Automatic cooperative regulation and control system and method for smart home equipment

Abstract

The invention relates to the technical field of smart home, and discloses an automatic coordinated regulation and control system and method for smart home equipment, and the system integrates a plurality of functional modules such as equipment access, environment perception and user behavior analysis. The equipment access module supports access of multiple protocols and endows the equipment with a unique identifier; the environment sensing module collects environment parameters in real time and compares the environment parameters with thresholds. The user behavior analysis module generates a user scene preference model by means of an algorithm, the cooperation strategy generation module synthesizes multiple factors to formulate a regulation and control strategy, and the dynamic priority management module allocates weights according to equipment and scene conditions. And the equipment control module executes the instruction and feeds back the state, and the abnormity detection module monitors the equipment state and triggers a strategy regeneration mechanism when abnormity occurs. The data storage module securely stores various data, and the user interaction module efficiently analyzes instructions. According to the invention, many problems of an existing smart home system are solved, automatic coordinated regulation and control of equipment can be realized, and user interaction experience is optimized.

Description

Technical Field

[0001] The present invention relates to the field of smart home technology, and in particular to a system and method for automatic coordinated control of smart home equipment. Background Art

[0002] With the booming development of Internet of Things technology, the field of smart home has made remarkable progress. More and more families are starting to use various smart home devices, such as smart lamps, home appliances, and sensors, hoping to improve the convenience and comfort of life. However, the current smart home system still has many defects in practical applications.

[0003] The existence of multiple IoT communication protocols is a major problem. Smart home devices of different brands and types use different communication protocols, such as ZigBee, Wi-Fi, Bluetooth, etc. This makes it difficult to connect devices to the system and difficult for devices to interact smoothly, which seriously restricts the overall performance and scalability of the smart home system. For example, if a user purchases a new smart sensor that uses a specific protocol, they may find that it is not compatible with the existing smart home system, making it unusable.

[0004] In terms of environmental perception, although current smart home devices can collect environmental parameters such as temperature, humidity, light intensity and air quality index, the use of data is relatively superficial. Most systems only display these data in real time and fail to automatically and intelligently adjust related equipment according to parameter changes. For example, when the indoor temperature is too high, the system cannot automatically link the air conditioner to cool down, and the user still needs to operate manually, and the degree of intelligence needs to be improved.

[0005] The lack of user behavior analysis capabilities is also prominent. Existing smart home systems have difficulty in deeply mining and analyzing users' historical operation data and real-time instructions, and cannot build an accurate user scenario preference model. This means that the system cannot understand users' personalized needs and habits, and cannot provide users with personalized device control services in different scenarios. For example, different users have different preferences for bedroom light brightness and air conditioning temperature when they rest at night, but the system cannot automatically adapt to these differences.

[0006] The device collaborative control strategy is also not perfect. Existing control strategies often only consider a single factor and lack comprehensive considerations. For example, when formulating device control strategies, you may only focus on device energy consumption, while ignoring important factors such as response delay and scene matching. This leads to either slow device response in actual use, affecting user experience, or the control plan does not match the actual scene and cannot achieve the desired use effect.

[0007] In addition, in terms of device priority management, current smart home systems usually use fixed priority settings, and do not dynamically adjust according to changes in device type, energy consumption level, and scene urgency. This will bring many inconveniences in real life. For example, in emergency scenarios, important devices may not be able to respond in time due to unreasonable priority settings, affecting user life.

[0008] There are also loopholes in the anomaly detection and processing mechanism. The existing system does not monitor the execution status of the device accurately enough. When the device execution status does not meet expectations, it cannot detect it in time and take effective measures. Moreover, once an anomaly occurs, there is a lack of an efficient strategy regeneration mechanism, and the system cannot quickly adjust the device control strategy, which greatly reduces the stability and reliability of the smart home system.

[0009] There are also certain problems with data storage and user interaction. On the one hand, the storage method of important information such as device identification codes and environmental parameter historical data is not secure enough, which is prone to the risk of data loss or tampering; on the other hand, the interaction between users and smart home systems is not convenient and intelligent enough. For example, voice command parsing is inaccurate, often misjudgments occur, and touch operations are also cumbersome, which affects users' enthusiasm for using smart home systems. Summary of the invention

[0010] The purpose of the present invention is to provide a smart home device automatic coordinated control system and method to solve the problems raised in the above background technology.

