Self-adaptive adjustment system for whole-house customized furniture under intelligent home environment perception

Abstract

The invention discloses a self-adaptive adjustment system for whole-house customized furniture under intelligent home environment perception, and relates to the technical field of furniture adjustment. The system comprises environment perception, furniture state monitoring, user behavior analysis, adaptive adjustment control, data storage and management, and a user interaction module. The advanced technology is applied, the environment and the user are accurately perceived, furniture is intelligently adjusted, and data safety and multi-element interaction are guaranteed. According to the method, the environment is accurately sensed by virtue of a nanoscale environment sensor, user behaviors are deeply analyzed by virtue of a multi-modal fusion technology, and self-adaptive adjustment of furniture is realized by virtue of model prediction control; data storage is safe and reliable, and multi-element interaction of a brain-computer interface and the like is supported; the state of the furniture can be monitored in real time, faults can be early warned in advance, and personalized, convenient, healthy and energy-saving home experience is provided for users.

Description

Technical Field

[0001] This invention relates to the field of furniture adjustment system technology, and more particularly to a whole-house customized furniture adaptive adjustment system under the perception of smart home environment. Background Technology

[0002] In the field of smart homes, customized furniture for the entire house is gradually becoming more common, but existing systems have many shortcomings in terms of environmental perception and adaptive furniture adjustment. Although early smart homes had environmental monitoring functions, the sensors had limited accuracy, only providing general data such as temperature, humidity, and air quality. They could not accurately capture subtle changes in the environment, making it difficult to meet the needs for precise furniture adjustment. For example, ordinary humidity sensors cannot detect extremely subtle fluctuations in humidity, causing delays in moisture-proofing measures for furniture and making wooden furniture prone to warping due to moisture.

[0003] Furniture condition monitoring is a weak point. Traditional methods rely heavily on regular manual inspections, making it difficult to monitor the wear and displacement of furniture components in real time. For example, the minor wear and tear on wardrobe door hinges caused by frequent opening and closing goes unnoticed until a malfunction occurs and affects usability. The lack of proactive monitoring and early maintenance mechanisms leads to a shortened lifespan for furniture.

[0004] The interaction between users and furniture is also not intelligent enough. Conventional smart home systems only support simple command operations and cannot provide users with a personalized, adaptive furniture adjustment experience based on user behavior, real-time status, and environmental changes. For example, users have different needs for temperature and humidity, lighting brightness inside bedroom wardrobes in different seasons and moods, but existing systems struggle to achieve intelligent adjustment, greatly limiting the comfort and convenience of home life. Therefore, developing a whole-house customized furniture adaptive adjustment system based on smart home environmental perception is of great significance for improving the level of home intelligence, and has the potential to break through existing limitations and create a better living environment for users. Summary of the Invention

[0005] The present invention proposes a smart home environment-sensing adaptive adjustment system for whole-house customized furniture to solve the problems mentioned in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a whole-house customized furniture adaptive adjustment system under intelligent home environment perception, comprising:

[0007] Environmental perception module: Employs a nanoscale environmental sensor array, utilizing imaging technology such as light and sound sensors to detect the distance and material information of surrounding objects; and employs sensor data fusion algorithms. Where F represents the fused environmental perception data, w i S represents the weight of the i-th sensor. iThe data collected by the i-th sensor enables the perception of environmental information. At the same time, environmental fingerprint recognition technology is introduced to generate a unique environmental fingerprint for each room or area through comprehensive analysis of environmental characteristics, which is used to identify environmental status and changes.

[0008] Furniture Condition Monitoring Module: Sensors based on microelectromechanical systems (MEMS) are installed in the whole-house customized furniture; fiber optic strain sensors are embedded at the cabinet structure connection points. By monitoring the wavelength change of the fiber optic grating, the strain of the cabinet structure is detected. Anomaly detection algorithms in machine learning are used to analyze the sensor data in real time. By monitoring the wear degree of furniture parts, the remaining service life of the parts is predicted using a wear prediction model.

