Sleep environment analysis and control method based on multi-source data analysis

Abstract

The present invention relates to a sleep environment analysis and control method based on multi-source data analysis, and belongs to the technical field of sleep environment control. The present invention collects the user's brain wave data information, constructs a sleep state recognition model based on the user's brain wave data information, thereby identifying the user's sleep state through the sleep state recognition model, and constructing a user sleep data knowledge graph. Finally, based on the actual sound energy data of the house within a preset range, the user's sleep data knowledge graph and the user's sleep state, the interference probability value of the current sound energy data is obtained, and the sleep environment is dynamically controlled based on the interference probability value of the current sound energy data. The present invention integrates the user's sleep state and external noise data to evaluate the interference probability value of the noise data on the user, and then evaluates the user's awakening situation, thereby improving the control accuracy of the sleep environment and ensuring the user's sleep quality.

Description

Technical Field

[0001] The present invention relates to the technical field of sleep environment control, and in particular to a sleep environment analysis and control method based on multi-source data analysis. Background Art

[0002] Sleep is an active process, a necessary rest for restoring energy. It promotes mental and physical recovery, is fundamental to maintaining health and strength, and guarantees high productivity. Modern medicine generally believes that sleep is a necessary process for the brain to distribute and consolidate stimuli and their connections to appropriate nerve cells (reorganize information). This process temporarily shifts the focus of attention to previously less excited nerve cells, allowing these cells to take over the majority of the body's vital activities. The more excited nerve cells that originally received and processed internal and external stimuli are largely shielded from these stimuli. During sleep, active activity decreases, allowing physical strength to recover. Compared to wakefulness, sleep reduces contact with the surrounding environment, diminishes conscious awareness, muscles relax, nerve reflexes weaken, body temperature drops, heart rate slows, blood pressure drops slightly, metabolism slows, and gastrointestinal motility is significantly reduced. The quality of sleep is closely related to the sleeping environment, and noise interference is also an important factor affecting the sleep state. In the existing technology, traditional sleep monitoring technology has three major defects: limited single signal dimension, inaccurate sleep environment adjustment, and insufficient personalized adaptation, which results in the user's sleep environment control not meeting the expected effect. Summary of the Invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a sleep environment analysis and control method based on multi-source data analysis.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is:

[0005] A first aspect of the present invention provides a sleep environment analysis and control method based on multi-source data analysis, comprising the following steps:

[0006] Collecting the type and coverage data of the sound barriers in the target area, and obtaining the actual sound energy data of the house within a preset range based on the type and coverage data of the sound barriers in the target area;

[0007] Collecting the user's brainwave data information through a brainwave data acquisition device, and building a sleep state recognition model based on the user's brainwave data information;

[0008] Identify the user's sleep state through the sleep state recognition model and build a user sleep data knowledge graph;

[0009] Obtaining an interference probability value of current sound energy data based on actual sound energy data of the house within a preset range, a user sleep data knowledge graph, and the user's sleep state, and dynamically adjusting the sleeping environment based on the interference probability value of the current sound energy data;

[0010] The actual sound energy data of the house within the preset range is obtained according to the type and coverage data of the sound barrier in the target area, specifically:

[0011] Acquire real-time acoustic energy data in the target area through acoustic energy data acquisition equipment, construct an acoustic energy simulation penetration scenario, and collect the sound absorption coefficient of the type of sound barrier in the target area;

[0012] Inputting the type of sound barrier in the target area and the real-time sound energy data information in the target area under the coverage data condition into the sound energy simulation penetration scenario, and performing sound energy penetration simulation in combination with the sound absorption coefficient of the type of sound barrier in the target area;

[0013] Acquiring the acoustic energy absorbed by the material and the acoustic energy reflected by the material through an acoustic energy penetration simulation, and calculating the acoustic energy data transmitted through the material based on the acoustic energy absorbed by the material and the acoustic energy reflected by the material;

[0014] An attenuation analysis is performed on the acoustic energy data transmitted through the material under the environmental data information in the house to obtain actual acoustic energy data of the house within a preset range, and the actual acoustic energy data of the house within the preset range is output.

