A method and system for controlling lighting parameters for circadian regulation

Abstract

The present application relates to the technical field of intelligent control, and particularly relates to a rhythm regulation light parameter control method and system. The system comprises: collecting biological sign data and environmental light data of a user through an intelligent wearable sensor, analyzing state differences in each biological rhythm mode according to the biological sign data, interacting features of the environmental light data and the state differences in the biological rhythm mode, generating light-mode state evaluation interaction feature data, constructing a light demand mapping relationship of the biological rhythm mode and learning optimization through a decision tree algorithm and the light-mode state evaluation interaction feature data, generating an optimized rhythm mode light demand decision tree model, transmitting the instant biological rhythm mode to the optimized rhythm mode light demand decision tree model to analyze intelligent rhythm light demand data, and executing rhythm regulation light parameter control work. The present application realizes dynamic intelligent light regulation control based on the instant biological rhythm mode of a user.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control, and in particular to a rhythm-adjusted light parameter control method and system. BACKGROUND

[0002] Circadian rhythm refers to the internal time regulation system of an organism, which controls the periodic changes in physiological processes and behaviors, such as sleep, body temperature, hormone secretion, etc. These changes are usually coordinated with external environmental periodic factors (such as day-night alternation). Light is a key factor affecting circadian rhythm, as light signals are transmitted through special receptors in the retina to regulate the biological clock in the brain. Circadian rhythm-adjusted light is the use of specific light conditions, such as light intensity, spectrum, and time distribution, to influence or synchronize the internal rhythms of an organism. The rhythm-adjusted light parameter control system adjusts the intensity, color temperature, spectral composition, and duration of light to adapt to the circadian rhythm needs of an organism, usually based on the operation mechanism of the biological clock, using artificial light sources to simulate the changes of natural light to help regulate the individual's physiological rhythm. However, existing rhythm-adjusted light parameter control systems cannot effectively collect user's biological sign data, such as heart rate, sleep state, etc., leading to a disconnection between light regulation and the user's actual circadian rhythm; light regulation is mostly based on preset schedules or fixed patterns, lacking in-depth analysis and prediction of individual circadian rhythms, and cannot achieve truly intelligent and personalized light control; and usually only adjusts the current circadian rhythm pattern, lacking in prediction of long-term trends and intelligent adjustment of future states, and cannot provide more accurate intervention and improvement for the user's rhythm changes. SUMMARY

[0003] Therefore, the present application provides a rhythm-adjusted light parameter control method and system to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a rhythm-adjusted light parameter control system includes the following modules:

[0005] A user perception data acquisition module is used to acquire user perception data based on intelligent wearable sensors, generating user perception data, wherein the user perception data includes environmental light data and biological sign data;

[0006] A biological rhythm pattern analysis module is used to analyze biological sign timing based on biological sign data, generating biological sign timing data; and to analyze and process biological rhythm patterns based on biological sign timing data, generating biological rhythm pattern data;

[0007] The biological rhythm mode state evaluation module is configured to perform steady state analysis on the biological rhythm mode based on the biological rhythm mode data, to generate biological rhythm mode steady state data; and perform state evaluation processing on each biological rhythm mode based on the biological rhythm mode steady state data, to generate biological rhythm mode state evaluation data.

[0008] The interaction feature analysis module is configured to perform interaction feature analysis on the ambient light and the biological rhythm mode state evaluation based on the ambient light data and the biological rhythm mode state evaluation data, to generate light-mode state evaluation interaction feature data.

[0009] The light demand model construction module is configured to establish a decision mapping relationship between the biological rhythm mode and the light demand based on a preset decision tree algorithm, to generate an initial rhythm mode light demand decision tree model; and perform multi-source decision tree pruning training optimization processing on the initial individualized biological rhythm light demand model by using the light-state interaction feature data, to generate an optimized rhythm mode light demand decision tree model.

[0010] The rhythm mode trend model construction module is configured to perform periodic feature analysis on the biological rhythm mode based on the biological rhythm mode data and the biological sign time series data, to generate biological rhythm mode periodic feature data; and design a biological rhythm mode change trend model based on the biological rhythm mode periodic feature data, to generate a biological rhythm mode change trend model.

[0011] The rhythm adjustment light parameter control module is configured to perform instant biological rhythm mode analysis processing based on the biological sign data, to generate instant biological rhythm mode data; perform instant biological rhythm mode change trend analysis processing on the instant biological rhythm mode data based on the biological rhythm mode change trend model, to generate instant biological rhythm mode change trend data; perform rhythm light demand intelligent analysis processing on the instant biological rhythm mode change trend data by using the optimized rhythm mode light demand decision tree model, to generate intelligent rhythm light demand data; and perform rhythm adjustment light parameter control operation based on the rhythm light intelligent adjustment control parameter.

[0012] The present application can accurately collect the user's biological sign data (such as heart rate, body temperature, sleep state, user behavior, etc.) and environmental illumination data through real-time sensing of intelligent wearable devices, ensuring that the system can monitor the user's physical condition and surrounding illumination conditions in real time. The collection of such data provides an accurate basis for subsequent analysis, helping to more accurately adjust the illumination and improve the matching degree of illumination control and user's actual needs. By analyzing the time sequence of the biological sign data, the user's biological sign change rule in different time periods can be accurately identified, and the biological rhythm pattern data with time sequence characteristics can be extracted, ensuring that the analysis of the biological rhythm pattern is closely related to the user's physiological state, providing personalized basis for subsequent illumination adjustment, making the illumination adjustment more synchronized with the user's biological rhythm. Combined with the interactive feature analysis of environmental illumination data and user's biological rhythm state evaluation data, the illumination adjustment is highly related to the current state of the user. Through this detailed interactive analysis, the state of the light feedback under each biological rhythm pattern is understood, and the light parameters can be dynamically adjusted, so that the light control is no longer a single-dimensional control, but fully considers the user's current biological rhythm state and the environment, improving the effectiveness and precision of light adjustment. By using the decision tree algorithm combined with the biological rhythm pattern data, the system can establish the light demand mapping relationship under different biological rhythm patterns, so that the system can provide the most suitable light environment for the user under different rhythm patterns. By further training and optimizing the decision tree model using light-state interactive feature data, the accuracy of the model is continuously improved. The optimization of the decision tree enables the system to adapt to the rhythm changes of the user and gradually adjust the light parameters, thereby providing more personalized and intelligent light adjustment, and the optimized rhythm pattern light demand decision tree model can more accurately predict and meet the user's light demand under different biological rhythm states, enhancing the system's adaptability to individual differences of users. By deeply analyzing the periodic characteristics of the biological rhythm pattern, the long-term change rule and periodic characteristics of the user's biological rhythm are identified, and the corresponding periodic characteristic data is generated to design a model reflecting the trend of the biological rhythm change. The trend model enables the system to predict future changes in the biological rhythm, so that it can prepare and adjust in advance when adjusting the light. Based on the periodic characteristic data, the light control better prevents rhythm disorders and provides more sustainable and dynamic light optimization solutions for users. Instant analysis of the user's biological sign data can update the user's biological rhythm state in real time, generating instant biological rhythm pattern data, so that the system can timely monitor the user's rhythm state and ensure the real-time nature of the light adjustment scheme. Combined with the biological rhythm pattern change trend model, the dynamic trend of the instant biological rhythm state is analyzed, and the light demand matched with it is generated through the optimized rhythm pattern light demand decision tree model. This intelligent analysis based on the current state greatly improves the response speed and adjustment accuracy of the system.The generation of the intelligent rhythm light illumination requirement data enables the system to intelligently adjust the light illumination parameters based on the current rhythm state of the user and the future trend prediction. By performing the light illumination control operation based on the instant analysis, the light illumination is effectively adjusted to adapt to the current biological rhythm requirement of the user, ensuring the rhythm balance and health of the user, ensuring the dynamic, personalized and intelligent light illumination adjustment, and greatly improving the user experience.

