Remote temperature control system based on artificial intelligence and control method thereof

Abstract

The application relates to the technical field of intelligent temperature control, and discloses a remote temperature control system based on artificial intelligence and a control method thereof, which comprises the following modules: an environment sensing module for collecting environment parameters; a temperature control device for executing adjustment and outputting operation state parameters; a user interactive terminal module for receiving temperature control setting information, comfort feedback information and user position data and uploading the same; a device control gateway module for communicating with the temperature control device and a cloud end, receiving and issuing control instructions, and uploading data; a data storage module for storing data to form historical data; an artificial intelligence model module for training based on the historical data and outputting target temperature control parameters and a control strategy; and a control strategy generation module for generating control instructions in combination with preset constraint conditions. When the target temperature control parameters and the control strategy are output, energy consumption reference and constraint basis can be provided, and in combination with preset constraints, executable instructions can be generated, precise remote temperature control can be realized, and the energy saving effect and the strategy rationality can be significantly improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent temperature control technology, specifically to a remote temperature control system and control method based on artificial intelligence. Background Technology

[0002] Remote temperature control systems are typically used to remotely monitor and regulate the temperature of target spaces such as residences and offices. They achieve cross-space control through temperature control equipment, user terminals, and network communication to improve management convenience and user experience. Artificial intelligence-based remote temperature control systems build upon the above-mentioned remote temperature control by introducing the ability to learn and analyze data such as environmental parameters, equipment operating status, and user input. This generates target temperature control parameters and control strategies, enabling temperature control to evolve from fixed thresholds or simple rules to data-driven adaptive control. Existing artificial intelligence-based remote temperature control implementations typically focus on multi-source data acquisition and model prediction to output target temperature control parameters and issue control commands.

[0003] However, current technologies often only use energy consumption as a statistical result or a post-hoc reference, lacking a mechanism to establish and maintain historical energy consumption baselines based on operating status parameters. This results in artificial intelligence models lacking stable energy consumption benchmarks when generating target temperature control parameters and control strategies, which can easily lead to problems such as unstable energy efficiency of control strategies or abnormal increases in energy consumption that are difficult to constrain in a timely manner. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an artificial intelligence-based remote temperature control system and its control method, solving the problems of insufficient historical energy consumption baseline leading to a lack of energy consumption benchmarks for control strategies and unstable energy-saving effects in existing technologies.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a remote temperature control system based on artificial intelligence, comprising: The environmental sensing module is used to collect environmental parameters of the target space and the operating environment parameters of the temperature control equipment; Temperature control equipment executes temperature adjustment operations according to control commands and outputs its own operating status parameters; The user interaction terminal module displays environmental parameters and the operating status of the temperature control equipment, receives temperature control settings and comfort feedback information input by the user, and obtains user location data and uploads the user location data to the cloud server. The device control gateway module communicates with the temperature control device and the cloud server, receives control commands from the cloud server and sends execution commands to the temperature control device, and uploads environmental parameters collected by the environmental sensing module and operating status parameters of the temperature control device to the cloud server. The data storage module is located on a cloud server and stores historical data formed by environmental parameters, operating status parameters, temperature control settings, comfort feedback information, and user location data. The artificial intelligence model module is located on a cloud server. It trains the artificial intelligence model based on historical data and outputs target temperature control parameters and control strategies according to current environmental parameters and user settings. The artificial intelligence model module has a version management and rollback / freeze governance mechanism for online model updates. It is used to roll back to the previous model version and enter the freeze window when the preset rollback conditions are met, and online updates of model parameters are prohibited in the freeze window. The control strategy generation module, located on a cloud server, generates control instructions for controlling the temperature control equipment based on the target temperature control parameters and control strategy, combined with preset constraints.

[0006] Preferably, the temperature control device includes: Adjust the execution power according to the control commands issued by the device control gateway module; Adjust the temperature within the target space based on changes in execution power; The operating status parameters generated by the changes in power and temperature are uploaded to the device control gateway module.

[0007] Preferably, the user interaction terminal module includes: A visual temperature control interface is generated based on the target temperature control parameters returned by the cloud server, and the target temperature control parameters and operating status information are presented. Temperature control settings are generated based on user adjustments or voice input and uploaded to the cloud server as input data for the artificial intelligence model module; Acquire user location data and upload it to a cloud server to determine the user's location status; The system acquires user-inputted comfort feedback information and uploads it to a cloud server as training data for the artificial intelligence model module.

[0008] Preferably, the device control gateway module includes: The operating status parameters uploaded by the temperature control equipment and the environmental parameters uploaded by the environmental sensing module are aggregated. Generate executable instructions for the temperature control device based on control commands issued by the cloud server; Send execution commands to the temperature control equipment to adjust the temperature within the target space.

[0009] Preferably, the data storage module includes: The environmental parameters uploaded by the environmental sensing module and the operating status parameters uploaded by the temperature control device are stored separately. Calculate the energy consumption statistics of temperature control equipment based on operating status parameters; Based on the energy consumption statistics, a historical energy consumption baseline is determined; Environmental parameters, user input data, and energy consumption statistics are provided to the artificial intelligence model module as training data.

