Spectral adaptive intelligent lighting system, method and equipment medium

Abstract

The invention discloses a spectrum self-adaptive intelligent lighting system and method and an equipment medium, relates to the technical field of intelligent home lighting, and is used for solving the problem that resource waste is possibly caused when an intelligent lighting system starts high-precision spectrum detectors of all monitoring points, the high-precision spectrum detectors usually need relatively large power support, and the cost is relatively high. Continuous starting of all monitoring points may consume a large amount of electric energy, especially under the condition of long-time operation. According to the spectrum data of the monitoring point and the distance between the monitoring point and the area adjusting center, the intelligent system can dynamically select to open or close the corresponding detector, and only when needed, the related detector can be started for spectrum adjustment, so that the energy consumption is reduced, and the energy utilization efficiency is improved. The system can determine whether global adjustment or local adjustment is needed or not according to the spectral data change rate and the time difference between the spectral data change rate and last adjustment, the AI technology can analyze environment changes in real time, illumination adjustment is conducted in good time, and the self-adaptive capacity and the response speed of intelligent home illumination are improved.

Description

Technical Field

[0001] The present invention relates to the field of smart home lighting technology, and more specifically, to a spectrum adaptive smart lighting method. Background Art

[0002] Traditional lighting systems usually use fixed light sources, such as incandescent lamps, fluorescent lamps or ordinary LED lamps, whose spectral output is relatively fixed. The spectrally adaptive lighting system can monitor the spectrum through high-precision spectral detectors at various monitoring points in the area, and dynamically adjust the spectral characteristics of the light source (such as color and color temperature) to adapt to different environments and needs. By adjusting the driving current and voltage of the LED lamp, or using a combination of different types of LEDs to generate the required spectrum, the system can simulate changes in natural light, such as cold light in the morning, white light at noon, and warm light in the evening. It can even adjust the light color according to specific scene requirements to improve the user experience.

[0003] The prior art has the following deficiencies:

[0004] Turning on high-precision spectral detectors at all monitoring points may lead to a waste of resources. High-precision spectral detectors usually require large amounts of power. Keeping all monitoring points turned on may consume a lot of power, especially when running for a long time. If each monitoring point continues to provide data, a large amount of redundant data may be generated, increasing the pressure on storage and processing. These data may not all provide useful information, resulting in higher costs for data storage and analysis. High-precision spectral detection requires powerful computing support, especially when processing and analyzing large amounts of data. If all monitoring points are turned on in real time, it may lead to excessive use of computing resources, which are not effectively used at all times. It is impossible to quickly respond to environmental changes and make timely adjustments by flexibly adjusting the strategy for turning on monitoring points.

[0005] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a spectrum adaptive intelligent lighting system, method and device medium to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A spectrum adaptive intelligent lighting method comprises the following steps:

[0009] Step S1, using a high-precision spectrum detector to measure the spectrum information of ambient light in the smart home at different monitoring points in different time periods, and measuring the distance between the high-precision spectrum detector at each monitoring point and the regional adjustment center;

[0010] Step S2, determining the monitoring point to be turned on in real time according to the time difference of the highest peak of the spectrum data of the monitoring points at different locations and the distance between the high-precision spectrum detector of each monitoring point and the regional adjustment center;

[0011] Step S3, determining whether it is necessary to turn on the detectors of all monitoring points for comprehensive adjustment according to the change rate of the spectral data measured at the monitoring point and the time difference from the last adjustment;

[0012] Step S4, using AI to comprehensively adjust the detectors at all monitoring points to the smart home lighting system.

[0013] In a preferred embodiment, step S1 includes the following contents:

[0014] Select spectral detectors with high precision and wide-band response to install them at multiple monitoring points in the smart home and record them. The spectral data collected at each monitoring point i is represented by S i (t), where t is the time and i is the number of the monitoring point;

[0015] The physical distance between the spectral detector and the regional adjustment center was recorded for each monitoring point.

[0016] In a preferred embodiment, step S2 includes the following contents:

[0017] Analyze the spectral data of each monitoring point and find the time point T with the highest spectral intensity. peak，i , the time difference between the peak appearance time of the spectrum at that point and the current time is expressed as: ΔT i =|T peak，i -T i |, where: ΔT i is the time difference between the spectral peak appearance time of monitoring point i and the current time, T peak，i is the spectral peak appearance time of monitoring point i, T i is the current time at point i;

[0018] The distance is taken into account via a weighting factor: Where: D i is the distance from the monitoring point to the adjustment center, W i is the weight associated with the distance;

[0019] For each monitoring point, a comprehensive “activation score” S is calculated based on the influence of time difference and distance. i , combining time difference and distance by weighted average: S i =α*ΔT i +β*W i , where: ΔTi is the time difference between the spectral peak appearance time of monitoring point i and the current time, W i is the weighted value of the distance between the monitoring point and the regional adjustment center. α and β are parameters for adjusting the time difference and distance weight, which are adjusted according to actual needs.