[0013] A rhythm-adjusted light illumination parameter control method is provided in the specification for executing the rhythm-adjusted light illumination parameter control system as described above, and the rhythm-adjusted light illumination parameter control method comprises:

[0014] Step S1: collecting user perception data according to intelligent wearable sensors to generate user perception data, wherein the user perception data includes environmental light data and biological sign data;

[0015] Step S2: performing biological sign timing identification analysis according to the biological sign data to generate biological sign timing data; performing biological rhythm pattern analysis processing according to the biological sign timing data to generate biological rhythm pattern data;

[0016] Step S3: performing stable state analysis of the biological rhythm pattern based on the biological rhythm pattern data to generate biological rhythm pattern stable state data; performing state evaluation processing of each biological rhythm pattern based on the biological rhythm pattern stable state data to generate biological rhythm pattern state evaluation data;

[0017] Step S4: performing interactive feature analysis of the environmental light and the biological rhythm pattern state evaluation based on the environmental light data and the biological rhythm pattern state evaluation data to generate light-pattern state evaluation interactive feature data;

[0018] Step S5: designing the light illumination requirement decision mapping relationship in each biological rhythm pattern based on the preset decision tree algorithm and the biological rhythm pattern data to generate a rhythm pattern light illumination requirement decision tree model; performing decision tree training optimization processing on the rhythm pattern light illumination requirement decision tree model using the light-state interactive feature data to generate an optimized rhythm pattern light illumination requirement decision tree model;

[0019] Step S6: performing periodicity feature analysis of the biological rhythm pattern based on the biological rhythm pattern data on the biological sign timing data to generate biological rhythm pattern periodicity feature data; performing biological rhythm pattern change trend model design based on the biological rhythm pattern periodicity feature data to generate a biological rhythm pattern change trend model;

[0020] Step S7: Perform real-time biorhythm pattern analysis and processing based on biological symptom data to generate real-time biorhythm pattern data; perform real-time biorhythm pattern change trend analysis and processing based on the biorhythm pattern change trend model to generate real-time biorhythm pattern change trend data; transmit the real-time biorhythm pattern change trend data to the optimized biorhythm pattern light demand decision tree model for intelligent analysis and processing of biorhythm light demand to generate intelligent biorhythm light demand data; execute biorhythm regulation light parameter control operation based on biorhythm light intelligent regulation control parameters.

[0021] The beneficial effects of this application are as follows: This invention collects users' biometric data through smart wearable devices, performs biometric time-series identification analysis and circadian rhythm pattern analysis, generates users' circadian rhythm patterns through multi-level data analysis and clustering methods, and performs precise light adjustment based on each user's circadian rhythm pattern. This real-time perception and analysis based on individual user biometrics effectively solves the problem of inaccurate collection and analysis of user biometric data, ensuring that light adjustment can accurately match the user's actual circadian rhythm needs. By designing a mathematical model of the light demand decision mapping relationship through decision tree algorithm and circadian rhythm pattern data, and training and optimizing the model, intelligent decision analysis of light adjustment needs is realized. Unlike fixed patterns or preset schedules, it can dynamically adjust light parameters according to the user's current biometric and circadian rhythm state, realizing personalized and intelligent light control. This intelligent decision-making capability ensures that light adjustment is more in line with the user's actual needs, improving the user experience. By using a biorhythm pattern change trend model and real-time updates of biorhythm pattern trend predictions, the system performs periodic feature analysis and trend prediction of users' biorhythms. This not only allows for adjustments to the current biorhythm but also enables proactive adjustments to light parameters based on future rhythm change trends, achieving adaptive light control. This solves the problem of only being able to make simple adjustments to the current state, providing users with more precise rhythm management and health interventions, especially with more significant effects in long-term light regulation. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the module flow of a rhythm-regulated illumination parameter control system according to the present invention;

[0023] Figure 2 for Figure 1 A detailed functional flowchart of the biological rhythm pattern state assessment module;

[0024] Figure 3 for Figure 1 A detailed functional flowchart of the medium-lighting demand model construction module;

[0025] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0026] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0027] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0028] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0029] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a rhythm-based illumination parameter control system, comprising the following modules:

[0030] To achieve the above objectives, a rhythm-based illumination parameter control system includes the following modules:

[0031] The user perception data acquisition module is used to collect user perception data based on the smart wearable sensor and generate user perception data, wherein the user perception data includes ambient light data and biometric data.

[0032] The circadian rhythm pattern analysis module is used to perform time-series identification analysis of biological signs based on biological sign data, and generate time-series data of biological signs; and to perform circadian rhythm pattern analysis and processing based on time-series data of biological signs, and generate circadian rhythm pattern data.

[0033] The circadian rhythm pattern state assessment module is used to perform stable state analysis of circadian rhythm patterns based on circadian rhythm pattern data, and generate stable state data of circadian rhythm patterns; and to perform state assessment processing of each circadian rhythm pattern based on the stable state data of circadian rhythm patterns, and generate state assessment data of circadian rhythm patterns.

[0034] The interaction feature analysis module is used to perform interaction feature analysis on the environmental light data and the biological rhythm pattern state assessment data, and generate light-pattern state assessment interaction feature data.

[0035] The light demand model construction module is used to establish a decision mapping relationship between biological rhythm patterns and light demand based on a preset decision tree algorithm, and generate an initial rhythm pattern light demand decision tree model; the initial personalized biological rhythm light demand model is optimized by multi-source decision tree pruning training using light-state interaction feature data, and an optimized rhythm pattern light demand decision tree model is generated.

[0036] The rhythm pattern trend model construction module is used to perform periodic feature analysis of biological rhythm pattern time series data based on biological rhythm pattern data, and generate periodic feature data of biological rhythm pattern; based on the periodic feature data of biological rhythm pattern, it designs a trend model of biological rhythm pattern change and generates a trend model of biological rhythm pattern change.

[0037] The circadian rhythm regulation lighting parameter control module is used to perform real-time biorhythm pattern analysis and processing based on biological symptom data to generate real-time biorhythm pattern data; to perform real-time biorhythm pattern change trend analysis and processing based on the biorhythm pattern change trend model to generate real-time biorhythm pattern change trend data; to transmit the real-time biorhythm pattern change trend data to the optimized circadian rhythm pattern lighting demand decision tree model for intelligent analysis and processing of circadian rhythm lighting demand to generate intelligent circadian rhythm lighting demand data; and to execute circadian rhythm regulation lighting parameter control operations based on the intelligent circadian rhythm lighting regulation control parameters.

[0038] This invention utilizes real-time sensing through smart wearable devices to accurately collect users' biometric data (such as heart rate, body temperature, sleep status, and user behavior) and ambient light data, ensuring the system can monitor the user's physical condition and surrounding lighting conditions in real time. This data collection provides a precise foundation for subsequent analysis, facilitating more accurate light adjustment and improving the match between light control and the user's actual needs. By performing time-series identification analysis on the biometric data, the system can accurately identify the changes in the user's biometrics at different times and extract time-series biorhythm pattern data, ensuring that the analysis of biorhythm patterns is closely related to the user's physiological state. This provides a personalized basis for subsequent light regulation, making light regulation more synchronized with the user's biorhythm. Combining ambient light data with the user's biorhythm state assessment data for interactive feature analysis ensures that light regulation is highly correlated with the user's current state. Through this refined interactive analysis, understanding the state of light feedback under various biorhythm patterns allows for dynamic adjustment of light parameters. This transforms light control from a single-dimensional control to one that fully considers the user's current biorhythm state and environment, improving the effectiveness and accuracy of light regulation. By using decision tree algorithms combined with circadian rhythm pattern data, the system can establish a mapping relationship between light requirements under different circadian rhythm patterns, enabling it to provide users with the most suitable light environment under different rhythm patterns. Further training and optimization of the decision tree model using light-state interaction feature data continuously improves the model's accuracy. The optimized decision tree processing allows the system to adapt to changes in the user's rhythm and gradually adjust light parameters, thus providing more personalized and intelligent light regulation. The generated optimized circadian rhythm pattern light requirement decision tree model can more accurately predict and meet the user's light requirements under different circadian rhythm states, enhancing the system's adaptability to individual user differences. In-depth analysis of the periodic characteristics of circadian rhythm patterns identifies the long-term variation patterns and periodic characteristics of the user's circadian rhythm, generating corresponding periodic feature data to design a model reflecting the trend of circadian rhythm changes. This trend model enables the system to predict future changes in circadian rhythms, allowing for advance preparation and adjustment during light regulation. Through data based on periodic features, light control better prevents rhythm disorders and provides users with more continuous and dynamic light optimization solutions. Real-time analysis of users' biometric data enables continuous updates to their circadian rhythm status, generating real-time circadian rhythm pattern data. This allows the system to monitor users' rhythm status in a timely manner, ensuring the real-time nature of light regulation plans. By combining a circadian rhythm pattern change trend model with dynamic trend analysis of real-time circadian rhythm status, and optimizing the circadian rhythm pattern light demand decision tree model to generate matching light requirements, this intelligent analysis and processing based on the current state significantly improves the system's response speed and regulation accuracy.The generation of intelligent circadian rhythm lighting demand data enables the system to intelligently adjust lighting parameters based on the user's current circadian rhythm status and future trend predictions. By executing lighting control operations based on real-time analysis, the system effectively adjusts lighting to adapt to the user's current biorhythmic needs, ensuring the user's circadian rhythm balance and health. This ensures the dynamic, personalized, and intelligent nature of lighting regulation, greatly enhancing the user experience.

[0039] In this embodiment of the invention, reference Figure 1 The above is a schematic diagram of the module flow of a rhythm-regulated illumination parameter control system according to the present invention. In this embodiment, the rhythm-regulated illumination parameter control system includes:

[0040] S1: User perception data acquisition module, used to collect user perception data based on smart wearable sensors and generate user perception data, wherein the user perception data includes ambient light data and biometric data.

