Intelligent fan temperature adjustment control system based on AI

Abstract

The invention relates to the technical field of smart home, and discloses an AI-based fan intelligent temperature regulation control system, which comprises a multi-dimensional data acquisition module for acquiring environment temperature, humidity, air quality, illumination intensity and human body state data; the data fusion and preprocessing module is used for carrying out denoising, standardization and time synchronization on the multi-dimensional data and fusing the multi-source data; the AI decision and control module is used for generating a speed regulation and mode control instruction of the fan based on the human body comfort evaluation model and an AI algorithm; and the feedback optimization and execution module is used for executing the control instruction. Through a multi-dimensional data perception and fusion technology, comprehensive monitoring of environment temperature and humidity, air quality and a human body state is realized. The AI intelligent regulation and control algorithm is combined, the fan speed and mode are dynamically adjusted, and the user comfort and energy consumption are balanced; through edge calculation and cloud optimization, real-time response and continuous adaptive optimization of the system are ensured, and the accuracy and stability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the field of smart home technology, and in particular to an AI-based intelligent temperature control system for a fan. Background Art

[0002] At present, fans are widely used in homes, offices, public places and other environments as a common indoor cooling and air circulation device. Traditional fans achieve temperature control by manually adjusting wind speed and operation mode, but most devices only have a fixed wind speed adjustment function and cannot be dynamically adjusted according to ambient temperature, humidity and actual user needs. With the development of smart home technology, some fans with simple automatic adjustment functions have appeared on the market, such as triggering fan start-up through temperature sensors or remotely controlling wind speed through mobile phone apps, but these systems still have great limitations in environmental adaptability, control accuracy and user experience.

[0003] In the prior art, although some smart fans have introduced temperature sensors or humidity sensors to monitor environmental data, they only use simple threshold judgment logic to trigger a fixed wind speed, lack the fusion and analysis of multi-dimensional data, and cannot fully reflect the user's true comfort state. In addition, the existing fan system lacks an effective artificial intelligence algorithm and cannot dynamically optimize the control strategy, resulting in inaccurate wind speed control and difficulty in balancing the relationship between user comfort and energy consumption. At the same time, traditional fan systems are insufficient in real-time response capabilities and cannot quickly adjust the wind speed according to the user's real-time status, affecting the user experience. Therefore, the existing technology cannot meet users' demand for efficient, intelligent, comfortable and energy-saving fan temperature control, and there is an urgent need for a fan control system that can realize multi-dimensional data perception, intelligent control and adaptive optimization. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides an AI-based fan intelligent temperature control system, which solves the problems in the prior art that fan control relies on a single data source, has low adjustment accuracy, and lacks real-time response and adaptive optimization.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an AI-based fan intelligent temperature control system, comprising: Multi-dimensional data acquisition module, used to collect environmental temperature, humidity, air quality, light intensity and human body status data; A data fusion and preprocessing module, used for denoising, standardizing and time synchronizing the multidimensional data, and fusing multi-source data; AI decision-making and control module, used to generate fan speed and mode control instructions based on human comfort evaluation model and AI algorithm; A feedback optimization and execution module, used to execute the control instructions and optimize the AI ​​algorithm based on user feedback data and cloud training results; The fan execution module is used to dynamically adjust the fan speed and operation mode according to the speed adjustment and mode control instructions.

[0006] Preferably, the multidimensional data acquisition module includes: Temperature and humidity sensor, used to collect environmental temperature and humidity data; Air quality sensor, used to detect air quality data such as CO2 concentration and PM2.5 concentration; Light sensor, used to collect ambient light intensity data; Infrared sensors and human body temperature sensors to detect user presence, proximity, and human body surface temperature; Distance sensor, used to measure the distance between the user and the fan.

[0007] Preferably, the data fusion and preprocessing module includes: Data cleaning unit, used to denoise multi-source data and remove outliers; Data normalization unit, used to normalize data of different dimensions; Data time synchronization unit, used to align the timing of various sensor data through timestamps; The data fusion unit is used to perform fusion processing on multi-source data based on weighted fusion and rule engine algorithm.

[0008] Preferably, the AI ​​decision and control module includes: a comfort evaluation unit for evaluating human thermal comfort based on ambient temperature, humidity, wind speed and human body status; The AI ​​decision-making unit is used to generate a speed regulation instruction and a mode switching instruction of the fan according to the thermal comfort evaluation result and the real-time sensor data by using a reinforcement learning algorithm.

