Intelligent control system and method based on AI intelligent home model

Abstract

The invention discloses an intelligent control system and method based on an AI intelligent home model, and the system comprises a data acquisition module which is used for collecting home related data in real time based on a multi-dimensional sensing system, and the data comprises indoor and outdoor environment parameters and habitant physiological feature data; the data analysis module is used for performing resident comfort demand prediction analysis on the home related data based on the AI smart home model to obtain the comfort demand of the resident; and the environment regulation and control module is used for carrying out working parameter setting and working mode adjustment on the household electrical equipment based on the comfort degree requirement, so as to realize personalized environment adjustment for different residents and scenes. The AI smart home model accurately predicts the comfort demands of the residents, and the system can provide customized environment adjustment schemes for individual differences of different residents, meets the differentiated demands of special groups such as old people, children, pregnant women and the like, and improves the living experience.

Description

Technical Field

[0001] The present invention relates to the field of smart home technologies, and in particular to an intelligent control system and method based on an AI smart home model. Background Art

[0002] With the rapid development of the Internet of Things (IoT) and artificial intelligence (AI), smart home systems have evolved from simple remote control devices to more intelligent living environment management systems. Traditional smart home systems primarily operate based on preset rules and simple trigger mechanisms, such as timed on / off functions and remote control, bringing convenience to people's daily lives.

[0003] However, existing smart home control systems suffer from the following technical shortcomings: Most existing systems only collect limited environmental parameters (such as temperature and humidity) and lack comprehensive sensing capabilities for occupants' physiological characteristics, making them unable to fully understand users' true needs. Traditional smart home systems often rely on fixed rules or simple algorithms for control, lacking deep learning capabilities and struggling to adapt to the personalized comfort needs of different occupants. Existing systems generally lack adaptive adjustment capabilities for different users and scenarios, making them unable to intelligently adjust to occupants' actual needs. Most systems lack effective monitoring and evaluation mechanisms after executing control commands, preventing them from promptly detecting environmental parameter deviations or device operating anomalies, and lack self-diagnosis and optimization capabilities. Data sharing and interoperability between subsystems are insufficient, making it difficult to achieve unified, intelligent control of the entire home. These technical shortcomings result in existing smart home systems often failing to provide a satisfactory living experience in practice. Users often need to frequently manually adjust various device parameters, making it difficult for the systems to truly achieve "intelligent" living environment control. This is particularly true for special groups such as the elderly and infants, as existing systems are unable to accurately adjust the environment to their specific physiological needs, resulting in significant user experience gaps.

[0004] Therefore, there is an urgent need for an intelligent control system and method based on the AI smart home model. Summary of the Invention

[0005] The present invention provides an intelligent control system and method based on an AI smart home model to solve the above-mentioned problems existing in the prior art.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] Intelligent control system based on AI smart home model, including:

[0008] A data acquisition module is used to collect home-related data in real time based on a multi-dimensional sensing system, including indoor and outdoor environmental parameters and resident physiological characteristics data;

[0009] The data analysis module is used to predict and analyze the comfort needs of residents based on the AI smart home model and obtain the comfort needs of residents;

[0010] The environmental control module is used to set the operating parameters and adjust the working mode of household electrical appliances based on comfort requirements, so as to achieve personalized environmental adjustment for different residents and scenarios.

[0011] Among them, also include:

[0012] The operation monitoring module is used to monitor and evaluate the environmental control results in real time, and to perform intelligent diagnosis and adaptive optimization adjustments when environmental parameters deviate from expectations or equipment operation is abnormal.

[0013] The data acquisition module includes:

[0014] AI configuration submodule, used for:

[0015] Using AI smart home models to analyze the layout of residential spaces and the needs of residents, it automatically generates configuration plans for multi-dimensional environmental monitoring sensors and physiological characteristics collection devices for different functional areas;

[0016] Based on historical data and user feedback, AI learning algorithms are used to predict sensing needs in different areas, continuously optimize the type, quantity, and location of sensors, and improve the accuracy and efficiency of data collection.

[0017] Sensor configuration submodule, used for:

[0018] According to the configuration plan generated by the AI configuration submodule, multi-dimensional environmental monitoring sensors and physiological characteristics collection equipment are configured for different functional areas;

[0019] Based on the real-time analysis of the functional attributes of different spaces by the AI smart home model, the acquisition frequency of the sensor configuration sub-module is dynamically adjusted in combination with the prediction model to optimize energy consumption and data quality, and achieve focused monitoring of key areas.

[0020] The data analysis module includes:

[0021] Model building submodule for:

[0022] Use AI algorithms to automatically extract key environmental parameter types from indoor and outdoor environmental parameters, and automatically identify behavioral parameter types from occupants' physiological characteristic data;

[0023] Through AI correlation analysis technology, a dynamic correlation mapping between environmental parameter types and behavioral parameter types is established to form a data group to be processed;

[0024] Based on the AI priority evaluation model, each data group to be processed is prioritized according to the importance and real-time changes of environmental parameter types to form home data to be analyzed;

[0025] Obtain smart home device specification data and use AI feature extraction technology to extract device performance standard data from it;

[0026] Perform AI-driven data preprocessing on device performance standard data to construct device feature vectors, while also obtaining device interconnection correlation data from smart home device specification data;

[0027] Based on the AI clustering algorithm, cluster analysis is performed on device feature vectors according to device interconnection data to generate a data group to be trained;

[0028] Apply the training data set to the initial model, and use AI learning technology to perform model training and parameter optimization to obtain an AI smart home model;

[0029] Using AI feature engineering technology to extract personalized comfort feature data from residents' historical behavioral preference data;

[0030] Perform AI data enhancement and adaptive noise reduction on personalized comfort feature data to construct multi-dimensional features;

[0031] Based on AI behavioral pattern analysis technology, it analyzes the patterns of multi-dimensional features in different life scenarios and generates deep scene correlation data;

[0032] Input the deep scene association data into the LLM large language model, perform collaborative calculations with the AI smart home model, and build a personalized comfort prediction model through deep learning of neural networks;

[0033] The scenario analysis submodule is used to:

[0034] Based on the AI smart home model, the user can identify the identity and physiological status of the residents, distinguish between family members, visitors, and changes in physiological parameters of different members;

[0035] Utilize AI scene recognition technology to analyze the current living space usage scenarios, including sleeping, working and studying, meeting and leisure, and cooking activities;

[0036] Based on the personalized comfort prediction model, combined with AI's in-depth analysis of occupants' identity characteristics, physiological status data and current space usage scenarios, real-time comfort demand prediction results are generated.

[0037] Among them, the environmental control module includes:

[0038] The device management submodule is used to:

[0039] Register and manage household electrical appliances, including negative oxygen ion generators, ventilation systems, humidifiers, M2 filtration systems, smart air conditioners, floor heating equipment, and lighting control devices;

[0040] Obtain the working status, energy consumption data and operating parameters of each device in real time;

[0041] Control strategy generation submodule, used to:

[0042] Generate environmental control targets based on predicted comfort requirements and current environmental parameters;

[0043] Formulate multi-device collaborative working strategies based on environmental control objectives and determine the working mode and parameter settings of each device;

[0044] The device control submodule is used to issue control instructions to each home appliance according to the collaborative work strategy and monitor the execution status of the instructions.

[0045] The operation monitoring module includes:

[0046] The environmental monitoring submodule is used to continuously monitor changes in indoor parameters after environmental control and evaluate whether the environmental control effect meets the expected goals;

[0047] Effect evaluation submodule, used to:

[0048] Collect residents' subjective feedback on environmental comfort and changes in physiological indicators;

[0049] Establish mapping relationships between environmental parameters, equipment operating status, and occupant comfort;

[0050] Generate an environmental control effect evaluation report based on the comfort assessment results and equipment operating status;

[0051] The self-optimization submodule is used to analyze the causes and automatically adjust the control strategy when environmental parameters deviate from expectations or equipment operates abnormally.

[0052] Among them, the self-optimization submodule includes:

[0053] Learning Optimization Unit for:

[0054] Generate optimized environmental control strategies through the continuous learning function of the DeepSeeK-R model;

[0055] Generate dynamic intelligent control parameters based on seasonal changes, changes in occupant health status, and equipment performance degradation;

[0056] Based on the behavior patterns and life patterns of residents, predict future changes in environmental needs, generate adjustment plans and optimize controls in advance;

[0057] Fault diagnosis subunit, used for:

[0058] When the equipment is operating abnormally or environmental parameters fail to meet expected targets, the cause is analyzed and intelligent fault diagnosis results are generated;

[0059] Identify the fault type based on the diagnosis results and generate a treatment plan from the maintenance strategy library;

[0060] For minor faults that can be automatically recovered, self-repair instructions are generated and executed;

[0061] For serious faults that require manual intervention, detailed maintenance recommendations are generated and maintenance personnel are contacted. At the same time, diagnostic data is fed back to the self-optimization submodule to optimize the control strategy.