[0011] To achieve the above object, the present invention provides the following technical solution: an automatic coordinated control system for smart home devices, the system comprising: The device access module is used to connect the lamps, home appliances and sensor devices in the home to the system through various IoT communication protocols, and assign a unique identification code to each device; Environmental sensing module, used to collect home environment parameters in real time, including temperature, humidity, light intensity and air quality index, and dynamically compare the parameters with preset threshold ranges; User behavior analysis module, which is used to perform cluster analysis on user historical operation data and real-time operation instructions through machine learning algorithms to generate user scenario preference models; A collaborative strategy generation module is used to generate a device collaborative control strategy based on a multi-objective optimization algorithm according to the environmental parameter comparison results, the user scenario preference model and the device operation status; Dynamic priority management module, which is used to assign dynamic priority weights to each device according to device type, energy consumption level and scene urgency, and update it to the collaborative control strategy; The device control module is used to send control instructions to the target device according to the collaborative control strategy and provide real-time feedback on the device execution status; The anomaly detection module is used to monitor the deviation between the execution status of the device and the expected status. When the deviation exceeds the preset fault tolerance threshold, the policy regeneration mechanism is triggered; Data storage module, used to store device identification code, environmental parameter historical data, user behavior model and collaborative strategy version information; The user interaction module is used to receive user voice or touch commands, and parse the commands into standardized operation commands and input them into the user behavior analysis module.

[0012] Preferably, the specific analysis method of the user behavior analysis module is: Obtain the user's historical operation time series data from the data storage module, and extract the operation frequency, device relevance, and scene trigger condition characteristics; The hidden Markov model is used to calculate the state transition probability of real-time operation instructions and predict the user's next operation intention; Combining time series characteristics with state transition probabilities, a user scenario preference model including device linkage rules and weight distribution is generated.

[0013] Preferably, the specific generation method of the collaborative strategy generation module is: Define equipment energy consumption, response delay and scene matching as optimization objectives, and construct multi-objective constraint equations; Using NSGA-II algorithm to perform non-dominated sorting on the constraint equations to generate a Pareto optimal solution set; The final collaborative control strategy is selected from the optimal solution set according to the dynamic priority weight.

[0014] Preferably, the specific allocation method of the dynamic priority management module is: Obtain the basic priority table corresponding to the device type from the local database, and map it to obtain the initial priority of each device; Calculate the dynamic energy consumption correction coefficient based on real-time energy consumption data and equipment load rate; The dynamic priority weight of the device is updated by combining the scene urgency and the correction coefficient through the fuzzy logic algorithm.

[0015] Preferably, the specific detection method of the anomaly detection module is: Obtain the difference Δt between the actual response time and the expected response time of the device from the device control module; The difference Δt is de-noised by using the Kalman filter algorithm to obtain a stable deviation value ΔT; If ΔT exceeds the preset fault tolerance threshold, the policy regeneration mechanism is triggered and the abnormal event is recorded in the data storage module.

[0016] Preferably, the specific implementation method of the strategy regeneration mechanism is: Freeze the current collaborative control strategy and back it up to the data storage module; Recall the real-time data of the environment perception module and the user behavior analysis module to generate alternative collaborative strategies; The deviation values ​​of the original strategy and the alternative strategy are compared through the A / B test algorithm, and the optimal strategy is selected to update to the device control module.

[0017] Preferably, the instruction parsing method of the user interaction module is: Convert voice commands into structured text through natural language processing technology, extract device operation verbs and target parameters; Use the knowledge graph to match the compatibility of device identification codes and operation verbs to generate standardized operation instructions; If the match fails, the instruction is corrected through the reinforcement learning model and fed back to the user for secondary confirmation.

[0018] Preferably, the storage method of the data storage module is: Use time series database partitions to store historical data of environmental parameters and create indexes based on equipment identification codes; The user scenario preference model and collaborative strategy version information are encrypted and stored through blockchain technology, and an unalterable hash check value is generated.