[0009] Preferably, the feedback optimization and execution module includes: The fan control execution unit is used to dynamically control the fan motor through PWM signals and adjust the fan speed and operation mode; A user feedback unit, used to receive user operation and feedback data; The cloud optimization unit is used to upload user feedback data and historical sensor data to the cloud for continuous training and model optimization of AI algorithms.

[0010] Preferably, the system includes an edge computing unit for deploying a lightweight AI model on the device side to achieve real-time analysis of sensor data and rapid response to fan control instructions.

[0011] Preferably, the air quality sensor includes a CO2 concentration detection module and a PM2.5 detection module, and the data can be used to automatically trigger a high-speed ventilation mode of the fan.

[0012] Preferably, the AI ​​decision-making unit adopts a reinforcement learning algorithm and uses the balance between fan energy consumption and human comfort as a reward function for optimization training.

[0013] Preferably, the cloud-based optimization unit pushes the updated AI model to the device through periodic model training based on user feedback data and device historical data.

[0014] Preferably, the fan execution module also includes a mode selection unit, which includes an automatic mode, an energy-saving mode and a silent mode, and a user can manually select and adjust the mode through a mobile App.

[0015] The present invention provides an AI-based intelligent temperature control system for fans, which has the following beneficial effects: 1. The present invention deploys multiple types of sensors such as temperature and humidity sensors, air quality sensors, light sensors, infrared human sensors and human temperature sensors to comprehensively collect environmental temperature and humidity, air quality, light intensity and human status data, ensuring that the system obtains rich and accurate environmental information. Combined with the data fusion and preprocessing module, denoising, data standardization, time synchronization and weighted fusion technologies are used to uniformly process multi-source heterogeneous data and generate high-quality data input. Through this multi-dimensional data fusion, the system can reflect the current environment and user status in real time, provide a reliable data basis for intelligent decision-making, and thus achieve precise control of the fan.

[0016] 2. The present invention adopts AI decision-making algorithms, especially the combination of reinforcement learning models and human comfort evaluation models, to evaluate the thermal comfort of the human body based on data such as temperature and humidity, air quality, and human surface temperature. The system dynamically generates fan wind speed, wind direction, and operation mode control instructions through intelligent algorithms, and can achieve a balance between human comfort and fan energy consumption. At the same time, through user feedback mechanisms and cloud model training, the system can continuously and adaptively optimize control strategies, form closed-loop feedback, achieve personalized regulation and long-term performance improvement, and meet the comfort needs of users in different environments.

[0017] 3. The present invention implements real-time reasoning of AI models and rapid execution of fan control instructions by deploying edge computing units on the device side, reducing data transmission delays and ensuring the rapid response capability of the system. Specifically, the fan execution module uses PWM signals to dynamically adjust the wind speed and operating mode, while providing wind speed feedback to ensure the stability and accuracy of control execution. In addition, the system can still operate independently when the network is unstable or disconnected, ensuring the continuity of user experience and the stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a system architecture diagram of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Example: Please refer to the attached Figure 1 , an embodiment of the present invention provides an AI-based fan intelligent temperature control system, comprising: Multi-dimensional data acquisition module, used to collect environmental temperature, humidity, air quality, light intensity and human body status data; A data fusion and preprocessing module, used for denoising, standardizing and time synchronizing the multidimensional data, and fusing multi-source data; AI decision-making and control module, used to generate fan speed and mode control instructions based on human comfort evaluation model and AI algorithm; A feedback optimization and execution module, used to execute the control instructions and optimize the AI ​​algorithm based on user feedback data and cloud training results; The fan execution module is used to dynamically adjust the fan speed and operation mode according to the speed adjustment and mode control instructions.

[0021] In this embodiment, the data fusion and preprocessing module is used to process the multi-source data obtained by the multi-dimensional data acquisition module to ensure the accuracy, consistency and availability of the data, and provide stable and reliable data input for the subsequent AI decision-making and control module.

[0022] Specifically, the data fusion and preprocessing module includes a data cleaning unit, a data standardization unit, a time synchronization unit, and a data fusion unit. Each unit cooperates with each other to form a complete data preprocessing process.

[0023] As an option, the data cleaning unit is mainly used to remove noise and outliers from the multi-source data collected by the sensor.

[0024] In this embodiment, the data collected by the temperature and humidity sensors, air quality sensors, human body status sensors, etc. may contain mutation noise or abnormal values, which may affect the accuracy of the data.

