Indoor thermal environment control method based on group thermal comfort voting autonomous learning

Abstract

The application discloses an indoor thermal environment control method based on group thermal comfort voting autonomous learning, aiming at realizing personalized indoor thermal environment control through intelligent regulation and control and artificial intelligence technology. The system is composed of three modules of data acquisition, data processing and terminal control. The data acquisition module is responsible for collecting user information, video information and indoor thermal environment data; the data processing module analyzes the data, processes user instructions and optimizes the control strategy through a machine learning algorithm; and the terminal control module adjusts air conditioner parameters according to the optimal strategy. The system can dynamically adjust the indoor thermal environment according to the thermal comfort voting of users and the cold and warm preferences of users. The system can also improve the response speed and control accuracy through wireless communication technology and reduce the maintenance cost. The method provided by the application can maximize the pursuit of the user to the thermal environment comfort, provide the thermal comfort experience of group and personalized users, and also achieve the effect of energy saving and emission reduction.

Description

Technical fields:

[0001] This invention relates to the interdisciplinary fields of intelligent air conditioning control and artificial intelligence, and in particular to an autonomous learning indoor thermal environment control method based on group thermal comfort voting. The aim is to comprehensively consider the thermal comfort needs of different users indoors through artificial intelligence methods, thereby achieving more personalized control. Background technology:

[0002] With the acceleration of urbanization and the improvement of people's living standards, people's requirements for indoor environmental comfort are also increasing. Indoor thermal environment, as one of the important factors affecting human comfort, directly relates to people's quality of life and health. However, traditional indoor thermal environment control methods are mostly based on individual thermal comfort perception or preset temperature and humidity setpoints, ignoring the differences and dynamic changes in the thermal comfort needs of different groups. This makes them inadequate when facing complex and ever-changing indoor environments and diverse user needs. Therefore, research on autonomous learning indoor thermal environment control methods based on group thermal comfort voting is particularly important.

[0003] Traditional indoor thermal environment control typically relies on fixed temperature and humidity setpoints, set by building managers or users themselves. This method ignores individual differences and dynamically changing needs among users. Different groups have varying sensitivities to thermal environments; for example, the elderly, children, and patients have more specific requirements for temperature and humidity. Furthermore, traditional control methods often operate according to preset values, failing to accurately perceive actual user needs and leading to over-adjustment and energy waste.

[0004] Group thermal comfort voting refers to collecting thermal comfort experience data from a certain number of users and using statistical methods to analyze the data to determine the group's thermal comfort needs. This method can more accurately reflect users' actual needs because each user can vote based on their own feelings, thus avoiding the problem that a single set value cannot meet diverse needs. At the same time, group thermal comfort voting can also enable dynamic adjustment of the indoor thermal environment, as the voting results can reflect changes in the indoor environment in real time, and the system can adjust its control strategies accordingly.

[0005] The rapid development of wireless communication technology has provided technical support for the research of this method. Wireless sensor networks can be easily deployed in various corners of an indoor space to collect environmental data in real time and transmit it to a data processing module for analysis. Through the wireless network, users can view the indoor environmental conditions and remotely control equipment such as air conditioners and heating devices anytime, anywhere via mobile devices such as smartphones and tablets. This convenient communication method not only improves the system's response speed and control accuracy but also reduces the system's maintenance and time costs.

[0006] The introduction of autonomous learning capability is a key innovation of this research. Traditional control systems often lack self-learning and optimization capabilities, requiring manual parameter adjustments and strategy optimization. However, the autonomous learning indoor thermal environment control method based on group thermal comfort voting introduces machine learning algorithms. Through the analysis and processing of historical and real-time feedback data, it continuously optimizes the control strategy, improving the system's intelligence level. The purpose of this invention is to provide an autonomous learning indoor thermal environment control method based on group thermal comfort voting, which dynamically adjusts indoor thermal environment parameters by real-time monitoring and analysis of indoor population thermal comfort perception voting data. This adaptive adjustment capability enables the system to better adapt to complex and changing indoor environments and diverse user needs, thereby providing a more comfortable indoor environment. While meeting the personalized needs of different users, it also avoids energy waste, further achieving the goal of energy conservation and emission reduction. Summary of the Invention:

[0007] To address the problems of existing technologies, this invention provides a self-learning indoor thermal environment control method based on group thermal comfort voting. The self-learning module incorporates machine learning algorithms to continuously optimize control strategies and improve the system's intelligence level through the analysis and processing of historical and real-time feedback data. This adaptive adjustment capability allows the system to better adapt to complex and changing indoor environments and diverse user needs, thereby providing a more comfortable and energy-efficient indoor environment. This avoids the problem of single setpoints failing to meet diverse needs. Furthermore, group thermal comfort voting enables dynamic adjustment of the indoor thermal environment; the system can adjust its control strategy promptly based on voting results. The wireless communication technology not only improves the system's response speed and control accuracy but also reduces maintenance and time costs.

[0008] This invention is achieved through the following technical solution:

[0009] An autonomous learning indoor thermal environment control method based on group thermal comfort voting includes a data acquisition module, a data processing module, and an end-point control module;

[0010] The data acquisition module is used to collect user information, image information, the user's specific location information indoors, and indoor thermal environment information in real time.

[0011] The data processing module is used to process the user information and video information collected by the data acquisition module, analyze and process user instructions and cycle settings, and store and learn user information.

[0012] The terminal control module is used to receive the control strategy output by the data processing module, and adjust the parameter settings of the air conditioner by combining the preliminary control strategy derived from autonomous learning and the valid instructions issued by the temporary user.

[0013] Preferably, the data acquisition module includes a user information acquisition module, an image acquisition module, an indoor positioning system (IPS) module, and an indoor thermal environment detection module;

[0014] The user information collection module is used to collect basic user information, including user-issued command information, user's current location information, and user's heating and cooling preferences. Command information includes temporary user voting on the current indoor thermal environment and air conditioning request commands. Location information includes the user's current location when voting and the distance between the user and the air conditioner when the user remotely controls the system. User heating and cooling preferences include the user's subjective feelings and preferences for temperature and humidity based on different environments, times, and weather.

[0015] The image acquisition module is used to acquire video information, including the user's state when the user votes on the current indoor thermal environment. The user state information includes the user's physical state when issuing the command, including the user's hot / cold posture when issuing the command and the user's hot / cold posture when entering the room after pre-cooling / pre-heating is turned on. The acquired video is combined with the IPS to accurately locate the user's specific location information in the room. By controlling the air volume of the nearest air outlet based on the user's specific location information, this adaptive adjustment capability is achieved, enabling the system to better adapt to complex and changing indoor environments and diverse user needs.

[0016] The Indoor Positioning System (IPS) module is specifically designed for indoor environments to address the problem that GPS signals cannot penetrate buildings. It collects radio signals (such as Wi-Fi, Bluetooth, Zigbee, RFID, UWB, and radar), optical signals (such as infrared sensing, visible light communication VLC, and lidar), and sound signals (such as ultrasound and audible sound) in real time, and processes the collected signals to provide users with real-time location information for positioning services.

[0017] The indoor thermal environment detection module is used to collect indoor thermal environment information and obtain the optimal control strategy.

[0018] Preferably, the data processing module includes a target detection module, a video processing module, an instruction parsing module, a data storage module, and a self-learning module;

[0019] The target detection module is used to detect the user's location information, monitor the user's location information in real time, detect the location of the resident user through GPS positioning, prepare the indoor environment for the resident user in advance for preheating / precooling, locate the user's specific location in the room through IPS, and adjust the air outlet closest to the user according to the user's command information to change the air volume of the air outlet to meet the user's personalized needs.

[0020] The video processing module is used to process the video and image information collected by the data acquisition module, analyze and process the collected information, obtain the current indoor user's hot and cold posture, judge the current user's hot and cold status, and judge the validity of the user's command based on the user's status information.