[0009] User behavior analysis module: Utilizes multimodal fusion behavior analysis technology, combined with wearable device data; wearable devices collect users' physiological and motion data, and through multimodal data fusion algorithms, integrate and analyze data from different sources to construct user behavior models;

[0010] Adaptive adjustment and control module: Based on environmental perception, furniture status monitoring and user behavior analysis results, the model predictive control (MPC) algorithm is used to achieve adaptive adjustment of the whole house customized furniture; for the adjustment of the internal environment of the furniture, a composite control algorithm based on fuzzy logic and neural network is used; at the same time, a smart material-driven furniture adjustment mechanism is introduced. When the ambient temperature or user operation is triggered, the shape memory alloy deforms, which drives the cabinet door to open or close automatically.

[0011] Data storage and management module: It adopts the distributed and decentralized InterPlanetary File System (IPFS) for data storage, uses homomorphic encryption technology to enable data calculation and analysis in encrypted state, and uses knowledge graph technology to associate and integrate different types of data to build a home knowledge graph, which facilitates data query, analysis and mining.

[0012] User interaction module: Provides an interaction method based on brain-computer interface (BCI). Users wear EEG acquisition devices, and the system recognizes the user's EEG signals, interprets the user's intentions, and realizes control of furniture; combined with gesture tracking and posture recognition technology, it uses depth cameras and inertial sensors to realize interaction.

[0013] Furthermore, it also includes:

[0014] Health Monitoring and Care Module: This module integrates multiple advanced biosensors into the furniture, including a protein sensor based on surface plasmon resonance (SPR) technology. It analyzes and diagnoses health data using medical knowledge graphs and machine learning algorithms. When an abnormality is detected in the user's health, it issues an alert and provides health advice. It also incorporates health risk assessment indicators. Where R is the health risk value, w i h represents the weight of the i-th health indicator. i Let represent the degree of abnormality of the i-th health indicator.

[0015] Energy Management Module: Installs smart meters for electricity, water, and gas to monitor household energy consumption in real time; predicts energy demand at different times through energy consumption models and allocates energy resources rationally; employs energy-saving control strategies to automatically turn off electrical appliances and furniture when no one is home; and introduces energy-saving rate indicators. Where E old E represents the energy consumption before optimization. new To optimize energy consumption, we continuously optimize energy-saving strategies to reduce energy consumption costs.

[0016] Furthermore, in the environmental perception module, edge computing technology is used to complete data processing tasks on sensor nodes or edge devices. At the same time, an environmental prediction model is introduced to predict environmental changes in the future by using historical environmental data and current environmental trends. An environmental prediction model based on Long Short-Term Memory (LSTM) network is adopted to predict changes in future environmental parameters by learning the time series features in historical environmental data.

[0017] Furthermore, in the furniture status monitoring module, IoT technology is used to realize the real-time uploading and remote monitoring of furniture component status data; a fault diagnosis algorithm is used to diagnose faults in furniture components; and a convolutional neural network (CNN) is used to jointly analyze image data and sensor data of furniture components to diagnose the type and location of faults.

[0018] Furthermore, in the user behavior analysis module, the dimensions of user behavior analysis are expanded by combining users' social network data and geolocation information; by analyzing users' lifestyle habits, interests, and geographical and time information shared on social networks, furniture adjustment services can be provided to users.

[0019] Furthermore, in the adaptive adjustment control module, a reinforcement learning algorithm is introduced to continuously optimize the control strategy based on user feedback on furniture adjustment; the Deep Q-Network (DQN) algorithm in deep reinforcement learning is adopted to enable the system to continuously learn the adjustment strategy during the interaction with the user.

[0020] Furthermore, in the data storage and management module, quantum key technology is used to ensure data security; environmental perception data, furniture status data, and user behavior data are stored on nodes, with each data block containing a timestamp and data hash value information, and traceability formulas are used to ensure data security. Where H iIt is the hash value of the i-th data block, which facilitates the tracing of the data's source and history; at the same time, big data analysis technology is used to mine the data and discover user needs and furniture usage patterns.