[0041] In this embodiment of the invention, a smart wearable device equipped with an ambient light sensor and biometric sensors (such as a heart rate sensor and a skin temperature sensor) is worn by the user. Sufficient contact between the device and the user's skin is ensured to improve data accuracy. The data acquisition module is triggered to start the acquisition process via a user interface or a preset schedule, automatically setting the acquisition interval based on the user's daily routine, or it can be triggered manually by the user. The ambient light sensor built into the smart wearable device continuously detects information such as external light intensity and color temperature, generating ambient light data. The light data should include a timestamp for time-series correlation with other data. The smart wearable device collects the user's biometric data (such as heart rate, body temperature, blood oxygen level, etc.) through sensors. This data is collected synchronously with the ambient light data and marked with the same timestamp to ensure consistency of time-series data. The collected ambient light data and biometric data are transmitted to a backend system or the cloud via a wireless connection (such as Wi-Fi or Bluetooth). The data undergoes preprocessing before uploading (e.g., noise reduction, outlier detection) to finally obtain the user-processed ambient light data and biometric data.

[0042] S2: The circadian rhythm pattern analysis module is used to perform circadian rhythm time sequence identification analysis based on circadian rhythm data to generate circadian rhythm time sequence data; and to perform circadian rhythm pattern analysis and processing based on circadian rhythm time sequence data to generate circadian rhythm pattern data.

[0043] In this embodiment of the invention, biometric data undergoes denoising, outlier cleaning, and interpolation to ensure data continuity and validity. Based on the temporal characteristics of the biometric data, temporal identifiers of biometrics are extracted using temporal analysis algorithms (such as Dynamic Time Warping (DTW) or smoothing filters). These biometric temporal identifiers include changes in key physiological indicators at specific times (such as periodic fluctuations in heart rate), used for subsequent analysis of biorhythms. By performing biorhythm pattern analysis on the biometric time-series data, clustering algorithms are applied to regularize various types of biometric time-series data. Pre-defined biorhythm pattern labels are used to identify the rhythmic patterns of biometrics. The generated biorhythm pattern data includes information such as peaks, troughs, and cycles of change in the user's biorhythm at different time periods, reflecting the user's daily physiological rhythm.

[0044] S3: Biorhythm Pattern State Assessment Module, used to perform stable state analysis of biorhythm patterns based on biorhythm pattern data, and generate stable state data of biorhythm patterns; and to perform state assessment processing of each biorhythm pattern based on the stable state data of biorhythm patterns, and generate biorhythm pattern state assessment data.

[0045] In this embodiment of the invention, the stability of circadian rhythms is assessed based on circadian rhythm pattern data. Statistical analysis methods (such as coefficient of variation (CV), mean, and standard deviation calculation) are used to analyze the fluctuation range and frequency of the rhythm to identify a stable circadian rhythm. If the circadian rhythm fluctuation exceeds the normal range, it means that the user's circadian rhythm is disturbed by external factors (such as changes in light exposure or irregular sleep patterns). Based on the stable state data, combined with preset health assessment standards (such as circadian rhythm stability index, health standard range, etc.), a health assessment of the circadian rhythm pattern is performed. Machine learning models are used to classify the data, such as support vector machines (SVM) or random forest classifiers, to generate circadian rhythm pattern state assessment data, evaluating the user's circadian rhythm state, including: normal, disordered, sub-healthy, or abnormal.

[0046] S4: Interactive Feature Analysis Module, used to perform interactive feature analysis of ambient light and circadian rhythm state assessment based on ambient light data and circadian rhythm state assessment data, and generate light-mode state assessment interactive feature data.

[0047] In this embodiment of the invention, the collected ambient light data and the analyzed circadian rhythm pattern state assessment data undergo temporal alignment and preprocessing to ensure consistency between the two types of data over time. Next, correlation and regression analysis techniques are used to analyze the relationship between light intensity, color temperature changes, and circadian rhythm pattern states (such as stability or volatility), extracting key light parameters that influence circadian rhythms. This analysis generates interactive characteristic data between ambient light and circadian rhythm pattern states, revealing how light conditions affect the user's circadian rhythm state and providing a basis for subsequent light regulation.

[0048] S5: Light demand model construction module, used to establish a decision mapping relationship between biological rhythm patterns and light demand based on a preset decision tree algorithm, generate an initial rhythm pattern light demand decision tree model; use light-state interaction feature data to perform multi-source decision tree pruning training optimization on the initial personalized biological rhythm light demand model, generate an optimized rhythm pattern light demand decision tree model.

[0049] In this embodiment of the invention, an initial light demand decision tree model is established based on a preset decision tree algorithm, utilizing circadian rhythm pattern data and historical light intensity data. This model generates an initial circadian rhythm pattern light demand decision tree by analyzing the mapping relationship between the user's circadian rhythm characteristics and light demand. Combining the light-state interaction feature data generated by the interaction feature analysis module, the initial model undergoes multi-source data optimization processing. Redundant decision paths are reduced through pruning algorithms, and the light demand under each circadian rhythm pattern is learned, improving the model's accuracy and efficiency, and generating a personalized, optimized circadian rhythm pattern light demand decision tree model.

[0050] S6: Rhythm Pattern Trend Model Construction Module, used to perform periodic characteristic analysis of biological rhythm pattern on biological vital sign time series data based on biological rhythm pattern data, and generate periodic characteristic data of biological rhythm pattern; and to design a trend model of biological rhythm pattern change based on the periodic characteristic data of biological rhythm pattern, and generate a trend model of biological rhythm pattern change.

[0051] In this embodiment of the invention, based on biorhythm pattern data, periodic analysis methods (such as Fourier transform and period extraction algorithms) are applied to analyze the time-series data of biological characteristics, extract periodic features of the rhythm, and generate periodic feature data reflecting the long-term changes in the user's rhythm. A trend model of biorhythm pattern change is constructed based on the periodic feature data of the biorhythm pattern to predict future trends in biorhythm changes. This model simulates the dynamic changes of biorhythms through time-series analysis and machine learning methods (such as Long Short-Term Memory networks, LSTM), providing trend references for personalized lighting needs adjustments.

[0052] S7: Circadian rhythm regulation illumination parameter control module, used to perform real-time biorhythm pattern analysis and processing based on biological symptom data to generate real-time biorhythm pattern data; perform real-time biorhythm pattern change trend analysis and processing based on the biorhythm pattern change trend model to generate real-time biorhythm pattern change trend data; transmit the real-time biorhythm pattern change trend data to the optimized circadian rhythm light demand decision tree model for intelligent analysis and processing of circadian light demand to generate intelligent circadian light demand data; and execute circadian rhythm regulation illumination parameter control operations based on intelligent circadian light regulation control parameters.

[0053] In this embodiment of the invention, based on the latest biometric data and by comparing the previously analyzed biorhythm pattern data with the latest biometric data, the user's current biorhythm pattern is analyzed in real time to generate instantaneous biorhythm pattern data. Based on the biorhythm pattern change trend model constructed in the previous step, trend analysis is performed on the current data to predict the future direction of changes in the user's biorhythm, generating instantaneous biorhythm change trend data. This trend data is then transmitted to an optimized biorhythm pattern lighting requirement decision tree model for intelligent analysis and processing of biorhythm lighting requirements, generating accurate intelligent biorhythm lighting requirement data. Based on the intelligent biorhythm lighting requirement data, lighting parameters (such as light intensity and color temperature) are automatically adjusted, and biorhythm-regulated lighting control is executed to ensure that the user's lighting environment matches their biorhythm requirements.

[0054] Preferably, the circadian rhythm pattern analysis module includes the following functions:

[0055] Based on the biological signs data, perform biological sign time series identification analysis to generate biological sign time series data;

[0056] Based on the time series data of biological signs, analyze the transient characteristic data of biological signs to generate transient characteristic data of biological signs;

[0057] Transient cluster analysis of biological signs was performed on transient characteristic data of biological signs to generate transient cluster data of biological signs;

[0058] Biological rhythm pattern analysis is performed based on transient clustering data of biological signs to generate biological rhythm pattern data.

