Intelligent household temperature control system based on artificial intelligence

Abstract

The invention relates to the technical field of artificial intelligence, in particular to an intelligent home temperature control system based on artificial intelligence, which comprises a multi-source data acquisition module, a user behavior analysis module, a mode decision engine, an instruction execution module and a feedback optimization module, and dynamically selecting a setting mode or a recovery mode according to the current environment parameters, the real-time position of the user and the time information. According to the method, the corresponding execution mode can be intelligently selected when the user initiates the regulation and control instruction through the global control interface (one-key starting), mode self-adaption is achieved, the accuracy of matching with the user intention is improved, the time nodes for adjusting the equipment parameters in all the functional areas are intelligently generated, and according to the home arrival time of all the members, the user experience is improved. The time node of each area for executing the regulation and control instruction is dynamically adjusted, the intelligence of the temperature control system is improved, and the resource waste is reduced.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, specifically a smart home temperature control system based on artificial intelligence. Background Art

[0002] The intelligent temperature control system realizes the automatic adjustment of indoor temperature by combining sensor technology, Internet of Things (IoT), artificial intelligence algorithms and user habit analysis. It avoids the idling of equipment and reduces energy waste. For example, it automatically turns off heating / cooling when detecting that there is no one at home. Through data-driven and automation technologies, the intelligent temperature control system upgrades temperature control from "passive response" to "active adaptation", reducing energy consumption while improving the quality of life, which is an important development direction for future home intelligence.

[0003] When the intelligent temperature control system is deployed, users can send control commands in advance, such as turning on smart devices such as air conditioners in advance. However, the system generally has two modes: the setting mode and the recovery mode. Among them, in the setting mode, after receiving the control command, it adjusts according to the pre-set device parameters; in the recovery mode, after receiving the control command, it restores the parameter settings of each smart device to the state before leaving home. When the user initiates a control command through the global control interface (one-key on), the system cannot perform mode self-adaptation according to the current environmental parameters (indoor-outdoor temperature difference, device operation duration), user behavior patterns (length of time away from home, season and time period), and real-time requirements (energy consumption optimization / rapid temperature adjustment), resulting in insufficient matching of user intentions. Summary of the Invention

[0004] The purpose of the present invention is to provide a smart home temperature control system based on artificial intelligence to solve the problems raised in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A smart home temperature control system based on artificial intelligence, the smart home temperature control system includes: a multi-source data acquisition module, a user behavior analysis module, a mode decision engine, an instruction execution module, and a feedback optimization module, where:

[0006] The multi-source data acquisition module is used to obtain user operation data, indoor and outdoor environmental parameters, and user location information in real time;

[0007] The user behavior analysis module, based on historical operation data and machine learning models, dynamically selects the setting mode or the recovery mode according to the current environmental parameters, user's real-time location, and time information;

[0008] The instruction execution module is used to send control commands to smart devices and monitor the execution status;

[0009] The feedback optimization module updates the parameters of the machine learning model according to the user feedback data and optimizes the mode decision logic.

[0010] Further, the multi-source data acquisition module includes:

[0011] The environmental perception unit: integrates a temperature sensor, a humidity sensor, a light sensor, and an outdoor meteorological data interface, and is used to obtain the values of indoor temperature, humidity, light intensity, and outdoor temperature;

[0012] The user behavior recording unit: stores the user's historical operation logs, including the setting mode parameter adjustment records, the recovery mode trigger time, and the device response delay data;

[0013] The position tracking unit: detects the user's real-time position through the GPS module or the home WiFi signal strength, and calculates the dynamic distance between the user and the home position.

[0014] Further, the user behavior analysis module includes:

[0015] Obtain the three-dimensional model of the indoor layout and divide it into each functional area;

[0016] Record the identity information of each family member, collect the activity trajectories of each family member, count the activity duration of each family member in each functional area, and set the family member with the longest activity duration as the main behavior object of this functional area;

[0017] Set the corresponding device parameters for each functional area to form a preset mode;

[0018] After the family member leaves, automatically switch to the away state, and the away state is to adjust each intelligent device to the standby state;

[0019] Record the device parameters corresponding to each functional area before switching to the away state to form a recovery mode;

[0020] After receiving the control instruction sent by the user terminal, send a mode selection instruction to the user terminal. The mode selection mode is an instruction to execute the recovery mode or the set mode. If the user terminal responds to the mode selection instruction, execute the mode selected by the user terminal;

[0021] If the user does not respond to the mode selection instruction, obtain the current indoor-outdoor temperature difference threshold and the device historical energy consumption data, dynamically allocate the priority weights of the recovery mode and the set mode, and automatically select the corresponding mode according to the priority weights:

