Home AI Weather Lighting System and Method

Abstract

This invention discloses a home AI weather lighting system and method, relating to the field of smart home technology. The system includes a data acquisition and processing module, a fusion decision-making module, an optimization control module, a joint control module, and a collaborative control module. By collecting and preprocessing outdoor weather data, a structured weather database is obtained; based on the structured weather database and user behavior logs, a hypergraph fusion decision is performed to obtain dynamic environmental coding; based on the dynamic environmental coding, a multi-objective optimization function is constructed to solve for the optimal control strategy, generating a Pareto optimal solution set; through the Pareto optimal solution set, spectral sound field collaborative optimization is performed to obtain a photoacoustic joint control strategy; based on the photoacoustic joint control strategy, multi-modal control signal analysis and collaborative optimization are performed to obtain the final multi-device collaborative control command. This invention achieves highly realistic simulation of outdoor weather environments, balancing user comfort and system energy consumption, and provides users with an immersive weather scenario experience.

Description

Technical Field

[0001] This invention relates to the field of smart home technology, and in particular to a home AI weather lighting system and method. Background Technology

[0002] With the rapid development of the Internet of Things, artificial intelligence, and sensing technologies, smart home systems are becoming increasingly popular, providing users with a more convenient, comfortable, and personalized living environment. Among them, lighting systems, as an important component of smart homes, have gradually evolved from simply meeting basic lighting needs to creating ambiance, providing healthy lighting, and simulating specific scenarios.

[0003] Existing technologies have explored several solutions that link outdoor weather conditions with indoor lighting. For example, sensors can be used to collect outdoor light intensity data, and then the brightness of indoor lights can be adjusted to simulate the outdoor lighting environment. However, these solutions have several shortcomings: 1. Limited Data Dimensions: Most existing solutions focus only on light intensity, neglecting changes in spectral distribution, color temperature, and accompanying acoustic environment (such as rain and wind sounds) caused by weather phenomena (e.g., sunshine, rain, snow, fog). Therefore, the simulated indoor environment differs significantly from the "realism" of actual outdoor weather.

[0004] 2. Simple decision-making mechanism: The decision-making logic of existing solutions is usually based on preset rules (such as "if the outdoor light is weak, then the indoor light is bright"), which cannot comprehensively consider the user's behavior habits, personalized preferences, and the collaborative work between multiple devices (such as lighting, sound, and curtains), resulting in rigid control strategies and poor user experience.

[0005] 3. Lack of multi-objective optimization: Existing systems usually only aim to achieve a single function (such as simulating weather), failing to simultaneously consider multiple conflicting objectives such as user visual comfort, thermal comfort, and system energy consumption, making it difficult to achieve the optimal balance between energy saving and user experience.

[0006] 4. Insufficient multimodal collaboration: Existing technologies give little consideration to the collaboration between hearing and vision. They usually control sound and light as two independent systems, lacking the ability to deeply integrate and optimize them at the perception level, and thus failing to provide users with an immersive and highly realistic weather scenario experience.

[0007] Therefore, there is an urgent need for a home AI weather lighting system that can deeply integrate multi-source environmental data, understand user behavior, and perform collaborative optimization control of multi-modal devices such as lighting and sound fields. Summary of the Invention

[0008] The present invention aims to solve the problems existing in the prior art and provide a home AI weather lighting system and method to achieve immersive and highly realistic simulation of outdoor weather environment, while taking into account user comfort and system energy consumption.

[0009] To achieve the above objectives, the present invention provides a home AI weather lighting system, comprising: The data acquisition and processing module is configured to collect and preprocess outdoor weather data to obtain a structured weather database. The fusion decision module is configured to perform hypergraph fusion decision-making based on the structured weather database and user behavior logs to obtain dynamic environmental codes; The optimized control module is configured to construct a multi-objective optimization function based on the dynamic encoding of the environment to solve for the optimal control strategy and generate a Pareto optimal solution set. The joint control module is configured to perform spectral acoustic field co-optimization through the Pareto optimal solution set to obtain a photoacoustic joint control strategy. The collaborative control module is configured to perform multi-modal control signal analysis and collaborative optimization based on the aforementioned photoacoustic joint control strategy to obtain the final multi-device collaborative control command.

