Household linkage system based on multi-dimensional fusion of personal physiology, environment and astronomy

Abstract

The invention discloses a household linkage system based on multi-dimensional fusion of personal physiology, environment and astronomy. A method for acquiring time series of physiological parameters (heartbeat frequency values, blood pressure values, blood oxygen saturation and blood glucose level values) of a user, time series of environmental parameters (temperature values, humidity values and illumination intensity values) and astronomical data (solar electromagnetic activity and lunar attraction) by using an AI-based data processing and analysis algorithm time sequence implicit correlation feature analysis and semantic correlation degree calculation are performed on the time sequence, so that whether the indoor illumination intensity is reduced or not is automatically judged according to multi-modal time sequence joint coding features among multiple parameters in the data. Thus, the system can collect and comprehensively analyze physiological data, indoor environment parameters and astronomical information of the user in real time, precise regulation and control and intelligent adjustment of the intelligent home equipment are achieved, a more precise and personalized home environment adjustment scheme is provided for the user, and the overall living experience is improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent control, and more specifically, to a home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy. Background Art

[0002] With the rapid development of science and technology, smart wearable devices and smart home systems have been deeply integrated into people's daily lives, providing people with unprecedented convenience and comfort. With its advanced sensing technology, smart wearable devices can monitor and record users' physiological parameters in real time, including but not limited to key indicators such as heart rate, blood pressure, blood oxygen, and blood sugar, providing users with personalized health management solutions. At the same time, smart home systems use automated control to achieve intelligent management of indoor environments such as temperature, humidity, lighting, and entertainment facilities, thereby creating a more comfortable and livable living environment.

[0003] However, although smart wearable devices and smart homes have made significant progress in their respective fields, the linkage and integration between the two is still in its infancy. Most of the products on the market currently focus on single-dimensional data monitoring and adjustment, such as intelligent adjustment based solely on indoor environmental parameters or the user's physiological state. This lacks comprehensive consideration of the user's living conditions, environmental changes, and time factors, resulting in the inability of smart home systems to accurately respond to user needs and to achieve truly personalized and intelligent services.

[0004] Therefore, an optimized smart home linkage system is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a method of performing time series implicit correlation feature analysis and semantic correlation calculation on the time series of user physiological parameters (heart rate value, blood pressure value, blood oxygen saturation and blood sugar level value), environmental parameters (temperature value, humidity value and light intensity value) and astronomical data (solar electromagnetic activity and lunar gravity) collected by using AI-based data processing and analysis algorithms, so as to automatically determine whether to reduce the indoor light intensity based on the multimodal time series joint coding features between multiple parameters in these data. In this way, the system can collect and comprehensively analyze the user's physiological data, indoor environmental parameters and astronomical information in real time, realize the precise control and intelligent adjustment of smart home devices, so as to provide users with more accurate and personalized home environment adjustment solutions and improve the overall living experience. .

[0006] According to one aspect of the present application, a home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy is provided, which includes:

[0007] A personal physiological data acquisition module, used to obtain a time series of the user's physiological parameters collected by the smart wearable device, wherein the physiological parameters include heart rate value, blood pressure value, blood oxygen saturation and blood sugar level value;

[0008] An environmental data acquisition module, used to acquire a time series of environmental parameters collected by an indoor sensor network, wherein the environmental parameters include temperature values, humidity values, and light intensity values;

[0009] An astronomical data acquisition module, used to acquire a time series of astronomical data collected by an astronomical observatory, wherein the astronomical data includes solar electromagnetic activity and lunar gravity;

[0010] The linkage control module is used to analyze the time series of the user's physiological parameters, the time series of the environmental parameters and the time series of the astronomical data to generate control instructions, and send the control instructions to the smart home device through wireless communication.

[0011] According to another aspect of the present application, a home linkage method based on multi-dimensional integration of personal physiology, environment and astronomy is provided, which includes:

[0012] Acquire a time series of physiological parameters of a user collected by a smart wearable device, wherein the physiological parameters include heart rate value, blood pressure value, blood oxygen saturation and blood sugar level value;

[0013] Acquire a time series of environmental parameters collected by an indoor sensor network, wherein the environmental parameters include temperature, humidity and light intensity;

[0014] Acquire a time series of astronomical data collected by an astronomical observatory, wherein the astronomical data includes solar electromagnetic activity and lunar gravity;

[0015] The time series of the user's physiological parameters, the time series of the environmental parameters and the time series of the astronomical data are analyzed to generate control instructions, and the control instructions are sent to the smart home device through wireless communication.

[0016] Compared with the prior art, the present application provides a home linkage system based on the multi-dimensional integration of personal physiology, environment and astronomy. It uses AI-based data processing and analysis algorithms to analyze the time series of user physiological parameters (heart rate values, blood pressure values, blood oxygen saturation and blood sugar level values), environmental parameters (temperature values, humidity values ​​and light intensity values) and astronomical data (solar electromagnetic activity and lunar gravity) collected, and perform time series implicit correlation feature analysis and semantic correlation calculation. In this way, it automatically determines whether to reduce the indoor light intensity based on the multimodal time series joint coding features between multiple parameters in these data. In this way, the system can collect and comprehensively analyze the user's physiological data, indoor environmental parameters and astronomical information in real time, realize the precise control and intelligent adjustment of smart home devices, so as to provide users with more accurate and personalized home environment adjustment solutions and improve the overall living experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 is a block diagram of a home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to an embodiment of the present application;

[0019] Figure 2 It is a data flow diagram of a home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to an embodiment of the present application;

[0020] Figure 3 It is a block diagram of the training phase of the home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to an embodiment of the present application;

[0021] Figure 4 is a block diagram of a linkage control module in a home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to an embodiment of the present application;

[0022] Figure 5 It is a flow chart of a home linkage method based on multi-dimensional integration of personal physiology, environment and astronomy according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0024] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0025] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0026] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0027] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0028] In the technical solution of the present application, a home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy is proposed. Figure 1 It is a block diagram of a home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to an embodiment of the present application. Figure 2 1 is a data flow diagram of a home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to an embodiment of the present application, a home linkage system 300 based on multi-dimensional fusion of personal physiology, environment and astronomy includes: a personal physiological data acquisition module 310, used to obtain a time series of user physiological parameters collected by a smart wearable device, wherein the physiological parameters include heart rate values, blood pressure values, blood oxygen saturation and blood sugar level values; an environmental data acquisition module 320, used to obtain a time series of environmental parameters collected by an indoor sensor network, wherein the environmental parameters include temperature values, humidity values ​​and light intensity values; an astronomical data acquisition module 330, used to obtain a time series of astronomical data collected by an astronomical observatory, wherein the astronomical data includes solar electromagnetic activity and lunar gravity; a linkage control module 340, used to analyze the time series of the user's physiological parameters, the time series of the environmental parameters and the time series of the astronomical data to generate a control instruction, and send the control instruction to the smart home device via wireless communication.