[0022] If the priority weight of the recovery mode is greater than the priority weight of the set mode, execute the recovery mode. If the priority weight of the recovery mode is less than the priority weight of the set mode, execute the selected mode;

[0023] Among them, the calculation process of the priority weight is as follows: extract the preferred temperature of family members;

[0024] Obtain the temperature difference between the set temperature and the preferred temperature in the selected mode, and the temperature difference between the set temperature and the preferred temperature in the recovery mode. Select the set temperature with the smaller temperature difference as the target temperature, mark the selected temperature as 1, and mark the unselected temperature as 0, which is marked as Data One;

[0025] Extract the number of times family members modify within the judgment time after arriving home after selecting a certain mode in the historical records. The judgment time is a certain period of time preset by the system after family members arrive home; calculate the proportion of the number of modifications to the number of issued regulation instructions, which is marked as Data Two;

[0026] Obtain the priority weight by normalizing Data One and Data Two.

[0027] Furthermore, obtain the information of family members who issue regulation instructions through the user terminal, match the functional areas corresponding to this family member as the main behavior object, calculate the time node when this family member arrives home, sort out the behavior route after this family member arrives home, extract each functional area involved in the behavior route, mark it as Adjustment Area One, and arrange each Adjustment Area One according to the sequence of the behavior route to generate response nodes, where the response nodes are the device parameters corresponding to the functional areas.

[0028] Obtain the arrival time of this family member, and calculate the starting time to execute the regulation instruction according to the maximum power of the device.

[0029] Obtain the return time nodes of the remaining family members, sort them according to the time line according to the return time nodes, and sequentially extract the functional areas of the remaining family members as the main behavior objects, which are marked as Adjustment Area Two;

[0030] If there are overlapping functional areas between Adjustment Area Two and Adjustment Area One, then remove the overlapping functional areas from Adjustment Area Two, obtain the response time nodes of the remaining Adjustment Area Two, and form an instruction response time sequence with the starting time of Adjustment Area One;

[0031] Execute the regulation instruction after selecting the mode according to the instruction response time sequence.

[0032] Furthermore, the user behavior analysis module further includes the following functional units:

[0033] Time series prediction unit: use the ARIMA or LSTM model to predict the time required for the user to arrive home and the best timing for device startup;

[0034] Abnormal Scenario Handling Unit: When the indoor-outdoor temperature difference exceeds the safety threshold or a device failure is detected, it forcibly switches to the set mode and calls the preset safety parameters;

[0035] User Intention Verification Unit: Before the mode execution, it pushes a parameter preview interface to the user terminal. If the user does not respond, the system-recommended mode is executed by default.

[0036] Furthermore, the instruction execution module includes:

[0037] Priority Scheduling Unit: Dynamically adjusts the multi-device control order according to the mode weight and the current device load;

[0038] Status Rollback Unit: If the device execution is abnormal or the user manually interrupts, it automatically restores to the state before regulation and triggers a fault alarm;

[0039] Energy Efficiency Optimization Unit: Combines the grid peak-valley electricity price data and selects the lowest energy consumption control strategy on the premise of meeting the user's needs.

[0040] Furthermore, the feedback optimization module includes:

[0041] Active Learning Mechanism: Dynamically adjusts the confidence threshold of the association rule according to the user's acceptance rate of the recommended mode and the manual correction operation data;

[0042] Cross-Scenario Migration Module: Migrates the trained user preference model to other intelligent devices under the same account and incrementally updates the model based on the new device data.

[0043] Furthermore, the environment perception unit further includes:

[0044] Infrared human body sensor, used to detect the presence status of indoor personnel;

[0045] Voice command parsing interface, supporting the user to directly modify the set mode parameters through natural language.

[0046] Furthermore, the user intention verification unit pushes the parameter preview interface in the following way:

[0047] Displays the expected regulation effect curve graph and energy consumption comparison data in the intelligent terminal APP;

[0048] Provides three-stage interactive buttons of "One-key Confirmation", "Fine-tuning Parameters", and "Cancel Operation", and updates the decision logic in real time according to the user's selection.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] 1. The present invention can intelligently select the execution mode corresponding to the regulation instruction initiated by the user through the global control interface (one-key activation), achieve mode self-adaptation, improve the accuracy of matching with the user's intention, and intelligently generate the time nodes for adjusting the device parameters in each functional area. According to the arrival time of each member, the time nodes for executing the regulation instruction in each area are dynamically adjusted, improving the intelligence of the temperature control system and reducing resource waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the modules of a smart home temperature control system based on artificial intelligence according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0053] Embodiment: As Figure 1 shown, the present invention provides a smart home temperature control system based on artificial intelligence. The smart home temperature control system includes: a multi-source data acquisition module, a user behavior analysis module, a mode decision engine, an instruction execution module, and a feedback optimization module, where:

[0054] The multi-source data acquisition module is used to obtain user operation data, indoor and outdoor environmental parameters, and user location information in real time;

[0055] The user behavior analysis module, based on historical operation data and a machine learning model, dynamically selects a set mode or a recovery mode according to the current environmental parameters, the user's real-time location, and time information;

[0056] The instruction execution module is used to send a regulation instruction to the smart device and monitor the execution status;

[0057] The feedback optimization module updates the parameters of the machine learning model according to the user feedback data and optimizes the mode decision logic.

[0058] Further, the multi-source data acquisition module includes:

[0059] An environmental perception unit: integrating a temperature sensor, a humidity sensor, a light sensor, and an outdoor meteorological data interface, and used to obtain the values of indoor temperature, humidity, light intensity, and outdoor temperature;

[0060] User Behavior Record Unit: Stores the user's historical operation logs, including the adjustment records of the set mode parameters, the triggering time of the recovery mode, and the device response delay data;

[0061] Location Tracking Unit: Detects the user's real-time location through the GPS module or the home WiFi signal strength, and calculates the dynamic distance between the user and the home location.

[0062] Furthermore, the user behavior analysis module includes:

[0063] Obtains the three-dimensional model of the indoor layout and divides it into each functional area;

[0064] Records the identity information of each family member, collects the activity trajectories of each family member, counts the activity duration of each family member in each functional area, and sets the family member with the longest activity duration as the main behavior object in that functional area;

[0065] Sets the corresponding device parameters for each functional area to form a preset mode;

[0066] After a family member leaves, automatically switches to the away state, and the away state is to adjust each smart device to the standby state;

[0067] Records the device parameters corresponding to each functional area before switching to the away state to form a recovery mode;

[0068] After receiving the regulation instruction sent by the user terminal, sends a mode selection instruction to the user terminal. The mode selection mode is an instruction to execute the recovery mode or the set mode. If the user terminal responds to the mode selection instruction, the mode selected by the user terminal is executed;

[0069] If the user does not respond to the mode selection instruction, obtains the current indoor-outdoor temperature difference threshold and the device historical energy consumption data, dynamically allocates the priority weights of the recovery mode and the set mode, and automatically selects the corresponding mode according to the priority weights:

[0070] If the priority weight of the recovery mode is greater than the priority weight of the set mode, the recovery mode is executed. If the priority weight of the recovery mode is less than the priority weight of the set mode, the selected mode is executed;

[0071] Among them, the calculation process of the priority weight is: extracts the preferred temperature of the family member;

[0072] Obtains the temperature difference between the set temperature and the preferred temperature in the selected mode, and the temperature difference between the set temperature and the preferred temperature in the recovery mode, selects the set temperature with the smaller temperature difference as the target temperature, marks the selected temperature as 1, and marks the unselected temperature as 0, which is marked as data one;

[0073] During the extraction of historical records, after selecting a certain mode, the number of times a family member makes modifications within the judgment time after arriving home is counted. The judgment time is a certain period set by the system after the family member arrives home; calculate the proportion of the number of modifications to the number of issued control commands, which is marked as Data Two;

[0074] The priority weights are obtained through normalization calculation of Data One and Data Two.

[0075] Among them, the smaller the temperature difference and the smaller the proportion, the greater the priority weight. Or the existing technology can be used to determine which mode to select. The parameters of each preset mode can be generated according to big data analysis and user habits, or the user can customize the device parameters of each intelligent device.

[0076] Obtain the family member information of the control command issued through the user terminal, match the functional area corresponding to this family member as the main behavior object, calculate the arrival time node of this family member at home, sort out the behavior route after this family member arrives home, extract each functional area involved in the behavior route, mark it as Adjustment Area One, and arrange each Adjustment Area One in the order of the behavior route to generate response nodes, where the response nodes are the device parameters corresponding to the functional areas.

[0077] Obtain the arrival time of this family member at home, and calculate the starting time to execute the control command according to the maximum power of the device.

[0078] Obtain the return time nodes of the remaining family members, sort them according to the time line according to the returned time nodes, and sequentially extract the functional areas of the remaining family members as the main behavior objects, and mark them as Adjustment Area Two;

[0079] If there are overlapping functional areas between Adjustment Area Two and Adjustment Area One, then remove the overlapping functional areas from Adjustment Area Two, obtain the response time nodes of the remaining Adjustment Area Two, and form an instruction response time sequence with the starting time of Adjustment Area One;

[0080] Execute the control command after selecting the mode according to the instruction response time sequence.