[0059] This invention performs time-series identification analysis on biometric data, extracting the changing trends of a user's physiological state over time, forming a series of time-series feature data. This ensures that the analysis of biometrics is not limited to the current state but also identifies physiological changes at different times, providing a more timely and dynamic foundation for subsequent biorhythm pattern analysis. This allows for more precise adjustment of light intensity based on the user's actual biorhythm changes. By performing transient feature analysis on the time-series biometric data, the invention captures the user's physiological characteristics at specific moments, such as heart rate, body temperature, and blood pressure. These transient feature data reflect the user's immediate physiological condition at different points in time, further understanding the user's current health status and biorhythm changes, making light intensity adjustment more dynamic and able to quickly respond to the user's immediate needs. Through cluster analysis of the transient feature data, the data is grouped according to different physiological characteristics of the user, generating meaningful cluster data. This helps identify potential patterns and correlations between user physiological characteristics, providing data support for more in-depth biorhythm pattern analysis. Cluster analysis can effectively extract individual physiological rhythm patterns of users, enabling lighting regulation to not only be based on general patterns but also to be finely adjusted according to individual user characteristics, further enhancing the personalization and intelligence of lighting control. Analyzing transient cluster data for biorhythm patterns allows for the derivation of users' biorhythm patterns based on their transient cluster characteristics. This identifies users' diurnal rhythm states, helping the system understand whether their current physiological rhythms are normal or disordered. This provides specific data support for subsequent lighting regulation based on individual user rhythms, ensuring that lighting regulation schemes accurately match users' current rhythm states and help them restore and maintain rhythm balance and stability.

[0060] In this embodiment of the invention, based on biometric data (such as heart rate, body temperature, and respiratory rate) collected by a smart wearable device, time-series analysis methods (such as Dynamic Time Warping (DTW) or smoothing filters) are used to process the data and generate biometric time-series data. By time-labeling and aligning these data, the patterns of biometric changes over time are identified, generating time-series data that reflects the dynamic changes in the user's physiological state over time, serving as the basis for subsequent analysis. Based on the biometric time-series data, transient feature analysis is performed on the biometric states at different time points. By extracting transient feature indicators (such as instantaneous heart rate and temperature fluctuations), data processing algorithms are used to analyze the user's biometric characteristics at a specific time point, generating transient biometric feature data to describe the user's physiological state at each time point and further capture key feature changes in circadian rhythms. Clustering algorithms (such as K-means and DBSCAN) are applied to perform cluster analysis on the transient biometric feature data. By classifying similar transient feature data, biometric patterns under different states are identified. By grouping time points with similar physiological characteristics together, transient biometric clustering data is generated. This clustering data helps analyze the changing trends of a user's physiological state over different time periods. Based on this transient biometric clustering data, further biorhythm pattern analysis is performed to identify the biorhythm patterns corresponding to different clusters. These patterns are then classified using pre-defined biorhythm pattern labels, generating biorhythm pattern data. This data describes various daily biorhythmic behavioral patterns of users, such as sleep and work.

[0061] Preferably, the biological rhythm pattern state assessment module includes the following functions:

[0062] Based on the transient clustering data of biological signs, baseline feature analysis of transient clustering of biological signs is performed to generate baseline feature data of transient clustering of biological signs.

[0063] Based on the biorhythm pattern data, the stable state analysis of the transient clustering baseline feature data of biological signs is performed to generate stable state data of biorhythm patterns.

[0064] Perform cluster subset bias quantification analysis on transient cluster data of biological signs to generate biological sign cluster subset bias quantification data;

[0065] Based on the stable state data of biological rhythm patterns, the deviation quantification data of biological characteristic cluster subsets are processed to evaluate the state of each biological rhythm pattern, generating biological rhythm pattern state evaluation data.

[0066] This invention performs baseline feature analysis on transient clustering data of biological characteristics, extracting stable physiological feature baselines from the transient clusters. These baseline feature data provide a physiological characteristic benchmark for users under normal conditions, helping to identify the user's health level and circadian rhythm balance. This provides a reference for subsequent circadian rhythm pattern analysis, enabling the determination of whether the user is in a stable circadian rhythm and allowing for corresponding light regulation adjustments based on baseline features, improving the system's accuracy and personalized adjustment capabilities. Analyzing the transient clustering baseline feature data based on biorhythm pattern data assesses whether the user's biorhythm is in a stable state, identifying whether the user's circadian rhythm is disturbed or fluctuates, generating stable circadian rhythm pattern data, providing a crucial basis for light control, enabling real-time optimization of light regulation based on the user's circadian rhythm state, helping users maintain or restore circadian rhythm stability, and improving health and comfort. By performing deviation quantification analysis on a subset of transient clustering data, the difference between the user's current physiological state and their normal baseline can be quantitatively assessed. This deviation analysis can accurately identify the degree of abnormal changes or deviations in a user's circadian rhythm, thus providing more refined data support for light regulation. It allows for corresponding light regulation measures tailored to different deviation amplitudes, ensuring the accuracy and effectiveness of the regulation. By combining stable state data of the rhythm pattern with clustered subset deviation quantification data, a comprehensive assessment of the user's state under different rhythm patterns is conducted to determine whether the user's circadian rhythm is in a normal, fluctuating, or other different states. This precise state assessment provides a highly personalized basis for subsequent light regulation, enabling the light regulation scheme to be finely adjusted according to different rhythm states, thereby improving the regulation effect and helping users maintain a healthy circadian rhythm.

[0067] As an example of the present invention, reference is made to... Figure 2 As shown, Figure 1 A functional flowchart of the circadian rhythm pattern state assessment module is shown. In this example, the functions of the circadian rhythm pattern state assessment module include:

[0068] S31: Perform baseline feature analysis of transient clustering of biological signs based on transient clustering data of biological signs, and generate baseline feature data of transient clustering of biological signs;

[0069] In this embodiment of the invention, key features are extracted from transient clustering data of biological signs, and baseline feature analysis is performed. By calculating the centroid or average value of each cluster group, baseline feature data is generated. This data is used to describe the typical state of each biological sign cluster group, ensuring that the generated baseline feature data can accurately represent the user's normal physiological state and providing a basis for subsequent stability analysis.

[0070] S32: Based on the biological rhythm pattern data, perform stable state analysis of the transient clustering baseline feature data of biological signs to generate stable state data of biological rhythm patterns.

[0071] In this embodiment of the invention, biorhythm pattern data is used to perform stability analysis on the transient clustering baseline characteristic data of biological signs. By calculating the changes in the clustering baseline within each time period, the volatility and consistency of the biorhythm are assessed, generating stable state data of the biorhythm pattern. This analysis determines the fluctuation amplitude and time span of each cluster state to indicate that the user's biorhythm pattern is in a stable state.

[0072] S33: Perform cluster subset bias quantification analysis on transient cluster data of biological signs to generate biological sign cluster subset bias quantification data;

[0073] In this embodiment of the invention, based on transient clustering data of biological characteristics, a deviation quantification analysis algorithm (such as mean squared error or coefficient of variation) is applied to quantify the deviation of each cluster subset. By comparing the data of each subset with its corresponding baseline features, the degree of deviation is calculated, generating deviation quantification data for the biological characteristic cluster subsets. This deviation data is used to assess whether each cluster subset is consistent with the baseline state or has a significant deviation.

[0074] S34: Based on the stable state data of the biological rhythm pattern, the deviation quantification data of the biological sign cluster subset is processed to evaluate the state of each biological rhythm pattern, and the biological rhythm pattern state evaluation data is generated.

[0075] In this embodiment of the invention, based on the stable state data of biorhythm patterns and combined with the deviation quantification data of biosignature cluster subsets, a state assessment process is performed on various biorhythm patterns. Specifically, through multi-dimensional analysis, the stability and deviation degree of biorhythms in each cluster state are calculated to generate biorhythm pattern state assessment data. This data provides an evaluation of the user's overall biorhythm state, such as whether the rhythm is normal, abnormal, or requires intervention.

[0076] Preferably, the interaction feature analysis module includes the following functions:

[0077] Based on ambient light data and circadian rhythm pattern state assessment data, synchronized processing of ambient light and circadian rhythm pattern state assessment data is performed to generate synchronized light-pattern state assessment data.

[0078] Based on the synchronous illumination-mode state assessment data, the interaction characteristics of ambient illumination and biological rhythm mode state assessment are analyzed to generate illumination-mode state assessment interaction characteristic data.

[0079] This invention synchronizes ambient light data with circadian rhythm state assessment data, organically combining current ambient light conditions with the user's circadian rhythm state. This synchronization process ensures that real-time light adjustments not only consider changes in ambient light but also simultaneously analyze the user's circadian rhythm state, enabling adaptive adjustments to ambient light conditions based on the user's current circadian rhythm needs. The synchronized light-pattern state assessment data provides integrated, multi-dimensional data support for subsequent analysis, contributing to improved personalization and accuracy of light regulation. Interactive feature analysis of the synchronized light-pattern state assessment data reveals the complex relationship between ambient light and the user's circadian rhythm pattern. This interactive feature analysis not only helps the system understand how changes in light affect the user's circadian rhythm state but also identifies sensitive time points or parameters related to light and circadian rhythm. For example, different light intensities or color temperatures may have different effects on a user's circadian rhythm. By analyzing interaction features, we can help find the most suitable light parameters for the user's current circadian rhythm. The generated interaction feature data provides more targeted guidance for light adjustment, ensuring that light adjustment better meets the user's immediate needs and maximizing the effectiveness of circadian rhythm regulation.