[0010] Furthermore, the acquisition and processing module includes: Multispectral sensors are used to collect outdoor spectral distribution and light intensity in real time. A weather sensor array is used to collect temperature, humidity, and fog levels. Microphone array, used to collect ambient sound waveform data; The acquisition and processing module is also used to perform time-series matching, calibration and denoising on the above-mentioned multi-source heterogeneous data to form a standardized weather data stream, and store it in the time-series database to form the structured weather database.

[0011] Furthermore, the fusion decision module includes: The hypergraph construction unit is configured to construct a weather configuration hypergraph structure based on user-defined functional parameters and the structured weather database. The node set of the hypergraph structure includes environmental measurement nodes, user configuration nodes, and derived feature nodes. The hyperedge set is used to capture unpaired association relationships between nodes. The hypergraph attention processing unit is configured to calculate the importance weight of each node to the hyperedge and the importance weight of each node to the global decision through the hypergraph attention network, so as to generate the hypergraph fusion decision signal. The temporal coding unit is configured to temporally combine the hypergraph fusion decision signal with multidimensional low-level static features extracted from the structured weather database, and input them into a spatiotemporal convolutional memory network to extract dynamic environmental evolution patterns, thereby forming the environmental dynamic code.

[0012] Furthermore, the optimization control module includes: The multi-objective building block is configured to construct a multi-objective optimization function that includes at least environmental realism indicators, comfort indicators, and energy consumption indicators. The constraint processing unit is configured to introduce device physical constraints as well as user and scenario constraints. The Pareto solver is configured to solve the multi-objective optimization function under the constraints and generate a Pareto optimal solution set consisting of multiple non-dominated solutions.

[0013] Furthermore, the environmental realism index is formed by fusing spectral realism and sound field realism, wherein, The spectral accuracy is obtained by constructing a spectral generation function based on a Gaussian function, and calculating the difference between the indoor generated spectrum and the target outdoor spectrum based on the spectral feature values ​​in the environmental dynamic coding to obtain the spectral accuracy. The sound field realism is obtained by calculating the similarity between the acoustic feature values ​​of the indoor generated sound and the outdoor acoustic feature values ​​in the dynamic environmental coding.

[0014] Furthermore, the comfort index is a fusion of thermal comfort and visual comfort, wherein, The thermal comfort index is obtained in the following way: based on the temporal feature value and predicted feature value in the dynamic coding of the environment, the thermal comfort index is obtained through the thermal comfort objective function and the predicted average voting model; The visual comfort is obtained by setting a dynamic weight for the timestamp, creating a time-varying visual comfort function based on the dynamic weight of the timestamp, and combining the temporal feature value and spectral feature value in the dynamic coding of the environment to obtain the visual comfort.

[0015] Furthermore, the joint control module includes: The light source driving vector generation unit is configured to transform the strategy selected in the Pareto optimal solution set into the optimal spectral driving vector based on the LED spectral response function; The three-dimensional sound field reconstruction unit is configured to match sound source parameters from a preset sound source library based on the acoustic feature values ​​in the dynamic environmental coding, and calculate complex sound pressure values ​​on discretized indoor space grid points using a sound field reconstruction algorithm to form a three-dimensional sound field distribution matrix. The joint optimization solution unit is configured to establish a regularized multiple loss function that includes spectral matching error terms and sound field reconstruction error terms, and solves it using the alternating direction multiplier method to obtain a photoacoustic joint control strategy that simultaneously includes illumination control components and sound field control components.

