Intelligent multi-dimensional air purification cloud intelligent control method and system

Abstract

The invention relates to an intelligent multi-dimensional air purification cloud intelligent control method and system, and solves the problem that the complexity of an indoor environment, such as temperature and humidity, concentration of other pollutants and factors of different sensitivities of a user to air quality, cannot be comprehensively considered due to the fact that operation of a purifier is adjusted simply according to a single air quality index. The method comprises the following steps: analyzing a voice control instruction to extract key information, analyzing the extracted key information, real-time air quality data and air quality trend information in a future preset time range, inputting the information into an intelligent decision algorithm model of a cloud end, and outputting operating parameters needing to be adjusted of the air purifier, forming a parameter adjustment decision instruction; the formed parameter adjustment decision instruction is issued to an air purifier terminal through a network; and the air purifier responds to the parameter adjustment decision instruction to adjust the operation parameters of the air purifier. The method has the following effects that the air quality is improved, and the requirement of a user for the indoor air environment is met.

Description

Technical Field

[0001] The present invention relates to the technical field of smart homes, and in particular to a smart multi-dimensional air purification cloud intelligent control method and system. Background Art

[0002] As people's requirements for quality of life continue to improve, the optimization of indoor air quality has received increasing attention. The intelligent multi-dimensional air purification system aims to create a more comfortable and healthy indoor environment for users, and realize intelligent control and environmental regulation of air purifiers through advanced technical means. In the field of smart home, intelligent air purification technology has gradually become a hot spot for research and application, which involves the integration of multiple technologies to meet the diverse needs of users.

[0003] At present, traditional air purification control methods mainly rely on simple sensor detection and preset operating modes. For example, some air purifiers can only adjust the operating wind speed according to a single air quality indicator (such as PM2.5 concentration) detected. When the PM2.5 concentration is detected to be excessive, the wind speed is automatically adjusted to a high level, and then adjusted back to a low level after the concentration decreases. There are also some smart air purifiers that can receive user voice commands, but can only perform simple command operations, such as turning on and off equipment, switching fixed modes, etc., and lack comprehensive consideration of complex environmental information and user's diverse needs.

[0004] With respect to the above-mentioned related technologies, the inventors found that there are the following defects: simply adjusting the operation of the purifier based on a single air quality indicator cannot fully consider the complexity of the indoor environment, such as temperature and humidity, concentrations of other pollutants, and users' different sensitivities to air quality. Summary of the invention

[0005] In order to improve air quality and meet users' needs for indoor air environment, the present application provides an intelligent multi-dimensional air purification cloud-based intelligent control method and system.

[0006] In the first aspect, the present application provides a cloud-based intelligent control method for intelligent multi-dimensional air purification, which adopts the following technical solutions: A cloud-based intelligent control method for intelligent multi-dimensional air purification, comprising: Obtain voice control instructions for the air purifier from users in the room; Based on the room's spatial layout information, historical air quality data, and real-time air quality data, a spatiotemporal data analysis algorithm based on deep learning is used to predict and obtain information on the air quality trend within a preset time range in the future; The voice control command is parsed to extract key information, and the extracted key information, real-time air quality data, and air quality trend information within a preset time range in the future are parsed and input into the cloud-based intelligent decision-making algorithm model, and the operating parameters that need to be adjusted for the air purifier are output to form parameter adjustment decision instructions; The formed parameter adjustment decision instruction is sent to the air purifier terminal through the network; The air purifier adjusts its own operating parameters in response to the parameter adjustment decision instruction.

[0007] By adopting the above technical solutions, the intelligent cloud-based intelligent control method of Zhilian Multi-dimensional Air Purification can understand user needs through voice control commands, predict the trend of air quality based on spatiotemporal data analysis algorithms, output operating parameters through intelligent decision-making algorithm models, and accurately control the air purifier. The technical effect is remarkable, realizing intelligent, precise, and personalized air purification control, improving air quality, and meeting users' needs for indoor air environment.

[0008] Optionally, the method further includes steps after obtaining the voice control instructions of the user in the room regarding the air purifier, which are as follows: Analyze whether relevant physiological data related to user emotions is obtained; If not, maintain the original setting; If yes, based on the obtained physiological data related to the user's emotions, a multimodal emotion fusion and feedback algorithm is used to analyze and obtain emotion-related data, and feature extraction is performed to obtain emotion feature data; The pre-processed voice command key information, emotional characteristic data, real-time air quality data and future air quality trend information are input together into an advanced decision-making module based on the fusion of fuzzy logic and expert system, and the operating parameters that need to be adjusted for the air purifier are output to form parameter adjustment decision instructions. The parameter adjustment decision instructions are then sent to the air purifier terminal through the network, and the air purifier adjusts its own operating parameters in response to the parameter adjustment decision instructions.

[0009] By adopting the above technical solutions, the intelligent multi-dimensional air purification cloud intelligent control method can understand user needs through voice control commands, predict the trend of air quality based on spatiotemporal data analysis algorithms, output operating parameters through intelligent decision-making algorithm models, and accurately control the air purifier. The technical effect is remarkable, realizing intelligent, precise, and personalized air purification control, improving air quality, and meeting users' needs for indoor air environment.

[0010] Optionally, the operation method of the advanced decision module based on the fusion of fuzzy logic and expert system is as follows: The fuzzy logic system fuzzifies the input data; Reasoning is performed based on the pre-set rule base in the expert system, which contains strategies for air purification and environmental regulation in various situations; After fuzzy logic reasoning, the decision module will generate detailed equipment control instructions, which cover the operating parameter adjustment of the air purifier and the linkage control of other smart devices.

[0011] By adopting the above technical solution, the operation method combines fuzzy logic and expert system, first fuzzifies the input data, and then infers based on the rule base. The technical effect is obvious, and it can use the experience of experts, comprehensively consider the various situations of air purification and environmental regulation, and accurately generate equipment control instructions, which not only optimizes the operating parameters of the air purifier, but also realizes the linkage of intelligent devices, improves the accuracy and intelligence of environmental regulation, and creates a more comfortable environment for users.

[0012] Optionally, the method further includes steps after obtaining the voice control instructions of the user in the room regarding the air purifier, which are as follows: Analyze whether the acquired voice control instructions come from multiple users; If not, maintain the original setting; If yes, each user is considered as an intelligent agent, and the voice commands of each user are parsed to extract key information. Then, the parsed key information, together with the room space layout, air quality data of each corner, and air quality trend information within a preset time range in the future, are used as the input of the multi-agent reinforcement learning algorithm. Each agent uses the decision-making model in the multi-agent reinforcement learning algorithm to select matching actions based on its own state information and the acquired environmental information. The scope of actions includes but is not limited to adjusting the operating parameters of the air purifier and controlling the switches or parameters of other smart devices.

[0013] By adopting the above technical solutions, the system's adaptability to multi-user scenarios is enhanced. The source of voice commands is analyzed, and the settings are maintained to ensure stability when there are only one user. When there are multiple users, the user is regarded as an intelligent agent, and a variety of environmental information is input into the multi-agent reinforcement learning algorithm. Each intelligent agent makes independent decisions and can accurately adjust the air purifier and other equipment to meet the needs of different users, thereby improving the personalization and intelligence of air purification and environmental regulation.

[0014] Optionally, it also includes the following steps where each agent selects a matching action based on its own state information and the acquired environmental information using a decision model in a multi-agent reinforcement learning algorithm, as follows: Various sensors distributed indoors are used to collect real-time environmental change data, and user satisfaction feedback on the current environment is collected regularly through user terminal devices; Summarize the collected environmental change data and user satisfaction feedback on the current environment to form environmental feedback information; According to the feedback from the environment, each agent is rewarded or punished according to the pre-set reward and punishment rules; The intelligent agent uses the learning mechanism of the reinforcement learning algorithm to update its own behavior strategy based on the reward or punishment feedback obtained.

[0015] By adopting the above technical solutions, a complete feedback optimization mechanism is constructed. Environmental changes and satisfaction feedback are collected through sensors and user terminals to form environmental feedback information. The rewards and punishments of the intelligent agents are determined according to the reward and punishment rules, so that they can update their behavior strategies. This allows the system to dynamically adjust according to the actual environment and user experience, continuously optimize the air purification and environmental regulation effects, improve user satisfaction, and make intelligent decisions more in line with actual needs.

