Intelligent switch control method and system based on Internet of Things

Abstract

The invention discloses an intelligent switch control method and system based on the Internet of Things, and relates to the technical field of smart home, and the method comprises the following specific steps: data collection: obtaining user operation records, equipment operation parameters, environment data and user basic configuration information through various collection assemblies, and forming an original data set; according to the method, a user behavior feature set and an equipment operation feature library are constructed, user operation records, equipment operation parameters and environment data are acquired in a data acquisition stage, deep analysis is performed on the acquired data in a self-adaptive learning stage, operation frequency distribution features, equipment operation features and environment-behavior mapping rules are extracted, and the operation frequency distribution features, the equipment operation features and the environment-behavior mapping rules are analyzed. And a periodic updating mechanism is set, and the parameter threshold is dynamically adjusted according to the newly added data, so that the intelligent switch can dynamically learn the user operation habit, the control strategy is adaptively adjusted along with the change of the user habit, and the problem of immobilization of the existing intelligent switch control strategy is effectively solved.

Description

Technical Field

[0001] This invention relates to the field of smart home technology, specifically to a smart switch control method and system based on the Internet of Things. Background Technology

[0002] With the rapid development of IoT technology, its application in the field of smart homes is increasing, greatly changing people's lifestyles and home experiences. Smart home systems connect various home devices into an organic whole through IoT technology, realizing interconnection and intelligent control between devices, providing users with a more convenient, comfortable, and efficient home living environment. As a key control terminal in the smart home system, smart switches bear the important responsibility of operating and managing home appliances, and their performance and functions directly affect the overall performance of the smart home system and the user experience.

[0003] However, existing smart switch technology still has many limitations. On the one hand, it lacks the ability to dynamically learn user operating habits. The control strategies of existing smart switches are often pre-set fixed modes, which cannot be adaptively adjusted according to the user's operating habits in actual use. As a result, its control strategies cannot meet the diverse needs of users in different scenarios. On the other hand, it has failed to establish a correlation model between environmental factors and device control behavior. Factors such as light intensity, temperature, humidity, and human activity in the home environment will affect the user's operation of electrical appliances. However, existing smart switches do not fully consider these environmental factors and cannot automatically adjust the device control strategy according to environmental changes. This makes the control strategy fixed and difficult to achieve precise control in different environments. In addition, existing smart switches have not formed a deep data interaction mechanism with the Internet of Things cloud platform. The cloud platform has powerful data processing and storage capabilities. Through deep interaction with the cloud platform, smart switches can achieve iterative optimization of control strategies and continuously improve control accuracy and adaptability. However, in the current technology, the data interaction between smart switches and the cloud platform is limited, and the advantages of the cloud platform cannot be fully utilized. As a result, smart switches are difficult to continuously optimize and improve when facing complex and ever-changing home environments. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an IoT-based intelligent switch control method and system. This system comprehensively acquires multi-dimensional data, including user operation records, device operating status parameters, environmental parameters, and basic configuration information, during the data acquisition phase. In the adaptive learning phase, it performs in-depth analysis of the acquired data, constructing a user behavior feature set, a device operating feature library, and an environment-behavior mapping rule set, and performs periodic updates. During the scene recognition and prediction phase, it presets multiple typical life scene labels and achieves accurate scene recognition through real-time data matching. In the control strategy execution phase, it automatically calls the associated control instruction set based on the scene recognition results and supports user strategy adjustments and feedback corrections. Finally, in the IoT cloud platform interaction phase, it realizes functions such as cloud storage of data, feature parameter and rule updates, remote control instruction reception, and device abnormal status monitoring, effectively improving the control accuracy and intelligence level of the intelligent switch and enhancing the system's adaptability and universality.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On one hand, an intelligent switch control method based on the Internet of Things, the method comprising the following specific steps:

[0006] Data Acquisition: Obtain user operation records, device operating parameters, environmental data, and basic user configuration information through various acquisition components to form a raw dataset;

[0007] Adaptive learning: Analyzes and processes the collected data to form a set of user behavior features, a database of device operation features, and a set of environment-behavior mapping rules, and dynamically adjusts relevant parameters through a periodic update mechanism;

[0008] Scene recognition and prediction: Preset multiple life scene labels and corresponding feature parameter combinations, and output scene recognition results through real-time data matching calculation;

[0009] Control strategy execution: Associate device control command sets for various scenarios, automatically execute corresponding control operations, update parameters according to user-adjusted commands and provide feedback to the learning phase;

[0010] IoT cloud platform interaction: Enables data uploading, model updates, command forwarding, and anomaly notification interaction between terminals and the cloud.

