitioned spectral data and environmental signals of the multispectral ring light source, and perform data fusion processing to generate an initial transmission flow matrix; Based on the initial transmission flow matrix, the spectral penetration relationships between partitions are clustered and grouped. The boundary features of the spectral penetration relationships are calculated based on the grouping results to obtain a set of penetration boundaries. If the permeation parameters in the permeation boundary set exceed the preset permeation threshold, then the light energy distribution characteristics are extracted and the volatility of the light energy distribution characteristics is evaluated to determine the flow path in the light energy distribution area with large volatility. Based on the traffic path, the interaction intensity data between adjacent partitions is obtained and compared with a preset standard spectral template to obtain a set of adjustment coefficients; Based on the set of adjustment coefficients, the transmission traffic is allocated in real time to determine the optimized partition optical intensity configuration; Real-time traffic data is acquired, the environmental signal is fused with the real-time traffic data, and based on the fusion result, the optimized partition light intensity configuration is iteratively adjusted to obtain a stable partition light intensity configuration.

2. The energy-saving luminous emission control method for a multispectral zoned ring light source as described in claim 1, characterized in that, The process of acquiring the partitioned spectral data and environmental signals of the multispectral ring light source, and performing data fusion processing to generate an initial transmission flow matrix includes: Acquire the partitioned spectral data and environmental signals of the multispectral ring light source to obtain the original spectral dataset and environmental signal dataset; If the original spectral dataset and environmental signal dataset contain noise, then the original spectral dataset and environmental signal dataset are denoised to obtain denoised spectral dataset and environmental signal dataset. Based on the denoised spectral dataset and the environmental signal dataset, the partitioned spectral data and environmental signals are integrated to obtain a comprehensive spectral feature dataset; Based on the comprehensive spectral feature dataset, an initial transport flow matrix is ​​calculated, and the spectral partition and permeation parameters corresponding to each element in the initial transport flow matrix are determined.

3. The energy-saving luminous emission control method for a multispectral zoned ring light source as described in claim 1, characterized in that, If the permeation parameters in the permeation boundary set exceed a preset permeation threshold, then the light energy distribution characteristics are extracted and the volatility of the light energy distribution characteristics is evaluated to determine the flow path in the light energy distribution region with high volatility, including: Obtain boundary point distribution data from the set of permeable boundaries, and extract features from the boundary point distribution data to obtain light energy distribution features; Based on the light energy distribution characteristics, the fluctuation of the characteristics is calculated, and the light energy distribution area with large fluctuation is determined; if the penetration parameter of the light energy distribution area exceeds the preset penetration threshold, the flow path in the light energy distribution area is divided to obtain a set of candidate flow paths; The candidate traffic path set is filtered to determine the traffic path with high volatility.

4. The energy-saving luminous emission control method for a multispectral zoned ring light source as described in claim 1, characterized in that, Based on the traffic path, the interaction intensity data between adjacent partitions is obtained and compared with a preset standard spectral template to obtain a set of adjustment coefficients, including: Interaction intensity data between adjacent partitions is obtained from the traffic path, and spectral analysis is performed on the interaction intensity data to obtain the spectral feature distribution; If the deviation between the spectral feature distribution and the preset standard spectral template exceeds a preset chromaticity threshold, then color distortion is determined, and the spectral features of the distorted region are identified. Based on the spectral characteristics of the distorted region, a set of adjustment coefficients is obtained by matching it with the preset standard spectral template.

5. The energy-saving luminous emission control method for a multispectral zoned ring light source as described in claim 1, characterized in that, The step of allocating transmission traffic in real time according to the set of adjustment coefficients and determining the optimized partition optical intensity configuration includes: Traffic data is obtained from the traffic path, and the traffic data is processed to analyze the traffic distribution characteristics between partitions; Based on the set of adjustment coefficients and the traffic distribution characteristics, a feedback loop mechanism is used to allocate the transmission traffic in real time, resulting in a preliminary optical intensity allocation scheme. If the deviation between the preliminary light intensity allocation scheme and the preset light intensity standard template exceeds the preset light intensity threshold, the allocation ratio of the transmission traffic is adjusted to determine the optimized partition light intensity configuration.

6. The energy-saving luminescence control method for a multispectral zoned ring light source as described in claim 1, characterized in that, The process of acquiring real-time traffic data, fusing the environmental signal with the real-time traffic data, and iteratively adjusting the optimized partitioned light intensity configuration based on the fusion result to obtain a stable partitioned light intensity configuration includes: The environmental signal is denoised and normalized to obtain a first signal dataset; The first signal dataset is fused with the real-time traffic data to obtain a fusion result; Based on the fusion result, the deviation between the current partition light intensity configuration and the preset standard template is calculated. If the deviation exceeds the preset stability threshold, the partition light intensity configuration is adjusted. Repeat the previous adjustment step until the deviation is no higher than the preset stable threshold, and finally obtain the stable partition light intensity configuration.

7. An energy-saving luminous control system for a multispectral zoned ring light source, characterized in that, include: Initial data processing module: acquires the partitioned spectral data and environmental signals of the multispectral ring light source, performs data fusion processing, and generates the initial transmission flow matrix; Spectral penetration analysis module: Based on the initial transmission flow matrix, the spectral penetration relationships between partitions are clustered and grouped, and the boundary features of the spectral penetration relationships are calculated to obtain a set of penetration boundaries; Flow path determination module: If the permeation parameters in the permeation boundary set exceed the preset permeation threshold, then extract the light energy distribution characteristics and evaluate the fluctuation of the light energy distribution characteristics to determine the flow path in the light energy distribution area with large fluctuations. Light intensity coefficient calculation module: Based on the traffic path, it obtains the interaction intensity data between adjacent partitions and compares it with a preset standard spectral template to obtain a set of adjustment coefficients; Partition optical intensity optimization module: Based on the set of adjustment coefficients, it allocates the transmission traffic in real time and determines the optimized partition optical intensity configuration; Light intensity iterative adjustment module: acquires real-time traffic data, fuses the environmental signal with the real-time traffic data, and iteratively adjusts the optimized partition light intensity configuration based on the fusion result to obtain a stable partition light intensity configuration.

Images