[0021] Furthermore, the hot and cold postures of the indoor occupants include: wiping sweat with raised hands, fanning themselves with raised hands, rolling up sleeves, hugging their arms, breathing on their hands to warm them, and hugging their necks with both hands; when the hot and cold postures of the indoor occupants are wiping sweat with raised hands, fanning themselves with raised hands, or rolling up their sleeves, the hot and cold state of the indoor occupants is that the indoor occupants feel hot; when the hot and cold postures of the indoor occupants are hugging their arms, breathing on their hands to warm them, or hugging their necks with both hands, the hot and cold state of the indoor occupants is that the indoor occupants feel cold.

[0022] The instruction parsing module is used to analyze and process user instructions and cycle settings;

[0023] Furthermore, the aforementioned cycle setting is used to set the automatic start and stop, automatic control and preset of the air conditioner timer. The cycle setting is one of the preliminary control strategies derived through self-learning based on the heating and cooling preferences and daily routines of resident users.

[0024] The data storage module is used to save information such as the user's temperature preference and the time period they usually spend in.

[0025] The self-learning module is used to train and optimize the self-learning model. The model is continuously trained and optimized based on the user's preferences for warmth and coldness and their daily routine.

[0026] Furthermore, the term "resident user" refers to a person who has stayed in this room three times or more, and the information of resident users includes, but is not limited to, their temperature preferences and real-time location information, such as their daily routine.

[0027] Furthermore, in order to achieve accurate precooling / preheating, when the precooling / preheating mode is first turned on, the user presets the arrival time, and the air conditioner calculates the pre-start time based on the indoor cooling, heating and humidity load. The air conditioner turns on before the user arrives according to the pre-start time. After the air conditioner runs, it records and stores the actual time of reaching the standard. When the air conditioner's precooling / preheating mode is turned on again, the air conditioner uses the previous actual time of reaching the standard as a basis and, in conjunction with the self-learning module, corrects the air conditioner's pre-start time.

[0028] The aforementioned end-point control module is used to combine the preliminary control strategy trained by autonomous learning with the instructions issued by temporary users to derive an optimal control strategy. The optimal control strategy includes the temporary users' votes on the current indoor thermal environment and the heating and cooling preferences of permanent users. The autonomous learning module combines the voting information of temporary users with the heating and cooling preferences of permanent users to calculate the optimal control strategy applicable to the current indoor thermal environment.

[0029] A self-learning indoor thermal environment control method based on group thermal comfort voting, comprising the system described above, including:

[0030] ① When there are no temporary users:

[0031] S10, the data acquisition module collects information on resident users and indoor thermal environment information in real time;

[0032] S20, the data processing module calculates the arrival time of resident users based on the collected user location information, heating and cooling preferences, and common time periods, and turns on the air conditioner in advance to pre-cool / preheat the room;

[0033] The S30, with its autonomous learning module and camera, continuously collects the arrival time of each resident user and the temperature of the room when the user first arrives. It records the user's daily routine and temperature preferences, and uses autonomous learning to continuously optimize the system's calculation of the resident user's travel time and temperature preferences, so that the resident user can have the best indoor thermal environment each time they arrive at the room, and obtain the best control strategy.

[0034] S40: The terminal control module adjusts the air conditioning settings according to the optimal control strategy. When the resident user arrives indoors, if the resident user's reply instruction is affirmative, the air conditioning settings will continue to be adjusted according to the optimal control strategy. If the resident user's reply instruction is negative, a new control strategy will be set and the process will return to S30. If the resident user does not reply to the instruction or replies with an irrelevant instruction within the set time, the air conditioning settings will continue to be adjusted according to this optimal control strategy.