[0029] In particular, the personal physiological data acquisition module 310, the environmental data acquisition module 320, and the astronomical data acquisition module 330 are used to obtain the time series of the user's physiological parameters collected by the smart wearable device, wherein the physiological parameters include heart rate values, blood pressure values, blood oxygen saturation and blood sugar level values; obtain the time series of environmental parameters collected by the indoor sensor network, wherein the environmental parameters include temperature values, humidity values ​​and light intensity values; and obtain the time series of astronomical data collected by the astronomical observatory, wherein the astronomical data includes solar electromagnetic activity and lunar gravity. That is, in the technical solution of the present application, the precise control and intelligent adjustment of smart home devices are achieved by comprehensively analyzing the user's physiological data, indoor environmental parameters and astronomical information.

[0030] In particular, the linkage control module 340 is used to analyze the time series of the user's physiological parameters, the time series of the environmental parameters and the time series of the astronomical data to generate control instructions, and send the control instructions to the smart home device via wireless communication. In particular, in a specific example of the present application, Figure 4As shown, the linkage control module 340 includes: a parameter data reorganization unit 341, which is used to reorganize the time series of the user's physiological parameters, the time series of the environmental parameters, and the time series of the astronomical data according to the time dimension to obtain a time series of heart rate values, a time series of blood pressure values, a time series of blood oxygen saturation, a time series of blood sugar level values, a time series of temperature values, a time series of humidity values, a time series of light intensity values, a time series of solar electromagnetic activity, and a time series of lunar gravity; a parameter data time series encoding unit 342, which is used to reorganize the time series of the heart rate values, the time series of the blood pressure values, the time series of the blood oxygen saturation, and the time series of the lunar gravity. The time series of the blood sugar level value, the time series of the temperature value, the time series of the humidity value, the time series of the light intensity value, the time series of the solar electromagnetic activity and the time series of the lunar gravity are respectively input into the sequence encoder to obtain the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector, and the lunar gravity time series associated implicit feature vector; the parameter data time series feature vector The feature association calculation unit 343 is used to calculate the semantic association between any two feature vectors among the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector and the lunar gravity time series associated implicit feature vector to obtain a semantic association adjacency matrix; the parameter data time series feature multimodal combination unit 344 is used to combine the heart rate time series associated implicit feature vector into a semantic association adjacency matrix; The characteristic vector, the blood pressure time series associated implicit characteristic vector, the blood oxygen saturation time series associated implicit characteristic vector, the blood sugar level time series associated implicit characteristic vector, the temperature time series associated implicit characteristic vector, the humidity time series associated implicit characteristic vector, the light intensity time series associated implicit characteristic vector, the solar electromagnetic activity time series associated implicit characteristic vector and the lunar gravity time series associated implicit characteristic vector are subjected to multimodal joint processing to obtain multimodal parameter time series joint graph coding features; an indoor light intensity control unit 345 is used to obtain a control instruction based on the multimodal parameter time series joint graph coding features, and the control instruction is used to indicate whether to reduce the indoor light intensity.

[0031] Specifically, the parameter data reorganization unit 341 is used to reorganize the time series of the user's physiological parameters, the time series of the environmental parameters, and the time series of the astronomical data according to the time dimension to obtain the time series of heart rate values, the time series of blood pressure values, the time series of blood oxygen saturation, the time series of blood sugar level values, the time series of temperature values, the time series of humidity values, the time series of light intensity values, the time series of solar electromagnetic activity, and the time series of lunar gravity. Considering that the time series of the user's physiological parameters, the time series of the environmental parameters, and the time series of the astronomical data are all parameters with a time dimension, and the parameter data of each dimension has a pattern and trend change feature in the time dimension, and there is also a correlation between the parameters in the time dimension. Therefore, in the technical solution of the present application, the time series of the user's physiological parameters, the time series of the environmental parameters, and the time series of the astronomical data are reorganized according to the time dimension to arrange the parameters in each dimension in time series to obtain a time series of heart rate values, a time series of blood pressure values, a time series of blood oxygen saturation, a time series of blood sugar level values, a time series of temperature values, a time series of humidity values, a time series of light intensity values, a time series of solar electromagnetic activity, and a time series of lunar gravity.

[0032] Specifically, the parameter data time series encoding unit 342 is used to input the time series of the heart rate value, the time series of the blood pressure value, the time series of the blood oxygen saturation, the time series of the blood sugar level value, the time series of the temperature value, the time series of the humidity value, the time series of the light intensity value, the time series of the solar electromagnetic activity and the time series of the lunar gravity into a sequence encoder to obtain the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector, and the lunar gravity time series associated implicit feature vector. Considering that each time series has different time series implicit association relationships between different time distances. Therefore, in the technical solution of the present application, the time series of the heart rate value, the time series of the blood pressure value, the time series of the blood oxygen saturation, the time series of the blood sugar level value, the time series of the temperature value, the time series of the humidity value, the time series of the light intensity value, the time series of the solar electromagnetic activity and the time series of the lunar gravity are respectively input into the sequence encoder based on the Bi-LSTM model to capture and extract the association between different local times respectively, thereby obtaining the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector, and the lunar gravity time series associated implicit feature vector. It is worth mentioning that Bi-LSTM is a recurrent neural network (RNN) that processes sequence data by simultaneously considering past and future contextual information. It consists of two LSTM layers, one processing the forward direction of the sequence and the other processing the reverse direction of the sequence.

[0033] Specifically, the parameter data time series feature correlation calculation unit 343 is used to calculate the semantic correlation between any two feature vectors of the heart rate time series correlation implicit feature vector, the blood pressure time series correlation implicit feature vector, the blood oxygen saturation time series correlation implicit feature vector, the blood sugar level time series correlation implicit feature vector, the temperature time series correlation implicit feature vector, the humidity time series correlation implicit feature vector, the light intensity time series correlation implicit feature vector, the solar electromagnetic activity time series correlation implicit feature vector and the lunar gravity time series correlation implicit feature vector to obtain a semantic correlation adjacency matrix. Considering that there is a temporal semantic association relationship between any two feature vectors among the heart rate time-series associated implicit feature vector, the blood pressure time-series associated implicit feature vector, the blood oxygen saturation time-series associated implicit feature vector, the blood sugar level time-series associated implicit feature vector, the temperature time-series associated implicit feature vector, the humidity time-series associated implicit feature vector, the light intensity time-series associated implicit feature vector, the solar electromagnetic activity time-series associated implicit feature vector and the lunar gravity time-series associated implicit feature vector, for example, under certain environmental conditions, the user's heart rate may increase, or the user's blood pressure may change during specific astronomical activities. Therefore, in the technical solution of the present application, the semantic correlation between any two feature vectors of the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector and the lunar gravity time series associated implicit feature vector is further calculated to obtain a semantic correlation adjacency matrix. That is, by calculating the semantic correlation between different feature vectors, it is possible to identify which physiological parameters, environmental parameters and astronomical data have strong semantic correlations, which helps to understand how different factors affect each other, thereby providing accurate decision support for personalized home control.

