A kettle control method based on internet of things change and a kettle

Abstract

The application discloses a hot water kettle control method based on Internet of Things change and a hot water kettle. The method comprises the following steps: inputting behavior characteristic data, trajectory characteristic data, time characteristic data and event characteristic data into a preset distillation LSTM model to obtain a target temperature; and controlling a heating start time and a heating power of the hot water kettle according to the target temperature and a local boiling point. The application realizes a distributed collaborative decision intelligent hot water kettle control method by fusing multi-source environmental information and behavior data, so as to improve environmental adaptability, network robustness and user service experience of the intelligent hot water kettle, and simultaneously optimize energy use efficiency.

Description

Technical Field

[0001] This invention relates to the field of home appliance control, and more particularly to a kettle control method and kettle based on Internet of Things (IoT) changes. Background Technology

[0002] With the rapid development of IoT technology, smart home appliances are gradually entering thousands of households. As a small appliance used very frequently in daily life, the intelligent upgrade of kettles has also attracted much attention. However, existing intelligent kettle control systems still have many limitations and cannot meet the growing personalized and intelligent needs of users.

[0003] The main problems with smart kettles currently on the market are as follows:

[0004] First, they have poor environmental adaptability. Traditional kettles only have a fixed boiling point (usually 100°C) and cannot automatically adjust according to environmental factors such as altitude and air pressure. In high-altitude areas, due to the reduced atmospheric pressure, the actual boiling point of water will drop significantly (e.g., at an altitude of 3000 meters, the boiling point is about 90°C), causing traditional kettles to exhibit a "false boiling" phenomenon. This not only fails to achieve the expected heating effect but may also lead to false triggering of the dry-boil protection, affecting the user experience. Although some high-end products are equipped with air pressure sensors, relying solely on the environmental perception of a single device limits the accuracy and reliability of the data.

[0005] Secondly, they are highly dependent on networks and lack collaborative capabilities. Most existing smart kettles rely on stable Wi-Fi connections for remote control; when the network fluctuates or is interrupted, their smart features are essentially disabled. Furthermore, each device operates in isolation, unable to effectively collaborate with other smart home devices. For example, multiple kettles may operate at high power simultaneously, overloading the home's electrical circuits; or they may fail to predict water demand based on the user's actual location and behavior patterns, resulting in energy waste. Summary of the Invention

[0006] This invention provides a kettle control method and kettle based on changes in the Internet of Things (IoT). By dynamically changing the IoT network topology and integrating multi-source environmental information and behavioral data, the method enables distributed collaborative decision-making for intelligent kettle control, thereby improving the environmental adaptability, network robustness, and user service experience of the intelligent kettle, while also optimizing energy efficiency.

[0007] To achieve the above objectives, a first aspect of this application provides a kettle control method based on Internet of Things (IoT) changes, specifically including:

[0008] Nearby stable devices are determined through a Bluetooth Mesh network; the nearby stable devices refer to IoT devices that respond to Mesh messages within a first preset time threshold and have an online time greater than a second preset time threshold within the same Bluetooth Mesh network.

[0009] Obtain environmental parameter messages broadcast by all nearby stable devices in the current area, and determine the local boiling point based on the environmental parameter messages;

[0010] Based on the usage record data of the kettle within the third preset time range and the usage record data of all the nearby stable devices, behavioral feature data is obtained; the number of dimensions of the behavioral feature data is equal to the number of regions, and each dimension corresponds to one region;

[0011] Based on the passive infrared sensors in each area, the user's dwell time, starting point, and ending point in each area are obtained to acquire trajectory feature data.

[0012] Based on the calendar data obtained from the Bluetooth Mesh hub, time feature data and event feature data are obtained;

[0013] The target temperature is obtained by inputting the behavioral feature data, trajectory feature data, time feature data, and event feature data into a preset distillation LSTM model.

[0014] The heating start time and heating power of the kettle are controlled according to the target temperature and the local boiling point.

[0015] In one possible implementation of the first aspect, obtaining all environmental parameter messages broadcast by the neighboring stable devices in the current area and determining the local boiling point based on the environmental parameter messages specifically includes:

[0016] Obtain all environmental parameter messages broadcast by the nearby stable devices in the current area, and extract the regional air pressure value and regional humidity value from the environmental parameter messages;

[0017] The theoretical boiling point is obtained based on the air pressure value of the region.

[0018] The local boiling point is obtained by adjusting the theoretical boiling point based on the regional humidity value and water quality information.

[0019] In one possible implementation of the first aspect, the extraction of the environmental parameter messages to obtain regional air pressure and regional humidity values ​​specifically includes:

[0020] Based on the Mesh messages of the kettle and the Mesh messages of the neighboring stable devices in each current region, calculate the functional similarity value between the kettle and the neighboring stable devices in each current region;

[0021] In the Bluetooth Mesh network, among the set of neighboring stable devices that broadcast barometric pressure messages in the current area, the barometric pressure value corresponding to the neighboring stable device with the greatest functional similarity is selected as the area barometric pressure value.

[0022] In the Bluetooth Mesh network, among the set of neighboring stable devices that broadcast humidity messages in the current area, the humidity value corresponding to the neighboring stable device with the greatest functional similarity is selected as the area humidity value.

[0023] In one possible implementation of the first aspect, calculating the functional similarity value between the kettle and the neighboring stable devices in each current area based on the kettle's mesh message and the mesh messages of the neighboring stable devices in each current area specifically includes:

[0024] Based on the switching parameter values, power outage recovery strategy values, numerical fluctuation values, and average power consumption of the kettle, a general model vector of the kettle is obtained as a first direction vector; based on the switching parameter values, power outage recovery strategy values, numerical fluctuation values, and average power consumption of the nearby stable devices in each current region, a general model vector of the nearby stable devices in each current region is obtained as a second direction vector; the projection value of the second direction vector onto the first direction vector is used as a first similarity value;

[0025] Based on the sensor parameter description values, sensor parameter measurement values, and reporting frequency values ​​of the kettle, the sensor model vector of the kettle is obtained as a third direction vector; based on the sensor parameter description values, sensor parameter measurement values, and reporting frequency values ​​of the nearby stable devices in each current region, the sensor model vector of the nearby stable devices in each current region is obtained as a fourth direction vector; the projection value of the fourth direction vector onto the third direction vector is used as a second similarity value;

[0026] The first similarity value and the second similarity value are weighted and summed to obtain the functional similarity value between the kettle and the neighboring stable devices in each current region.