[0035] ②When there are temporary users:

[0036] S10, the data acquisition module collects temporary users' voting information, location information, instruction information, current status information, and current indoor thermal environment information through the mobile terminal;

[0037] S20, the data processing module processes the collected temporary user information, current indoor thermal environment information and information of permanent indoor users, especially based on the collected image and video information and the status information of temporary users, to determine whether the temporary user's command information is valid;

[0038] S30, the self-learning module combines the instructions of temporary users with the period settings of resident users, and uses the self-learning method to continuously train the control strategy under the current situation to obtain the optimal control strategy.

[0039] S40: The terminal control module adjusts the air conditioning settings according to the optimal control strategy. If the indoor occupant's response is affirmative, the air conditioning settings are adjusted according to the optimal control strategy. If the indoor occupant's response is negative, the system returns to S30. If the indoor occupant does not respond or responds with an irrelevant command within the set time, the air conditioning settings are adjusted according to the optimal control strategy.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] This invention relates to an autonomous learning indoor thermal environment control system based on group thermal comfort voting. Users vote on the current indoor thermal environment using mobile terminals. By neutralizing and adjusting users' heating and cooling preferences, the system obtains the optimal control strategy. This approach can meet the diverse thermal comfort needs of different groups while reducing energy consumption and costs, building upon existing technologies. Compared to other control methods, the autonomous learning indoor thermal environment control method based on group thermal comfort voting proposed in this invention better addresses the personalized thermal comfort requirements of the public, maximizing user satisfaction. Attached image description:

[0042] Figure 1 This is a block diagram of the system of the autonomous learning indoor thermal environment control method based on group thermal comfort voting of the present invention;

[0043] Figure 2 This is a flowchart of the autonomous learning indoor thermal environment control based on group thermal comfort voting in this invention.

[0044] Figure 3 This is a flowchart illustrating the system operation of the present invention.

[0045] Figure 4 Flowchart for the self-directed learning module;

[0046] Figure 5 A flowchart for determining the validity of user commands;

[0047] Figure 6 A self-learning process for determining the validity of instructions. Detailed implementation method:

[0048] To further understand the present invention, the present invention will be described below in conjunction with embodiments. These descriptions are only for further explaining the features and advantages of the present invention, and are intended to facilitate a clearer and more complete description of the present invention, and are not intended to limit the claims of the present invention.

[0049] likeFigure 1 This invention relates to a self-learning indoor thermal environment control method system based on group thermal comfort voting, specifically including a data acquisition module, a data processing module, and an end-point control module.

[0050] The data acquisition module is used to collect user information, video information, the user's specific location information indoors, and indoor environmental information in real time, providing data for the data processing module. The data acquisition module mainly includes a user information acquisition module, an image acquisition module, an indoor positioning system (IPS) module, and an indoor thermal environment detection module.

[0051] The user information collection module is used to collect user information and mainly includes a mobile terminal, a GPS positioning system, and an IPS.

[0052] The image acquisition module is used to acquire video information in real time, and mainly consists of a camera;

[0053] The indoor thermal environment detection module is used to detect the current indoor thermal environment status in real time, and mainly includes temperature and humidity sensors and air flow sensors.

[0054] The Indoor Positioning System (IPS) module is specifically designed for accurate location tracking of users indoors, while also providing navigation and tracking capabilities. It primarily includes a Wi-Fi access point, a Bluetooth beacon, and an ultra-wideband (UWB) sensor.

[0055] The data processing module is used to obtain the current indoor conditions and the user's presence in the room based on the user information, video information, and indoor thermal environment information collected by the data acquisition module, and to analyze and process user commands based on this information. The data processing module mainly includes a target detection module, a video processing module, a command parsing module, a data storage module, and a self-learning module.

[0056] The target detection module is used to detect user location information. When a temporary user votes on the current indoor thermal environment, the module detects the temporary user's location information and adjusts the airflow at the nearest air outlet based on the IPS positioning system, thereby changing the air volume to meet personalized needs. Simultaneously, it also makes timely equipment adjustments based on the location information of resident users. For example, if a resident user is detected approaching from three kilometers away, the air conditioner will pre-cool / preheat based on the resident user's heating / cooling preferences.

