Intelligent Device Management Cloud Platform and Method Based on IoT and Big Data Analytics

Abstract

This invention provides a cloud platform and method for intelligent device management based on the Internet of Things (IoT) and big data analytics. The platform includes: a device data acquisition subsystem for connecting multiple intelligent devices requiring management and acquiring their device data based on IoT technology; a big data analytics subsystem for analyzing the device data using big data analytics techniques and generating analysis results; and a dynamic display subsystem for dynamically displaying the device data and analysis results on a smart dashboard. This invention's cloud platform and method for intelligent device management, based on IoT and big data analytics, connects to multiple intelligent devices requiring management to acquire device data; introduces big data analytics techniques to analyze the device data and obtain analysis results, improving the accuracy of the analysis results; and dynamically displays the device data and analysis results on a smart dashboard for a more intuitive viewing experience.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and in particular to a cloud platform and method for managing intelligent devices based on IoT and big data analytics. Background Technology

[0002] The Internet of Things (IoT) is a technology that connects physical devices (such as sensors, controllers, and smart devices) via the internet, enabling data interaction and collaborative work between these devices. Its core lies in the interconnection of things, allowing devices to autonomously sense and collect data, and process it intelligently, thereby achieving intelligent management and control. Big data analytics is used to process and extract value from this data, including data cleaning, storage, analysis, and visualization. Through this combination, IoT systems can achieve more efficient decision support and intelligent management.

[0003] However, the data uploaded by smart devices is multimodal and multidimensional. Analyzing complex data types, whether manually or through large models, can easily lead to analytical errors, thus affecting the accuracy of the analysis results.

[0004] In view of this, there is an urgent need for a cloud platform and method for intelligent device management based on the Internet of Things and big data analytics, in order to at least address the above-mentioned shortcomings. Summary of the Invention

[0005] One of the objectives of this invention is to provide a cloud platform and method for managing intelligent devices based on the Internet of Things and big data analysis, which connects to multiple intelligent devices that need to be managed to obtain device data; introduces big data analysis technology to analyze the device data and obtain analysis results, thereby improving the accuracy of the analysis results; and dynamically displays the device data and analysis results on an intelligent dashboard for a more intuitive experience.

[0006] The intelligent device management cloud platform based on the Internet of Things and big data analysis provided in this embodiment of the invention includes:

[0007] The device data acquisition subsystem is used to connect multiple smart devices that need to be managed based on Internet of Things (IoT) technology and acquire device data from these smart devices. The device data includes: working status, operating parameters, and maintenance information.

[0008] The big data analytics subsystem is used to analyze equipment data based on big data analytics technology and generate analysis results, including fault warnings, energy efficiency reports, and operating trends.

[0009] The dynamic display subsystem is used to dynamically display device data and analysis results based on intelligent dashboards.

[0010] Preferably, the device data acquisition subsystem is based on Internet of Things (IoT) technology and connects to multiple smart devices that need to be managed, including:

[0011] Obtain the API call request from the required access device;

[0012] Obtain the parsing information of open API calls to API;

[0013] The parsed information is processed accordingly to generate response data;

[0014] The response data will be sent back via an open API.

[0015] Preferably, the big data analytics subsystem analyzes device data based on big data analytics technology and generates analysis results, including:

[0016] Data tags for acquiring device data;

[0017] Connect to the big data platform to obtain analytical big data corresponding to the data tags in the big data platform;

[0018] Based on the analysis of big data, a deep learning model is trained to obtain a large-scale analytical model for the data type corresponding to the data labels.

[0019] Input the device data into the large-scale analysis model corresponding to the data type of the data label according to the corresponding data label, and obtain the analysis results.

[0020] Preferably, the big data analytics subsystem interfaces with the big data platform to obtain analytical big data corresponding to the data tags in the big data platform, including:

[0021] Obtain page tags from the big data platform;

[0022] Input page tags, data tags, and tag relevance query statements into the large language model to obtain tag relevance;

[0023] If the tag relevance is greater than or equal to the preset tag relevance threshold, retrieve the page content of the subordinate pages of the page tag;

[0024] Predict the browsing intent of users accessing page content;

[0025] If the access intent is to query the analysis process of the target device data corresponding to the data tag, obtain the historical page browsing information of the user with the corresponding access intent;

[0026] Based on historical page browsing information, big data is analyzed to pinpoint the location.

[0027] Preferably, the big data analytics subsystem locates and analyzes big data based on historical page browsing information, including:

[0028] Determine the preset page browsing progress markers based on historical page browsing information, and calculate the movement rate of the page browsing progress markers;

[0029] If the movement speed is less than the average movement speed within a preset time period, locate the first browsing page;

[0030] Determine the horizontal line where the virtual pointer of the first browsing page is located;

[0031] Analyze the first line of sight information on the first browsing page to obtain the trajectory of the line of sight points on both sides of the horizontal line;

[0032] Based on the trajectory of the line of sight, attempt to obtain the scanning features;

[0033] If the attempt to acquire the data is successful, the partial page in the first browsing page is determined based on the scanning point trajectory and used as the second browsing page.

[0034] If the attempt to obtain the data fails, determine the first viewing area above the horizontal line and the second viewing area below the horizontal line on the first browsing page based on the trajectory of the line of sight.