[0027] In one possible implementation of the first aspect, before obtaining all environmental parameter messages broadcast by the neighboring stable devices in the current area and determining the local boiling point based on the environmental parameter messages, the process specifically includes:

[0028] Users can divide all IoT devices into zones based on the room where the IoT devices belong.

[0029] In one possible implementation of the first aspect, obtaining behavioral feature data based on the usage record data of the kettle within a third preset time range and the usage record data of all nearby stable devices specifically includes:

[0030] Within the third preset time range, the most recent usage time, usage frequency, and average power consumption of the kettle and all nearby stable devices in the area where the kettle is located are statistically analyzed to obtain a behavioral feature sub-data.

[0031] The most recent usage time, usage frequency, and average power consumption of each of the nearby stable devices in other areas are statistically analyzed to obtain multiple behavioral feature sub-data; each behavioral feature sub-data corresponds to one area.

[0032] All the behavioral feature sub-data are merged to obtain the behavioral feature data.

[0033] In one possible implementation of the first aspect, time feature data and event feature data are obtained based on calendar data acquired from the Bluetooth Mesh hub, specifically including:

[0034] Convert numerical information in calendar data that conforms to preset standards into time feature data;

[0035] The remaining calendar data is divided into blocks according to preset natural language processing rules, and each block is mapped to a quantized value; all the quantized values ​​are merged to obtain event feature data.

[0036] In one possible implementation of the first aspect, the step of inputting the behavioral feature data, trajectory feature data, time feature data, and event feature data into a pre-defined distillation LSTM model to obtain the target temperature and target usage time specifically includes:

[0037] The Bluetooth Mesh hub obtains the preset running parameters of the distilled LSTM model from the cloud;

[0038] The distillation LSTM model is updated while the kettle is not in operation;

[0039] The behavioral feature data, trajectory feature data, time feature data, and event feature data are input into the distillation LSTM model to obtain the user's desired expected temperature, expected probability, and target usage time.

[0040] The target temperature is the product of the user's desired temperature and the expected probability.

[0041] In one possible implementation of the first aspect, controlling the heating start time and heating power of the kettle based on the target temperature, the local boiling point, and the target usage time specifically includes:

[0042] Calculate the time difference between the target usage time and the current time.

[0043] The product of the time difference and the expected probability is used as the heating time;

[0044] Obtain the theoretical boiling point parameter value from the distillation LSTM model, and use the difference between the theoretical boiling point parameter value and the local boiling point as the boiling point deviation value; adjust the target temperature value according to the boiling point deviation value.

[0045] Before the heating time expires, while ensuring that the water temperature in the kettle reaches the target temperature value, the average heating power of the kettle is reduced to the maximum extent.

[0046] A second aspect of this application provides a kettle based on Internet of Things (IoT) changes, comprising:

[0047] The device determination module is used to determine nearby stable devices through a Bluetooth Mesh network; the nearby stable devices refer to IoT devices that respond to Mesh messages within a first preset time threshold and have an online time of more than a second preset time threshold within the same Bluetooth Mesh network.

[0048] The boiling point determination module is used to obtain environmental parameter messages broadcast by all nearby stable devices in the current area, and determine the local boiling point based on the environmental parameter messages.

[0049] The behavior feature module is used to obtain behavior feature data based on the usage record data of this device and the usage record data of all the nearby stable devices within a third preset time range; the number of dimensions of the behavior feature data is equal to the number of regions, and each dimension corresponds to one region;

[0050] The trajectory feature module is used to obtain the user's dwell time, starting point, and ending point in each area based on the passive infrared sensors in each area, and to obtain trajectory feature data.

[0051] The calendar feature module is used to obtain time feature data and event feature data based on the calendar data obtained from the Bluetooth Mesh hub;

[0052] The temperature calculation module is used to input the behavioral feature data, trajectory feature data, time feature data and event feature data into a preset distillation LSTM model to obtain the target temperature.

[0053] The equipment control module is used to control the heating start time and heating power of the equipment according to the target temperature and the local boiling point.

[0054] Compared to existing technologies, this embodiment provides a kettle control method and kettle based on IoT-based changes. It constructs a home IoT collaborative system via a Bluetooth Mesh network, first selecting stable devices that respond promptly and remain online for extended periods as data sources. Utilizing environmental parameters broadcast by these devices, the system dynamically calculates the local boiling point, taking into account altitude, air pressure, and humidity, thus addressing the "false boiling" problem in high-altitude areas. Simultaneously, it integrates usage records from multiple regions to construct behavioral feature data matching the number of household areas. This data, combined with user activity trajectories obtained from infrared sensors and time event features analyzed from calendar data, is input into a lightweight distillation LSTM model to predict user demand. This model runs efficiently on edge devices, outputting a target temperature that considers usage probability. Finally, based on the difference between the target temperature and the local boiling point, the system intelligently plans the heating start time and segmented power strategies, optimizing energy consumption while ensuring timely water supply. The technical effects are significant: improved boiling point calculation accuracy, increased demand prediction accuracy, reduced energy consumption, reduced circuit overload risk, truly achieving personalized drinking water services tailored to users, and greatly enhancing the kettle's adaptability in complex environments and the user experience. Attached Figure Description

[0055] Figure 1 This is a flowchart illustrating a method for controlling a kettle based on changes in the Internet of Things, according to an embodiment of the present invention.

[0056] Figure 2 This is a schematic diagram of the module structure of a kettle based on Internet of Things (IoT) according to an embodiment of the present invention. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be clearly and completely described below 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.

[0058] To resolve the above issues, please refer to [link / reference]. Figure 1 An embodiment of the present invention provides a kettle control method based on Internet of Things (IoT) changes, specifically including:

[0059] S10. Determine nearby stable devices through a Bluetooth Mesh network; the nearby stable devices refer to IoT devices that respond to Mesh messages within a first preset time threshold and have an online time greater than a second preset time threshold within the same Bluetooth Mesh network.

[0060] S11. Obtain all environmental parameter messages broadcast by the nearby stable devices in the current area, and determine the local boiling point based on the environmental parameter messages.

[0061] S12. Based on the usage record data of the kettle within the third preset time range and the usage record data of all the nearby stable devices, behavioral feature data is obtained; the number of dimensions of the behavioral feature data is equal to the number of regions, and each dimension corresponds to one region.

[0062] S13. Based on the passive infrared sensors in each area, obtain the user's dwell time, starting point, and ending point in each area to obtain trajectory feature data.

[0063] S14. Based on the calendar data obtained from the Bluetooth Mesh hub, obtain time feature data and event feature data.

[0064] S15. Input the behavioral feature data, trajectory feature data, time feature data and event feature data into the preset distillation LSTM model to obtain the target temperature.

