Equipment linkage control method and system of intelligent Internet of Things

Abstract

The invention relates to the technical field of Internet of Things control, and discloses an intelligent Internet of Things equipment linkage control method and system, and the method comprises the steps: collecting Internet of Things terminal data, converting the Internet of Things terminal data into a standardized space-time matrix, and carrying out the feature decoupling processing, and obtaining a high-dimensional feature vector; querying equipment information to calculate the energy transfer efficiency and the communication delay sensitivity of the equipment by using the high-dimensional feature vector so as to construct a dynamic knowledge graph of the Internet of Things equipment; carrying out fault signal injection on the dynamic knowledge graph, analyzing an abnormal propagation path of the equipment and the anti-interference capability of the equipment, and carrying out risk assessment on the Internet of Things equipment to obtain an equipment risk degree; an initial linkage strategy of the Internet of Things equipment is constructed, feasible strategies in the initial linkage strategy are identified, and a feasible strategy subset is obtained; and performing directed acyclic graph decomposition on the feasible strategy subset, performing dynamic resource allocation on decomposition task nodes, and executing a target task. According to the invention, the equipment linkage control collaboration of the intelligent Internet of Things can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things control, and provides a device linkage control method and system for an intelligent Internet of Things. Background Art

[0002] In today's era of rapid development of digitalization and intelligence, the application of device linkage control of the intelligent Internet of Things in various fields is becoming more and more extensive and critical. In many scenarios such as industrial production, smart home, and smart transportation, the interconnection and collaborative control between devices to achieve efficient and intelligent operation mode has become the core demand for improving productivity, optimizing quality of life, and improving traffic management. The device linkage control of the intelligent Internet of Things can integrate scattered devices into an organic whole, automatically perform various complex tasks based on preset rules and real-time data interaction, greatly improve the system's automation and response speed, and thus significantly improve the overall efficiency.

[0003] At present, the realization of device linkage control of intelligent Internet of Things mainly relies on traditional network architecture and control algorithms. Usually, wired or wireless networks are built to connect various sensors, actuators and other devices, and with the help of centralized or distributed control systems, devices are linked according to pre-set logic and instructions. For example, in smart homes, smart home appliances are connected through gateways, and users can issue instructions on their mobile phones to achieve device linkage. However, with the rapid growth in the number of IoT devices and the increasing complexity of application scenarios, traditional methods have exposed many drawbacks. On the one hand, when facing the access of massive devices, traditional network architectures are prone to network congestion and excessive delays, which seriously affect the real-time and stability of linkage control; on the other hand, fixed control algorithms are difficult to adapt to dynamically changing environments and diverse user needs, resulting in the inability of device linkage control of intelligent Internet of Things to flexibly and accurately achieve efficient collaboration between devices. Summary of the invention

[0004] The present invention provides a device linkage control method and system for an intelligent Internet of Things, the main purpose of which is to improve the coordination of device linkage control of the intelligent Internet of Things.

[0005] To achieve the above-mentioned purpose, the present invention provides a device linkage control method of the intelligent Internet of Things, comprising: Collecting IoT terminal data, converting the IoT terminal data into a standardized space-time matrix, and performing feature decoupling processing on the standardized space-time matrix to obtain a high-dimensional feature vector; Query the device information of the IoT device corresponding to the IoT terminal, calculate the energy transfer efficiency and communication delay sensitivity of the IoT device based on the device information and using the high-dimensional feature vector, obtain device state association information, and construct a dynamic knowledge graph of the IoT device based on the device state association information; Injecting a fault signal into the dynamic knowledge graph to obtain fault simulation information, analyzing the abnormal propagation path and the anti-interference capability of the IoT device based on the fault simulation information, and performing a risk assessment on the IoT device based on the abnormal propagation path and the anti-interference capability of the device to obtain a device risk degree; Query the current energy consumption and task information of the IoT terminal, construct an initial linkage strategy for the IoT device based on the device risk, the current energy consumption and the task information, perform digital twinning on the IoT device to obtain a simulated IoT device, and use the simulated IoT device to identify feasible strategies in the initial linkage strategy to obtain a subset of feasible strategies; The feasible strategy subset is decomposed into a directed acyclic graph to obtain decomposed task nodes, and resource information of the Internet of Things terminal is obtained to construct a resource matching matrix of the Internet of Things device. The resource matching matrix is ​​used to dynamically allocate resources to the decomposed task nodes to obtain a target task, and the target task is executed to achieve linkage control of the Internet of Things device.

[0006] Optionally, converting the IoT terminal data into a standardized space-time matrix includes: Performing multimodal data fusion on the IoT terminal data to obtain fused data; Performing spatiotemporal mapping on the fused data to obtain mapping data; Performing time-space calibration on the mapping data to obtain calibration data; Resampling the calibration data using a preset spatiotemporal resolution to obtain target data; A space-time matrix of the target data is constructed to obtain a standardized space-time matrix.

[0007] Optionally, performing feature decoupling processing on the standardized spatiotemporal matrix to obtain a high-dimensional feature vector includes: Performing complex domain conversion on the standardized space-time matrix to obtain a conversion matrix; Extracting matrix features of the conversion matrix, performing independent component decomposition on the matrix features, and obtaining decomposition features; Extracting device attribute features and environment attribute features related to the Internet of Things from the decomposed features to obtain refined features; Performing wavelet decomposition on the refined features to obtain wavelet decomposition features; Performing feature fusion on the wavelet decomposition features to obtain fused features; Calculating the mutual information gain of the fusion feature; The fused features are screened based on the mutual information gain to obtain a high-dimensional feature vector.