[0035] Compare the density of line-of-sight points in the first and second viewing areas, and determine the first or second viewing area with a higher density of line-of-sight points as the third browsing page;

[0036] The second or third browsing page is used as the fourth browsing page. Text recognition is performed on the fourth browsing page to determine the text recognition result.

[0037] If the gaze duration of the recognized character in the text recognition result is greater than or equal to the preset duration threshold, the corresponding text recognition result will be used as big data for analysis.

[0038] Preferably, the dynamic display subsystem dynamically displays device data and analysis results based on a smart dashboard, including:

[0039] Based on device data and a pre-set virtual entity network construction template, construct a virtual entity network;

[0040] Based on the analysis results and the correspondence between virtual entities in the virtual entity network, associate the virtual entities with the analysis results;

[0041] Configure the trigger conditions for viewing the analysis results of virtual entity associations;

[0042] Get the current operation information of the smart dashboard;

[0043] The system dynamically displays information based on the trigger conditions and current operations on the smart dashboard.

[0044] Preferably, the triggering conditions for viewing the analysis results of the dynamic display subsystem configuration virtual entity association include:

[0045] When the warning characteristics of the smart device corresponding to the virtual entity meet the first target warning characteristics, and the current operation characteristics of the virtual entity meet any standard operation characteristics, the analysis results of the association of the first target virtual entity are displayed. The first target virtual entity is: the virtual entity that affects access to the cloud platform as determined by the first target warning characteristics.

[0046] When the warning features of the smart device corresponding to the virtual entity meet the warning features of the second target, and the current operation features of the virtual entity meet any standard operation features, the analysis results of the association of the second target virtual entity are displayed. The second target virtual entity is: the virtual entity and the traceable virtual entity. The traceable virtual entity is: the virtual entity of the smart device that caused the smart device to generate the warning features of the second target.

[0047] When the smart device corresponding to the virtual entity does not have any warning information, and the current operation characteristics of the virtual entity meet any standard operation characteristics, the analysis results associated with the corresponding virtual entity are displayed.

[0048] The standard operating characteristics include: the viewing duration of the virtual entity is greater than or equal to the preset viewing duration threshold and the virtual entity is clicked by the Kanban control pointer.

[0049] Preferably, the dynamic display subsystem obtains the current operation information of the smart dashboard, including:

[0050] Obtain second-line view information from the smart dashboard;

[0051] Obtain pointer operations for Kanban control pointers;

[0052] The second line of sight information and pointer operation are used together as the current operation information.

[0053] Preferably, the dynamic display subsystem dynamically displays information based on viewing trigger conditions and the current operation information of the smart dashboard, including:

[0054] Get the template for extracting conditional factors;

[0055] Based on the preset factor extraction timing determination rules, determine whether the virtual entity meets the viewing condition factor extraction timing.

[0056] If the conditions are met, the corresponding virtual entity will be used as the third target virtual entity;

[0057] Based on the third target virtual entity, current operation information, and viewing condition factor extraction template, extract the first viewing condition factor;

[0058] Based on the viewing trigger condition configuration information and viewing condition factor extraction template of the third target virtual entity, extract the second viewing condition factor;

[0059] Calculate the factor matching degree of the first viewing condition factor and the second viewing condition factor. If the factor matching degree is greater than or equal to the preset factor matching degree threshold, then determine the display content according to the viewing trigger condition of the corresponding third target virtual entity.

[0060] Preferably, the display is dynamically based on the viewing trigger conditions and the current operation information of the smart dashboard, and also includes:

[0061] Capture the gestures used by users when viewing the analysis results text box on the dashboard.

[0062] If successful, the operation gesture will be matched with the preset information capture operation gesture.

[0063] If the gesture is successfully matched, a pre-export reminder will be generated based on the preset pre-export reminder rules, and the result data in the corresponding analysis result text box will be pre-exported.

[0064] Get confirmation actions for the pre-export reminder. Confirmation actions include: confirm export and export cancellation.

[0065] If the confirmation operation is to confirm the export, the corresponding result data will be stored in the export form;

[0066] If the operation is confirmed to be an export withdrawal, retrieve the preceding gestures within a preset time period before the person viewing the dashboard made the operation gesture.

[0067] Extract target features from previous gesture operations. These target features include: different gesture actions, the sequential relationship between different gesture actions, and the time interval between them.

[0068] Clustering of preceding gesture operations based on target features;

[0069] If the clustering result meets the clustering conditions, the corresponding preceding gesture operation will be used as the pre-judgment action set of the operation gesture. The clustering conditions are: the number of clustering targets of the preceding gesture operation in the clustering result is greater than the first threshold, and the number of actions of the preceding gesture operation in the clustering result is greater than the second threshold.

[0070] Get the confirmation form for the exported form;

[0071] Based on the analytical entities and analytical entity knowledge graphs that confirm the result data of the form, extract graph features;

[0072] Train and analyze the knowledge model corresponding to the entity based on the characteristics of the graph;

[0073] Input the result data from the confirmed export form into the knowledge model, obtain the analysis results output by the knowledge model, and display the analysis results to the dashboard viewers;

[0074] The system dynamically displays information based on the trigger conditions and current operations on the smart dashboard, and also includes:

[0075] After the pre-judgment action set of the operation gesture is recorded, before the corresponding operation gesture is successfully matched and a pre-export reminder is generated in the future, the necessity analysis of the pre-export reminder is performed based on the pre-judgment action set of the corresponding operation gesture. If it is necessary, the pre-export reminder is performed; otherwise, no reminder is given.