[0016] Optionally, the agent uses the learning mechanism of the reinforcement learning algorithm to update its own behavior strategy based on the reward or penalty feedback obtained, including: Analyze whether relevant physiological data related to user emotions is obtained; If not, the agent uses the learning mechanism of the reinforcement learning algorithm to update its behavior strategy based on the reward or punishment feedback obtained; If yes, based on the obtained physiological data related to the user's emotions, a multimodal emotion fusion and feedback algorithm is used to analyze and obtain emotion-related data, and feature extraction is performed to obtain emotion feature data; Assign weights based on the contribution of different modal data to emotional judgment, fuse the weighted multimodal data, and input them into the deep learning-based emotional analysis model to determine the emotional state of each user; Count the number and proportion of agents that want to reduce noise and increase wind speed to deal with pollution, and analyze the preliminary strategies associated with agents whose proportion exceeds the preset threshold; Combine the sentiment analysis results with the preliminary strategy derived from the agent proportion analysis.

[0017] By adopting the above technical solutions, the update logic of the intelligent agent behavior strategy is enriched. By analyzing whether there is user physiological data, it is decided whether to enable the multimodal emotion fusion algorithm. Extracting emotional features from physiological data, combining deep learning models to judge emotions, statistically analyzing intelligent agent strategies, and then integrating emotions with strategies can make the intelligent agent behavior strategy more comprehensively consider user emotions and actual needs, achieve more humane and precise environmental control, and improve user experience.

[0018] Optionally, the steps after fusing the sentiment analysis results with the preliminary strategy derived from the agent proportion analysis are also included, as follows: Input the current indoor environment parameters, the existing status of the smart devices and the control instructions in the fusion strategy to simulate the environmental changes after the strategy is implemented; According to the simulation results and the preset optimization goals, the fusion strategy is further optimized by referring to the success and failure experience of the intelligent agent in similar historical scenarios; Convert the optimized strategy into specific equipment control instructions.

[0019] By adopting the above technical solutions, this step improves the strategy implementation process. By simulating environmental changes after the implementation of the strategy, the effect can be estimated in advance. The strategy is optimized and integrated by combining preset goals and historical experience to make the strategy more scientific and reasonable. The optimization strategy is converted into control instructions to accurately control the equipment. The effectiveness and accuracy of air purification and environmental regulation are improved overall, ensuring that the system decision can better meet the actual environmental needs and user expectations.

[0020] In the second aspect, the present application provides a smart multi-dimensional air purification cloud intelligent control system, which adopts the following technical solutions: A smart multi-dimensional air purification cloud intelligent control system includes a memory, a processor, and a program stored in the memory and executable on the processor. The program can be loaded and executed by the processor to implement the smart multi-dimensional air purification cloud intelligent control method as described in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flow chart of an intelligent multi-dimensional air purification cloud-based intelligent control method according to an embodiment of the present application.

[0022] Figure 2 It is a flowchart diagram of the steps after obtaining the voice control instructions of the user in the room regarding the air purifier in another embodiment of the present application.

[0023] Figure 3 It is a flowchart of an operation method of an advanced decision-making module based on the fusion of fuzzy logic and expert system according to another embodiment of the present application.

[0024] Figure 4 It is a flowchart diagram of the steps after obtaining the voice control instructions of the user in the room regarding the air purifier in another embodiment of the present application. DETAILED DESCRIPTION

[0025] The present application is further described in detail below in conjunction with the accompanying drawings.

[0026] Reference Figure 1 , is a cloud-based intelligent control method for intelligent multi-dimensional air purification disclosed in this application, comprising: Step S100, obtaining voice control instructions of the user in the room regarding the air purifier.

[0027] Voice control instructions refer to the operating instructions conveyed by the user to the air purifier in the form of voice, which are intended to enable the air purifier to perform specific tasks to meet the user's needs for air quality and related environmental adjustments.

[0028] Acquisition method: Arrange voice collection equipment at appropriate locations in the room, such as the microphone array of a smart speaker or the sound pickup device of the air purifier itself. When the user issues a voice command, these devices convert the sound signal into an electrical signal, which is then converted into a digital signal through analog-to-digital conversion and transmitted to the system for processing.

[0029] For example: When a user says "adjust the air purifier's wind speed to the maximum" or "turn on the air purifier's sterilization mode", the system obtains such voice commands for subsequent operations.

[0030] Step S200, based on the spatial layout information of the room, historical air quality data, and real-time air quality data, a spatiotemporal data analysis algorithm based on deep learning is used to predict and obtain air quality trend information within a preset time range in the future.

[0031] Room space layout information: describes the room's structure, size, furniture placement, and the location of vents, doors, and windows, which affects the flow of air in the room and the diffusion of pollutants.

[0032] Historical air quality data: recorded data of various indicators of indoor air quality (such as PM2.5, formaldehyde, TVOC concentration, etc.) over the past period of time, reflecting the changing patterns of air quality.

[0033] Real-time air quality data: The air quality index data in the room monitored in real time by the air quality sensor at the current moment.

[0034] Spatiotemporal data analysis algorithm based on deep learning: Utilize deep learning models, combine data features in time and space dimensions, explore the trend of air quality changes over time and its distribution patterns in space, and thus predict future air quality.

[0035] The specific process is as follows: First, collect the room space layout information, which can be obtained through architectural drawings or measured and modeled using technologies such as laser scanning.

[0036] Historical air quality data is derived from long-term monitoring records and stored in the database. Real-time air quality data is collected in real time by sensors installed at different locations in the room.

[0037] The data is input into a trained spatiotemporal data analysis algorithm model based on deep learning. The model will learn the spatiotemporal characteristics in the data, such as the change pattern of air quality in different seasons and time periods, and the impact of spatial location on air quality. Through these learned patterns, the air quality trend information within the preset time range (such as the next 1 hour, 3 hours, etc.) is predicted, such as whether the PM2.5 concentration will rise, fall or remain stable.

[0038] In step S300, the voice control command is parsed to extract key information, and the extracted key information, real-time air quality data, and air quality trend information within a preset time range in the future are parsed and input into the intelligent decision-making algorithm model in the cloud, and the operating parameters that need to be adjusted for the air purifier are output to form parameter adjustment decision instructions.

[0039] Voice control command analysis: The voice commands input by the user are converted into text form that can be understood by the computer through natural language processing technology, and the core operation information and target parameters are extracted.

[0040] Key information: Specific operating requirements of the air purifier extracted from voice commands, such as adjusting wind speed, switching modes, setting operating time, and other key content.

[0041] Real-time air quality data: The current air quality index values, such as formaldehyde concentration, PM10 value, etc., are collected and transmitted in real time by various air quality sensors in the room.

[0042] Air quality trend information within a preset time range in the future: Through a deep learning-based spatiotemporal data analysis algorithm, the changing trend and indicator estimate of air quality for a period of time in the future (such as 1-2 hours) are predicted.

[0043] Intelligent decision-making algorithm model: An algorithm model that is trained based on a large amount of data and integrates multiple factors for intelligent analysis and decision-making. It can output the optimal operating parameters of the air purifier based on the input information.

[0044] The specific process is as follows: With the help of speech recognition and natural language processing tools, voice control commands are parsed. For example, the key information "increase the wind speed of the air purifier" can be extracted.

[0045] Real-time air quality data and information on future air quality trends are acquired through sensors and prediction algorithms and transmitted to the cloud.

[0046] These key information, real-time air quality data, and information on future air quality trends are input together into the trained intelligent decision-making algorithm model in the cloud. The model comprehensively considers these factors and, after complex calculations, outputs the operating parameters that need to be adjusted for the air purifier, such as fan speed and filter working intensity, to form parameter adjustment decision instructions for precise control of the air purifier.

[0047] Step S400, sending the formed parameter adjustment decision instruction to the air purifier terminal through the network.

[0048] Parameter adjustment decision instructions: generated by the cloud-based intelligent decision-making algorithm model, containing information on various operating parameters that need to be adjusted for the air purifier, such as adjusting wind speed, changing purification mode, and other specific instruction content, which are used to accurately control the operating status of the air purifier.

[0049] Network: covers a variety of network communication methods, such as wireless communication networks such as Wi-Fi, Bluetooth, ZigBee, and wired networks such as Ethernet, responsible for data transmission between the cloud and the air purifier terminal.

[0050] Air purifier terminal: the air purification device actually used by the user, which receives instructions from the cloud and performs corresponding operations to adjust the indoor air quality.