[0021] Furthermore, the user interaction module supports voice interaction to meet user needs; at the same time, virtual reality and augmented reality technologies are used to provide users with a real-time furniture adjustment experience.

[0022] Furthermore, the entire system is equipped with self-learning and evolution capabilities; by collecting user feedback, environmental change data, and furniture usage, it uses deep learning algorithms to update and optimize the system's model and parameters. At the same time, it supports online software upgrades and function expansions, providing users with continuous high-quality services.

[0023] Compared with existing technologies, the beneficial effects of this invention are:

[0024] This system leverages advanced environmental sensing technology to monitor real-time multi-dimensional information such as indoor temperature, humidity, light intensity, and human activity. Based on this precise data, customized furniture throughout the house can achieve adaptive adjustments. For example, when it senses an increase in indoor temperature, the smart mattress can automatically adjust its firmness and breathability to ensure a comfortable sleep environment; the study chair can adjust its backrest angle and support in real time according to changes in posture and fatigue levels, effectively alleviating fatigue from prolonged sitting and creating a highly personalized and comfortable home environment for users, significantly improving their quality of life.

[0025] Traditional furniture layouts are fixed and difficult to adapt to diverse spatial needs. The furniture in this system can automatically adjust its shape and position based on environmental changes and user behavior. The living room sofa can automatically retract when not in use, freeing up more space; when guests arrive, it can quickly unfold to accommodate multiple people. Wardrobes can also automatically adjust their internal compartment layout according to seasonal changes and the amount of clothing, greatly improving space utilization and making the home more flexible and adaptable to different living scenarios. Attached Figure Description

[0026] Figure 1 This is a schematic block diagram of the whole-house customized furniture adaptive adjustment system under the intelligent home environment perception proposed in this invention;

[0027] Figure 2 This is a schematic diagram illustrating the trend of early warning rate for furniture malfunctions in the whole-house customized furniture adaptive adjustment system under the intelligent home environment perception proposed in this invention. Detailed Implementation

[0028] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0030] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection or an indirect connection through an intermediate medium; and they may refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The invention will now be described in further detail with reference to the accompanying drawings.

[0031] Reference Figures 1-2 : A specific implementation method for a whole-house customized furniture adaptive adjustment system under smart home environment perception

[0032] I. System Overall Architecture

[0033] This system comprises modules that work collaboratively: environmental sensing, furniture status monitoring, user behavior analysis, adaptive adjustment and control, data storage and management, user interaction, health monitoring and care, and energy management. These modules work closely together, enabling the customized furniture to intelligently adjust based on the environment, user behavior, and health status, creating a convenient, comfortable, healthy, and energy-efficient home environment for users.

[0034] II. Specific Implementation Details of Each Module

[0035] 1. Environmental perception module

[0036] Sensor Deployment: Nanoscale environmental sensor arrays are evenly distributed throughout key areas of the house, such as the living room and bedrooms. Nanoparticle humidity sensors, with their large specific surface area and unique activity, can accurately detect humidity fluctuations as small as 0.01% RH; carbon nanotube gas sensors utilize unique electrical properties to achieve ppb-level accuracy in detecting harmful gases; super-resolution illumination sensors, through multiple low-resolution imaging and reconstruction algorithms, achieve illumination distribution information resolution far exceeding that of traditional sensors under low light conditions; and sound sensors that simulate bat echolocation can not only measure noise but also detect the distance, shape, and material of surrounding objects.

[0037] Data fusion: Data from each sensor is transmitted to the edge computing node. A multi-sensor data fusion algorithm is then employed. Data processing. First, based on the importance and reliability of sensors in different environmental scenarios, the weights w are determined through extensive experiments and data analysis. i For example, the weight of the gas sensor in the kitchen can be high, while the weight of the sound sensor in the bedroom can be adjusted for use during rest. Then, the data from each sensor (S) can be... i The fused data F is obtained by weighted summation. At the same time, environmental fingerprinting technology is introduced to extract features from the fused data to generate an environmental fingerprint. This fingerprint is then compared with a historical fingerprint database to quickly and accurately identify environmental conditions and changes, providing precise environmental information for furniture adjustment.