[0025] Exemplarily, the data cleaning unit may use a sliding average filtering method to smooth the sensor data to eliminate high-frequency noise in the data.

[0026] It should be noted that for noise data with random mutation characteristics, the median filtering method can also be used to remove data points that deviate from the middle value to further improve the stability of the data.

[0027] In some embodiments, the system may dynamically adjust the size of the sliding window or the threshold range of the median filter according to the actual application scenario.

[0028] In one possible implementation, the data normalization unit is used to map data of different dimensions into a unified range to facilitate subsequent data fusion and input of AI decisions.

[0029] Specifically, the data output by various sensors in the multi-dimensional data acquisition module have different value ranges, for example: The temperature range of the temperature and humidity sensor output is -20℃60℃, and the humidity range is 0100%; The air quality sensor outputs a CO2 concentration of 400 ppm-5000 ppm and a PM2.5 concentration of 0-500 μg / m³; The light intensity range of the light sensor output is 0~65535 Lux.

[0030] It should be noted that the data standardization unit processes the above data by a normalization method or a mean-variance normalization method to ensure that various types of data are within the same numerical range. For example, the data can be mapped to the range of [0, 1] to achieve data standardization.

[0031] Exemplarily, the time synchronization unit is used to add a unified timestamp to multi-source data to ensure the consistency of the data in the time dimension.

[0032] In this embodiment, the multi-dimensional data acquisition module in the system collects data through different types of sensors, and the data acquisition cycles of the sensors may be different, for example: Temperature, humidity and human body status data are collected once per second; Air quality data is collected every 10 seconds.

[0033] As an option, the time synchronization unit controls the main control unit (such as ESP32 or Raspberry Pi) to add a collection timestamp to each set of data to ensure the validity of the data at the same time point.

[0034] It should be noted that the time synchronization unit can also be combined with data resampling technology to interpolate or fill in data with different sampling periods to form a data stream with a unified time series.

[0035] In this embodiment, the data fusion unit is the core of the data fusion and preprocessing module, which is used to fuse the cleaned and standardized multi-source data to generate a unified data set that can be input into the AI ​​model.

[0036] Specifically, the data fusion unit adopts the weighted fusion method and the rule engine fusion method to realize the comprehensive processing of data.

[0037] In a possible implementation, the weighted fusion method performs weighted summation processing on multi-dimensional data by assigning different weight coefficients to different types of data.

[0038] It should be noted that the weight coefficient can be dynamically adjusted according to the sensor type, data real-time and confidence.

[0039] Exemplarily, air quality data has a higher weight when CO2 concentration is higher, while human body temperature data has a higher weight when the human body is close to a fan.

[0040] As an option, the rule engine fusion method processes multi-source data based on preset logical rules, for example: When the ambient temperature is higher than 28°C and the human state sensor detects the approach of a user, it outputs a high wind speed control command for the fan; When the light sensor data is lower than the set threshold and the current time is at night, a silent mode control command for the fan is output; When the PM2.5 concentration in the air quality data exceeds the set upper limit, the fan's high-speed ventilation mode is triggered.

[0041] It should be noted that the rule engine in the data fusion unit can be implemented through logical judgment and simple decision trees to ensure the interpretability and real-time performance of the fusion results.

[0042] In some embodiments, in order to further improve the stability and reliability of data fusion, the data fusion unit can be combined with a Kalman filter algorithm to dynamically estimate and correct errors in the data collected by the sensor to ensure that the output data is smoother and more accurate.

[0043] It can be understood that the data fusion and preprocessing module plays a role in ensuring data flow and data reliability in the present invention.

[0044] Through data cleaning, standardization, time synchronization and data fusion processing, the data such as ambient temperature, humidity, air quality, light intensity and human body status obtained by the multi-dimensional data acquisition module are converted into an input data set in a unified format, providing high-quality data input for the AI ​​decision-making and control module.

[0045] As a possible extension, the data fusion unit can further combine real-time data with historical data to achieve short-term trend analysis and abnormal data detection through a sliding time window to improve the robustness and adaptability of the system.

[0046] In summary, the data fusion and preprocessing module in this embodiment ensures the accuracy and consistency of input data through the synergy of data cleaning, standardization, time synchronization and fusion algorithm, and provides reliable data support for the intelligent control of the fan.