[0011] Furthermore, in the data acquisition step, the acquisition components deployed on the smart switch continuously acquire user operation records of the switch, including the timestamp of each operation, the physical triggering method, and the unique identifier of the controlled device; synchronously acquire the operating status parameters of the connected electrical equipment, including on / off status, continuous running time, and real-time power value; acquire light intensity, temperature, humidity, and human activity detection signals in the space through the environmental sensing components; and collect basic configuration information through the user interface, including preset work and rest periods, number of users, and device linkage relationship table.

[0012] Furthermore, in the adaptive learning step, the collected operation records are subjected to time-series analysis to extract the operation frequency distribution characteristics and the correlation of operation methods in different time periods, forming a user behavior feature set; the device operating parameters are statistically processed to determine the start probability and running time range of each device in different time periods, establishing a device operating feature library; by comparing the correlation between environmental parameters and operation records, the correspondence between the environmental parameter threshold range and the device control behavior is identified, and the correlation strength is calculated through the environment-behavior correlation strength model to form an environment-behavior mapping rule set; a periodic update mechanism is set to dynamically adjust the parameter thresholds of the feature set, feature library and rule set according to the newly added data.

[0013] Furthermore, in the adaptive learning step, the association strength is calculated using an environment-behavior association strength model to form an environment-behavior mapping rule set, the calculation formula of which is: ,in, Environmental parameters With control behavior The strength of the association, It is the first Environmental parameters in the second sampling The standardized value, It is the first Control behavior in subsampling The quantization value is 1, indicating execution, and 0, indicating no execution. It is the total number of samples. This is a sample size correction term.

[0014] Furthermore, in the scene recognition and prediction step, multiple typical life scene labels are preset, including but not limited to waking up, leaving home, returning home, sleeping, and entertainment scenes; and feature parameter combinations are defined for each scene label, including time interval range, environmental parameter threshold range, device status combination mode, and human activity characteristics; by acquiring the current environmental parameters, device operating status, and human activity signals in real time, the matching degree is calculated with the feature parameter combinations of the preset scene; when the matching degree reaches the set threshold, the corresponding scene label is output as the recognition result.

[0015] Furthermore, in the scene recognition and prediction step, the matching degree is calculated by acquiring current environmental parameters, device operating status, and human activity signals in real time, and combining them with the feature parameters of the preset scene. The calculation formula is as follows: ,in, For the current scene and the preset scene The degree of matching, For time matching factor, As an environmental matching factor, For device status matching factor, As a matching factor for human activities, These are dynamic weighting coefficients, which are iteratively optimized using user behavior data.

[0016] Furthermore, in the control strategy execution step, a corresponding set of device control instructions is associated with each preset scene label, including device on / off instructions, light brightness adjustment values, and device operating mode parameters; after the scene recognition result is output, the associated control instruction set is automatically invoked, and an execution signal is sent to the target device through the control component; the strategy adjustment instructions issued by the user through the interactive interface are received, the control parameters of the corresponding scene are updated according to the adjustment content, and the adjustment record is fed back to the adaptive learning step for correcting the feature set and rule set.

[0017] Furthermore, during the interaction phase of the IoT cloud platform, the collected operation records, device operating parameters, environmental data, and scene recognition results are uploaded to the cloud storage unit at a set cycle through the communication component; feature parameter correction values ​​and rule update data sent from the cloud are received and replaced with the corresponding content stored locally; user remote control commands verified by the cloud are received, parsed, and converted into control signals that can be executed by the device; and device abnormal status notifications sent from the cloud are received and the preset protection control process is triggered.