[0038] Environmental Prediction: Edge computing nodes utilize an environmental prediction model based on Long Short-Term Memory (LSTM) networks to predict environmental changes. Historical environmental data over a period of time, such as temperature, humidity, and air quality, is input into the model. The LSTM model learns time-series characteristics, grasping the trends and patterns of environmental parameter changes, such as learning temperature variations at different times of the day and the seasonal impact on temperature. By predicting future environmental parameters, furniture settings can be adjusted in advance; for example, if humidity is predicted to rise, the dehumidifier function in wardrobes can be activated ahead of time.

[0039] 2. Furniture Status Monitoring Module

[0040] Sensor Installation: Smart sensors based on Microelectromechanical Systems (MEMS) are installed in key parts of the custom-made furniture throughout the house. The MEMS pressure sensor at the cabinet door hinges utilizes the piezoresistive effect; the resistance changes when subjected to pressure, allowing for precise pressure measurement with an accuracy of 0.01N. The MEMS displacement sensor in the drawer slides is based on the principle of capacitance change; the capacitance changes when the drawer slides, enabling precise displacement measurement with an accuracy of 0.01mm. Fiber Bragg grating strain sensors embedded at the cabinet structure connections accurately detect strain with a resolution of 1με as the fiber Bragg grating wavelength changes during cabinet deformation.

[0041] Fault Diagnosis and Lifespan Prediction: Furniture component status data collected by sensors is transmitted in real-time to the data storage and management module. This module uses a deep learning-based fault diagnosis model to analyze the data. A convolutional neural network (CNN) is used to jointly analyze furniture component images (acquired by cameras on the furniture surface) and sensor data. Sensor data is first preprocessed to a format suitable for CNN input, and features such as shape, color, and texture are extracted from the image data. The processed data is then input into the CNN, and through multiple convolutional and pooling layers, it learns the characteristics of normal and faulty furniture states. When the model detects a match between the current data and fault features, it determines a component fault, analyzes the fault features to determine its location and severity. A particle filter-based wear prediction algorithm is used to monitor and predict furniture component wear. This algorithm updates particle states and weights, simulating the wear process of components under different usage conditions. For example, when predicting drawer slide wear, factors such as opening and closing frequency, load weight, and usage time are considered to predict future wear and estimate the remaining lifespan in advance. When the predicted remaining lifespan of a component is lower than a set threshold, the system automatically sends a maintenance reminder to reduce the furniture failure rate.

[0042] 3. User Behavior Analysis Module

[0043] Data Acquisition: User behavior data is collected through multimodal fusion technology, combining depth cameras, millimeter-wave radar, and wearable devices. Depth cameras use structured light or time-of-flight (ToF) technology to acquire the user's three-dimensional pose, accurately identifying actions such as standing and walking; millimeter-wave radar transmits and receives millimeter waves to monitor the user's movement speed, direction, and distance in real time, and is effective even in low light or obstructed conditions; wearable devices such as smart bracelets collect physiological and motion data such as the user's heart rate, blood pressure, steps, and sleep.

[0044] Data Fusion and Analysis: Data collected from various devices is transmitted to the user behavior analysis module. It is then integrated and analyzed using a multimodal data fusion algorithm. First, the data from each device is preprocessed to remove noise and outliers. Then, a feature-level fusion method is used to fuse the posture features of depth cameras, the motion features of millimeter-wave radar, and the physiological and motion features of wearable devices. For example, combining standing posture and movement speed can determine whether a user is stationary or walking. A spatiotemporal convolutional neural network (ST-CNN) is used for deep learning of the fused user behavior data. The spatial convolutional layer of ST-CNN extracts spatial dimension features, while the temporal convolutional layer captures temporal dimension changes. Through extensive data learning, the model uncovers user behavior patterns and habits across different times and spaces. For example, if it detects that a user spends a long time watching movies in the living room on weekend evenings, the system automatically adjusts the sofa angle and lighting brightness. Simultaneously, user social network data and geolocation information are combined to enrich the analysis dimensions. Natural language processing techniques are used to analyze the sentiment and semantics of social network texts, combined with geolocation and time information, to provide users with contextualized furniture adjustment suggestions. For example, when a user is traveling, the system adjusts the furniture to energy-saving mode and periodically checks the furniture status.