[0016] Furthermore, the light source driving vector generation unit includes: The spectral response modeling subunit is configured to experimentally measure the output spectrum under different combinations of driving currents and use a multivariate regression method to fit the LED spectral response function. The decision subunit is configured to define an ideal policy point, calculate the standardized Euclidean distance from each policy in the Pareto optimal solution set to the ideal policy point, and select the policy with the smallest distance as the selected policy. The spectral reconstruction subunit is configured to obtain the optimal spectral driving vector by solving the spectral reconstruction optimization model based on the target spectrum of the selected strategy and the LED spectral response function.

[0017] Furthermore, the collaborative control module includes: The driving instruction generation unit is configured to obtain adaptive spectral driving instructions by constructing and solving a spectral mapping optimization function based on the illumination control component in the photoacoustic joint control strategy. The multimodal fusion unit is configured to coordinate and format the adaptive spectral driving command, the sound field control component in the photoacoustic joint control strategy, and other environmental device control commands to generate a preliminary multi-device control signal set. The discrete event scheduling unit is configured to assign a timestamp and transition duration to each control instruction in the initial multi-device control signal set, forming a conflict-free instruction sequence, which is then output as the final multi-device collaborative control instruction.

[0018] To achieve the above objectives, the present invention also provides a home AI weather lighting method, applied to the home AI weather lighting system as described above, comprising the following steps: Outdoor weather data is collected and preprocessed to obtain a structured weather database; Based on the structured weather database and user behavior logs, a hypergraph fusion decision is made to obtain a dynamic environmental code; Based on the dynamic coding of the environment, a multi-objective optimization function is constructed to solve the optimal control strategy and generate a Pareto optimal solution set. The spectral acoustic field is co-optimized using the Pareto optimal solution set to obtain a photoacoustic joint control strategy; Based on the aforementioned photoacoustic joint control strategy, multimodal control signal analysis and collaborative optimization are performed to obtain the final multi-device collaborative control command.

[0019] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. High-fidelity weather simulation: This invention collects multi-dimensional outdoor multispectral, meteorological, and acoustic data, and utilizes hypergraph fusion decision-making and dynamic environment coding technologies to accurately capture and characterize the dynamic characteristics of complex weather environments, providing rich and accurate inputs for subsequent control, thereby significantly improving the realism of outdoor weather simulation. 2. Personalization and Multi-Objective Optimization: This invention constructs a multi-objective optimization function that incorporates environmental realism, comfort, and energy consumption by combining user behavior logs, and solves for the Pareto optimal solution set. This enables the system to find the optimal balance between different objectives based on the user's personalized preferences, providing a personalized control strategy that is both comfortable and energy-efficient. 3. Immersive Multimodal Collaborative Experience: This invention achieves deep collaborative optimization of the illumination spectrum and sound field through a joint control module and a collaborative control module. By solving and finely scheduling the light and sound joint control strategy, an immersive weather scenario with a high degree of visual and auditory unity can be created, greatly enhancing the user's sensory experience; 4. Efficient and robust control: This invention employs advanced optimization algorithms such as Pareto optimal solution set and alternating direction multiplier method, which can efficiently handle high-dimensional and multi-constraint optimization problems, ensuring that the generated final control commands can meet the requirements of complex scenarios and conform to the physical constraints of the equipment, and the system has strong robustness. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the overall architecture of the home AI weather lighting system provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of the fusion decision module provided in an embodiment of the present invention; Figure 3 A schematic diagram illustrating the workflow of the optimized control module provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of the joint control module provided in an embodiment of the present invention; Figure 5 This is a flowchart illustrating the home AI weather lighting method provided in an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it.

[0022] Example 1 Please refer to Figure 1This embodiment provides a home AI weather lighting system. The system includes: a data acquisition and processing module 100, a fusion decision-making module 200, an optimization control module 300, a joint control module 400, and a collaborative control module 500.

[0023] The data acquisition and processing module 100 is responsible for acquiring and preprocessing outdoor weather data. Specifically, the data acquisition and processing module 100 includes a multispectral sensor, a meteorological sensor array, and a microphone array. The multispectral sensor is used to acquire the spectral power distribution and light intensity of the outdoor environment in real time; the meteorological sensor array is used to acquire temperature, humidity, and haze values ​​used to characterize the concentration of fog or haze; and the microphone array is used to acquire waveform data of ambient sound.