[0010] Preferably, the artificial intelligence model module includes: Extract user temperature control preference characteristics by combining historical temperature control settings and comfort feedback information; A temperature control prediction model is trained based on user temperature control preferences and environmental parameters, and the model parameters are adjusted based on the training error. Target temperature control parameters and control strategies are generated based on the trained temperature control prediction model; The model parameters of the temperature control prediction model are updated online based on the comfort feedback information and the operating status parameters. The previous model version information is stored and a new model version information is generated during each online update. At the same time, the change in model parameters during this online update is limited based on a preset update magnitude threshold.

[0011] Preferably, the control strategy generation module includes: Based on the target temperature control parameters output by the artificial intelligence model module, the upper and lower temperature limit parameters and the operating capacity parameters of the temperature control equipment in the preset constraints are read; When the distance between the user's location data and the temperature control device meets the preset distance conditions, the target temperature control parameters are adjusted. Based on the adjusted target temperature control parameters and the operating capability parameters of the temperature control equipment, control commands are generated for the temperature control equipment to execute.

[0012] Preferably, the artificial intelligence model module further includes: Predict future environmental change trends based on the latest uploaded time-series environmental parameters; The weights of user temperature control preferences are adjusted based on predicted environmental change trends. The target temperature control parameters are updated based on the adjusted weights to adapt to future environmental changes; During the online update process, when the preset rollback conditions are met, the model is rolled back to the model version corresponding to the previous model version information, and a freeze window is started. In the freeze window, online updates of the model parameters of the temperature control prediction model are prohibited, and the target temperature control parameters and control strategy are generated using the rolled-back model version. The rollback conditions include: the comfort index calculated based on comfort feedback information deteriorates beyond a threshold within a continuous preset period; the temperature tracking error or temperature oscillation index calculated based on the target temperature control parameters and the target space environment parameters exceeds a threshold; or the increase in the energy consumption statistics result relative to the historical energy consumption baseline exceeds a threshold.

[0013] A remote temperature control method based on artificial intelligence, the method comprising: S1. Collect temperature and humidity parameters within the target space and generate time series data, and collect temperature parameters outside the target space and remove outliers from the collected results; S2. Adjust the temperature in the target space according to the change in the execution power of the temperature control equipment, and upload the operating status parameters formed by the change in execution power and temperature; S3. Display the target temperature control parameters and operating status information on the user interaction terminal, receive the user's temperature control setting information, comfort feedback information and user location data and upload them to the cloud server; S4. At the device control gateway, environmental parameters and operating status parameters are aggregated, and execution instructions that can be executed by the temperature control device are generated according to the control instructions issued by the cloud server and sent to the temperature control device. S5. Store environmental parameters, operating status parameters, user input data and user location data in the cloud server respectively, calculate energy consumption statistics based on the operating status parameters, and determine the historical energy consumption baseline based on the energy consumption statistics. S6. Train an artificial intelligence model based on the historical data, and output target temperature control parameters and control strategies according to the current environmental parameters and user settings. During operation, update the artificial intelligence model online according to comfort feedback information and operating status parameters. The online update adopts a model version management mechanism, and rolls back to the previous model version and enters a freeze window when the preset rollback conditions are met. Online updates of the model are prohibited in the freeze window. S7. Based on the target temperature control parameters and control strategy, and combined with preset constraints, generate control instructions for controlling the temperature control device; and when the distance between the user location data and the temperature control device meets the preset distance conditions, adjust the target temperature control parameters and then generate the control instructions.

[0014] This invention provides a remote temperature control system and its control method based on artificial intelligence. It has the following beneficial effects: 1. This invention integrates multiple sources in the cloud and calculates energy consumption statistics based on operating status parameters to determine historical energy consumption baselines. This enables the AI ​​model to have energy consumption references and constraints when outputting target temperature control parameters and control strategies. Combined with preset constraints, it generates executable instructions to achieve precise remote temperature control and significantly improve energy-saving effects and strategy rationality.

[0015] 2. This invention utilizes temperature control settings and comfort feedback to extract and correct user preferences, making the target temperature control parameters more aligned with individual needs, improving comfort and personalization, and enhancing adaptability to long-term use and environmental changes.

[0016] 3. This invention dynamically adjusts the target temperature control parameters based on the user's location and the distance to the device, making the strategy more timely when the user approaches or moves away, reducing unnecessary energy consumption and improving temperature control response efficiency and actual usability. Attached Figure Description

[0017] Figure 1 This is an architecture diagram of a remote temperature control system based on artificial intelligence according to the present invention; Figure 2 This is a flowchart of a remote temperature control method based on artificial intelligence according to the present invention. Detailed Implementation

[0018] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see the appendix Figure 1 This invention provides an artificial intelligence-based remote temperature control system, comprising: The environmental sensing module is used to collect environmental parameters of the target space and the operating environment parameters of the temperature control equipment; Furthermore, the environmental perception module includes: Temperature and humidity parameters within the target space are collected at preset time intervals to form time series data. Collect temperature parameters outside the target space and remove outliers from the collected results; The processed environmental parameters are uploaded to the cloud server as input data for the artificial intelligence model module.