[0076] The intelligent device management method based on the Internet of Things and big data analysis provided in this invention includes:

[0077] Step 1: Based on IoT technology, connect multiple smart devices that need to be managed and obtain device data from the smart devices; device data includes: working status, operating parameters and maintenance information;

[0078] Step 2: Analyze equipment data using big data analytics to generate analysis results; the results include: fault warnings, energy efficiency reports, and operational trends;

[0079] Step 3: Dynamically display device data and analysis results based on a smart dashboard.

[0080] The beneficial effects of this invention are as follows:

[0081] This invention connects to multiple smart devices that require management to acquire device data; it introduces big data analytics technology to analyze the device data and obtain analysis results, thereby improving the accuracy of the analysis results; and it dynamically displays the device data and analysis results on a smart dashboard for a more intuitive presentation.

[0082] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0083] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0084] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0085] Figure 1This is a schematic diagram of a smart device management cloud platform based on the Internet of Things and big data analysis in an embodiment of the present invention;

[0086] Figure 2 This is a schematic diagram of an intelligent device management method based on the Internet of Things and big data analysis in an embodiment of the present invention. Detailed Implementation

[0087] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0088] This invention provides a cloud platform for intelligent device management based on the Internet of Things and big data analytics, such as... Figure 1 As shown, it includes:

[0089] The device data acquisition subsystem 1 is used to connect multiple smart devices that need to be managed based on Internet of Things (IoT) technology and acquire device data from the smart devices. The device data includes: working status (e.g., "normal operation", "abnormal operation"), operating parameters (e.g., internal component temperature, electrical parameters (e.g., current, voltage)) and maintenance information (e.g., historical maintenance records (including: historical maintenance fault types, historical maintenance strategies, fault time and maintenance time)).

[0090] The device data acquisition subsystem is based on Internet of Things (IoT) technology and connects to multiple smart devices that need to be managed, including:

[0091] Obtain the API call request from the required access device; where the API call request is: the request sent by the required access device to the server (smart device management cloud platform), such as: the required access device sending a data reporting request to the smart device management cloud platform;

[0092] Obtain the parsing information of open API calls to APIs; where open APIs are: API interfaces publicly provided by the smart device management cloud platform that allow external systems to access the platform's functions and services; parsing information is: information obtained by the cloud platform after receiving the API call request and parsing the request content, the parsing process including verifying the request format and extracting request parameters;

[0093] The parsed information is processed accordingly to generate response data. The processing according to the parsed information includes: verifying the request and determining whether to execute the request, such as verifying the API key. If the verification is successful, the data reported by the device is stored in the database. The response data is the data returned by the cloud platform to the smart device after processing the API call request, such as returning a confirmation message that the data storage was successful.

[0094] The response data will be sent back via an open API;

[0095] Big Data Analytics Subsystem 2 is used to analyze equipment data based on big data analytics technology and generate analysis results, including: fault warnings, energy efficiency reports, and operating trends.

[0096] The dynamic display subsystem 3 is used to dynamically display device data and analysis results based on intelligent dashboards.

[0097] The working principle and beneficial effects of the above technical solution are as follows:

[0098] This invention relates to an intelligent device management cloud platform based on the Internet of Things (IoT) and big data analytics. This platform includes a range of management functions such as device management, user management, agent management, order management, after-sales management, and data analysis. The platform supports multi-role login and unified management of devices from multiple brands and categories. Users can independently edit dashboards to view the operational data of different products. It also features an open API for integration with enterprise management systems. The platform uses IoT technology to collect real-time data on the working status, operating parameters, and maintenance information of various devices and transmits this data to the cloud platform for storage. Through big data analytics, the platform analyzes the device's operational data to generate fault warnings, energy efficiency reports, and operational trends, helping users optimize device operation and maintenance strategies.

[0099] This invention connects to multiple smart devices that require management to acquire device data; it introduces big data analytics technology to analyze the device data and obtain analysis results, thereby improving the accuracy of the analysis results; and it dynamically displays the device data and analysis results on a smart dashboard for a more intuitive presentation.

[0100] In one embodiment, the big data analytics subsystem analyzes device data based on big data analytics technology and generates analysis results, including:

[0101] Obtain data tags for equipment data; where data tags are keywords used to classify and label equipment data, such as "working status data", "operating parameter data", and "maintenance information data", etc.

[0102] Connect to the big data platform to obtain analytical big data corresponding to the data tags. The analytical big data refers to the process data of analyzing data of the corresponding data type of the data tags in the big data platform. For example, if the big data platform is the Internet of Things exchange forum and the data tag is: working status data of remotely connected cameras, then the analytical big data would be: the discussion content in the discussion thread about handling experience of remotely connected cameras going offline, the discussion content in the discussion thread about handling experience of image upload delay of remotely connected cameras, etc.