[0024] The acquisition and processing module 100 utilizes multi-source data synchronous acquisition technology to ensure time consistency for these multi-source heterogeneous data, and employs a timestamp alignment algorithm for time series matching to obtain a multimodal weather synchronous dataset. Subsequently, a multi-sensor fusion algorithm based on extended Kalman filtering is used to establish a system state-space model including light intensity, spectral distribution, temperature and humidity, fog, and sound characteristics. The raw data is iteratively optimized through a prediction-correction loop to suppress noise interference and statistically detect and remove outliers, forming a standardized weather data stream. This data stream is stored in a time-series database, employing a time-partitioned storage strategy and establishing a multi-dimensional joint index structure based on spatial location, sensor type, and physical quantity dimensions, ultimately forming a structured weather database that provides a data foundation for subsequent modules.

[0025] The fusion decision module 200 has the following internal structure: Figure 2 As shown, this module is used for hypergraph fusion decision-making based on a structured weather database and user behavior logs to obtain dynamic environmental codes. Specifically, this module includes: Hypergraph Construction Unit 210: Extracts user-defined function parameters (such as color temperature, brightness level, air conditioning temperature, speaker volume level, etc. at time t) from user behavior logs, and matches these parameters with the structured weather database along the time dimension using a timestamp alignment algorithm. Based on this, a weather configuration hypergraph structure is constructed. The node set of this hypergraph contains three types of nodes: Environmental measurement nodes: These represent raw physical quantities collected at a specific moment in a structured weather database, such as K-dimensional spectral intensity vectors, illuminance values, temperature values, humidity values, and fog values.

[0026] User configuration node: Represents a function parameter that is actively set by the user.

[0027] Derived feature nodes: High-level semantic features derived from the original physical quantities and user-defined functional parameters, such as time period nodes derived from timestamps.

[0028] Hyperedge sets are used to capture the complex unpaired associations between these nodes. A hypergraph clustering algorithm is used to analyze all node data, identifying which nodes frequently co-occur or are strongly correlated, and generating a hyperedge from the set of these nodes.

[0029] Hypergraph Attention Processing Unit 220: Processes the constructed hypergraph through a hypergraph attention network. For a hyperedge α, the importance weights of all nodes inside it are calculated using the following formula: ; Where, α c,α Let represent the importance weight of node c to hyperedge α; b is the learnable attention vector; W is the learnable weight matrix; h c h is the feature vector of node c; α α is the initial feature of the hyperedge, usually taken as the average feature of the nodes connected by the hyperedge; σ is the activation function (such as LeakyReLU). This indicates a splicing operation.

[0030] Then, all hyperedge information containing node v is aggregated to update the node features: ; in, is the updated feature vector of node c; E(c) is the set of all hyperedges containing node c; It is the importance weight of hyperedge α to node c, which is calculated through another attention mechanism.

[0031] Finally, a learnable global query vector q is introduced, which interacts with the attention of all updated node features to calculate the importance weight of each node for the final global decision: ; in, It is the importance weight of node c to the global decision; V is a learnable weight matrix; V is the set of all nodes. The final hypergraph fusion decision signal G is obtained by weighted summation of the features of all nodes: ; The signal G is a dense vector of fixed dimensions that integrates key information from three aspects: real-time environmental state, long-term user preferences, and current context.

[0032] Temporal coding unit 230: First, based on a structured weather database, multi-dimensional low-level static features are calculated and obtained through standardized algorithms and specifications. These features include: Spectral characteristics: such as outdoor light correlated color temperature (calculated according to CIE1931 standard), deviation from blackbody radiation, wavelength and intensity characteristics of main spectral peaks, and melanopsin ratio that affects human circadian rhythm (calculated according to CIES026:2018 standard).