[0080] In this embodiment of the invention, ambient light data acquired from an ambient light sensor in a smart wearable device and analyzed biorhythm pattern state assessment data are synchronized. Timestamp matching technology is used to align the light data and biorhythm pattern data, ensuring consistency in their time series and generating synchronized light-pattern state assessment data. This step aims to provide an accurate data foundation for subsequent interaction feature analysis. After data synchronization, interaction feature analysis is performed based on the synchronized light-pattern state assessment data. Algorithms such as multivariate regression analysis and correlation analysis are used to analyze the relationship between ambient light intensity, color temperature, and other light characteristics and biorhythm pattern states. Key features reflecting the interaction between the two are extracted, such as the state differences caused by different ambient light levels under the same biorhythm pattern, generating light-pattern state assessment interaction feature data.

[0081] Preferably, the illumination demand model construction module includes the following functions:

[0082] A decision mapping relationship between biological rhythm patterns and light demand is established based on a pre-defined decision tree algorithm, generating an initial decision tree model of light demand for the rhythm pattern.

[0083] Based on the biological rhythm pattern data, the initial biological rhythm pattern light demand decision tree model is pruned for each biological rhythm pattern to generate a biological rhythm pattern light demand decision tree model.

[0084] The light-mode state assessment interaction feature data of light-mode state are weighted by light parameters of each biological rhythm mode to generate weighted light-mode state assessment interaction feature data.

[0085] Based on the weighted illumination-mode state evaluation interaction feature data, the rhythmic mode illumination demand decision tree model is trained and optimized to generate an optimized rhythmic mode illumination demand decision tree model.

[0086] This invention establishes a mapping relationship between light requirements based on different circadian rhythm patterns using a pre-defined decision tree algorithm. This generates an initial circadian rhythm pattern light requirement decision tree model, clearly reflecting the light conditions (such as light intensity and duration) required by the user under different circadian rhythm states. This model provides a basic framework for light regulation, ensuring the system can effectively adjust light according to circadian rhythm patterns, laying the foundation for subsequent optimization and personalized adjustments. Decision tree pruning is a key step in optimizing the decision tree structure. Pruning removes redundant or unnecessary branches, simplifying the decision model. Pruning the initial decision tree model based on circadian rhythm pattern data improves model efficiency and prediction accuracy, enabling faster responses to changes in circadian rhythm states while maintaining decision accuracy. This ensures a more efficient and smoother light regulation process, avoiding unnecessary complexity. Weighting the interactive feature data of light-mode state assessment allows for the assignment of appropriate weights to various light parameters (such as intensity and color temperature) under different circadian rhythm modes. This enables a more rational balance and adjustment of light parameters based on the user's current circadian rhythm state. The weighting process makes light regulation more flexible and precise, allowing for the prioritization of different parameters according to actual needs, thus achieving more accurate light control. Training and optimizing the weighted interactive feature data improves the adaptability and accuracy of the circadian rhythm light demand decision tree model. The optimized model better predicts the user's light needs under different circadian rhythm states and dynamically adjusts based on actual data. This optimization process makes the light control process more intelligent and adaptive, ensuring that the system can provide continuously optimized light solutions as the user's circadian rhythm changes.

[0087] As an example of the present invention, reference is made to... Figure 3 As shown, Figure 1 A functional flowchart of the illumination demand model construction module is shown below. In this example, the functions of the illumination demand model construction module include:

[0088] S51: Establish a decision mapping relationship between biological rhythm patterns and light demand based on a preset decision tree algorithm, and generate an initial rhythm pattern light demand decision tree model.

[0089] In this embodiment of the invention, a mapping relationship between circadian rhythm patterns and light requirements is established based on a preset decision tree algorithm. Specifically, the preset decision tree algorithm, along with collected circadian rhythm pattern data (such as periodic changes in biological characteristics, peak and trough times, etc.) and historical light requirement data (such as light intensity, color temperature, etc.), generates an initial circadian rhythm pattern light requirement decision tree model. Through the branching structure in the decision tree, each node represents the light requirement of a specific circadian rhythm pattern, thereby achieving a preliminary mapping between circadian rhythm patterns and light requirements.

[0090] S52: Based on the biological rhythm pattern data, the initial biological rhythm pattern light demand decision tree model is pruned for each biological rhythm pattern to generate a biological rhythm pattern light demand decision tree model.

[0091] In this embodiment of the invention, based on circadian rhythm pattern data, the initial circadian rhythm pattern light demand decision tree model is pruned to optimize its complexity. By analyzing the light demand characteristics of each circadian rhythm pattern, it is determined whether certain branches contain redundant or irrelevant information. Using pruning algorithms (such as CART or ID3 pruning algorithms), nodes and branches with minimal impact on the model's prediction performance are removed, reducing overfitting and generating a simplified circadian rhythm pattern light demand decision tree model. The pruned model is more efficient, reducing computational resource consumption and improving the model's generalization ability.

[0092] S53: Perform weighted processing of the light parameters of each biological rhythm pattern on the light-mode state assessment interaction feature data to generate weighted light-mode state assessment interaction feature data.

[0093] In this embodiment of the invention, the illumination parameters of each circadian rhythm pattern are weighted based on the illumination-pattern state assessment interaction feature data. By analyzing the correlation between parameters such as light intensity and color temperature and the state of the circadian rhythm pattern, a weighted average method or information gain method is used to assign different weight values ​​to each illumination parameter. The magnitude of the weight value is determined according to the degree of influence of the illumination parameter on the circadian rhythm pattern. The weighted illumination-pattern state assessment interaction feature data is generated, including the ambient illumination corresponding to each circadian rhythm pattern state and the weight information corresponding to each ambient illumination feature information, ensuring that the influence of different illumination parameters on each circadian rhythm pattern can be accurately captured in the subsequent model optimization process.

[0094] S54: Based on the weighted illumination-mode state evaluation interaction feature data, the rhythmic mode illumination demand decision tree model is trained and optimized to generate an optimized rhythmic mode illumination demand decision tree model.

[0095] In this embodiment of the invention, a circadian rhythmic lighting demand decision tree model is trained and optimized based on weighted illumination-pattern state assessment interaction feature data. The system employs supervised learning methods, such as gradient descent, and utilizes historical data for iterative model training, gradually adjusting the weights and branch structure in the decision tree. By repeatedly optimizing the model parameters, it is made to better adapt to individualized lighting needs, ultimately generating an optimized circadian rhythmic lighting demand decision tree model. The optimized model can more accurately predict users' lighting needs under different biological rhythm states, thereby improving the overall system's accuracy and responsiveness.

[0096] Preferably, the weighting process for the illumination parameters of each circadian rhythm pattern in the illumination-pattern state assessment interaction feature data includes:

[0097] Temporal data extraction is performed on the interaction feature data of illumination-mode state assessment to generate time-series data of illumination-mode state assessment interaction features;

[0098] Based on the time-series data of the interaction characteristics of light-pattern state assessment, the time-series light sensitivity factor analysis of each biological rhythm pattern is performed to generate time-series light sensitivity factor data;

[0099] Based on time-series light sensitivity factor data, the light parameters of each biological rhythm pattern are weighted in the light-mode state assessment interaction feature data to generate weighted light-mode state assessment interaction feature data.

[0100] This invention extracts time-series data from the interaction characteristics of light-mode state assessment, identifying the time dependency between light and circadian rhythm patterns. This allows the system to analyze how light conditions affect user rhythm changes over time and capture the dynamic characteristics of light parameters. The generated time-series data lays the foundation for subsequent analysis, ensuring the system can make more precise adjustments based on historical changes in light and circadian rhythm states, making light regulation schemes more timely. Through time-series light sensitivity factor analysis, the sensitivity of users to specific light parameters under different circadian rhythm patterns is identified, helping to discover the impact of certain light conditions (such as intensity and color temperature) on various circadian rhythm patterns at different times. This extraction of sensitivity factors allows for more precise light regulation of different circadian rhythm patterns, making light regulation more targeted and better suited to users' actual needs, avoiding unnecessary intervention. Based on the time-series light sensitivity factor data, different light parameters are weighted, and the priority of each light parameter is dynamically adjusted under different circadian rhythm patterns. Weighted processing ensures that the system can make corresponding weight adjustments for significantly influential lighting factors during illumination regulation, improving the flexibility and accuracy of the adjustment. The generated weighted data makes illumination regulation more personalized, fully adapting to the specific needs of users under different rhythmic states.