[0103] The analysis model is trained by analyzing big data to obtain the analysis data type corresponding to the data label. The analysis data type is the analysis data category of the big data corresponding to the data label. For example, if the data label is: remotely connected camera working status data, then the analysis data category is: analyzing the working status of remotely connected cameras. During training, historical device data in the big data is used as the model input and the analysis results are used as the model output to train the preset neural network model until convergence.

[0104] Input the equipment data into the large analysis model corresponding to the data type of the data label according to the corresponding data label, and obtain the analysis results;

[0105] The connection to the big data platform, obtaining analytical big data corresponding to the data tags in the big data platform, includes:

[0106] Obtain page tags from the big data platform; these page tags are pre-defined labels within the big data platform page used to categorize page content, such as "Industry", "Logistics", "Smart City", and "Contact Us", which are then crawled.

[0107] Input page tags, data tags, and tag relevance query statements into a large language model to obtain tag relevance scores. The large language model can be GPT4, Claude2, etc. Tag relevance scores represent the degree of association between tag content, obtained by asking questions through the large language model. For example: Question: Please help me query the tag relevance score between "remotely connected camera working status data" and "industry," the model answers: 90 / 100; Another example: Question: Please help me query the tag relevance score between "remotely connected camera working status data" and "contact us," the model answers: 20 / 100.

[0108] If the tag relevance is greater than or equal to the preset tag relevance threshold, retrieve the page content of the subordinate pages of the page tag; where the preset tag relevance threshold is manually set, for example: 85;

[0109] Predict the access intent of users who access the page content; where users refer to those who have accessed the page content in the past; access intent is inferred from the search content entered by the users in the search box, which is the retrieval box of the web page content on the big data platform;

[0110] If the access intent is to query the analysis process of the target device data corresponding to the data tag, obtain the historical page browsing information of the user with the corresponding access intent; among which, the historical page browsing information is: the browsing action information of the user, including: page dwell data, page gaze data and operation data (such as: mouse click, selection, etc.);

[0111] Based on historical page browsing information, locate and analyze big data;

[0112] The method of locating and analyzing big data based on historical page browsing information includes:

[0113] Determine the preset page browsing progress markers based on historical page browsing information, and calculate the movement rate of the page browsing progress markers; wherein, the preset page browsing progress markers are: markers used to track the browsing progress of users when accessing pages, such as the progress indicators in the page browsing progress bar.

[0114] If the movement rate is less than the average movement rate within a preset time period, the first browsing page is located; wherein, the preset time period is: within 10 seconds before the determination time of the movement rate, and the determination time period is preset manually, for example: 1 second; the first browsing page is: the content displayed by the corresponding page display device when the movement rate of the page browsing progress mark is less than the average movement rate within the preset time period.

[0115] Determine the horizontal line where the virtual pointer of the first browsing page is located; where the virtual pointer is: a pointer that represents the current browsing focus of the accessing user, such as: the mouse pointer; the horizontal line where the virtual pointer is located is: a straight line that passes through the display point of the display device selected by the pointer, is perpendicular to the direction of change of the content displayed on the display device, and is in the plane where the display screen of the display device is located;

[0116] Analyze the first gaze information of the first browsing page to obtain the gaze point trajectory on both sides of the horizontal line; where the gaze point trajectory is: the gaze points of the visiting user on both sides of the horizontal line of the first browsing page.

[0117] Based on the trajectory of the gaze points, attempt to obtain the scanning features; where the scanning features are: the features of the content viewed by the user when browsing the first browsing page, such as: the point interval between adjacent gaze points is less than or equal to a preset point interval threshold, and the variance of the point interval is less than or equal to a preset variance threshold.

[0118] If the attempt to acquire the data is successful, the partial page in the first browsing page is determined based on the scanning point trajectory and used as the second browsing page.

[0119] If the attempt to obtain the data fails, determine the first viewing area above the horizontal line and the second viewing area below the horizontal line on the first browsing page based on the trajectory of the line of sight.

[0120] Compare the density of line-of-sight points in the first and second viewing areas, and determine the first or second viewing area with a higher density of line-of-sight points as the third browsing page; where the density of line-of-sight points is the result obtained by dividing the number of line-of-sight points by the area of ​​the viewing area.

[0121] The second or third browsing page is used as the fourth browsing page. Text recognition is performed on the fourth browsing page to determine the text recognition result. The text recognition result is the result of analyzing and recognizing the text content on the page.

[0122] If the gaze duration of the recognized character in the text recognition result is greater than or equal to a preset duration threshold, the corresponding text recognition result will be used as big data for analysis. The preset duration threshold is set manually in advance.

[0123] The working principle and beneficial effects of the above technical solution are as follows:

[0124] This invention introduces the movement rate of page browsing progress markers. When the movement rate is less than the average movement rate, the current first browsing page is determined. The first gaze information of the first browsing page is analyzed to obtain the gaze point trajectory on both sides of the horizontal line. Based on the gaze point trajectory, scanning features are attempted to be obtained. When scanning features are identified, it indicates that the target user has not yet located the word they want to view. Therefore, the second browsing page corresponding to all scanning point trajectories is used as the big data extraction page for analysis. When scanning features are not identified, it indicates that the viewer has initially located the area to be viewed. The density of gaze points in the first and second viewing areas is compared, and the first or second viewing area with a higher density of gaze points is determined as the third browsing page. Text recognition is performed on the second or third browsing page to determine the text recognition results. Text recognition results whose gaze dwell time for the recognized characters is greater than or equal to a preset duration threshold are determined as big data for analysis, thus improving the selection accuracy of big data for analysis.