[0008] Optionally, the calculating the mutual information gain of the fusion feature includes: The mutual information gain of the fusion feature is calculated using the following formula: ; ; ; in, represents the mutual information gain, X represents the feature set of fusion features, represents the i-th feature of the fusion feature, and Y is expressed as The corresponding target variable is (Y) represents the entropy of Y, Indicates that the known features Under the condition of, the conditional entropy of Y is, represents the probability that the target variable Y takes the value y, Representation characteristics The probability of taking the value x, Indicated in Under the condition of x, Y The conditional probability of y.

[0009] Optionally, based on the device information, using the high-dimensional feature vector, calculating the energy transfer efficiency and communication delay sensitivity of the IoT device to obtain device state association information includes: Query the device transmit power and the device receive power in the device information; Based on the device transmission power and the device reception power, the energy transfer efficiency of the IoT device is calculated using the following formula: ; Where E represents the energy transfer efficiency, represents the device receiving power of the jth device in the IoT device, represents the device transmission power of the i-th device in the IoT device; The communication delay in the device information is queried, the communication load of the IoT device is analyzed using the high-dimensional feature vector, and based on the communication delay and the communication load, the communication delay sensitivity of the IoT device is calculated using the following formula: ; in, Indicates the sensitivity of communication delay represents the communication delay from IoT device i to IoT device j, represents the communication load between IoT devices i and j; Based on the energy transfer efficiency and the communication delay sensitivity, device state association information of the Internet of Things device is identified.

[0010] Optionally, constructing a dynamic knowledge graph of the IoT device based on the device state association information includes: Extracting entity information, relationship information and attribute information from the device status association information to obtain knowledge data; Performing entity alignment on the knowledge data to obtain entity alignment data; Performing knowledge merging on the entity alignment data to obtain knowledge merging data; The knowledge is combined with data to construct a dynamic knowledge graph of the IoT device.

[0011] Optionally, analyzing the abnormal propagation path and anti-interference capability of the IoT device based on the fault simulation information includes: Using the dynamic knowledge graph of the IoT device, a propagation model of the IoT device is constructed; Setting propagation attributes of the propagation model to obtain a target propagation model; Performing time series analysis on the fault simulation information to obtain an abnormal propagation path; Query the performance indicator data and recovery time of the IoT device during fault simulation; Performing indicator quantification processing on the performance indicator data and the recovery time to obtain a quantitative indicator; The quantitative index is used to analyze the device anti-interference capability of the Internet of Things device.

[0012] Optionally, constructing an initial linkage strategy for the IoT device based on the device risk, the current energy consumption and the task information includes: Identify the risk level of the equipment; Based on the risk level, build a risk mitigation strategy for IoT devices; Identify the risk type of the IoT device; Based on the risk type, construct a risk prevention and control strategy for the IoT device; Based on the current energy consumption, construct a resource allocation strategy for the networked device; Constructing a collaborative operation strategy for the IoT devices according to the task information; Based on the risk mitigation strategy, the risk prevention and control strategy, the resource allocation strategy and the collaborative operation strategy, an initial linkage strategy for the Internet of Things devices is constructed.

[0013] Optionally, performing directed acyclic graph decomposition on the subset of feasible strategies to obtain decomposition task nodes includes: Decomposing the feasible strategy subset into elements to obtain task node elements; Analyzing the task execution relationship of the task node elements; Based on the task execution relationship, construct a directed acyclic graph of the feasible strategy subset; Performing node merging optimization on the directed acyclic graph to obtain an optimized directed acyclic graph; The element nodes of the optimized directed acyclic graph are extracted to obtain decomposition task nodes.

[0014] In order to solve the above problems, the present invention also provides a device linkage control system for the intelligent Internet of Things, the system comprising: A data processing module is used to collect IoT terminal data, convert the IoT terminal data into a standardized space-time matrix, and perform feature decoupling processing on the standardized space-time matrix to obtain a high-dimensional feature vector; A dynamic knowledge graph construction module is used to query the device information of the IoT device corresponding to the IoT terminal, calculate the energy transfer efficiency and communication delay sensitivity of the IoT device based on the device information and the high-dimensional feature vector, obtain device state association information, and construct a dynamic knowledge graph of the IoT device based on the device state association information; The device anti-interference analysis module is used to inject fault signals into the dynamic knowledge graph to obtain fault simulation information, analyze the abnormal propagation path and device anti-interference capability of the IoT device based on the fault simulation information, and conduct risk assessment on the IoT device based on the abnormal propagation path and the device anti-interference capability to obtain the device risk level; A control strategy analysis module is used to query the current energy consumption and task information of the IoT terminal, construct an initial linkage strategy for the IoT device based on the device risk, the current energy consumption and the task information, perform digital twinning on the IoT device to obtain a simulated IoT device, and use the simulated IoT device to identify feasible strategies in the initial linkage strategy to obtain a subset of feasible strategies; The intelligent linkage control module is used to perform directed acyclic graph decomposition on the subset of feasible strategies to obtain decomposed task nodes, obtain resource information of the Internet of Things terminal to construct a resource matching matrix of the Internet of Things device, use the resource matching matrix to dynamically allocate resources to the decomposed task nodes, obtain the target task, and execute the target task to achieve linkage control of the Internet of Things device.