[0051] The specific process is as follows: When the cloud completes the generation of the parameter adjustment decision instruction, the instruction is encapsulated in a specific data format, which contains key information such as device identification, instruction type, parameter value, etc., to ensure accurate transmission to the target air purifier.

[0052] The encapsulated instructions are sent out through a pre-set network connection using a network protocol, such as TCP / IP. If it is a home environment, it is generally transmitted through a home Wi-Fi network; in some commercial places, it may be transmitted using an enterprise-level wired or wireless network.

[0053] The air purifier terminal has a built-in corresponding network receiving module, which continuously monitors the network signal. Once it receives an instruction that matches its own device identification, it immediately parses the instruction, extracts the adjustment parameters, and prepares for the next step of executing the instruction.

[0054] Step S500: The air purifier adjusts its own operating parameters in response to the parameter adjustment decision instruction.

[0055] Parameter adjustment decision instructions: Instructions sent from the cloud carry information on adjusting the operating parameters of the air purifier, including specific parameters such as wind speed, air volume, working mode, and filter cleaning frequency.

[0056] Operating parameters: various indicators that determine the working status of the air purifier. The adjustment of these parameters directly affects the air purification effect and energy consumption, etc.

[0057] The specific process is as follows: The control chip or microprocessor inside the air purifier is responsible for receiving and processing instructions. When receiving the parameter adjustment decision instruction, the instruction is first checked to ensure that the instruction is complete and error-free.

[0058] After the verification is passed, the control chip analyzes the operating parameters that need to be adjusted according to the command content. For example, if the command requires the wind speed to be adjusted from level 3 to level 5, the control chip will identify the two key information "wind speed" and "level 5".

[0059] The control chip sends a control signal to the corresponding execution component based on the analysis results. For wind speed adjustment, it will control the fan motor to change the speed; if the working mode is switched, it will control the circuit to switch to the corresponding working circuit module, thereby adjusting the operating parameters of the air purifier to meet the indoor air quality adjustment requirements.

[0060] Reference Figure 2 A smart multi-dimensional air purification cloud intelligent control method also includes steps after obtaining the voice control instructions of the user in the room about the air purifier, which are as follows: Step SA00, analyzing whether relevant physiological data that feedbacks the user's emotions is obtained. If not, executing step SB00; if yes, executing step SC00.

[0061] Physiological data related to user emotions: refers to human physiological indicators that can reflect the user's current emotional state, such as heart rate, blood pressure, skin galvanic response, facial expression muscle movement, etc. Changes in these data are often associated with emotional fluctuations and can be used as a basis for judging user emotions.

[0062] The analysis method is as follows: the system will query various physiological data collection devices or related data storage areas connected to it to confirm whether there is user physiological data that meets the requirements. For example, check whether there are connected smart bracelets (which can collect heart rate, skin galvanic response and other data), facial expression recognition cameras and other devices, and whether these devices transmit data to the system normally. If there is data transmission and meets the format requirements, it is determined that the relevant physiological data that reflects the user's emotions has been obtained; otherwise, it is determined that it has not been obtained.

[0063] Step SB00, maintain the original settings.

[0064] When the system determines that relevant physiological data to reflect the user's emotions has not been obtained, the operating parameters and control strategies of the current air purifier and related smart devices will not be changed, and the system will continue to operate according to the existing settings.

[0065] Step SC00, based on the acquired physiological data related to the feedback of the user's emotions, adopt a multimodal emotion fusion and feedback algorithm to analyze and acquire emotion-related data, and perform feature extraction to acquire emotion feature data.

[0066] Multimodal emotion fusion and feedback algorithm: an algorithm that comprehensively processes multiple different types of physiological data to more accurately analyze user emotions. These data modalities include but are not limited to heart rate, blood pressure, facial expressions, etc. By fusing different modal data, the emotional information contained therein is mined and corresponding feedback is given.

[0067] Emotion-related data: data directly related to emotional expression analyzed from the user's physiological data, such as the amplitude of heart rate changes, muscle movement characteristics of facial expressions, etc. These data can provide a basis for judging emotions.

[0068] Emotional feature data: data obtained after feature extraction of emotion-related data. It is a further refinement and abstraction of emotion-related data, is more representative, and is helpful for subsequent emotion analysis and decision-making.

[0069] The specific process is as follows: First, the collected multimodal physiological data such as heart rate, blood pressure, facial expressions, etc. are input into the multimodal emotion fusion and feedback algorithm.

[0070] The algorithm preprocesses the data of each modality, such as filtering the heart rate data to remove noise interference; normalizing the facial expression images to unify the image size.

[0071] Next, the algorithm analyzes and extracts information closely related to emotions from the preprocessed data, such as analyzing the degree of emotional excitement from changes in heart rate, identifying basic emotion types such as joy, anger, sorrow, and happiness from facial expressions, and obtaining emotion-related data.

[0072] Finally, feature extraction techniques, such as principal component analysis (PCA) and linear discriminant analysis (LDA), are used to extract the features that best represent the emotional state from emotion-related data to form emotion feature data, preparing for subsequent more accurate emotion analysis.

[0073] Step SD00, input the pre-processed voice command key information, emotional characteristic data, real-time air quality data and future air quality trend information into the advanced decision-making module based on the fusion of fuzzy logic and expert system, output the operating parameters that need to be adjusted for the air purifier, form parameter adjustment decision instructions, and send the formed parameter adjustment decision instructions to the air purifier terminal through the network. The air purifier adjusts its own operating parameters in response to the parameter adjustment decision instructions.

[0074] Key information of pre-processed voice commands: The core operation content extracted after voice recognition and natural language processing of user voice commands, such as adjusting wind speed, switching modes, etc., removes redundant and interfering information.

[0075] Emotional feature data: Feature data extracted from the user's multimodal physiological data that can reflect the user's emotional state and provide a reference for the emotional dimension for decision-making.

[0076] Real-time air quality data: The current indoor air quality index values ​​​​obtained in real time through air quality sensors, such as PM2.5, formaldehyde concentration, etc.

[0077] Information on future air quality trends: Use spatiotemporal data analysis algorithms to predict the changing trends and estimated indicators of indoor air quality in the future.

[0078] Advanced decision-making module based on the integration of fuzzy logic and expert system: a module that combines the ability of fuzzy logic to process uncertain information with the rich domain knowledge of expert system and integrates multi-source information to make intelligent decisions.

[0079] The specific process is as follows: The pre-processed voice command key information, emotional characteristic data, real-time air quality data and future air quality trend information are organized in a specific format and input into the advanced decision-making module.

[0080] The fuzzy logic system first fuzzifies these input data, converting precise numerical values ​​into fuzzy concepts, such as quantifying fuzzy expressions such as "high concentration of formaldehyde" and "low wind speed requirement".

[0081] The expert system makes inferences based on the pre-set rule base and fuzzy information. The rule base stores strategies for air purification and environmental regulation under various environmental conditions and user needs.

[0082] After fuzzy logic reasoning and expert system rule matching, the decision module outputs the operating parameters that need to be adjusted for the air purifier, such as fan speed, filter replacement reminder, etc., to form parameter adjustment decision instructions.

[0083] Decision instructions are transmitted to the air purifier terminal via Wi-Fi, Bluetooth and other networks.

[0084] The air purifier terminal receives and analyzes the instructions, and adjusts its own operating parameters according to the instructions, such as changing the wind speed and switching the purification mode, to achieve precise control of indoor air quality.

[0085] Reference Figure 3 , the operation method of the advanced decision-making module based on the fusion of fuzzy logic and expert system is as follows: In step SD10, the fuzzy logic system performs fuzzification processing on the input data.

[0086] Fuzzy logic system: A mathematical tool that simulates the way human thinking processes uncertainty and fuzzy information. It can convert precise input data into fuzzy concepts that conform to human language habits, thereby solving problems that are difficult to accurately model in complex systems.

[0087] Fuzzy processing: converting clear and precise values ​​or information into elements of fuzzy sets, describing them with language variables and membership degrees, so that computers can process data in a way similar to human cognition.

[0088] Specific process: When the pre-processed voice command key information, emotional characteristic data, real-time air quality data and future air quality trend information are input into the fuzzy logic system, the system will process it according to the pre-defined fuzzy sets and membership functions.

[0089] Taking the PM2.5 concentration in real-time air quality data as an example, suppose the PM2.5 concentration is divided into three fuzzy sets: "low", "medium" and "high". If the current PM2.5 concentration is 30μg / m³, the system calculates the membership of the concentration in the "low" set as 0.8, the membership in the "medium" set as 0.2, and the membership in the "high" set as 0 according to the membership function. In this way, the precise value 30μg / m³ is fuzzified into the language variables "low (membership 0.8), medium (membership 0.2)".