[0125] In one embodiment, the dynamic display subsystem dynamically displays device data and analysis results based on a smart dashboard, including:

[0126] A virtual entity network is constructed based on device data and a preset virtual entity network construction template. The virtual entity network construction template includes multiple smart devices and digital mapping rules for the relationships between devices. The virtual entity network is an architecture that simulates the behavior and state of physical devices by referring to the virtual entity network construction template based on device data.

[0127] Based on the analysis results and the correspondence between virtual entities in the virtual entity network, associate the virtual entities with the analysis results;

[0128] Configure the trigger conditions for viewing the analysis results of virtual entity associations;

[0129] Obtain the current operation information of the smart dashboard; the current operation information includes: the second line of sight information of the smart dashboard (view the line of sight of the smart dashboard user) and the pointer operation of the dashboard control pointer;

[0130] The system dynamically displays information based on the trigger conditions and current operations on the smart dashboard.

[0131] The trigger conditions for viewing the analysis results of the configured virtual entity association include:

[0132] When the warning characteristics of the smart device corresponding to the virtual entity meet the first target warning characteristics, and the current operation characteristics of the virtual entity satisfy any standard operation characteristics, the analysis results associated with the first target virtual entity are displayed. The first target virtual entity is: the virtual entity that affects access to the cloud platform as determined by the first target warning characteristics; wherein, the warning characteristics are: the characteristic representation of the warning information when the smart device has warning information in the analysis results, and the first target warning characteristics are: device failure and communication interruption; the first target virtual entity includes: the virtual entity itself corresponding to the first target warning characteristics and the virtual entities corresponding to other smart devices it affects, such as: the lower-level smart device whose communication is interrupted due to the failure of the smart device corresponding to the first target warning characteristics;

[0133] When the warning characteristics of the smart device corresponding to the virtual entity meet the warning characteristics of the second target, and the current operation characteristics of the virtual entity satisfy any standard operation characteristics, the analysis results of the association of the second target virtual entity are displayed. The second target virtual entity is: the virtual entity and the traceable virtual entity. The traceable virtual entity is: the virtual entity of the smart device that caused the smart device to generate the second target warning characteristics; wherein, the second target warning characteristics are: communication interruption.

[0134] When the smart device corresponding to the virtual entity does not have any warning information, and the current operation characteristics of the virtual entity meet any standard operation characteristics, the analysis results associated with the corresponding virtual entity are displayed.

[0135] The standard operating characteristics include: the viewing time of the virtual entity is greater than or equal to the preset viewing time threshold and the virtual entity is clicked by the Kanban control pointer; the preset viewing time threshold is set manually, for example: 8 seconds;

[0136] The dynamic display based on the viewing trigger conditions and the current operation information of the smart dashboard includes:

[0137] Obtain the viewing condition factor extraction template; where the viewing condition factor extraction template is: a template used to extract viewing condition factors, and the viewing condition factor is: a description vector of the viewing condition, such as: [communication interrupted, view for 8 seconds, ...];

[0138] Based on the preset factor extraction timing judgment rules, it is determined whether the virtual entity meets the viewing condition factor extraction timing; the preset factor extraction timing judgment rules are set manually in advance, such as: the gaze stays on the virtual entity for 3 seconds, or the pointer stays for 2 seconds;

[0139] If the conditions are met, the corresponding virtual entity will be used as the third target virtual entity;

[0140] Based on the third target virtual entity, current operation information, and viewing condition factor extraction template, the first viewing condition factor is extracted; wherein, the first viewing condition factor is: the viewing condition description vector of the smart dashboard viewer viewing the third target virtual entity;

[0141] Based on the viewing trigger condition configuration information and viewing condition factor extraction template of the third target virtual entity, the second viewing condition factor is extracted; the second viewing condition factor is: the viewing condition description vector extracted by comparing the viewing trigger condition of the third target virtual entity with the viewing condition factor extraction template.

[0142] Calculate the factor matching degree between the first viewing condition factor and the second viewing condition factor. If the factor matching degree is greater than or equal to a preset factor matching degree threshold, then determine the displayed content based on the viewing trigger condition of the corresponding third target virtual entity. The factor matching degree is the vector cosine value of the first and second viewing condition factors; the preset factor matching degree threshold is manually set, for example, 0.85.

[0143] The working principle and beneficial effects of the above technical solution are as follows:

[0144] This invention introduces a preset virtual entity network construction template to visualize smart devices, obtain virtual entity networks, correlate analysis results and virtual entities in the virtual entity network, and configure the viewing trigger conditions for the analysis results associated with virtual entities.