[0020] In a preferred embodiment, step S3 includes the following contents:

[0021] The change rate is obtained by calculating the change of spectral data between adjacent time points. The formula is expressed as: Where: G i (t) represents the spectral data value of the i-th monitoring point at time t, Δt is the time interval, which represents the time difference between two measurement points, and R i It represents the change rate of the spectrum data of the i-th monitoring point within the time interval Δt. When the current spectrum change rate is large and the time since the last adjustment is long, the system needs to perform a new adjustment as soon as possible. The time difference can be expressed as: D = tt last , where: t is the current time, t last is the time of the last adjustment. When D is large, the detectors of all monitoring points need to be turned on again;

[0022] Based on the spectral change rate and time difference, a decision rule is designed, which is expressed as: Where: θ1 is a threshold value of the change rate of spectral data, θ2 is a time difference threshold value. When the above conditions are met, the system needs to turn on the detectors of all monitoring points for comprehensive adjustment. If any condition is met (large change rate or large time difference), the comprehensive adjustment mechanism is triggered, and the detectors of all monitoring points are turned on for real-time adjustment. After the comprehensive adjustment, the new adjustment time t is recorded. last For the next decision.

[0023] In a preferred embodiment, step S4 includes the following contents:

[0024] First, the received spectral data λ weighted (t) and environmental data E(t) are standardized: X = λ weighted (t), E(t)];

[0025] The standardized spectral data λ weighted (t) is input into the CNN model. This vector represents the spatial feature representation of the input data. The formula can be expressed as F CNN =CNN(X);

[0026] Time series modeling: RNN predicts the spectral changes at future moments based on the input spectral data sequence, and the network outputs the spectral data prediction λ at the future moment t+1 pred(t+1), based on the results of time series analysis, the model outputs the spectral data prediction λ at the future moment pred (t+1) and the spectral variation trend;

[0027] According to the prediction results of RNN, adjust the color temperature and brightness, and calculate the optimal color temperature and brightness: color temperature T optimal : Based on the predicted spectral data λ pred (t+1), by matching the color temperature characteristics of different light sources, determine the color temperature that best suits the current environment, brightness L optimal :The calculation of brightness is usually based on the predicted light intensity, combined with the environmental data E normalized (t), adjust the indoor brightness value by considering the changes in ambient light sources;

[0028] Generate adjustment plan: According to the above optimal values ​​of color temperature and brightness, generate lighting adjustment plan P(t+1)=[T optimal ,L optimal ] and transmit the adjustment plan to the actual lighting system for adjustment;

[0029] To make the prediction results more personalized and accurate, the model will combine user feedback for self-learning and optimize the network weights through back-propagation.

[0030] A spectrally adaptive intelligent lighting system includes: an information acquisition module, a processing and monitoring module, a comprehensive analysis module and an AI adjustment module.

[0031] Information acquisition module: Use high-precision spectrum detectors to measure the spectrum information of ambient light in the smart home at different monitoring points in different time periods, and measure the distance between the high-precision spectrum detectors at each monitoring point and the regional adjustment center;

[0032] Processing monitoring module: determines the monitoring point to be opened in real time based on the time difference of the highest peak of the spectral data of the monitoring points at different locations and the distance between the high-precision spectral detector of each monitoring point and the regional adjustment center;

[0033] Comprehensive analysis module: determines whether it is necessary to turn on the detectors of all monitoring points for comprehensive adjustment based on the change rate of the spectral data measured at the monitoring point and the time difference from the last adjustment;

[0034] AI adjustment module: Use AI to comprehensively adjust the detectors at all monitoring points of the smart home lighting system.

[0035] A spectrum-adaptive intelligent lighting device medium, the device comprises a detector and a regional adjustment center, the detector is used to detect spectrum data of a monitoring point, and the regional adjustment center is used to adjust light based on the integrated spectrum data.