[0062] Among them, the intelligent control method based on the AI smart home model includes:

[0063] S101: Real-time collection of home-related data based on a multi-dimensional sensing system, including indoor and outdoor environmental parameters and resident physiological characteristics data;

[0064] S102: Based on the AI smart home model, perform resident comfort demand forecasting and analysis on home-related data to obtain resident comfort needs;

[0065] S103: Based on comfort requirements, the operating parameters of household electrical appliances are set and the operating modes are adjusted to achieve personalized environmental adjustment for different residents and scenarios.

[0066] Wherein, after step S103, the following steps are included:

[0067] S104: Monitor and evaluate environmental control results in real time, and perform intelligent diagnosis and adaptive optimization adjustments when environmental parameters deviate from expectations or equipment operates abnormally.

[0068] Wherein, step S101 includes:

[0069] According to the layout of the living space and the needs of the residents, multi-dimensional environmental monitoring sensors and physiological characteristics collection equipment are configured in different functional areas;

[0070] The acquisition frequency of the sensor configuration submodule is dynamically adjusted based on different spatial functional attributes to achieve focused monitoring of key areas.

[0071] Compared with the prior art, the present invention has the following advantages:

[0072] The intelligent control system based on the AI smart home model includes: a data acquisition module for collecting home-related data in real time based on a multi-dimensional sensing system, including indoor and outdoor environmental parameters and resident physiological characteristics; a data analysis module for predicting and analyzing resident comfort needs based on the AI smart home model to determine resident comfort needs; and an environmental control module for setting operating parameters and adjusting operating modes for home appliances based on comfort needs, achieving personalized environmental adjustments for different residents and scenarios. By accurately predicting resident comfort needs through the AI smart home model, the system can provide customized environmental adjustment solutions based on the individual differences of different residents, meeting the differentiated needs of special groups such as the elderly, children, and pregnant women, and improving the living experience.

[0073] Other features and advantages of the present invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the present invention.

[0074] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0076] Figure 1 This is a structural diagram of an intelligent control system based on an AI smart home model in an embodiment of the present invention;

[0077] Figure 2 This is a structural diagram of a data analysis module in an embodiment of the present invention;

[0078] Figure 3 This is a flow chart of an intelligent control method based on an AI smart home model in an embodiment of the present invention. DETAILED DESCRIPTION

[0079] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0080] The present invention is implemented as follows Figure 1 The figure provides an intelligent control system based on the AI smart home model, including:

[0081] A data acquisition module is used to collect home-related data in real time based on a multi-dimensional sensing system, including indoor and outdoor environmental parameters and resident physiological characteristics data;

[0082] The data analysis module is used to predict and analyze the comfort needs of residents based on the AI smart home model and obtain the comfort needs of residents;

[0083] The environmental control module is used to set the operating parameters and adjust the working mode of household electrical appliances based on comfort requirements, so as to achieve personalized environmental adjustment for different residents and scenarios.

[0084] The working principle of the above technical solution is as follows: The data acquisition module collects home-related data in real time through a multi-dimensional sensor system distributed throughout the home environment. These sensors include environmental parameter monitoring devices such as temperature and humidity sensors, light sensors, air quality sensors, and noise sensors, as well as wearable devices or non-contact sensors to collect physiological characteristic data of residents. Indoor and outdoor environmental parameters include indoor temperature, humidity, carbon dioxide concentration, PM2.5 content, formaldehyde concentration, light intensity, noise level, and outdoor temperature, humidity, air quality index, and weather conditions. Physiological characteristic data includes body temperature, heart rate, respiratory rate, activity level, and sleep quality.

[0085] The data analysis module receives the collected data and feeds it into a pre-trained AI smart home model for processing. Based on machine learning algorithms, this model analyzes a combination of historical and current data to identify occupants' behavioral patterns, living habits, and comfort preferences, thereby predicting their comfort needs at different times and in different scenarios.

[0086] Based on the comfort needs predictions generated by the data analysis module, the environmental control module intelligently controls various home appliances, such as the air conditioner, humidifier, curtains, lighting, and fresh air system, adjusting their operating parameters and modes to provide a home environment that best suits the resident's current needs. For example, based on the comfort needs analysis for resident A, the system automatically performs the following controls: the air conditioner is activated and set to 23°C (slightly lower than the preferred 24°C to help lower body temperature); the humidifier is adjusted to maintain 50% indoor humidity; the curtains are half-drawn to reduce natural light intensity while retaining some; the living room lighting is adjusted to a soft warm yellow setting; and the fresh air system is activated to provide fresh air. The system also prepares the resident's favorite TV channels or music playlist, but remains in standby mode awaiting further instructions.

[0087] When there are multiple residents in a home, the system analyzes each resident's preferences to find the optimal balance, or adjusts the environmental parameters of each area based on the activities of the residents. For example, the living room can be kept warmer while an elderly person watches TV, while the study can be kept cooler while a young person works.

[0088] The beneficial effects of this technical solution include: the system can provide a customized home environment based on the individual differences and real-time status of each resident, significantly improving living comfort. Through precise prediction and control, unnecessary energy consumption is avoided, such as only turning on air conditioning, lighting, and other equipment in areas where they are needed. This can save 20-30% of energy compared to traditional homes. The system can adjust environmental parameters based on the resident's physiological state. For example, if it detects that a resident is sleeping poorly, it will automatically adjust the bedroom temperature, humidity, and lighting to promote a healthier lifestyle. Residents do not need to manually adjust various devices; the system proactively senses their needs and makes adjustments, reducing manual operation and providing a "senseless but effective" intelligent service experience. Through continuous learning, the system continuously improves its understanding of resident preferences, making environmental control more precise and adaptable to changing preferences.

[0089] In another embodiment, further comprising:

[0090] The operation monitoring module is used to monitor and evaluate the environmental control results in real time, and to perform intelligent diagnosis and adaptive optimization adjustments when environmental parameters deviate from expectations or equipment operation is abnormal.

[0091] The working principle of this technical solution is to deploy a multi-source sensor network to comprehensively monitor the indoor environment. The system installs sensors for environmental parameters such as temperature, humidity, CO2 concentration, PM2.5, light intensity, and noise in various areas of the space. Simultaneously, power monitoring devices record real-time energy consumption data and operating status of equipment such as air conditioners, fresh air, and lighting. Sensor data is collected every five seconds and transmitted to the central processing unit via a wireless network, ensuring real-time monitoring of environmental conditions. The system integrates environmental parameters and energy consumption data with user-defined expectations to calculate an environmental comfort index. For example, the comfortable range for a summer working environment is a temperature of 24-26°C, humidity of 40-60%, and a CO2 concentration below 1000 ppm. The system compares real-time monitored values with these target values to generate a deviation rate indicator, which is used to assess the accuracy of environmental control. If the monitored temperature exceeds the set point by more than 2°C for 15 minutes, the system initiates an abnormality diagnosis process to determine whether it is due to equipment failure, external environmental interference, or improper settings. By comparing historical operating data with equipment response characteristic curves, the system can determine specific issues such as insufficient air conditioning cooling capacity or a blockage in the air supply system. Based on the abnormal diagnosis results, the system automatically executes corresponding adjustment measures; for example, when it detects that the temperature in a certain area is too high and the air conditioner is operating normally, the system will first increase the air volume by 10% and slightly reduce the outlet temperature by 1°C, rather than simply lowering the temperature setting value. This fine-tuning can not only quickly improve the environmental comfort level, but also avoid energy waste. The system collects user subjective feedback through mobile applications, and at the same time combines the user's physiological indicators (such as heart rate, skin temperature, etc.) collected by wearable devices to establish a personalized comfort model; as the data accumulates, the system uses machine learning algorithms to continuously optimize the mapping relationship between environmental parameters and comfort, so that the control is more in line with the needs of specific user groups. The system generates an environmental quality report every day, showing key indicators such as the fluctuation trend of each parameter, the compliance rate, and energy efficiency; at the same time, based on historical data analysis, it predicts possible environmental changes and equipment loads in the next 24 hours, adjusts the operation strategy in advance, and realizes forward-looking control.