[0057] The video processing module is used to process the video and image information collected by the data acquisition module in real time, analyze and process the collected information, and determine the user's current location and current status information based on the pictures and video information when the user votes, so as to ensure the precise control of the air outlet and the effectiveness of the user's vote.

[0058] The instruction parsing module is used to analyze and process user instructions and cycle settings. The cycle settings refer to setting a set of perfectly applicable personalized air conditioning control parameters for resident users based on the resident user information stored in the data acquisition module.

[0059] The data storage module is used to store user information, mainly storing common command control information of the mobile terminal and GPS location information.

[0060] The self-learning module is used to train and optimize the model, continuously optimizing air conditioning control. Based on user information and commonly used user commands, it continuously learns and trains to achieve optimal air conditioning control for resident users.

[0061] The term "regular users" refers to people who have stayed in this room three times or more. Information on regular users includes, but is not limited to, their preferences for temperature and real-time location, such as their daily routines.

[0062] The cycle setting module is used to set the automatic start and stop, automatic control and pre-cooling / preheating automatic start settings of the air conditioner. The cycle setting is usually based on the heating and cooling preferences and work and rest times of the resident users collected by the data storage module, and sets the air conditioner control settings specifically for the resident users.

[0063] The aforementioned end-point control module is used to combine the preliminary control strategy trained by autonomous learning with the instructions issued by temporary users to derive an optimal control strategy. The optimal control strategy combines the voting information of temporary users on the current indoor thermal environment with the preliminary control strategy derived by autonomous learning based on the heating and cooling preferences of permanent users. Through the autonomous learning module, the voting information of temporary users and the heating and cooling preferences of permanent users are simulated and optimized to calculate the optimal control strategy applicable to the current indoor thermal environment. The control parameters of the air conditioner (such as temperature, humidity and fan speed) are adjusted accordingly to provide an indoor thermal environment that satisfies the indoor users.

[0064] like Figure 2 This invention discloses a self-learning indoor thermal environment control method system based on group thermal comfort voting. Based on the above system, the main steps include:

[0065] ① When there are no temporary users:

[0066] S10, the data acquisition module collects information on resident users and indoor thermal environment information in real time;

[0067] S20, the data processing module calculates the arrival time of resident users based on the collected user location information, heating and cooling preferences, and common time periods, and turns on the air conditioner in advance to pre-cool / preheat the room;

[0068] The S30, with its autonomous learning module and camera, continuously collects the arrival time of each resident user and the temperature status of the room when the user first arrives. It records the user's daily routine, temperature preferences, and uses autonomous learning to continuously optimize the system's calculation of the resident user's travel time and temperature preferences, so that the resident user can have the optimal indoor thermal environment each time they arrive at the room, and obtain the optimal control strategy.

[0069] S40: The terminal control module adjusts the air conditioning settings according to the optimal control strategy. When the resident user arrives indoors, if the resident user's response is affirmative, the air conditioning settings will continue to be adjusted according to the optimal control strategy. If the resident user's response is negative and a new control strategy is set, the process returns to S30. If the resident user does not respond to the instruction or responds with an irrelevant instruction within the set time, the air conditioning settings will continue to be adjusted according to this optimal control strategy.

[0070] ②When there are temporary users:

[0071] S10, the data acquisition module collects temporary users' voting information, location information, instruction information, and current indoor thermal environment information through the mobile terminal;

[0072] S20, the data processing module processes the collected user information, current indoor thermal environment information and information of resident users, especially based on the collected image and video information, to determine whether the command information of temporary users is valid;

[0073] S30, the self-learning module combines the instructions of temporary users with the period settings of resident users, and uses the self-learning method to continuously train the control strategy under the current situation to obtain the optimal control strategy.

[0074] S40: The terminal control module adjusts the air conditioning settings according to the optimal control strategy. If the indoor occupant's response is affirmative, the air conditioning settings are adjusted according to the optimal control strategy. If the indoor occupant's response is negative, the system returns to S30. If the indoor occupant does not respond or responds with an irrelevant command within the set time, the air conditioning settings are adjusted according to the optimal control strategy.

