Intelligent scene linkage control platform and equipment

Abstract

The invention relates to the technical field of scene linkage control, and discloses an intelligent scene linkage control platform and equipment, which comprises an environment monitoring model and a user behavior analysis model, the user behavior analysis model identifies a scene and predicts a user behavior through target detection data, and the user interface module comprises a user interface, a feedback unit, a scene management unit, an intelligent recommendation unit, a monitoring maintenance unit and a security privacy unit. The fault detection module realizes fault detection of the intelligent scene linkage control platform through the fault model, the complexity of user configuration is reduced, the usability of the system is improved, and missing report or false report caused by the fact that all types of faults are difficult to identify is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of scene linkage control and discloses a control platform and equipment for intelligent scene linkage. Background Art

[0002] Common deficiencies of intelligent scene linkage control platforms in their development process: Although many platforms provide powerful functions, their setup and configuration process may be too complicated for ordinary users and require professional knowledge or a lot of time to adjust and optimize; Creating and managing scenes (such as automation rules) may involve complex logic and multiple steps, which may be difficult for ordinary users to master; Intelligent control platforms usually rely on network connections, and network interruptions may cause system failure or inability to control devices normally; Software failures or untimely updates may cause system instability, affecting user experience; Scene trigger conditions are overly complex, and setting trigger conditions may require multiple configurations, making it difficult for users to set and manage; Sensor data may have delays, affecting real-time and accuracy; Intelligent recommendations are limited by the quality and quantity of data and are difficult to be completely accurate; Self-learning recommendations may require a large amount of data to train the model and are easily interfered by abnormal data; Monitoring and maintenance, real-time display and monitoring require efficient system support, otherwise there may be delays; Fault detection models may have difficulty identifying all types of faults, resulting in missed or false alarms; Limitations of automatic repair: The automatic repair function may not be able to handle complex fault situations and still requires manual intervention.

[0003] For example, the existing Chinese patent application with publication number CN116700035A discloses a smart scene linkage control method based on a smart home platform, which makes it more flexible for users to set smart scenes and more efficient to execute smart scenes. The method includes: A. Creating smart scenes and saving them on a cloud platform: the smart scene creation includes creating trigger scenes, creating linkage execution scenes, and configuring the effective time of smart scenes; B. The cloud platform starts device trigger detection tasks and timed trigger detection tasks through parallel threads to perform trigger detection; C. The cloud platform starts linkage scene execution task generation and linkage scene execution task execution through parallel threads to execute linkage scenes. The present invention is applicable to smart home scenes.

[0004] However, the above patents have the following problems: no specific scene recognition logic and method is provided, no real-time monitoring and control of environmental parameters and target detection within the scene is performed, and the scene control cannot be handled without a network. Faults within the scene cannot be adaptively adjusted, and no fault detection buffer is set, which greatly increases the losses caused by all types of faults that are difficult to identify. Connection to the cloud platform can easily lead to missed reports and false alarms. Summary of the invention

[0005] In order to solve the above technical problems, the main purpose of the present invention is to provide a control platform and device for intelligent scene linkage, wherein the control platform for intelligent scene linkage comprises: An analysis module, including an environment monitoring model and a user behavior analysis model, wherein the environment monitoring model is used to predict environmental changes, and the user behavior analysis model identifies scenes and predicts user behaviors through target detection data; User interface module, including user interface, feedback unit, scene management unit, intelligent recommendation unit, monitoring and maintenance unit and security and privacy unit; The fault detection module realizes fault detection of a control platform for intelligent scene linkage through a fault model.

[0006] As a preferred solution of a control platform for intelligent scene linkage of the present invention, wherein: Useless data in the environmental data is filtered out by the built-in data cleaning unit of the environmental monitoring model, and the environmental data in the filtered scene is monitored and predicted in real time; Adaptive adjustment of reliability weights Dynamically adjust adaptive weights to allow environmental monitoring models to focus on specific environmental parameters; The environmental data includes temperature data, humidity data and light data; The special environmental parameters include scene internal environmental parameter imbalance data.