[0092] The beneficial effects of this technical solution include: through precise monitoring and adaptive adjustment, environmental parameters are consistently maintained within the optimal range, improving user comfort and satisfaction. Intelligent diagnosis avoids over-adjustment and ineffective operation, saving energy and reducing operating costs compared to traditional control systems. Early detection of abnormalities and preventive maintenance reduce equipment failure rates and avoid losses caused by overload operation.

[0093] In another embodiment, the data acquisition module includes:

[0094] AI configuration submodule, used for:

[0095] Using AI smart home models to analyze the layout of residential spaces and the needs of residents, it automatically generates configuration plans for multi-dimensional environmental monitoring sensors and physiological characteristics collection devices for different functional areas;

[0096] Based on historical data and user feedback, AI learning algorithms are used to predict sensing needs in different areas, continuously optimize the type, quantity, and location of sensors, and improve the accuracy and efficiency of data collection.

[0097] Sensor configuration submodule, used for:

[0098] According to the configuration plan generated by the AI configuration submodule, multi-dimensional environmental monitoring sensors and physiological characteristics collection equipment are configured for different functional areas;

[0099] Based on the real-time analysis of the functional attributes of different spaces by the AI smart home model, the acquisition frequency of the sensor configuration sub-module is dynamically adjusted in combination with the prediction model to optimize energy consumption and data quality, and achieve focused monitoring of key areas.

[0100] The working principle of this technical solution is as follows: The AI configuration submodule comprehensively analyzes the living space using a deep learning neural network model, identifying the functional attributes of different areas and their interrelationships. The system then uses three-dimensional spatial mapping technology to create a digital twin of the living environment, precisely marking the boundaries of functional areas, flow paths between areas, and the location of furniture and equipment. Based on this, the AI configuration submodule combines a resident profile database (including multi-dimensional information such as age, occupation, health status, and lifestyle habits) with an association rule mining algorithm to identify the frequency and usage patterns of residents in different areas, generating a preliminary map of sensor configuration requirements. The AI learning algorithm plays a central role in the configuration optimization process, primarily implemented through an iterative multi-objective optimization approach. A multi-layer decision tree is constructed, taking living space characteristics (such as area, orientation, and lighting conditions) and resident needs (such as safety and health monitoring needs and comfort preferences) as input variables. The accuracy, comprehensiveness, energy efficiency, and cost-effectiveness of the sensor configuration are the optimization objectives. In each iteration, the system generates multiple candidate configurations using a genetic algorithm. Monte Carlo simulation techniques are then used to predict the performance of each option in different scenarios, evaluate the options, and retain the best-performing configuration for the next round of iterative optimization. As the system's operating time increases, the AI learning algorithm automatically collects actual sensor operating data and user feedback, identifying the spatiotemporal patterns of data collection through time series analysis. The algorithm applies seasonal decomposition techniques to isolate hourly, daily, and seasonal periodic patterns, while utilizing anomaly detection algorithms to identify events that deviate from normal patterns. When the system identifies redundancies in the sensor configuration (e.g., multiple sensors collecting highly correlated data) or blind spots (e.g., missing data in specific areas or time periods), it automatically triggers a reconfiguration process to adjust the type, quantity, or location of sensors. This adaptive optimization mechanism, based on a Bayesian network inference model, selects the optimal sensor layout under current conditions by calculating the posterior probabilities of different configuration options.

[0101] The sensor configuration submodule implements specific hardware deployment based on the configuration scheme generated by the AI configuration submodule. This submodule adopts a hierarchical configuration strategy to identify functional areas such as living rooms, bedrooms, kitchens, and bathrooms based on the layout of the living space, and then differentiates the configuration of sensors based on the specific needs of the residents (such as the elderly, children, and patients with chronic diseases). For example, for bedrooms frequently used by the elderly, the system will deploy a microwave detection system for contactless monitoring of breathing and heart rate, and infrared detection equipment to monitor activity status and fall risks; while for children's rooms, the system will focus on indoor air quality monitors and noise level sensors to ensure a healthy growth environment. The sensor configuration submodule uses a dynamic sampling algorithm to adjust the data collection strategy. When a resident is detected entering a specific area, the sensor equipment in that area automatically increases the collection frequency to capture more detailed changes in environmental and physiological data; when there is no one in the area, the collection frequency is reduced to save energy and computing resources. For example, in the kitchen area, when human activity is detected, the collection frequency of the gas sensor and temperature sensor will be increased from once per minute to once every 10 seconds, monitoring any abnormal conditions that may occur during the cooking process in real time; and during the night sleep period, the bedroom environmental sensors will operate in low-interference mode, maintaining a low collection frequency but retaining the ability to respond quickly to abnormal events.

[0102] The sensor configuration submodule uses a diverse network of sensor devices to comprehensively monitor the living environment and occupant health. Temperature and humidity sensors provide real-time monitoring of indoor climate comfort; air quality monitors continuously analyze indoor carbon dioxide concentrations, PM2.5 levels, and formaldehyde levels; and physiological parameter monitoring devices collect occupants' body temperature, heart rate, and respiratory rate data contactlessly. All collected data is assigned an identity tag (identifying the occupant) and a spatial tag (indicating the specific location of data collection) to construct a structured, multidimensional database. The sensor configuration submodule provides personalized monitoring strategies tailored to the specific needs of different family members. For elderly residents requiring health monitoring, the system deploys a higher density of physiological parameter monitoring devices in their activity areas and increases the frequency of monitoring. For residents with allergies, the system strengthens monitoring of indoor allergens (such as dust mites and pollen). For infants and young children, the system focuses on temperature, humidity, and air quality changes in the sleeping environment.

[0103] Through continuous interaction with the AI configuration submodule, the sensor configuration submodule continuously receives optimization suggestions and implements adjustments. The deep learning algorithm in the system analyzes environmental change trends through typical convolutional neural networks, predicts resident behavior patterns through recurrent neural network models, and comprehensively generates dynamic optimization solutions for sensor configuration. For example, when the algorithm detects that the contribution of specific sensor data in a certain area to the prediction of environmental status is low, it will automatically reduce the weight of the sensor or recommend replacing it with a more suitable type; when it identifies seasonal changes in resident activity patterns, the system will adjust the spatial distribution of sensors accordingly to ensure that key areas are fully monitored. This data-driven adaptive configuration mechanism ensures that the entire sensor network always maintains its optimal state, maximizing resource utilization efficiency while meeting monitoring needs.

[0104] The beneficial effects of this technical solution are as follows: by accurately monitoring environmental parameters in different functional areas, comfort indicators such as temperature, humidity, and air quality in the living space are continuously optimized, significantly improving the quality of life of residents. The system's continuous monitoring of residents' physiological status provides a health warning mechanism, enabling timely detection and appropriate measures for abnormal conditions (such as abnormal heart rate and falls), thereby improving residential safety. By dynamically adjusting the sensor acquisition frequency, the system achieves a reasonable allocation of energy and computing resources, avoiding resource waste and extending equipment life.

[0105] In another embodiment, Figure 2 As shown, the data analysis module includes:

[0106] Model building submodule for:

[0107] Use AI algorithms to automatically extract key environmental parameter types from indoor and outdoor environmental parameters, and automatically identify behavioral parameter types from occupants' physiological characteristic data;

[0108] Through AI correlation analysis technology, a dynamic correlation mapping between environmental parameter types and behavioral parameter types is established to form a data group to be processed;

[0109] Based on the AI priority evaluation model, each data group to be processed is prioritized according to the importance and real-time changes of environmental parameter types to form home data to be analyzed;

[0110] Obtain smart home device specification data and use AI feature extraction technology to extract device performance standard data from it;

[0111] Perform AI-driven data preprocessing on device performance standard data to construct device feature vectors, while also obtaining device interconnection correlation data from smart home device specification data;

[0112] Based on the AI clustering algorithm, cluster analysis is performed on device feature vectors according to device interconnection data to generate a data group to be trained;

[0113] Apply the training data set to the initial model, and use AI learning technology to perform model training and parameter optimization to obtain an AI smart home model;

[0114] Using AI feature engineering technology to extract personalized comfort feature data from residents' historical behavioral preference data;

[0115] Perform AI data enhancement and adaptive noise reduction on personalized comfort feature data to construct multi-dimensional features;

[0116] Based on AI behavioral pattern analysis technology, it analyzes the patterns of multi-dimensional features in different life scenarios and generates deep scene correlation data;

[0117] Input the deep scene association data into the LLM large language model, perform collaborative calculations with the AI smart home model, and build a personalized comfort prediction model through deep learning of neural networks;

[0118] The scenario analysis submodule is used to:

[0119] Based on the AI smart home model, the user can identify the identity and physiological status of the residents, distinguish between family members, visitors, and changes in physiological parameters of different members;

[0120] Utilize AI scene recognition technology to analyze the current living space usage scenarios, including sleeping, working and studying, meeting and leisure, and cooking activities;

[0121] Based on the personalized comfort prediction model, combined with AI's in-depth analysis of occupants' identity characteristics, physiological status data and current space usage scenarios, real-time comfort demand prediction results are generated.