[0034] In an embodiment of the present application, the semantic association between any two feature vectors of the heart rate timing-associated implicit feature vector, the blood pressure timing-associated implicit feature vector, the blood oxygen saturation timing-associated implicit feature vector, the blood sugar level timing-associated implicit feature vector, the temperature timing-associated implicit feature vector, the humidity timing-associated implicit feature vector, the light intensity timing-associated implicit feature vector, the solar electromagnetic activity timing-associated implicit feature vector and the lunar gravity timing-associated implicit feature vector is calculated to obtain a semantic association adjacency matrix, including: transposing the transposed vector of the heart rate timing-associated implicit feature vector with the blood pressure timing-associated implicit feature vector; The implicit feature vectors are vector multiplied to obtain a heart rate-blood pressure semantic association value; the norms of the heart rate timing-associated implicit feature vector and the blood pressure timing-associated implicit feature vector are calculated respectively to obtain a heart rate timing-associated semantic norm value and a blood pressure timing-associated semantic norm value; the heart rate timing-associated semantic norm value and the blood pressure timing-associated semantic norm value are multiplied to obtain a heart rate-blood pressure semantic interaction value; the heart rate-blood pressure semantic association value is divided by the heart rate-blood pressure semantic interaction value to obtain the semantic association degree between the heart rate timing-associated implicit feature vector and the blood pressure timing-associated implicit feature vector.

[0035] In summary, in the above embodiment, the semantic association between any two feature vectors of the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector and the lunar gravity time series associated implicit feature vector is calculated to obtain a semantic association adjacency matrix, including: The semantic association degree between any two feature vectors among the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector and the lunar gravity time series associated implicit feature vector is calculated by a related calculation formula to obtain a semantic association degree adjacency matrix; wherein the semantic association calculation formula is:

[0036]

[0037] Among them, v i and u jare any two feature vectors of the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector and the lunar gravity time series associated implicit feature vector, ||·|| is the one-norm of the feature vector, (·) T is the transposed vector of the eigenvector, M ij is the eigenvalue of the (i, j)th position in the semantic association adjacency matrix.

[0038] Specifically, the parameter data timing feature multimodal joint unit 344 is used to perform multimodal joint processing on the heart rate timing-related implicit feature vector, the blood pressure timing-related implicit feature vector, the blood oxygen saturation timing-related implicit feature vector, the blood sugar level timing-related implicit feature vector, the temperature timing-related implicit feature vector, the humidity timing-related implicit feature vector, the light intensity timing-related implicit feature vector, the solar electromagnetic activity timing-related implicit feature vector and the lunar gravity timing-related implicit feature vector to obtain the multimodal parameter timing-related graph encoding feature. In the technical solution of the present application, in order to further process and learn the complex relationship and interaction between each parameter data, so as to better understand the overall state and time series changes, in the technical solution of the present application, the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector and the lunar gravity time series associated implicit feature vector are used as a set of nodes and the semantic association degree adjacency matrix is ​​used as a set of edges between nodes, and the set of nodes and the set of edges are input into the graph neural network model to obtain a multimodal parameter time series joint graph encoding matrix. Specifically, the time series implicit features of each time series associated implicit feature vector parameter node and the semantic association degree adjacency matrix are regarded as feature associations between nodes to form a graph structure. Integrating these two features into graph neural networks can help obtain a more comprehensive and integrated feature representation to better capture the temporal correlation interaction information between nodes, which can improve the ability to understand the overall control parameters and thus improve the accuracy of indoor light intensity.

[0039] Specifically, the indoor light intensity control unit 345 is used to obtain a control instruction based on the multimodal parameter time series joint graph coding feature, and the control instruction is used to indicate whether to reduce the indoor light intensity. In a specific example of the present application, the multimodal parameter time series joint graph coding matrix is ​​input into a control instruction generator based on a classifier to obtain the control instruction, and the control instruction is used to indicate whether to reduce the indoor light intensity. That is, the multimodal parameter time series joint graph coding features obtained by graph joint coding using the respective time series associated implicit feature vectors and the semantic association degree adjacency matrix are classified and processed to automatically determine whether to reduce the indoor light intensity. In this way, the system can collect and comprehensively analyze the user's physiological data, indoor environmental parameters and astronomical information in real time, realize accurate control and intelligent adjustment of smart home devices, so as to provide users with more accurate and personalized home environment adjustment solutions and improve the overall living experience.

[0040] Considering that the heart rate time series association feature vector, the blood pressure time series association implicit feature vector, the blood oxygen saturation time series association implicit feature vector, the blood sugar level time series association implicit feature vector, the temperature time series association implicit feature vector, the humidity time series association implicit feature vector, the light intensity time series association implicit feature vector, the solar electromagnetic activity time series association implicit feature vector and the lunar gravity time series association implicit feature vector respectively represent the short-range and long-range bidirectional time series association characteristics of heart rate value, blood pressure value, blood oxygen saturation, blood sugar level value, temperature value, humidity value, light intensity value, solar electromagnetic activity and lunar gravity, therefore, when they are used as graph nodes and input into the graph neural network model based on the semantic association adjacency topology for topological association, the multimodal parameter time series joint graph encoding matrix will also have difficulties in multimodal graph fusion classification and regression recognition due to the relative independence of node topological association features, thereby affecting the accuracy of the classification results.