[0020] Specifically, the environmental sensing module is installed within the target space to acquire environmental parameters used for temperature control. The module can collect temperature and humidity parameters within the target space via internally integrated temperature and humidity sensors, and can also collect temperature parameters outside the target space via externally installed temperature sensors, providing a data foundation for the subsequent artificial intelligence model module to generate target temperature control parameters. The environmental perception module collects temperature and humidity parameters within the target space at preset time intervals and correlates the collected data with corresponding time information to form time series data reflecting environmental change trends. This time series data enables the artificial intelligence model module to make predictions based on the patterns of environmental changes over time, thereby improving the accuracy of target temperature control parameters. After collecting temperature parameters outside the target space, the environmental perception module can process the collected results according to preset outlier removal rules. For example, when the trend of a certain collected value deviates from that of adjacent collected values ​​beyond a preset range, the collected value can be identified as an outlier and removed. By removing outlier collected results, the quality of environmental data uploaded to the cloud server can be improved, and the impact of data fluctuations on artificial intelligence models can be reduced.

[0021] Temperature control equipment executes temperature adjustment operations according to control commands and outputs its own operating status parameters; Furthermore, temperature control equipment includes: Adjust the execution power according to the control commands issued by the device control gateway module; Adjust the temperature within the target space based on changes in execution power; The operating status parameters generated by the changes in power and temperature are uploaded to the device control gateway module.

[0022] Specifically, the temperature control device is installed in the target space and performs temperature adjustment operations according to the control commands issued by the system. The temperature control device can adjust its execution power and obtain its own operating status in real time during the adjustment process, thereby achieving effective control of the space temperature and providing operating feedback data to the system. After receiving control commands from the device control gateway module, the temperature control device can adjust its own execution power. The control commands may include information such as the target execution power or the working status of the target temperature control device. The temperature control device performs corresponding power adjustment operations according to the control commands to make its output power consistent with the temperature control target set by the system. By adjusting the execution power according to the system commands, precise control of the heat output of the target space can be achieved. When the execution power of the temperature control device changes, it will directly affect the temperature change trend in the target space. After the power is adjusted, the temperature control device can generate temperature data reflecting the temperature change of the space based on the internal operating temperature or the external ambient temperature. Based on the impact of the execution power change on the temperature in the target space, the system can understand the temperature control capability of the device under different power levels, thereby providing a basis for the prediction and strategy generation of the subsequent artificial intelligence model. Therefore, while performing adjustment operations, the temperature control device also generates operating status parameters based on the execution power and the temperature change information caused by that execution power, and uploads them to the device control gateway module. The operating status parameters may include the current execution power, temperature change amplitude, temperature change rate, etc. After receiving the operating status parameters, the device control gateway module can upload them to the cloud server, enabling the data storage module and artificial intelligence model module to update the model and optimize the control strategy based on the actual operating conditions. By uploading the operating status parameters to the system, a data closed loop can be formed, improving the responsiveness and adaptability of the temperature control strategy to the real operating conditions.

[0023] The user interaction terminal module displays environmental parameters and the operating status of the temperature control equipment, receives temperature control settings and comfort feedback information input by the user, and obtains user location data and uploads the user location data to the cloud server. Furthermore, the user interaction terminal module includes: A visual temperature control interface is generated based on the target temperature control parameters returned by the cloud server, and the target temperature control parameters and operating status information are presented. Temperature control settings are generated based on user adjustments or voice input and uploaded to the cloud server as input data for the artificial intelligence model module; Acquire user location data and upload it to a cloud server to determine the user's location status; The system acquires user-inputted comfort feedback information and uploads it to a cloud server as training data for the artificial intelligence model module.

[0024] Specifically, the user interaction terminal module can be a mobile terminal device, such as a smartphone or tablet, used to communicate with the cloud server and display system operation information to the user. When the cloud server generates the target temperature control parameters, the user interaction terminal module can obtain the target temperature control parameters and generate a visual temperature control interface based on the preset interface layout. In this interface, the target temperature control parameters and the operating status information of the temperature control device can be presented simultaneously, such as the current execution power, operating mode or target space temperature change trend. Through visualization, users can intuitively understand the system's prediction results and current operating status, improving the system's transparency and user comprehensibility. After presenting the target temperature control parameters, the user interaction terminal module allows users to adjust the parameters according to their needs, such as adjusting the desired temperature, changing the operating mode, or setting temperature control requirements for a specific time period. The user interaction terminal module can structure the user's adjustment behavior into temperature control setting information and upload it to the cloud server. The cloud server can then use this temperature control setting information as one of the input data for the artificial intelligence model module, enabling the AI ​​model to continuously optimize the generation process of the target temperature control parameters based on the user's real-time adjustment behavior. By acquiring user adjustment behavior, the system's adaptability to user habits is enhanced, and the personalization of the temperature control strategy is improved. The user interaction terminal module simultaneously acquires user location data, which can be obtained from the positioning function of the mobile terminal, such as obtaining the user's current location based on satellite positioning or network positioning. The user interaction terminal module can upload the acquired user location data to the cloud server so that the system can determine the positional status between the user and the target space. When the distance between the user's position and the temperature control device changes, the temperature control strategy can be adjusted based on the positional status. For example, when the user is gradually approaching the target space, the temperature control strategy can be prepared in advance to ensure that the desired temperature is reached when the user enters the space. By uploading user location data, the response speed and initiative of the control strategy to user behavior can be improved. Simultaneously, the system acquires user-inputted comfort feedback. Users can provide simple feedback on the temperature control effect through the interface, such as selecting options like "too cold," "moderate," or "too hot." This feedback is uploaded to the cloud server, allowing the AI ​​model module to update its preference features based on the user's subjective comfort experience. As comfort feedback data accumulates, the AI ​​model can more accurately match user preferences, improving the accuracy of the system's generated target temperature control parameters, thus making the temperature control results more in line with user needs. Furthermore, users can issue temperature control-related commands via voice input. The user interaction terminal module collects the voice signals and converts them into corresponding control semantic information. The control semantic information includes at least the target temperature adjustment command, the operating mode switching command, or the on / off control command. The converted control semantic information is uploaded to the cloud server as temperature control setting information. It adopts the same data processing flow as the temperature control setting information entered by the user through the interface. The artificial intelligence model module combines the current environmental parameters and historical data to generate the corresponding target temperature control parameters and control strategies. The control strategy generation module generates the corresponding control commands to realize remote temperature control based on voice input. Meanwhile, the user interaction terminal module supports centralized management and remote control of a single temperature control device or multiple temperature control devices. The cloud server can assign a unified number and logical association to the temperature control devices connected to the system based on the device identification information uploaded by the device control gateway module, and obtain the corresponding environmental parameters and operating status parameters for different temperature control devices. When generating target temperature control parameters and control strategies, the artificial intelligence model module can analyze the target space environment parameters corresponding to different temperature control devices. The control strategy generation module can then generate corresponding control commands for single or multiple devices, thereby enabling remote temperature control and adjustment of one or more target spaces. When multiple temperature control devices are connected to the system, the control commands for each device can be generated and issued independently, giving the system the ability to expand from controlling a single device to simultaneously controlling multiple or all devices.