[0122] The working principle of this technical solution is as follows: the model-building submodule obtains indoor and outdoor environmental parameters (such as temperature, humidity, light intensity, air quality, etc.) and identifies their parameter types. At the same time, it extracts behavioral parameter types from the occupant's physiological characteristics data (such as heart rate, body temperature, activity status, etc.). For example, if the system monitors that the indoor air is slightly damp and the light is dim, and observes that the occupant's activity level is reduced and the body is slightly tired, the system will classify these data as environmental parameters related to "humidity comfort" and behavioral parameters related to "fatigue".

[0123] The system uses AI correlation analysis technology to establish a dynamic correlation mapping between environmental parameter types and behavioral parameter types to form a data group to be processed. The "correlation analysis technology" here refers to the use of artificial intelligence algorithms (such as deep learning or correlation analysis) to find the causal or correlation relationship between environmental factors and human reactions. Specifically, the system will analyze a large amount of data to identify patterns such as "high indoor humidity may cause occupants to feel uncomfortable" or "lack of light may affect occupants' attention." For example, the system will map humid air and the occupants' slight discomfort into a "wet comfort" data group; and map dim light and the occupants' fatigue into a "visual comfort" data group. These mapping relationships are not fixed, but are dynamically adjusted as data accumulates.

[0124] According to the importance of the environmental parameter type to human comfort, the system uses the AI priority evaluation model to prioritize each data group to be processed to form the home data to be analyzed. The "priority evaluation model" here is an artificial intelligence-based decision-making tool that determines which environmental factors have a greater impact on comfort by analyzing historical data and real-time changes. For example, in humid seasons, the system may consider the "humidity comfort" data group to be more important than the "light comfort" data group, and therefore prioritizes humidity-related data; on nights with insufficient light, visual comfort may be ranked first. The model construction process includes: collecting a large number of environmental and comfort data samples, training the model through machine learning algorithms (such as decision trees or neural networks), and letting it learn to automatically assign priorities according to different scenarios. Finally, the system stores the environmental parameters and behavioral parameters in the corresponding data groups to be processed according to their importance, generating home data to be analyzed.

[0125] The system obtains specification data of smart home devices, including detailed parameters of air conditioners, air purifiers, smart lamps and other equipment, and uses AI feature extraction technology to extract equipment performance standard data from it. The "feature extraction technology" here refers to the process of picking out key information from complex data through artificial intelligence algorithms, such as extracting "cooling speed" from the parameters of air conditioners and extracting "purification efficiency" from air purifiers. These data are pre-processed (such as removing invalid data, unifying the format, etc.) and converted into device feature vectors, which are concise forms of representing device performance in numbers. At the same time, the system obtains device interconnection related data from the device specification data, such as the collaborative operation mode of air conditioners and air purifiers, or the linkage relationship between smart lamps and curtains.

[0126] Based on the AI clustering algorithm, the system performs cluster analysis on the device feature vectors based on the device interconnection data to generate a data set to be trained. The clustering algorithm is a method of grouping similar things, and is used here to find a combination of devices with complementary functions or good synergistic effects. For example, the system groups air conditioners and air purifiers into a group of "air conditioning clusters" and smart lamps and curtains into a group of "light environment adjustment clusters." This process is automatically completed by analyzing the correlation between devices (such as overlapping operating times or mutual functional assistance) to ensure a more efficient device combination.

[0127] The system inputs the generated training data set into the initial model and uses AI learning technology to perform model training and parameter optimization, ultimately obtaining an AI smart home model. "Learning technology" here primarily refers to supervised learning, but can also refer to unsupervised training. The initial model is based on a deep neural network architecture. By repeatedly adjusting parameters (such as through a backpropagation algorithm), the model can accurately predict the comfort needs of residents. For example, the model will learn that "when the air purifier is turned on, the resident's breathing comfort improves."

[0128] The system uses AI feature engineering to extract personalized comfort characteristics from historical resident behavior preference data. "Feature engineering" here refers to using algorithms to uncover key information reflecting resident habits, such as a preference for a quiet environment or soft lighting. This data undergoes AI data enhancement and adaptive noise reduction, filling in missing information and removing unusual interference, to construct a multidimensional feature map encompassing environmental preferences, physiological responses, and other aspects.

[0129] Based on AI behavioral pattern analysis technology, the system analyzes the patterns of multi-dimensional features in different life scenarios and generates deep scene-related data. For example, the system finds that residents prefer bright light when working and warm colors when resting, thereby establishing a mapping relationship between scenes and comfort needs. These related data are input into the LLM large language model (LLM is a technology based on artificial intelligence), and are calculated in collaboration with the AI smart home model. Through deep learning of neural networks, a personalized comfort prediction model is constructed, which can accurately predict the needs of different residents in different scenarios.

[0130] The scene analysis submodule, based on an AI smart home model, analyzes sensor data (such as camera, sound, and temperature and humidity sensors) to identify residents and determine their physiological status, differentiating between family members and visitors, and the physiological changes of different members. For example, the system can identify "This is a child in the home, currently very active" or "This is a visitor, sensitive to noise." Using AI scene recognition technology, the system analyzes the current usage scenarios of the living space, including sleeping, working or studying, meeting and leisure, and so on. This technology accurately determines the space's purpose by integrating resident behavior (such as lying down or sitting), location, and environmental characteristics (such as dim lighting or large crowds). For example, if the system detects a resident lying in bed in soft lighting, it identifies the resident as "sleeping." If it detects multiple people chatting or eating, it identifies the scenario as "meeting and leisure." Based on a personalized comfort prediction model, the system combines resident identity characteristics (such as age and health status), real-time physiological status data (such as heart rate and activity level), and the current space usage scenario to generate a real-time comfort demand forecast. For example, when the system detects an elderly person entering a rest area, it predicts they will need a warm environment with moderate lighting. When children are playing, it predicts they will need fresh air and relaxed temperature control. Based on these predictions, the system automatically adjusts home appliances to meet individual comfort needs.

[0131] The beneficial effects of this technical solution include: a shift from passive response to proactive prediction. The system can proactively adjust environmental parameters before occupants experience discomfort, significantly enhancing the living comfort experience. It also provides customized services tailored to the individual needs of different family members, avoiding the "one-size-fits-all" control approach of traditional smart home systems and significantly improving user satisfaction. Through scene recognition and identity recognition technologies, the system can understand complex home usage scenarios, balancing the needs of multiple people in a room and finding the optimal comfort solution.

[0132] In another embodiment, the environment control module includes:

[0133] The device management submodule is used to:

[0134] Register and manage household electrical appliances, including negative oxygen ion generators, ventilation systems, humidifiers, M2 filtration systems, smart air conditioners, floor heating equipment, and lighting control devices;

[0135] Obtain the working status, energy consumption data and operating parameters of each device in real time;

[0136] Control strategy generation submodule, used to:

[0137] Generate environmental control targets based on predicted comfort requirements and current environmental parameters;

[0138] Formulate multi-device collaborative working strategies based on environmental control objectives and determine the working mode and parameter settings of each device;

[0139] The device control submodule is used to issue control instructions to each home appliance according to the collaborative work strategy and monitor the execution status of the instructions.

[0140] Among them, the device control submodule includes:

[0141] Scene mode control unit, used to:

[0142] Based on the occupant's identity and the current scene, select a preset scene mode, including baby sleeping mode, elderly living mode, guest mode or master bedroom sleeping mode;

[0143] When the baby is detected to be sleeping, the temperature in the baby room is automatically maintained at 23-25℃ and the humidity is controlled at 50-60%. At the same time, the negative oxygen ion generator is activated and the light is adjusted to a weak warm light.

[0144] When elderly residents are identified, the room temperature is increased and floor heating is activated, while indoor lighting brightness is increased.