[0045] 4. Adaptive Adjustment Control Module

[0046] Physical Structure Adjustment: For adjusting the physical structure of furniture, such as the height of electric table legs and the space of deformable cabinets, a Model Predictive Control (MPC) algorithm is used. First, a dynamic model of the furniture system is established, considering physical parameters and external inputs. The model predicts future system outputs. Based on the prediction results and set goals, such as the user's desired table leg height and a cabinet layout adapted to the environment, the current control inputs, such as motor speed and driver current, are optimized. Taking electric table legs as an example, the MPC algorithm continuously adjusts the motor control signal, allowing the table legs to quickly and stably reach the target height. During adjustment, it adapts to environmental changes and dynamic user needs, such as adjusting the support force of the table legs in time when there is environmental vibration.

[0047] Internal Environment Regulation: Furniture internal environment regulation, such as wardrobe temperature and humidity control and shoe cabinet deodorization and ventilation, employs a composite control algorithm based on fuzzy logic and neural networks. Fuzzy logic handles the uncertainty of environmental parameters and user needs. For example, in wardrobe humidity regulation, humidity data is used as a fuzzy input, and user humidity preferences are categorized into fuzzy sets such as "dry," "moderate," and "humid." According to fuzzy control rules, the dehumidifier's workload is increased when humidity is "humid," and the status quo is maintained when humidity is "moderate." The neural network optimizes control strategies by learning from a large amount of historical data. Historical environmental parameters, user needs, and control outputs are input into the neural network as training data to learn complex relationships and generate precise control commands based on current environmental parameters and user needs. For example, users have different wardrobe temperature and humidity requirements in different seasons, and the neural network automatically adjusts the control strategy after learning historical data. Simultaneously, intelligent material-driven regulation mechanisms are introduced, such as shape memory alloys used for automatic cabinet door opening and closing. Shape memory alloys have a memory effect within a certain temperature range; when triggered by ambient temperature or specific user actions, the alloy deforms, causing the cabinet door to open and close. For example, in summer, when the wardrobe temperature exceeds a threshold, the alloy deforms to open the cabinet door for heat dissipation; when a user approaches the wardrobe, a human infrared sensor triggers the alloy to open the door.

[0048] 5. Data Storage and Management Module

[0049] Data storage: Data is stored using the distributed, decentralized InterPlanetary File System (IPFS). IPFS divides data into small blocks and stores them on different nodes. Each data block has a unique hash value, which allows for quick data location and retrieval. It offers higher reliability than traditional distributed databases, and the failure of some nodes does not affect data retrieval. Homomorphic encryption is used to encrypt and store data, allowing computation and analysis in encrypted form, ensuring data security. For example, statistical analysis of environmental perception data can be performed directly on the encrypted data, with results identical to those in plaintext. Massive amounts of data are compressed using a highly efficient compression algorithm based on trie and Huffman coding. A trie index is first built to index duplicate data, then Huffman coding is used, with short codes for high-frequency data and long codes for low-frequency data, saving storage space.

[0050] Data Management and Mining: Knowledge graph technology is used to connect and integrate different types of data to construct a smart home knowledge graph. First, entity extraction and relationship identification are performed on data such as environmental perception, furniture status, and user behavior. For example, "temperature sensor," "wardrobe," and "user" are entities, and "sensor measures ambient temperature" and "user uses wardrobe" are relationships, which are then organized into a knowledge graph. Through the knowledge graph, furniture adjustment records and user behavior patterns related to environmental status can be quickly found, such as querying user habits regarding bedroom furniture usage at a specific temperature. Simultaneously, big data analytics are used to mine massive amounts of data in IPFS. Cluster analysis reveals user preferences for the internal space layout of furniture, such as the required proportion of different areas inside wardrobes for different user groups. Association rule mining identifies changes in furniture environmental adjustment needs across different seasons, such as increased demand for dehumidification and ventilation in summer and increased demand for warmth in winter, allowing for optimization of system functions and provision of personalized services.