[0041] Therefore, when the multimodal parameter timing joint diagram encoding matrix is ​​input into a classifier-based control instruction generator to obtain a control instruction, the present application optimizes the multimodal parameter timing joint diagram encoding matrix, including the steps of:

[0042] Expanding the multimodal parameter temporal joint graph coding matrix into a multimodal parameter temporal joint graph coding vector;

[0043] Subtract the zero norm of the multimodal parameter temporal joint graph coding matrix from the length of the multimodal parameter temporal joint graph coding vector to obtain a multimodal parameter temporal joint graph coding isolated representation value;

[0044] Calculate the logarithm value with base 2 of the sum of the square of the multimodal parameter temporal joint graph coding isolated representation value and the multimodal parameter temporal joint graph coding isolated representation value to obtain the multimodal parameter temporal joint graph coding information order value;

[0045] Calculate a power function with each eigenvalue of the multimodal parameter time series joint graph coding vector as the base, the difference value of the multimodal parameter time series joint graph coding isolated representation value minus one as the exponent, and perform point multiplication with the multimodal parameter time series joint graph coding information order value to obtain a multimodal parameter time series joint graph coding leading vector;

[0046] After performing a dot multiplication on the multimodal parameter temporal joint graph coding vector and the multimodal parameter temporal joint graph coding isolated representation value minus one, the dot multiplication is performed on the reciprocal of the multimodal parameter temporal joint graph coding isolated representation value to obtain a multimodal parameter temporal joint graph coding field constraint vector;

[0047] Calculate an exponential function with a natural constant as the base and each eigenvalue of the multimodal parameter time series joint graph coding field constraint vector as the exponent to obtain a multimodal parameter time series joint graph coding field bias vector;

[0048] The multimodal parameter timing joint graph coding leader vector and the multimodal parameter timing joint graph coding field bias vector are dot-added to obtain an optimized multimodal parameter timing joint graph coding vector. Finally, the optimized multimodal parameter timing joint graph coding vector is input into a control instruction generator based on a classifier to obtain a control instruction.

[0049] The optimization process of the multimodal parameter temporal joint graph coding vector obtained by expanding the multimodal parameter temporal joint graph coding matrix, for example, denoted as V, is expressed as:

[0050] V'=V ⊙(n-1) ⊙log2(n 2 +n)⊕exp[V⊙(n-1)⊙n -1 ]

[0051] n=L-||V||0

[0052] Wherein, V represents the multimodal parameter temporal joint graph coding vector, ||·||0 represents the zero norm of the vector, L represents the length of the multimodal parameter temporal joint graph coding vector, n represents the multimodal parameter temporal joint graph coding isolated representation value, ⊙ is the position point multiplication, ⊕ is the position addition, (·) ⊙(n-1) A power function with the difference value of the multimodal parameter temporal joint graph coding isolated representation value minus one as the exponent for calculating the feature value of each position in the feature vector is represented, and V' represents the optimized multimodal parameter temporal joint graph coding vector.

[0053] Therefore, the high-dimensional feature manifold of the multimodal parameter time series joint graph coding vector obtained by expanding the multimodal parameter time series joint graph coding matrix is ​​represented by the eigenvalues ​​of the feature set as the vector field of the aggregation dimension, and the superposition value of the vector field of the multimodal parameter time series joint graph coding vector at the isolated zero position is used as the order information to fix the local position of the eigenvalue of its feature set, and a bias for the reversibility of the feature regression distribution field of the multimodal parameter time series joint graph coding vector is added as a reward to achieve the mapping target tracking of the regression distribution of the multimodal parameter time series joint graph coding vector for the eigenvalue position, so that the feature set of the multimodal parameter time series joint graph coding vector perceives the mapping migration to the aggregate distribution, thereby improving the accuracy of the control instructions obtained by the classifier-based control instruction generator input to the multimodal parameter time series joint graph coding matrix by improving the multimodal semantic fusion regression comprehensibility of the multimodal parameter time series joint graph coding vector.

[0054] It should be understood that before using the above-mentioned neural network model for inference, it is necessary to train the sequence encoder based on the Bi-LSTM model, the graph neural network model and the control instruction generator based on the classifier. That is to say, according to the home linkage system 300 based on multi-dimensional fusion of personal physiology, environment and astronomy of the present application, it also includes a training phase for training the sequence encoder based on the Bi-LSTM model, the graph neural network model and the control instruction generator based on the classifier.

[0055] Figure 3 FIG. 1 is a block diagram of the training phase of the home linkage system based on the multi-dimensional integration of personal physiology, environment and astronomy according to an embodiment of the present application. Figure 3As shown, according to the embodiment of the present application, the home linkage system 300 based on the multi-dimensional integration of personal physiology, environment and astronomy includes: a training module 400, including: a training data acquisition unit 410, used to acquire training data, the training data including a time series of physiological parameters of the training user collected by the smart wearable device, wherein the physiological parameters include heart rate value, blood pressure value, blood oxygen saturation and blood sugar level value; acquiring a time series of training environment parameters collected by an indoor sensor network, wherein the environment parameters include temperature value, humidity value and light intensity value; and acquiring a time series of training astronomical data collected by an astronomical observatory, wherein the astronomical data includes solar electromagnetic activity and lunar gravity; training The data reorganization unit 420 is used to reorganize the time series of the training user's physiological parameters, the time series of the training environment parameters, and the time series of the training astronomical data according to the time dimension to obtain the time series of the training heart rate value, the time series of the training blood pressure value, the time series of the training blood oxygen saturation, the time series of the training blood sugar level value, the time series of the training temperature value, the time series of the training humidity value, the time series of the training light intensity value, the time series of the training solar electromagnetic activity, and the time series of the training lunar gravity; the training data time series encoding unit 430 is used to encode the time series of the training heart rate value, the time series of the training blood pressure value, the time series of the training blood oxygen saturation, and the time series of the training The time series of the training blood sugar level values, the time series of the training temperature values, the time series of the training humidity values, the time series of the training light intensity values, the time series of the training solar electromagnetic activity and the time series of the training lunar gravity are respectively input into a sequence encoder based on a Bi-LSTM model to obtain a training heart rate time series associated implicit feature vector, a training blood pressure time series associated implicit feature vector, a training blood oxygen saturation time series associated implicit feature vector, a training blood sugar level time series associated implicit feature vector, a training temperature time series associated implicit feature vector, a training humidity time series associated implicit feature vector, a training light intensity time series associated implicit feature vector, a training solar electromagnetic activity time series associated implicit feature vector a training data time series feature correlation degree calculation unit 440, used to calculate the semantic correlation degree between any two feature vectors of the training heart rate time series correlation implicit feature vector, the training blood pressure time series correlation implicit feature vector, the training blood oxygen saturation time series correlation implicit feature vector, the training blood sugar level time series correlation implicit feature vector, the training temperature time series correlation implicit feature vector, the training humidity time series correlation implicit feature vector, the training light intensity time series correlation implicit feature vector, the training solar electromagnetic activity time series correlation implicit feature vector and the training lunar gravity time series correlation implicit feature vector to obtain a training semantic correlation degree adjacency matrix;The training data time series feature multimodal association unit 450 is used to use the training heart rate time series associated implicit feature vector, the training blood pressure time series associated implicit feature vector, the training blood oxygen saturation time series associated implicit feature vector, the training blood sugar level time series associated implicit feature vector, the training temperature time series associated implicit feature vector, the training humidity time series associated implicit feature vector, the training light intensity time series associated implicit feature vector, the training solar electromagnetic activity time series associated implicit feature vector and the training lunar gravity time series associated implicit feature vector as a set of nodes and the training semantic association adjacency matrix as a set of edges between nodes, and input the set of nodes and the set of edges into a graph neural network model to obtain training multimodal parameters. a time-series joint graph encoding matrix for training multimodal parameters; a classification loss unit 460, for inputting the training multimodal parameter time-series joint graph encoding matrix into a control instruction generator based on a classifier to obtain a classification loss function value; a multimodal parameter time-series joint graph encoding loss calculation unit 470, for calculating a multimodal parameter time-series joint graph encoding loss function value based on the multimodal parameter time-series joint graph encoding loss term; a weighting unit 480, for calculating a weighted sum of the classification loss function value and the multimodal parameter time-series joint graph encoding loss function value to obtain a final loss function value; a training unit 490, for training the sequence encoder based on the Bi-LSTM model, the graph neural network model, and the control instruction generator based on the classifier based on the final loss function value. ;