[0065] S16. Control the heating start time and heating power of the kettle according to the target temperature and the local boiling point.

[0066] The S10 utilizes Bluetooth Mesh networking's multi-hop communication capability to periodically send probe messages to the network and record responses. Simultaneously, it continuously monitors the online status of each device, calculating daily online time. Two key thresholds are set: the first is a message response time threshold (devices must respond to probe messages within a specified time), and the second is a daily online time threshold (devices must meet a minimum online time requirement). Only devices that simultaneously meet both conditions are considered "stable proximity devices." This step effectively filters out devices that are consistently online and have reliable communication, eliminating interference from temporarily connected devices and devices with unstable signals. By establishing a device trust mechanism, the reliability of subsequent data collection is significantly improved, providing a high-quality data source for environmental perception and behavioral analysis. Furthermore, it accurately distinguishes between resident home devices and temporary visitor devices, avoiding decision-making errors caused by device instability.

[0067] S11 collects environmental parameters broadcast by all nearby stable devices in the current area, focusing on extracting air pressure and humidity values. Based on physical principles, the theoretical boiling point is first calculated based on the air pressure value (the lower the air pressure, the lower the boiling point). Then, the theoretical boiling point is corrected by combining humidity values ​​and known water quality information (hard water / soft water). Data fusion methods such as weighted averaging are used to process the measurement results from multiple devices, improving the accuracy of the final boiling point calculation.

[0068] Before executing S12, users need to divide their IoT devices into zones based on their actual living space (e.g., kitchen, living room, bedroom, etc.). In S12, the system applying this method statistically analyzes the usage data of devices in each zone over a preset time period. The behavioral data for each zone is used as one dimension of a feature vector, forming a multi-dimensional behavioral feature dataset that reflects differences in users' water usage habits across different zones. This step achieves refined division of the home space, accurately identifying differences in water usage habits across different living areas. For example, the kitchen area may prefer hot water for cooking, while the bedroom area may prefer warm water for drinking before bed. This regionalized behavioral modeling avoids incorrectly applying usage habits from one area to others, supports personalized services for multi-user households, and provides accurate behavioral data for subsequent predictions.

[0069] The S13 system is equipped with passive infrared sensors in each functional area to detect human activity in real time. The system accurately records the time of entry and exit for each user, calculating the duration of stay and the sequence of movement. By analyzing users' daily activity patterns, such as the fixed route from the bedroom to the kitchen in the morning, it constructs user activity trajectory characteristics, reflecting behavioral patterns and spatial movement characteristics. This step accurately captures user activity patterns and location changes, enabling the prediction of the user's next possible action (e.g., the need for hot water when walking from the living room to the kitchen). By identifying user activity patterns, the system can prepare water at the appropriate temperature in advance, reducing waiting time. Simultaneously, it can detect abnormal behavior patterns, providing extra attention to special groups (such as the elderly), improving the real-time performance and accuracy of hot water demand prediction.

[0070] S14 retrieves schedule data from the user-authorized calendar service. It converts time information (such as day of the week and time period) in the calendar into numerical features to reflect periodic patterns. Natural language processing (NLP) techniques are used to parse event descriptions, extract key information, and categorize and quantify different types of events, such as meetings, exercise, and rest. Time and event feature vectors are constructed to reveal the correlation between schedules and water demand. This step transforms unstructured calendar data into computable feature vectors, effectively identifying the correlation between schedules and water demand. It can predict the impact of specific schedules (such as meetings and exercise) on drinking water demand; for example, a user might need more warm water after exercise. By integrating schedule information, the system can more accurately predict the user's future water demand and provide more personalized services.

[0071] S15 employs knowledge distillation technology to compress large-scale cloud-based LSTM models into smaller models suitable for edge devices. It integrates behavioral features, trajectory features, temporal features, and event features into a unified input vector. The model output includes expected temperature, usage probability, and target usage time. The target temperature is calculated by weighting the expected temperature with the usage probability, reflecting the likelihood of the user's actual needs. The model performs lightweight updates when the device is idle, adapting to changes in user habits. This step achieves high-precision prediction on resource-constrained embedded devices, with prediction accuracy significantly higher than traditional rule engines. By processing sensitive data locally, it effectively protects user privacy and reduces reliance on network connectivity. Even in unstable network or offline states, the system can still provide intelligent services, ensuring the continuity and reliability of the user experience.

[0072] S16 calculates the difference between the target time and the current time, and determines the actual heating time based on usage probability. It dynamically adjusts the heating power according to the difference between the local boiling point and the theoretical boiling point of the model, considering the remaining heating time and the household circuit load, optimizing energy consumption and circuit safety while ensuring timely water supply. This step significantly reduces user waiting time and improves user satisfaction. By optimizing energy usage strategies, overall energy consumption is reduced, achieving green energy saving. It effectively avoids the risk of circuit overload caused by multiple devices operating at high power simultaneously, improving household electrical safety. The segmented heating strategy extends the equipment's lifespan, reduces losses caused by frequent start-stop cycles, and ensures precise water temperature control.

[0073] The above solution innovatively proposes a kettle control method based on a Bluetooth Mesh network. First, a dual-threshold mechanism is used to filter out stable and reliable neighboring devices, establishing a high-quality data source network. Then, environmental parameters from multiple devices are integrated to accurately calculate the local boiling point, solving the "false boiling" problem in high-altitude areas. Simultaneously, a multi-dimensional behavioral feature model is constructed by combining regional segmentation, user trajectory analysis, and calendar event parsing. The system employs a lightweight distillation LSTM model to fuse various feature data, achieving high-precision water demand prediction and intelligently planning heating strategies accordingly. Through multi-device collaborative sensing and edge intelligent decision-making, environmental adaptability and prediction accuracy are significantly improved, effectively reducing energy consumption and avoiding circuit overload risks. While ensuring a good user experience, a truly "seamless intelligent" service model is achieved.

[0074] For example, obtaining all environmental parameter messages broadcast by the nearby stable devices in the current area, and determining the local boiling point based on the environmental parameter messages, specifically includes:

[0075] Obtain all environmental parameter messages broadcast by the nearby stable devices in the current area, and extract the regional air pressure value and regional humidity value from the environmental parameter messages;

[0076] The theoretical boiling point is obtained based on the air pressure value of the region.

[0077] The local boiling point is obtained by adjusting the theoretical boiling point based on the regional humidity value and water quality information.

[0078] Taking a typical home environment as an example, the kettle is located in the kitchen area. The system first identifies nearby stable devices in the kitchen area via a Bluetooth Mesh network: a kitchen environment monitor and a smart refrigerator. These devices all meet the conditions of responding to Mesh messages within five minutes and being online for more than eighteen hours per day.