[0145] When configuring the trigger conditions, if a virtual entity meets the first target warning characteristic and the dashboard viewer wants to view the corresponding device information, the analysis results of the device and its affected devices will be displayed to the dashboard viewer. This allows for a clear understanding of the impact range of the first target warning characteristic, facilitating subsequent maintenance. If a virtual entity meets the second target warning characteristic and the dashboard viewer wants to view the corresponding device information, the device and the analysis results that caused the device's failure will be displayed to the dashboard viewer. This helps the dashboard viewer quickly locate the source of the failure and improves the efficiency of fault handling. If the smart device corresponding to the virtual entity does not have warning information and the dashboard viewer wants to view the corresponding device information, it is a routine check, and the analysis results associated with the corresponding virtual entity will be displayed.

[0146] Next, the system acquires the current operation information of the smart dashboard in real time, introduces a viewing condition factor extraction template and a factor extraction timing determination rule, and extracts the first viewing condition factor of the third target virtual entity that the dashboard viewer may be interested in at the appropriate time. Then, it performs factor matching with the second viewing condition factor of the third target virtual entity. Based on the real-time matching results, it presents the corresponding triggered content that meets the viewing trigger conditions, making it more intelligent and intuitive.

[0147] In one embodiment, dynamically displaying information based on viewing trigger conditions and the current operation information of the smart dashboard further includes:

[0148] Capture the gestures used by users when viewing the analysis results text box on the dashboard; the gestures are hand gestures, such as grabbing or pinching.

[0149] If successful, the operation gesture will be matched with the preset information grabbing operation gesture; the preset information grabbing operation gesture is preset by humans, such as: grab;

[0150] If the gesture match is successful, a pre-export reminder will be generated based on the preset pre-export reminder rules, and the result data in the corresponding analysis result text box will be pre-exported. The preset pre-export reminder rules are: pop up a notification box asking "Do you want to export the analysis result you are viewing?" The notification box also includes "Yes" and "No" option buttons.

[0151] Get confirmation actions for the pre-export reminder. Confirmation actions include: confirm export and export cancellation.

[0152] If the confirmation operation is to confirm the export, the corresponding result data will be stored in the export form; the export form is a data table used to temporarily store the system data that the user wants to export.

[0153] If the operation is confirmed to be an export and withdrawal, retrieve the preceding gestures performed by the person viewing the dashboard within a preset time period before they made the operation gesture; the preset time period is manually set, for example: 5 minutes;

[0154] Extract target features from previous gesture operations. These target features include: different gesture actions, the sequential relationship between different gesture actions, and the time interval between them.

[0155] Clustering of preceding gesture operations is performed based on target features; during clustering, preceding gesture operations with a similarity to the target features greater than a similarity threshold are clustered.

[0156] If the clustering result meets the clustering conditions, the corresponding preceding gesture operation is used as the pre-judged action set of the operation gesture. The clustering conditions are: the number of clustering targets of the preceding gesture operation in the clustering result is greater than the first threshold, and the number of actions of the preceding gesture operation in the clustering result is greater than the second threshold. The first threshold and the second threshold are both preset manually. The number of clustering targets is the number of preceding gesture operations in the same clustering result. The number of actions is the number of gestures in the preceding gesture operation.

[0157] Get the confirmed export form; the confirmed export form is the data table containing all analysis results that the user finally confirms after viewing all the virtual entities they want to see.

[0158] Based on the analyzed entities and their knowledge graphs that led to the form's results, graph features are extracted. The analyzed entities are the actual physical devices corresponding to the results data, and the analyzed entity knowledge graph is a knowledge graph of these actual physical devices. The graph features are the entities (actual physical devices) and their relationships (the associations between actual physical devices, such as the mutual influence of different connection methods) in the analyzed entity knowledge graph.

[0159] The knowledge model corresponding to the entity is trained and analyzed based on the characteristics of the graph; the knowledge model is: an AI model that stores relevant knowledge of the entity being analyzed (e.g., the fault principles of different fault types);

[0160] Input the results data from the confirmed export form into the knowledge model, obtain the analysis results output by the knowledge model, and display the analysis results to the dashboard viewers;

[0161] The system dynamically displays information based on the trigger conditions and current operations on the smart dashboard, and also includes:

[0162] After the pre-determined action set of the operation gesture is recorded, before a successful gesture match and pre-export reminder are generated for the corresponding operation gesture in the future, a necessity analysis of the pre-export reminder is performed based on the pre-determined action set of the corresponding operation gesture. If it is necessary, a pre-export reminder is generated; otherwise, no reminder is generated. During the necessity analysis of the pre-export reminder, if the action matching degree between the future preceding gesture set corresponding to the action duration of the pre-determined action set and the pre-determined action set is greater than or equal to a preset action matching degree threshold, then it is unnecessary; otherwise, it is necessary. The preset action matching degree threshold is set manually.