[0145] Energy-optimized control unit for:

[0146] Optimize energy efficiency of equipment operation while ensuring living comfort;

[0147] Automatically adjust equipment start and stop times based on energy consumption data and usage habits to reduce peak energy consumption;

[0148] When it detects that the occupants are away from home, it activates energy-saving mode and restores to optimal comfort 30 minutes before the expected return;

[0149] Special scenario response unit for:

[0150] When multiple visitors are detected, the ventilation frequency of the fresh air system is automatically increased and the cooling power of the air conditioner is adjusted;

[0151] When cooking activity is detected, the M2 filtration system is activated in advance and the exhaust capacity is enhanced;

[0152] When it is recognized that the occupant is in a sleeping state, the room environment parameters are automatically adjusted to a state conducive to deep sleep.

[0153] The working principle of the above technical solution is as follows: the device management submodule uniformly registers and manages various electrical appliances in the home through the smart home central control system to achieve centralized control of the devices. When a new device, such as a negative oxygen ion generator, is connected to the home network, the submodule automatically identifies the device type and obtains its model information, functional parameters and communication protocol, and adds it to the home device list; the system performs regular heartbeat detection on registered devices such as ventilation systems, humidifiers, M2 filtration systems (i.e., air purification devices using M2-level high-efficiency filtration technology), smart air conditioners, floor heating equipment and lighting control devices to ensure that the device connection status is normal; the submodule obtains and records the working status data of each device in real time, such as the current temperature setting, operating mode, and wind speed of the air conditioner, the temperature setting value and actual temperature of the floor heating system, the water level status and humidification amount of the humidifier, as well as key parameters such as energy consumption data and operating time of each device. This information is stored in the local database and regularly synchronized to the cloud to provide data support for subsequent device linkage and strategy formulation.

[0154] The control strategy generation submodule generates precise environmental control targets through intelligent algorithms based on the comfort demand prediction results provided by the scenario analysis submodule and the current environmental parameters collected by the environmental perception submodule. For example, when the system recognizes that the occupant is in a working state and his body temperature is slightly higher than normal, the submodule will combine the current indoor temperature, humidity and air quality data to generate a control target of "lowering the room temperature to a comfortable range and improving air freshness". The control strategy generation submodule takes into account the working characteristics, energy consumption ratio and response speed of each device based on the determined environmental control targets and formulates the optimal strategy for the coordinated operation of multiple devices. For example, it can simultaneously adjust the air conditioning cooling mode and reduce the wind speed, start the ventilation system to increase the fresh air volume, turn on the negative oxygen ion generator to improve air quality and turn off the floor heating system, thereby achieving comprehensive control of temperature, humidity and air quality. At the same time, this submodule continuously optimizes the control strategy through machine learning algorithms, and adjusts the parameter configuration weights of each device based on historical control effects and occupant feedback, making environmental control more intelligent and precise.

[0155] The device control submodule receives the collaborative work strategy issued by the control strategy generation submodule and converts it into control instructions that can be recognized by each device; this submodule sends control instructions to household electrical appliances in sequence according to the preset priority order. For example, it first sends instructions to the smart air conditioner to adjust the temperature and wind speed, then starts the enhanced mode of the ventilation system, then adjusts the humidity output of the humidifier, and finally controls the brightness and color temperature of the lighting equipment to achieve coordinated changes in environmental parameters; during the instruction sending process, the device control submodule monitors the response status and execution progress of each device in real time. When it detects that a device such as the M2 filtration system cannot respond to the instruction normally, it immediately issues a fault reminder to the user and automatically adjusts the control strategy, and selects alternative solutions such as increasing the working intensity of other air purification equipment to ensure the realization of environmental control goals; in addition, this submodule also records the instruction execution history and response time of each device to provide data support for continuous optimization of the system.

[0156] The beneficial effects of the above technical solution are as follows: the environmental control module realizes intelligent and precise control of the home environment through the collaborative work of three sub-modules, thereby improving the living comfort and living experience of the residents; the module can automatically adjust the home environment parameters according to the personalized needs of different family members, eliminating the tedious steps of manually adjusting multiple devices; the control strategy based on the AI smart home model and the personalized comfort prediction model makes the environmental control more in line with the actual needs of the residents, reduces unnecessary energy consumption, and improves the efficiency of equipment use; the multi-device collaborative work mechanism avoids energy waste and equipment loss caused by excessive operation of a single device, and extends the service life of the equipment; the real-time monitoring and fault handling mechanism enhances the stability and reliability of the system, and reduces the interruption of environmental control due to equipment failure.

[0157] In another embodiment, the operation monitoring module includes:

[0158] The environmental monitoring submodule is used to continuously monitor changes in indoor parameters after environmental control and evaluate whether the environmental control effect meets the expected goals;

[0159] Effect evaluation submodule, used to:

[0160] Collect residents' subjective feedback on environmental comfort and changes in physiological indicators;

[0161] Establish mapping relationships between environmental parameters, equipment operating status, and occupant comfort;

[0162] Generate an environmental control effect evaluation report based on the comfort assessment results and equipment operating status;

[0163] The self-optimization submodule is used to analyze the causes and automatically adjust the control strategy when environmental parameters deviate from expectations or equipment operates abnormally.

[0164] The working principle of the above technical solution is as follows: the environmental monitoring submodule continuously collects indoor environmental parameter data through a sensor network composed of multiple types of sensors distributed in various areas of the home; after the smart home system executes the environmental control instructions, the environmental monitoring submodule will track the changing trends of key environmental parameters such as temperature, humidity, air quality index, light intensity, and negative oxygen ion concentration in real time; the system will compare and analyze the collected environmental data with the preset environmental control target values and generate an environmental parameter deviation report; when it is detected that the environmental parameters have reached the expected target, the environmental monitoring submodule sends a compliance signal to the system core, and when the parameters continue to deviate from the expected target and exceed the preset threshold, the early warning mechanism is triggered and the abnormal condition data is pushed to the self-optimization submodule.

[0165] The effect evaluation submodule collects residents' subjective feedback information through the smart home APP interface, smart voice assistant and residents' wearable devices; when residents enter a specific living space, the system will collect residents' evaluation data on the current environmental comfort through pop-up windows or voice inquiries at appropriate times; at the same time, the system obtains residents' physiological indicators such as heart rate, skin resistance, and body surface temperature from wearable devices as a basis for objective evaluation; the effect evaluation submodule associates and stores these subjective feedback and objective physiological data with current environmental parameters and equipment operating status to construct a multi-dimensional mapping relationship matrix; based on this mapping relationship matrix, the system calculates the environmental control effect score and generates a detailed evaluation report, which includes key indicators such as environmental compliance, resident satisfaction, and equipment operating efficiency; when the system recognizes that residents express environmental discomfort through body language (such as frequently adjusting clothes, wiping sweat, rubbing hands, etc.), even if there is no active feedback, the effect evaluation submodule will incorporate this information into the evaluation system and improve the mapping relationship matrix.

[0166] Upon receiving a signal that environmental parameters deviate from expectations or that occupants' comfort ratings are poor, the self-optimization submodule initiates a multi-level analysis process. First, the system checks the operating status of the equipment to confirm whether there is any equipment failure or performance degradation. Second, it analyzes whether external environmental factors (such as extreme weather and outdoor pollution) have had an impact on the indoor environment that exceeds the system's expectations. Third, it compares historical data to identify whether the current environmental control strategy is applicable to the current scenario. Based on the analysis results, the self-optimization submodule generates an adjustment plan, including adjusting equipment operating parameters, changing equipment coordination strategies, or redefining environmental target values. When it is identified that the environmental parameters of a specific area repeatedly fail to meet the expected goals, the self-optimization submodule initiates a deep learning algorithm to model the environmental characteristics of the area and optimize the control logic accordingly. For recurring comfort deviations, the system records the associated scenario features and forms an optimization knowledge base. When encountering similar scenarios in the future, it actively applies the optimized strategy to reduce the number of adjustment iterations.

[0167] The beneficial effects of the above technical solution are: it realizes closed-loop management of smart home environment control, ensuring that environmental parameters are always maintained within the comfort range of residents; through continuous environmental monitoring and effect evaluation, the system can promptly detect and correct environmental control deviations, avoiding residents from being in an uncomfortable environment for a long time; the effect evaluation submodule combines the subjective feelings of residents with objective physiological data, and can comprehensively evaluate the effect of environmental control, preventing the situation of relying solely on environmental parameter data and ignoring the real feelings of the human body; the self-learning ability of the self-optimization submodule enables the system to adapt to different seasons, different weather and the personalized needs of different residents, and gradually improve the accuracy and adaptability of environmental control.