[0090] For emotional feature data, such as the "pleasure" value obtained through facial expression recognition, it will also be mapped to fuzzy sets such as "very happy", "pleasant", "average", "unpleasant", and "very unpleasant" in a similar way, and the corresponding membership will be determined.

[0091] After such fuzzification processing, the originally precise data is converted into fuzzy information that is easy to understand and process, laying the foundation for subsequent reasoning based on fuzzy logic.

[0092] Step SD20, reasoning is performed based on a rule base preset in the expert system, which contains strategies for air purification and environmental regulation in various situations.

[0093] Expert system: A knowledge-based intelligent system that stores the expertise and experience of domain experts in a specific form and simulates the thinking process of experts to solve problems.

[0094] Rule base: The core component of the expert system is a collection of "IF-THEN" rules. These rules are based on professional knowledge and practical experience in the field of air purification, covering strategies for air purification and environmental regulation in various situations.

[0095] Reasoning: The expert system searches for matching rules in the rule base based on the input fuzzy data and draws conclusions according to a certain reasoning mechanism.

[0096] The specific process is as follows: When the fuzzy logic system completes the fuzzification of the input data, the fuzzy information is passed to the expert system.

[0097] The expert system searches for matching rules one by one in the rule base based on the fuzzified key information of voice commands, emotional feature data, real-time air quality data, and future air quality trend information. For example, if there is a rule such as "IF (real-time PM2.5 concentration is high) AND (user emotion is irritable) THEN (increase the air purifier wind speed and turn on the deodorization function)".

[0098] The system determines whether the current input data meets the prerequisites of the rule. If the real-time PM2.5 concentration is indeed "high" after fuzzy processing, and the user's emotion is "irritated" through emotional feature data analysis, then this rule is triggered.

[0099] According to the preset reasoning mechanism, which may be based on forward reasoning (inferring conclusions from conditions), reverse reasoning (finding conditions from goals) or mixed reasoning, the system draws corresponding conclusions based on the triggered rules, that is, the air purifier needs to increase the wind speed and turn on the deodorization function, providing a basis for the subsequent generation of detailed equipment control instructions.

[0100] Step SD30, after fuzzy logic reasoning, the decision module will generate detailed equipment control instructions, which cover the adjustment of operating parameters of the air purifier and the linkage control of other smart devices.

[0101] Fuzzy logic reasoning: The process of calculating and deducing fuzzified input data based on fuzzy logic rules to draw reasonable conclusions. In this scenario, the fuzzy logic system and expert system rule base are used to conduct a comprehensive analysis of various types of information related to air purification.

[0102] Decision-making module: A functional unit that integrates fuzzy logic reasoning results and generates specific operation instructions based on the overall system goals. It is the core part of the entire intelligent multi-dimensional air purification system to make decisions.

[0103] Device control instructions: A set of commands that clearly instruct a device to perform specific operations, detailing the parameters that the device needs to adjust or the actions that need to be performed.

[0104] Operating parameter adjustment: For air purifiers, change the parameter settings that determine their operating status and purification effect, such as wind speed, air volume, working mode, filter replacement cycle, etc.

[0105] Smart device linkage control: enables the air purifier to work together with other related smart devices (such as smart windows, fresh air fans, air conditioners, etc.), uniformly coordinate equipment operation according to the environment and user needs, and achieve more optimized indoor environment regulation.

[0106] The specific process is as follows: After the fuzzy logic reasoning is completed, the conclusions drawn from the reasoning are passed to the decision module. For example, if the reasoning result shows that the current air quality is poor and the user has a demand for higher wind speed, and considering the indoor temperature and humidity conditions, the decision module will integrate this information.

[0107] For air purifiers, the decision module generates specific operating parameter adjustment instructions, such as adjusting the wind speed from the current medium to high, changing the working intensity of the filter to improve purification efficiency, etc.

[0108] Considering the coordination with other smart devices, if the indoor air quality is poor and the carbon dioxide concentration is high, the decision module may generate instructions to start the new air blower to increase the introduction of fresh air; or if the outdoor air quality is good, it will open the smart window for natural ventilation. These instructions are summarized into detailed device control instructions to accurately control each device to meet the indoor environment optimization and user needs.

[0109] Reference Figure 4 A smart multi-dimensional air purification cloud intelligent control method also includes steps after obtaining the voice control instructions of the user in the room about the air purifier, which are as follows: Step Sa00, analyzing whether the acquired voice control instruction comes from multiple users. If not, executing step Sb00; if yes, executing step Sc00.

[0110] Voice control instructions: Users give voice commands to the air purification system to control the operation of the air purifier, such as adjusting the wind speed, turning on and off equipment, etc.

[0111] Multiple users: Two or more users who issue control commands to the air purifier in the same space.

[0112] The analysis method is as follows: The system uses voice recognition technology and related identity recognition functions to make judgments. For example, through voice and voiceprint recognition technology, the received voice command is compared with the voiceprint features of multiple users registered in the system. If multiple different voiceprint features can be matched, it means that the voice control command comes from multiple users; if only one voiceprint feature is matched or no registered voiceprint can be matched (it can be regarded as a single unknown user), it is determined that the voice control command does not come from multiple users.

[0113] Step Sb00, maintain the original settings.

[0114] When the system determines that the voice control instructions obtained do not come from multiple users, the air purifier and related smart devices maintain the current operating status and parameter settings without making additional adjustments.

[0115] In step Sc00, each user is regarded as an intelligent agent, and the voice commands of each user are parsed to extract key information. Then, the parsed key information together with the room space layout, air quality data of each corner, and air quality trend information within a preset time range in the future are used as the input of the multi-agent reinforcement learning algorithm.

[0116] Agent: In a multi-agent system, it can be regarded as an individual with autonomous decision-making ability, able to perceive the environment and take actions to achieve goals. Here, each user is regarded as an agent because each user has different needs and decisions for the air environment.

[0117] Voice command analysis: Use speech recognition and natural language processing technology to convert the user's voice into text that the computer can understand, and extract key operational intentions, such as adjusting the wind speed, changing the purification mode, etc.

[0118] Key information: The core operational requirements extracted from the voice command to clarify the user’s specific expectations for the air purifier or other smart device.

[0119] Multi-agent reinforcement learning algorithm: A machine learning algorithm in which multiple agents interact in a shared environment and optimize their decision-making strategies to achieve common or individual goals by continuously trying and learning from environmental feedback.

[0120] The specific process is as follows: When it is determined that the voice control instructions come from multiple users, the system abstracts each user into an intelligent agent.

[0121] The system analyzes each user's voice commands. For example, if a user says "increase the speed of the air purifier and turn on the humidifier", the system extracts key information such as "increase the speed of the air purifier" and "turn on the humidifier" through voice recognition and natural language processing technology.

[0122] At the same time, the system obtains room space layout information, such as the shape, size, door and window positions of the room, which affects air flow and purification effects.

[0123] Collect air quality data from every corner to fully understand the distribution of indoor air quality.

[0124] Combined with the air quality trend information within a preset future time range predicted by the spatiotemporal data analysis algorithm.

[0125] The above analyzed key information, room space layout, air quality data in each corner, and information on future air quality trends are integrated as input to the multi-agent reinforcement learning algorithm to provide comprehensive information for subsequent agent decision-making.

[0126] In step Sd00, each intelligent agent uses the decision-making model in the multi-agent reinforcement learning algorithm to select matching actions based on its own state information and the acquired environmental information. The scope of actions includes but is not limited to adjusting the operating parameters of the air purifier and controlling the switches or parameters of other intelligent devices.

[0127] Agent state information: refers to the relevant data representing the specific needs and preferences of each agent (i.e. each user), such as the user's sensitivity to air quality, the habitual air purifier operation mode, etc. This information determines the agent's decision-making tendency in a specific environment.

[0128] Environmental information: covers external environmental conditions such as the room space layout, air quality data in each corner, and the trend of air quality within a preset time range in the future. The intelligent agent uses this information to judge the current environment and make decisions.

[0129] Decision model in multi-agent reinforcement learning algorithm: A model based on the multi-agent reinforcement learning framework, which can calculate the expected return of each possible action based on the state information and environmental information of the agent by learning from past experience, thereby guiding the agent to make the best decision.