[0019] Preferably, the protocol compatibility method of the device access module is: The ZigBee, Wi-Fi and Bluetooth communication protocols are uniformly mapped to the MQTT standard data format through the protocol conversion middleware; Assign a virtual communication port to each device and monitor the port connection status in real time to trigger the device reconnection mechanism.

[0020] Preferably, the present invention also includes a method for automatic coordinated control of smart home devices, comprising the following steps: Step S1: Connect the home device to the system through the device access module and assign a unique identification code; Step S2: using the environment perception module to collect environmental parameters in real time and dynamically compare them with preset thresholds; Step S3: Generate a user scenario preference model through a user behavior analysis module; Step S4: Generate a device collaborative control strategy based on a multi-objective optimization algorithm; Step S5: using the dynamic priority management module to update the policy weight; Step S6: Execute the control strategy through the device control module and feedback the device status; Step S7: When it is detected that the execution deviation exceeds the threshold, a policy regeneration mechanism is triggered; Step S8: Store all operation data and policy versions, and receive real-time instructions through the user interaction module.

[0021] Compared with the prior art, the present invention has the following beneficial effects: The device access module supports multiple IoT communication protocols, and uses protocol conversion middleware to uniformly map ZigBee, Wi-Fi and Bluetooth communication protocols to the MQTT standard data format, and assigns a unique identification code and virtual communication port to each device, which enables convenient access and unified management of various lamps, home appliances and sensor devices in the home. This not only improves system compatibility and reduces the difficulty of device expansion, but also triggers the device reconnection mechanism by real-time monitoring of port connection status, ensuring the stability of communication between the device and the system, greatly facilitating users' use and maintenance of smart home devices.

[0022] The environmental sensing module collects home environmental parameters such as temperature, humidity, light intensity and air quality index in real time, and dynamically compares them with the preset threshold range. Based on this, the system can detect environmental changes in a timely manner, combine information from other modules, and automatically and accurately coordinate related equipment for regulation. For example, when the temperature is too high, the air conditioner is automatically linked to cool down to maintain a comfortable indoor environment, improving the comfort and intelligence of the home environment.

[0023] The user behavior analysis module uses machine learning algorithms to deeply analyze the user's historical operation data and real-time operation instructions. By extracting the operation frequency, device relevance and scene trigger condition characteristics, combined with the hidden Markov model to predict the user's next operation intention, a user scene preference model containing device linkage rules and weight distribution is generated. This enables the system to deeply understand the user's personalized habits and needs, provide users with customized device control services, significantly improve the user experience, and make smart homes truly adapt to the unique lifestyles of different users. The collaborative strategy generation module defines device energy consumption, response delay and scene matching as optimization goals, constructs a multi-objective constraint equation, uses the NSGA-II algorithm to generate the Pareto optimal solution set, and selects the final collaborative control strategy based on dynamic priority weights. This method comprehensively considers multiple factors and effectively balances device energy consumption, response speed and scene adaptability. It not only reduces energy waste, but also ensures that the device can respond to instructions quickly, and the control scheme is highly consistent with the actual scene, improving the overall operation efficiency and performance of the smart home system.

[0024] The dynamic priority management module assigns dynamic priority weights to devices based on device type, energy consumption level and scene urgency. By obtaining the basic priority from the local database, combining the real-time energy consumption data and the device load rate to calculate the dynamic energy consumption correction coefficient, and then using the fuzzy logic algorithm to update the priority weight, the system can reasonably allocate resources according to the actual situation in complex and changeable home scenes. In emergency scenarios, key devices can respond quickly with high priority to ensure that users' lives are not affected, thereby improving the system's resilience and reliability. The anomaly detection module uses the Kalman filter algorithm to monitor the deviation between the device execution status and the expected status, and can accurately identify device anomalies. Once the deviation exceeds the preset fault tolerance threshold, the strategy regeneration mechanism is immediately triggered. This mechanism freezes the current strategy and backs it up, calls the real-time data to generate an alternative strategy, and selects the optimal strategy through the A / B test algorithm to update it to the device control module. This series of operations ensures that the system can quickly resume stable operation when the device fails or anomalies occur, reducing interference with users' lives and enhancing the stability and reliability of the system.