[0075] Example 1:

[0076] In one specific embodiment, the implementation environment is an office space, and the office workers are resident users.

[0077] In the initial stage of use, the system continuously records the commands of resident users to collect their temperature preferences, and processes the data on the usage stages of resident users to obtain their daily routines.

[0078] By collecting information from resident users, a self-learning process is used to train and optimize the model. Based on the training results, a cycle setting suitable for resident users is derived. Combining GPS positioning system and IPS, the air conditioning is turned on in advance to pre-cool / preheat the indoor environment before resident users arrive, based on the estimated pre-cooling / preheating time provided by the self-learning module. The air conditioning is turned off in advance a period of time before resident users leave, usually set according to their commuting hours.

[0079] Based on the information preferences of all resident users, the optimal control strategy that suits the majority of resident users is calculated and then implemented.

[0080] If a user is not satisfied with the optimal control strategy, they can directly reply "no" to the command on their mobile terminal and issue a new command. The system will automatically combine the new command with the temperature preferences of other resident users to obtain a new optimal control strategy.

[0081] Example 2:

[0082] In one specific embodiment, the implementation environment is a commercial area, where the mall staff are resident users and transient customers are temporary users.

[0083] When temporary users are dissatisfied with the current indoor thermal environment, they can vote and issue desired commands on their mobile devices.

[0084] The system uses images and video information captured by cameras to determine whether the instructions from temporary users are valid. If valid, it combines information from resident users and instructions from temporary users to obtain the optimal control strategy and output it.

[0085] If a user is not satisfied with the new optimal control strategy, they can reply "no" to the command on their mobile terminal and issue a new command. The system will automatically combine the new command with the temperature preferences of other resident users to obtain a new optimal control strategy.

Claims

1. A self-learning indoor thermal environment control system based on group thermal comfort voting, characterized in that... It includes a data acquisition module, a data processing module, and a terminal processing module; The data acquisition module is used to collect user information, image information, the user's specific location information indoors, and indoor thermal environment information in real time. The data processing module is used to process the user information and video information collected by the data acquisition module, analyze and process user instructions and cycle settings, and store and learn user information. The terminal control module is used to receive the control strategy output by the data processing module, and adjust the parameter settings of the air conditioner by combining the preliminary control strategy derived from autonomous learning and the valid instructions issued by the temporary user.

2. The autonomous learning indoor thermal environment control system based on group thermal comfort voting as described in claim 1, characterized in that, The data acquisition module includes a user information acquisition module, an image acquisition module, an indoor positioning system (IPS) module, and an indoor thermal environment detection module; The user information collection module is used to collect basic user information, including user-issued command information, user's current location information, and user's heating and cooling preferences. Command information includes temporary user voting on the current indoor thermal environment and air conditioning request commands. Location information includes the user's current location when voting and the distance between the user and the air conditioner when remotely controlling the system. User status information includes the user's physical state when issuing commands. User heating and cooling preferences include the user's subjective feelings and preferences for temperature based on different environments, times, and weather conditions. The image acquisition module is used to acquire video information, including the user's location and status when the user votes on the current indoor thermal environment. The user status information includes the user's body state when issuing the command, including the user's hot / cold posture when issuing the command and the user's hot / cold posture when entering the room after pre-cooling / pre-heating is turned on. The acquired video is combined with the IPS to accurately locate the user's specific location information in the room. By controlling the air volume of the nearest air outlet based on the user's specific location information, this adaptive adjustment capability is achieved, enabling the system to better adapt to complex and changing indoor environments and diverse user needs. The Indoor Positioning System (IPS) module is specifically designed for indoor environments to address the problem that GPS signals cannot penetrate buildings. It collects radio signals (such as Wi-Fi, Bluetooth, Zigbee, RFID, UWB, and radar), optical signals (such as infrared sensing, visible light communication VLC, and lidar), and sound signals (such as ultrasound and audible sound) in real time, and processes the collected signals to provide users with real-time location information for positioning services. The indoor thermal environment detection module is used to collect indoor thermal environment information and obtain the optimal control strategy.