[0007] As a preferred solution of a control platform for intelligent scene linkage of the present invention, wherein: Establish the number of labels detected by the target in the scene, and use the cross-optimization function to optimize the probabilities of the left and right positions of the target detected labels, so that the network distribution focuses on the label value. The calculation expression of the cross-optimization function is as follows: ; in, The network distribution focus value optimized for the cross optimization function, is the left position value of the target label detected by the user behavior analysis model, p is the correct target label detected by the user behavior analysis model, The right position value of the target label detected by the user behavior analysis model. The probability of focusing on the left side of the correct target label for the network distribution, The probability of focusing on the right side of the correct target label volume for the network distribution; Predict environmental data and user behavior through regression models.

[0008] As a preferred solution of a control platform for intelligent scene linkage of the present invention, wherein: The scene management unit includes scene creation and editing and scene triggering conditions; Intelligent recommendations include personalized recommendations and self-learning recommendations; Personalized recommendations are used to recommend scene settings based on user historical behavior and preferences; Identify user behavior patterns and preferences through environmental monitoring models and user behavior analysis models to provide customized scenario recommendations; Self-learning recommendations optimize scenario settings and recommendations by automatically learning user behavior patterns and environmental changes.

[0009] As a preferred solution of a control platform for intelligent scene linkage of the present invention, wherein: The fault detection model includes a data buffer, a buffer simulation unit and a buffer correction unit; The data buffer includes a buffer range and a fault buffer; The buffer simulation unit includes fault models, fault simulation applications, and fault evaluation; The buffer correction unit is used for fault evaluation.

[0010] As a preferred solution of a control platform for intelligent scene linkage of the present invention, wherein: Buffer scope determines the system components and functions that need to be monitored and sets the buffer capacity to cope with different types and sizes of failures; The fault model uses the environmental monitoring model and the user behavior analysis model to monitor the fault value as the data basis, establishes a buffer fault monitoring neural convolution network model, monitors and predicts the scene environment parameters and user behavior abnormal parameters in real time, and outputs the scene fault value; The scenario fault value is input into the fault simulation application, and a virtual environment is created through the fault simulation application to simulate the scenario fault value, and the response of the control platform of the intelligent scenario linkage is recorded.

[0011] As a preferred solution of a control platform for intelligent scene linkage of the present invention, wherein: Fault assessment receives the response from the control platform linked to the intelligent scenario, predicts the system status by training the fault assessment model, and determines whether a fault will occur. If the scenario fault value simulates a fault, the fault buffer parameters, system environment parameters or user behavior parameters are adjusted.

[0012] A control device for intelligent scene linkage, comprising: Environmental monitoring interface, environmental parameter control button, environmental change control button, target identification switch button, target detection parameter adjustment button, fault warning light, fault alarm light, equipment normal indicator light, equipment fault monitoring interface, target identification and monitoring interface.

[0013] As a preferred solution of a control device for intelligent scene linkage of the present invention, wherein: The environment monitoring interface, the equipment fault monitoring interface and the target identification and monitoring interface are located on the same plane, and the environment monitoring interface and the equipment fault monitoring interface are located on both sides of the target identification and monitoring interface respectively; The environmental parameter control button and the environmental change control button are located at the bottom of the environmental monitoring interface; The target recognition switch button and the target detection parameter adjustment button are located at the bottom of the target recognition and monitoring interface; The fault warning light, fault alarm light and equipment normal indicator light are located at the bottom of the equipment fault monitoring interface.

[0014] As a preferred solution of a control device for intelligent scene linkage of the present invention, wherein: The environment parameter control button is used to manually control the scene environment parameters; The environment change control button is used to manually switch scenes; The target recognition switch button is used to manually switch the target of scene target detection; The target detection parameter adjustment button is used to manually control the parameters of target detection; The fault warning light is used to respond when the platform reaches a critical fault value; The fault alarm light is used to sound an alarm when a platform fault occurs; The device normal indicator light is used to respond to the normal operation of the platform and the device.