[0051] 6. User Interaction Module

[0052] Brain-Computer Interface Interaction: Provides an interaction method based on a brain-computer interface (BCI). Users wear an EEG acquisition device; electrodes collect EEG signals from the scalp, which are then amplified, filtered, and preprocessed before being transmitted to the BCI analysis module. This module uses machine learning algorithms to classify and identify EEG signals. First, it collects EEG signals when the user performs different actions (such as imagining opening a wardrobe) and extracts features such as frequency and amplitude. Then, it uses algorithms such as Support Vector Machines (SVM) to train and establish a mapping relationship between EEG signals and the user's intentions. When the user imagines an action, the system identifies the EEG signals according to the trained model, interprets the intention, and transmits it to the adaptive adjustment and control module to control the furniture. For example, if the user imagines opening a wardrobe, the system can control the wardrobe to open, providing a convenient and natural interactive experience.

[0053] Gesture and posture interaction: Combining gesture tracking and posture recognition technologies, this system utilizes depth cameras and inertial sensors to achieve natural and intuitive interaction. The depth camera captures user gestures in real time, matching them with preset templates using image recognition algorithms, such as a waving gesture. Inertial sensors (accelerometers and gyroscopes) are mounted on the user's wearable device to monitor changes in body posture in real time. By fusing data from the depth camera and inertial sensors, the system more accurately identifies user posture and intentions. For example, if a user clenches their fist, points at furniture, and moves their arm up and down, the system will recognize the gesture and generate corresponding furniture adjustment commands.

[0054] Multilingual Interaction: Supports multilingual interaction, employing neural machine translation technology for real-time translation of different languages. A large-scale multilingual corpus is first established, and a neural network model (such as the Transformer model) is used to learn and build a translation model. When a user interacts in a non-default language, the system first identifies the language type, then uses the translation model to translate the input text into a language the system can understand for instruction parsing. The processing result is then translated back into the user's language feedback. For example, if a foreign user issues a command in English, the system translates and processes it in real time and provides the English result. Simultaneously, speech synthesis technology is used to provide natural and fluent voice feedback, enhancing the user interaction experience.

[0055] 7. Health Monitoring and Care Module

[0056] Biosensor Integration: Integrating advanced biosensors into furniture. A quantum dot-based blood glucose sensor utilizes the fluorescence properties of quantum dots; changes in fluorescence intensity occur when blood glucose molecules bind to surface antibodies, enabling non-invasive real-time blood glucose monitoring. A surface plasmon resonance (SPR)-based protein sensor uses the plasmon resonance phenomenon on a metal surface to detect proteins secreted by the human body for early disease diagnosis. These biosensors collect health data in real time and transmit it to a health monitoring and care module.

[0057] Health Analysis and Recommendations: The system analyzes and diagnoses health data using a medical knowledge graph and machine learning algorithms. The medical knowledge graph integrates a vast amount of medical knowledge, while machine learning algorithms (such as random forest algorithms) classify and predict biosensor data. For example, by combining the medical knowledge graph with user data such as blood sugar, blood pressure, and heart rate, the system assesses health status. When an abnormality is detected, the system promptly issues a warning, such as a voice reminder that the user's blood pressure is high, and provides corresponding health recommendations based on the medical knowledge graph, such as suggesting appropriate exercise and dietary adjustments. Simultaneously, the system automatically adjusts furniture usage modes based on the user's health condition; for example, it automatically adjusts the height and angle of chairs for users with lumbar spine problems to provide better support and improve the user's health experience.

[0058] 8. Energy Management Module

[0059] Energy Monitoring: Smart meters, water meters, and gas meters are installed to monitor household energy consumption in real time. Smart meters use high-precision current and voltage sensors to accurately measure electricity consumption with an accuracy of 0.01 kWh; smart water meters accurately measure water consumption using ultrasonic flow measurement technology with a resolution of 0.01 L; smart gas meters use infrared sensing technology to monitor gas flow in real time with an accuracy of 0.01 m³ / s. 3 This energy data is transmitted to the energy management module in real time.