[0025] The device control gateway module communicates with the temperature control device and the cloud server, receives control commands from the cloud server and sends execution commands to the temperature control device, and uploads environmental parameters collected by the environmental sensing module and operating status parameters of the temperature control device to the cloud server. Furthermore, the device control gateway module includes: The operating status parameters uploaded by the temperature control equipment and the environmental parameters uploaded by the environmental sensing module are aggregated. Generate executable instructions for the temperature control device based on control commands issued by the cloud server; Send execution commands to the temperature control equipment to adjust the temperature within the target space.

[0026] Specifically, the device control gateway module performs data interaction and command transmission between the temperature control device and the cloud server. It can receive operating status parameters from the temperature control device and environmental parameters from the environmental sensing module, and aggregate data from different sources. This enables the cloud server to perform temperature control strategy analysis and decision-making based on complete data input. In other words, by aggregating the data, it can ensure that the data obtained by the cloud server is structured and continuous, thereby improving the reliability of temperature control strategy generation. Since the control commands issued by the cloud server are usually abstract target temperature control parameters or equipment operation strategies, the equipment control gateway module can convert the control commands into specific instructions that the temperature control equipment can directly execute, such as target execution power, working mode or running time, according to the operating capabilities of the temperature control equipment. By converting the cloud control commands into executable instructions, the system can ensure that the system accurately executes the temperature control strategy generated by the cloud server at the device level. The device control gateway module can send execution commands to the temperature control device via wired or wireless means. For example, it can control the device based on short-range wireless communication. When the temperature control device receives the execution command, it can adjust the power according to the command content, thereby regulating the temperature in the target space. By sending execution commands to the temperature control device, the system can respond in real time to the target temperature control parameters generated by the cloud server, improving the timeliness of temperature control.

[0027] The data storage module is located on a cloud server and stores historical data formed by environmental parameters, operating status parameters, temperature control settings, comfort feedback information, and user location data. Furthermore, the data storage module includes: The environmental parameters uploaded by the environmental sensing module and the operating status parameters uploaded by the temperature control device are stored separately. Calculate the energy consumption statistics of temperature control equipment based on operating status parameters; Determine the historical energy consumption baseline based on energy consumption statistics; Environmental parameters, user input data, and energy consumption statistics are provided to the artificial intelligence model module as training data.

[0028] Specifically, the data storage module is set up in the cloud server to centrally store and manage various types of data during the operation of the temperature control system. The data storage module receives environmental parameters from the environmental sensing module, operating status parameters from the temperature control equipment, and temperature control setting information, comfort feedback information, user location data, etc. uploaded by the user interaction terminal, and records them in chronological order to form a historical dataset, so as to provide the data foundation required for training and inference of the artificial intelligence model module. The data storage module can store environmental parameters uploaded by the environmental sensing module and operating status parameters uploaded by the temperature control device, and classify and manage them according to data source or data type. For example, environmental parameters form environmental time series data, and operating status parameters record information such as execution power and corresponding temperature change amplitude according to the device reporting time, so as to improve data retrieval and calling efficiency. In this embodiment, the data storage module also calculates the energy consumption statistics of the temperature control device based on the operating status parameters. For example, it calculates the energy consumption based on the execution power of different time periods and the preset power conversion coefficient, and records it by time to form an energy consumption sequence. For example, if the execution power is 50% of the rated power and the device runs for 1 hour, the energy consumption during that time period can be recorded according to the preset energy consumption calculation method. Meanwhile, the data storage module can determine the historical energy consumption baseline based on the energy consumption statistics results. The historical energy consumption baseline can be the benchmark value of the energy consumption statistics results within the preset statistical window, used to characterize the typical energy consumption level under the same or similar operating conditions. The preset statistical window can be set by day / week / month, and can aggregate the energy consumption statistics results within the same season, similar outdoor temperature range, same target temperature control parameter range, or same equipment operating capacity range to calculate the corresponding historical energy consumption baseline. The data storage module can associate and store the historical energy consumption baseline with its corresponding condition labels (such as outdoor temperature range, target temperature control parameter range, equipment operating capacity range, and time period identifier) ​​for subsequent model training, strategy generation, or rollback judgment calls. In one embodiment, the data storage module provides environmental parameters, user input data, energy consumption statistics, and historical energy consumption baselines to the artificial intelligence model module as training data. This enables the artificial intelligence model module to learn user temperature control preferences and environmental change patterns, and to generate more reasonable target temperature control parameters and control strategies based on energy consumption performance. It also supports the identification of abnormal increases in energy consumption.