[0047] In this embodiment, the decision and control module is used to make intelligent decisions on human comfort assessment and fan control strategy based on multi-dimensional data input, and output wind speed, wind direction and mode control instructions to the execution module to achieve dynamic intelligent regulation of the fan.

[0048] As an option, the decision-making and control module includes a human comfort evaluation unit and an AI decision-making unit, which work together to realize the complete processing flow from data input to control command output.

[0049] Specifically, the human comfort evaluation unit is used to evaluate the thermal comfort of input data such as ambient temperature, humidity, wind speed, and human body temperature.

[0050] In one possible implementation, the human comfort evaluation is based on the PMV-PPD (predicted mean comfort and probability of thermal discomfort) model, which generates the user's comfort evaluation result by mapping parameters such as temperature, humidity, and human body status to comfort indicators.

[0051] Exemplarily, human body temperature is collected by an infrared sensor, humidity data is provided by a humidity sensor, and wind speed is fed back by the current fan speed.

[0052] It should be noted that the output results of the PMV model can be divided into multiple levels, such as "hot", "slightly hot", "comfortable", "slightly cool", etc., corresponding to different fan speed control instructions.

[0053] As an option, the human comfort evaluation unit can dynamically adjust the comfort assessment results based on real-time environmental data and user surface temperature data, combined with user historical comfort feedback.

[0054] In this embodiment, the AI ​​decision unit is used to generate fan speed regulation and mode control instructions based on human comfort evaluation results and real-time sensor data.

[0055] In one possible implementation, the AI ​​decision-making unit uses a reinforcement learning (RL) algorithm to achieve a balance between human comfort and fan energy consumption through dynamic learning control strategies.

[0056] Specifically, reinforcement learning algorithms are designed with the following elements: State space: Input parameters include current temperature, humidity, light intensity, air quality, human body surface temperature, and user distance.

[0057] Action space: Output parameters include fan speed (such as high, medium, and low) and fan mode (such as automatic mode, energy-saving mode, and silent mode).

[0058] Reward mechanism: The reward function is set based on human comfort index and fan energy consumption. The system continuously optimizes the fan control strategy by evaluating the balance between comfort and energy consumption.

[0059] It should be noted that the training of the reinforcement learning model can be completed in the cloud, and the trained lightweight model is deployed on the device side for real-time reasoning to ensure the system's rapid response capability.

[0060] In some embodiments, the AI ​​decision-making unit can also be combined with a simple logic rule engine as a supplement to reinforcement learning decision-making. For example: When the temperature is higher than the set threshold and a human body is close to it, the fan speed will be automatically increased; Automatically switch to silent mode when light intensity is low and ambient noise needs to be reduced; When the air quality sensor detects that PM2.5 exceeds the standard, it enters high wind speed ventilation mode.

[0061] As an option, the AI ​​decision unit can receive real-time feedback from users on manual adjustments, record the user's preferred wind speed and mode data, and further use it for model optimization and decision-making refinement.

[0062] For example, the user manually adjusts the fan speed through the mobile app, and the system uses the feedback data as one of the inputs for the next round of reinforcement learning training to improve the personalized control effect.

[0063] In this embodiment, the decision and control module also includes a mode switching unit, which is used to realize automatic switching of the fan operation mode according to the AI ​​decision result and preset conditions.

[0064] Specifically, the operating modes include: Automatic mode: Automatically adjust wind speed and direction according to current environmental conditions and human comfort indicators; Energy-saving mode: prioritizes reducing fan power consumption and runs at a low wind speed while meeting the user's basic comfort needs; Silent Mode: Ideal for nighttime or quiet scenes, it reduces fan noise for low-intrusion operation.

[0065] It should be noted that the mode switching unit can dynamically select the operating mode by combining sensor data and AI output results to adapt to different environments and user needs.

[0066] In one possible implementation, in order to improve the real-time responsiveness of the system, the decision-making and control module can be deployed on the edge computing unit.

[0067] Specifically, the edge computing unit runs a lightweight AI inference model, processes data from the multi-dimensional data acquisition module, and quickly generates control instructions.

[0068] For example, edge computing can use ESP32 or Raspberry Pi platform to reduce data transmission delay through local processing and ensure the real-time performance of fan control.

[0069] It can be understood that the AI ​​decision-making unit and the human comfort evaluation unit in the decision-making and control module work together to form the intelligent control core of the system.

[0070] Through human comfort evaluation, reinforcement learning decision-making and mode switching functions, the system can achieve precise control and adaptive optimization of the fan.