[0025] The data storage module uses time-series database partitions to store historical data of environmental parameters, and creates indexes according to device identification codes to facilitate data query and management; blockchain technology is used to encrypt and store user scenario preference models and collaborative strategy version information, generating tamper-proof hash check values ​​to ensure data security and integrity. The user interaction module uses natural language processing technology to parse voice commands, uses knowledge graphs to generate standardized operation commands, and uses reinforcement learning models to correct commands and feedback to users for confirmation when matching fails, thus achieving convenient and intelligent interaction between users and the system and improving the accuracy and convenience of operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a working principle diagram of the automatic coordinated control system for smart home devices according to the present invention; Figure 2 This is the workflow diagram of the user behavior analysis module; Figure 3 Workflow diagram for the collaborative strategy generation module. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] See also Figure 1-3The present invention provides a technical solution: an automatic coordinated control system for smart home devices, the system comprising: Device access module: The basic entrance of the system, responsible for the key task of connecting home lamps, home appliances and sensor devices to the system. It supports a variety of IoT communication protocols, such as ZigBee, Wi-Fi and Bluetooth. During the access process, the system assigns a unique identification code to each device, just like giving the device an "ID card" to facilitate subsequent accurate identification and management.

[0029] Environmental perception module: responsible for real-time collection of home environment parameters, including temperature, humidity, light intensity and air quality index, etc. The collected data will be dynamically compared with the preset threshold range to detect environmental changes in time and provide a basis for subsequent regulation.

[0030] User behavior analysis module: With the help of machine learning algorithms, cluster analysis is performed on user historical operation data and real-time operation instructions. Its core purpose is to generate a user scenario preference model, so as to accurately grasp user habits and needs and make system regulation more in line with user preferences.

[0031] Collaborative strategy generation module: Based on the environmental parameter comparison results, user scenario preference model and device operation status, a multi-objective optimization algorithm is used to generate a device collaborative control strategy. This strategy comprehensively considers multiple factors to achieve collaborative operation between devices.

[0032] Dynamic priority management module: Assign dynamic priority weights to each device based on device type, energy consumption level, and scenario urgency, and update them to the collaborative control strategy to ensure the rationality and efficiency of device control in complex situations.

[0033] Equipment control module: Sends control instructions to the target equipment according to the collaborative control strategy, and provides real-time feedback on the equipment execution status to achieve actual control and status monitoring of the equipment.

[0034] Anomaly detection module: monitors the deviation between the execution status of the device and the expected status. Once the deviation exceeds the preset fault tolerance threshold, it triggers the policy regeneration mechanism to ensure stable operation of the system.

[0035] Data storage module: stores device identification codes, environmental parameter historical data, user behavior models, and collaborative strategy version information, providing data support and historical backtracking for system operation.

[0036] User interaction module: receives user voice or touch commands, parses them into standardized operation instructions, and then inputs them into the user behavior analysis module, building a convenient interaction bridge between users and the system.

[0037] The present invention will be further described below in conjunction with Examples 1 to 5: Embodiment 1: This embodiment elaborates on how the user behavior analysis module uses specific methods to deeply mine user operation data and generate an accurate user scenario preference model, providing strong support for the system to achieve personalized intelligent control. In this system, the user behavior analysis module obtains the user's historical operation time series data from the data storage module. This data contains the user's operation records on various devices at different times, such as the specific time points of turning on lights, adjusting air-conditioning temperature, starting washing machines, etc. The operation frequency, device association and scenario trigger condition features are extracted. The operation frequency is calculated as follows: the number of times the user operates device i within a certain time period T. , then the operating frequency . The device association is determined by counting the order and frequency of operations on different devices. If the probability of device B being operated within a short period of time after device A is operated is high, it means that devices A and B are highly associated. Scene trigger condition feature extraction is to analyze under what environment or operation combination a specific scene will be triggered, such as turning on the bedroom light at night when the light intensity is below a certain threshold.

[0038] The hidden Markov model is used to calculate the state transition probability of real-time operation instructions and predict the user's next operation intention. The hidden Markov model contains the state set S, the observation set O, the initial state probability vector , state transition probability matrix A and observation probability matrix B. Assume that there are n possible device operation states in the system, and the state set , observation set For real-time operation instructions, use it as an observation sequence , through the formula (in are model parameters, represents the state at time k, Status Observed below The probability of state transition is calculated to predict the user’s next possible operation state, that is, the next operation intention.