[0015] Furthermore, in the sleep environment analysis and control method based on multi-source data analysis, a sleep state recognition model is constructed according to the user's brain wave data information, specifically:

[0016] Constructing a search tag based on the user's brainwave data information, searching through big data based on the search tag to obtain sleep state information underlying each brainwave data, and constructing a sleep state recognition model based on a deep neural network;

[0017] The first layer of the deep neural network is a two-stream temporal CNN to process temporal physiological signals, and the second layer is a graph convolutional network to process the sleep state information under each brain wave data;

[0018] Construct a topological structure diagram of EEG data and sleep state information, and introduce a Transformer encoder in the third layer to capture long-term dependencies;

[0019] Based on long-term dependencies, the EEG data of each timestamp is used as the model input, and each sleep state information is used as the model output. The probability distribution of each sleep state information is generated, and the prediction results are output based on the probability analysis.

[0020] Furthermore, in the sleep environment analysis and control method based on multi-source data analysis, the sleep state recognition model is used to identify the user's sleep state, specifically:

[0021] Acquiring real-time brainwave data information of the user, inputting the real-time brainwave data information of the user into the sleep state recognition model for recognition, and obtaining a probability distribution of each sleep state;

[0022] The sleeping state with the maximum probability distribution is obtained as the sleeping state of the user, and the sleeping state of the user is output.

[0023] Furthermore, in the sleep environment analysis and control method based on multi-source data analysis, a user sleep data knowledge graph is constructed, specifically:

[0024] Obtaining non-interference events and interference events of each sound energy data for the user in different sleep states;

[0025] Setting the number of non-interference events and interference events, and performing statistics on the non-interference events and interference events of each sound energy data for the user in different sleep states based on the statistical number;

[0026] Obtain, through statistics, the interference probability values ​​of each sound energy data for the user in different sleep states, construct a user sleep data knowledge graph, construct a first graph node based on the sound energy data, the sleep state as the second graph node, and the interference probability value as the third graph node;

[0027] A directed description relationship is constructed, and the first graph node, the second graph node, and the third graph node are connected based on the directed description relationship to construct a user sleep data topology structure graph, and the user sleep data topology structure graph is input into the user sleep data knowledge graph for embedding representation.

[0028] Furthermore, in the sleep environment analysis and control method based on multi-source data analysis, the interference probability value of the current sound energy data is obtained based on the actual sound energy data of the house within a preset range, the user's sleep data knowledge graph, and the user's sleep state, specifically including:

[0029] Taking the actual sound energy data of the house within a preset range and the sleeping state of the user as verification data, and inputting the verification data into the user sleep data knowledge graph for data node embedding analysis;

[0030] Calculating the Euclidean distance between the verification data and the data node, setting a Euclidean distance threshold, and determining whether the Euclidean distance between the verification data and the data node is greater than the Euclidean distance threshold;

[0031] When the Euclidean distance value is greater than the Euclidean distance threshold, the data node corresponding to the Euclidean distance value greater than the Euclidean distance threshold is obtained, and the interference probability value of the current acoustic energy data is obtained through the data node corresponding to the Euclidean distance value greater than the Euclidean distance threshold.

[0032] Furthermore, in the sleep environment analysis and control method based on multi-source data analysis, the sleep environment is dynamically regulated based on the interference probability value of the current sound energy data, specifically:

[0033] Setting an interference probability threshold, and determining whether the interference probability value of the current acoustic energy data is greater than the interference probability threshold;

[0034] When the interference probability value of the current sound energy data is greater than the interference probability threshold, controlling a device associated with noise control through the Internet of Things, and dynamically adjusting the sleeping environment through the device associated with noise control;

[0035] When the interference probability value of the current sound energy data is not greater than the interference probability threshold, the current working state of the device associated with noise control is maintained unchanged.

[0036] A second aspect of the present invention provides a sleep environment analysis and control system based on multi-source data analysis, including a memory and a processor. The memory includes a sleep environment analysis and control method program based on multi-source data analysis. When the sleep environment analysis and control method program based on multi-source data analysis is executed by the processor, the steps of any one of the sleep environment analysis and control methods based on multi-source data analysis are implemented.