[0033] Acoustic features include sound category coding (output by an acoustic event detection model), Mel frequency cepstral coefficients (MFCC), and equivalent sound level (calculated according to IEC 61672:2013).

[0034] Basic environmental parameters: such as spectral intensity vector, illuminance value, temperature value, humidity value, and haze value.

[0035] Subsequently, the hypergraph fusion decision signal G is combined with the aforementioned multidimensional low-level static features in chronological order to form a multivariate time series. This multivariate time series is then input into a spatiotemporal convolutional memory network (such as ConvLSTM) for deep temporal modeling. This network captures short-term local dependency patterns in the sequence through one-dimensional causal convolution, learns long-term dynamic evolution patterns and controls noise using gated recurrent units, and focuses on key time steps using an attention mechanism. The network ultimately outputs the environmental dynamic encoding E. t It is a dynamic vector that integrates recent historical trends with current characteristics, specifically including: 1. Temporal characteristic values: the changing trend of encoding environment parameters (such as the first / second derivative of temperature). Periodicity (such as the intensity of circadian rhythm Et) period ) and volatility (Et) volat ); 2. Spectral eigenvalues: These encode key attributes of the outdoor spectrum, such as correlated color temperature estimates. Deviation from blackbody radiation, wavelength and intensity of major spectral peaks Black-pixel luminous efficacy ratio ; 3. Acoustic features: key attributes that encode outdoor sound fields, such as sound category encoding. Mel frequency cepstral coefficients Et MFCC Equivalent sound level Et loud ; 4. Predictive Feature Values: The network's predictions of short-term future states, such as the predicted temperature within the next few minutes. and predicted values ​​of illuminance .

[0036] The optimized control module 300 has the following workflow: Figure 3 As shown, this module is used to construct and solve multi-objective optimization functions based on dynamic environment encoding, generating a Pareto optimal solution set. Specifically, this module includes: Multi-objective building block 310: Constructs an optimization function containing multiple conflicting objectives, with the goal of maximizing environmental realism and comfort while minimizing energy consumption.

[0037] Environmental realism metrics: A spectral generation function is constructed based on a Gaussian function to simulate the target's outdoor spectrum. ; Among them, amplitude A i and center wavelength From the dynamic coding of the environment Determine the standard deviation σ i and Related. Spectral Authenticity spec By calculating the indoor generated spectrum S indoor (λ) and the target spectrum S model The differences were obtained as follows: ; Sound field realism is calculated by comparing the MFCC coefficients of the generated sound in the room with Et. MFCC The cosine similarity is obtained. The environmental realism index F1 is a weighted fusion of spectral realism and sound field realism.

[0038] Comfort indicators include thermal comfort and visual comfort.

[0039] Thermal comfort index: Based on the temporal and predicted feature values ​​in the environmental dynamic coding, the target temperature T is calculated through the thermal comfort objective function. target : ; Among them, T out，pred and They are and α and β are compensation coefficients. The thermal comfort index Comf is then calculated using the Predicted Average Voting (PMV) model. ther : ; Visual comfort: Set a dynamic weight w for the timestamp and create a time-varying visual comfort function to obtain the target color temperature (CCT). target : ; in, The preset day and night comfortable color temperature value, Weight ; The final comfort index F2 is a weighted fusion of thermal comfort index and visual comfort index.

[0040] Energy consumption index F3: Used to assess the power consumed by equipment during operation. ; The first item is air conditioning energy consumption, and the second item is LED lighting energy consumption.

[0041] Constraint processing unit 320: Introduces various constraints into the optimization process, including: Equipment physical constraints: such as the LED current needing to be between the minimum and maximum current; the air conditioning temperature setting value needing to be within the equipment's allowable range; and the color temperature setting value needing to fall within the physical color gamut of the LED chip.

[0042] User and scenario constraints: such as user-preset temperature range; dynamic preference boundaries. Dynamic illuminance constraint: Indoor illuminance not less than The η ratio.