[0168] In another embodiment, the self-optimization submodule includes:

[0169] Learning Optimization Unit for:

[0170] Generate optimized environmental control strategies through the continuous learning function of the DeepSeeK-R model;

[0171] Generate dynamic intelligent control parameters based on seasonal changes, changes in occupant health status, and equipment performance degradation;

[0172] Based on the behavior patterns and life patterns of residents, predict future changes in environmental needs, generate adjustment plans and optimize controls in advance;

[0173] Fault diagnosis subunit, used for:

[0174] When the equipment is operating abnormally or environmental parameters fail to meet expected targets, the cause is analyzed and intelligent fault diagnosis results are generated;

[0175] Identify the fault type based on the diagnosis results and generate a treatment plan from the maintenance strategy library;

[0176] For minor faults that can be automatically recovered, self-repair instructions are generated and executed;

[0177] For serious faults that require manual intervention, detailed maintenance recommendations are generated and maintenance personnel are contacted. At the same time, diagnostic data is fed back to the self-optimization submodule to optimize the control strategy.

[0178] The working principle of the above technical solution is as follows: the learning optimization unit continuously analyzes environmental data and occupant feedback based on the DeepSeeK-R model to generate an intelligent control strategy. When the system detects that the indoor temperature fluctuation exceeds the comfort range, the DeepSeeK-R model will analyze historical temperature change data, outdoor weather forecast information and the occupants' past temperature preferences to generate a predictive temperature control plan; for example, after detecting that the occupants are accustomed to getting up at 7 o'clock in the morning, the system will automatically adjust the bedroom temperature from the night energy-saving mode to the occupants' preferred comfort temperature at 6:30, so that the occupants are in an ideal environment when they get up.

[0179] In response to seasonal and health changes, the system dynamically adjusts control parameters. When external meteorological data indicates a change in seasons, the system automatically adjusts the indoor humidity target range based on changing trends in humidity, air pressure, and other factors. For example, during seasonal changes, if the system detects mild allergic symptoms in residents, it will automatically increase the filtration level of the M2 filtration system and increase the working time of the negative oxygen ion generator, while reducing indoor temperature fluctuations to create a living environment more suitable for people with allergies.

[0180] To achieve early optimization based on the prediction of residents' behavior patterns, the system continuously collects residents' activity data through the sensor network and identifies behavioral patterns such as weekday schedules and weekend activity patterns. For example, after the system recognizes that residents have the habit of inviting friends for dinner every Friday night, it will increase the ventilation frequency of the fresh air system in advance on Friday afternoon and optimize the temperature and humidity in the living room area to ensure air quality and comfort when multiple people are active. At the same time, it pre-adjusts the indoor lighting brightness according to outdoor weather conditions to create an ideal social atmosphere.

[0181] The fault diagnosis subunit identifies system anomalies through multi-source data analysis. When the output temperature of the smart air conditioner continuously deviates from the set value by more than a threshold, the system will simultaneously analyze multi-dimensional data such as the air conditioner operating parameters, power status, and outdoor temperature changes. For example, in hot summer weather, if the room temperature drops slowly after the air conditioner is running, the system will analyze parameters such as the air conditioner refrigerant pressure, evaporator temperature, and compressor current, and combine historical operating data to determine whether it is a common fault such as insufficient refrigerant or a clogged filter. Based on the diagnostic results, the system intelligently retrieves processing solutions and maintains a dynamic maintenance strategy library that contains common faults of various types of equipment and corresponding solutions. For example, when the diagnostic results show that the humidifier water level sensor is abnormal, the system will retrieve the corresponding processing flow from the strategy library, first detecting the humidifier circuit connection status, then attempting to reset the water level sensor, and at the same time suspending the humidifier to prevent potential water leakage risks, and sending the corresponding fault prompt to the user. Execute self-repair instructions for minor faults. When the system detects that the communication module of the floor heating equipment is intermittently disconnected, the self-repair program will automatically restart the communication module and re-establish the connection. For example, after the network fluctuation causes the smart lighting controller to be temporarily offline, the system will automatically execute the network reconnection protocol, restore communication and verify the device status. No user intervention is required throughout the process to ensure that the lighting system continues to operate normally. Provide professional maintenance suggestions for serious faults. When the system detects that the ventilation system motor is abnormally hot, it will immediately reduce the operating power and generate a detailed fault report. For example, after detecting that the pressure difference of the M2 filter system is abnormal and automatic cleaning is ineffective, the system will generate a diagnostic report including the filter status, usage time, and pollutant accumulation, and automatically contact the preset professional maintenance personnel. At the same time, it will push maintenance suggestions to the user, such as "It is recommended to replace the main filter element. The current filter element has been running for more than the recommended service life, and the accumulation of pollutants has affected the filtration efficiency", to ensure that the user understands the cause of the fault and obtains timely and professional maintenance support.

[0182] The beneficial effects of this technical solution include: the system can proactively sense changes in environmental needs and make adjustments, avoiding the discomfort caused by the delayed response of traditional systems and ensuring that residents remain in a comfortable environment. By continuously learning resident preferences and behavioral patterns, the system can provide highly personalized environmental control to meet the unique needs of different residents at different times. The system can automatically adjust environmental parameters based on changes in resident health, providing a healthier living environment for those with allergies, respiratory sensitivity, and other special needs. Through intelligent fault diagnosis and self-repair capabilities, the system can promptly detect and address minor faults, preventing minor issues from escalating into serious damage and extending the life of the equipment.

[0183] In another embodiment, Figure 3 As shown, the intelligent control method based on the AI smart home model includes:

[0184] S101: Real-time collection of home-related data based on a multi-dimensional sensing system, including indoor and outdoor environmental parameters and resident physiological characteristics data;

[0185] S102: Based on the AI smart home model, perform resident comfort demand forecasting and analysis on home-related data to obtain resident comfort needs;

[0186] S103: Based on comfort requirements, the operating parameters of household electrical appliances are set and the operating modes are adjusted to achieve personalized environmental adjustment for different residents and scenarios.

[0187] The working principle of the above technical solution is as follows: Step S102 includes:

[0188] Obtaining environmental parameter types from indoor and outdoor environmental parameters, and obtaining behavioral parameter types from occupant physiological characteristic data;

[0189] Establishing an association mapping between environmental parameter types and behavioral parameter types to form a data group to be processed;

[0190] According to the importance of the environmental parameter type, each data group to be processed is prioritized, and the corresponding environmental parameter and behavior parameter data are stored in the data group to be processed to obtain the home data to be analyzed;

[0191] Obtain smart home device specification data and extract device performance standard data from it;

[0192] Preprocess the device performance standard data to construct the device feature vector, and obtain the device interconnection association data from the smart home device specification data;

[0193] Perform cluster analysis on device feature vectors based on device interconnection association data to generate a data group to be trained;

[0194] Apply the data set to be trained to the initial model for supervised training to obtain an AI smart home model;

[0195] Extract personalized comfort feature data from residents' historical behavior preference data;

[0196] Perform data enhancement and noise reduction on personalized comfort feature data to construct a multimodal feature vector;

[0197] Analyze resident behavior patterns in different living scenarios based on multimodal feature vectors to generate scenario-related data;

[0198] The DeepSeeK-R model is integrated into the AI smart home model, and multimodal feature vectors are deeply clustered based on scene-related data to build a personalized comfort prediction model.

[0199] Based on the AI smart home model, the user can identify the identity and physiological status of the residents, distinguish between family members, visitors, and changes in physiological parameters of different members;

[0200] Use AI smart home models to identify the current living space usage scenarios, including sleeping, working and studying, meeting guests and leisure, cooking activities, and other life scenarios;

[0201] Based on the personalized comfort prediction model, combined with the occupant's identity characteristics, physiological status data and current space usage scenarios, real-time comfort demand prediction results are generated.

[0202] Step S103 includes:

[0203] Register and manage household electrical appliances, including negative oxygen ion generators, ventilation systems, humidifiers, M2 filtration systems, smart air conditioners, floor heating equipment, and lighting control devices;

[0204] Obtain the working status, energy consumption data and operating parameters of each device in real time;

[0205] Generate environmental control targets based on predicted comfort requirements and current environmental parameters;

[0206] Formulate multi-device collaborative working strategies based on environmental control objectives and determine the working mode and parameter settings of each device;

[0207] According to the collaborative work strategy, control instructions are issued to each home appliance and the execution status of the instructions is monitored.