[0039] Combining time series features with state transition probabilities, a user scenario preference model containing device linkage rules and weight assignments is generated. For example, if analysis finds that after turning on the air conditioner in the summer evening, the user will most likely turn on the bedroom fan within 10 minutes, and the air conditioner is operated frequently, then in the user scenario preference model, the linkage rule weights between the air conditioner and the fan can be set to have a high probability (such as 0.8) of turning on the fan when the air conditioner is turned on, and the weight of the air conditioner in this scenario is relatively large to highlight its importance in the scenario.

[0040] Embodiment 2: This embodiment describes in detail how the collaborative strategy generation module defines optimization objectives, constructs constraint equations, and uses specific algorithm screening strategies to achieve the generation of equipment collaborative control strategies that comprehensively consider multiple factors and ensure the efficiency and rationality of system operation.

[0041] The collaborative strategy generation module first defines device energy consumption, response delay and scene matching as optimization objectives and constructs a multi-objective constraint equation. Assume that the device set is ,equipment The energy consumption is , the response delay is , the scene matching degree is The objective function can be expressed as:

[0042] At the same time, the actual constraints of equipment operation should be considered, such as Power Limit , then the power constraint equation is ( For equipment The actual power of the device Maximum working time limit , then there is a working time constraint equation ( For equipment cumulative working hours), etc.

[0043] The NSGA-II algorithm is used to perform non-dominated sorting on the above constraint equations to generate a Pareto optimal solution set. The NSGA-II algorithm divides the population into non-dominated frontiers of different levels through fast non-dominated sorting, and introduces crowding calculation to maintain population diversity. In each generation of evolution, the non-dominated level and crowding degree of each individual are calculated. For two individuals p and q, if individual p is not better than individual q in all objective functions, and is better than individual q in at least one objective function, then individual p is said to dominate individual q. Through continuous evolution, a set of Pareto optimal solutions are obtained, which achieve a balance between different objectives.

[0044] The final collaborative control strategy is selected from the optimal solution set according to the dynamic priority weight. Assume that the dynamic priority weight vector is The final collaborative control strategy is selected from the optimal solution set according to the dynamic priority weight. Assume that the dynamic priority weight vector is , for each solution in the Pareto optimal solution set ( Representation solution Medium Equipment The comprehensive score is calculated , and select the solution with the highest score as the final collaborative control strategy.

[0045] Embodiment 3: This embodiment describes in detail how the dynamic priority management module allocates dynamic priority weights to devices based on multiple factors, so that the system can flexibly adjust strategies according to actual conditions when regulating devices.

[0046] The dynamic priority management module obtains the basic priority table corresponding to the device type from the local database and maps it to obtain the initial priority of each device. For example, the initial priority of the lighting device specified in the basic priority table is , the initial priority of the air conditioning equipment is ,and , because air conditioning is more critical than lighting equipment in regulating indoor comfort.

[0047] According to the real-time energy consumption data and equipment load rate, the dynamic energy consumption correction coefficient is calculated. Assume that the real-time energy consumption of equipment i is , rated energy consumption is , the equipment load rate is , then the dynamic energy consumption correction factor The calculation formula is:

[0048] in, and It is a weight coefficient, which can be adjusted according to actual conditions to balance the impact of real-time energy consumption and equipment load rate on the correction coefficient.

[0049] The dynamic priority weight of the device is updated by combining the scene urgency and the correction coefficient through the fuzzy logic algorithm. In the fuzzy logic algorithm, the scene urgency is divided into three fuzzy sets of "low", "medium" and "high", the dynamic energy consumption correction coefficient is divided into three fuzzy sets of "small", "medium" and "large", and the dynamic priority weight of the device is divided into three fuzzy sets of "low", "medium" and "high". For example, when the scene urgency is "high" and the dynamic energy consumption correction coefficient is "small", the output device dynamic priority weight is "high" through fuzzy rule reasoning. The fuzzy rules are determined through expert experience and actual tests. After fuzzy reasoning and defuzzification, the specific device dynamic priority weight update value is obtained.