[0060] Energy Management Strategy: A reinforcement learning-based energy management strategy is adopted. First, an energy consumption model is established, such as an energy demand forecasting model based on grey prediction. By analyzing historical energy consumption data and considering factors such as season, time, and user behavior, energy demand is predicted for different time periods. For example, the model might learn that air conditioning use is frequent in summer evenings, increasing energy demand. Based on the forecast results, energy resources are allocated rationally, such as prioritizing power supply to essential electrical appliances during peak hours and intelligently controlling non-essential equipment such as landscape lighting. Energy-saving control strategies are implemented, such as automatically adjusting indoor lighting brightness based on ambient light levels and automatically turning off unnecessary electrical appliances and furniture when no one is in the room. An energy saving rate index is introduced. Assess energy-saving effects, continuously optimize energy-saving strategies, improve household energy efficiency, and reduce energy consumption costs. Simultaneously, utilize renewable energy management technologies, such as intelligent control of solar panels and wind turbines, to rationally integrate renewable energy into the household energy supply system based on weather and sunlight conditions, further reducing dependence on traditional energy sources. For example, during periods of ample sunshine, prioritize using solar panels to generate electricity for the household, storing excess power in batteries for backup.

[0061] III. Characterization of Beneficial Effects Data

[0062]

[0063] Based on 50 sets of experimental data, it was found that this system has advantages over traditional systems in terms of detection accuracy, early warning rate, identification accuracy, user satisfaction, and energy saving rate.

[0064] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A whole-house customized furniture adaptive adjustment system under intelligent home environment perception, characterized in that, include: Environmental perception module: It adopts a nanoscale environmental sensor array and uses light sensors and sound sensors with imaging technology to detect the distance and material information of surrounding objects; Through sensor data fusion algorithms Where F represents the fused environmental perception data, w i S represents the weight of the i-th sensor. i The data collected by the i-th sensor enables the perception of environmental information; at the same time, environmental fingerprint recognition technology is introduced to generate an environmental fingerprint for each room or area through comprehensive analysis of environmental characteristics, which is used to identify the environmental status and changes. Furniture Condition Monitoring Module: Sensors based on microelectromechanical systems (MEMS) are installed in the whole-house customized furniture; fiber optic strain sensors are embedded at the cabinet structure connection points. By monitoring the wavelength change of the fiber optic grating, the strain of the cabinet structure is detected. Anomaly detection algorithms in machine learning are used to analyze the sensor data in real time. By monitoring the wear degree of furniture parts, the remaining service life of the parts is predicted using a wear prediction model. User behavior analysis module: Utilizes multimodal fusion behavior analysis technology, combined with wearable device data; wearable devices collect users' physiological and motion data, and through multimodal data fusion algorithms, integrate and analyze data from different sources to construct user behavior models; Adaptive adjustment and control module: Based on environmental perception, furniture status monitoring and user behavior analysis results, the model predictive control (MPC) algorithm is used to achieve adaptive adjustment of the whole house customized furniture; for the adjustment of the internal environment of the furniture, a composite control algorithm based on fuzzy logic and neural network is used; at the same time, a material-driven furniture adjustment mechanism is introduced. When the ambient temperature or user operation is triggered, the shape memory alloy deforms, which drives the cabinet door to open or close automatically.