[0208] The beneficial effects of the above technical solution are as follows: Real-time, multi-dimensional data collection enables the system to fully understand the actual state of the living environment and the physiological responses of residents, thereby more accurately identifying the true needs of different residents in different scenarios and avoiding the comfort mismatch caused by traditional systems that rely on single parameter settings. By accurately predicting resident comfort needs through the AI smart home model, the system can provide customized environmental adjustment solutions based on individual differences, meeting the differentiated needs of special groups such as the elderly, children, and pregnant women, and improving the living experience. Based on accurate comfort demand predictions, the system can rationally adjust the operating parameters and modes of various household appliances to avoid excessive operation or functional overlap, reduce energy waste, and lower overall energy consumption while ensuring living comfort. The system can autonomously analyze the correlation between environmental data and resident feedback, continuously optimize control strategies, reduce the frequency of manual intervention, and enhance the intelligent and adaptive capabilities of home environment management. By continuously learning resident behavior patterns and preferences, the system can quickly adjust control strategies as environmental conditions change, adapting to complex factors such as seasonal changes, weather changes, and changes in resident health, to maintain a stable and comfortable living environment.

[0209] In another embodiment, the step S103 includes:

[0210] S104: Monitor and evaluate environmental control results in real time, and perform intelligent diagnosis and adaptive optimization adjustments when environmental parameters deviate from expectations or equipment operates abnormally.

[0211] Step S104 includes:

[0212] Continuously monitor changes in indoor parameters after environmental control and evaluate whether the environmental control effect meets the expected goals;

[0213] Collect residents' subjective feedback on environmental comfort and changes in physiological indicators;

[0214] Establish mapping relationships between environmental parameters, equipment operating status, and occupant comfort;

[0215] Generate an environmental control effect evaluation report based on the comfort assessment results and equipment operating status;

[0216] When environmental parameters deviate from expectations or equipment operates abnormally, the cause is analyzed and the control strategy is automatically adjusted.

[0217] Analyzing causes and automatically adjusting control strategies include:

[0218] Generate optimized environmental control strategies through the continuous learning function of the DeepSeeK-R model;

[0219] Generate dynamic intelligent control parameters based on seasonal changes, changes in occupant health status, and equipment performance degradation;

[0220] Based on the behavior patterns and life patterns of residents, predict future changes in environmental needs, generate adjustment plans and optimize controls in advance;

[0221] When the equipment is operating abnormally or environmental parameters fail to meet expected targets, the cause is analyzed and intelligent fault diagnosis results are generated;

[0222] Identify the fault type based on the diagnosis results and generate a treatment plan from the maintenance strategy library;

[0223] For minor faults that can be automatically recovered, generate and execute self-repair instructions (such as restarting the device or adjusting parameters);

[0224] For serious faults that require manual intervention, detailed maintenance recommendations are generated and maintenance personnel are contacted. At the same time, diagnostic data is fed back to the self-optimization submodule to optimize the control strategy.

[0225] The working principle of the above technical solution is as follows: after the system completes the initial parameter setting of air conditioning, fresh air, lighting and other equipment, the multi-dimensional sensor network continuously collects environmental parameters such as indoor temperature, humidity, CO2 concentration, illumination, etc., and transmits real-time data to the central control unit; at the same time, the system collects physiological indicators such as heart rate, body temperature, movement status of residents through wearable devices and smart furniture, and collects residents' subjective comfort evaluation through mobile applications; the central control unit compares these data with the preset comfort targets, calculates the deviation between the environmental parameters and the target values, and generates an environmental control effect evaluation report.

[0226] When the system detects an abnormal situation, the intelligent diagnosis and adaptive optimization mechanism is triggered. For example, when the temperature in an elderly resident's bedroom remains high and the resident's heart rate is slightly elevated, but the air conditioning operating parameters are normal, the DeepSeeK-R model (a deep reinforcement learning model) will analyze historical data and discover that the resident's sensitivity to temperature increases after taking specific medications. The system will then adjust the room's temperature setting to a lower level and increase the air supply frequency to improve the cooling effect. At the same time, the system will record this association pattern and store it in the knowledge base as the resident's personalized parameter for prediction and early adjustment of similar situations in the future.

[0227] In response to equipment abnormalities, the system performs intelligent fault diagnosis and processing. When the air supply volume of the fresh air system is continuously lower than the set value but the power consumption is normal, the fault diagnosis module determines that the filter is clogged by analyzing the vibration sensor and wind pressure data; the system first tries to automatically switch to the backup filter channel and increase the fan power as a temporary solution; at the same time, the system generates a detailed fault analysis report, including the cause, the validity period of the temporary solution and the recommended maintenance operations; if this is a minor fault that can be handled automatically, the system directly executes the processing instructions; if manual intervention is required, the system sends a notification to the resident or property through the home management application, with detailed maintenance suggestions and possible repair service provider information.

[0228] The system continuously learns and optimizes environmental control strategies. Over time, it records the behavioral patterns and preference changes of residents in different seasons and weather conditions. For example, it finds that residents tend to sleep late and prefer lower temperatures on weekend mornings in the summer. Based on this, the system adjusts the bedroom temperature to a lower level in advance on weekend mornings and delays the automatic opening time of the curtains. When it monitors the physiological signals of the residents who are about to wake up, the system gradually adjusts the environmental parameters to the resident's preferred wakefulness state configuration. Through this predictive adjustment, the system can proactively create a comfortable environment before the residents have a demand, thereby reducing the time of discomfort.

[0229] The beneficial effects of the above technical solution are: through real-time monitoring and evaluation of environmental control effects, it can accurately identify the deviation between environmental parameters and comfort targets and automatically adjust them, avoiding the comfort mismatch problem that may be caused by traditional systems relying solely on preset parameters, making environmental control more accurate and intelligent. By continuously learning the physiological reactions and subjective feedback of residents, the system can establish an individualized mapping relationship between environmental parameters and comfort, dynamically adjust the control strategy according to factors such as the health status of residents and seasonal changes, provide environmental conditions that are truly suitable for individual needs, and enhance the living experience. The intelligent fault diagnosis function enables the system to detect and handle equipment anomalies in a timely manner, automatically execute repair measures for minor faults, and provide detailed diagnostic results and handling suggestions for problems that require manual intervention, greatly reducing the impact of system failures on living comfort, while improving the accuracy and efficiency of maintenance.

[0230] In another embodiment, step S101 includes:

[0231] According to the layout of the living space and the needs of the residents, multi-dimensional environmental monitoring sensors and physiological characteristics collection equipment are configured in different functional areas;

[0232] The acquisition frequency of the sensor configuration submodule is dynamically adjusted based on different spatial functional attributes to achieve focused monitoring of key areas.

[0233] The working principle of the above technical solution is: differentiated sensor configuration and data collection are carried out according to the characteristics of different functional areas. The system first divides the living space into functional areas, such as living room, bedroom, kitchen, bathroom, etc., and then configures corresponding sensor equipment according to the characteristics of each area and the needs of the residents. During the living space layout analysis phase, the system will identify the frequency of use and importance of each functional area. For example, the kitchen area is used as a food processing and cooking area and is equipped with temperature, humidity, gas (carbon monoxide, natural gas) concentration sensors and smoke detectors; the bedroom area is used as a resting area and is equipped with temperature, humidity, light, noise sensors and mattress pressure sensors to monitor sleep quality; the bathroom area is equipped with humidity, temperature and water level sensors to prevent flooding.

[0234] For elderly residents or those with special health needs, the system will add devices to collect physiological characteristics such as heart rate and blood pressure in areas where they frequently move. For example, heart rate monitoring sensors can be installed in seats frequently used by the elderly, respiratory monitoring devices can be installed under mattresses in bedrooms, and motion capture cameras can be installed in hallways and corridors to detect fall risks.

[0235] The system intelligently adjusts data collection frequency based on the functional attributes of the space. For example, in the kitchen, if the system detects that the stove is on, it automatically increases the collection frequency of the gas and temperature sensors from the standard rate of once per minute to once every 10 seconds, thereby promptly identifying potential safety hazards. In the bedroom, the system reduces the collection frequency of the light sensor while increasing the sensitivity of the noise and pressure sensors to monitor sleep status without disturbing normal rest.