[0079] The system collects environmental parameter messages broadcast by two nearby stable devices within the kitchen area. The kitchen environmental monitor reports an air pressure of 975 hPa and a humidity of 55%; the smart refrigerator reports an air pressure of 976 hPa and a humidity of 52%. Since both devices are located in the same kitchen area, the system merges these data to calculate an average air pressure of 975.5 hPa and an average humidity of 53.5% for the kitchen area.

[0080] Based on the air pressure in the kitchen area, the system calculates the theoretical boiling point. According to physical principles, for every 33 hPa decrease in atmospheric pressure, the boiling point of water decreases by about 1 degree Celsius. Compared to the standard sea level pressure of 1013.25 hPa, the current air pressure in the kitchen area is about 37.75 hPa lower, therefore the theoretical boiling point is approximately 98.7 degrees Celsius.

[0081] Subsequently, the system adjusted the theoretical boiling point by combining the humidity level of the kitchen area and the water quality information set by the user. Considering that the higher humidity in the kitchen area would slightly lower the actual boiling point, and that the user-set water quality to medium hardness would slightly raise the boiling point, the system made comprehensive adjustments and finally determined the local boiling point of the kitchen area to be 98.1 degrees Celsius.

[0082] This process ensures that the kettle can precisely control the heating process based on the actual environmental conditions of its specific location, avoiding errors caused by using environmental data from other regions. By using data from equipment in only the same region, the system significantly improves the accuracy and reliability of boiling point calculation, providing users with more precise hot water service, and effectively solving the "false boiling" problem, especially in high-altitude areas.

[0083] For example, the step of extracting the environmental parameter messages to obtain regional air pressure and regional humidity values ​​specifically includes:

[0084] Based on the Mesh messages of the kettle and the Mesh messages of the neighboring stable devices in each current region, calculate the functional similarity value between the kettle and the neighboring stable devices in each current region;

[0085] In the Bluetooth Mesh network, among the set of neighboring stable devices that broadcast barometric pressure messages in the current region, the barometric pressure value corresponding to the neighboring stable device with the greatest functional similarity is selected as the regional barometric pressure value.

[0086] In the Bluetooth Mesh network, among the set of neighboring stable devices that broadcast humidity messages in the current area, the humidity value corresponding to the neighboring stable device with the greatest functional similarity is selected as the area humidity value.

[0087] For example, calculating the functional similarity value between the kettle and the neighboring stable devices in each current area based on the kettle's mesh message and the mesh messages of each neighboring stable device in the current area specifically includes:

[0088] Based on the switch parameter values, power outage recovery strategy values, numerical fluctuation values, and average power consumption of the kettle, a general model vector of the kettle is obtained as a first direction vector; based on the switch parameter values, power outage recovery strategy values, numerical fluctuation values, and average power consumption of the nearby stable devices in each current region, a general model vector of the nearby stable devices in each current region is obtained as a second direction vector; the projection value of the second direction vector onto the first direction vector is used as a first similarity value.

[0089] Based on the sensor parameter description values, sensor parameter measurement values, and reporting frequency values ​​of the kettle, the sensor model vector of the kettle is obtained as a third direction vector; based on the sensor parameter description values, sensor parameter measurement values, and reporting frequency values ​​of the nearby stable devices in each current region, the sensor model vector of the nearby stable devices in each current region is obtained as a fourth direction vector; the projection value of the fourth direction vector onto the third direction vector is used as a second similarity value;

[0090] The first similarity value and the second similarity value are weighted and summed to obtain the functional similarity value between the kettle and the neighboring stable devices in each current region.

[0091] Continuing with the previous example, in a kitchen environment, a kettle needs to determine its local boiling point. The system first identifies three nearby stable devices located in the same kitchen area: a smart refrigerator, an environmental monitor, and a smart socket. All of these devices are stably online and broadcast environmental parameter messages.

[0092] Kettle features:

[0093] General model characteristics: Frequent switching (approximately 5 times per day), power outage recovery strategy is "restore last settings", small numerical fluctuations (precise water temperature control), average power consumption is 1200W; Sensor model characteristics: Primarily monitors water temperature, high measurement accuracy, reporting frequency is once every 30 seconds.

[0094] Features of smart refrigerators:

[0095] General model characteristics: infrequent switching (approximately 3 times per day), power outage recovery strategy is "shutdown", moderate numerical fluctuation, average power consumption is 150W; Sensor model characteristics: monitors internal temperature and humidity, moderate measurement accuracy, reporting frequency is once every 5 minutes.

[0096] Features of environmental monitoring instruments:

[0097] General model characteristics: almost no switching on and off, power failure recovery strategy is "automatic restart", small numerical fluctuations, average power consumption is 5W; Sensor model characteristics: monitors multiple environmental parameters, high measurement accuracy, and reports every 1 minute.

[0098] Smart socket functional characteristics: General model characteristics: Frequent switching (about 8 times a day), power outage recovery strategy is "off", large value fluctuations, average power consumption is 10W; Sensor model characteristics: No environmental sensor, only monitors current and voltage.

[0099] The similarity scores of the kettle with various devices were calculated: 0.45 with the smart refrigerator (general model similarity 0.35, sensor model similarity 0.55); 0.82 with the environmental monitor (general model similarity 0.75, sensor model similarity 0.89); and 0.38 with the smart socket (general model similarity 0.52, but sensor model similarity 0). In the set of devices broadcasting barometric pressure messages (environmental monitor and smart refrigerator), the system selected the barometric pressure value of 975 hPa from the environmental monitor (similarity value 0.82) as the regional barometric pressure value. Similarly, in the set of devices broadcasting humidity messages (again, environmental monitor and smart refrigerator), the system selected the humidity value of 52% from the environmental monitor (again, the device with the highest functional similarity) as the regional humidity value.

[0100] This method avoids the errors that may arise from simple averaging. Through functional similarity assessment, it selects the environmental parameters of the device most closely matching the kettle's operating characteristics. Since the environmental monitoring instrument and the kettle are highly similar in sensor accuracy and usage patterns, its environmental parameters better reflect the kettle's actual operating environment, ensuring the accuracy of local boiling point calculations. Compared to random selection or simple averaging, this method reduces boiling point calculation errors in high-altitude areas, significantly improving the kettle's adaptability and safety under different environmental conditions.

[0101] For example, before obtaining all environmental parameter messages broadcast by the nearby stable devices in the current area and determining the local boiling point based on the environmental parameter messages, the process specifically includes:

[0102] Users can divide all IoT devices into zones based on the room where the IoT devices belong.