[0043] Pareto Solver 330: Constructs a multi-objective optimization problem: Minimize [-F1(X), -F2(X), F3(X)]. Under the above constraints, a multi-objective optimization algorithm (such as NSGA-II) is used to solve the problem, generating a Pareto optimal solution set = {w1, w2, ..., w...} consisting of multiple non-dominated solutions. n}, where each solution w represents a set of control parameter vectors.

[0044] Joint control module 400, such as Figure 4 As shown, this module is used for spectral acoustic field co-optimization using the Pareto optimal solution set to obtain a photoacoustic joint control strategy. Specifically, this module includes: Light source driving vector generation unit 410: Spectral response modeling subunit: Experimentally measuring different combinations of driving currents I (k) The output spectrum S (k) , constitute the dataset The LED spectral response function φ(I) was obtained by fitting using a multiple regression method (such as Gaussian process regression).

[0045] Decision Subunit: Define an ideal policy point [F1] F2 F3 (These represent the theoretical maximum comfort, realism, and theoretical minimum energy consumption, respectively). Calculate the standardized Euclidean distance from each strategy w in w to the ideal point, and select the strategy with the smallest distance as the chosen strategy w. : .

[0046] Spectral reconstruction subunit: Target spectrum S based on selected strategy target (w Based on the LED spectral response function φ(I), a spectral reconstruction optimization model is constructed: ; Where R(I) is the regularization term and x is the regularization coefficient. Solving this model yields the optimal spectral driving vector I. .

[0047] 3D sound field reconstruction unit 420: Based on acoustic feature values ​​in dynamic environmental coding (such as...) Match the corresponding sound source amplitude A from the preset sound source library. q and phase parameter φ q On the discretized indoor spatial grid points (x, y, z), the complex sound pressure level at each point is calculated using a sound field reconstruction algorithm: ; Where, r q (x, y, z) represents the distance from the sound source q to the spatial point, k is the wave number, and j is the imaginary unit. The complex sound pressure amplitudes |P(x, y, z)| at all grid points are combined into a three-dimensional matrix, thus forming the three-dimensional sound field distribution matrix P. 3d .

[0048] Joint optimization solver 430: Establishing a regularized multiple loss function that includes spectral matching error terms and acoustic field reconstruction error terms: R(U)]; Among them, U=[U (o) U (v) ] is the joint control vector, containing the lighting control component U. (o) Harmony sound field control component U (v) ;ψU (v) U is the sound field response function used to calculate the spatial sound field distribution corresponding to the loudspeaker array drive signal; u1 and u2 are weighting coefficients; R(U) is the joint regularization term. The solution is obtained by using the Alternating Direction Multiplier Method (ADMM), fixing the sound field control to optimize the lighting control, then fixing the lighting control to optimize the sound field control, iterating until convergence, to obtain the photoacoustic joint control strategy U that simultaneously satisfies spectral fidelity and accurate sound field reconstruction. joint .

[0049] The collaborative control module 500 is used for multi-modal control signal analysis and collaborative optimization based on a photoacoustic joint control strategy to obtain the final multi-device collaborative control commands. This module includes: Drive command generation unit: Illumination control component based on photoacoustic joint control strategy Construct the spectral mapping optimization function: ; Where I is the driving current vector, D is the modulation parameter vector (such as PWM duty cycle), and w(I, D) is the regularization term. The adaptive spectrum driving command (optimal driving current vector I' and modulation parameter vector D') is obtained by iteratively solving the problem using the projection gradient descent method.

[0050] Multimodal fusion unit: integrates the adaptive spectrum driving command and the sound field control component in the photoacoustic joint control strategy. And the temperature setpoint T in the decision rule selection strategy. set Humidity setpoint H set The process involves coordination and formatting via a lightweight multimodal fusion network. This network, based on an attention mechanism, resolves potential micro-conflicts (such as avoiding rapid cooling while playing high-decibel sounds), generating a preliminary set of multi-device control signals [D]. led D audio D ac D humid ] T .