[0018] On the other hand, there is an IoT-based intelligent switch control system, which includes: an intelligent switch terminal, an IoT cloud platform, and a user interaction terminal;

[0019] The intelligent switch terminal includes a main control module, a communication module, an execution control module, an environmental sensing module, a storage module, and a power supply module;

[0020] The main control module employs an embedded processing unit to receive and process data transmitted from various modules, run data processing programs, and generate device control commands. The communication module integrates a short-range wireless communication unit and a wide-area network communication unit, supporting bidirectional data transmission with the cloud platform, controlled devices, and user terminals. The execution control module uses relay components to receive command signals from the main control module, execute circuit switching operations, and realize start-stop control of electrical equipment. The environmental sensing module consists of a light detection element, a temperature and humidity sensor, and a human infrared sensing element, used to collect environmental physical parameters and human activity signals. The storage module uses non-volatile storage media to save collected data, feature parameters, rule sets, and control strategies. The power supply module adopts a wide-range AC input design, outputs a stable DC voltage to power each module, and has a built-in backup power unit to ensure basic function operation after a power outage.

[0021] The IoT cloud platform includes a data storage subsystem, a data processing subsystem, an instruction forwarding subsystem, an anomaly monitoring subsystem, and a remote upgrade subsystem.

[0022] The data storage subsystem adopts a distributed storage architecture to centrally store various types of data uploaded by multiple smart switch terminals; the data processing subsystem summarizes and analyzes the uploaded data to generate feature parameter correction values ​​and rule update data; the instruction forwarding subsystem receives control instructions sent by user terminals, verifies permissions, and forwards them to the target smart switch terminal; the anomaly monitoring subsystem continuously monitors the device's operating parameters and generates an anomaly notification when the parameters exceed the safety threshold; the remote upgrade subsystem pushes program update packages to the smart switch terminals to achieve terminal function iteration.

[0023] The user interaction terminal includes a device management interface, a parameter configuration interface, a remote control interface, and a data display interface;

[0024] The device management interface provides functions for adding, deleting, and displaying the status of smart switches; the parameter configuration interface supports manual setting of user basic information, scene feature parameters, and control strategies; the remote control interface provides manual control buttons and parameter adjustment sliders; and the data display interface presents collected data, scene recognition records, and device operation statistics in the form of charts.

[0025] Compared with existing technologies, this IoT-based intelligent switch control method and system have the following advantages:

[0026] I. This invention constructs a user behavior feature set and a device operation feature library. During the data acquisition phase, it obtains user operation records, device operation parameters, and environmental data. During the adaptive learning phase, it performs in-depth analysis on the collected data, extracts operation frequency distribution features, device operation features, and environment-behavior mapping rules, and sets up a periodic update mechanism to dynamically adjust parameter thresholds based on new data. This enables the smart switch to dynamically learn user operation habits and allows the control strategy to adaptively adjust as user habits change. This effectively solves the problem of fixed control strategies in existing smart switches and improves control accuracy and intelligence.

[0027] Second, this invention constructs a scene recognition mechanism based on multi-dimensional parameters, presets multiple typical life scene labels and defines feature parameter combinations, and acquires current environment, equipment and human activity signals in real time to calculate the matching degree, accurately determine the life scene, and provide a reliable basis for automated equipment control. At the same time, with the help of the Internet of Things cloud platform for collaborative processing, the uploaded data is summarized and analyzed to generate correction values ​​and updated data, realizing remote optimization of the control model; it can also receive remote control commands and abnormal status notifications verified by the cloud, trigger corresponding processes, improve the continuous adaptability of the system, and enhance the practicality and reliability of the system.

[0028] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0030] Figure 1 This is a schematic diagram of an IoT-based intelligent switch control system.

[0031] Figure 2 This is a flowchart of an IoT-based intelligent switch control method. Detailed Implementation

[0032] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structure, features and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0033] Example 1

[0034] This embodiment provides an IoT-based intelligent switch control system, which includes three parts: an intelligent switch terminal, an IoT cloud platform, and a user interaction terminal. It can effectively improve the convenience and security of home control and is adaptable to diverse smart home scenarios.