[0050] Embodiment 4: This embodiment describes in detail how the anomaly detection module accurately monitors the deviation between the execution status of the device and the expected status, processes data through a specific algorithm and promptly triggers the policy regeneration mechanism to ensure stable and reliable operation of the system and reduce control errors caused by equipment anomalies.

[0051] The anomaly detection module obtains the difference between the actual response time and the expected response time of the device from the device control module For example, after sending a command to turn on a lamp, the time difference between the actual time the lamp turns on and the expected time the lamp turns on is recorded as .

[0052] The difference is calculated by Kalman filtering algorithm Perform noise elimination to obtain a stable deviation value The Kalman filter algorithm is an optimal linear recursive filter algorithm. Assume that the system state equation is , the observation equation is ,in is the system state at time k, is the state transfer matrix, is the control input matrix, is the control input, is the process noise, is the observed value at time k is the observation matrix, is the observation noise. sequence, taking it as the observation sequence , through the prediction and update steps of the Kalman filter algorithm, the system state is continuously estimated to obtain a stable deviation value .

[0053] like If the preset fault tolerance threshold is exceeded, the policy regeneration mechanism is triggered and the abnormal event is recorded in the data storage module. The preset fault tolerance threshold is determined according to the device type and the actual application scenario. For example, for lighting equipment with high response speed requirements, the preset fault tolerance threshold may be set to 500ms. When the lamp responds abnormally, the strategy regeneration mechanism is triggered. At the same time, the time of the abnormality, device identification code, abnormality type and other information are recorded in the data storage module to facilitate subsequent troubleshooting and system optimization.

[0054] Embodiment 5: This embodiment describes in detail the specific implementation process of the policy regeneration mechanism, the user interaction module instruction parsing method, the data storage module storage method and the device access module protocol compatibility method, so as to make the functional integrity and interaction process of the entire system clearer and improve the system's usability, stability and compatibility.

[0055] In terms of the policy regeneration mechanism, when the anomaly detection module triggers the mechanism, the current collaborative control policy is first frozen and backed up to the data storage module. The system will record all parameters and status information of the current policy, such as the control instructions and execution progress of each device. The backup file can be named in the form of a timestamp plus a policy version number to facilitate subsequent query and tracing.

[0056] Recall the real-time data of the environment perception module and the user behavior analysis module to generate an alternative collaborative strategy. The environment perception module recollects environmental parameters, the user behavior analysis module updates the user scenario preference model according to the latest user operation instructions, and the collaborative strategy generation module regenerates an alternative collaborative control strategy based on these new data according to the method in Example 2.

[0057] The A / B test algorithm compares the deviation between the original strategy and the alternative strategy, and selects the optimal strategy to update to the device control module. The A / B test algorithm uses the original strategy as version A and the alternative strategy as version B. The two strategies are run in the same environment and simulated user operation conditions to compare their deviation values ​​in indicators such as device energy consumption, response delay, and scene matching. For example, calculate the difference in total device energy consumption under the two strategies , the difference in average response delay And the difference in scene matching , comprehensively consider these differences to determine which strategy is better, and update the instructions of the optimal strategy to the device control module for execution.

[0058] The instruction parsing method of the user interaction module is to convert the voice instruction into structured text through natural language processing technology, and extract the device operation verb and target parameter. For example, for the voice instruction "adjust the brightness of the living room light to 50%", the natural language processing technology first converts it into text form, and then extracts the operation verb "adjust to", the target parameter "50%" and the device related information "living room light".

[0059] The knowledge graph is used to match the compatibility of device identification codes and operation verbs to generate standardized operation instructions. The knowledge graph stores the relationship between device types, device identification codes, and various operation verbs. For example, lighting equipment is compatible with operation verbs such as "turn on", "turn off", and "adjust brightness". By querying the knowledge graph, the identification code of "living room lamp" is matched with the operation verb "adjust to", and standardized operation instructions such as "[living room lamp device identification code, adjust brightness, 50%]" are generated.