[0037] A third aspect of the present invention provides a computer-readable storage medium, comprising a sleep environment analysis and control method program based on multi-source data analysis. When the sleep environment analysis and control method program based on multi-source data analysis is executed by a processor, the program implements any one of the steps of the sleep environment analysis and control method based on multi-source data analysis.

[0038] The present invention solves the defects existing in the background technology and has the following beneficial effects:

[0039] The present invention collects the type and coverage data of sound barriers in the target area, obtains the actual sound energy data of the house within a preset range based on the type and coverage data of sound barriers in the target area, and then collects the user's brain wave data information through a brain wave data acquisition device, and constructs a sleep state recognition model based on the user's brain wave data information, thereby identifying the user's sleep state through the sleep state recognition model and constructing a user sleep data knowledge graph. Finally, based on the actual sound energy data of the house within the preset range, the user's sleep data knowledge graph and the user's sleep state, the interference probability value of the current sound energy data is obtained, and the sleep environment is dynamically controlled based on the interference probability value of the current sound energy data. The present invention evaluates the interference probability value of the noise data on the user by integrating the user's sleep state and external noise data, and then evaluates the user's awakening situation, thereby improving the control accuracy of the sleep environment and ensuring the user's sleep quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, without paying any creative work, they can also obtain drawings of other embodiments based on these drawings.

[0041] Figure 1 The overall flow chart of the sleep environment analysis and control method based on multi-source data analysis is shown;

[0042] Figure 2 A system block diagram of a sleep environment analysis and control system based on multi-source data analysis is shown. DETAILED DESCRIPTION

[0043] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0045] like Figure 1 As shown, the first aspect of the present invention provides a sleep environment analysis and control method based on multi-source data analysis, comprising the following steps:

[0046] S102: Collect the type and coverage data of the sound barriers in the target area, and obtain the actual sound energy data of the house within a preset range based on the type and coverage data of the sound barriers in the target area;

[0047] S104: Collecting the user's brain wave data information through a brain wave data acquisition device, and building a sleep state recognition model based on the user's brain wave data information;

[0048] S106: Identify the user's sleep state through a sleep state recognition model and construct a user sleep data knowledge graph;

[0049] S108: Obtain the interference probability value of the current sound energy data based on the actual sound energy data of the house within a preset range, the user's sleep data knowledge graph, and the user's sleep state, and dynamically adjust the sleeping environment based on the interference probability value of the current sound energy data.

[0050] It should be noted that this invention integrates the user's sleep state and external noise data to assess the probability of noise interference with the user, and then estimates the user's awakening situation, thereby improving the control accuracy of the sleep environment and ensuring the user's sleep quality. The scenarios involved in this invention include hospital scenarios, home scenarios, etc.

[0051] Furthermore, in the sleeping environment analysis and control method based on multi-source data analysis, the actual sound energy data of the house within a preset range is obtained according to the type and coverage data of the sound barrier in the target area, specifically:

[0052] Acquire real-time acoustic energy data (sound volume, sound direction, etc.) in the target area through acoustic energy data acquisition equipment, construct an acoustic energy simulation penetration scenario, and collect the sound absorption coefficient of the type of sound barrier in the target area;

[0053] Input the real-time sound energy data in the target area under the conditions of the type of sound barrier in the target area and coverage data (sound barrier coverage area, thickness data, etc.) into the sound energy simulation penetration scenario, and perform sound energy penetration simulation in combination with the sound absorption coefficient of the type of sound barrier in the target area;

[0054] Through the sound energy penetration simulation, the sound energy absorbed by the material and the sound energy reflected by the material are obtained, and the sound energy data transmitted through the material is calculated based on the sound energy absorbed by the material and the sound energy reflected by the material;

[0055] Perform attenuation analysis on the sound energy data passing through the material under the environmental data information in the house (such as temperature data and humidity data) (due to changes in the environment, the propagation medium will undergo certain changes, such as the attenuation degree of sound energy in different environments), obtain the actual sound energy data of the house within the preset range, and output the actual sound energy data of the house within the preset range.