[0029] The artificial intelligence model module is located on a cloud server. It trains the artificial intelligence model based on historical data and outputs target temperature control parameters and control strategies according to current environmental parameters and user settings. Among them, the artificial intelligence model module has a version management and rollback / freeze governance mechanism for online model updates. It is used to roll back to the previous model version and enter the freeze window when the preset rollback conditions are met, and online updates of model parameters are prohibited in the freeze window. Furthermore, the artificial intelligence model module includes: Extract user temperature control preference characteristics by combining historical temperature control settings and comfort feedback information; A temperature control prediction model is trained based on user temperature control preferences and environmental parameters, and the model parameters are adjusted based on the training error. Target temperature control parameters and control strategies are generated based on the trained temperature control prediction model; The model parameters of the temperature control prediction model are updated online based on comfort feedback information and operating status parameters. The previous model version information is stored and a new model version information is generated during each online update. At the same time, the change in model parameters during this online update is limited based on a preset update magnitude threshold.

[0030] Specifically, the artificial intelligence model module is located on a cloud server and is used to build temperature control prediction capabilities based on historical data. It outputs target temperature control parameters and control strategies according to current environmental parameters and user settings. That is, the artificial intelligence model module can learn by comparing historical temperature control settings, comfort feedback information, environmental parameters, and temperature control equipment operating status parameters to form a temperature control prediction model that reflects user preferences and the thermal dynamics of the space. Through this prediction model, the system can generate control strategies that meet user experience and energy-saving requirements. The artificial intelligence model module has a version management and rollback / freeze governance mechanism for online model updates. When preset rollback conditions are met, it rolls back to the previous model version and enters a freeze window, during which online updates of model parameters are prohibited.

[0031] The AI ​​model module can extract user temperature control preference features based on historical temperature control setting information and comfort feedback information. For example, when a user repeatedly adjusts the temperature control setting to a target temperature under a specific ambient temperature, the AI ​​model module can identify this behavior as a user preference trend. At the same time, comfort feedback information can be used to correct user preferences, making the model more accurately meet user experience needs. For example, when a user frequently selects "too cold" feedback within a certain period of time, the temperature control device can automatically adjust its preference features, causing the model to favor higher target temperature control parameters during prediction. By analyzing and improving temperature control preferences based on user historical behavior, the system can generate target temperature control parameters that better meet user expectations in subsequent predictions. Meanwhile, the AI ​​model module can train a temperature control prediction model based on user temperature control preferences and environmental parameters, and adjust the model parameters based on training errors. In one optional approach, the AI ​​model module can optimize the model using backpropagation. For example, the temperature control prediction error can be measured using the following loss function: ; in: The target temperature control parameters are predicted by the model based on environmental parameters and preference characteristics. The temperature control parameters currently set by the user or the temperature control target after correction based on comfort feedback; The AI ​​model module can update the model parameters of the temperature control prediction model online based on comfort feedback information and operating status parameters. It stores the previous model version information and generates a new model version information with each online update, while limiting the amount of change in model parameters during the current online update based on a preset update magnitude threshold. For example, when an online update is triggered, the current model parameters are recorded as... And save the parameters from the previous version. Candidate parameters are obtained based on the new samples. Post-calculation And limit the amplitude by setting a preset update amplitude threshold. In one optional method, the following limiting method can be used: ; in: To preset the update amplitude threshold, This is the stable term; and new model parameters are obtained: ; In addition, the AI ​​model module can be configured with rollback conditions and a freeze window mechanism. When the preset rollback conditions are met, the model will roll back to the previous version and activate the freeze window. During the freeze window, online updates to the model parameters are prohibited, and the target temperature control parameters and control strategies will be output using the rolled-back model version. Rollback conditions may include one or a combination of the following: the comfort index calculated based on comfort feedback information deteriorates beyond a threshold within a consecutive preset period; the temperature tracking error or temperature oscillation index calculated based on the target temperature control parameters and environmental parameters exceeds a threshold; or the increase in energy consumption statistics calculated based on operating status parameters relative to the historical energy consumption baseline exceeds a threshold.

[0032] The control strategy generation module, located on a cloud server, generates control commands for controlling temperature control equipment based on target temperature control parameters and control strategies, combined with preset constraints.

[0033] Furthermore, the control strategy generation module includes: Based on the target temperature control parameters output by the artificial intelligence model module, the upper and lower temperature limit parameters and the operating capacity parameters of the temperature control equipment in the preset constraints are read; When the distance between the user's location data and the temperature control device meets the preset distance conditions, the target temperature control parameters are adjusted. Based on the adjusted target temperature control parameters and the operating capability parameters of the temperature control equipment, control commands are generated for the temperature control equipment to execute.