[0060] If the match fails, the reinforcement learning model corrects the instruction and feeds back to the user for secondary confirmation. The reinforcement learning model learns the optimal instruction correction strategy through continuous trial and error. When the match fails, the model tries to generate possibly correct instructions based on the current instructions and knowledge graph information. For example, if the user says "turn on the TV fan", the model recognizes that "TV fan" may be incorrect, and generates a correction instruction of "turn on the fan next to the TV" based on learning experience, and feeds back to the user "Do you want to turn on the fan next to the TV? Please confirm", and then proceeds with subsequent processing after the user confirms.

[0061] The storage method of the data storage module is to use time series database partitions to store historical environmental parameter data and to create indexes based on device identification codes. For example, based on time, daily environmental parameter data is stored in different partitions, and indexes are created in each partition based on device identification codes. This allows rapid location and acquisition of environmental parameter data for a device within a specific time period, thereby improving data query efficiency.

[0062] The user scenario preference model and collaborative strategy version information are encrypted and stored through blockchain technology, and a tamper-proof hash check value is generated. The blockchain adopts a decentralized distributed ledger structure. Each block contains a certain amount of transaction data (here, the user scenario preference model and collaborative strategy version information), a timestamp, and the hash value of the previous block. For each updated user scenario preference model and collaborative strategy version information, its hash value is calculated and stored in the blockchain. If the data is tampered with, the hash value will change, thereby ensuring the security and integrity of the data.

[0063] The protocol compatibility method of the device access module is to uniformly map the ZigBee, Wi-Fi and Bluetooth communication protocols to the MQTT standard data format through the protocol conversion middleware. The protocol conversion middleware has built-in conversion rules between different protocols and the MQTT protocol. For example, for the device data of the ZigBee protocol, it is converted into data in the MQTT protocol format according to specific rules, which is convenient for the system to uniformly process and transmit.

[0064] Assign a virtual communication port to each device and monitor the port connection status in real time to trigger the device reconnection mechanism. The allocation of virtual communication ports uses a unique identification algorithm to ensure that the port number of each device is not repeated. The real-time monitoring program continuously checks the port connection status. When it is found that the port connection of a device is disconnected, the device reconnection mechanism is triggered to try to reestablish the connection, such as sending a reconnection request instruction. According to the device response, it is decided whether to try to reconnect again to ensure stable communication between the device and the system.

[0065] The present invention also includes a method for automatic coordinated control of smart home devices, comprising the following steps: Step S1: Connect the home device to the system through the device access module and assign a unique identification code; Step S2: using the environment perception module to collect environmental parameters in real time and dynamically compare them with preset thresholds; Step S3: Generate a user scenario preference model through a user behavior analysis module; Step S4: Generate a device collaborative control strategy based on a multi-objective optimization algorithm; Step S5: using the dynamic priority management module to update the policy weight; Step S6: Execute the control strategy through the device control module and feedback the device status; Step S7: When it is detected that the execution deviation exceeds the threshold, a policy regeneration mechanism is triggered; Step S8: Store all operation data and policy versions, and receive real-time instructions through the user interaction module.

[0066] The implementation of this method refers to the above embodiment and will not be described in detail in the specification.

[0067] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0068] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An automatic coordinated control system for smart home devices, characterized in that: The system comprises: The device access module is used to connect the lamps, home appliances and sensor devices in the home to the system through various IoT communication protocols, and assign a unique identification code to each device; Environmental sensing module, used to collect home environment parameters in real time, including temperature, humidity, light intensity and air quality index, and dynamically compare the parameters with preset threshold ranges; User behavior analysis module, which is used to perform cluster analysis on user historical operation data and real-time operation instructions through machine learning algorithms to generate user scenario preference models; A collaborative strategy generation module is used to generate a device collaborative control strategy based on a multi-objective optimization algorithm according to the environmental parameter comparison results, the user scenario preference model and the device operation status; Dynamic priority management module, which is used to assign dynamic priority weights to each device according to device type, energy consumption level and scene urgency, and update it to the collaborative control strategy; The device control module is used to send control instructions to the target device according to the collaborative control strategy and provide real-time feedback on the device execution status; The anomaly detection module is used to monitor the deviation between the execution status of the device and the expected status. When the deviation exceeds the preset fault tolerance threshold, the policy regeneration mechanism is triggered; Data storage module, used to store device identification code, environmental parameter historical data, user behavior model and collaborative strategy version information; The user interaction module is used to receive user voice or touch commands, and parse the commands into standardized operation commands and input them into the user behavior analysis module.