[0051] Discrete Event Scheduling Unit: For each control command in the initial multi-device control signal set, based on the response characteristics of its controlled object and the preset transition effect, it assigns a precise timestamp and transition duration (e.g., a smooth change in lights within 5 seconds, an immediate sound switch). By checking for time conflicts between commands, it generates a conflict-free, ordered sequence of commands, which serves as the final multi-device collaborative control command D. final Output to various execution devices.

[0052] Example 2 Please refer to Figure 5 This embodiment provides a home AI weather lighting method, which can be applied to any of the home AI weather lighting systems described in the above embodiments. The method includes the following steps: Step S100: Collect and preprocess outdoor weather data to obtain a structured weather database.

[0053] Step S200: Perform hypergraph fusion decision based on the structured weather database and user behavior logs to obtain dynamic environmental coding.

[0054] Step S300: Based on the dynamic encoding of the environment, construct a multi-objective optimization function to solve for the optimal control strategy and generate a Pareto optimal solution set.

[0055] Step S400: Perform spectral acoustic field co-optimization using the Pareto optimal solution set to obtain a photoacoustic joint control strategy.

[0056] Step S500: Based on the aforementioned photoacoustic joint control strategy, perform multi-modal control signal analysis and collaborative optimization to obtain the final multi-device collaborative control command.

[0057] The specific implementation details of each step of the method have been described in detail in the system embodiments, and will not be repeated here.

[0058] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A home AI weather lighting system, characterized in that, include: The data acquisition and processing module is configured to collect and preprocess outdoor weather data to obtain a structured weather database. The fusion decision module is configured to perform hypergraph fusion decision-making based on the structured weather database and user behavior logs to obtain dynamic environmental codes; The optimized control module is configured to construct a multi-objective optimization function based on the dynamic encoding of the environment to solve for the optimal control strategy and generate a Pareto optimal solution set. The joint control module is configured to perform spectral acoustic field co-optimization through the Pareto optimal solution set to obtain a photoacoustic joint control strategy. The collaborative control module is configured to perform multi-modal control signal analysis and collaborative optimization based on the aforementioned photoacoustic joint control strategy to obtain the final multi-device collaborative control command.

2. The home AI weather lighting system according to claim 1, characterized in that, The acquisition and processing module includes: Multispectral sensors are used to collect outdoor spectral distribution and light intensity in real time. A weather sensor array is used to collect temperature, humidity, and fog levels. Microphone array, used to collect ambient sound waveform data; The acquisition and processing module is also used to perform time-series matching, calibration and denoising on the above-mentioned multi-source heterogeneous data to form a standardized weather data stream, and store it in the time-series database to form the structured weather database.

3. The home AI weather lighting system according to claim 1, characterized in that, The fusion decision module includes: The hypergraph construction unit is configured to construct a weather configuration hypergraph structure based on user-defined functional parameters and the structured weather database. The node set of the hypergraph structure includes environmental measurement nodes, user configuration nodes, and derived feature nodes. The hyperedge set is used to capture unpaired association relationships between nodes. The hypergraph attention processing unit is configured to calculate the importance weight of each node to the hyperedge and the importance weight of each node to the global decision through the hypergraph attention network, so as to generate the hypergraph fusion decision signal. The temporal coding unit is configured to temporally combine the hypergraph fusion decision signal with multidimensional low-level static features extracted from the structured weather database, and input them into a spatiotemporal convolutional memory network to extract dynamic environmental evolution patterns, thereby forming the environmental dynamic code.

4. The home AI weather lighting system according to claim 1, characterized in that, The optimization control module includes: The multi-objective building block is configured to construct a multi-objective optimization function that includes at least environmental realism indicators, comfort indicators, and energy consumption indicators. The constraint processing unit is configured to introduce device physical constraints as well as user and scenario constraints. The Pareto solver is configured to solve the multi-objective optimization function under the constraints and generate a Pareto optimal solution set consisting of multiple non-dominated solutions.