[0056] In particular, in a preferred embodiment of the present application, in training the sequence encoder based on the Bi-LSTM model, the graph neural network model and the classifier-based control instruction generator based on the classification loss function value and through back propagation of gradient descent, a multimodal parameter temporal joint graph encoding loss function value other than the classification loss function value is further introduced, that is, training the model through gradient back propagation based on the weighted sum of the classification loss function value and the multimodal parameter temporal joint graph encoding loss function value as the loss function value includes the following steps:

[0057] Expanding the training multimodal parameter temporal joint graph encoding matrix into a training multimodal parameter temporal joint graph encoding vector;

[0058] Calculate a training multimodal parameter temporal joint graph code sum matrix and a training multimodal parameter temporal joint graph code difference matrix based on the training multimodal parameter temporal joint graph code vector, wherein the eigenvalue of the (i, j)th position of the training multimodal parameter temporal joint graph code sum matrix is ​​the average of the i-th eigenvalue and the j-th eigenvalue of the training multimodal parameter temporal joint graph code vector, and the eigenvalue of the (i, j)th position of the training multimodal parameter temporal joint graph code difference matrix is ​​half of the absolute value of the difference between the i-th eigenvalue and the j-th eigenvalue of the training multimodal parameter temporal joint graph code vector;

[0059] Performing query matrix multiplication of the training multimodal parameter temporal joint graph coding vector with the multimodal parameter temporal joint graph coding sum matrix and the multimodal parameter temporal joint graph coding difference matrix respectively to obtain a multimodal parameter temporal joint graph coding query sum vector and a multimodal parameter temporal joint graph coding query difference vector;

[0060] Calculating the vector inner product of the multimodal parameter temporal joint graph coding query and the vector and the multimodal parameter temporal joint graph coding query difference vector to obtain a first multimodal parameter temporal joint graph coding loss term;

[0061] Performing matrix multiplication on the multimodal parameter temporal joint graph coding sum matrix and the multimodal parameter temporal joint graph coding difference matrix, and calculating the Frobenius norm of the product matrix to obtain a second multimodal parameter temporal joint graph coding loss term; and

[0062] Subtract the product of a predetermined weight hyperparameter and the second multimodal parameter temporal joint graph coding loss term from the first multimodal parameter temporal joint graph coding loss term to obtain the multimodal parameter temporal joint graph coding loss function value;

[0063] The Bi-LSTM model-based sequence encoder, the graph neural network model and the classifier-based control instruction generator are trained based on the weighted sum of the classification loss function value and the multimodal parameter temporal joint graph encoding loss function value as the loss function value and through back propagation of gradient descent.

[0064] Among them, the multimodal parameter temporal joint graph encoding loss function value is specifically expressed as:

[0065]

[0066] Among them, M μ (i,j)=v i +v j / 2, and M σ (i,j)=|v i -v j| / 2, V is the training multimodal parameter temporal joint graph encoding vector, v i and v j are the eigenvalues ​​of the i-th and j-th positions in the training multimodal parameter temporal joint graph encoding vector, M μ (i,j) and M σ (i, j) are the eigenvalues ​​of the (i, j)th position in the multimodal parameter temporal joint graph coding sum matrix and the multimodal parameter temporal joint graph coding difference matrix, respectively. μ and M σ are the multimodal parameter temporal joint graph encoding sum matrix and the multimodal parameter temporal joint graph encoding difference matrix respectively, α is the predetermined weight hyperparameter, ||·|| F is the norm of the matrix being calculated, is matrix multiplication, is the multimodal parameter temporal joint graph encoding loss function value.

[0067] Here, considering that the training heart rate time series association implicit feature vector, the training blood pressure time series association implicit feature vector, the training blood oxygen saturation time series association implicit feature vector, the training blood sugar level time series association implicit feature vector, the training temperature time series association implicit feature vector, the training humidity time series association implicit feature vector, the training light intensity time series association implicit feature vector, the training solar electromagnetic activity time series association implicit feature vector and the training lunar gravity time series association implicit feature vector respectively represent the short-range and long-range bidirectional time series association characteristics of heart rate value, blood pressure value, blood oxygen saturation, blood sugar level value, temperature value, humidity value, light intensity value, solar electromagnetic activity and lunar gravity, therefore, when they are used as graph nodes and input into the graph neural network model based on the semantic association adjacency topology for topological association, the training multimodal parameter time series joint graph encoding matrix will also have classification regression recognition difficulties due to the relative independence of the node topological association features, which is attributed to the complex node topological association feature representation, thereby affecting the accuracy of the classification results.

[0068] Therefore, the present application uses the structured feature representation of the short-range-long-range cross-scale detail links of the training multimodal parameter time series joint graph coding matrix to perform the query composition of the detail inner product space in the training multimodal parameter time series joint graph coding matrix, so as to approximate the low-rank independent observable composition of the link detail composition provided by the structured detail interaction of the training multimodal parameter time series joint graph coding matrix. In this way, by training with the multimodal parameter time series joint graph coding loss function, the distributed detail group of the training multimodal parameter time series joint graph coding matrix can be used to decompose the detail group on the basis of detail complexity, so as to promote the classification regression decomposition recognition of the training multimodal parameter time series joint graph coding matrix based on complex feature representation, and improve the accuracy of the classification results. In this way, the system can collect and comprehensively analyze the user's physiological data, indoor environmental parameters and astronomical information in real time, realize the precise control and intelligent adjustment of smart home devices, so as to provide users with more accurate and personalized home environment adjustment solutions and improve the overall living experience.