[0034] Specifically, the control strategy generation module is located in the cloud server. It is used to generate control instructions for controlling the temperature control equipment based on the target temperature control parameters output by the artificial intelligence model module and in combination with preset constraints. The control strategy generation module can obtain the target temperature control parameters output by the artificial intelligence model module, as well as the preset upper and lower temperature limit parameters and the operating capacity parameters of the temperature control equipment. This ensures that the generated final control instructions can meet the user's comfort needs and avoid problems such as overload or energy waste caused by the temperature control equipment operating beyond its capacity. By introducing preset constraints in the process of generating control instructions, the safety and energy-saving effect of the system temperature control can be improved. The control strategy generation module can read the upper and lower temperature limits and the operating capacity parameters of the temperature control equipment from the preset constraints based on the target temperature control parameters output by the artificial intelligence model module. For example, the system can set a maximum temperature limit for the target temperature control parameters. With the lowest temperature limit And set the maximum operating power for the temperature control equipment. and minimum execution power When the artificial intelligence model module generates the target temperature control parameters Subsequently, the control strategy generation module can adopt the following constraint processing method: ; in, These are the target temperature control parameters after being constrained by upper and lower temperature limits; These are the lower temperature limit and the upper temperature limit, respectively.

[0035] By using the above methods, it can be ensured that the target temperature control parameters are within the system's allowable range, avoiding overheating or overcooling, and improving the safety and rationality of temperature control commands. The control strategy generation module can adjust the target temperature control parameters based on the distance between the user's location data and the temperature control device. The user's location data is uploaded by the user interaction terminal module, and the cloud server can calculate the distance between the user's current location and the target space where the temperature control device is located. When this distance meets the system's preset distance conditions, such as a distance threshold... The control strategy generation module can further adjust the constrained target temperature control parameters. For example, when a user approaches the target space, the system can set the temperature control parameters slightly higher than the normal requirement so that the user reaches a comfortable temperature when entering the space. When the user moves away from the target space, the temperature control parameters can be set slightly lower to achieve energy-saving control. The adjustment can be done in the following ways: ; in: The target temperature control parameters after final position adjustment; For user distance The adjustment function can be a monotonically decreasing function, for example, when When the time is positive, The value is zero or negative. Based on the above implementation method, the control strategy generation module can generate control commands for the temperature control equipment to execute based on the adjusted target temperature control parameters and the operating capability parameters of the temperature control equipment. For example, the control strategy generation module can calculate the required execution power based on the difference between the target temperature control parameters and the current space temperature, so that the execution power is within the specified range. and Within a certain range, the generated control commands can include the target execution power, operating mode, or running time, and are sent to the temperature control device for execution by the device control gateway module. By generating control commands based on the final adjusted target temperature control parameters, the response speed and temperature control accuracy of the temperature control device in actual operation can be improved, enabling users to obtain a more stable and energy-saving temperature control experience.

[0036] Furthermore, the artificial intelligence model module also includes: Predict future environmental change trends based on the latest uploaded time-series environmental parameters; The weights of user temperature control preferences are adjusted based on predicted environmental change trends. The target temperature control parameters are updated based on the adjusted weights to adapt to future environmental changes; During the online update process, when the preset rollback conditions are met, the model is rolled back to the model version corresponding to the previous model version information, and a freeze window is started. In the freeze window, online updates of the model parameters of the temperature control prediction model are prohibited, and the target temperature control parameters and control strategy are generated using the rolled-back model version. The rollback conditions include: the comfort index calculated based on comfort feedback information deteriorates beyond a threshold within a continuous preset period; the temperature tracking error or temperature oscillation index calculated based on the target temperature control parameters and the target space environment parameters exceeds a threshold; or the increase in energy consumption statistics relative to the historical energy consumption baseline exceeds a threshold.

[0037] Specifically, the AI ​​model module can also predict future environmental change trends based on the latest uploaded time-series environmental parameters. These time-series environmental parameters can include data on changes in indoor temperature, indoor humidity, and outdoor temperature over time. The AI ​​model module can use time-series prediction methods to obtain the environmental change trend results within a preset future period, such as the rising or falling trend of outdoor temperature and its magnitude over a future period. The artificial intelligence model module can adjust the weights of user temperature control preference features based on predicted environmental change trends. These weights characterize the degree of influence of user preferences during the generation of target temperature control parameters. When predicted environmental change trends indicate significant changes in the external environment, the weights of user preferences can be increased or decreased accordingly, thereby making the subsequently generated target temperature control parameters more aligned with comfort requirements under future environmental conditions. Furthermore, the artificial intelligence model module updates the target temperature control parameters based on the adjusted weights to adapt to future environmental changes. The target temperature control parameters can be the target temperature or the target temperature range. For example, the "target temperature obtained from user preference" and the "compensation temperature obtained based on future environmental changes" can be fused to obtain the updated target temperature control parameters, and a control strategy can be formed accordingly. Meanwhile, during the online update process, the artificial intelligence model module can set rollback conditions and execute the rollback and freeze window mechanism when the rollback conditions are met. When the preset rollback conditions are met, the artificial intelligence model module rolls back to the model version corresponding to the previous model version information and starts the freeze window. In the freeze window, online updates of the model parameters of the temperature control prediction model are prohibited, and the target temperature control parameters and control strategy are generated by using the rolled-back model version. In one implementation, the rollback conditions may include one or a combination of the following: The comfort index obtained based on comfort feedback information deteriorates beyond a threshold within a continuous preset period of time; the comfort index can be obtained by statistical analysis of comfort feedback within a preset period of time and compared with the threshold. The temperature tracking error or temperature oscillation index calculated based on the target temperature control parameters and target space environment parameters exceeds the threshold; for ease of implementation, the temperature tracking error can be calculated in the following way: ; The increase in energy consumption statistics relative to the historical energy consumption baseline exceeds a threshold; for ease of implementation, the energy consumption increase ratio can be calculated: ; in: This refers to the energy consumption statistics within the current statistical window. As the historical energy consumption baseline; when When the preset threshold is exceeded, a rollback is triggered and the window is frozen.