[0056] It should be noted that acoustic energy is a form of mechanical energy that propagates through media such as air, water, or solids, emitted by the vibration of objects, and has properties such as frequency, amplitude, and duration. It manifests itself in various forms, including audible sound within the frequency range perceptible to the human ear (20 Hz to 20 kHz), infrasound below this range, and ultrasound above this range. Acoustic energy data is a collection of various information related to acoustic energy, obtained through the measurement and analysis of acoustic energy. Examples include acoustic energy density (the amount of acoustic energy per unit volume, measured in J / m³) and acoustic power (the amount of acoustic energy flux per unit time through a given area perpendicular to the direction of sound wave propagation, measured in W).

[0057] It should be noted that the types of sound barriers include walls, windows and other sound insulation materials. The sound energy passing through the material satisfies the following relationship:

[0058]

[0059]

[0060] in, is the total incident sound energy, is the sound energy absorbed by the material, is the sound energy that passes through the material, is the sound energy reflected by the material, is the absorption coefficient of the material.

[0061] By constructing a sound energy simulation penetration scenario through virtual reality technology or digital twin technology, the type of sound barrier in the target area and the real-time sound energy data information in the target area under the coverage data conditions are input into the sound energy simulation penetration scenario, and the sound absorption coefficient of the type of sound barrier in the target area is combined to perform sound energy penetration simulation, so as to more accurately estimate the noise data in the house.

[0062] Furthermore, in the sleep environment analysis and control method based on multi-source data analysis, a sleep state recognition model is constructed based on the user's brain wave data information, specifically:

[0063] Build retrieval tags based on the user's brainwave data, search through big data based on the retrieval tags, obtain sleep state information under each brainwave data, and build a sleep state recognition model based on a deep neural network;

[0064] The first layer of the deep neural network is a two-stream temporal CNN to process temporal physiological signals, and the second layer is a graph convolutional network to process the sleep state information under each brain wave data;

[0065] Construct a topological structure diagram of EEG data and sleep state information, and introduce a Transformer encoder in the third layer to capture long-term dependencies;

[0066] Based on long-term dependencies, the EEG data of each timestamp is used as the model input, and each sleep state information is used as the model output. The probability distribution of each sleep state information is generated, and the prediction results are output based on the probability analysis.

[0067] It should be noted that the first layer of the deep neural network is a dual-stream Temporal CNN to process time-series brain wave signals, the second layer is a graph convolutional network (GCN) to model the brain wave data-sleep state information topological structure diagram, and the third layer introduces a Transformer encoder to capture long-term dependencies. The model input is 5-minute sliding window data, and the output is the probability distribution of the five stages of sleep: falling asleep N1, light sleep N2, deep sleep N3, REM, and Wake. A sleep state recognition model is formed, and the sleep state information corresponding to the largest probability is selected as the output result. For example, when the probability distribution value of the Wake sleep state is the largest during the prediction process, the Wake sleep state is output as the prediction result.

[0068] Furthermore, in the sleep environment analysis and control method based on multi-source data analysis, the sleep state of the user is identified through a sleep state recognition model, specifically:

[0069] Obtain the user's real-time brainwave data information, input the user's real-time brainwave data information into the sleep state recognition model for recognition, and obtain the probability distribution of each sleep state;

[0070] The sleeping state with the maximum probability distribution is obtained as the sleeping state of the user, and the sleeping state of the user is output.

[0071] Furthermore, in the sleep environment analysis and control method based on multi-source data analysis, a user sleep data knowledge graph is constructed, specifically:

[0072] Obtaining non-interference events and interference events of each sound energy data for the user in different sleep states;

[0073] Set the number of non-interference events and interference events (e.g., 10,000 times), and count the non-interference events and interference events of each sound energy data for the user in different sleep states based on the statistical number;

[0074] Through statistics, the interference probability values ​​of each sound energy data for users in different sleep states are obtained, and the user sleep data knowledge graph is constructed. The first graph node is constructed based on the sound energy data, the sleep state is used as the second graph node, and the interference probability value is used as the third graph node;

[0075] Construct a directed description relationship, connect the first graph node, the second graph node, and the third graph node based on the directed description relationship, construct a user sleep data topology structure graph, and input the user sleep data topology structure graph into the user sleep data knowledge graph for embedding representation.

[0076] It should be noted that due to different sleep states, the user will be affected differently by the sound energy data. For example, the deep sleep state N3 is not easily affected by noise. This method can form a topological structure diagram of the user's sleep data, thereby combining the noise data with the user's interference probability value in the current sleep state. When the interference probability value is greater than the preset interference probability value, it means that the probability of the user waking up is high.