[0035] like Figure 1 As shown, the intelligent switch terminal integrates a main control module, a communication module, an execution control module, an environmental sensing module, a storage module, and a power supply module. The main control module uses an STM32L476RGT6 microprocessor with a processing speed of 80MHz, enabling it to quickly process data from multiple modules and ensure that control commands are executed without delay. The communication module integrates Bluetooth Mesh and Wi-Fi modules. The Bluetooth Mesh module supports linkage with nearby devices such as smart door locks and smart curtains, with a response speed of ≤0.5 seconds. The Wi-Fi module supports the 2.4GHz band, ensuring a stable connection and preventing control failure due to network outages. The execution control module uses two magnetic latching relays, supporting a maximum load control of 2500W and can simultaneously connect to chandeliers. The device is compatible with devices such as televisions; the environmental sensing module's light sensor, temperature and humidity sensor, and human infrared sensor all use high-precision components, with light detection accuracy reaching ±20%, temperature detection error ≤0.5℃, and human infrared sensor detection distance of 3-7 meters, accurately capturing user activity status; the storage module uses a 16GBeMMC flash memory chip, which can store more than 1 year of local data, avoiding the loss of historical data due to cloud failures; the power supply module supports AC100-240V wide voltage input, adapting to different regional power grids, and outputs a stable DC5V / 1A voltage. The built-in 300mAh lithium battery serves as a backup power source, and can still maintain data acquisition and Bluetooth communication for 4 hours after a power outage, allowing users to check the device status before the power outage.

[0036] The IoT cloud platform is deployed on cloud servers and adopts a distributed architecture design, featuring high concurrency and high reliability. The data storage subsystem's MySQL database supports tens of millions of data storage entries, while Redis caches real-time data with a query response time of ≤100ms. The data analysis subsystem aggregates and processes data uploaded by devices through scheduled tasks, identifying changes in user habits and dynamically optimizing control strategies. The command forwarding subsystem uses the MQTT protocol to ensure stable bidirectional communication between devices and the cloud, with a command transmission success rate of ≥99.9%. The anomaly monitoring subsystem presets safe ranges for device operating parameters; for example, if the TV power exceeds the normal range of 200-300W, it immediately triggers an alert to prevent device malfunctions from causing safety hazards. The remote upgrade subsystem supports OTA firmware update packages, allowing users to upgrade functions without disassembling the device, improving the user experience throughout the system's lifecycle.

[0037] The user interface of the mobile APP is simple and easy to operate. The device list clearly displays the status of all associated appliances. The scene setting function supports one-click creation of custom scenes such as "coming home" and "sleep". The remote control function has a latency of ≤1 second. The data statistics page presents power consumption and device usage time in chart form to help users understand energy consumption. The management platform is for administrators to view the operating status of all devices in the area. The batch parameter configuration function reduces repetitive operations. Firmware upgrade management supports batch updates to avoid network congestion.

[0038] Example 2

[0039] This embodiment provides an IoT-based intelligent switch control method, applied to the system described in Embodiment 1 above. It can accurately match user habits, achieve automated control, and improve living comfort and energy efficiency. Figure 2 As shown, the specific steps are as follows:

[0040] Data acquisition phase: After the smart switch terminal is powered on, the environmental sensing module collects data on light intensity, temperature and humidity in the living room every 10 seconds, and collects the operating status (on / off) and running time of the connected chandelier and TV every 5 seconds; when the user operates the device through the switch panel or mobile APP, the operation time, operation method (single click / long press) and the identification of the controlled device are recorded in real time; the user enters family routine information through the mobile APP, including the approximate time range of family members getting up, leaving home, returning home and going to sleep.

[0041] Adaptive learning phase: The system processes the previous day's collected data at 2 AM daily, statistically analyzing the frequency of user operation of the chandelier and television at different times, identifying a pattern where users typically turn on the television between 7:00 PM and 9:00 PM and turn off the chandelier after 10:30 PM; it analyzes the relationship between environmental data and device operation, finding that the probability of users turning on the chandelier increases significantly when the light intensity is below 50 lux; and it calculates the correlation strength using an environment-behavior correlation strength model, forming an environment-behavior mapping rule set, the calculation formula of which is: ,in, Environmental parameters With control behavior The strength of the association, It is the first Environmental parameters in the second sampling The standardized value, It is the first Control behavior in subsampling The quantization value is 1, indicating execution, and 0, indicating no execution. It is the total number of samples. This is a sample size correction term, and it is stored in the local storage module.