[0103] This zoning based on the actual house layout allows for precise acquisition of environmental parameters down to the specific living space. When the kettle needs to determine its local boiling point, the system uses only the environmental parameters of the equipment within the kitchen area, avoiding interference from data from living room air conditioners or bedroom equipment, ensuring the accuracy and specificity of the boiling point calculation. Especially in multi-story residential buildings, where there may be pressure differences between floors, precise zoning enables the kettle to acquire environmental data that truly reflects its location, significantly improving the accuracy and safety of heating control.

[0104] For example, obtaining behavioral feature data based on the usage record data of the kettle within a third preset time range and the usage record data of all nearby stable devices specifically includes:

[0105] Within the third preset time range, the most recent usage time, usage frequency, and average power consumption of the kettle and all nearby stable devices in the area where the kettle is located are statistically analyzed to obtain a behavioral feature sub-data.

[0106] The most recent usage time, usage frequency, and average power consumption of each of the nearby stable devices in other areas are statistically analyzed to obtain multiple behavioral feature sub-data; each behavioral feature sub-data corresponds to one area.

[0107] All the behavioral feature sub-data are merged to obtain the behavioral feature data.

[0108] Taking a smart home environment for a family of four as an example, the system divides the house into four functional areas: the kitchen (where the kettle is located), the living room, the master bedroom, and the secondary bedroom. The system sets a third preset time threshold of seven days to collect and analyze device usage data.

[0109] The system first compiles statistics on the usage of all equipment in the kitchen area over the past seven days:

[0110] Smart kettle: Last used at 07:15 today, used 3.2 times per day, average power consumption per use is 0.15 kWh; Smart refrigerator: Last door opened at 06:45 today, used 28 times per day, average power consumption is 1.2 kWh / day; Environmental monitor: Continuously running, latest data updated in real time, continuous usage frequency, average power consumption is 0.02 kWh / day; Smart socket: Last triggered at 07:10 today, used 4.5 times per day, average power consumption is 0.05 kWh / day; The system weights these data, giving the kettle a higher weight as the target device, forming the behavioral characteristic sub-data of the kitchen area: [07:15, 3.2 times / day, 0.15 kWh].

[0111] The system then compiled statistics on device usage in the other three areas:

[0112] Living room area:

[0113] Smart TV: Last used time yesterday at 20:30, usage frequency 1.8 times per day, average power consumption 0.8 kWh / day; Air conditioner: Last used time today at 09:15, usage frequency 2.5 times per day, average power consumption 2.4 kWh / day; Smart bulb: Last used time today at 07:00, usage frequency 4.2 times per day, average power consumption 0.12 kWh / day; Behavioral characteristic sub-data: [20:30, 1.8 times / day, 0.8 kWh].

[0114] Master bedroom area:

[0115] Smart bedside lamp: Last used time today 06:45, usage frequency 2.1 times per day, average power consumption 0.08 kWh / day; Air conditioner: Last used time today 06:30, usage frequency 1.3 times per day, average power consumption 1.6 kWh / day; Smart socket: Last used time today 07:00, usage frequency 0.9 times per day, average power consumption 0.03 kWh / day; Behavioral characteristic sub-data: [06:45, 2.1 times / day, 0.08 kWh];

[0116] Secondary bedroom area:

[0117] Smart bulb: Last used time yesterday at 21:15, usage frequency 1.5 times per day, average power consumption 0.06 kWh / day; Smart socket: Last used time yesterday at 19:30, usage frequency 0.7 times per day, average power consumption 0.02 kWh / day; Behavioral characteristic sub-data: [21:15, 1.5 times / day, 0.06 kWh].

[0118] The system integrates the behavioral feature sub-data from the four regions in regional order to form complete behavioral feature data:

[0119] [07:15,3.2,0.15], / / Kitchen area;

[0120] [20:30,1.8,0.80], / / Living room area;

[0121] [06:45,2.1,0.08], / / Master bedroom area;

[0122] [21:15,1.5,0.06] / / Secondary bedroom area;

[0123] This multidimensional behavioral data accurately reflects the differences in water usage habits across different areas of the home: the kitchen area is used frequently with moderate electricity consumption, indicating high daily cooking and drinking water needs; the living room area sees concentrated evening use and higher electricity consumption, reflecting water usage during leisure time; and the bedroom area sees concentrated morning and bedtime use with low electricity consumption, reflecting a preference for warm water. Through this regionalized behavioral modeling, the system can accurately distinguish water demand in different living scenarios, avoiding the misappliance of living room users' habits to kitchen scenarios. For example, after identifying the pattern of delayed use in the master bedroom on weekend mornings, the system will adjust the preheating strategy of the kitchen kettle accordingly. This refined behavioral analysis improves the accuracy of hot water demand prediction, significantly enhancing user experience and energy efficiency.

[0124] For example, based on calendar data obtained from the Bluetooth Mesh hub, time feature data and event feature data are obtained, specifically including:

[0125] Convert numerical information in calendar data that conforms to preset standards into time feature data.

[0126] The remaining calendar data is divided into blocks according to preset natural language processing rules, and each block is mapped to a quantized value; all the quantized values ​​are merged to obtain event feature data.

[0127] Through this processing method, the system can transform unstructured calendar data into structured features that can be used for predictive models. Time features accurately capture the periodic patterns of user activities, while event features reflect the type and importance of the activities. For example, the system recognizes that there is usually a need for hot water after a "family dinner" and a preference for a slightly lower temperature (around 80°C); while after a "morning run," users usually need water at a higher temperature (around 85°C).

[0128] In practical applications, this calendar data feature extraction improves the accuracy of hot water demand forecasting, particularly in identifying the impact of special occasions (such as family gatherings or post-exercise activities) on drinking water demand. The system can predict users' water needs up to 30 minutes in advance, significantly reducing user waiting time and avoiding unnecessary heating, thus achieving efficient energy utilization.

[0129] For example, the step of inputting the behavioral feature data, trajectory feature data, time feature data, and event feature data into a preset distillation LSTM model to obtain the target temperature and target usage time specifically includes:

[0130] The Bluetooth Mesh hub obtains the preset running parameters of the distilled LSTM model from the cloud.

[0131] The distillation LSTM model is updated while the kettle is not in operation.

[0132] The behavioral feature data, trajectory feature data, time feature data, and event feature data are input into the distillation LSTM model to obtain the user's desired expected temperature, expected probability, and target usage time.

[0133] The target temperature is the product of the user's desired temperature and the expected probability.