[0036] The technical effects and advantages of a spectrally adaptive intelligent lighting method of the present invention are as follows: According to the spectral data of the monitoring point and the distance from the regional adjustment center, the intelligent system can dynamically choose to turn on or off the corresponding detector to avoid unnecessary energy waste. Only when necessary, the relevant detectors will be enabled for spectral adjustment to reduce energy consumption and improve energy efficiency. The system can decide whether global or local adjustment is required based on the rate of change of spectral data and the time difference from the last adjustment. AI technology can analyze environmental changes in real time and adjust the light in a timely manner, thereby improving the adaptive ability and response speed of smart home lighting. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 The figure is a schematic diagram of the structure of a spectrum adaptive intelligent lighting method of the present invention.

[0038] Figure 2 The figure is a schematic diagram of the structure of a spectrum adaptive intelligent lighting system of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] Example 1

[0041] The present invention Figure 1 As shown, a spectrum adaptive intelligent lighting method is provided, comprising the following steps:

[0042] Step S1, using a high-precision spectrum detector to measure the spectrum information of the ambient light in each time period in the smart home at different monitoring points, and measuring the distance between the high-precision spectrum detector at each monitoring point and the regional adjustment center

[0043] Step S2: determine the monitoring point to be activated in real time according to the time difference of the highest peak of the spectrum data of the monitoring points at different locations and the distance between the high-precision spectrum detector of each monitoring point and the regional adjustment center.

[0044] Step S3: Determine whether to turn on the detectors at all monitoring points for comprehensive adjustment based on the change rate of the spectral data measured at the monitoring point and the time difference from the last adjustment.

[0045] Step S4, using AI to comprehensively adjust the detectors at all monitoring points. The specific implementation of the smart home lighting system is as follows:

[0046] In step S1, the spectral information of the ambient light in each time period in the smart home is measured at different monitoring points by using a high-precision spectral detector, and the distance between the high-precision spectral detector at each monitoring point and the regional adjustment center is measured. The specific contents include:

[0047] Select a spectral detector with high precision and wide-band response, which can cover a wide spectral range from visible light to near infrared. Common technologies include spectral sensors (such as CCD, CMOS sensors) or fiber optic spectral detectors, which are installed at multiple monitoring points in the smart home to measure and record the environmental spectral data in real time at different time periods. Assume that the spectral data collected at each monitoring point i can be expressed as S i (t), where t is the time and i is the number of the monitoring point.

[0048] The physical distance between the spectral detector at each monitoring point and the regional adjustment center needs to be accurately recorded, which is crucial for subsequent data analysis and adjustment decisions. The distance between the detector and the adjustment center may affect the signal transmission delay and the timeliness of data processing. To measure the distance between each monitoring point and the regional adjustment center, you can use a fixed measurement device (such as a laser rangefinder) or a network protocol (such as Wi-Fi, Bluetooth, etc.) to estimate the relative distance between each detector and the adjustment center and record it.

[0049] In step S2, the monitoring points to be activated in real time are determined according to the time difference of the highest peak of the spectral data of the monitoring points at different locations and the distance between the high-precision spectral detectors of each monitoring point and the regional adjustment center. The specific contents include:

[0050] Analyze the spectral data of each monitoring point to find the time point with the highest spectral intensity (i.e., "spectral peak"). For the spectral data of each monitoring point, calculate the time difference between the occurrence time of the spectral peak at that point and the current time. This time difference refers to the delay time from when the spectral data at that point is recorded to the current time point. In order to improve the accuracy of detection, select points with smaller time differences for subsequent detection. Select the monitoring point with the smallest time difference from the current time to ensure that the spectral data closest to real time can be captured. The time difference between each monitoring point and the current time can be calculated, and the monitoring point with the smallest time difference is preferentially selected for subsequent operations. The time difference formula can be expressed as: ΔT i =|T peak，i -T i |, where: ΔT i is the time difference between the spectral peak appearance time of monitoring point i and the current time, T peak，i is the spectral peak appearance time of monitoring point i, T i is the current time at point i.

[0051] Distance is an important factor affecting signal transmission and processing speed. Therefore, the distance between each monitoring point and the regional regulation center should be considered. Monitoring points closer to the regional regulation center may be opened first because their data is more real-time and effective. A weight value can be assigned to each monitoring point, and the weighted average of the time difference and distance can be combined to decide which monitoring point to open. The distance can be considered through a weighting coefficient: Where: D i is the distance from the monitoring point to the adjustment center, W i is a distance-related weight, where monitoring points farther away may experience greater delays and thus have smaller weights.