2. The automatic coordinated control system for smart home devices according to claim 1, characterized in that: The specific analysis method of the user behavior analysis module is: Obtain the user's historical operation time series data from the data storage module, and extract the operation frequency, device relevance, and scene trigger condition characteristics; The hidden Markov model is used to calculate the state transition probability of real-time operation instructions and predict the user's next operation intention; Combining time series characteristics with state transition probabilities, a user scenario preference model including device linkage rules and weight distribution is generated.

3. The automatic coordinated control system for smart home devices according to claim 2, characterized in that: The specific generation method of the collaborative strategy generation module is: Define equipment energy consumption, response delay and scene matching as optimization objectives, and construct multi-objective constraint equations; Using NSGA-II algorithm to perform non-dominated sorting on the constraint equations to generate a Pareto optimal solution set; The final collaborative control strategy is selected from the optimal solution set according to the dynamic priority weight.

4. The automatic coordinated control system for smart home devices according to claim 3 is characterized in that: The specific allocation method of the dynamic priority management module is: Obtain the basic priority table corresponding to the device type from the local database, and map it to obtain the initial priority of each device; Calculate the dynamic energy consumption correction coefficient based on real-time energy consumption data and equipment load rate; The dynamic priority weight of the device is updated by combining the scene urgency and the correction coefficient through the fuzzy logic algorithm.

5. The automatic coordinated control system for smart home devices according to claim 4, characterized in that: The specific detection method of the anomaly detection module is: Obtain the difference Δt between the actual response time and the expected response time of the device from the device control module; The difference Δt is de-noised by using the Kalman filter algorithm to obtain a stable deviation value ΔT; If ΔT exceeds the preset fault tolerance threshold, the policy regeneration mechanism is triggered and the abnormal event is recorded in the data storage module.

6. The automatic coordinated control system for smart home devices according to claim 5, characterized in that: The specific implementation method of the strategy regeneration mechanism is: Freeze the current collaborative control strategy and back it up to the data storage module; Recall the real-time data of the environment perception module and the user behavior analysis module to generate alternative collaborative strategies; The deviation values ​​of the original strategy and the alternative strategy are compared through the A / B test algorithm, and the optimal strategy is selected to update to the device control module.

7. The automatic coordinated control system for smart home devices according to claim 6, characterized in that: The instruction parsing method of the user interaction module is: Convert voice commands into structured text through natural language processing technology, extract device operation verbs and target parameters; Use the knowledge graph to match the compatibility of device identification codes and operation verbs to generate standardized operation instructions; If the match fails, the instruction is corrected through the reinforcement learning model and fed back to the user for secondary confirmation.

8. The automatic coordinated control system for smart home devices according to claim 7, characterized in that: The storage method of the data storage module is: Use time series database partitions to store historical data of environmental parameters and create indexes based on equipment identification codes; The user scenario preference model and collaborative strategy version information are encrypted and stored through blockchain technology, and an unalterable hash check value is generated.

9. The automatic coordinated control system for smart home devices according to claim 8, characterized in that: The protocol compatibility method of the device access module is: The ZigBee, Wi-Fi and Bluetooth communication protocols are uniformly mapped to the MQTT standard data format through the protocol conversion middleware; Assign a virtual communication port to each device and monitor the port connection status in real time to trigger the device reconnection mechanism.

10. A method for automatic coordinated control of smart home devices, characterized in that: The following steps are involved: Step S1: Connect the home device to the system through the device access module and assign a unique identification code; Step S2: using the environment perception module to collect environmental parameters in real time and dynamically compare them with preset thresholds; Step S3: Generate a user scenario preference model through a user behavior analysis module; Step S4: Generate a device collaborative control strategy based on a multi-objective optimization algorithm; Step S5: using the dynamic priority management module to update the policy weight; Step S6: Execute the control strategy through the device control module and feedback the device status; Step S7: When it is detected that the execution deviation exceeds the threshold, a policy regeneration mechanism is triggered; Step S8: Store all operation data and policy versions, and receive real-time instructions through the user interaction module.

Images