[0130] Action: The actions taken by the agent to optimize the environment to meet its own needs. In this scenario, it mainly involves controlling the air purifier and other smart devices.

[0131] The specific process is as follows: Each agent first collects its own status information. For example, a user is accustomed to setting the air purifier to low wind speed and silent mode when the air quality reaches a certain standard. This preference is part of the agent's status information.

[0132] At the same time, the intelligent agent obtains shared environmental information, such as the high PM2.5 concentration in a corner of the room and predicts that the air quality will deteriorate in the future.

[0133] The agent then inputs its own state information and environment information into the decision model in the multi-agent reinforcement learning algorithm. Based on the learning mechanism of the algorithm, the model calculates the expected return of each possible action in the current situation based on the rewards obtained from taking different actions in similar situations in the past (for example, after turning up the air purifier's wind speed, user satisfaction increased, which is a positive return).

[0134] Based on the expected return, the agent chooses the action with the highest expected return. For example, an agent may choose to increase the speed of the air purifier, or control the smart window to close to prevent outdoor pollutants from entering; or control the fresh air fan to increase the amount of fresh air introduced, etc., in order to improve the environment and meet its own needs.

[0135] A smart multi-dimensional air purification cloud intelligent control method also includes the following steps where each intelligent agent uses a decision model in a multi-agent reinforcement learning algorithm to select a matching action based on its own state information and the acquired environmental information, as follows: Step Se00, using various sensors distributed indoors to collect environmental change data in real time, and at the same time obtaining user satisfaction feedback on the current environment regularly collected through user terminal devices.

[0136] Various types of sensors distributed indoors: refers to various types of sensors installed at different locations indoors to monitor environmental parameters. For example, air quality sensors monitor the concentration of pollutants in the air, temperature and humidity sensors detect indoor temperature and humidity, and light sensors measure indoor light intensity. These sensors can sense environmental changes in real time and transmit data to the system.

[0137] Environmental change data: Information collected by various sensors that reflects real-time changes in the indoor environment, such as fluctuations in air quality indicators, changes in temperature and humidity, changes in light intensity, etc. These data can help the system understand environmental dynamics and provide a basis for subsequent decision-making.

[0138] User terminal devices: Electronic devices used by users in daily life, such as smartphones, tablets, etc., which interact with the air purification system by installing specific applications.

[0139] User satisfaction feedback on the current environment: Users use terminal devices to provide subjective evaluations of the current indoor environment (including air quality, comfort, etc.). Feedback can be provided by scoring (e.g., 1-5 points), describing in text, or selecting preset options (e.g., satisfied, average, dissatisfied).

[0140] The specific process is as follows: Sensors at various locations in the room work continuously to capture environmental information in real time. For example, the air quality sensor detects and records the concentration data of pollutants such as PM2.5 and formaldehyde in the air at regular intervals (such as every minute), and the temperature and humidity sensor simultaneously monitors and transmits temperature and humidity values. These data are transmitted to the system's data collection module in real time via wired or wireless means.

[0141] At the same time, the system regularly pushes satisfaction surveys to users through the application on the user's terminal device. For example, every once in a while (such as once a day), when the user opens the application, a simple questionnaire will pop up, asking the user about the current indoor air quality, temperature, humidity and other environmental factors. The user evaluates and submits according to their own feelings. The system collects these feedback data and integrates it with the environmental change data collected by the sensor to provide comprehensive information for subsequent analysis.

[0142] Step Sf00, summarizing the collected environmental change data and user satisfaction feedback on the current environment to form environmental feedback information.

[0143] Environmental change data: Information about the indoor environmental status collected in real time by various indoor sensors, covering dynamic data on air quality (such as PM2.5, formaldehyde concentration), temperature and humidity, light intensity, etc.

[0144] User satisfaction feedback on the current environment: The user's subjective feelings about the overall condition of the current indoor environment expressed through the terminal device, presented in the form of ratings, text descriptions or option selections.

[0145] Environmental feedback information: A comprehensive information collection formed by integrating environmental change data and user satisfaction feedback, which is used to reflect the actual state of the indoor environment and the user's acceptance of this state, providing a key basis for subsequent system decision-making.

[0146] The specific process is as follows: The system first organizes the environmental change data collected from various sensors. For example, the air quality sensor data, temperature and humidity sensor data, and light sensor data at different time points and locations are arranged in order according to time series and spatial locations to form an environmental change data group.

[0147] At the same time, collect satisfaction feedback information from user terminal devices. If there are multiple user feedbacks, summarize them. For example, collect the rating of user A, the text description of user B, and the satisfaction option selected by user C.

[0148] Then, the sorted environmental change data is integrated with user satisfaction feedback. Using time as a clue, the environmental change data within the same time period is associated with the corresponding user satisfaction feedback. For example, the environmental data such as air quality, temperature and humidity at a certain moment are integrated with the satisfaction feedback submitted by users near that moment, and finally form environmental feedback information that fully reflects the actual situation of the indoor environment and the user's feelings, providing rich data support for the subsequent reward and punishment judgment and strategy optimization of the intelligent agent.

[0149] Step Sg00, based on the environmental feedback information and in accordance with the pre-set reward and punishment rules, rewards and punishments are determined for each intelligent agent.

[0150] Environmental feedback information: It integrates indoor environmental change data (such as air quality, temperature and humidity, etc.) and comprehensive information on user satisfaction feedback on the current environment, fully reflecting the actual environmental conditions and user feelings.

[0151] Reward and Penalty Rules: A set of pre-defined criteria for judging the impact of the agent's actions on the environment and user satisfaction. Actions that meet user expectations and improve environmental conditions are rewarded, while actions that do not are punished to guide the agent to optimize its decision-making.

[0152] Agent: Here, each user is regarded as an agent, and each agent affects the operation of the air purification and environmental conditioning system through its own decisions.

[0153] The specific process is as follows: After the system obtains environmental feedback information, it compares each data in it with the conditions in the reward and penalty rules one by one. For example, if the rule is set to give rewards when the air quality improves and the user satisfaction score reaches 4 points or above, the system will check whether the air quality data in the environmental feedback information shows an upward trend and check the user satisfaction score at the same time.

[0154] If the air quality data shows that the PM2.5 concentration has decreased, the content of harmful gases has decreased, and the user satisfaction score reaches 4 points or higher, the corresponding agent will be rewarded. The reward may be in the form of increasing the weight of the agent in the system decision, or providing some virtual rewards, such as points, to the user corresponding to the agent.

[0155] On the contrary, if the air quality deteriorates and the user satisfaction score is low, such as less than 3 points, the agent will be punished according to the rules. The punishment may be manifested as reducing its influence in system decision-making or reducing the right to use certain specific functions. Through this reward and punishment judgment mechanism, the agent is encouraged to continuously adjust its decision-making to achieve better environmental regulation effects and user satisfaction.

[0156] In step Sh00, the intelligent agent uses the learning mechanism of the reinforcement learning algorithm to update its own behavior strategy based on the reward or punishment feedback obtained.

[0157] Reinforcement learning algorithm: A machine learning algorithm that learns the optimal behavior strategy from the reward or penalty signal fed back by the environment through the interaction between the agent and the environment. The agent takes actions in the environment, and the environment gives rewards or penalties based on the results of the actions. The goal of the agent is to maximize the long-term cumulative rewards through continuous attempts.

[0158] Learning mechanism: The method and rules used in reinforcement learning algorithms to update the agent's behavior strategy. It adjusts the probability or method of the agent choosing different actions in different states based on the reward or penalty feedback received by the agent.

[0159] Behavioral strategy: The rules or methods by which an agent chooses actions in a specific state, which determines how the agent makes decisions when faced with various environmental situations.

[0160] The specific process is as follows: When the agent receives reward or penalty feedback, the learning mechanism of the reinforcement learning algorithm kicks in. For example, if the agent is rewarded for increasing the wind speed of the air purifier, which improves the air quality and achieves higher user satisfaction, then the learning mechanism will strengthen the agent's tendency to choose to increase the wind speed in similar environmental conditions.

[0161] The specific implementation method is usually to adjust the parameters in the behavior strategy. Taking the policy gradient-based reinforcement learning algorithm as an example, the algorithm will calculate the policy gradient based on the reward signal and update the policy parameters in the direction of maximizing future rewards, so that the agent is more likely to choose actions that can bring rewards when it encounters similar environmental states in the future.