[0077] Furthermore, in the sleep environment analysis and control method based on multi-source data analysis, the interference probability value of the current sound energy data is obtained based on the actual sound energy data of the house within a preset range, the user's sleep data knowledge graph, and the user's sleep state, specifically including:

[0078] The actual sound energy data of the house within the preset range and the user's sleeping state are used as verification data, and the verification data is input into the user's sleep data knowledge graph for data node embedding analysis;

[0079] Calculate the Euclidean distance between the verification data and the data node, set the Euclidean distance threshold, and determine whether the Euclidean distance between the verification data and the data node is greater than the Euclidean distance threshold;

[0080] When the Euclidean distance value is greater than the Euclidean distance threshold, the data node corresponding to the Euclidean distance value greater than the Euclidean distance threshold is obtained, and the interference probability value of the current acoustic energy data is obtained through the data node corresponding to the Euclidean distance value greater than the Euclidean distance threshold.

[0081] Furthermore, in the sleep environment analysis and control method based on multi-source data analysis, the sleep environment is dynamically regulated based on the interference probability value of the current sound energy data, specifically:

[0082] Set an interference probability threshold and determine whether the interference probability value of the current sound energy data is greater than the interference probability threshold;

[0083] When the interference probability value of the current sound energy data is greater than the interference probability threshold, the device associated with the noise control is controlled through the Internet of Things, and the sleeping environment is dynamically adjusted through the device associated with the noise control;

[0084] When the interference probability value of the current sound energy data is not greater than the interference probability threshold, the working state of the current device associated with the noise control is maintained unchanged.

[0085] It should be noted that the equipment associated with noise control includes sound insulation panels, automatic windows and other equipment. This method can ensure the user's sleep quality.

[0086] In addition, the method further comprises:

[0087] Construct a Markov decision model, using sleep state and environmental parameters as dynamic parameters, and device operating parameters as parameters to be optimized. Increased deep sleep duration is considered a positive reward, while increased awakening times are considered a negative reward.

[0088] Inputting the dynamic parameter into the Markov decision model, calculating the state transition probability of the dynamic parameter transferring to another dynamic parameter within a preset time, and setting a state transition probability threshold;

[0089] When the state transition probability is greater than the state transition probability threshold, it is updated to another dynamic parameter, and the proximal policy optimization (PPO) algorithm is used to train the agent. A virtual environment is constructed through digital twin technology, and a preset number of operations are performed in the virtual environment to search for the number of positive rewards and the number of negative rewards for the current parameter to be optimized;

[0090] Calculate the ratio between the number of positive rewards and the number of negative rewards for the current parameter to be optimized. When the ratio is greater than a preset ratio, maintain the parameter to be optimized unchanged. When the ratio is not greater than the preset ratio, adjust the parameter to be optimized until it is greater than the preset ratio.

[0091] It should be noted that when the state transition probability exceeds the state transition probability threshold, it is updated to another dynamic parameter, and the parameter is updated in real time. When the ratio is greater than a preset ratio, it indicates that the user's sleep quality is high; otherwise, it indicates that the user is prone to waking up. This method can timely update the user's sleep state and sleep environment parameters, thereby further searching for parameters to be optimized, thereby improving the user's sleep quality.

[0092] In addition, the method further comprises:

[0093] LiDAR is deployed in the bedroom to scan and generate a 3D point cloud model, marking the locations of doors, windows, heaters, and air conditioners. Environmental sensors transmit temperature, humidity, light, and noise data in real time, and wearable devices upload physiological data. All data is collected and compiled to construct indoor data.

[0094] Based on the indoor data, an interactive virtual bedroom is constructed using Unreal Engine. A 3D point cloud is imported to generate a precise spatial model. An air flow model is established using computational fluid dynamics (CFD) within the precise spatial model generated by the 3D point cloud. The diffusion path of the air conditioning airflow is simulated, and the temperature field distribution is calculated using thermodynamic equations.