[0236] The system can sense changes in residents' daily behavior patterns and adjust its monitoring strategy. For example, if the system detects that a resident has recently increased their time in the study, it will automatically increase the frequency and accuracy of sensor data collection in that area to ensure real-time monitoring of environmental comfort and health indicators. For areas that have been unoccupied for extended periods, the system will reduce the frequency of sensor data collection to conserve energy.

[0237] In certain situations, the system triggers a coordinated monitoring mechanism. For example, if a bathroom humidity sensor detects abnormally high humidity for a period exceeding a preset threshold, while no human activity is detected, the system immediately increases the acquisition frequency of all sensors in that area and sends a warning message to the user indicating a possible water leak.

[0238] The beneficial effects of this technical solution include: through differentiated sensor configuration strategies, the system can provide precise monitoring tailored to the characteristics of different functional areas, avoiding the "one-size-fits-all" drawbacks of traditional monitoring systems and improving the pertinence and practicality of monitoring. A mechanism for dynamically adjusting data collection frequency effectively balances monitoring accuracy with system resource consumption, ensuring uninterrupted security monitoring of critical areas while avoiding unnecessary data redundancy and energy waste. Intelligent monitoring of occupant physiological characteristics provides seamless support for health management, particularly suitable for the elderly and those with special health needs, and enhancing the safety and care of living spaces.

[0239] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. Intelligent control system based on AI smart home model, characterized by: include: A data acquisition module is used to collect home-related data in real time based on a multi-dimensional sensing system, including indoor and outdoor environmental parameters and resident physiological characteristics data; The data analysis module is used to predict and analyze the comfort needs of residents based on the AI smart home model and obtain the comfort needs of residents; The environmental control module is used to set the operating parameters and adjust the working mode of household electrical appliances based on comfort requirements, so as to achieve personalized environmental adjustment for different residents and scenarios.

2. The intelligent control system based on the AI smart home model according to claim 1 is characterized in that: Also includes: The operation monitoring module is used to monitor and evaluate the environmental control results in real time, and to perform intelligent diagnosis and adaptive optimization adjustments when environmental parameters deviate from expectations or equipment operation is abnormal.

3. The intelligent control system based on the AI smart home model according to claim 1 is characterized in that: The data acquisition module includes: AI configuration submodule, used for: Using AI smart home models to analyze the layout of residential spaces and the needs of residents, it automatically generates configuration plans for multi-dimensional environmental monitoring sensors and physiological characteristics collection devices for different functional areas; Based on historical data and user feedback, AI learning algorithms are used to predict sensing needs in different areas, continuously optimize the type, quantity, and location of sensors, and improve the accuracy and efficiency of data collection. Sensor configuration submodule, used for: According to the configuration plan generated by the AI configuration submodule, multi-dimensional environmental monitoring sensors and physiological characteristics collection equipment are configured for different functional areas; Based on the real-time analysis of the functional attributes of different spaces by the AI smart home model, the acquisition frequency of the sensor configuration sub-module is dynamically adjusted in combination with the prediction model to optimize energy consumption and data quality, and achieve focused monitoring of key areas.

4. The intelligent control system based on the AI smart home model according to claim 1, characterized in that: Data analysis modules include: Model building submodule for: Use AI algorithms to automatically extract key environmental parameter types from indoor and outdoor environmental parameters, and automatically identify behavioral parameter types from occupants' physiological characteristic data; Through AI correlation analysis technology, a dynamic correlation mapping between environmental parameter types and behavioral parameter types is established to form a data group to be processed; Based on the AI priority evaluation model, each data group to be processed is prioritized according to the importance and real-time changes of environmental parameter types to form home data to be analyzed; Obtain smart home device specification data and use AI feature extraction technology to extract device performance standard data from it; Perform AI-driven data preprocessing on device performance standard data to construct device feature vectors, while also obtaining device interconnection correlation data from smart home device specification data; Based on the AI clustering algorithm, cluster analysis is performed on device feature vectors according to device interconnection data to generate a data group to be trained; Apply the training data set to the initial model, and use AI learning technology to perform model training and parameter optimization to obtain an AI smart home model; Using AI feature engineering technology to extract personalized comfort feature data from residents' historical behavioral preference data; Perform AI data enhancement and adaptive noise reduction on personalized comfort feature data to construct multi-dimensional features; Based on AI behavioral pattern analysis technology, it analyzes the patterns of multi-dimensional features in different life scenarios and generates deep scene correlation data; Input the deep scene association data into the LLM large language model, perform collaborative calculations with the AI smart home model, and build a personalized comfort prediction model through deep learning of neural networks; The scenario analysis submodule is used to: Based on the AI smart home model, the user can identify the identity and physiological status of the residents, distinguish between family members, visitors, and changes in physiological parameters of different members; Utilize AI scene recognition technology to analyze the current living space usage scenarios, including sleeping, working and studying, meeting and leisure, and cooking activities; Based on the personalized comfort prediction model, combined with AI's in-depth analysis of occupants' identity characteristics, physiological status data and current space usage scenarios, real-time comfort demand prediction results are generated.

5. The intelligent control system based on the AI smart home model according to claim 1 is characterized in that: The environmental control module includes: The device management submodule is used to: Register and manage household electrical appliances, including negative oxygen ion generators, ventilation systems, humidifiers, M2 filtration systems, smart air conditioners, floor heating equipment, and lighting control devices; Obtain the working status, energy consumption data and operating parameters of each device in real time; Control strategy generation submodule, used to: Generate environmental control targets based on predicted comfort requirements and current environmental parameters; Formulate multi-device collaborative working strategies based on environmental control objectives and determine the working mode and parameter settings of each device; The device control submodule is used to issue control instructions to each home appliance according to the collaborative work strategy and monitor the execution status of the instructions.

6. The intelligent control system based on the AI smart home model according to claim 2, characterized in that: The operation monitoring module includes: The environmental monitoring submodule is used to continuously monitor changes in indoor parameters after environmental control and evaluate whether the environmental control effect meets the expected goals; Effect evaluation submodule, used to: Collect residents' subjective feedback on environmental comfort and changes in physiological indicators; Establish mapping relationships between environmental parameters, equipment operating status, and occupant comfort; Generate an environmental control effect evaluation report based on the comfort assessment results and equipment operating status; The self-optimization submodule is used to analyze the causes and automatically adjust the control strategy when environmental parameters deviate from expectations or equipment operates abnormally.

7. The intelligent control system based on the AI smart home model according to claim 6 is characterized in that: The self-optimization submodules include: Learning Optimization Unit for: Generate optimized environmental control strategies through the continuous learning function of the DeepSeeK-R model; Generate dynamic intelligent control parameters based on seasonal changes, changes in occupant health status, and equipment performance degradation; Based on the behavior patterns and life patterns of residents, predict future changes in environmental needs, generate adjustment plans and optimize controls in advance; Fault diagnosis subunit, used for: When the equipment is operating abnormally or environmental parameters fail to meet expected targets, the cause is analyzed and intelligent fault diagnosis results are generated; Identify the fault type based on the diagnosis results and generate a treatment plan from the maintenance strategy library; For minor faults that can be automatically recovered, self-repair instructions are generated and executed; For serious faults that require manual intervention, detailed maintenance recommendations are generated and maintenance personnel are contacted. At the same time, diagnostic data is fed back to the self-optimization submodule to optimize the control strategy.

8. The intelligent control method based on the AI smart home model is characterized by: include: S101: Real-time collection of home-related data based on a multi-dimensional sensing system, including indoor and outdoor environmental parameters and resident physiological characteristics data; S102: Based on the AI smart home model, perform resident comfort demand forecasting and analysis on home-related data to obtain resident comfort needs; S103: Based on comfort requirements, the operating parameters of household electrical appliances are set and the operating modes are adjusted to achieve personalized environmental adjustment for different residents and scenarios.

9. The intelligent control method based on the AI smart home model according to claim 8, characterized in that: The following steps are included after step S103: S104: Monitor and evaluate environmental control results in real time, and perform intelligent diagnosis and adaptive optimization adjustments when environmental parameters deviate from expectations or equipment operates abnormally.

10. The intelligent control system based on the AI smart home model according to claim 8, characterized in that: Step S101 includes: According to the layout of the living space and the needs of the residents, multi-dimensional environmental monitoring sensors and physiological characteristics collection equipment are configured in different functional areas; The acquisition frequency of the sensor configuration submodule is dynamically adjusted based on different spatial functional attributes to achieve focused monitoring of key areas.

Images