[0015] Beneficial effects of the present invention: The preset templates and dynamic adjustments of the present invention reduce the complexity of user configuration and improve the ease of use of the system. Data caching and edge computing reduce delays, making the system respond more quickly and accurately to real-time events. The combination of multiple data sources and adaptive algorithms improves the accuracy and stability of intelligent recommendations to meet personalized needs. The hybrid detection method improves the accuracy of fault identification. The multi-layer automatic repair strategy ensures timely handling of faults and reduces the need for manual intervention. The fault detection buffer is set, which greatly reduces the difficulty in identifying all types of faults, resulting in missed reports or false reports. Faults can be adjusted adaptively; The present invention collects data parameters in real time to form network-free adaptive adjustment parameters to cope with the control of scenes without a network. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them: Figure 1 This is a system composition diagram of a control platform for intelligent scene linkage according to the present invention; Figure 2 A topological diagram of a control device for intelligent scene linkage according to the present invention; Figure 3 This is a flow chart of the cross-optimization function of the present invention for optimizing the network distribution focus value; Figure 4 The present invention completes the flow chart of correct target position identification.

[0017] Figure markings: 1. Environmental monitoring interface; 2. Environmental parameter control button; 3. Environmental change control button; 4. Target recognition switching button; 5. Target detection parameter adjustment button; 6. Fault warning light; 7. Fault alarm light; 8. Equipment normal indicator light; 9. Equipment fault monitoring interface; 10. Target recognition and monitoring interface. DETAILED DESCRIPTION

[0018] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0019] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0020] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0021] Example 1 like Figure 1 As shown, a control platform for intelligent scene linkage includes: Collection module, used to collect environmental data and user behavior data in real time; Among them, the acquisition module includes temperature sensor, humidity sensor, light sensor, motion sensor, etc.; The temperature sensor collects scene temperature data; The humidity sensor collects scene humidity data; The light sensor collects scene light data; The image sensor collects scene image data; Environmental data includes temperature data, humidity data, light data, etc. User behavior data includes object detection data identified by image data; A processing module, including a data storage unit and a data preprocessing unit; Data storage is used to store collected data and historical records by using a database or data warehouse; Data preprocessing is used to clean, standardize, and format environmental data and user behavior data for further analysis; Analysis module, including environment monitoring model and user behavior analysis model; Among them, the analysis module is used to analyze and predict data; Environmental monitoring models are used to predict environmental changes, such as temperature changes, energy consumption, etc. Since the environmental changes such as temperature change and energy consumption collect long time series data, the environmental monitoring model filters out useless data in the environmental data and monitors and predicts the environment by adaptively adjusting the reliability weight; Useless data in the environmental data is filtered out by the built-in data cleaning unit of the environmental monitoring model, and feature extraction is performed on the environmental data in the filtered scene, and characteristic values ​​such as temperature, humidity, and light in the environmental data are extracted respectively, and initial weights are set by convolution operation to obtain characteristic vectors of temperature, humidity, and light in the environmental data. The initial weights are dynamically adjusted according to the deviation between the environmental data output by the environmental monitoring model and the actual value until the difference between the environmental data output by the environmental monitoring model and the real data in the scene is minimized, and the training of the environmental monitoring model is completed. The environmental parameters in the scene are monitored and predicted in real time through the trained environmental monitoring model; After obtaining the predicted values ​​of the environmental parameters in the scene, the environment is judged. If the predicted values ​​of the environmental parameters are displayed as special environmental data, the temperature, humidity and light are compensated respectively through the scene linkage system. If the predicted values ​​of the environmental parameters are displayed as normal environmental data, the scene linkage system does not output any control instructions; Dynamically adjust weights for temperature, humidity, light and other data in the environment, so that the environmental monitoring model focuses on specific environmental parameters; Environmental data includes temperature data, humidity data and light data; Special environment parameters include the imbalance data of the environment parameters inside the scene.

[0022] The user behavior analysis model uses target detection data to identify scenarios and predict user behavior; like Figure 3As shown, the steps of optimizing the network distribution focus value by the cross-optimization function include marking the approximate rectangular area of ​​the object position in the image through the initial bounding box, detecting the left position value and the right position value of the label quantity detected by the user behavior analysis model, calculating the probability that the left position value is the left position of the correct label quantity of the target label quantity and the probability that the right position value is the right position of the correct label quantity of the target label quantity through the cross-optimization function, and adjusting the position of the bounding box by direction with reference to the pre-set correct target label quantity, and finally calculating the error weight between the actual position and the correct position to adjust the initial bounding box.