[0095] Acquiring historical physiological data information of the user, and constructing a personalized physiological response model based on the historical physiological data information of the user, wherein the personalized physiological response model includes a thermal comfort model, a sound sensitivity model, and a light awakening model;

[0096] obtaining sleeping position information and sleeping state information of the user, evaluating the user's sleep sensitivity based on the personalized physiological response model in combination with the temperature field distribution, the user's sleeping position information and the sleeping state information, and obtaining a sleep sensitivity evaluation membership degree;

[0097] A sleep sensitivity evaluation membership evaluation index is set. When the sleep sensitivity evaluation membership is greater than the sleep sensitivity evaluation membership evaluation index, the current indoor environmental parameters are maintained unchanged. When the sleep sensitivity evaluation membership is not greater than the sleep sensitivity evaluation membership evaluation index, the current indoor environmental parameters are adjusted.

[0098] It should be noted that the thermal comfort model includes skin temperature, metabolic rate, and thermal sensation data; the acoustic sensitivity model includes the auditory filter group (equivalent rectangular bandwidth ERB) and user sensitivity to sound energy; the light arousal model includes the association between the ipRGC cell response curve (480nm peak) and sleep stages, and user sensitivity to light (such as sensitivity to a 10lx light intensity range). By building a personalized physiological response model based on sensitivity and combining it with the sleep environment, a personalized sleep environment can be developed for the user, further improving the user's sleep quality. Among them, the sleep sensitivity evaluation membership includes low sensitivity, medium sensitivity, and high sensitivity.

[0099] like Figure 2 As shown, the second aspect of the present invention provides a sleep environment analysis and control system 4 based on multi-source data analysis, including a memory 41 and a processor 42. The memory 41 includes a sleep environment analysis and control method program based on multi-source data analysis. When the sleep environment analysis and control method program based on multi-source data analysis is executed by the processor 42, any step of the sleep environment analysis and control method based on multi-source data analysis is implemented.

[0100] A third aspect of the present invention provides a computer-readable storage medium, comprising a sleep environment analysis and control method program based on multi-source data analysis. When the sleep environment analysis and control method program based on multi-source data analysis is executed by a processor, any step of the sleep environment analysis and control method based on multi-source data analysis is implemented.

[0101] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0102] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0103] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0104] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0105] Alternatively, if the integrated units described above are implemented as software modules and sold or used as standalone products, they can also be stored on a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product, stored on a storage medium, includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute all or part of the methods of the various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as removable storage devices, ROM, RAM, magnetic disks, or optical disks.

[0106] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A sleep environment analysis and control method based on multi-source data analysis, characterized in that: The following steps are involved: Collecting the type and coverage data of the sound barriers in the target area, and obtaining the actual sound energy data of the house within a preset range based on the type and coverage data of the sound barriers in the target area; Collecting the user's brainwave data information through a brainwave data acquisition device, and building a sleep state recognition model based on the user's brainwave data information; Identify the user's sleep state through the sleep state recognition model and build a user sleep data knowledge graph; Obtaining an interference probability value of current sound energy data based on actual sound energy data of the house within a preset range, a user sleep data knowledge graph, and the user's sleep state, and dynamically adjusting the sleeping environment based on the interference probability value of the current sound energy data; The actual sound energy data of the house within the preset range is obtained according to the type and coverage data of the sound barrier in the target area, specifically: Acquire real-time acoustic energy data in the target area through acoustic energy data acquisition equipment, construct an acoustic energy simulation penetration scenario, and collect the sound absorption coefficient of the type of sound barrier in the target area; Inputting the type of sound barrier in the target area and the real-time sound energy data information in the target area under the coverage data condition into the sound energy simulation penetration scenario, and performing sound energy penetration simulation in combination with the sound absorption coefficient of the type of sound barrier in the target area; Acquiring the acoustic energy absorbed by the material and the acoustic energy reflected by the material through an acoustic energy penetration simulation, and calculating the acoustic energy data transmitted through the material based on the acoustic energy absorbed by the material and the acoustic energy reflected by the material; An attenuation analysis is performed on the acoustic energy data transmitted through the material under the environmental data information in the house to obtain actual acoustic energy data of the house within a preset range, and the actual acoustic energy data of the house within the preset range is output.