[0069] In another embodiment of the present application, there is also provided a home linkage system based on the multi-dimensional integration of personal physiology, environment and astronomy, which includes: a personal physiological data acquisition module for collecting user physiological parameters and transmitting data wirelessly. An environmental data acquisition module for collecting indoor environmental parameters and transmitting data wirelessly. An astronomical data acquisition module for collecting astronomical data and transmitting data via the Internet. A linkage control module for comprehensively analyzing all collected data, generating control instructions, and adjusting smart home devices. More specifically, in another embodiment, the personal physiological data acquisition module is responsible for obtaining the user's physiological parameters from the smart wearable device, including heart rate, blood pressure, blood oxygen saturation and blood sugar level. These data are transmitted to the system via Bluetooth or Wi-Fi. The environmental data acquisition module obtains environmental parameters such as temperature, humidity and light intensity from the indoor sensor network. The data is transmitted to the system via a wireless network. The astronomical data acquisition module obtains solar electromagnetic activity and lunar gravitational data from the astronomical observatory and transmits them to the system via the Internet. The linkage control module analyzes the collected physiological, environmental and astronomical data, generates control instructions through AI algorithms, and adjusts the operation of smart home devices, such as adjusting light intensity. The home linkage system based on the multi-dimensional integration of personal physiology, environment and astronomy includes the following smart home devices in specific application scenarios: Smart wearable devices: users wear smart watches to monitor physiological parameters in real time and transmit them to the system via Bluetooth. Indoor environmental sensors: sensor networks are arranged in the home to collect environmental parameters and transmit data via Wi-Fi. Astronomical data acquisition: the system regularly obtains astronomical observatory data from the Internet. Smart home devices: such as smart lighting systems, which automatically adjust the light intensity by receiving control instructions to improve user comfort.

[0070] In another embodiment, the AI ​​algorithm includes: data preprocessing: reorganizing physiological, environmental and astronomical data into time series. Time series implicit correlation feature analysis: using the Bi-LSTM model to encode each time series data and extract implicit feature vectors. Semantic association calculation: calculating the semantic association between each feature vector and generating a semantic association adjacency matrix. Modal joint processing: inputting the feature vector and the adjacency matrix into the graph neural network model to generate multi-modal parameter time series joint graph encoding features. Control instruction generation: based on the joint graph encoding features, a control instruction on whether to adjust the light intensity is generated by a classifier.

[0071] Compared with the prior art, another embodiment of the present application has the following innovations: 1. Multi-dimensional data fusion: Fusion analysis of personal physiological, environmental and astronomical data to improve the accuracy of smart home control. 2. Real-time adjustment: Real-time data processing and analysis based on AI algorithms to achieve real-time control of smart home devices. 3. Personalized solutions: Provide personalized home environment adjustment solutions to improve users' living experience. Compared with the prior art that only focuses on single-dimensional data, this system integrates multi-dimensional data and has a higher level of intelligence and accuracy.

[0072] Further, for another embodiment described above, a verification field 1 is also provided, and the verification field 1 is a home environment, and its goal is to verify the application effect of the system in an actual home environment. The following steps are provided for the verification field 1, and the steps include: 1. Installation and configuration: Install smart wearable devices, indoor environmental sensor networks and smart home devices in the home. Configure data transmission channels to ensure that physiological parameters are transmitted via Bluetooth, environmental parameters are transmitted via Wi-Fi, and astronomical data are obtained via the Internet. 2. Data collection: Let users wear smart watches to monitor heart rate, blood pressure, blood oxygen saturation and blood sugar levels in real time. Environmental sensors collect temperature, humidity and light intensity data. The system regularly obtains solar electromagnetic activity and lunar gravity data from astronomical observatories. 3. Data processing and analysis: The collected multidimensional data is reorganized in time series and analyzed for associated features. Use the Bi-LSTM model and graph neural network to jointly encode the data to generate multimodal parameter time series joint graph encoding features. 4. System regulation: Based on the analysis results, generate control instructions to adjust the operation of smart home devices, such as adjusting the indoor light intensity. 5. Effect verification: record user feedback and the actual control effect of the system, and evaluate the accuracy of the system and user satisfaction.

[0073] For the other embodiment above, a verification field 2 is also provided, and the verification field 2 is a laboratory environment, and its goal is to perform precise verification in a controlled environment and eliminate external interference factors. The following steps are provided for verification field 2, and the steps include: 1. Laboratory configuration: simulate the home environment in the laboratory, and install smart wearable devices, environmental sensors and smart home devices. 2. Data acquisition and transmission: perform data acquisition and transmission in the same way as the home environment. 3. Simulation experiment: design a variety of experimental scenarios to simulate different user physiological states and environmental conditions. Record the system's control instructions and actual execution effects in different scenarios. 4. Data analysis: verify the system's analysis accuracy and response speed by comparing experimental data and system output. 5. Result evaluation: evaluate according to the experimental results, adjust and optimize the system algorithm and configuration.

[0074] For another embodiment described above, a verification field three is also provided, and the verification field three is a public place, and its goal is to verify the stability and adaptability of the system in a complex environment. The following steps are provided for verification field three, and the steps include: 1. Installation and configuration: Install smart wearable devices, environmental sensors and smart home devices in public places (such as office buildings and shopping malls). 2. Data collection: Collect physiological parameters and environmental parameters of multiple users and obtain astronomical data. 3. Data processing and analysis: Perform real-time data processing and analysis to generate control instructions. 4. System regulation: Adjust the operation of smart devices in public places, such as lights, air conditioners, etc. 5. Effect verification: Record environmental changes and user feedback in public places, and evaluate the performance of the system in complex environments. The expected result is that through verification in different fields, the practicality, stability and accuracy of the system can be comprehensively evaluated to ensure its reliability and user satisfaction in practical applications.

[0075] As described above, the home linkage system 300 based on the multi-dimensional fusion of personal physiology, environment and astronomy according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a home linkage algorithm based on the multi-dimensional fusion of personal physiology, environment and astronomy. In a possible implementation, the home linkage system 300 based on the multi-dimensional fusion of personal physiology, environment and astronomy according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the home linkage system 300 based on the multi-dimensional fusion of personal physiology, environment and astronomy can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the home linkage system 300 based on the multi-dimensional fusion of personal physiology, environment and astronomy can also be one of the many hardware modules of the wireless terminal.

[0076] Alternatively, in another example, the home linkage system 300 based on multi-dimensional fusion of personal physiology, environment and astronomy and the wireless terminal may also be separate devices, and the home linkage system 300 based on multi-dimensional fusion of personal physiology, environment and astronomy may be connected to the wireless terminal via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0077] Furthermore, a home linkage method based on multi-dimensional integration of personal physiology, environment and astronomy is also provided.