[0162] On the contrary, if the agent's actions lead to environmental degradation or reduced user satisfaction and are punished, the learning mechanism will reduce the probability of the agent taking that action in similar situations, prompting the agent to explore other actions that may improve the environment and increase user satisfaction, and gradually optimize its own behavioral strategy to better adapt to the environment and meet user needs.

[0163] The agent uses the learning mechanism of the reinforcement learning algorithm to update its own behavior strategy based on the reward or penalty feedback obtained, including: Step Sh10, analyzing whether relevant physiological data that feeds back the user's emotions is obtained.

[0164] Step Sh20, if no, the agent uses the learning mechanism of the reinforcement learning algorithm to update its own behavior strategy based on the reward or punishment feedback obtained.

[0165] Step Sh30, if yes, then based on the acquired physiological data related to the feedback of the user's emotions, a multimodal emotion fusion and feedback algorithm is used to analyze and acquire emotion-related data, and feature extraction is performed to acquire emotion feature data.

[0166] Overview of Multimodal Emotion Fusion and Feedback Algorithm: This algorithm aims to comprehensively process physiological data of multiple different modalities to achieve a more accurate analysis of emotions. It not only considers the emotional information contained in each data modality, but also focuses on the relationship and complementarity between different modal data. By integrating this information, it can more comprehensively understand user emotions, and give corresponding feedback based on the analysis results to guide the subsequent actions of the intelligent agent.

[0167] Data preprocessing is as follows: 1. For different types of physiological data, standardization is first performed. For example, heart rate data may range from tens to more than a hundred, while skin electrical response data may range from a few microsieverts to tens of microsieverts. Through standardization, these data are unified into similar numerical ranges for subsequent analysis. 2. Remove noise and outliers. Physiological data may be subject to various interferences during the collection process, such as the impact of motion artifacts on heart rate monitoring. Through filtering, statistical tests and other methods, these noise and outliers are identified and removed to improve data quality.

[0168] The analysis and acquisition of emotion-related data are as follows: 1. Based on heart rate data: The algorithm analyzes the change pattern of heart rate. For example, a sudden increase in heart rate may be related to emotions such as tension and excitement, while a steady and low heart rate may indicate relaxation. Calculate heart rate variability indicators, such as the standard deviation of adjacent heartbeat intervals (SDNN), which can reflect the dynamic changes in the autonomic nervous system's regulation of the heart and thus be associated with emotional states. 2. Based on skin galvanic response data: Skin galvanic response is closely related to the degree of emotional arousal. When emotional arousal increases, skin conductivity usually increases. The algorithm analyzes the amplitude change and frequency of skin galvanic response to determine the level of emotional arousal. 3. Based on facial expression data (if any): Identify the movement of key facial muscles, such as the contraction and relaxation of the corrugator supercilii, orbicularis oculi, and zygomatic major. Different muscle combinations correspond to different basic emotions. For example, frowning and squinting may indicate dissatisfaction or worry, and raising the corners of the mouth may indicate pleasure. By analyzing the intensity and duration of these muscle movements, facial expression information related to emotions is obtained.

[0169] Feature extraction obtains emotional feature data as follows: 1. Time domain feature extraction: For data that changes over time, such as heart rate and skin electrical response, extract time domain features such as mean, variance, peak, etc. The mean reflects the average level of physiological indicators over a period of time, the variance reflects the degree of fluctuation of the data, and the peak may be related to the occurrence of specific emotional events. 2. Frequency domain feature extraction: Convert time domain data to the frequency domain through methods such as Fourier transform, and extract features such as power spectral density of different frequency components. For example, the power ratio of heart rate variability in different frequency bands (such as low frequency band and high frequency band) is related to the balance of sympathetic and parasympathetic nerve activity in the autonomic nervous system, and can be used as part of the emotional characteristics. 3. Feature selection based on machine learning: Use machine learning algorithms (such as principal component analysis PCA, linear discriminant analysis LDA, etc.) to screen and reduce the dimensionality of the extracted features. PCA can convert high-dimensional feature data into a set of linearly unrelated principal components, retaining the main information while reducing the data dimension; LDA selects the most discriminative features based on the differences between different emotion categories, making the extracted emotion feature data more representative and discriminative, which is convenient for subsequent emotion analysis and intelligent agent decision-making.

[0170] Step Sh40, assigning weights according to the contribution of different modal data to emotion judgment, fusing the weighted multimodal data, and inputting them into a deep learning-based emotion analysis model to judge the emotional state of each user.

[0171] The weight can be determined by a data-driven method, as follows: Analyze a large amount of multimodal physiological data with labeled emotional states. For example, collect thousands of sets of samples that contain heart rate, galvanic skin response, facial expression and other data and have known corresponding emotional labels (such as happiness, sadness, anger, etc.). Use machine learning algorithms, such as regression analysis or decision tree algorithms, to calculate the correlation between each modal data and the final emotional label. The higher the correlation, the greater the contribution of the modal data to emotional judgment, and the higher the weight is assigned accordingly. For example, it was found through analysis that the correlation between facial expression data and the actual emotional label when judging "happy" and "angry" emotions is as high as 0.8, while the correlation of heart rate data is 0.6. In this case, the weight of facial expression data can be set to 0.5, and the weight of heart rate data can be set to 0.3 (the sum of the weights is 1, and the remaining weights are assigned to other modal data in the same way).

[0172] Weighted fusion of multimodal data: For each modality of emotion-related data, multiply each eigenvalue by the corresponding weight. For example, suppose that after feature extraction, facial expression data has an eigenvalue of 0.8, and its weight is 0.5; and a eigenvalue of heart rate data is 0.6, and its weight is 0.3. Then the weighted value of facial expression data is 0.4, and the weighted value of heart rate data is 0.18.

[0173] The weighted eigenvalues ​​of all modal data are combined to form a comprehensive feature vector, which integrates the information of multiple modal data and takes into account the relative importance of each modal data to emotion judgment.

[0174] Enter the deep learning sentiment analysis model: Deep learning model selection: Convolutional neural network (CNN), recurrent neural network (RNN) or its variants are usually used, such as long short-term memory network (LSTM), gated recurrent unit (GRU), etc. For example, for processing facial expression image data with spatial structure, CNN can effectively extract local features in the image; while for processing heart rate and skin electrical response data with time series characteristics, RNN and its variants can better capture the information of data changes over time.

[0175] Model training: Before using weighted fusion data, the model needs to be trained on a large amount of multimodal data with labeled emotions. The model parameters (such as the weights and biases of the neural network) are continuously adjusted through the back-propagation algorithm to minimize the error between the emotional state predicted by the model and the actual labeled emotional label. For example, the cross-entropy loss function is used to measure the difference between the predicted value and the true value, and the model parameters are updated through optimization algorithms such as stochastic gradient descent. After multiple iterations of training, the model can accurately learn the mapping relationship between emotional features and emotional states from the input data.

[0176] Emotional judgment: The weighted fused feature vector is input into the trained deep learning model. The model calculates and infers based on the learned patterns, and outputs the possible emotional state of each user, such as "happy", "sad", "calm" and other categories, or gives the probability value of each emotional state, thereby providing a quantitative basis for the intelligent agent to understand the user's emotional state, so that the intelligent agent can adjust its behavior strategy more accurately to meet user needs.

[0177] Step Sh50, count the number and proportion of intelligent agents that want to reduce noise and require increased wind speed to deal with pollution, and analyze the preliminary strategies associated with intelligent agents whose proportion exceeds a preset threshold.

[0178] In a multi-agent system, different agents (representing different users) have different needs for air purification and environmental regulation. Counting the number and proportion of agents with two representative needs, namely, those who want to reduce noise and those who require increased wind speed to deal with pollution, can help understand the demand trends of user groups in these two key aspects. Analyzing the preliminary strategies associated with agents whose proportion exceeds the preset threshold can help discover the environmental regulation methods expected by most users, provide direction for subsequent strategy optimization, and make the final regulation strategy more in line with the interests of most users.

[0179] The number and proportion of statistical agents are as follows: Determine the criteria for judging demand: The system first determines whether the agent wants to reduce noise or request to increase the wind speed to deal with pollution. For example, by analyzing the actions previously chosen by the agent (such as repeatedly choosing to reduce the wind speed of the air purifier, which may indicate a desire to reduce noise; frequent requests to increase the wind speed and air quality data showing high pollution indicators may indicate a request to increase the wind speed to deal with pollution), or directly analyzing the relevant voice commands issued by the agent (such as "turn up the wind speed, the air is too dirty" and "turn down the volume") to determine its needs.