[0015] Compared with the problems described in the background technology, the embodiment of the present invention first provides a comprehensive and rich data foundation by collecting IoT terminal data, including multimodal sensor data such as temperature and pressure, and performs feature decoupling processing on the networked terminal data to improve data quality, thereby avoiding the misleading of subsequent analysis by interference factors, and making the analysis results more reliable; further, the present invention helps users to deeply understand the operating status and performance bottlenecks of the device through the steps of device status analysis and knowledge graph construction, provides a basis for subsequent optimization and management, and can clearly display the overall architecture and operating dynamics of the IoT device, thereby helping users to quickly understand the relationship and influence between devices; the present invention can understand the chain reaction of faults and potential problems in a virtual environment by performing fault simulation and risk assessment on IoT devices, so as to prevent faults in advance and reduce the overall risk level; further, the present invention reduces the time and cost of actual strategy attempts and improves the success rate of device linkage control by constructing an initial control strategy for IoT devices and performing simulation verification; further, the present invention ensures that the task is executed in a suitable resource environment by performing task decomposition and resource allocation on the initial linkage strategy, improves the efficiency and quality of task execution, and implements device linkage control after the task resource configuration is completed, thereby improving the synergy of device linkage control of the intelligent IoT. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a flow chart of a device linkage control method of a smart Internet of Things provided by an embodiment of the present invention; Figure 2 A schematic diagram of a module for implementing a device linkage control method of the smart Internet of Things provided by an embodiment of the present invention.

[0017] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0018] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0019] The embodiment of the present application provides a device linkage control method for an intelligent Internet of Things. The execution subject of the device linkage control method for the intelligent Internet of Things includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the device linkage control method for the intelligent Internet of Things can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0020] Embodiment 1: Reference Figure 1FIG. 1 is a flow chart of a device linkage control method of an intelligent Internet of Things provided by an embodiment of the present invention. In this embodiment, the device linkage control method of the intelligent Internet of Things includes: S1. Collect IoT terminal data, convert the IoT terminal data into a standardized space-time matrix, perform feature decoupling processing on the standardized space-time matrix, and obtain a high-dimensional feature vector.

[0021] The embodiments of the present invention can provide a comprehensive and rich data basis for subsequent data analysis and processing by collecting IoT terminal data.

[0022] Among them, the IoT terminal data refers to data from multimodal sensors such as temperature, pressure, motion trajectory, energy consumption and communication signal strength, which can be obtained through edge computing nodes deployed on IoT terminals.

[0023] Furthermore, the embodiment of the present invention can eliminate the spatiotemporal deviation between devices by converting the IoT terminal data into a standardized spatiotemporal matrix, so that the data of different devices can be compared and analyzed under the same spatiotemporal reference.

[0024] The standardized space-time matrix refers to a matrix used to represent the distribution and characteristics of data in the time and space dimensions in the field of the Internet of Things.

[0025] As an embodiment of the present invention, the converting of the IoT terminal data into a standardized space-time matrix includes: performing multimodal data fusion on the IoT terminal data to obtain fused data, performing space-time mapping on the fused data to obtain mapping data, performing space-time calibration on the mapping data to obtain calibration data, resampling the calibration data using a preset space-time resolution to obtain target data, constructing a space-time matrix of the target data, and obtaining a standardized space-time matrix.

[0026] The preset spatiotemporal resolution refers to the precision of sampling data in the time and space dimensions set during the process of converting IoT terminal data into a standardized spatiotemporal matrix.

[0027] Optionally, the fused data can be obtained by fusing IoT terminal data from different types of sensors such as temperature, humidity, and position using a multimodal fusion model based on deep learning. The mapping data can be obtained by mapping the fused data using a coordinate transformation algorithm. The calibration data can be obtained by time-calibrating the mapping data using high-precision atomic clock synchronization technology, and then spatially calibrating the time-calibrated mapping data using ultrasonic ranging technology. The target data can be obtained by resampling using linear interpolation based on a preset spatiotemporal resolution. The standardized spatiotemporal matrix can be obtained by constructing an all-zero matrix, and then filling the corresponding values ​​of the target data into the all-zero matrix using the time of the target data as the row index and the spatial position as the column index.

[0028] Furthermore, the embodiment of the present invention can remove data noise by performing feature decoupling processing on the standardized spatiotemporal matrix to obtain a high-dimensional feature vector, thereby improving data quality and avoiding misleading subsequent analysis by interference factors, making the analysis result more reliable.

[0029] As an embodiment of the present invention, the feature decoupling processing of the standardized space-time matrix to obtain a high-dimensional feature vector includes: performing complex domain conversion on the standardized space-time matrix to obtain a transformation matrix, extracting matrix features of the transformation matrix, performing independent component decomposition on the matrix features to obtain decomposition features, extracting device attribute features and environmental attribute features about the Internet of Things from the decomposition features to obtain refined features, performing wavelet decomposition on the refined features to obtain wavelet decomposition features, performing feature fusion on the wavelet decomposition features to obtain fused features, calculating the mutual information gain of the fused features, and performing feature screening on the fused features based on the mutual information gain to obtain a high-dimensional feature vector.

[0030] Among them, the device attribute feature refers to the feature that can reflect the inherent characteristics of the IoT device itself, such as the device model, hardware parameters and other features. The environmental attribute feature refers to the feature that reflects the external environment of the device, such as temperature, humidity, etc. The mutual information gain refers to the degree to which the uncertainty of the target information is reduced after a certain feature is known. For example, in the IoT scenario, it is a correlation measure between the fusion feature and the target information such as the device status (normal / faulty, etc.). If a feature is highly correlated with the device failure, the uncertainty of the device failure status is greatly reduced after knowing the feature value, and the mutual information gain of this feature is high, which can be used to judge the value of the feature to the analysis of the target information. The high-dimensional feature vector refers to the key feature composition selected from the original standardized spatiotemporal matrix after a series of complex processing.