[0052] For each monitoring point, a comprehensive “activation score” S is calculated based on the influence of time difference and distance. i , used to determine whether to open the monitoring point for detection. The time difference and distance can be combined by weighted average: S i =α*ΔT i +β*W i , where: ΔT i is the time difference between the spectral peak appearance time of monitoring point i and the current time, W i is the weighted value of the distance between the monitoring point and the regional adjustment center. α and β are parameters for adjusting the time difference and distance weight, which can be adjusted according to actual needs.

[0053] In all monitoring points, select Enable Score i Small monitoring points, because the smallest score represents the most appropriate time synchronization and distance advantage of the monitoring point, can provide the most accurate and timely ambient light data. The real-time system can dynamically update and decide which monitoring point to enable for ambient light detection and subsequent processing based on this score. The system can dynamically adjust α and β according to the actual measured data changes, so as to better adapt to the changes in real-time ambient light. According to the spectral data and time synchronization of the monitoring point, the enabled monitoring point is continuously adjusted to ensure that the entire system always operates in the optimal state under different environmental conditions.

[0054] In step S3, it is determined whether the detectors of all monitoring points need to be turned on for comprehensive adjustment based on the change rate of the spectral data measured at the monitoring point and the time difference from the last adjustment. The specific contents include:

[0055] The change rate of spectral data refers to the speed and amplitude of the change of the spectral signal of a monitoring point over time. A monitoring point with a large change rate means that the spectral signal in this area has changed significantly, and the regulatory response may need to be adjusted or enhanced. The change rate can usually be obtained by calculating the change in spectral data between adjacent time points. The formula can be expressed as: Where: G i(t) represents the spectral data value of the i-th monitoring point at time t, Δt is the time interval, which represents the time difference between two measurement points. i It represents the rate of change of the spectral data of the i-th monitoring point within the time interval Δt.

[0056] The time difference from the last adjustment reflects the response frequency of the system. When the time difference is large, it may indicate that the environmental changes need to be re-evaluated and new monitoring needs to be performed. If the current spectrum change rate is large and the time since the last adjustment is long, the system may need to perform new adjustments as soon as possible. The time difference can be expressed as: D = tt last , where: t is the current time, t last is the time of the last adjustment. If D is large, it means that the environment may have changed significantly, so it is necessary to restart the detectors of all monitoring points.

[0057] Based on the spectral change rate and time difference, a decision rule can be designed to determine whether it is necessary to turn on the detectors of all monitoring points for comprehensive adjustment. A threshold system can be used as follows: Among them: θ1 is a threshold value of the change rate of spectral data. If the maximum change rate exceeds this threshold, it indicates that the spectrum has changed significantly. is a time difference threshold. If the time difference between the last adjustment and the current moment is greater than this value, it means that the system may not respond in time. If the above conditions are met, the system needs to turn on the detectors of all monitoring points for comprehensive adjustment. If any condition is met (large change rate or large time difference), the comprehensive adjustment mechanism is triggered, and the detectors of all monitoring points are turned on for real-time adjustment. After the comprehensive adjustment, the new adjustment time t is recorded. last For the next decision.

[0058] In step S4, AI is used to comprehensively adjust the smart home lighting system using the detectors at all monitoring points, including:

[0059] The AI ​​module calculates the optimal indoor lighting parameters (color temperature, brightness) based on real-time spectral data and prediction models, and generates adjustment plans. Standardization can help eliminate scale differences between different data sources and ensure the effectiveness and stability of the algorithm. First, the received spectral data λ weighted (t) and environmental data E(t) are standardized: X = [λ weighted (t),E(t)].

[0060] Use convolutional neural networks (CNNs) to extract spatial features of spectral data:

[0061] The standardized spectral data λ weighted(t) Input CNN model, CNN extracts features from spectral data through multiple convolutional layers, pooling layers, etc. The convolutional layer will learn the spatial pattern of the input spectral data (for example, the correlation of spectral data at different wavelengths), these features will be abstracted through the multi-layer network, and finally output a feature vector, which represents the feature representation of the input data in space, and the formula can be expressed as F CNN =CNN(X)

[0062] The extracted spatial features f(t) are subjected to time series analysis, and a recurrent neural network (RNN) is used to predict future spectral change trends. RNN is suitable for processing sequence data with time correlation and can effectively capture time series patterns.

[0063] Time series modeling: RNN predicts the spectral changes at future moments based on the input spectral data sequence. Specifically, the network outputs the spectral data prediction λ at the future moment t+1 pred (t+1). Based on the results of time series analysis, the model outputs the spectral data prediction λ at the future moment pred (t+1), and possible changing trends.