2. The whole-house customized furniture adaptive adjustment system under intelligent home environment perception as described in claim 1, characterized in that, Also includes: Data storage and management module: It adopts the distributed and decentralized InterPlanetary File System (IPFS) for data storage, uses homomorphic encryption technology to enable data calculation and analysis in encrypted state, and uses knowledge graph technology to associate and integrate different types of data to build a home knowledge graph, which facilitates data query, analysis and mining. User interaction module: Provides an interaction method based on brain-computer interface (BCI). Users wear EEG acquisition devices, and the system recognizes the user's EEG signals, interprets the user's intentions, and realizes control of furniture; Combines gesture tracking and posture recognition technology, and uses depth cameras and inertial sensors to realize interaction; Health Monitoring and Care Module: This module integrates multiple advanced biosensors into the furniture, including a protein sensor based on surface plasmon resonance (SPR) technology. It analyzes and diagnoses health data using medical knowledge graphs and machine learning algorithms. When an abnormality is detected in the user's health, it issues an alert and provides health advice. It also incorporates health risk assessment indicators. Where R is the health risk value, w i h is the weight of the i-th health indicator. i Let represent the degree of abnormality of the i-th health indicator.

3. The whole-house customized furniture adaptive adjustment system under intelligent home environment perception according to claim 1, characterized in that, Also includes: Energy Management Module: Installs smart meters for electricity, water, and gas to monitor household energy consumption in real time; predicts energy demand at different times through energy consumption models and allocates energy resources rationally; employs energy-saving control strategies to automatically turn off electrical appliances and furniture when no one is home; and introduces energy-saving rate indicators. Where E old E represents the energy consumption before optimization. new To optimize energy consumption, we continuously optimize energy-saving strategies to reduce energy consumption costs.

4. The whole-house customized furniture adaptive adjustment system under intelligent home environment perception according to claim 1, characterized in that, In the environmental perception module, edge computing technology is used to complete data processing tasks on sensor nodes or edge devices. At the same time, an environmental prediction model is introduced to predict environmental changes in the future by using historical environmental data and current environmental trends. An environmental prediction model based on Long Short-Term Memory (LSTM) network is used to predict changes in future environmental parameters by learning the time series features in historical environmental data.

5. The whole-house customized furniture adaptive adjustment system under intelligent home environment perception according to claim 1, characterized in that, In the furniture condition monitoring module, IoT technology is used to realize the real-time uploading and remote monitoring of furniture component status data; a fault diagnosis algorithm is used to diagnose the faults of furniture components; and a convolutional neural network (CNN) is used to jointly analyze the image data and sensor data of furniture components to diagnose the fault type and location.

6. The whole-house customized furniture adaptive adjustment system under intelligent home environment perception according to claim 2, characterized in that, In the user behavior analysis module, the dimensions of user behavior analysis are expanded by combining users' social network data and geolocation information; by analyzing users' lifestyle habits, interests and hobbies shared on social networks and users' geographical location and time information, furniture adjustment services are provided to users.

7. The whole-house customized furniture adaptive adjustment system under intelligent home environment perception according to claim 1, characterized in that, In the adaptive adjustment control module, a reinforcement learning algorithm is introduced to continuously optimize the control strategy based on user feedback on furniture adjustment; the Deep Q-Network (DQN) algorithm in deep reinforcement learning is adopted to enable the system to learn the adjustment strategy during the interaction with the user.

8. The whole-house customized furniture adaptive adjustment system under intelligent home environment perception according to claim 2, characterized in that, In the data storage and management module, quantum key technology is used to ensure data security. Environmental perception data, furniture status data, and user behavior data are stored on nodes. Each data block contains a timestamp and data hash value information, and data traceability formulas are used to ensure data security. Where H i It is the hash value of the i-th data block, which facilitates the tracing of the data's source and history; at the same time, big data analysis technology is used to mine the data and discover user needs and furniture usage patterns.

9. The whole-house customized furniture adaptive adjustment system under intelligent home environment perception according to claim 2, characterized in that, The user interaction module supports voice interaction to meet user needs; at the same time, it uses virtual reality and augmented reality technologies to provide users with a real-time furniture adjustment experience.

10. The whole-house customized furniture adaptive adjustment system under intelligent home environment perception according to claim 1, characterized in that, The entire system is equipped with self-learning and evolution capabilities; by collecting user feedback, environmental change data, and furniture usage, it uses deep learning algorithms to update and optimize the system's model and parameters. At the same time, it supports online software upgrades and function expansions, providing users with continuous high-quality services.

Images