[0031] Optionally, the conversion matrix can be obtained by converting the standardized space-time matrix from the real domain to the complex domain using Fourier transform technology. The matrix features can be obtained by extracting key features such as eigenvalues ​​and eigenvectors of the conversion matrix using matrix analysis methods, such as singular value decomposition. The decomposed features can be obtained by reducing the dimension of the matrix features using principal component analysis and then removing redundant information. The refined features can be obtained by screening out device attribute features (such as device model, hardware parameter related features) and environmental attribute features (such as temperature, humidity and other environmental factors related features) from the reduced dimensional features according to the characteristics of the IoT devices and the environment. The wavelet decomposition features can be obtained by decomposing the refined features at different frequency scales using wavelet transform technology. The fusion features can be obtained by fusing the wavelet decomposition features using methods such as splicing or weighted summation. The high-dimensional feature vector can be obtained by screening out features with gains higher than the threshold according to a preset mutual information gain threshold, and the preset mutual information gain threshold can be set to 0.1, which needs to be set in combination with actual applications.

[0032] Further, as another optional embodiment of the present invention, the calculating the mutual information gain of the fused features includes: calculating the mutual information gain of the fused features using the following formula: ; ; ; in, represents the mutual information gain, X represents the feature set of fusion features, represents the i-th feature of the fusion feature, and Y is expressed as The corresponding target variable is (Y) represents the entropy of Y, Indicates that the known features Under the condition of, the conditional entropy of Y is, represents the probability that the target variable Y takes the value y, Representation characteristics The probability of taking the value x, Indicated in Under the condition of x, Y The conditional probability of y.

[0033] It should be further explained that the core principle of the above mutual information gain calculation formula is based on information theory, which aims to quantify the contribution of fused features to the target variable information and reflect the dependency between features and target variables from a statistical perspective. By calculating mutual information gain through this formula, the most explanatory features for the target variable can be effectively screened out. In IoT device analysis scenarios, such as equipment failure prediction tasks, those fused features that are closely related to the equipment failure status can be accurately found, and a large number of fused features can be sorted according to their contribution to the target variable, which helps to remove redundant features and highlight key features, providing strong support for building efficient equipment analysis models and improving the accuracy of analysis and prediction.

[0034] S2. Query the device information of the IoT device corresponding to the IoT terminal, calculate the energy transfer efficiency and communication delay sensitivity of the IoT device based on the device information and using the high-dimensional feature vector to obtain device state association information, and construct a dynamic knowledge graph of the IoT device based on the device state association information.

[0035] The embodiment of the present invention can provide necessary data support for a comprehensive understanding of the Internet of Things devices by querying the device information of the Internet of Things device corresponding to the Internet of Things terminal.

[0036] The device information refers to various information including the model, specifications, functional characteristics and communication protocol of the device.

[0037] Optionally, the device information can be obtained through a commonly used IoT communication protocol, such as MQTT.

[0038] Furthermore, the embodiment of the present invention calculates the energy transfer efficiency and communication delay sensitivity of the IoT device based on the device information and utilizes the high-dimensional feature vector to obtain device status association information, which can obtain important indicator data for measuring the performance of networked devices, thereby helping users to gain an in-depth understanding of the operating status and performance bottlenecks of the device, providing a basis for subsequent optimization and management.

[0039] Among them, the energy transfer efficiency refers to the effective utilization of the device in the energy input and output process during the operation of the IoT device, and the communication delay sensitivity refers to the sensitivity of the IoT device to the delay in the communication process.

[0040] As an embodiment of the present invention, based on the device information, using the high-dimensional feature vector, calculating the energy transfer efficiency and communication delay sensitivity of the IoT device to obtain device state association information includes: querying the device transmit power and the device receive power in the device information, and based on the device transmit power and the device receive power, calculating the energy transfer efficiency of the IoT device using the following formula: ; Where E represents the energy transfer efficiency, represents the device receiving power of the jth device in the IoT device, represents the device transmission power of the i-th device in the IoT device; The communication delay in the device information is queried, the communication load of the IoT device is analyzed using the high-dimensional feature vector, and based on the communication delay and the communication load, the communication delay sensitivity of the IoT device is calculated using the following formula: ; in, Indicates the sensitivity of communication delay represents the communication delay from IoT device i to IoT device j, represents the communication load between IoT devices i and j; Based on the energy transfer efficiency and the communication delay sensitivity, device state association information of the Internet of Things device is identified.

[0041] Furthermore, the embodiment of the present invention can clearly display the overall architecture and operation dynamics of the IoT device by constructing the dynamic knowledge graph of the IoT device based on the device status association information, thereby helping users quickly understand the relationships and impacts between devices.

[0042] Among them, the dynamic knowledge graph refers to a structured knowledge representation form that can reflect the changes in the status and relationships of IoT devices over time in real time. It organizes information in a graph manner. Nodes represent various entities in the IoT, such as devices, environmental factors, and users, and edges represent the relationships between entities, such as communication connections between devices, energy transfer relationships, and the influence relationships between devices and environmental factors.

[0043] As an embodiment of the present invention, the dynamic knowledge graph of the Internet of Things device is constructed based on the device state association information, including: extracting entity information, relationship information and attribute information in the device state association information to obtain knowledge data, performing entity alignment on the knowledge data to obtain entity alignment data, performing knowledge merging on the entity alignment data to obtain knowledge merging data, and using the knowledge merging data to construct the dynamic knowledge graph of the Internet of Things device.

[0044] Among them, the entity information refers to the unique identification of the device, such as device ID, serial number, etc. The relationship information refers to the connection between IoT device entities, such as communication relationship (such as the existence of a data transmission link between device A and device B) and the impact of ambient temperature changes on the operating status of the device. The attribute information refers to the characteristics used to describe entities and relationships in detail, such as the computing power, storage capacity and communication bandwidth of the device. The dynamic knowledge graph refers to a structured knowledge representation that can reflect the changes in the status of IoT devices and their relationships over time in real time.