[0180] Count and calculate the proportion: Traverse all agents and count the agents that meet the two requirements of reducing noise and increasing wind speed to deal with pollution.

[0181] The preliminary strategies associated with agents whose proportion exceeds the preset threshold are as follows: Set a preset threshold: Based on actual application scenarios and experience, a percentage threshold (such as 30%) is set in advance. This threshold is used to determine whether the number of agents with a certain type of demand is sufficient to affect the overall strategy formulation.

[0182] If the proportion of agents exceeds 30%, that is, the proportion of agents who want to reduce noise exceeds the preset threshold, the system will analyze the actions, related instructions, and environmental feedback of these agents before, and summarize the preliminary strategies associated with them. For example, it may be found that these agents tend to set the air purifier to low wind speed silent mode when the air quality is acceptable, and may also turn on some auxiliary purification equipment (such as small air purifiers with low noise characteristics) to maintain air quality.

[0183] Similarly, if the proportion of agents is less than 30%, we can analyze the behavior patterns of agents that request to increase the wind speed to deal with pollution. We may find that when the air quality detects that the pollution index is rising, these agents will immediately request to adjust the air purifier wind speed to the maximum, and may also want to turn on the fresh air system to enhance air circulation.

[0184] Step Sh60, integrates the sentiment analysis results with the preliminary strategy derived from the agent proportion analysis.

[0185] The fusion method is as follows: Strategy adjustment based on emotion type: Positive emotion: If the emotion analysis results show that most users are in a positive emotion state, and the proportion of agents who want to reduce noise is high (obtained through agent proportion analysis), the low-noise-related adjustment measures are further strengthened in the preliminary strategy. For example, in the selection of air purifier operation mode, priority is given to low-noise modes that can maintain the current air quality; for other equipment that may generate noise, such as air conditioners, their operating power is appropriately reduced or the operating frequency is adjusted to reduce noise. Negative emotion: When it is found that some users are in a negative emotion and the proportion of agents who require increased wind speed to cope with pollution is high, the preliminary strategy is optimized. For example, while increasing the wind speed of the air purifier, according to the special needs of users for the environment that may be reflected in the emotion analysis (such as the desire for rapid air circulation to bring a cool feeling due to irritability), the operating angle and speed of the indoor fan are appropriately adjusted to strengthen the circulation of air, and the indoor temperature is adjusted to create a more comfortable environment.

[0186] Consider the intensity of emotions in strategy weight allocation: Emotional analysis can not only determine the type of emotions, but also the intensity of emotions. For situations with higher emotional intensity, the corresponding strategy adjustment is more forceful. For example, if the user sentiment analysis in a certain area shows strong dissatisfaction, and the proportion of agents in this area who require increased wind speed to cope with pollution is high, then in the fusion strategy, the magnitude of the increase in wind speed will be greater than usual, and the ventilation frequency of the fresh air system may be accelerated, improving air quality more quickly to alleviate users' negative emotions.

[0187] 3. Strategy complementation and synergy: For the non-conflicting parts of the preliminary strategy obtained from the sentiment analysis results and the agent proportion analysis, complementation and synergy are carried out. For example, sentiment analysis found that users were in a bad mood due to air quality issues, while agent proportion analysis showed that many users required increasing wind speed to deal with pollution. At this time, in the fusion strategy, in addition to increasing wind speed, other demands that may be implied in sentiment analysis (such as higher requirements for air freshness) can also be combined to increase the negative ion release of the air purifier, further improve air quality, and meet users' potential needs.

[0188] Forming a fusion strategy: Through the above-mentioned fusion methods, the sentiment analysis results are integrated with the preliminary strategy obtained from the agent proportion analysis to form a fusion strategy that comprehensively considers user sentiment and group needs. This fusion strategy will serve as the basis for subsequent simulation of environmental changes, optimization strategies, and generation of device control instructions to ensure that the final environmental adjustment measures can meet the actual needs of users to the greatest extent and improve user satisfaction with the indoor environment.

[0189] A smart multi-dimensional air purification cloud intelligent control method also includes steps after integrating the sentiment analysis results with the preliminary strategy obtained by the intelligent agent proportion analysis, which are as follows: Step Sh70, inputting current indoor environmental parameters, the existing status of the smart device and the control instructions in the fusion strategy, and simulating the environmental changes after the strategy is implemented.

[0190] Current indoor environmental parameters: covers a wide range of environmental information, such as air quality parameters (PM2.5, PM10, formaldehyde, TVOC and other pollutant concentrations), temperature and humidity, carbon dioxide concentration, light intensity, etc. These parameters reflect the actual conditions of the current indoor environment and are the basic data for simulating environmental changes. For example, a high PM2.5 concentration indicates that the current air quality is poor. The simulation process will take this as a starting point and consider the impact of the control instructions in the fusion strategy on it.

[0191] Current status of smart devices: refers to the current operating status of smart devices related to indoor environment regulation, such as air purifiers, air conditioners, fresh air blowers, smart windows, etc. This includes the on / off status of the device, operating mode (such as the automatic mode, sleep mode, and strong mode of the air purifier), set parameters (such as the set temperature and wind speed of the air conditioner), etc. For example, if the air purifier is currently in automatic mode and the wind speed is medium, this information will affect the initial conditions of the device's environmental regulation in the simulation.

[0192] Control instructions in the fusion strategy: In the strategy based on the fusion of sentiment analysis and agent proportion analysis, the operation instructions for smart devices. For example, the instructions may require the air purifier to be adjusted to a high speed, the fresh air fan to be turned on and set to the maximum ventilation volume, the south-facing smart windows to be closed, etc. These instructions are the driving factors for simulating environmental changes.

[0193] The simulation process is as follows: Establish environmental model: Use mathematical models and algorithms to describe the relationship between various factors in the indoor environment and the impact of smart devices on the environment. For example, use the law of conservation of mass and diffusion equations to simulate the spread and purification process of pollutants in the air; use the principle of heat transfer to simulate changes in temperature and humidity; consider the impact of ventilation and air exchange on carbon dioxide concentration, etc.

[0194] Device response simulation: According to the control instructions in the fusion strategy, simulate the response of smart devices. For example, if the instruction requires the air purifier to adjust the wind speed to a high level, the model will calculate its purification effect on indoor air quality based on the performance parameters of the air purifier (such as purification efficiency and air volume at different wind speeds). At the same time, consider the impact of the amount of fresh air introduced after the new fan is turned on on parameters such as indoor air quality and carbon dioxide concentration, as well as the impact of closing smart windows on indoor and outdoor air exchange and temperature.

[0195] Environmental parameter update: In the process of simulating device response, indoor environmental parameters are updated in real time according to the environmental model. For example, as the air purifier purifies pollutants, the PM2.5 concentration gradually decreases; after the new fan introduces fresh air, the carbon dioxide concentration decreases; the operation of the air conditioner changes the indoor temperature and humidity, etc. Through continuous iterative calculations, the dynamic changes of indoor environmental parameters within a certain period of time as the fusion strategy is implemented are simulated.

[0196] Simulation result presentation: The simulation results are presented in an intuitive way, such as generating a chart showing the change of environmental parameters over time (including the change curves of air quality indicators, temperature and humidity, and carbon dioxide concentration over time), showing how the indoor environment will evolve after the fusion strategy is implemented. These results provide a clear reference for the subsequent optimization of the fusion strategy based on preset goals and historical experience.

[0197] Step Sh80, based on the simulation results and the preset optimization goals, and referring to the success and failure experiences of the intelligent agent in similar historical scenarios, the fusion strategy is further optimized.

[0198] The evaluation strategy based on the simulation results is as follows: Compare with preset optimization targets: Compare the simulated environmental change results with the preset optimization targets in detail. The preset optimization targets may include air quality indicators (such as PM2.5 concentration below a certain value, formaldehyde content within a safe range), temperature and humidity comfort range (such as temperature between 22-26 degrees Celsius, humidity between 40%-60%), etc. For example, if the simulation results show that the PM2.5 concentration has decreased after the implementation of the fusion strategy, but is still higher than the preset target value, it means that the strategy may not be effective in purifying air quality.

[0199] Analyze the changing trends of key indicators: not only pay attention to whether the final environmental indicators reach the target, but also analyze the changing trends of key indicators during the simulation process. For example, observe whether the air quality indicators are approaching the target quickly or changing slowly. If the change is slow, it may mean that the response speed of the strategy needs to be improved; for example, whether the temperature and humidity fluctuate excessively during the adjustment process. If so, the control logic of related smart devices may need to be adjusted.