2. The sleep environment analysis and control method based on multi-source data analysis according to claim 1, characterized in that: Constructing a sleep state recognition model based on the user's brain wave data information, specifically: Constructing a search tag based on the user's brainwave data information, searching through big data based on the search tag to obtain sleep state information underlying each brainwave data, and constructing a sleep state recognition model based on a deep neural network; The first layer of the deep neural network is a dual-stream temporal CNN that processes temporal physiological signals, and the second layer is a graph convolutional network that processes the sleep state information underlying each brain wave data. Construct a topological structure diagram of EEG data and sleep state information, and introduce a Transformer encoder in the third layer to capture long-term dependencies; Based on long-term dependencies, the EEG data of each timestamp is used as the model input, and each sleep state information is used as the model output. The probability distribution of each sleep state information is generated, and the prediction results are output based on the probability analysis.

3. The sleep environment analysis and control method based on multi-source data analysis according to claim 1, characterized in that: The sleep state recognition model is used to identify the user's sleep state, specifically: Acquiring real-time brainwave data information of the user, inputting the real-time brainwave data information of the user into the sleep state recognition model for recognition, and obtaining a probability distribution of each sleep state; The sleeping state of the maximum probability distribution is obtained as the sleeping state of the user, and the sleeping state of the user is output.

4. The sleep environment analysis and control method based on multi-source data analysis according to claim 1, characterized in that: Build a user sleep data knowledge graph, specifically: Obtaining non-interference events and interference events of each sound energy data for the user in different sleep states; Setting the number of non-interference events and interference events, and performing statistics on the non-interference events and interference events of each sound energy data for the user in different sleep states based on the statistical number; Obtain, through statistics, the interference probability values ​​of each sound energy data for the user in different sleep states, construct a user sleep data knowledge graph, construct a first graph node based on the sound energy data, the sleep state as the second graph node, and the interference probability value as the third graph node; A directed description relationship is constructed, and the first graph node, the second graph node, and the third graph node are connected based on the directed description relationship to construct a user sleep data topology structure graph, and the user sleep data topology structure graph is input into the user sleep data knowledge graph for embedding representation.

5. The sleep environment analysis and control method based on multi-source data analysis according to claim 1, characterized in that: Obtaining an interference probability value of current sound energy data based on actual sound energy data of the house within a preset range, a user's sleep data knowledge graph, and the user's sleep state, specifically includes: Taking the actual sound energy data of the house within a preset range and the sleeping state of the user as verification data, and inputting the verification data into the user sleep data knowledge graph for data node embedding analysis; Calculating the Euclidean distance between the verification data and the data node, setting a Euclidean distance threshold, and determining whether the Euclidean distance between the verification data and the data node is greater than the Euclidean distance threshold; When the Euclidean distance value is greater than the Euclidean distance threshold, the data node corresponding to the Euclidean distance value greater than the Euclidean distance threshold is obtained, and the interference probability value of the current acoustic energy data is obtained through the data node corresponding to the Euclidean distance value greater than the Euclidean distance threshold.

6. The sleep environment analysis and control method based on multi-source data analysis according to claim 1, characterized in that: The sleeping environment is dynamically regulated based on the interference probability value of the current sound energy data, specifically: Setting an interference probability threshold, and determining whether the interference probability value of the current acoustic energy data is greater than the interference probability threshold; When the interference probability value of the current sound energy data is greater than the interference probability threshold, controlling a device associated with noise control through the Internet of Things, and dynamically adjusting the sleeping environment through the device associated with noise control; When the interference probability value of the current sound energy data is not greater than the interference probability threshold, the current working state of the device associated with noise control is maintained unchanged.

7. Sleep environment analysis and control system based on multi-source data analysis, characterized in that: The system includes a memory and a processor, wherein the memory includes a sleep environment analysis and control method program based on multi-source data analysis, and when the sleep environment analysis and control method program based on multi-source data analysis is executed by the processor, the steps of the sleep environment analysis and control method based on multi-source data analysis are implemented.

8. A computer-readable storage medium, characterized in that The invention also includes a sleep environment analysis and control method program based on multi-source data analysis, which, when executed by a processor, implements the steps of the sleep environment analysis and control method based on multi-source data analysis according to any one of claims 1 to 6.

Images