[0078] Figure 5 FIG. 1 is a flow chart of a home linkage method based on multi-dimensional integration of personal physiology, environment and astronomy according to an embodiment of the present application. Figure 5 As shown, according to the home linkage method based on multi-dimensional fusion of personal physiology, environment and astronomy according to the embodiment of the present application, the method includes the following steps: S1, obtaining a time series of user physiological parameters collected by a smart wearable device, wherein the physiological parameters include heart rate values, blood pressure values, blood oxygen saturation and blood sugar level values; S2, obtaining a time series of environmental parameters collected by an indoor sensor network, wherein the environmental parameters include temperature values, humidity values ​​and light intensity values; S3, obtaining a time series of astronomical data collected by an astronomical observatory, wherein the astronomical data includes solar electromagnetic activity and lunar gravity; S4, analyzing the time series of the user's physiological parameters, the time series of the environmental parameters and the time series of the astronomical data to generate a control instruction, and sending the control instruction to the smart home device via wireless communication.

[0079] In summary, according to the embodiment of the present application, the home linkage method based on the multi-dimensional integration of personal physiology, environment and astronomy is explained, which uses AI-based data processing and analysis algorithms to analyze the time series of user physiological parameters (heart rate values, blood pressure values, blood oxygen saturation and blood sugar level values), environmental parameters (temperature values, humidity values ​​and light intensity values) and astronomical data (solar electromagnetic activity and lunar gravity) collected. The time series performs implicit correlation feature analysis and semantic correlation calculation, thereby automatically judging whether to reduce the indoor light intensity based on the multimodal time series joint coding features between multiple parameters in these data. In this way, the system can collect and comprehensively analyze the user's physiological data, indoor environmental parameters and astronomical information in real time, realize the precise control and intelligent adjustment of smart home devices, so as to provide users with more accurate and personalized home environment adjustment solutions and improve the overall living experience.

[0080] In one example, in actual application, the user needs to wear a smart wearable device, which will monitor the user's physiological parameters in real time and transmit the data to the central control system via wireless transmission. At the same time, the indoor environment sensor will also collect indoor environment parameters in real time and send them to the central control system synchronously.

[0081] After receiving the data, the central control system will use the preset algorithms and models to analyze and process it to determine the user's physiological state and indoor environment requirements. Based on the judgment results, the system will generate corresponding control instructions and send them to smart home devices through wireless communication. After receiving the instructions, the smart home devices will automatically perform corresponding operations, such as adjusting the air conditioning temperature, opening or closing windows, adjusting the opening and closing of curtains, changing the brightness and color temperature of lights, playing specific music, etc., to meet the user's personalized needs.

[0082] During the implementation process, the central control system will also make corresponding adjustments based on astronomical information. For example, when the moon's gravity is strong, the system may automatically adjust the angle and hardness of the bed to provide a more comfortable sleeping environment; when the solar electromagnetic activity is strong, the system may automatically lower the indoor light intensity to reduce visual stimulation to the user.

[0083] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A home linkage system based on the multi-dimensional integration of personal physiology, environment and astronomy, characterized in that: include: A personal physiological data acquisition module, used to obtain a time series of the user's physiological parameters collected by the smart wearable device, wherein the physiological parameters include heart rate value, blood pressure value, blood oxygen saturation and blood sugar level value; An environmental data acquisition module, used to acquire a time series of environmental parameters collected by an indoor sensor network, wherein the environmental parameters include temperature, humidity and light intensity values; An astronomical data acquisition module, used to acquire a time series of astronomical data collected by an astronomical observatory, wherein the astronomical data includes solar electromagnetic activity and lunar gravity; The linkage control module is used to analyze the time series of the user's physiological parameters, the time series of the environmental parameters and the time series of the astronomical data to generate control instructions, and send the control instructions to the smart home device through wireless communication.

2. The home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to claim 1 is characterized in that: The linkage control module comprises: a parameter data reorganization unit, for reorganizing the time series of the user's physiological parameters, the time series of the environmental parameters, and the time series of the astronomical data according to the time dimension to obtain a time series of heart rate values, a time series of blood pressure values, a time series of blood oxygen saturation, a time series of blood sugar level values, a time series of temperature values, a time series of humidity values, a time series of light intensity values, a time series of solar electromagnetic activity, and a time series of lunar gravity; a parameter data time series encoding unit, used for inputting the time series of the heart rate value, the time series of the blood pressure value, the time series of the blood oxygen saturation, the time series of the blood sugar level value, the time series of the temperature value, the time series of the humidity value, the time series of the light intensity value, the time series of the solar electromagnetic activity and the time series of the lunar gravity into a sequence encoder respectively to obtain a heart rate time series associated implicit feature vector, a blood pressure time series associated implicit feature vector, a blood oxygen saturation time series associated implicit feature vector, a blood sugar level time series associated implicit feature vector, a temperature time series associated implicit feature vector, a humidity time series associated implicit feature vector, a light intensity time series associated implicit feature vector, a solar electromagnetic activity time series associated implicit feature vector, and a lunar gravity time series associated implicit feature vector; a parameter data time series feature association degree calculation unit, used for calculating the semantic association degree between any two feature vectors of the heart rate time series association implicit feature vector, the blood pressure time series association implicit feature vector, the blood oxygen saturation time series association implicit feature vector, the blood sugar level time series association implicit feature vector, the temperature time series association implicit feature vector, the humidity time series association implicit feature vector, the light intensity time series association implicit feature vector, the solar electromagnetic activity time series association implicit feature vector and the lunar gravity time series association implicit feature vector to obtain a semantic association degree adjacency matrix; a parameter data time series feature multimodal joint unit, for performing multimodal joint processing on the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector and the lunar gravity time series associated implicit feature vector to obtain a multimodal parameter time series joint graph encoding feature; The indoor light intensity control unit is used to obtain a control instruction based on the multi-modal parameter time series joint graph encoding feature, and the control instruction is used to indicate whether to reduce the indoor light intensity.

3. The home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to claim 2 is characterized in that: The parameter data time series encoding unit is used to: input the time series of the heart rate value, the time series of the blood pressure value, the time series of the blood oxygen saturation, the time series of the blood sugar level value, the time series of the temperature value, the time series of the humidity value, the time series of the light intensity value, the time series of the solar electromagnetic activity and the time series of the lunar gravity into a sequence encoder based on the Bi-LSTM model respectively to obtain the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector, and the lunar gravity time series associated implicit feature vector.