[0101] In this embodiment of the invention, when processing the light-mode state assessment interaction feature data, it is necessary to extract the time series of these interaction feature data. The light data and circadian rhythm mode state data are rearranged and segmented according to the time dimension to generate time series data of the light-mode state assessment interaction features. This captures the correlation between light features and circadian rhythm mode states that change over time, ensuring that the time series data reflects the dynamic relationship between light and circadian rhythm modes. Time series data extraction is the foundation for subsequent analysis, ensuring that feature data is aligned and processed in chronological order. After generating the time series data of the light-mode state assessment interaction features, the system will perform time series light sensitivity factor analysis on each circadian rhythm mode based on this time series data. The system uses time series analysis methods (such as autocorrelation analysis and time series regression analysis) to identify the sensitivity factors between light parameters (such as light intensity, color temperature, and light duration) and circadian rhythm mode states. By analyzing the influence of light parameters on circadian rhythms at different times, time series light sensitivity factor data is generated. Each light sensitivity factor represents the degree of influence of a certain light parameter on a specific circadian rhythm pattern within a specific time period. This sensitivity factor data will serve as a crucial basis for subsequent weight assignment. After obtaining the time-series light sensitivity factor data, the light parameters in the light-pattern state assessment interaction feature data will be weighted based on these sensitivity factors. Different weights are assigned to each light parameter according to the magnitude of the time-series light sensitivity factors, and the weight values ​​reflect the degree of influence of the light parameter on a specific circadian rhythm pattern. Weight assignment employs weighted analysis methods, such as information gain analysis or weighted regression analysis, to ensure that the weight of the light parameter is proportional to its sensitivity factor data. The generated weighted light-pattern state assessment interaction feature data includes the relative importance of different light parameters under each circadian rhythm pattern, providing accurate input data for the subsequent optimization and training of the decision tree model. This ensures that the influence of light parameters is accurately modeled, thereby improving the model's ability to predict the light demand of individual circadian rhythms.

[0102] Preferably, the model training and optimization process for the rhythmic pattern illumination demand decision tree model based on weighted illumination-pattern state assessment interaction feature data includes:

[0103] The weighted illumination-mode state assessment interaction feature data is transmitted to the circadian rhythm light demand decision tree model. In the circadian rhythm light demand decision tree model, the gradient descent algorithm is used to learn the optimal light demand node for each biological circadian rhythm model from the weighted illumination-mode state assessment interaction feature data, thereby generating an optimized circadian rhythm light demand decision tree model.

[0104] This invention transmits weighted illumination-mode state assessment interactive feature data to a decision tree model. By comprehensively considering the weight information of different illumination parameters, the decision tree model can make more accurate illumination demand predictions based on each weighted parameter. This ensures that the illumination adjustment model rationally allocates the importance of each illumination parameter during the decision-making process, improving the responsiveness and targeting of the decision tree model, thus making illumination adjustment more precise and personalized. Through the gradient descent algorithm, the decision tree model learns based on the weighted illumination-mode state assessment data to find the optimal illumination demand node adapted to different circadian rhythm modes. The application of the gradient descent algorithm allows the model to continuously optimize its illumination demand prediction, quickly converging to the optimal solution, improving the accuracy and adaptability of the decision tree model. This enables the system to better adjust illumination parameters according to individual user differences and dynamic changes, providing more precise illumination adjustment solutions. After learning and optimization through the gradient descent algorithm, the generated optimized circadian rhythm mode illumination demand decision tree model can more accurately predict the optimal illumination demand under different circadian rhythm modes, making the illumination adjustment process more intelligent and personalized, and better meeting the actual needs of users under different circadian rhythm states.

[0105] In this embodiment of the invention, during the model training and optimization process, the weighted illumination-mode state evaluation interaction feature data, after weight assignment, is transmitted to the circadian rhythm illumination demand decision tree model. This data is loaded into the model according to a predefined input format, including weight information for different illumination parameters (such as light intensity, color temperature, and duration) under each circadian rhythm state. The decision tree models corresponding to various circadian rhythm states receive this weighted data for further use in the illumination demand node learning process. Data transmission must ensure consistent format and time series alignment to ensure accurate input of the correlation information between different circadian rhythm states and illumination conditions into the decision tree model. After receiving the weighted illumination-mode state evaluation interaction feature data, the decision tree model enters the training and optimization phase, using the gradient descent algorithm to learn the model's nodes. The gradient descent algorithm calculates the error between the model's predicted output and the actual illumination demand data (usually calculated using a loss function), and adjusts the parameters of the decision tree in reverse according to the gradient of the error, gradually updating the weight and decision path of each node. Through a continuous iterative learning process, the optimal light demand node for each biorhythm pattern is found at each branch of the tree. The model gradually approaches the optimal decision path, enabling it to more accurately predict a user's light demand under specific biorhythmic states. After multiple iterations of the gradient descent algorithm, each node of the model has undergone sufficient learning, generating an optimized biorhythmic pattern light demand decision tree model based on the learning results. This optimized model has adjusted the light demand for different biorhythmic patterns through the node learning process, ensuring that each node accurately reflects the relationship between light parameters and the biorhythmic pattern. The final optimized biorhythmic pattern light demand decision tree model possesses more accurate and efficient predictive capabilities, providing users with more personalized and dynamic light demand suggestions.

[0106] Preferably, the rhythm pattern trend model construction module includes the following functions:

[0107] Based on the biological rhythm pattern data, the biological vital signs time series data are extracted to generate biological rhythm pattern time series feature data.

[0108] Perform periodicity feature analysis on the time-series characteristic data of circadian rhythm patterns to generate periodic characteristic data of circadian rhythm patterns;

[0109] A model for the changing trend of biological rhythm patterns is designed based on the periodic characteristic data of biological rhythm patterns, and a model for the changing trend of biological rhythm patterns is generated.

[0110] This invention combines and analyzes biorhythm pattern data with biosignature time-series data to extract time-series characteristic data of user biorhythm patterns. This data reflects the changing trends of the user's physiological state at different time points, enabling the system to capture the dynamic changes in the user's biorhythm. This provides a more comprehensive temporal dimension for subsequent light regulation, making light regulation more adaptable to the user's actual rhythm fluctuations and improving the accuracy of regulation. Analysis of the periodic characteristics of the time-series characteristic data can identify the periodic patterns of the user's biorhythm patterns, understand the changing trends of the user's biorhythm in different time periods, and predict the patterns of changes in the user's physiological state over time. This ensures that light regulation can be adjusted in advance according to the user's periodic rhythm needs, making light regulation more accurate and forward-looking. Using the periodic characteristic data of biorhythm patterns, a trend model reflecting changes in the user's biorhythm is designed to predict future changes in the user's biorhythm, providing advance strategic guidance for light regulation. This trend model design enables the anticipation of user rhythm changes and corresponding adjustments, improving the intelligence and adaptability of the light system, making light regulation more predictive and accurate.

[0111] In this embodiment of the invention, time-series features are extracted from biorhythm pattern data. Time-series analysis techniques (such as autocorrelation analysis and Fast Fourier Transform, FFT) are used to identify key change patterns in the biorhythm data, such as the fluctuation trends of heart rate and body temperature over time. These biorhythm time-series data are preprocessed, including denoising, smoothing, and normalization operations, to ensure data continuity and accuracy. Then, feature extraction algorithms (such as peak detection and local extremum analysis) and biorhythm pattern data are used to identify key feature points in the biorhythm patterns, such as peak periods, trough periods, and cycle start points, thereby generating biorhythm pattern time-series feature data that reflects the periodic changes in the user's physiological rhythm at different times, serving as the basis for subsequent periodic analysis. Periodic feature analysis is performed on the biorhythm pattern time-series feature data, using periodic detection algorithms (such as periodogram analysis and Discrete Wavelet Transform, DWT) to identify periodic change features in the biorhythm patterns. By analyzing the repetitive fluctuations of biorhythms over time, the main cycle length and frequency of each biorhythm pattern are determined, and its inherent periodic structure is identified. The generated circadian rhythm pattern periodic feature data contains information on the cycle length, phase, and amplitude of different physiological states (such as sleep, activity, and rest), further revealing how the user's circadian rhythm cycle changes over time. This periodic feature data is used to model the long-term trend of circadian rhythm changes. Based on the circadian rhythm pattern periodic feature data, a circadian rhythm pattern change trend model is designed. Time-series prediction algorithms (such as ARIMA models and Long Short-Term Memory networks LSTM) are used to analyze the periodic feature data and predict future trends in circadian rhythm changes. By learning from the trajectory of past circadian rhythm pattern changes, a trend model capable of predicting future rhythm changes is constructed. This trend model captures long-term change signals in circadian rhythm patterns, such as cycle lengthening, shortening, or phase shifts, ultimately generating a circadian rhythm pattern change trend model.

[0112] Preferably, the rhythm-regulating illumination parameter control module includes the following functions:

[0113] Real-time circadian rhythm pattern analysis and processing are performed based on biosign data to generate real-time circadian rhythm pattern data.

[0114] Real-time circadian rhythm pattern data is transmitted to a circadian rhythm pattern change trend model for circadian rhythm pattern change trend prediction processing, generating circadian rhythm pattern change trend data.

[0115] The time-series characteristic data of circadian rhythm patterns are processed to extract circadian rhythm pattern change features, generating circadian rhythm pattern change feature data; based on the circadian rhythm pattern change feature data, the real-time update time period of each circadian rhythm pattern is designed, generating real-time update time period data of circadian rhythm patterns.