[0042] Scene recognition and prediction stage: The system presets scene types such as homecoming scene, entertainment scene, and sleep scene. The entertainment scene is characterized by the time period of 19:00-21:00, human activity, and moderate light intensity. When the system detects that the current time is 19:30, the human infrared sensor detects human activity, and the light intensity is 150 lux, the matching degree is calculated using the following formula: ,in, For the current scene and the preset scene The degree of matching, For time matching factor, As an environmental matching factor, For device status matching factor, As a matching factor for human activities, The dynamic weighting coefficient determines whether the current scene matches the entertainment scene to a set threshold, and identifies it as an entertainment scene.

[0043] Control strategy execution phase: The control strategy configured by the system for the entertainment scene is to turn on the TV power and adjust the brightness of the chandelier to 70%; after scene recognition, the smart switch terminal automatically sends control commands to the TV and the chandelier; if the user manually adjusts the brightness of the chandelier to 50%, the system records the adjustment operation, updates the control parameters of the entertainment scene, and feeds the adjustment information back to the adaptive learning phase to optimize the relevant rules.

[0044] During the IoT cloud platform interaction phase: Every 30 minutes, the smart switch terminal uploads the collected environmental data, device operating status, and scene recognition results to the cloud platform; after analyzing data from multiple users, the cloud platform discovers that the power fluctuation of this model of TV increases after running for more than 4 hours, and sends this pattern to all associated terminals; when a user remotely sends a command to turn off the TV via a mobile APP, the command is verified by the cloud platform and forwarded to the smart switch terminal to execute the shutdown operation; if the TV power exceeds the safe range, the cloud platform sends an abnormal warning message, and the smart switch terminal automatically cuts off the TV power.

[0045] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A smart switch control method based on the Internet of Things, characterized in that, The method includes the following specific steps: Data Acquisition: Obtain user operation records, device operating parameters, environmental data, and basic user configuration information through various acquisition components to form a raw dataset; Adaptive learning: Analyzes and processes the collected data to form a set of user behavior features, a database of device operation features, and a set of environment-behavior mapping rules, and dynamically adjusts relevant parameters through a periodic update mechanism; Scene recognition and prediction: Preset multiple life scene labels and corresponding feature parameter combinations, and output scene recognition results through real-time data matching calculation; Control strategy execution: Associate device control command sets for various scenarios, automatically execute corresponding control operations, update parameters according to user-adjusted commands and provide feedback to the learning phase; IoT cloud platform interaction: Enables data uploading, model updates, command forwarding, and anomaly notification interaction between terminals and the cloud.

2. The IoT-based intelligent switch control method according to claim 1, characterized in that, In the data acquisition step, the user's operation records of the switch are continuously acquired through the acquisition components deployed on the smart switch, including the timestamp of each operation, the physical triggering method and the unique identifier of the controlled device. The system synchronously collects operating status parameters of connected electrical devices, including on / off status, continuous operating time, and real-time power values; it also collects light intensity, temperature, humidity, and human activity detection signals in the space through environmental sensing components; and it gathers basic configuration information through the user interface, including preset work and rest periods, number of users, and device linkage relationship tables.

3. The IoT-based intelligent switch control method according to claim 1, characterized in that, In the adaptive learning step, the collected operation records are analyzed over time to extract the operation frequency distribution characteristics and the correlation of operation methods in different time periods, forming a set of user behavior features. Statistical processing of equipment operating parameters is performed to determine the start-up probability and operating time range of each device at different time periods, and a device operating feature library is established. By comparing the correlation between environmental parameters and operation records, the correspondence between environmental parameter threshold ranges and equipment control behaviors is identified. The correlation strength is calculated through an environment-behavior correlation strength model to form an environment-behavior mapping rule set. A periodic update mechanism is set to dynamically adjust the parameter thresholds of the feature set, feature library, and rule set based on new data.

4. The IoT-based intelligent switch control method according to claim 3, characterized in that, In the adaptive learning step, the association strength is calculated using the environment-behavior association strength model to form an environment-behavior mapping rule set, the calculation formula of which is: ,in, Environmental parameters With control behavior The strength of the association, It is the first Environmental parameters in the second sampling The standardized value, It is the first Control behavior in subsampling The quantization value is 1, indicating execution, and 0, indicating no execution. It is the total number of samples. This is a sample size correction term.