[0023] Establish the number of labels detected by the target in the scene. Due to the uneven density of scene labels and unclear object boundaries, the network distribution cannot focus on the label quantity. Therefore, the cross optimization function is used to optimize the probabilities of the left and right positions of the target detected labels. The probability is converted through the logarithmic function, and the probability value is mapped to [-∞, 0] to avoid overfitting, ensure numerical stability and determine continuous gradient smoothing. By estimating the probability that the left and right positions of the detection boundary box are correct values, the network distribution is focused near the label value. The calculation expression of the cross optimization function is as follows: ; in, The network distribution focus value optimized for the cross optimization function, is the left position value of the target label detected by the user behavior analysis model, p is the correct target label detected by the user behavior analysis model, The right position value of the target label detected by the user behavior analysis model. The probability of focusing on the left side of the correct target label for the network distribution, The probability of focusing on the right side of the correct target label volume for the network distribution; Furthermore, the left position value of the label is the left position coordinate of the correct target bounding box, and the right position is the right position coordinate of the correct target bounding box. The bounding box is a rectangular area used to mark the position of the object in the image; Further, determine the position (bounding box) of the object and the category to which the object belongs, and optimize the position of the detected target label so that the detection result more accurately reflects the position of the actual target, wherein the position of the actual target includes the left position and the right position, the label value difference represents the gap between the detected position and the actual position, the label probability includes the probability that the network distribution focuses on the left side of the correct target label amount and the probability that the network distribution focuses on the right side of the correct target label amount, the label probability is used to describe the probability that the detected position is the correct position, and the label difference is multiplied by the label probability to obtain the directional error weight; Furthermore, the direction is used to describe whether the label value difference is positive or negative. If it is positive, it is the left position, and if it is negative, it is the right position. The error weight is used to describe the influence of the label probability on the error. The influence of the label probability on the error reflects the product of the uncertainty of the target detection for the detection position and the actual error.

[0024] Predict environmental data and user behavior through regression models; Control module, including device control and action execution; Device control controls various devices such as lights, air conditioners, curtains, etc. through interfaces.

[0025] Action execution controls the device according to the environmental monitoring model and executes corresponding actions or scene settings according to the user behavior analysis model; User interface module, including user interface, feedback unit, scene management unit, intelligent recommendation unit, monitoring and maintenance unit and security and privacy unit; The user interface includes mobile applications, web interfaces, voice control, etc., allowing users to view and manage smart scenes; The feedback unit is used to provide real-time feedback and notifications to let users know the current system status and events; The scene management unit includes scene creation and editing and scene triggering conditions; User scenario creation and editing Create, edit and delete various scenarios, such as "Leaving Home Mode", "Returning Home Mode" and so on.

[0026] Set the scene trigger conditions by setting trigger conditions, such as time, sensor data threshold, user behavior, etc. For example: set a specific time or periodic time (such as daily, weekly) to trigger a scene, trigger a scene based on a calendar event or appointment, for example, start "Morning Mode" at 7 am on weekdays.

[0027] Use GPS or Wi-Fi positioning to trigger scenarios. For example, when a user is more than 5 kilometers away from home, the "away mode" is automatically enabled, and the user behavior analysis model monitors when the user enters or leaves a specific geographic area to trigger a scenario.

[0028] The scenario data is monitored based on the user behavior analysis model to trigger the scenario. For example, the "home mode" is automatically activated when the door is opened.

[0029] The environmental monitoring model monitors the power changes in the scene, such as triggering a specific scene when the mobile phone battery is low.

[0030] Intelligent recommendations include personalized recommendations and self-learning recommendations; Personalized recommendations are used to recommend suitable scene settings based on user historical behavior and preferences; Identify user behavior patterns and preferences through environmental monitoring models and user behavior analysis models to provide customized scene recommendations. For example, if the user often dims the lights at night, the system will recommend similar night mode settings; Users select or set preferences, and the system recommends scenes that meet their needs. For example, if a user marks "prefers a warm environment", the system will recommend the corresponding temperature setting scene; Self-learning recommendations optimize scenario settings and recommendations by automatically learning user behavior patterns and environmental changes.

[0031] The platform uses a user behavior analysis model to analyze user behavior in real time, and an environmental monitoring model to monitor environmental changes, in order to automatically adjust recommendations. For example, if the system detects that a user frequently turns on music during a specific time period, it will automatically recommend music-related scene settings.