[0163] The working principle and beneficial effects of the above technical solution are as follows:

[0164] When users view the analysis results of virtual entities on a smart dashboard, a situation arises where they want to export the analysis results they are currently viewing. Manual exporting requires selecting each data point individually, which is inconvenient. This invention matches the gestures of users viewing the analysis results text boxes with preset information capture gestures. When a match is successful, a pre-export reminder is generated to prevent misdirection. Next, the confirmation action of the pre-export reminder is obtained to determine whether the corresponding result data should be stored in the export form. If confirmed, since users may want to export multiple analysis results for comparison, a temporary form is created for temporary storage of the result data. If the export is cancelled, there is a possibility of accidental triggering of the information capture gesture. Therefore, the preceding gestures within a preset time period before the user made the gesture are obtained, and the target features of these historical preceding gestures are extracted and clustered. Clustering of preceding gesture operations is performed based on the characteristics of the data. A first threshold and a second threshold are introduced to limit the clustering conditions: the number of cluster targets for preceding gesture operations in the clustering results is greater than the first threshold, ensuring that the same situation (recognition of grabbing and subsequent withdrawal) occurs a sufficient number of times in the past; the number of actions in the preceding gesture operations in the clustering results is greater than the second threshold, ensuring the accuracy of the pre-judgment action set used for subsequent pre-judgment. When the clustering results meet the clustering conditions, they are used as the pre-judgment action set for the operation gestures. After the user confirms that all the necessary data is pre-exported to the export form, the real-time export form is used as the confirmed export form for export. Based on the analysis entity corresponding to the form, the analysis entity knowledge graph is retrieved to extract graph features. Based on the graph characteristics, an AI model that stores relevant knowledge of the analysis entity is trained. The result data in the confirmed export form is input into the knowledge model, which can output analysis results specifically according to the needs of the user, making it more user-friendly. In addition, after the pre-judgment action set of the operation gesture is recorded, before the corresponding operation gesture is successfully matched and a pre-export reminder is generated in the future, the necessity analysis of the pre-export reminder is performed based on the pre-judgment action set of the corresponding operation gesture, which reduces the number of false reminders and improves the user's system experience.

[0165] This invention provides a method for managing intelligent devices based on the Internet of Things and big data analytics, such as... Figure 2 As shown, it includes:

[0166] Step 1: Based on IoT technology, connect multiple smart devices that need to be managed and obtain device data from the smart devices; device data includes: working status, operating parameters and maintenance information;

[0167] Step 2: Analyze equipment data using big data analytics to generate analysis results; the results include: fault warnings, energy efficiency reports, and operational trends;

[0168] Step 3: Dynamically display device data and analysis results based on a smart dashboard.

[0169] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A cloud platform for intelligent device management based on the Internet of Things and big data analytics, characterized in that: include: The device data acquisition subsystem is used to connect multiple smart devices that need to be managed based on Internet of Things (IoT) technology and acquire device data from these smart devices. The device data includes: working status, operating parameters, and maintenance information. The big data analytics subsystem is used to analyze equipment data based on big data analytics technology and generate analysis results, including fault warnings, energy efficiency reports, and operating trends. Data tags for acquiring device data; Obtain page tags from the big data platform; Input page tags, data tags, and tag relevance query statements into the large language model to obtain tag relevance; If the tag relevance is greater than or equal to the preset tag relevance threshold, retrieve the page content of the subordinate pages of the page tag; Predict the browsing intent of users accessing page content; If the access intent is to query the analysis process of the target device data corresponding to the data tag, obtain the historical page browsing information of the user with the corresponding access intent; Based on historical page browsing information, locate and analyze big data; Determine the preset page browsing progress markers based on historical page browsing information, and calculate the movement rate of the page browsing progress markers; If the movement speed is less than the average movement speed within a preset time period, locate the first browsing page; Determine the horizontal line where the virtual pointer of the first browsing page is located; Analyze the first line of sight information on the first browsing page to obtain the trajectory of the line of sight points on both sides of the horizontal line; Based on the trajectory of the line of sight, attempt to obtain the scanning features; If the attempt to acquire the data is successful, the partial page in the first browsing page is determined based on the scanning point trajectory and used as the second browsing page. If the attempt to obtain the data fails, determine the first viewing area above the horizontal line and the second viewing area below the horizontal line on the first browsing page based on the trajectory of the line of sight. Compare the density of line-of-sight points in the first and second viewing areas, and determine the first or second viewing area with a higher density of line-of-sight points as the third browsing page; The second or third browsing page is used as the fourth browsing page. Text recognition is performed on the fourth browsing page to determine the text recognition result. If the gaze duration of the recognized character corresponding to the text recognition result is greater than or equal to the preset duration threshold, the corresponding text recognition result will be used as big data for analysis; the dynamic display subsystem is used to dynamically display device data and analysis results based on the smart dashboard; Based on device data and a pre-set virtual entity network construction template, construct a virtual entity network; Based on the analysis results and the correspondence between virtual entities in the virtual entity network, associate the virtual entities with the analysis results; Configure the trigger conditions for viewing the analysis results of virtual entity associations; Get the current operation information of the smart dashboard; The system dynamically displays information based on the trigger conditions and current operations on the smart dashboard.

2. The intelligent device management cloud platform based on the Internet of Things and big data analysis as described in claim 1, characterized in that, The device data acquisition subsystem is based on IoT technology and connects to multiple smart devices that need to be managed, including: Obtain the API call requests from the required access devices; Obtain the parsing information of the open API for API call requests; The parsed information is processed accordingly to generate response data; The response data will be sent back via the open API.