[0134] The following is an example of a pre-set scenario: Mr. Zhang lives in a high-altitude city at 2500 meters above sea level. His home has a smart kettle that supports Bluetooth Mesh networking and several IoT devices. The system has divided the devices into three areas—kitchen, living room, and bedroom—based on the house layout. Mr. Zhang has a fixed morning routine: he gets up at 6:45, enters the kitchen at 7:15 to prepare breakfast, and needs hot water at around 85°C to make tea at 7:30.

[0135] The system identified two nearby stable devices in the kitchen area: an environmental monitor (online for 22 hours daily, with a 100% response rate within 5 minutes) and a smart refrigerator (online for 20 hours daily, with a 98% response rate within 5 minutes). The two devices reported air pressure values ​​of 748 hPa and 749 hPa, and humidity values ​​of 45% and 47%, respectively.

[0136] The system integrates the following data:

[0137] Behavioral characteristics: The kitchen area is used 3.2 times / day in the morning, with an average preferred temperature of 85°C; Trajectory characteristics: Users usually enter the kitchen at 7:15 and stay for 15-20 minutes; Time characteristics: Monday morning, weekday mode; Event characteristics: The calendar shows a "team meeting" at 8:00, with an importance score of 0.85.

[0138] After distillation and prediction using the LSTM model, the following data was obtained:

[0139] Expected temperature: 83°C (considering a slight preference for lower temperatures before the meeting); Expected probability: 0.92; Target usage time: 7:30. Therefore, target temperature = 83°C × 0.92 = 76.4°C ≈ 76°C.

[0140] Based on functional similarity calculations, the system determined that the environmental monitor (similarity value 0.85) was more suitable than the smart refrigerator (similarity value 0.42) for providing environmental parameters. Finally, using data from the environmental monitor: air pressure 748 hPa, humidity 45%, combined with information on medium hardness water, the local boiling point was calculated to be 91.5°C.

[0141] The system calculates that heating should begin at 7:10 (target time 7:30 minus 20 minutes of heating time, considering a probability adjustment of 0.92). Based on the local boiling point of 91.5°C, the system sets the heating target to 76°C, allowing for a segmented heating strategy.

[0142] 7:10-7:20: 600W low-power preheating; 7:20-7:25: 1200W main heating; 7:25-7:30: pulse-type fine adjustment to 76°C.

[0143] In Mr. Zhang's scenario, when he entered the kitchen at 7:15, the kettle had already precisely controlled the water temperature at 76°C. This avoided the false alarms caused by "false boiling" in traditional kettles used in high-altitude areas, and provided a suitable temperature to meet his current needs. Because the system predicted his water demand in advance and optimized the heating strategy, Mr. Zhang could enjoy perfectly warm tea without waiting, preparing for his upcoming meeting.

[0144] Of particular note is that on a cold winter morning, the system detected that Mr. Zhang entered the kitchen 10 minutes later than usual (7:25 AM) through trajectory characteristics. It immediately adjusted the heating time to start at 7:20 AM, ensuring the water temperature reached 76°C by 7:35 AM. This dynamic adaptability allows the system to provide precise service even when the user's schedule changes, a flexibility that traditional kettles cannot achieve. This technological solution not only solves the special problems of using kettles in high-altitude areas but also achieves truly "seamless intelligence" service through multi-device collaboration and edge intelligence prediction, significantly improving user experience and energy efficiency.

[0145] For example, controlling the heating start time and heating power of the kettle based on the target temperature, the local boiling point, and the target usage time specifically includes:

[0146] Calculate the time difference between the target usage time and the current time.

[0147] The product of the time difference and the expected probability is used as the heating time.

[0148] Obtain the theoretical boiling point parameter value from the distillation LSTM model, and use the difference between the theoretical boiling point parameter value and the local boiling point as the boiling point deviation value; adjust the target temperature value according to the boiling point deviation value.

[0149] Before the heating time expires, while ensuring that the water temperature in the kettle reaches the target temperature value, the average heating power of the kettle is reduced to the maximum extent.

[0150] In a typical high-altitude home environment, the system predicts that the user will need drinking water at a suitable temperature at 07:30. The current time is 07:00, and the system has determined the following key parameters: Target usage time: 07:30; Expected probability: 0.88 (based on a distillation LSTM model); Local boiling point: 91.5°C (calculated through the fusion of environmental parameters from multiple devices); Theoretical boiling point parameter value: 100°C (the internal parameters of the distillation LSTM model are generally the standard sea-level boiling point); Original target temperature: 78°C (based on user preferences and scenario analysis).

[0151] The system first calculates the time difference between the target usage time and the current time as 30 minutes. Then, it multiplies this time difference by the expected probability to obtain the actual heating time: 30 minutes multiplied by 0.88, approximately 26.4 minutes. This means the system has 26.4 minutes to complete the heating task. Considering the certainty of users' water demand, the available heating time is appropriately shortened to reserve a buffer for possible unforeseen circumstances.

[0152] The theoretical boiling point parameter value of 100°C was then obtained from the distillation LSTM model and compared with the currently measured local boiling point of 91.5°C, resulting in a boiling point deviation of 8.5°C. Based on this deviation, the system adjusted the original target temperature: considering the physical properties of water in high-altitude areas, the perceived "heat" of water at the same temperature would be slightly different. The system reduced the original target temperature of 78°C by 2.3°C, resulting in an adjusted target temperature of 75.7°C. This adjustment ensures that the water temperature perceived by the user is consistent with the experience in plains areas, avoiding temperature perception differences caused by the reduction in boiling point.

[0153] During the 26.4-minute heating period, the system implemented a segmented power control strategy to minimize the average heating power while ensuring the water temperature reached 75.7°C on time.

[0154] Preheating stage (0-10 minutes): Using a low-power 600W heater, the water temperature is steadily increased from 25°C to 45°C. This stage focuses on reducing thermal shock, extending equipment life, and avoiding a sudden increase in circuit load caused by initial high power.

[0155] Main heating phase (10-22 minutes): The power is dynamically adjusted between 800-1000W based on real-time water temperature feedback. When the ambient temperature is low or the water flow is high, the power is automatically increased; when the target temperature is approached, the power is gradually reduced to minimize the risk of temperature overshoot.

[0156] Fine-tuning stage (22-26 minutes): Pulse heating (heating for 5 seconds every 10 seconds) is used to precisely control the water temperature within the range of 75.7°C ± 0.5°C. During this stage, the power is maintained at only around 400W to ensure temperature stability and avoid energy waste caused by repeated heating.