[0032] The platform optimizes recommendations by adapting to users’ new behavior patterns through machine learning algorithms and adaptive overlays of new user behaviors and habits. For example, if a user starts working from home, the system may recommend a “work mode” scenario.

[0033] The monitoring and maintenance unit is used to display the status and environmental data of each device in real time, detect equipment failures or abnormal conditions, and provide maintenance suggestions or automatic repairs; The fault detection module realizes fault detection of a control platform for intelligent scene linkage through a fault model; The fault detection model includes a data buffer, a buffer simulation unit and a buffer correction unit; The data buffer includes a buffer range and a fault buffer; The buffer range determines the system components and functions that need to be monitored, such as the real-time status of the equipment (including the startup, operation, and stop status of the equipment and the operating parameters of key components, etc.), network connection status (such as network bandwidth, packet loss rate, connection stability, etc.), data processing flow (such as data collection, transmission, storage, and processing efficiency and accuracy, etc.); The fault buffer focuses on setting the capacity of the buffer to cope with faults of different types and sizes. The capacity setting needs to comprehensively consider factors such as the complexity of the system, the frequency of data generation, and the severity of the impact of potential faults on the system. The buffer simulation unit includes fault models, fault simulation applications, and fault evaluation; The fault model uses the fault values ​​monitored by the environment monitoring model and the user behavior analysis model as basic data input. The environment monitoring model is used to collect and analyze scene environmental parameters in real time, such as the changes in environmental indicators such as temperature, humidity, and air quality in the system operation environment. The user behavior analysis model obtains the abnormal behavior parameters generated during the interaction between the user and the system, such as whether the user's operation frequency, operation path, operation time, etc. deviate from the normal mode; Furthermore, after weighted data fusion of the preprocessed environmental parameters and user behavior data, fault features are extracted, and the correlation between each fault feature and whether a fault occurs is obtained through correlation analysis, and whether the fault feature needs to be retained is determined based on the correlation; Furthermore, a tree-like decision structure is constructed to make branch judgments based on different values ​​of fault characteristics and output whether a fault exists. For example, when judging whether a scene has a fault, branches are first made based on the temperature characteristics in the scene. If the temperature is too high, the judgment is further subdivided based on other characteristics such as the air conditioning usage rate in the scene to gradually determine the fault situation. The scenario fault values ​​output by the fault model are input into the fault simulation application; Furthermore, the fault simulation application creates a virtual environment, takes the received scenario fault value as the basic data, and simulates the fault scenario of the scenario fault value. Through the virtual environment, the scenario linkage system completely reproduces the various states and conditions when the fault occurs, and records the response of the intelligent scenario linkage control platform when facing the simulated fault; The buffer correction unit is used to receive the result of fault assessment and perform correction and optimization compensation on the system environmental parameters; The fault assessment receives the response of the control platform linked with the intelligent scene, and judges whether the scene environment data and behavior data will cause a scene fault by training the fault assessment model. If the scene fault value input fault assessment calculation is abnormal, the fault buffer parameters are adjusted until the scene environment data and behavior data change without a fault, then the instructions for adjusting the system environment parameters or user behavior parameters are output to avoid actual system faults. The fault assessment model inputs the fault model through the scenario fault value, and determines whether the scenario fault value will cause platform failure through the fault model recall rate or mean square error; Update the model regularly based on the latest system data and fault records, collect the actual effects of system operation and fault buffering, and optimize the model and algorithm based on feedback information; Security and privacy include data encryption and permission management; Data encryption is used to ensure the security of data in transmission and storage.

[0034] Permission management is used to set permissions for different users to protect the security and privacy of the system.