5. The home AI weather lighting system according to claim 4, characterized in that, The environmental realism index is a fusion of spectral realism and sound field realism, wherein, The spectral accuracy is obtained by constructing a spectral generation function based on a Gaussian function, and calculating the difference between the indoor generated spectrum and the target outdoor spectrum based on the spectral feature values ​​in the environmental dynamic coding to obtain the spectral accuracy. The sound field realism is obtained by calculating the similarity between the acoustic feature values ​​of the indoor generated sound and the outdoor acoustic feature values ​​in the dynamic environmental coding.

6. The home AI weather lighting system according to claim 4, characterized in that, The comfort index is a fusion of thermal comfort and visual comfort, wherein... The thermal comfort index is obtained in the following way: based on the temporal feature value and predicted feature value in the dynamic coding of the environment, the thermal comfort index is obtained through the thermal comfort objective function and the predicted average voting model; The visual comfort is obtained by setting a dynamic weight for the timestamp, creating a time-varying visual comfort function based on the dynamic weight of the timestamp, and combining the temporal feature value and spectral feature value in the dynamic coding of the environment to obtain the visual comfort.

7. The home AI weather lighting system according to claim 1, characterized in that, The joint control module includes: The light source driving vector generation unit is configured to transform the strategy selected in the Pareto optimal solution set into the optimal spectral driving vector based on the LED spectral response function; The three-dimensional sound field reconstruction unit is configured to match sound source parameters from a preset sound source library based on the acoustic feature values ​​in the dynamic environmental coding, and calculate complex sound pressure values ​​on discretized indoor space grid points using a sound field reconstruction algorithm to form a three-dimensional sound field distribution matrix. The joint optimization solution unit is configured to establish a regularized multiple loss function that includes spectral matching error terms and sound field reconstruction error terms, and solves it using the alternating direction multiplier method to obtain a photoacoustic joint control strategy that simultaneously includes illumination control components and sound field control components.

8. The home AI weather lighting system according to claim 7, characterized in that, The light source driving vector generation unit includes: The spectral response modeling subunit is configured to experimentally measure the output spectrum under different combinations of driving currents and use a multivariate regression method to fit the LED spectral response function. The decision subunit is configured to define an ideal policy point, calculate the standardized Euclidean distance from each policy in the Pareto optimal solution set to the ideal policy point, and select the policy with the smallest distance as the selected policy. The spectral reconstruction subunit is configured to obtain the optimal spectral driving vector by solving the spectral reconstruction optimization model based on the target spectrum of the selected strategy and the LED spectral response function.

9. The home AI weather lighting system according to claim 1, characterized in that, The collaborative control module includes: The driving instruction generation unit is configured to obtain adaptive spectral driving instructions by constructing and solving a spectral mapping optimization function based on the illumination control component in the photoacoustic joint control strategy. The multimodal fusion unit is configured to coordinate and format the adaptive spectral driving command, the sound field control component in the photoacoustic joint control strategy, and other environmental device control commands to generate a preliminary multi-device control signal set. The discrete event scheduling unit is configured to assign a timestamp and transition duration to each control instruction in the initial multi-device control signal set, forming a conflict-free instruction sequence, which is then output as the final multi-device collaborative control instruction.

10. A home AI weather lighting method, characterized in that, The system is applied to a home AI weather lighting system as described in any one of claims 1-9, comprising the following steps: Outdoor weather data is collected and preprocessed to obtain a structured weather database; Based on the structured weather database and user behavior logs, a hypergraph fusion decision is made to obtain a dynamic environmental code; Based on the dynamic coding of the environment, a multi-objective optimization function is constructed to solve the optimal control strategy and generate a Pareto optimal solution set. The spectral acoustic field is co-optimized using the Pareto optimal solution set to obtain a photoacoustic joint control strategy; Based on the aforementioned photoacoustic joint control strategy, multimodal control signal analysis and collaborative optimization are performed to obtain the final multi-device collaborative control command.

Images