3. The intelligent device management cloud platform based on the Internet of Things and big data analysis as described in claim 1, characterized in that, The big data analytics subsystem analyzes equipment data based on big data analytics technology and generates analysis results, including: Connect to the big data platform to obtain analytical big data corresponding to the data tags in the big data platform; Based on the analysis of big data, a deep learning model is trained to obtain a large-scale analytical model for the data type corresponding to the data labels. Input the device data into the large-scale analysis model corresponding to the data type of the data label according to the corresponding data label, and obtain the analysis results.

4. The intelligent device management cloud platform based on the Internet of Things and big data analysis as described in claim 1, characterized in that, The trigger conditions for viewing the analysis results of the dynamic display subsystem configuration virtual entity association include: When the warning characteristics of the smart device corresponding to the virtual entity meet the first target warning characteristics, and the current operation characteristics of the virtual entity meet any standard operation characteristics, the analysis results of the association of the first target virtual entity are displayed. The first target virtual entity is: the virtual entity that affects access to the cloud platform as determined by the first target warning characteristics. When the warning features of the smart device corresponding to the virtual entity meet the warning features of the second target, and the current operation features of the virtual entity meet any standard operation features, the analysis results of the association of the second target virtual entity are displayed. The second target virtual entity is: the virtual entity and the traceable virtual entity. The traceable virtual entity is: the virtual entity of the smart device that caused the smart device to generate the warning features of the second target. When the smart device corresponding to the virtual entity does not have any warning information, and the current operation characteristics of the virtual entity meet any standard operation characteristics, the analysis results associated with the corresponding virtual entity are displayed. The standard operating characteristics include: the viewing duration of the virtual entity is greater than or equal to the preset viewing duration threshold and the virtual entity is clicked by the Kanban control pointer.

5. The intelligent device management cloud platform based on the Internet of Things and big data analysis as described in claim 1, characterized in that, The dynamic display subsystem obtains the current operation information of the smart dashboard, including: Obtain second-line view information from the smart dashboard; Obtain pointer operations for Kanban control pointers; The second line of sight information and pointer operation are used together as the current operation information.

6. The intelligent device management cloud platform based on the Internet of Things and big data analysis as described in claim 1, characterized in that, The dynamic display subsystem dynamically displays information based on viewing trigger conditions and the current operation information of the smart dashboard, including: Get the template for extracting conditional factors; Based on the preset factor extraction timing determination rules, determine whether the virtual entity meets the viewing condition factor extraction timing. If the conditions are met, the corresponding virtual entity will be used as the third target virtual entity; Based on the third target virtual entity, current operation information, and viewing condition factor extraction template, extract the first viewing condition factor; Based on the viewing trigger condition configuration information and viewing condition factor extraction template of the third target virtual entity, extract the second viewing condition factor; Calculate the factor matching degree of the first viewing condition factor and the second viewing condition factor. If the factor matching degree is greater than or equal to the preset factor matching degree threshold, then determine the display content according to the viewing trigger condition of the corresponding third target virtual entity.

7. A smart device management method based on the Internet of Things and big data analytics, characterized in that: include: Step 1: Based on IoT technology, connect multiple smart devices that need to be managed and obtain device data from the smart devices; Equipment data includes: operating status, operating parameters, and maintenance information; Step 2: Analyze equipment data using big data analytics to generate analysis results; the results include: fault warnings, energy efficiency reports, and operational trends; Data tags for acquiring device data; Obtain page tags from the big data platform; Input page tags, data tags, and tag relevance query statements into the large language model to obtain tag relevance; If the tag relevance is greater than or equal to the preset tag relevance threshold, retrieve the page content of the subordinate pages of the page tag; Predict the browsing intent of users accessing page content; If the access intent is to query the analysis process of the target device data corresponding to the data tag, obtain the historical page browsing information of the user with the corresponding access intent; Based on historical page browsing information, locate and analyze big data; Determine the preset page browsing progress markers based on historical page browsing information, and calculate the movement rate of the page browsing progress markers; If the movement speed is less than the average movement speed within a preset time period, locate the first browsing page; Determine the horizontal line where the virtual pointer of the first browsing page is located; Analyze the first line of sight information on the first browsing page to obtain the trajectory of the line of sight points on both sides of the horizontal line; Based on the trajectory of the line of sight, attempt to obtain the scanning features; If the attempt to acquire the data is successful, the partial page in the first browsing page is determined based on the scanning point trajectory and used as the second browsing page. If the attempt to obtain the data fails, determine the first viewing area above the horizontal line and the second viewing area below the horizontal line on the first browsing page based on the trajectory of the line of sight. Compare the density of line-of-sight points in the first and second viewing areas, and determine the first or second viewing area with a higher density of line-of-sight points as the third browsing page; The second or third browsing page is used as the fourth browsing page. Text recognition is performed on the fourth browsing page to determine the text recognition result. If the gaze duration of the recognized character corresponding to the text recognition result is greater than or equal to the preset duration threshold, the corresponding text recognition result will be used as big data for analysis. Step 3: Dynamically display device data and analysis results based on a smart dashboard; Based on device data and a pre-set virtual entity network construction template, construct a virtual entity network; Based on the analysis results and the correspondence between virtual entities in the virtual entity network, associate the virtual entities with the analysis results; Configure the trigger conditions for viewing the analysis results of virtual entity associations; Get the current operation information of the smart dashboard; The system dynamically displays information based on the trigger conditions and current operations on the smart dashboard.

Images