[0200] The following are some historical experiences of similar scenarios: Retrieve historical data: The system searches historical data records for cases similar to the current simulation scenario. Similar scenarios are judged based on initial indoor environmental parameters, smart device status, and major environmental issues faced. For example, if the current scenario is a high indoor PM2.5 concentration and a slightly higher temperature, the system will search for historical records of scenarios with similar PM2.5 pollution and high temperatures.

[0201] Extracting successful and failed experiences: Extract the actions taken by the intelligent agent and the results they bring from the historical similar scene records found. Successful experience may be a certain combination of intelligent device operations or control parameter settings that effectively improve environmental conditions; failed experience may be that certain actions cause environmental deterioration or fail to achieve the expected improvement effect. For example, in similar high PM2.5 concentration scenarios in the past, turning on the powerful mode of the air purifier and the fresh air fan at the same time, and adjusting the opening angle of the smart window appropriately, the air quality was quickly improved. This is a successful experience; and there have been cases where the indoor air quality deteriorated due to excessive reliance on air conditioning and cooling and unreasonable ventilation adjustment. This is a failed experience.

[0202] The optimized fusion strategy is as follows: Adjust control instructions: Based on the difference between the simulation results and the preset targets and historical experience, adjust the control instructions in the fusion strategy. For example, if the simulation shows that the air quality is improving slowly, based on historical successful experience, the operating power of the air purifier can be increased or the operating time can be extended, while the operating mode of the fresh air fan can be optimized to purify the air more effectively.

[0203] Optimize control logic: In addition to adjusting the control instructions of specific devices, it is also possible to optimize the control logic of the entire strategy. For example, if it is found that the temperature and humidity adjustment during the simulation is not stable enough, the collaborative control logic between smart devices can be improved to enable air conditioners, humidifiers, dehumidifiers and other devices to better cooperate with each other when adjusting temperature and humidity, avoiding over-adjustment or adjustment lag.

[0204] Comprehensively weigh multiple objectives: During the optimization process, it is necessary to comprehensively consider the balance between multiple optimization objectives. For example, improving air purification efficiency may increase equipment noise and affect the user's acceptance of noise. Therefore, when optimizing the strategy, it is necessary to weigh the relationship between air quality improvement and noise control to ensure that while meeting the main optimization objectives, it does not seriously affect other aspects of user experience. Through the above steps, the fusion strategy is comprehensively and meticulously optimized to make it more in line with actual needs, laying the foundation for the final conversion into specific equipment control instructions and effectively improving the indoor environment.

[0205] Step Sh90, converting the optimized strategy into specific device control instructions.

[0206] Based on the same inventive concept, an embodiment of the present invention provides a smart multi-dimensional air purification cloud intelligent control system, including a memory and a processor, wherein the memory stores a program that can be executed on the processor to implement the following Figures 1 to 4 Procedure for either method.

[0207] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A cloud-based intelligent control method for intelligent multi-dimensional air purification, characterized in that: include: Obtain voice control instructions for the air purifier from users in the room; Based on the room's spatial layout information, historical air quality data, and real-time air quality data, a spatiotemporal data analysis algorithm based on deep learning is used to predict and obtain information on the air quality trend within a preset time range in the future; Analyze the voice control command to extract key information, and analyze the extracted key information, real-time air quality data, and air quality trend information within a preset time range in the future, and input them into the cloud-based intelligent decision-making algorithm model, output the operating parameters that need to be adjusted for the air purifier, and form parameter adjustment decision instructions; The formed parameter adjustment decision instruction is sent to the air purifier terminal through the network; The air purifier adjusts its own operating parameters in response to the parameter adjustment decision instruction.

2. According to claim 1, a smart multi-dimensional air purification cloud intelligent control method is characterized in that: The steps after obtaining the voice control instructions of the user in the room regarding the air purifier are also included, which are as follows: Analyze whether relevant physiological data related to user emotions is obtained; If not, maintain the original setting; If yes, based on the obtained physiological data related to the user's emotions, a multimodal emotion fusion and feedback algorithm is used to analyze and obtain emotion-related data, and feature extraction is performed to obtain emotion feature data; The pre-processed voice command key information, emotional characteristic data, real-time air quality data and future air quality trend information are input together into an advanced decision-making module based on the fusion of fuzzy logic and expert system, and the operating parameters that need to be adjusted for the air purifier are output to form parameter adjustment decision instructions. The parameter adjustment decision instructions are then sent to the air purifier terminal through the network, and the air purifier adjusts its own operating parameters in response to the parameter adjustment decision instructions.

3. According to claim 2, a smart multi-dimensional air purification cloud intelligent control method is characterized in that: The operation method of the advanced decision-making module based on the fusion of fuzzy logic and expert system is as follows: The fuzzy logic system fuzzifies the input data; Reasoning is performed based on the pre-set rule base in the expert system, which contains strategies for air purification and environmental regulation in various situations; After fuzzy logic reasoning, the decision module will generate detailed equipment control instructions, which cover the operating parameter adjustment of the air purifier and the linkage control of other smart devices.

4. A cloud-based intelligent control method for intelligent multi-dimensional air purification according to any one of claims 1 to 3, characterized in that: The steps after obtaining the voice control instructions of the user in the room regarding the air purifier are also included, which are as follows: Analyze whether the acquired voice control instructions come from multiple users; If not, maintain the original setting; If yes, each user is considered as an intelligent agent, and the voice commands of each user are parsed to extract key information. Then, the parsed key information, together with the room space layout, air quality data of each corner, and air quality trend information within a preset time range in the future, are used as the input of the multi-agent reinforcement learning algorithm. Each agent uses the decision-making model in the multi-agent reinforcement learning algorithm to select matching actions based on its own state information and the acquired environmental information. The scope of actions includes but is not limited to adjusting the operating parameters of the air purifier and controlling the switches or parameters of other smart devices.

5. The intelligent multi-dimensional air purification cloud intelligent control method according to claim 4 is characterized in that: It also includes the steps after each agent selects the matching action based on its own state information and the acquired environmental information using the decision model in the multi-agent reinforcement learning algorithm, as follows: Various sensors distributed indoors are used to collect real-time environmental change data, and user satisfaction feedback on the current environment is collected regularly through user terminal devices; Summarize the collected environmental change data and user satisfaction feedback on the current environment to form environmental feedback information; According to the feedback from the environment, each agent is rewarded or punished according to the pre-set reward and punishment rules; The intelligent agent uses the learning mechanism of the reinforcement learning algorithm to update its own behavior strategy based on the reward or punishment feedback obtained.

6. The intelligent multi-dimensional air purification cloud intelligent control method according to claim 5, characterized in that: The agent uses the learning mechanism of the reinforcement learning algorithm to update its own behavior strategy based on the reward or penalty feedback obtained, including: Analyze whether relevant physiological data related to user emotions is obtained; If not, the agent uses the learning mechanism of the reinforcement learning algorithm to update its behavior strategy based on the reward or punishment feedback obtained; If yes, based on the obtained physiological data related to the user's emotions, a multimodal emotion fusion and feedback algorithm is used to analyze and obtain emotion-related data, and feature extraction is performed to obtain emotion feature data; Assign weights based on the contribution of different modal data to emotional judgment, fuse the weighted multimodal data, and input them into the deep learning-based emotional analysis model to determine the emotional state of each user; Count the number and proportion of agents that want to reduce noise and increase wind speed to deal with pollution, and analyze the preliminary strategies associated with agents whose proportion exceeds the preset threshold; Combine the sentiment analysis results with the preliminary strategy derived from the agent proportion analysis.

7. The intelligent multi-dimensional air purification cloud intelligent control method according to claim 6 is characterized in that: It also includes the steps after fusing the sentiment analysis results with the preliminary strategy derived from the agent share analysis, as follows: Input the current indoor environment parameters, the existing status of the smart devices and the control instructions in the fusion strategy to simulate the environmental changes after the strategy is implemented; According to the simulation results and the preset optimization goals, the fusion strategy is further optimized by referring to the success and failure experience of the intelligent agent in similar historical scenarios; Convert the optimized strategy into specific equipment control instructions.

8. A smart multi-dimensional air purification cloud intelligent control system, characterized in that: It includes a memory, a processor, and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement a smart multi-dimensional air purification cloud intelligent control method as claimed in any one of claims 1 to 7.

Images