[0045] The optional knowledge data uses the text extraction technology in natural language processing to parse the device status association information. From the data such as energy transfer efficiency and communication delay sensitivity, the device entities, the relationships between devices (such as communication relationships, energy transfer relationships) and the attributes of devices and relationships (such as device models, energy transfer efficiency values) are identified. The entity alignment data can be obtained by using a method based on rule matching and machine learning, such as the cosine similarity algorithm combined with clustering technology, comparing the device entities in the knowledge data from different sources, and unifying the different expressions referring to the same device, such as identifying "device A" and "device numbered 001 produced by a certain factory" as the same entity. The knowledge merging data can use graph database related technologies, such as the merge operation of Neo4j. The data after entity alignment is integrated, the relationships and attributes of the same entities are summarized, duplicate information is removed, and the scattered knowledge is integrated into a complete knowledge system. The dynamic knowledge graph can be based on the knowledge merging data and use a knowledge graph construction tool (such as Stardog). By taking device entities as nodes, relationships as edges, and attributes as descriptions of nodes and edges, a knowledge graph reflecting the status of IoT devices and their relationships is constructed.

[0046] S3. Inject fault signals into the dynamic knowledge graph to obtain fault simulation information. Based on the fault simulation information, analyze the abnormal propagation path and the anti-interference capability of the IoT device. Based on the abnormal propagation path and the anti-interference capability of the device, conduct a risk assessment on the IoT device to obtain the device risk level.

[0047] The embodiment of the present invention injects fault signals into the dynamic knowledge graph to obtain fault simulation information, which can comprehensively and systematically understand the chain reactions that may be caused by various faults in IoT devices in a virtual environment and discover potential problems and hidden dangers in advance.

[0048] The fault simulation information can refer to historical fault data to analyze the patterns and rules of fault occurrence. Through data mining algorithms, such as association rule mining, the fault characteristics and related equipment status that often occur are found. Then, in the dynamic knowledge graph, the simulation information is injected into the corresponding entities (equipment, environment, etc.) and relationships according to these rules.

[0049] The embodiment of the present invention can deeply understand the vulnerability and fault propagation law of the Internet of Things system by analyzing the abnormal propagation path and the anti-interference ability of the Internet of Things device based on the fault simulation information, so as to help users focus on the device when performing Internet of Things linkage control to reduce the failure probability of the device.

[0050] Among them, the abnormal propagation path refers to the diffusion path when a device fails or is in an abnormal state in a complex network composed of IoT devices, and the abnormality spreads to other related devices along the connections and relationships between devices. The device anti-interference capability refers to the ability to maintain its normal operation, ensure stable performance and accurately perform tasks in the face of interference such as aging of the device's own hardware, software vulnerabilities, electromagnetic interference, changes in ambient temperature and humidity, and network attacks.

[0051] As an embodiment of the present invention, the abnormal propagation path and device anti-interference capability of the Internet of Things device are analyzed based on the fault simulation information, including: using the dynamic knowledge graph of the Internet of Things device to construct a propagation model of the Internet of Things device, setting propagation attributes for the propagation model to obtain a target propagation model, performing time series analysis on the fault simulation information to obtain the abnormal propagation path, querying the performance indicator data and recovery time of the Internet of Things device during fault simulation, quantifying the performance indicator data and the recovery time to obtain quantitative indicators, and using the quantitative indicators to analyze the device anti-interference capability of the Internet of Things device.

[0052] Among them, the propagation model refers to a model based on a graph structure, which is used to simulate the process of abnormal propagation between IoT devices. The propagation attribute refers to the characteristic description of the connection edge in the propagation model, such as propagation probability and propagation delay. The performance indicator data refers to various types of data used to measure the operating status and performance of the device during the IoT device fault simulation, such as data transmission rate, packet loss rate, etc.

[0053] Optionally, the propagation model can be based on the dynamic knowledge graph of IoT devices, with device nodes in the graph as model nodes, and relationships between devices as connecting edges, and a graph theory algorithm, such as the Dijkstra algorithm, can be used to construct a propagation model. The target propagation model can be obtained by setting propagation attributes, such as propagation probability and propagation delay, on the edges in the propagation model based on the fault simulation scenario. The abnormal propagation path can be determined by using the ARIMA model, according to the chronological order of the faults, combined with the propagation rules of the target propagation model, and analyzing the propagation trajectory of the fault signal between devices over time. The quantitative index can be obtained by quantifying the performance index data and the recovery time using normalization technology. The anti-interference ability of the device can be evaluated by using a linear regression model based on the quantitative index to construct an evaluation model, using the quantitative index as input to analyze the degree of performance impact and recovery ability of the device when facing fault interference, and then evaluating it.

[0054] The embodiment of the present invention performs risk assessment on the IoT device based on the abnormal propagation path and the anti-interference capability of the device, and obtains the device risk level, which can determine the devices with higher risks in the system, focus on monitoring, maintaining and upgrading these devices, and prevent the occurrence of failures in advance, so as to reduce the overall risk level of system equipment linkage control.

[0055] The device risk refers to a numerical indicator that comprehensively and quantitatively evaluates the possibility of failure of IoT devices in the system and the degree of impact of the failure.

[0056] Optionally, the equipment risk level is first determined based on the abnormal propagation path to determine the impact range and propagation speed caused by the equipment failure, such as how many production line links are affected by the failure of key equipment. Then, the equipment anti-interference ability is combined, such as the performance fluctuation of the equipment in an electromagnetic interference environment. The two factors are weighted to obtain the risk level.