5. The IoT-based intelligent switch control method according to claim 1, characterized in that, In the scene recognition and prediction step, multiple typical life scene labels are preset, including but not limited to waking up, leaving home, returning home, sleeping, and entertainment scenes; and feature parameter combinations are defined for each scene label, including time interval range, environmental parameter threshold range, device status combination mode, and human activity characteristics; by acquiring the current environmental parameters, device operating status, and human activity signals in real time, the matching degree is calculated with the feature parameter combinations of the preset scene; when the matching degree reaches the set threshold, the corresponding scene label is output as the recognition result.

6. The IoT-based intelligent switch control method according to claim 5, characterized in that, In the scene recognition and prediction step, the matching degree is calculated by acquiring current environmental parameters, equipment operating status, and human activity signals in real time, and combining them with the feature parameters of the preset scene. The calculation formula is as follows: ,in, For the current scene and the preset scene The degree of matching, For time matching factor, As an environmental matching factor, For device status matching factor, As a matching factor for human activities, These are dynamic weighting coefficients, which are iteratively optimized using user behavior data.

7. The IoT-based intelligent switch control method according to claim 1, characterized in that, In the control strategy execution step, a corresponding set of device control instructions is associated with each preset scene label, including device on / off instructions, light brightness adjustment values, and device operating mode parameters. After the scene recognition result is output, the associated control instruction set is automatically called, and an execution signal is sent to the target device through the control component. The strategy adjustment instructions issued by the user through the interactive interface are received, the control parameters of the corresponding scene are updated according to the adjustment content, and the adjustment record is fed back to the adaptive learning step to correct the feature set and rule set.

8. The IoT-based intelligent switch control method according to claim 1, characterized in that, During the interaction phase of the IoT cloud platform, the collected operation records, device operating parameters, environmental data, and scene recognition results are uploaded to the cloud storage unit at a set cycle through the communication component; feature parameter correction values ​​and rule update data sent from the cloud are received and replaced with the corresponding content stored locally; user remote control commands verified by the cloud are received, parsed, and converted into control signals that can be executed by the device; and device abnormal status notifications sent from the cloud are received and the preset protection control process is triggered.

9. An IoT-based intelligent switch control system, applicable to the IoT-based intelligent switch control system as described in any one of claims 1-8, characterized in that, The system includes: a smart switch terminal, an Internet of Things cloud platform, and a user interaction terminal; The intelligent switch terminal includes a main control module, a communication module, an execution control module, an environmental sensing module, a storage module, and a power supply module; The main control module employs an embedded processing unit to receive and process data transmitted from various modules, run data processing programs, and generate device control commands. The communication module integrates a short-range wireless communication unit and a wide-area network communication unit, supporting bidirectional data transmission with the cloud platform, controlled devices, and user terminals. The execution control module uses relay components to receive command signals from the main control module, execute circuit switching operations, and realize start-stop control of electrical equipment. The environmental sensing module consists of a light detection element, a temperature and humidity sensor, and a human infrared sensing element, used to collect environmental physical parameters and human activity signals. The storage module uses non-volatile storage media to save collected data, feature parameters, rule sets, and control strategies. The power supply module adopts a wide-range AC input design, outputs a stable DC voltage to power each module, and has a built-in backup power unit to ensure basic function operation after a power outage. The IoT cloud platform includes a data storage subsystem, a data processing subsystem, an instruction forwarding subsystem, an anomaly monitoring subsystem, and a remote upgrade subsystem. The data storage subsystem adopts a distributed storage architecture to centrally store various types of data uploaded by multiple smart switch terminals; the data processing subsystem summarizes and analyzes the uploaded data to generate feature parameter correction values ​​and rule update data; the instruction forwarding subsystem receives control instructions sent by user terminals, verifies permissions, and forwards them to the target smart switch terminal; the anomaly monitoring subsystem continuously monitors the device's operating parameters and generates an anomaly notification when the parameters exceed the safety threshold; the remote upgrade subsystem pushes program update packages to the smart switch terminals to achieve terminal function iteration. The user interaction terminal includes a device management interface, a parameter configuration interface, a remote control interface, and a data display interface; The device management interface provides functions for adding, deleting, and displaying the status of smart switches; the parameter configuration interface supports manual setting of user basic information, scene feature parameters, and control strategies; the remote control interface provides manual control buttons and parameter adjustment sliders; and the data display interface presents collected data, scene recognition records, and device operation statistics in the form of charts.

Images