[0071] In summary, the decision-making and control module in this embodiment dynamically generates fan control instructions by combining human comfort assessment with AI decision-making algorithms, thereby realizing intelligent and precise control of the fan.

[0072] Through the synergy of reinforcement learning algorithms and logic rule engines, the system can take into account both user comfort and energy saving, ensuring the system's adaptability in different environments.

[0073] It should be noted that the design of the decision-making and control module plays a key role in the present invention and provides core technical support for the intelligent operation of the fan.

[0074] In this embodiment, the feedback optimization and execution module is used to receive the wind speed, wind direction and operation mode control instructions generated by the AI ​​decision and control module, dynamically execute fan control operations, and continuously optimize system performance through user feedback and cloud data training to achieve a closed control loop.

[0075] As an option, the feedback optimization and execution module includes a fan execution unit, a user feedback unit, an edge computing unit and a cloud optimization unit. The units work together to ensure system operation stability and adaptive optimization capabilities.

[0076] Specifically, the fan execution unit is used to dynamically adjust the operating status of the fan, including wind speed, wind direction and operating mode, according to the control instructions output by the AI ​​decision and control module.

[0077] In a possible implementation, the fan execution unit controls the speed of the fan motor through a PWM (pulse width modulation) signal, and the PWM signal frequency can be adjusted according to actual needs to achieve smooth adjustment of high, medium and low wind speeds.

[0078] As an option, the fan actuator can implement multiple operating modes, including: Automatic mode: The system automatically adjusts wind speed and direction based on environmental data and human comfort evaluation results; Energy-saving mode: prioritizes low-power operation of the fan, reducing energy consumption by reducing wind speed and operating time; Silent mode: Suitable for night scenes, reducing fan noise by controlling the wind speed to the lowest level.

[0079] It should be noted that the fan execution unit can monitor the current wind speed in real time through the built-in fan speed feedback mechanism, and feed back the wind speed information to the AI ​​decision-making and control module to ensure the accuracy of the regulation execution.

[0080] In this embodiment, the user feedback unit is used to collect the user's manual operation data and usage feedback to further optimize the system control strategy.

[0081] Specifically, users can manually adjust the wind speed, wind direction and operating mode through the mobile app or local control panel that comes with the fan, and the system records these operating data in real time.

[0082] In some embodiments, the user feedback unit further includes a status feedback interface for displaying the current operating status and environmental parameters of the fan, including information such as the current wind speed, temperature, humidity, and fan operating mode.

[0083] For example, users can view current temperature and humidity data and fan control results in the mobile App, and provide satisfaction evaluation or manual adjustment operations.

[0084] It should be noted that the data collected by the user feedback unit will serve as one of the important inputs for optimizing the AI ​​model and will be used for subsequent cloud training and strategy updates.

[0085] In this embodiment, the edge computing unit is used to perform real-time inference and execution of the AI ​​model on the device side, reducing data transmission delays and improving the response speed of the system.

[0086] Specifically, the edge computing unit can deploy a lightweight AI model (for example, through TensorFlow Lite model deployment), process sensor data from the multi-dimensional data acquisition module in real time, and generate local decision-making instructions.

[0087] In one possible implementation, the edge computing unit works together with the fan execution unit to ensure that the fan response time is controlled at the millisecond level.

[0088] It is understandable that the edge computing unit can run independently on the device side to ensure that the system can still work normally when the network is unstable or disconnected.

[0089] In this embodiment, the cloud optimization unit is used for long-term performance optimization of the system and iterative update of the AI ​​model.

[0090] Specifically, the cloud optimization unit uploads user feedback data, environmental sensor data and system operation status data to the cloud server, and the cloud performs model training and optimization through big data analysis and reinforcement learning algorithms.

[0091] As an option, the cloud optimization unit adopts a periodic update mechanism to regularly send the latest trained model to the device for replacement, so as to achieve continuous optimization and adaptive update of the AI ​​control strategy.

[0092] It should be noted that the cloud optimization unit can perform cluster analysis on users' historical operation data, identify users' usage preferences, and provide support for personalized system control.