[0057] S4. Query the current energy consumption and task information of the IoT terminal, construct an initial linkage strategy for the IoT device based on the device risk, the current energy consumption and the task information, perform digital twinning on the IoT device to obtain a simulated IoT device, use the simulated IoT device to identify feasible strategies in the initial linkage strategy, and obtain a subset of feasible strategies.

[0058] The embodiment of the present invention can help users evaluate the operating cost and energy efficiency of the device and understand whether the device is in an abnormal state of high energy consumption by querying the current energy consumption and task information of the Internet of Things terminal.

[0059] Among them, the current energy consumption refers to the energy value consumed by the IoT terminal during the current operation process, which can be obtained through the device's own energy consumption monitoring module or related sensors, such as the power consumption data recorded in real time by a smart meter. The task information refers to the details of the task that the IoT terminal is executing or is about to execute, such as the task type (data collection, transmission, analysis, etc.), task priority and estimated completion time, which can be obtained by querying the device task scheduling system.

[0060] The embodiment of the present invention can balance device risk, energy consumption and task requirements by constructing the initial linkage strategy of the Internet of Things device based on the device risk, the current energy consumption and the task information, and strive to ensure the completion of the task while improving the overall reliability and energy utilization efficiency of the device, reducing the system operating cost and the probability of failure, and improving the comprehensive performance of the Internet of Things system.

[0061] As an embodiment of the present invention, the initial linkage strategy of the Internet of Things device is constructed based on the device risk degree, the current energy consumption and the task information, including: identifying the risk level of the device risk degree, constructing a risk mitigation strategy for the Internet of Things device based on the risk level, identifying the risk type of the Internet of Things device, constructing a risk prevention and control strategy for the Internet of Things device based on the risk type, constructing a resource allocation strategy for the networked device based on the current energy consumption, constructing a collaborative operation strategy for the Internet of Things device according to the task information, and constructing the initial linkage strategy for the Internet of Things device based on the risk mitigation strategy, the risk prevention and control strategy, the resource allocation strategy and the collaborative operation strategy.

[0062] Among them, the risk level refers to different levels divided according to the risk level of the equipment, which is used to intuitively reflect the degree of risk of equipment failure. The risk mitigation strategy refers to the strategy formulated for reducing the impact of failures for high-risk level equipment. The risk type refers to the different categories of causes that cause IoT devices to fail or operate abnormally, such as hardware failure, network failure, and environmental factors. The risk prevention and control strategy refers to the prevention and response strategy formulated for different risk types. For example, for hardware failure risks, a regular hardware inspection and replacement mechanism can be established, and high-quality hardware equipment can be used to improve reliability. For software failures, software patches can be updated in a timely manner and vulnerability scans and repairs can be performed. The deployment strategy refers to a strategy for the reasonable allocation of energy, computing resources, etc. based on the current energy consumption and task priority of IoT devices. The collaborative operation strategy refers to a strategy for coordinating the orderly execution of tasks by various devices based on IoT device task information.

[0063] Optionally, the risk level can be divided into risk levels such as low, medium and high by setting a risk threshold interval. For high-risk level equipment, redundant backup technology can be used to prepare spare equipment. When the main equipment is at high risk, it can be switched in time to reduce the impact of the failure and achieve risk mitigation. The risk type can be identified by using data mining algorithms based on fault simulation information and historical data, such as hardware failure, network attack, etc. Then, according to the risk type, such as for network attack risk, firewall, encrypted communication and other technologies are used to build a prevention and control strategy. The resource allocation strategy can be constructed by analyzing the current energy consumption data. If the energy consumption is too high, dynamic voltage and frequency adjustment technology can be used to reduce the operating frequency of the equipment to reduce energy consumption. At the same time, energy resources can be allocated according to the priority of the equipment. The collaborative operation strategy can analyze the task information, clarify the task process and dependency, and then use the task scheduling algorithm, such as priority scheduling, to arrange the order of equipment execution tasks and coordinate the collaborative operation between equipment. The initial linkage strategy can be formed by system integration technology so that each strategy cooperates with each other.

[0064] The embodiment of the present invention performs digital twinning on the IoT device to obtain a simulated IoT device, which can provide a safe and controllable virtual environment for testing and verifying the initial linkage strategy.

[0065] Optionally, the digital twinning of the IoT device to obtain a simulated IoT device can be implemented through multi-physics field simulation software, such as COMSOL Multiphysics.

[0066] The embodiment of the present invention utilizes the simulated Internet of Things device to identify feasible strategies in the initial linkage strategy, and obtains a feasible strategy subset, which can quickly find practical solutions from a large number of initially designed linkage strategies, reduce the time and cost of strategy attempts on actual Internet of Things devices, and thus improve the success rate of strategy implementation.

[0067] Optionally, the feasible strategy subset can use optimization algorithms such as genetic algorithms and particle swarm optimization algorithms. Encode each strategy in the initial linkage strategy as an individual in the algorithm, and use device performance, energy consumption, task completion rate, etc. as evaluation indicators of the fitness function. Through algorithm iterative optimization, the simulated IoT device can find the feasible strategy that can optimize the fitness function value, that is, the one that best meets the actual needs, through continuous attempts.

[0068] S5. Decompose the feasible strategy subset into a directed acyclic graph to obtain decomposed task nodes, obtain resource information of the Internet of Things terminal to construct a resource matching matrix of the Internet of Things device, dynamically allocate resources to the decomposed task nodes using the resource matching matrix to obtain a target task, and execute the target task to achieve linkage control of the Internet of Things device.

[0069] The embodiment of the present invention performs directed acyclic graph decomposition on the feasible strategy subset to obtain decomposed task nodes. When there are many IoT devices and complex tasks, task dependencies can be clarified to avoid task execution conflicts and improve execution efficiency.