[0038] Through the above implementation methods, the artificial intelligence model module can dynamically adjust the preference weights and update the target temperature control parameters based on the prediction of future environmental change trends. At the same time, the system's operational stability and reliability are improved through rollback and window freezing mechanisms during the online update process.

[0039] Please see the appendix Figure 2 A remote temperature control method based on artificial intelligence, the method includes: S1. Collect temperature and humidity parameters within the target space and generate time series data, and collect temperature parameters outside the target space and remove outliers from the collected results; S2. Adjust the temperature in the target space according to the change in the execution power of the temperature control equipment, and upload the operating status parameters formed by the change in execution power and temperature; S3. Display the target temperature control parameters and operating status information on the user interaction terminal, receive the user's temperature control setting information, comfort feedback information and user location data and upload them to the cloud server; S4. At the device control gateway, environmental parameters and operating status parameters are aggregated, and execution instructions that can be executed by the temperature control device are generated according to the control instructions issued by the cloud server and sent to the temperature control device. S5. Store environmental parameters, operating status parameters, user input data and user location data in the cloud server, calculate energy consumption statistics based on operating status parameters, and determine historical energy consumption baseline based on energy consumption statistics. S6. Train an artificial intelligence model based on historical data, and output target temperature control parameters and control strategies according to current environmental parameters and user settings. During operation, update the artificial intelligence model online based on comfort feedback information and operating status parameters. The online update adopts a model version management mechanism, and rolls back to the previous model version and enters a freeze window when the preset rollback conditions are met. Online model updates are prohibited in the freeze window. S7. Based on the target temperature control parameters and control strategy, and combined with preset constraints, generate control commands for controlling the temperature control equipment; and when the distance between the user's location data and the temperature control equipment meets the preset distance conditions, adjust the target temperature control parameters and then generate control commands.

[0040] Specifically, indoor temperature and humidity are collected and compiled into a time series, while outdoor temperature is collected and outliers are removed; temperature control devices adjust the space temperature according to the execution power and upload operating status parameters; user terminals display information and upload temperature control settings, comfort feedback, and location data; the gateway aggregates data and issues execution commands according to cloud instructions; the cloud stores various types of data, calculates energy consumption statistics, and determines historical energy consumption baselines; an artificial intelligence model is trained based on historical data and outputs target temperature control parameters and control strategies, which are updated online during operation based on feedback and operating status, using version management and rolling back when rollback conditions are met and entering a freeze window to prevent updates; finally, control commands are generated based on constraints, and when distance conditions are met, the target temperature control parameters are corrected before generating control commands.

[0041] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A remote temperature control system based on artificial intelligence, characterized in that, include: The environmental sensing module is used to collect environmental parameters of the target space and the operating environment parameters of the temperature control equipment; Temperature control equipment executes temperature adjustment operations according to control commands and outputs its own operating status parameters; The user interaction terminal module displays environmental parameters and the operating status of the temperature control equipment, receives temperature control settings and comfort feedback information input by the user, and obtains user location data and uploads the user location data to the cloud server. The device control gateway module communicates with the temperature control device and the cloud server, receives control commands from the cloud server and sends execution commands to the temperature control device, and uploads environmental parameters collected by the environmental sensing module and operating status parameters of the temperature control device to the cloud server. The data storage module is located on a cloud server and stores historical data formed by environmental parameters, operating status parameters, temperature control settings, comfort feedback information, and user location data. The artificial intelligence model module is located on a cloud server. It trains the artificial intelligence model based on historical data and outputs target temperature control parameters and control strategies according to current environmental parameters and user settings. The artificial intelligence model module has a version management and rollback / freeze governance mechanism for online model updates. It is used to roll back to the previous model version and enter the freeze window when the preset rollback conditions are met, and online updates of model parameters are prohibited in the freeze window. The control strategy generation module, located on a cloud server, generates control instructions for controlling the temperature control equipment based on the target temperature control parameters and control strategy, combined with preset constraints.

2. The remote temperature control system based on artificial intelligence according to claim 1, characterized in that, The environment sensing module includes: Temperature and humidity parameters within the target space are collected at preset time intervals to form time series data. Collect temperature parameters outside the target space and remove outliers from the collected results; The processed environmental parameters are uploaded to the cloud server as input data for the artificial intelligence model module.