[0035] Example 2 like Figure 2 Shown: A control device for intelligent scene linkage, comprising: Environmental monitoring interface 1, environmental parameter control button 2, environmental change control button 3, target identification switch button 4, target detection parameter adjustment button 5, fault warning light 6, fault alarm light 7, equipment normal indicator light 8, equipment fault monitoring interface 9, target identification and monitoring interface 10; The environment monitoring interface 1, the equipment fault monitoring interface 9 and the target identification and monitoring interface 1 are located on the same plane, and the environment monitoring interface 1 and the equipment fault monitoring interface 9 are located on both sides of the target identification and monitoring interface 1 respectively; The environmental parameter control button 2 and the environmental change control button 3 are located at the bottom of the environmental monitoring interface 1; The target recognition switching button 4 and the target detection parameter adjustment button 5 are located at the bottom of the target recognition and monitoring interface 1; The fault warning light 6, the fault alarm light 7 and the equipment normal indicator light 8 are located at the bottom of the equipment fault monitoring interface 9; Furthermore, the environment parameter control button 2 is used to manually control the scene environment parameters; The environment change control button 3 is used to manually switch scenes; The target recognition switch button 4 is used to manually switch the target of scene target detection; The target detection parameter adjustment button 5 is used to manually control the target detection parameters; The fault warning light 6 is used to respond when the platform reaches a critical fault value; The fault alarm light 7 is used to sound an alarm when a fault occurs on the platform; The device normal indicator light 8 is used to respond to the normal operation of the platform and the device.

[0036] Example 3 like Figure 4 As shown in the figure, the triangular area is the correct target position in the scene linkage recognition scene, and the rectangular area is the recognition area. Figure 4 In the first step, the recognition border is on the right side of the correct target position, so it is adaptively adjusted to the correct one. In the second step, the recognition border is on the left side of the correct target position, and it is continuously corrected until the fourth step recognizes the correct target position. After identifying the correct target position, the platform monitors and triggers the scene linkage. The control system and control equipment of the intelligent scene linkage are combined with the environmental monitoring model to monitor the scene environmental parameters and user personalization and other factors to control the scene target.

[0037] It is important to note that the construction and arrangement of the present application shown in a plurality of different exemplary embodiments are only exemplary. Although only two embodiments are described in detail in this disclosure, it should be readily understood by those who refer to this disclosure that many modifications are possible, for example, the size, scale, structure, shape and ratio of various elements, and parameter values ​​(e.g., temperature, pressure, etc.), installation arrangement, use of materials, color, directional changes, etc., without substantially departing from the novel teachings and advantages of the subject matter described in the application. For example, the element shown as integrally formed can be composed of multiple parts or elements, the position of the element can be inverted or otherwise changed, and the nature or number or position of the discrete element can be changed or changed. Therefore, all such modifications are intended to be included in the scope of the present invention. The order or sequence of any process or method steps can be changed or reordered according to alternative embodiments. Any "device plus function" clause is intended to cover the structure of the execution function described in this article, and is not only structurally equivalent but also equivalent structure. Without departing from the scope of the present invention, other replacements, modifications, changes and omissions can be made in the design, operating conditions and arrangement of the exemplary embodiment. Therefore, the invention is not limited to a specific embodiment, but extends to numerous modifications still falling within the scope of the appended claims.

[0038] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment may not be described (ie, those features that are not relevant to the best mode presently contemplated for carrying out the invention or those features that are not relevant to implementing the invention).

[0039] It should be understood that in the development of any actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but for those of ordinary skill having the benefit of this disclosure, the development effort will be a routine task of design, fabrication, and production without undue experimentation.

[0040] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A control platform for intelligent scene linkage, characterized in that: include: An analysis module, including an environment monitoring model and a user behavior analysis model, wherein the environment monitoring model is used to predict environmental changes, and the user behavior analysis model identifies scenes and predicts user behaviors through target detection data; User interface module, including user interface, feedback unit, scene management unit, intelligent recommendation unit, monitoring and maintenance unit and security and privacy unit; The fault detection module realizes fault detection of a control platform for intelligent scene linkage through a fault model.

2. According to claim 1, a control platform for intelligent scene linkage is characterized in that: Useless data in the environmental data is filtered out by the built-in data cleaning unit of the environmental monitoring model, and the environmental data in the filtered scene is monitored and predicted in real time. After the predicted value of the environmental parameter in the scene is obtained, the environment is judged. If the predicted value of the environmental parameter is displayed as special environmental data, the temperature, humidity and light are compensated respectively through the scene linkage system. If the predicted value of the environmental parameter is displayed as normal environmental data, the scene linkage system does not output any control instructions; The environmental data includes temperature data, humidity data and light data; The special environmental parameters include scene internal environmental parameter imbalance data.