[0070] As an embodiment of the present invention, the directed acyclic graph decomposition of the feasible strategy subset to obtain decomposed task nodes includes: element decomposition of the feasible strategy subset to obtain task node elements, analyzing the task execution relationship of the task node elements, constructing a directed acyclic graph of the feasible strategy subset based on the task execution relationship, performing node merging optimization on the directed acyclic graph to obtain an optimized directed acyclic graph, extracting element nodes of the optimized directed acyclic graph to obtain decomposed task nodes.

[0071] Among them, the task execution relationship refers to a method for describing the temporal and logical order and dependency relationship between task node elements, and the directed acyclic graph refers to a special graph structure consisting of vertices (i.e., task node elements) and directed edges. In IoT device strategy analysis, vertices represent independent task units, such as device A performing an operation and device B performing data processing. Directed edges reflect the task execution relationship, and their direction indicates the order of tasks, from the first executed task node to the later executed task node. For example, if device A must complete the task before device B can start, a directed edge is drawn from the task node of device A to the task node of device B.

[0072] Optionally, the task node element can treat each strategy in the subset of feasible strategies as a whole, and use text parsing technology to split the strategy content according to the task execution steps. For example, if the strategy contains a collaborative workflow between devices, it can be split into task nodes such as device A executing tasks and device B receiving data. The directed acyclic graph can be constructed by analyzing the execution order and dependencies between task node elements, determining the direction of directed edges, and then using a graph construction algorithm with task node elements as vertices and task execution relationships as directed edges. The optimized directed acyclic graph can merge multiple task node elements with the same predecessor and successor nodes in the directed acyclic graph using a graph optimization algorithm. For example, if two consecutive tasks of device c have the same impact on subsequent tasks, they can be merged into one node to simplify the graph structure.

[0073] The embodiment of the present invention obtains the resource information of the IoT terminal to construct the resource matching matrix of the IoT device, so as to clearly understand the resource status of each device and the resource requirement of each task, thereby achieving accurate matching of resources and tasks.

[0074] The resource matching matrix refers to a two-dimensional matrix used to intuitively present the corresponding relationship between the resource status of IoT terminal devices and the task resource requirements, which can be constructed through the NumPy library in Python.

[0075] Furthermore, the embodiment of the present invention uses the resource matching matrix to dynamically allocate resources to the decomposed task nodes to obtain the target task, which can ensure that each task can be executed in the most suitable resource environment, thereby improving the efficiency and quality of task execution.

[0076] Optionally, the target task can first clearly decompose the resource requirements of the task nodes, and through the resource matching matrix, compare the resource status of each device, and reasonably allocate the device resources to the task nodes according to the task priority and resource margin. Then, the task execution process can be monitored in real time. If resources are insufficient or the task priority changes, the resource allocation can be readjusted with the help of the matrix until the task is completed.

[0077] Furthermore, the embodiments of the present invention can convert the pre-formulated strategies, planned tasks and allocated resources into actual device actions by executing the target tasks to realize the linkage control of the IoT devices, thereby realizing efficient collaborative control of the IoT devices.

[0078] Embodiment 2: like Figure 2 The figure shows a functional module diagram of the device linkage control system of the intelligent Internet of Things of the present invention.

[0079] The device linkage control system 200 of the intelligent Internet of Things of the present invention can be installed in an electronic device. According to the functions to be implemented, the device linkage control system of the intelligent Internet of Things may include a data processing module 201, a dynamic knowledge graph construction module 202, a device anti-interference analysis module 203, a control strategy analysis module 204 and an intelligent linkage control module 205. The module of the present invention may also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, and is stored in the memory of the electronic device.

[0080] In the embodiment of the present invention, the functions of each module / unit are as follows: The data processing module 201 is used to collect IoT terminal data, convert the IoT terminal data into a standardized space-time matrix, and perform feature decoupling processing on the standardized space-time matrix to obtain a high-dimensional feature vector; The dynamic knowledge graph construction module 202 is used to query the device information of the IoT device corresponding to the IoT terminal, calculate the energy transfer efficiency and communication delay sensitivity of the IoT device based on the device information and the high-dimensional feature vector, obtain device state association information, and construct the dynamic knowledge graph of the IoT device based on the device state association information; The device anti-interference analysis module 203 is used to inject fault signals into the dynamic knowledge graph to obtain fault simulation information, analyze the abnormal propagation path and device anti-interference capability of the IoT device based on the fault simulation information, and conduct risk assessment on the IoT device based on the abnormal propagation path and the device anti-interference capability to obtain the device risk level; The control strategy analysis module 204 is used to query the current energy consumption and task information of the IoT terminal, build the initial linkage strategy of the IoT device based on the device risk, the current energy consumption and the task information, perform digital twinning on the IoT device to obtain a simulated IoT device, and use the simulated IoT device to identify feasible strategies in the initial linkage strategy to obtain a subset of feasible strategies; The intelligent linkage control module 205 is used to perform directed acyclic graph decomposition on the feasible strategy subset to obtain decomposed task nodes, obtain resource information of the Internet of Things terminal to construct a resource matching matrix of the Internet of Things device, use the resource matching matrix to dynamically allocate resources to the decomposed task nodes, obtain the target task, and execute the target task to achieve linkage control of the Internet of Things device.

[0081] In detail, each module in the device linkage control system 200 of the smart Internet of Things in the embodiment of the present invention is used in the same manner as described above. Figure 1 The same technical means are used as the device linkage control method of the smart Internet of Things described in, and can produce the same technical effects, so they will not be repeated here.