[0081] Specifically, after the user terminal issues a control command, first determine the mode to be executed, select the preset mode or the recovery mode. The preset mode is to adjust the device parameters of each functional area to the temperature preset by the user, and the recovery mode is to adjust the device parameters of each functional area to the final temperature before all family members leave;

[0082] After selecting the execution mode, obtain the device parameters of each functional area. At this time, analyze the information of the family member who issued the instruction. For example, if mom issues a regulation instruction, generate the behavior trajectory after mom comes home. For example, mom goes to the living room to rest first after coming home, and then cooks in the kitchen. At this time, although the functional area corresponding to mom is the bedroom, it is not the start of the behavior trajectory. At this time, according to mom's behavior trajectory, adjust the functional areas involved, and then when the preset time arrives, first adjust the device parameters of the functional areas involved. Then, obtain the second and third family members who come back. For example, when dad comes back second, at this time, the device parameters of the corresponding functional area, the bathroom, can be adjusted according to the time when dad comes back, and so on. After determining the mode, obtain the device parameters of each functional area, and then dynamically adjust the time points for starting to adjust the device parameters of each functional area.

[0083] The user behavior analysis module further includes the following functional units:

[0084] Time series prediction unit: Use the ARIMA or LSTM model to predict the time required for the user to reach home and the best timing for device startup;

[0085] Abnormal scenario handling unit: When it is detected that the indoor-outdoor temperature difference exceeds the safety threshold or a device failure occurs, forcibly switch to the set mode and call the preset safety parameters;

[0086] User intention verification unit: Push a parameter preview interface to the user terminal before the mode is executed. If the user does not respond, the system-recommended mode is executed by default.

[0087] The instruction execution module includes:

[0088] Priority scheduling unit: Dynamically adjust the multi-device regulation order according to the mode weight and the current device load;

[0089] Status rollback unit: If the device execution is abnormal or the user manually interrupts, automatically restore to the state before regulation and trigger a fault alarm;

[0090] Energy efficiency optimization unit: Combine the grid peak-valley electricity price data and select the lowest energy consumption regulation strategy on the premise of meeting the user's needs.

[0091] The feedback optimization module includes:

[0092] Active learning mechanism: Dynamically adjust the confidence threshold of the association rule according to the user's acceptance rate of the recommended mode and the manual correction operation data;

[0093] Cross-scenario migration module: Migrate the trained user preference model to other intelligent devices under the same account and incrementally update the model based on the new device data.

[0094] The environmental perception unit further includes:

[0095] An infrared human body sensor for detecting the presence status of indoor personnel;

[0096] A voice command parsing interface that supports users to directly modify the set mode parameters through natural language.

[0097] The user intention verification unit pushes the parameter preview interface in the following way:

[0098] Display the expected regulation effect curve graph and energy consumption comparison data in the intelligent terminal APP;

[0099] Provide three-stage interactive buttons of "one-key confirmation", "fine-tuning parameters", and "canceling the operation", and update the decision-making logic in real time according to the user's selection.

[0100] By intelligently selecting the execution mode corresponding to the regulation instruction initiated by the user through the global control interface (one-key activation), the mode self-adaptation is realized, the accuracy of matching with the user intention is improved, and the time nodes for adjusting the device parameters of each functional area are intelligently generated. According to the arrival time of each member, the time nodes for executing the regulation instruction in each area are dynamically adjusted, the intelligence of the temperature control system is improved, and resource waste is reduced.

[0101] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

Claims

1. An artificial intelligence-based smart home temperature control system, characterized in that: The smart home temperature control system includes: a multi-source data acquisition module, a user behavior analysis module, a mode decision engine, an instruction execution module, and a feedback optimization module, where: The multi-source data acquisition module is used to obtain user operation data, indoor and outdoor environmental parameters, and user location information in real time; The user behavior analysis module, based on historical operation data and a machine learning model, dynamically selects a set mode or a recovery mode according to current environmental parameters, the user's real-time location, and time information; The instruction execution module is used to send control instructions to smart devices and monitor the execution status; The feedback optimization module updates the parameters of the machine learning model according to user feedback data and optimizes the mode decision logic.

2. The smart home temperature control system based on artificial intelligence according to claim 1, wherein: The multi-source data acquisition module includes: An environmental perception unit: integrating a temperature sensor, a humidity sensor, a light sensor, and an outdoor meteorological data interface, and used to obtain the values of indoor temperature, humidity, light intensity, and outdoor temperature; A user behavior recording unit: storing the user's historical operation logs, including setting mode parameter adjustment records, recovery mode trigger times, and device response delay data; A location tracking unit: detecting the user's real-time location through a GPS module or the strength of the home WiFi signal, and calculating the dynamic distance between the user and the home location.