3. According to claim 2, a control platform for intelligent scene linkage is characterized in that: Establish the number of labels detected by the target in the scene, and use the cross-optimization function to optimize the probabilities of the left and right positions of the target detected labels, so that the network distribution focuses on the label value. The calculation expression of the cross-optimization function is as follows: ； in, The network distribution focus value optimized for the cross optimization function, is the left position value of the target label detected by the user behavior analysis model, p is the correct target label detected by the user behavior analysis model, The right position value of the target label detected by the user behavior analysis model. The probability of focusing on the left side of the correct target label for the network distribution, The probability of focusing on the right side of the correct target label volume for the network distribution; Predict environmental data and user behavior through regression models.

4. The control platform for intelligent scene linkage according to claim 3 is characterized by: The scene management unit includes scene creation and editing and scene triggering conditions; Intelligent recommendations include personalized recommendations and self-learning recommendations; Personalized recommendations are used to recommend scene settings based on user historical behavior and preferences; Identify user behavior patterns and preferences through environmental monitoring models and user behavior analysis models to provide customized scenario recommendations; Self-learning recommendations optimize scenario settings and recommendations by automatically learning user behavior patterns and environmental changes.

5. The control platform for intelligent scene linkage according to claim 4 is characterized by: The fault detection model includes a data buffer, a buffer simulation unit and a buffer correction unit; The data buffer includes a buffer range and a fault buffer; The buffer simulation unit includes fault models, fault simulation applications, and fault evaluation; Extract fault features through the fault model, and obtain the correlation between each fault feature and whether the fault occurs through correlation analysis, and determine whether the fault feature is retained based on the correlation; A tree-like decision structure is constructed to make branch judgments based on different values ​​of fault characteristics and output whether a fault occurs or not. The scenario fault value output by the fault model is input into the fault simulation application.

6. The control platform for intelligent scene linkage according to claim 5, characterized in that: The fault simulation application creates a virtual environment and simulates the fault scenario of the scenario fault value based on the received scenario fault value; The buffer correction unit is used to receive the result of fault assessment and perform correction and optimization compensation on the system environment parameters.

7. The control platform for intelligent scene linkage according to claim 6, characterized in that: By training the fault assessment model, we can determine whether changes in scene environment data and behavior data will cause scene faults. If the scene fault value input into the fault assessment calculation is abnormal, the fault buffer parameters are adjusted until no fault occurs after changes in the scene environment data and behavior data. Then, we output instructions for adjusting system environment parameters or user behavior parameters.

8. A control device for intelligent scene linkage, applied to a control platform for intelligent scene linkage as claimed in any one of claims 1 to 7, characterized in that: include: Environmental monitoring interface (1), environmental parameter control button (2), environmental change control button (3), target recognition switching button (4), target detection parameter adjustment button (5), fault warning light (6), fault alarm light (7), equipment normal indicator light (8), equipment fault monitoring interface (9), target recognition and monitoring interface (10).

9. The intelligent scene linkage control device according to claim 8, characterized in that: The environment monitoring interface (1), the equipment fault monitoring interface (9) and the target identification and monitoring interface (10) are located on the same plane, and the environment monitoring interface (1) and the equipment fault monitoring interface (9) are located on both sides of the target identification and monitoring interface (10); The environmental parameter control button (2) and the environmental change control button (3) are located at the bottom of the environmental monitoring interface (1); The target recognition switching button (4) and the target detection parameter adjustment button (5) are located at the bottom of the target recognition and monitoring interface (10); The fault warning light (6), the fault alarm light (7) and the equipment normal indicator light (8) are located at the bottom of the equipment fault monitoring interface (9).

10. The intelligent scene linkage control device according to claim 8, characterized in that: The environment parameter control button (2) is used to manually control the scene environment parameters; The environment change control button (3) is used to manually switch scenes; The target recognition switch button (4) is used to manually switch the target of scene target detection; The target detection parameter adjustment button (5) is used to manually control the target detection parameters; The fault warning light (6) is used to respond when the platform reaches a critical fault value; The fault alarm light (7) is used to sound an alarm when a fault occurs on the platform; The device normal indicator light (8) is used to respond to the normal operation of the platform and the device.

Images