[0064] Based on the prediction results of RNN, the optimal indoor lighting adjustment plan is generated, which involves adjusting the color temperature and brightness. Calculation of the optimal color temperature and brightness: Color temperature T optimal : Based on the predicted spectral data λ pred (t+1), by matching the color temperature characteristics of different light sources, determine the color temperature that best suits the current environment. Brightness L optimal :The calculation of brightness is usually based on the predicted light intensity, combined with the environmental data E normalized (t), for example, by adjusting the brightness value in the room by taking into account changes in the ambient light source.

[0065] Generate adjustment plan: According to the above optimal values ​​of color temperature and brightness, generate lighting adjustment plan P(t+1)=[T optimal ,L optimal ] and transmit the adjustment plan to the actual lighting system for adjustment.

[0066] In order to make the prediction results more personalized and accurate, the model will conduct self-learning in combination with user feedback and optimize the network weights through back propagation. Feedback data is obtained based on the user's satisfaction with the lighting adjustment (such as whether it feels comfortable, whether the expected brightness and color temperature are achieved). Feedback can be collected through user ratings, questionnaires, or direct sensor data (such as adjusted lighting data). User feedback is used as training data, combined with loss functions (such as mean square error loss, classification loss, etc.), and the back propagation algorithm is used to update the model's weights. In this way, the system can gradually learn the user's preferences and needs, making the prediction model more personalized. After each adjustment, the model will update its weights based on the new feedback, thereby optimizing the subsequent prediction and adjustment process to ensure that future adjustments are more in line with user needs.

[0067] Example 2

[0068] Figure 2 The present invention provides a spectrum adaptive intelligent lighting system, comprising: an information acquisition module, a processing and monitoring module, a comprehensive analysis module and an AI adjustment module.

[0069] Information acquisition module: By using high-precision spectral detectors to measure the spectral information of ambient light in the smart home at different time periods at different monitoring points, the distance between the high-precision spectral detectors at each monitoring point and the regional adjustment center is measured.

[0070] Processing monitoring module: The monitoring points to be opened in real time are determined based on the time difference of the highest peak of the spectral data of the monitoring points at different locations and the distance between the high-precision spectral detectors of each monitoring point and the regional adjustment center.

[0071] Comprehensive analysis module: Determines whether it is necessary to turn on the detectors at all monitoring points for comprehensive adjustment based on the rate of change of the spectral data measured at the monitoring point and the time difference from the last adjustment.

[0072] AI adjustment module: Use AI to comprehensively adjust the detectors at all monitoring points of the smart home lighting system.

[0073] A spectrum-adaptive intelligent lighting device medium, the device comprises a detector and a regional adjustment center, the detector is used to detect spectrum data of a monitoring point, and the regional adjustment center is used to adjust light based on the integrated spectrum data.

[0074] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0076] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0077] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0078] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0079] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0080] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Based on a spectrum adaptive intelligent lighting method, characterized in that: Includes steps: Step S1, using a high-precision spectrum detector to measure the spectrum information of ambient light in the smart home at different monitoring points in different time periods, and measuring the distance between the high-precision spectrum detector at each monitoring point and the regional adjustment center; Step S2, determining the monitoring point to be turned on in real time according to the time difference of the highest peak of the spectrum data of the monitoring points at different locations and the distance between the high-precision spectrum detector of each monitoring point and the regional adjustment center; Step S3, determining whether it is necessary to turn on the detectors of all monitoring points for comprehensive adjustment according to the change rate of the spectral data measured at the monitoring point and the time difference from the last adjustment; Step S4, using AI to comprehensively adjust the detectors at all monitoring points to the smart home lighting system.

2. The spectrum adaptive intelligent lighting method according to claim 1, characterized in that: Select spectral detectors with high precision and wide-band response to install them at multiple monitoring points in the smart home and record them. The spectral data collected at each monitoring point i is represented by S i (t), where t is the time and i is the number of the monitoring point; The physical distance between the spectral detector and the regional adjustment center was recorded for each monitoring point.