3. The remote temperature control system based on artificial intelligence according to claim 1, characterized in that, The temperature control device includes: Adjust the execution power according to the control commands issued by the device control gateway module; Adjust the temperature within the target space based on changes in execution power; The operating status parameters generated by the changes in power and temperature are uploaded to the device control gateway module.

4. The remote temperature control system based on artificial intelligence according to claim 1, characterized in that, The user interaction terminal module includes: A visual temperature control interface is generated based on the target temperature control parameters returned by the cloud server, and the target temperature control parameters and operating status information are presented. Temperature control settings are generated based on user adjustments or voice input and uploaded to the cloud server as input data for the artificial intelligence model module; Acquire user location data and upload it to a cloud server to determine the user's location status; The system acquires user-inputted comfort feedback information and uploads it to a cloud server as training data for the artificial intelligence model module.

5. The remote temperature control system based on artificial intelligence according to claim 1, characterized in that, The device control gateway module includes: The operating status parameters uploaded by the temperature control equipment and the environmental parameters uploaded by the environmental sensing module are aggregated. Generate executable instructions for the temperature control device based on control commands issued by the cloud server; Send execution commands to the temperature control equipment to adjust the temperature within the target space.

6. The remote temperature control system based on artificial intelligence according to claim 1, characterized in that, The data storage module includes: The environmental parameters uploaded by the environmental sensing module and the operating status parameters uploaded by the temperature control device are stored separately. Calculate the energy consumption statistics of temperature control equipment based on operating status parameters; Based on the energy consumption statistics, a historical energy consumption baseline is determined; Environmental parameters, user input data, and energy consumption statistics are provided to the artificial intelligence model module as training data.

7. The remote temperature control system based on artificial intelligence according to claim 6, characterized in that, The artificial intelligence model module includes: Extract user temperature control preference characteristics by combining historical temperature control settings and comfort feedback information; A temperature control prediction model is trained based on user temperature control preferences and environmental parameters, and the model parameters are adjusted based on the training error. Target temperature control parameters and control strategies are generated based on the trained temperature control prediction model; The model parameters of the temperature control prediction model are updated online based on the comfort feedback information and the operating status parameters. The previous model version information is stored and a new model version information is generated during each online update. At the same time, the change in model parameters during this online update is limited based on a preset update magnitude threshold.

8. The remote temperature control system based on artificial intelligence according to claim 1, characterized in that, The control strategy generation module includes: Based on the target temperature control parameters output by the artificial intelligence model module, the upper and lower temperature limit parameters and the operating capacity parameters of the temperature control equipment in the preset constraints are read; When the distance between the user's location data and the temperature control device meets the preset distance conditions, the target temperature control parameters are adjusted. Based on the adjusted target temperature control parameters and the operating capability parameters of the temperature control equipment, control commands are generated for the temperature control equipment to execute.

9. A remote temperature control system based on artificial intelligence according to claim 7, characterized in that, The artificial intelligence model module also includes: Predict future environmental change trends based on the latest uploaded time-series environmental parameters; The weights of user temperature control preferences are adjusted based on predicted environmental change trends. The target temperature control parameters are updated based on the adjusted weights to adapt to future environmental changes; During the online update process, when the preset rollback conditions are met, the model is rolled back to the model version corresponding to the previous model version information, and a freeze window is started. In the freeze window, online updates of the model parameters of the temperature control prediction model are prohibited, and the target temperature control parameters and control strategy are generated using the rolled-back model version. The rollback conditions include: the comfort index calculated based on comfort feedback information deteriorates beyond a threshold within a continuous preset period; the temperature tracking error or temperature oscillation index calculated based on the target temperature control parameters and the target space environment parameters exceeds a threshold; or the increase in the energy consumption statistics result relative to the historical energy consumption baseline exceeds a threshold.

10. A remote temperature control method based on artificial intelligence, characterized in that, For use in an artificial intelligence-based remote temperature control system according to any one of claims 1-9, the method comprises: S1. Collect temperature and humidity parameters within the target space and generate time series data, and collect temperature parameters outside the target space and remove outliers from the collected results; S2. Adjust the temperature in the target space according to the change in the execution power of the temperature control equipment, and upload the operating status parameters formed by the change in execution power and temperature; S3. Display the target temperature control parameters and operating status information on the user interaction terminal, receive the user's temperature control setting information, comfort feedback information and user location data and upload them to the cloud server; S4. At the device control gateway, environmental parameters and operating status parameters are aggregated, and execution instructions that can be executed by the temperature control device are generated according to the control instructions issued by the cloud server and sent to the temperature control device. S5. Store environmental parameters, operating status parameters, user input data and user location data in the cloud server respectively, calculate energy consumption statistics based on the operating status parameters, and determine the historical energy consumption baseline based on the energy consumption statistics. S6. Train an artificial intelligence model based on the historical data, and output target temperature control parameters and control strategies according to the current environmental parameters and user settings. During operation, update the artificial intelligence model online according to comfort feedback information and operating status parameters. The online update adopts a model version management mechanism, and rolls back to the previous model version and enters a freeze window when the preset rollback conditions are met. Online updates of the model are prohibited in the freeze window. S7. Based on the target temperature control parameters and control strategy, and combined with preset constraints, generate control instructions for controlling the temperature control device; and when the distance between the user location data and the temperature control device meets the preset distance conditions, adjust the target temperature control parameters and then generate the control instructions.

Images