[0157] The aforementioned heating control strategy achieves significant technical benefits. Through segmented power optimization, the average heating power is significantly reduced compared to traditional full-power heating, minimizing energy consumption while maintaining a good user experience. The off-peak heating strategy staggers the high-power periods of the kettle with other appliances (such as microwaves and ovens), reducing peak load on the household circuit and effectively preventing circuit breaker tripping. Pulse-based fine-tuning achieves a final water temperature control accuracy of ±0.5°C, a 90% improvement over the ±5°C accuracy of traditional kettles, ensuring users receive the perfect drinking water temperature. Furthermore, it avoids frequent full-power start-stop cycles, reducing thermal stress on the heating element and effectively extending the kettle's lifespan.

[0158] Compared to existing technologies, this embodiment provides a kettle control method and kettle based on IoT-based changes. It constructs a home IoT collaborative system via a Bluetooth Mesh network, first selecting stable devices that respond promptly and remain online for extended periods as data sources. Utilizing environmental parameters broadcast by these devices, the system dynamically calculates the local boiling point, taking into account altitude, air pressure, and humidity, thus addressing the "false boiling" problem in high-altitude areas. Simultaneously, it integrates usage records from multiple regions to construct behavioral feature data matching the number of household areas. This data, combined with user activity trajectories obtained from infrared sensors and time event features analyzed from calendar data, is input into a lightweight distillation LSTM model to predict user demand. This model runs efficiently on edge devices, outputting a target temperature that considers usage probability. Finally, based on the difference between the target temperature and the local boiling point, the system intelligently plans the heating start time and segmented power strategies, optimizing energy consumption while ensuring timely water supply. The technical effects are significant: improved boiling point calculation accuracy, increased demand prediction accuracy, reduced energy consumption, reduced circuit overload risk, truly achieving personalized drinking water services tailored to users, and greatly enhancing the kettle's adaptability in complex environments and the user experience.

[0159] See Figure 2 One embodiment of this application provides an IoT-based electric kettle, comprising: a device determination module 20, a boiling point determination module 21, a behavior feature module 22, a trajectory feature module 23, a calendar feature module 24, a temperature calculation module 25, and a device control module 26.

[0160] The device determination module 20 is used to determine nearby stable devices through a Bluetooth Mesh network; the nearby stable devices refer to IoT devices that respond to Mesh messages within a first preset time threshold and have an online time of more than a second preset time threshold within the same Bluetooth Mesh network.

[0161] Boiling point determination module 21 is used to obtain environmental parameter messages broadcast by all nearby stable devices in the current area, and determine the local boiling point based on the environmental parameter messages.

[0162] The behavior feature module 22 is used to obtain behavior feature data based on the usage record data of this device and the usage record data of all the nearby stable devices within a third preset time range; the number of dimensions of the behavior feature data is equal to the number of regions, and each dimension corresponds to one region.

[0163] The trajectory feature module 23 is used to obtain the user's dwell time, dwell start point and dwell end point in each area based on the passive infrared sensors in each area, and obtain trajectory feature data.

[0164] The calendar feature module 24 is used to obtain time feature data and event feature data based on the calendar data obtained from the Bluetooth Mesh hub.

[0165] The temperature calculation module 25 is used to input the behavioral feature data, trajectory feature data, time feature data and event feature data into a preset distillation LSTM model to obtain the target temperature.

[0166] The equipment control module 26 is used to control the heating start time and heating power of the equipment according to the target temperature and the local boiling point.

[0167] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the formula-based full-line R&D and production management system described above can be referred to the corresponding process in the foregoing method embodiments, and will not be elaborated further here.

[0168] Compared to existing technologies, the IoT-based kettle provided in this embodiment constructs a home IoT collaborative system through a Bluetooth Mesh network. First, it selects stable devices that respond promptly and remain online for extended periods as data sources. Utilizing environmental parameters broadcast by these devices, the system dynamically calculates the local boiling point, taking into account altitude, air pressure, and humidity, thus resolving the "false boiling" problem in high-altitude areas. Simultaneously, it integrates usage records from multiple regions to construct behavioral feature data matching the number of household areas. This data, combined with user activity trajectories obtained from infrared sensors and time event features analyzed from calendar data, is input into a lightweight distillation LSTM model to predict user demand. This model runs efficiently on edge devices, outputting a target temperature that considers usage probability. Finally, based on the difference between the target temperature and the local boiling point, the system intelligently plans the heating start time and segmented power strategies, optimizing energy consumption while ensuring timely water supply. The technical effects are significant: improved boiling point calculation accuracy, increased demand prediction accuracy, reduced energy consumption, reduced circuit overload risk, truly achieving personalized drinking water services tailored to users, and greatly enhancing the kettle's adaptability in complex environments and the user experience.

[0169] One embodiment of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for controlling a kettle based on Internet of Things (IoT) changes.

[0170] The computer device may be a smartphone, tablet, desktop computer, or cloud server, among other computing devices. This computer device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the figures are merely examples of computer devices and do not constitute a limitation on the computer device. It may include more or fewer components than illustrated, or a combination of certain components, or different components, such as input / output devices, network access devices, etc.

[0171] The processor referred to can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0172] In some embodiments, the memory may be an internal storage unit of the computer device, such as a hard drive or RAM. In other embodiments, the memory may be an external storage device of the computer device, such as a plug-in hard drive, SmartMediaCard (SMC), SecureDigital (SD) card, or FlashCard. Furthermore, the memory may include both internal and external storage units of the computer device. The memory is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory can also be used to temporarily store data that has been output or will be output.

[0173] This application provides a computer program product that, when run on a computer device, enables the computer device to execute the steps described in the various method embodiments above.

[0174] In the several embodiments provided in this application, it will be understood that each block in the flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved.

[0175] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0176] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for controlling a kettle based on Internet of Things (IoT) changes, characterized in that: include: Identify nearby stable devices via Bluetooth Mesh network; The nearby stable device refers to an IoT device that responds to Mesh messages within a first preset time threshold and has an online time of more than a second preset time threshold within the same Bluetooth Mesh network; Obtain all environmental parameter messages broadcast by the nearby stable devices in the current area, and determine the local boiling point based on the environmental parameter messages; Based on the usage record data of the kettle within the third preset time range and the usage record data of all the nearby stable devices, behavioral feature data is obtained; the number of dimensions of the behavioral feature data is equal to the number of regions, and each dimension corresponds to one region; Based on the passive infrared sensors in each area, the user's dwell time, starting point, and ending point in each area are obtained to acquire trajectory feature data. Based on the calendar data obtained from the Bluetooth Mesh hub, time feature data and event feature data are obtained; The target temperature is obtained by inputting the behavioral feature data, trajectory feature data, time feature data, and event feature data into a preset distillation LSTM model. The heating start time and heating power of the kettle are controlled according to the target temperature and the local boiling point.