3. The autonomous learning indoor thermal environment control system based on group thermal comfort voting as described in claim 1, characterized in that, The data processing module includes a target detection module, a video processing module, an instruction parsing module, a data storage module, and a self-learning module; The target detection module is used to detect the user's location information, monitor the user's location information in real time, detect the location of the resident user through GPS positioning, prepare the indoor environment for the resident user in advance for preheating / precooling, locate the user's specific location in the room through IPS, and adjust the air outlet closest to the user according to the user's command information to change the air volume of the air outlet to meet the user's personalized needs. The video processing module is used to process the video and image information collected by the data acquisition module, analyze and process the collected information, obtain the current indoor user's hot and cold posture, judge the current user's hot and cold status, and judge the validity of the user's command based on the user's status information. The instruction parsing module is used to analyze and process user instructions and cycle settings; The data storage module is used to save information such as the user's temperature preference and the time period they usually spend in. The self-learning module is used to train and optimize the self-learning model. The model is continuously trained and optimized based on the user's preferences for warmth and coldness and their daily routine.

4. The autonomous learning indoor thermal environment control system based on group thermal comfort voting as described in claim 1, characterized in that, The aforementioned end-point control module is used to combine the preliminary control strategy trained by autonomous learning with the instructions issued by temporary users to derive an optimal control strategy. The optimal control strategy includes the temporary users' votes on the current indoor thermal environment and the heating and cooling preferences of permanent users. The autonomous learning module calculates the optimal control strategy applicable to the current indoor thermal environment.

5. A self-learning indoor thermal environment control method based on group thermal comfort voting, characterized in that, The system based on claim 1 includes: ① When there are no temporary users: S10, the data acquisition module collects information on resident users and indoor thermal environment information in real time; S20, the data processing module calculates the arrival time of resident users based on the collected user location information, heating and cooling preferences, and common time periods, and turns on the air conditioner in advance to pre-cool / preheat the room; The S30, with its autonomous learning module and camera, continuously collects the arrival time of each resident user and the temperature status of the room when the user first arrives. It records the user's daily routine, temperature preferences, and uses autonomous learning to continuously optimize the system's calculation of the resident user's travel time and temperature preferences, so that the resident user can have the optimal indoor thermal environment each time they arrive at the room, and obtain the optimal control strategy. S40: The terminal control module obtains the optimal control strategy and adjusts the air conditioning settings according to the optimal control strategy. When the resident user arrives indoors, if the resident user's reply instruction is affirmative, the air conditioning settings will continue to be adjusted according to the optimal control strategy. If the resident user's reply instruction is negative and a new control strategy is set, the process returns to S30. If the resident user does not reply with the instruction or replies with an irrelevant instruction within the set time, the air conditioning settings will continue to be adjusted according to this optimal control strategy in the future. ②When there are temporary users: S10, the data acquisition module collects temporary users' voting information, location information, instruction information, current status information, and current indoor thermal environment information through the mobile terminal; S20, the data processing module processes the collected user information, current indoor thermal environment information and information of permanent indoor users, especially based on the collected image and video information and the status information of temporary users, to determine whether the command information of temporary users is valid; S30, the self-learning module combines the instructions of temporary users with the period settings of resident users, and uses the self-learning method to continuously train the control strategy under the current situation to obtain the optimal control strategy. S40: The terminal control module obtains the optimal control strategy and adjusts the air conditioning settings according to the optimal control strategy. If the indoor occupant's response command is affirmative, the air conditioning settings are adjusted according to the optimal control strategy. If the indoor occupant's response command is negative, the process returns to S30. If the indoor occupant does not respond or responds with an irrelevant command within the set time, the air conditioning settings are adjusted according to the optimal control strategy.

Images