[0093] In some embodiments, the cloud optimization unit can also optimize the basic operating parameters of the system according to regional climate and environmental characteristics. For example, in high humidity areas, the system can adjust parameters according to humidity-sensitive comfort requirements. For example, the feedback optimization and execution modules form a complete closed-loop control mechanism: The AI ​​decision-making and control module outputs control instructions; The fan execution unit performs speed regulation and mode adjustment in real time; The user feedback unit collects operation data and satisfaction information; The edge computing unit ensures the real-time response of the system; The cloud-based optimization unit performs model optimization and strategy updates based on feedback data to complete closed-loop optimization.

[0094] It can be understood that this module ensures efficient and stable operation of the system in different usage scenarios through a combination of fan execution, user feedback and cloud optimization.

[0095] Through real-time collection of user feedback and long-term training in the cloud, the system can achieve personalized and adaptive adjustment of control strategies, thereby improving users' actual experience and system operation effects.

[0096] To summarize, the feedback optimization and execution module in this embodiment realizes closed-loop control and continuous optimization of the system through the coordination of fan dynamic execution, user feedback collection and cloud optimization mechanism, ensures the accuracy and intelligence of fan control results, and provides users with a stable and personalized fan control experience.

[0097] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An AI-based fan intelligent temperature control system, characterized in that: include: Multi-dimensional data acquisition module, used to collect environmental temperature, humidity, air quality, light intensity and human body status data; A data fusion and preprocessing module, used for denoising, standardizing and time synchronizing the multidimensional data, and fusing multi-source data; AI decision-making and control module, used to generate fan speed and mode control instructions based on human comfort evaluation model and AI algorithm; A feedback optimization and execution module, used to execute the control instructions and optimize the AI ​​algorithm based on user feedback data and cloud training results; The fan execution module is used to dynamically adjust the fan speed and operation mode according to the speed adjustment and mode control instructions.

2. The AI-based fan intelligent temperature control system according to claim 1, characterized in that: The multi-dimensional data acquisition module comprises: Temperature and humidity sensor, used to collect environmental temperature and humidity data; Air quality sensor, used to detect air quality data such as CO2 concentration and PM2.5 concentration; Light sensor, used to collect ambient light intensity data; Infrared sensors and human body temperature sensors to detect user presence, proximity, and human body surface temperature; Distance sensor, used to measure the distance between the user and the fan.

3. The AI-based fan intelligent temperature control system according to claim 1 is characterized in that: The data fusion and preprocessing module includes: Data cleaning unit, used to denoise multi-source data and remove outliers; Data normalization unit, used to normalize data of different dimensions; Data time synchronization unit, used to align the timing of various sensor data through timestamps; The data fusion unit is used to perform fusion processing on multi-source data based on weighted fusion and rule engine algorithm.

4. The AI-based fan intelligent temperature control system according to claim 1, characterized in that: The AI ​​decision and control module includes: a comfort evaluation unit for evaluating human thermal comfort based on ambient temperature, humidity, wind speed and human body status; The AI ​​decision-making unit is used to generate a speed regulation instruction and a mode switching instruction of the fan according to the thermal comfort evaluation result and the real-time sensor data by using a reinforcement learning algorithm.

5. The AI-based fan intelligent temperature control system according to claim 1, characterized in that: The feedback optimization and execution module includes: The fan control execution unit is used to dynamically control the fan motor through PWM signals and adjust the fan speed and operation mode; A user feedback unit, used to receive user operation and feedback data; The cloud optimization unit is used to upload user feedback data and historical sensor data to the cloud for continuous training and model optimization of AI algorithms.

6. The AI-based fan intelligent temperature control system according to claim 1, characterized in that: The system includes an edge computing unit for deploying a lightweight AI model on the device side to achieve real-time analysis of sensor data and rapid response to fan control instructions.

7. The AI-based fan intelligent temperature control system according to claim 2, characterized in that: The air quality sensor includes a CO2 concentration detection module and a PM2.5 detection module, and the data can be used to automatically trigger a high-speed ventilation mode of the fan.

8. The AI-based fan intelligent temperature control system according to claim 4, characterized in that: The AI ​​decision-making unit adopts a reinforcement learning algorithm and uses the balance between fan energy consumption and human comfort as a reward function for optimization training.

9. The AI-based fan intelligent temperature control system according to claim 5, characterized in that: The cloud optimization unit pushes the updated AI model to the device through periodic model training based on user feedback data and device historical data.

10. The AI-based fan intelligent temperature control system according to claim 1, characterized in that: The fan execution module also includes a mode selection unit, which includes an automatic mode, an energy-saving mode and a silent mode. Users can manually select and adjust the mode through a mobile App.

Images