2. The method for controlling a kettle based on IoT changes as described in claim 1, characterized in that, The step of obtaining all environmental parameter messages broadcast by the nearby stable devices in the current area and determining the local boiling point based on the environmental parameter messages specifically includes: Obtain all environmental parameter messages broadcast by the nearby stable devices in the current area, and extract the regional air pressure value and regional humidity value from the environmental parameter messages; The theoretical boiling point is obtained based on the air pressure value of the region. The local boiling point is obtained by adjusting the theoretical boiling point based on the regional humidity value and water quality information.

3. The method for controlling a kettle based on IoT changes as described in claim 2, characterized in that, The extraction of the environmental parameter messages to obtain the regional air pressure value and regional humidity value specifically includes: Based on the Mesh messages of the kettle and the Mesh messages of the neighboring stable devices in each current region, calculate the functional similarity value between the kettle and the neighboring stable devices in each current region; In the Bluetooth Mesh network, among the set of neighboring stable devices that broadcast barometric pressure messages in the current area, the barometric pressure value corresponding to the neighboring stable device with the greatest functional similarity is selected as the area barometric pressure value. In the Bluetooth Mesh network, among the set of neighboring stable devices that broadcast humidity messages in the current area, the humidity value corresponding to the neighboring stable device with the greatest functional similarity is selected as the area humidity value.

4. The method for controlling a kettle based on Internet of Things changes as described in claim 3, characterized in that, The step of calculating the functional similarity value between the kettle and the neighboring stable devices in each current area based on the kettle's mesh message and the mesh messages of each neighboring stable device in the current area specifically includes: Based on the switching parameter values, power outage recovery strategy values, numerical fluctuation values, and average power consumption of the kettle, a general model vector of the kettle is obtained as a first direction vector; based on the switching parameter values, power outage recovery strategy values, numerical fluctuation values, and average power consumption of the nearby stable devices in each current region, a general model vector of the nearby stable devices in each current region is obtained as a second direction vector; the projection value of the second direction vector onto the first direction vector is used as a first similarity value; Based on the sensor parameter description values, sensor parameter measurement values, and reporting frequency values ​​of the kettle, the sensor model vector of the kettle is obtained as a third direction vector; based on the sensor parameter description values, sensor parameter measurement values, and reporting frequency values ​​of the nearby stable devices in each current region, the sensor model vector of the nearby stable devices in each current region is obtained as a fourth direction vector; the projection value of the fourth direction vector onto the third direction vector is used as a second similarity value; The first similarity value and the second similarity value are weighted and summed to obtain the functional similarity value between the kettle and the neighboring stable devices in each current region.

5. The method for controlling a kettle based on Internet of Things changes as described in claim 1, characterized in that, Before obtaining all environmental parameter messages broadcast by the nearby stable devices in the current area and determining the local boiling point based on the environmental parameter messages, the process specifically includes: Users can divide all IoT devices into zones based on the room where the IoT devices belong.

6. The method for controlling a kettle based on Internet of Things changes as described in claim 1, characterized in that, The behavioral characteristic data obtained based on the usage record data of the kettle within the third preset time range and the usage record data of all nearby stable devices specifically includes: Within the third preset time range, the most recent usage time, usage frequency, and average power consumption of the kettle and all nearby stable devices in the area where the kettle is located are statistically analyzed to obtain a behavioral feature sub-data. The most recent usage time, usage frequency, and average power consumption of each of the nearby stable devices in other areas are statistically analyzed to obtain multiple behavioral feature sub-data; each behavioral feature sub-data corresponds to one area. All the behavioral feature sub-data are merged to obtain the behavioral feature data.

7. The method for controlling a kettle based on Internet of Things changes as described in claim 1, characterized in that, Based on calendar data obtained from the Bluetooth Mesh hub, time feature data and event feature data are obtained, specifically including: Convert numerical information in calendar data that conforms to preset standards into time feature data; The remaining calendar data is divided into blocks according to preset natural language processing rules, and each block is mapped to a quantized value; all the quantized values ​​are merged to obtain event feature data.

8. The method for controlling a kettle based on Internet of Things changes as described in claim 1, characterized in that, The target temperature is obtained by inputting the behavioral feature data, trajectory feature data, time feature data, and event feature data into a preset distillation LSTM model, specifically including: The Bluetooth Mesh hub obtains the preset running parameters of the distilled LSTM model from the cloud; The distillation LSTM model is updated while the kettle is not in operation; The behavioral feature data, trajectory feature data, time feature data, and event feature data are input into the distillation LSTM model to obtain the user's desired expected temperature, expected probability, and target usage time. The target temperature is the product of the user's desired temperature and the expected probability.

9. The method for controlling a kettle based on Internet of Things changes as described in claim 8, characterized in that, The step of controlling the heating start time and heating power of the kettle based on the target temperature, the local boiling point, and the target usage time specifically includes: Calculate the time difference between the target usage time and the current time. The product of the time difference and the expected probability is used as the heating time; Obtain the theoretical boiling point parameter value from the distillation LSTM model, and use the difference between the theoretical boiling point parameter value and the local boiling point as the boiling point deviation value; adjust the target temperature value according to the boiling point deviation value. Before the heating time expires, while ensuring that the water temperature in the kettle reaches the target temperature value, the average heating power of the kettle is reduced to the maximum extent.

10. A kettle based on Internet of Things (IoT) changes, characterized in that: include: The device identification module is used to identify nearby stable devices via a Bluetooth Mesh network. The nearby stable device refers to an IoT device that responds to Mesh messages within a first preset time threshold and has an online time of more than a second preset time threshold within the same Bluetooth Mesh network; The boiling point determination module is used to obtain environmental parameter messages broadcast by all nearby stable devices in the current area, and determine the local boiling point based on the environmental parameter messages. The behavior feature module is used to obtain behavior feature data based on the usage record data of this device and the usage record data of all the nearby stable devices within a third preset time range; the number of dimensions of the behavior feature data is equal to the number of regions, and each dimension corresponds to one region; The trajectory feature module is used to obtain the user's dwell time, starting point, and ending point in each area based on the passive infrared sensors in each area, and to obtain trajectory feature data. The calendar feature module is used to obtain time feature data and event feature data based on the calendar data obtained from the Bluetooth Mesh hub; The temperature calculation module is used to input the behavioral feature data, trajectory feature data, time feature data and event feature data into a preset distillation LSTM model to obtain the target temperature. The equipment control module is used to control the heating start time and heating power of the equipment according to the target temperature and the local boiling point.

Images