[0116] Real-time update processing of biorhythm pattern change trend data is performed based on real-time update time period data of biorhythm pattern to generate real-time biorhythm pattern change trend data.

[0117] Real-time biological rhythm pattern change trend data is transmitted to the optimized rhythm pattern light demand decision tree model for intelligent analysis and processing of rhythm light demand, generating intelligent rhythm light demand data.

[0118] Based on the intelligent rhythmic lighting demand data, design intelligent rhythmic lighting control parameters and generate intelligent rhythmic lighting control parameters.

[0119] The operation of rhythmic lighting parameter control is performed based on intelligent rhythmic lighting control parameters.

[0120] This invention provides real-time analysis of biometric data, capturing the user's current circadian rhythm state and rapidly adapting to the user's immediate needs. This ensures that light regulation accurately responds to the user's current rhythm. The generation of real-time circadian rhythm pattern data provides real-time data support for subsequent light regulation, enhancing its dynamism and flexibility. By transmitting real-time circadian rhythm pattern data to a trend model, it predicts future trends in the user's circadian rhythm, making light regulation no longer a passive response to the current state but proactive, allowing for advance adjustments to address impending changes in the user's circadian rhythm and improving the system's predictability and regulatory effectiveness. Furthermore, by extracting the changing characteristics of circadian rhythm patterns, identifying potential trends in the user's circadian rhythm, and adjusting the understanding of the user's circadian rhythm in a timely manner, and by rationally designing the real-time update time periods for each circadian rhythm pattern based on the changing characteristics, adjustments are made at appropriate time periods according to the user's latest state, further improving the accuracy and timeliness of light regulation. By dynamically updating data over real-time time periods, the system corrects the trend data of the circadian rhythm pattern, ensuring that predictions more closely reflect the user's current actual state. This real-time update process enables rapid response to changes in the circadian rhythm and adjusts lighting requirements based on the latest trend predictions, guaranteeing the real-time and forward-looking nature of lighting regulation. The changed trend data is transmitted to an optimized decision tree model, which performs intelligent analysis based on the user's current state and trend predictions. Combining historical data and real-time trends, it provides highly accurate guidance for lighting regulation schemes, ensuring that the prediction of lighting requirements better matches the user's personalized circadian rhythm changes, thus improving the accuracy and intelligence of the lighting regulation scheme. Lighting control parameters are designed based on the intelligent analysis results, generating a lighting regulation scheme best suited to the user's current circadian rhythm state. The intelligent parameter design ensures a high degree of personalization and accuracy in lighting regulation, enabling rapid adjustment of lighting parameters to ensure the user receives the best lighting environment, thereby improving user comfort and experience. The system executes adjustment operations based on the intelligently designed lighting control parameters, ensuring the effective implementation of the lighting regulation scheme. This allows the lighting environment to adapt to the user's circadian rhythm needs in real time, further improving the response speed and accuracy of lighting control, and helping users better adapt to the environment.

[0121] In this embodiment of the invention, real-time biometric data (such as heart rate, respiratory rate, body temperature, etc.) are collected and combined with existing biorhythm pattern analysis methods to perform real-time biorhythm pattern analysis and processing. The current biorhythm pattern is extracted from the real-time data to generate real-time biorhythm pattern data. This data reflects the user's current physiological state, ensuring that light regulation can be adjusted based on the user's real-time circadian rhythm needs. The generated real-time biorhythm pattern data is transmitted to a biorhythm pattern change trend model for biorhythm pattern change trend prediction processing. Based on the previously designed trend model (such as ARIMA or LSTM model), the trend of the real-time biorhythm pattern data is predicted, calculating the direction and magnitude of biorhythm changes over a future period, generating biorhythm pattern change trend data. This helps the system predict the possible future trends of the user's biorhythm, providing forward-looking guidance for subsequent light regulation. This system extracts circadian rhythm change features from existing circadian rhythm time-series data. By analyzing key nodes in the circadian rhythm time series (such as periodic peaks, troughs, and turning points) and combining them with trend model predictions, key feature data reflecting circadian rhythm changes is extracted, generating circadian rhythm change feature data. This feature data reveals significant potential changes in the user's circadian rhythm in the future, providing a basis for further lighting control strategy design. After acquiring the circadian rhythm change feature data, the system designs real-time update time periods for each circadian rhythm pattern based on these features. According to the periodicity and current state of the rhythm, suitable key time periods for rhythm regulation are designed. These time periods typically include turning points or periods of greatest fluctuation in the circadian rhythm, generating real-time update time period data for circadian rhythm patterns. This data determines the optimal time window for lighting regulation to ensure that lighting intervention effectively affects the user's circadian rhythm. After obtaining real-time updated time-segment data of the circadian rhythm pattern, the system updates the previously generated circadian rhythm pattern change trend data based on these time periods. This is achieved by recalculating the real-time impact of light parameters on the circadian rhythm and dynamically adjusting the prediction model to generate real-time circadian rhythm pattern change trend data. This update ensures the model can promptly reflect the user's current circadian rhythm changes and provides the latest rhythm trend information for subsequent light demand analysis. The real-time circadian rhythm pattern change trend data is then transmitted to the optimized circadian rhythm pattern light demand decision tree model for intelligent analysis of circadian light demand. Based on the current circadian rhythm change trend and historical data, the model intelligently analyzes the user's light demand parameters (such as light intensity, duration, color temperature, etc.) under different circadian rhythm states, generating intelligent circadian light demand data to ensure that users can obtain the optimal light environment at different times and under different circadian rhythm states.Based on intelligent rhythmic lighting demand data, specific intelligent rhythmic lighting control parameters are designed. According to parameters in the lighting demand data (such as light intensity, color temperature, and time), combined with the user's current environmental lighting conditions, the lighting control strategy is dynamically adjusted to generate intelligent rhythmic lighting control parameters, ensuring that lighting adjustment matches the user's current biorhythm needs. After the intelligent rhythmic lighting control parameters are determined, the final execution of the rhythmic lighting parameter control operation is carried out. By controlling lighting equipment (such as adjusting the brightness and color temperature of lamps), the ambient lighting conditions are automatically adjusted to match the user's current biorhythm state. The lighting effect is monitored periodically, and the lighting parameters are continuously adjusted based on real-time feedback to ensure that the adjustment effect achieves the expected results, helping the user maintain a healthy biorhythm.

[0122] This specification provides a method for controlling rhythm-regulated illumination parameters, used to execute the rhythm-regulated illumination parameter control system as described above. The method includes:

[0123] Step S1: Collect user perception data based on smart wearable sensors to generate user perception data, wherein the user perception data includes ambient light data and biometric data.

[0124] Step S2: Perform biological sign time series identification analysis based on biological sign data to generate biological sign time series data; perform biological rhythm pattern analysis and processing based on biological sign time series data to generate biological rhythm pattern data.

[0125] Step S3: Perform steady-state analysis of the circadian rhythm patterns based on the circadian rhythm pattern data to generate steady-state data of the circadian rhythm patterns; perform state assessment processing of each circadian rhythm pattern based on the steady-state data of the circadian rhythm patterns to generate state assessment data of the circadian rhythm patterns.

[0126] Step S4: Analyze the interaction features between ambient light data and circadian rhythm pattern state assessment data to generate light-pattern state assessment interaction feature data.

[0127] Step S5: Based on the preset decision tree algorithm and the biological rhythm pattern data, design the light demand decision mapping relationship under each biological rhythm pattern to generate a rhythm pattern light demand decision tree model; use the light-state interaction feature data to perform decision tree training and optimization on the rhythm pattern light demand decision tree model to generate an optimized rhythm pattern light demand decision tree model.

[0128] Step S6: Analyze the periodic characteristics of the biological rhythm pattern based on the biological rhythm pattern data and generate periodic characteristic data of the biological rhythm pattern; design a trend model of biological rhythm pattern change based on the periodic characteristic data of the biological rhythm pattern and generate a trend model of biological rhythm pattern change.

[0129] Step S7: Perform real-time biorhythm pattern analysis and processing based on biological symptom data to generate real-time biorhythm pattern data; perform real-time biorhythm pattern change trend analysis and processing based on the biorhythm pattern change trend model to generate real-time biorhythm pattern change trend data; transmit the real-time biorhythm pattern change trend data to the optimized biorhythm pattern light demand decision tree model for intelligent analysis and processing of biorhythm light demand to generate intelligent biorhythm light demand data; execute biorhythm regulation light parameter control operation based on biorhythm light intelligent regulation control parameters.