3. The spectrum adaptive intelligent lighting method according to claim 2, characterized in that: Analyze the spectral data of each monitoring point and find the time point T with the highest spectral intensity. peak，i , the time difference between the peak appearance time of the spectrum at that point and the current time is expressed as: ΔT i =|T peak，i -T i |, where: ΔT i is the time difference between the spectral peak appearance time of monitoring point i and the current time, T peak，i is the spectral peak appearance time of monitoring point i, T i is the current time at point i; The distance is taken into account via a weighting factor: Where: D i is the distance from the monitoring point to the adjustment center, W i is the weight associated with the distance; For each monitoring point, a comprehensive "activation score" S is calculated based on the influence of time difference and distance. i , combining time difference and distance by weighted average: S i =α*ΔT i +β*W i , where: ΔT i is the time difference between the spectral peak appearance time of monitoring point i and the current time, W i is the weighted value of the distance between the monitoring point and the regional adjustment center. α and β are parameters for adjusting the time difference and distance weight, which are adjusted according to actual needs.

4. The spectrum adaptive intelligent lighting method according to claim 3, characterized in that: The change rate is obtained by calculating the change of spectral data between adjacent time points. The formula is expressed as: Where: G i (t) represents the spectral data value of the i-th monitoring point at time t, Δt is the time interval, which represents the time difference between two measurement points, and R i It represents the change rate of the spectrum data of the i-th monitoring point within the time interval Δt. When the current spectrum change rate is large and the time since the last adjustment is long, the system needs to perform a new adjustment as soon as possible. The time difference can be expressed as: D = tt last , where: t is the current time, t last is the time of the last adjustment. When D is large, the detectors of all monitoring points need to be turned on again; Based on the spectral change rate and time difference, a decision rule is designed, which is expressed as: Where: θ1 is a threshold value of the change rate of spectral data, θ2 is a time difference threshold value. When the above conditions are met, the system needs to turn on the detectors of all monitoring points for comprehensive adjustment. If any condition is met (large change rate or large time difference), the comprehensive adjustment mechanism is triggered, and the detectors of all monitoring points are turned on for real-time adjustment. After the comprehensive adjustment, the new adjustment time t is recorded. last For the next decision.

5. The spectrum adaptive intelligent lighting method according to claim 4, characterized in that: First, the received spectral data λ weighted (t) and environmental data E(t) are standardized: X = [λ weighted (t), E(t)]; The standardized spectral data λ weighted (t) is input into the CNN model. This vector represents the spatial feature representation of the input data. The formula can be expressed as F CNN =CNN(X); Time series modeling: RNN predicts the spectral changes at future moments based on the input spectral data sequence, and the network outputs the spectral data prediction λ at the future moment t+1 pred (t+1), based on the results of time series analysis, the model outputs the spectral data prediction λ at the future moment pred (t+1) and the spectral variation trend; According to the prediction results of RNN, adjust the color temperature and brightness, and calculate the optimal color temperature and brightness: color temperature T optimal : Based on the predicted spectral data λ pred (t+1), by matching the color temperature characteristics of different light sources, determine the color temperature that best suits the current environment, brightness L optimal :The calculation of brightness is usually based on the predicted light intensity, combined with the environmental data E normalized (t), adjust the indoor brightness value by considering the changes in ambient light sources; Generate adjustment plan: According to the above optimal values ​​of color temperature and brightness, generate lighting adjustment plan P(t+1)=[T optimal ,L optimal ] and transmit the adjustment plan to the actual lighting system for adjustment; To make the prediction results more personalized and accurate, the model will combine user feedback for self-learning and optimize the network weights through back-propagation.

6. A spectrum adaptive intelligent lighting system, used to implement the algorithm-based spectrum adaptive intelligent lighting method according to any one of claims 1 to 5, characterized in that: include: Information acquisition module, processing monitoring module, comprehensive analysis module and AI adjustment module; Information acquisition module: Use high-precision spectrum detectors to measure the spectrum information of ambient light in the smart home at different monitoring points in different time periods, and measure the distance between the high-precision spectrum detectors at each monitoring point and the regional adjustment center; Processing monitoring module: determines the monitoring point to be opened in real time based on the time difference of the highest peak of the spectral data of the monitoring points at different locations and the distance between the high-precision spectral detector of each monitoring point and the regional adjustment center; Comprehensive analysis module: determines whether it is necessary to turn on the detectors of all monitoring points for comprehensive adjustment based on the change rate of the spectral data measured at the monitoring point and the time difference from the last adjustment; AI adjustment module: Use AI to comprehensively adjust the detectors at all monitoring points of the smart home lighting system.

7. A spectrum-adaptive intelligent lighting system, method and device medium, used to implement a spectrum-adaptive intelligent lighting method based on an algorithm as described in any one of claims 1 to 5, characterized in that: The device comprises a detector and a regional adjustment center. The detector is used to detect the spectrum data of the monitoring point, and the regional adjustment center is used to adjust the light according to the integrated spectrum data.

Images