3. The artificial intelligence-based smart home temperature control system according to claim 1, wherein: The user behavior analysis module includes: Obtaining a three-dimensional model of the indoor layout and dividing it into each functional area; Recording the identity information of each family member, collecting the activity trajectories of each family member, counting the activity duration of each family member in each functional area, and setting the family member with the longest activity duration as the main behavior object in this functional area; Setting corresponding device parameters for each functional area to form a preset mode; After a family member leaves, automatically switch to the away state, and the away state is to adjust each smart device to the standby state; Recording the device parameters corresponding to each functional area before switching to the away state to form a recovery mode; After receiving a control instruction sent by the user terminal, sending a mode selection instruction to the user terminal, and the mode selection mode is an instruction to execute the recovery mode or the set mode. If the user terminal responds to the mode selection instruction, execute the mode selected by the user terminal; If the user does not respond to the mode selection instruction, obtain the current indoor and outdoor temperature difference threshold and the device historical energy consumption data, dynamically allocate the priority weights of the recovery mode and the set mode, and automatically select the corresponding mode according to the priority weights: If the priority weight of the recovery mode is greater than the priority weight of the set mode, execute the recovery mode. If the priority weight of the recovery mode is less than the priority weight of the set mode, execute the selected mode.

4. The artificial intelligence-based smart home temperature control system according to claim 3, wherein: Obtain the family member information for which the regulation instruction is issued by the client, match the family member as the main behavior object to the corresponding functional area, calculate the arrival time node of the family member, sort out the behavior route after the family member arrives home, extract each functional area involved in the behavior route, mark it as adjustment area one, arrange each adjustment area one in the order of the behavior route to generate a response node, where the response node is the device parameter corresponding to the functional area. Obtain the arrival time of the family member, and calculate the starting time for executing the regulation instruction according to the maximum power of the device. Obtain the return time nodes of the other family members, sort them according to the time line according to the returned time nodes, and sequentially extract the functional areas of the other family members as the main behavior objects, and mark them as adjustment area two. If there are overlapping functional areas between adjustment area two and adjustment area one, remove the overlapping functional areas from adjustment area two, obtain the response time nodes of the remaining adjustment area two, and form an instruction response time sequence with the starting time of adjustment area one. Execute the regulation instruction after selecting the mode according to the instruction response time sequence.

5. The smart home temperature control system based on artificial intelligence according to claim 1, characterized in that: The user behavior analysis module further includes the following functional units: Time series prediction unit: Use the ARIMA or LSTM model to predict the time required for the user to arrive home and the best timing for device startup. Abnormal scenario processing unit: When it is detected that the indoor-outdoor temperature difference exceeds the safety threshold or the device fails, forcefully switch to the set mode and call the preset safety parameters. User intention verification unit: Push a parameter preview interface to the user terminal before the mode is executed. If the user does not respond, the system recommended mode will be executed by default.

6. The smart home temperature control system based on artificial intelligence according to claim 1, characterized in that: The instruction execution module includes: Priority scheduling unit: Dynamically adjust the multi-device regulation order according to the mode weight and the current device load. Status rollback unit: If the device execution is abnormal or the user manually interrupts, automatically restore to the state before regulation and trigger a fault alarm. Energy efficiency optimization unit: Combine the grid peak-valley electricity price data and select the lowest energy consumption regulation strategy on the premise of meeting the user's needs.

7. The intelligent home temperature control system based on artificial intelligence according to claim 1, characterized in that: The feedback optimization module includes: Active learning mechanism: Dynamically adjust the confidence threshold of the association rule according to the acceptance rate of the recommended mode by the user and the manual correction operation data. Cross-scenario migration module: Migrate the trained user preference model to other intelligent devices under the same account, and incrementally update the model based on the new device data.

8. The smart home temperature control system based on artificial intelligence according to claim 2, characterized in that: The environment perception unit further includes: Infrared human body sensor, used to detect the presence status of indoor personnel. Voice command parsing interface, which supports the user to directly modify the set mode parameters through natural language.

9. The smart home temperature control system based on artificial intelligence according to claim 5, wherein: The user intention verification unit pushes the parameter preview interface in the following way: Display the expected regulation effect curve graph and energy consumption comparison data in the intelligent terminal APP. Provide three-stage interactive buttons of "one-key confirmation", "fine-tuning parameters", and "cancel operation", and update the decision logic in real time according to the user's selection.

Images