4. The home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to claim 3 is characterized in that: The parameter data temporal feature correlation calculation unit comprises: A heart rate-blood pressure semantic association subunit, used for performing vector multiplication of the transposed vector of the heart rate time series association implicit feature vector and the blood pressure time series association implicit feature vector to obtain a heart rate-blood pressure semantic association value; A semantic norm value calculation subunit, used for respectively calculating a norm of the heart rate time series associated implicit feature vector and the blood pressure time series associated implicit feature vector to obtain a heart rate time series associated semantic norm value and a blood pressure time series associated semantic norm value; A heart rate-blood pressure semantic interaction subunit, used for multiplying the heart rate time series associated semantic norm value and the blood pressure time series associated semantic norm value to obtain a heart rate-blood pressure semantic interaction value; The semantic association calculation subunit is used to divide the heart rate-blood pressure semantic association value by the heart rate-blood pressure semantic interaction value to obtain the semantic association between the heart rate time series association implicit feature vector and the blood pressure time series association implicit feature vector.

5. The home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to claim 4 is characterized in that: The parameter data time series feature multimodal joint unit is used to: use the heart rate time series associated implicit feature vector, the blood pressure time series associated implicit feature vector, the blood oxygen saturation time series associated implicit feature vector, the blood sugar level time series associated implicit feature vector, the temperature time series associated implicit feature vector, the humidity time series associated implicit feature vector, the light intensity time series associated implicit feature vector, the solar electromagnetic activity time series associated implicit feature vector and the lunar gravity time series associated implicit feature vector as a set of nodes and use the semantic association adjacency matrix as a set of edges between nodes, input the set of nodes and the set of edges into a graph neural network model to obtain a multimodal parameter time series joint graph encoding matrix as the multimodal parameter time series joint graph encoding feature.

6. The home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to claim 5 is characterized in that: The indoor light intensity control unit is used to: input the multimodal parameter time series joint graph encoding matrix into a classifier-based control instruction generator to obtain the control instruction, and the control instruction is used to indicate whether to reduce the indoor light intensity.

7. The home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to claim 6 is characterized in that: It also includes a training module for training the Bi-LSTM model-based sequence encoder, the graph neural network model and the classifier-based control instruction generator.

8. The home linkage system based on multi-dimensional integration of personal physiology, environment and astronomy according to claim 7 is characterized in that: The training module comprises: A training data acquisition unit is used to acquire training data, wherein the training data includes a time series of physiological parameters of a training user collected by a smart wearable device, wherein the physiological parameters include heart rate values, blood pressure values, blood oxygen saturation, and blood sugar level values; acquire a time series of training environment parameters collected by an indoor sensor network, wherein the environment parameters include temperature values, humidity values, and light intensity values; and acquire a time series of training astronomical data collected by an astronomical observatory, wherein the astronomical data includes solar electromagnetic activity and lunar gravity; a training data reorganization unit, for reorganizing the time series of the training user's physiological parameters, the time series of the training environment parameters, and the time series of the training astronomical data according to the time dimension to obtain a time series of training heart rate values, a time series of training blood pressure values, a time series of training blood oxygen saturation values, a time series of training blood sugar level values, a time series of training temperature values, a time series of training humidity values, a time series of training light intensity values, a time series of training solar electromagnetic activity, and a time series of training lunar gravity; A training data time series encoding unit, used for inputting the time series of the training heart rate value, the time series of the training blood pressure value, the time series of the training blood oxygen saturation, the time series of the training blood sugar level value, the time series of the training temperature value, the time series of the training humidity value, the time series of the training light intensity value, the time series of the training solar electromagnetic activity and the time series of the training lunar gravity into a sequence encoder based on a Bi-LSTM model to obtain a training heart rate time series associated implicit feature vector, a training blood pressure time series associated implicit feature vector, a training blood oxygen saturation time series associated implicit feature vector, a training blood sugar level time series associated implicit feature vector, a training temperature time series associated implicit feature vector, a training humidity time series associated implicit feature vector, a training light intensity time series associated implicit feature vector, a training solar electromagnetic activity time series associated implicit feature vector, and a training lunar gravity time series associated implicit feature vector; a training data time series feature association degree calculation unit, used for calculating the semantic association degree between any two feature vectors of the training heart rate time series association implicit feature vector, the training blood pressure time series association implicit feature vector, the training blood oxygen saturation time series association implicit feature vector, the training blood sugar level time series association implicit feature vector, the training temperature time series association implicit feature vector, the training humidity time series association implicit feature vector, the training light intensity time series association implicit feature vector, the training solar electromagnetic activity time series association implicit feature vector and the training lunar gravity time series association implicit feature vector to obtain a training semantic association degree adjacency matrix; A training data time series feature multimodal joint unit, used for taking the training heart rate time series associated implicit feature vector, the training blood pressure time series associated implicit feature vector, the training blood oxygen saturation time series associated implicit feature vector, the training blood sugar level time series associated implicit feature vector, the training temperature time series associated implicit feature vector, the training humidity time series associated implicit feature vector, the training light intensity time series associated implicit feature vector, the training solar electromagnetic activity time series associated implicit feature vector and the training lunar gravity time series associated implicit feature vector as a set of nodes and taking the training semantic association adjacency matrix as a set of edges between nodes, inputting the set of nodes and the set of edges into a graph neural network model to obtain a training multimodal parameter time series joint graph encoding matrix; A classification loss unit, used for inputting the training multimodal parameter temporal joint graph encoding matrix into a classifier-based control instruction generator to obtain a classification loss function value; A multimodal parameter temporal joint graph coding loss calculation unit, used to calculate a multimodal parameter temporal joint graph coding loss function value based on the multimodal parameter temporal joint graph coding loss term; A weighting unit, used to calculate a weighted sum of the classification loss function value and the multimodal parameter temporal joint graph encoding loss function value to obtain a final loss function value; A training unit is used to train the Bi-LSTM model-based sequence encoder, the graph neural network model and the classifier-based control instruction generator based on the final loss function value.

9. A home linkage method based on multi-dimensional integration of personal physiology, environment and astronomy, characterized in that: include: Acquire a time series of physiological parameters of a user collected by a smart wearable device, wherein the physiological parameters include heart rate value, blood pressure value, blood oxygen saturation and blood sugar level value; Acquire a time series of environmental parameters collected by an indoor sensor network, wherein the environmental parameters include temperature, humidity and light intensity; Acquire a time series of astronomical data collected by an astronomical observatory, wherein the astronomical data includes solar electromagnetic activity and lunar gravity; The time series of the user's physiological parameters, the time series of the environmental parameters and the time series of the astronomical data are analyzed to generate control instructions, and the control instructions are sent to the smart home device through wireless communication.

Images