[0130] The beneficial effects of this application are as follows: This invention collects users' biometric data through smart wearable devices, performs biometric time-series identification analysis and circadian rhythm pattern analysis, generates users' circadian rhythm patterns through multi-level data analysis and clustering methods, and performs precise light adjustment based on each user's circadian rhythm pattern. This real-time perception and analysis based on individual user biometrics effectively solves the problem of inaccurate collection and analysis of user biometric data, ensuring that light adjustment can accurately match the user's actual circadian rhythm needs. By designing a mathematical model of the light demand decision mapping relationship through decision tree algorithm and circadian rhythm pattern data, and training and optimizing the model, intelligent decision analysis of light adjustment needs is realized. Unlike fixed patterns or preset schedules, it can dynamically adjust light parameters according to the user's current biometric and circadian rhythm state, realizing personalized and intelligent light control. This intelligent decision-making capability ensures that light adjustment is more in line with the user's actual needs, improving the user experience. By using a biorhythm pattern change trend model and real-time updates of biorhythm pattern trend predictions, the system performs periodic feature analysis and trend prediction of users' biorhythms. This not only allows for adjustments to the current biorhythm but also enables proactive adjustments to light parameters based on future rhythm change trends, achieving adaptive light control. This solves the problem of only being able to make simple adjustments to the current state, providing users with more precise rhythm management and health interventions, especially with more significant effects in long-term light regulation.

[0131] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0132] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A circadian-modulating lighting parameter control system, characterized by, The method comprises the following modules: A user perception data acquisition module is configured to acquire user perception data based on an intelligent wearable sensor, and generate user perception data, wherein the user perception data comprises environmental illumination data and biological sign data; A biological rhythm pattern analysis module is configured to analyze biological sign timing based on the biological sign data, and generate biological sign timing data; analyze biological sign instantaneous state feature data based on the biological sign timing data, and generate biological sign instantaneous state feature data; perform biological sign instantaneous state clustering analysis on the biological sign instantaneous state feature data, and generate biological sign instantaneous state clustering data; and perform biological rhythm pattern analysis based on the biological sign instantaneous state clustering data, and generate biological rhythm pattern data; A biological rhythm pattern state evaluation module is configured to perform biological sign instantaneous state clustering baseline feature analysis based on the biological sign instantaneous state clustering data, and generate biological sign instantaneous state clustering baseline feature data; Perform stable state analysis of the biological rhythm pattern based on the biological rhythm pattern data on the biological sign instantaneous state clustering baseline feature data, and generate biological rhythm pattern stable state data; perform clustering subset bias quantification analysis on the biological sign instantaneous state clustering data, and generate biological sign clustering subset bias quantification data; and perform state evaluation processing of each biological rhythm pattern based on the biological rhythm pattern stable state data on the biological sign clustering subset bias quantification data, and generate biological rhythm pattern state evaluation data; An interaction feature analysis module is configured to perform data synchronization processing of environmental illumination and biological rhythm pattern state evaluation based on the environmental illumination data and the biological rhythm pattern state evaluation data, and generate synchronized illumination-pattern state evaluation data; and perform interaction feature analysis of environmental illumination and biological rhythm pattern state evaluation based on the synchronized illumination-pattern state evaluation data, and generate illumination-pattern state evaluation interaction feature data; An illumination demand model construction module is configured to establish a decision mapping relationship between biological rhythm patterns and illumination demands based on a preset decision tree algorithm, and generate an initial rhythm pattern illumination demand decision tree model; and perform decision tree pruning processing of each biological rhythm pattern on the initial rhythm pattern illumination demand decision tree model based on the biological rhythm pattern data, and generate a rhythm pattern illumination demand decision tree model; Perform illumination parameter weight assignment processing of each biological rhythm pattern on the illumination-pattern state evaluation interaction feature data, and generate weighted illumination-pattern state evaluation interaction feature data; Perform model training and optimization processing on the rhythm pattern illumination demand decision tree model based on the weighted illumination-pattern state evaluation interaction feature data, and generate an optimized rhythm pattern illumination demand decision tree model; A rhythm pattern trend model construction module is configured to extract biological rhythm pattern timing feature data from the biological rhythm pattern data based on the biological sign timing data, and generate biological rhythm pattern timing feature data; Perform periodic feature analysis of the biological rhythm pattern based on the biological rhythm pattern timing feature data, and generate biological rhythm pattern periodic feature data; Perform biological rhythm pattern change trend model design based on the biological rhythm pattern periodic feature data, and generate a biological rhythm pattern change trend model; and Perform biological rhythm pattern change trend model design based on the biological rhythm pattern periodic feature data, and generate a biological rhythm pattern change trend model. The rhythm regulation light parameter control module performs real-time biological rhythm pattern analysis processing according to the biological sign data, and generates real-time biological rhythm pattern data; The real-time biological rhythm pattern data is transmitted to the biological rhythm pattern change trend model for biological rhythm pattern change trend prediction processing, and biological rhythm pattern change trend data is generated; The biological rhythm pattern change feature data is extracted from the biological rhythm pattern time sequence feature data, and biological rhythm pattern change feature data is generated; According to the biological rhythm pattern change feature data, the real-time update time period data of each biological rhythm pattern is designed, and the biological rhythm pattern real-time update time period data is generated; Based on the biological rhythm pattern real-time update time period data, the real-time update processing is performed on the biological rhythm pattern change trend data, and the real-time biological rhythm pattern change trend data is generated; The real-time biological rhythm pattern change trend data is transmitted to the optimized rhythm pattern light demand decision tree model for intelligent analysis processing of the rhythm light demand, and the intelligent rhythm light demand data is generated; According to the intelligent rhythm light demand data, the intelligent rhythm light control parameter design is performed, and the intelligent rhythm light control parameter is generated; Based on the intelligent rhythm light control parameter, the rhythm regulation light parameter control operation is performed.

2. The circadian-modulating lighting parameter control system of claim 1, wherein, The light parameter weight assignment processing of each biological rhythm pattern on the light-mode state evaluation interaction feature data includes: The time sequence data extraction is performed on the light-mode state evaluation interaction feature data, and the light-mode state evaluation interaction feature time sequence data is generated; According to the light-mode state evaluation interaction feature time sequence data, the time sequence light sensitive factor data is generated by analyzing the time sequence light sensitive factor of each biological rhythm pattern; Based on the time sequence light sensitive factor data, the light parameter weight assignment processing of each biological rhythm pattern on the light-mode state evaluation interaction feature data is performed, and the weighted light-mode state evaluation interaction feature data is generated.

3. The circadian rhythm modulating lighting parameter control system of claim 1, wherein, The model training and optimization processing of the rhythm pattern light demand decision tree model based on the weighted light-mode state evaluation interaction feature data includes: The weighted light-mode state evaluation interaction feature data is transmitted to the rhythm pattern light demand decision tree model, and the best light demand node of each biological rhythm pattern is learned by the gradient descent algorithm in the rhythm pattern light demand decision tree model based on the weighted light-mode state evaluation interaction feature data, and the optimized rhythm pattern light demand decision tree model is generated.

4. A method for controlling circadian rhythm illumination parameters, characterized in that, The rhythm regulation light parameter control system for performing the method according to claim 1, the method comprising: Step S1: collecting user perception data according to intelligent wearable sensors, and generating user perception data, wherein the user perception data includes environmental light data and biological sign data; Step S2: performing biological sign time sequence identification analysis according to the biological sign data, and generating biological sign time sequence data; performing biological rhythm pattern analysis processing according to the biological sign time sequence data, and generating biological rhythm pattern data; Step S3: Based on the biological rhythm mode data, a steady state analysis of the biological rhythm mode is performed to generate biological rhythm mode steady state data; based on the biological rhythm mode steady state data, a state evaluation process of each biological rhythm mode is performed to generate biological rhythm mode state evaluation data; Step S4: According to the environmental light data and the biological rhythm mode state evaluation data, an interactive feature analysis of the environmental light and the biological rhythm mode state evaluation is performed to generate light-mode state evaluation interactive feature data; Step S5: Based on the preset decision tree algorithm and the biological rhythm mode data, a light demand decision mapping relationship design under each biological rhythm mode is performed to generate an initial rhythm mode light demand decision tree model; the light-mode state evaluation interactive feature data is used to perform decision tree training optimization processing on the initial rhythm mode light demand decision tree model to generate an optimized rhythm mode light demand decision tree model; Step S6: According to the biological rhythm mode data, a periodicity feature analysis of the biological sign time series data is performed to generate biological rhythm mode periodicity feature data; based on the biological rhythm mode periodicity feature data, a biological rhythm mode change trend model is designed to generate a biological rhythm mode change trend model; Step S7: According to the biological sign data, an instant biological rhythm mode analysis process is performed to generate instant biological rhythm mode data; based on the biological rhythm mode change trend model, an instant biological rhythm mode change trend analysis process is performed on the instant biological rhythm mode data to generate instant biological rhythm mode change trend data; the instant biological rhythm mode change trend data is transmitted to the optimized rhythm mode light demand decision tree model to perform a rhythm light demand intelligent analysis process to generate intelligent rhythm light demand data; based on the intelligent rhythm light demand data, a rhythm adjustment light parameter control job is performed.

Images