[0082] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution 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 solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A device linkage control method for the intelligent Internet of Things, characterized in that: The method comprises: Collecting IoT terminal data, converting the IoT terminal data into a standardized space-time matrix, and performing feature decoupling processing on the standardized space-time matrix to obtain a high-dimensional feature vector; Query the device information of the IoT device corresponding to the IoT terminal, calculate the energy transfer efficiency and communication delay sensitivity of the IoT device based on the device information and using the high-dimensional feature vector, obtain device state association information, and construct a dynamic knowledge graph of the IoT device based on the device state association information; Injecting a fault signal into the dynamic knowledge graph to obtain fault simulation information, analyzing the abnormal propagation path and the anti-interference capability of the IoT device based on the fault simulation information, and performing a risk assessment on the IoT device based on the abnormal propagation path and the anti-interference capability of the device to obtain a device risk degree; Query the current energy consumption and task information of the IoT terminal, construct an initial linkage strategy for the IoT device based on the device risk, the current energy consumption and the task information, perform digital twinning on the IoT device to obtain a simulated IoT device, and use the simulated IoT device to identify feasible strategies in the initial linkage strategy to obtain a subset of feasible strategies; The feasible strategy subset is decomposed into a directed acyclic graph to obtain decomposed task nodes, and resource information of the Internet of Things terminal is obtained to construct a resource matching matrix of the Internet of Things device. The resource matching matrix is ​​used to dynamically allocate resources to the decomposed task nodes to obtain a target task, and the target task is executed to achieve linkage control of the Internet of Things device.

2. The device linkage control method of the intelligent Internet of Things according to claim 1, characterized in that: The converting the IoT terminal data into a standardized space-time matrix includes: Performing multimodal data fusion on the IoT terminal data to obtain fused data; Performing spatiotemporal mapping on the fused data to obtain mapping data; Performing time-space calibration on the mapping data to obtain calibration data; Resampling the calibration data using a preset spatiotemporal resolution to obtain target data; A space-time matrix of the target data is constructed to obtain a standardized space-time matrix.

3. The device linkage control method of the intelligent Internet of Things according to claim 1, characterized in that: The step of performing feature decoupling processing on the standardized spatiotemporal matrix to obtain a high-dimensional feature vector includes: Performing complex domain conversion on the standardized space-time matrix to obtain a conversion matrix; Extracting matrix features of the conversion matrix, performing independent component decomposition on the matrix features, and obtaining decomposition features; Extracting device attribute features and environment attribute features related to the Internet of Things from the decomposed features to obtain refined features; Performing wavelet decomposition on the refined features to obtain wavelet decomposition features; Performing feature fusion on the wavelet decomposition features to obtain fused features; Calculating the mutual information gain of the fusion feature; The fused features are screened based on the mutual information gain to obtain a high-dimensional feature vector.

4. The device linkage control method of the intelligent Internet of Things as claimed in claim 3, characterized in that: The calculating the mutual information gain of the fusion feature includes: The mutual information gain of the fusion feature is calculated using the following formula: ； ； ； in, represents the mutual information gain, X represents the feature set of fusion features, represents the i-th feature of the fusion feature, and Y is expressed as The corresponding target variable is (Y) represents the entropy of Y, Indicates that the known features Under the condition of, the conditional entropy of Y is, represents the probability that the target variable Y takes the value y, Representation characteristics The probability of taking the value x, Indicated in Under the condition of x, Y The conditional probability of y.

5. The device linkage control method of the intelligent Internet of Things according to claim 1, characterized in that: The method of calculating the energy transfer efficiency and communication delay sensitivity of the IoT device based on the device information and using the high-dimensional feature vector to obtain device state association information includes: Query the device transmit power and the device receive power in the device information; Based on the device transmission power and the device receiving power, the energy transfer efficiency of the IoT device is calculated using the following formula: ； Where E represents the energy transfer efficiency, represents the device receiving power of the jth device in the IoT device, represents the device transmission power of the i-th device in the IoT device; The communication delay in the device information is queried, the communication load of the IoT device is analyzed using the high-dimensional feature vector, and based on the communication delay and the communication load, the communication delay sensitivity of the IoT device is calculated using the following formula: ； in, represents the communication delay sensitivity, represents the communication delay from IoT device i to IoT device j, represents the communication load between IoT devices i and j; Based on the energy transfer efficiency and the communication delay sensitivity, device state association information of the Internet of Things device is identified.

6. The device linkage control method of the intelligent Internet of Things according to claim 1, characterized in that: The step of constructing a dynamic knowledge graph of the IoT device based on the device state association information includes: Extracting entity information, relationship information and attribute information from the device status association information to obtain knowledge data; Performing entity alignment on the knowledge data to obtain entity alignment data; Performing knowledge merging on the entity alignment data to obtain knowledge merging data; The knowledge is combined with data to construct a dynamic knowledge graph of the IoT device.

7. The device linkage control method of the intelligent Internet of Things according to claim 1, characterized in that: The analyzing the abnormal propagation path and the anti-interference capability of the IoT device based on the fault simulation information includes: Using the dynamic knowledge graph of the IoT device, a propagation model of the IoT device is constructed; Setting propagation attributes of the propagation model to obtain a target propagation model; Performing time series analysis on the fault simulation information to obtain an abnormal propagation path; Query the performance indicator data and recovery time of the IoT device during fault simulation; Performing indicator quantification processing on the performance indicator data and the recovery time to obtain a quantitative indicator; The quantitative index is used to analyze the device anti-interference capability of the Internet of Things device.

8. The device linkage control method of the intelligent Internet of Things according to claim 1, characterized in that: The constructing the initial linkage strategy of the IoT device based on the device risk, the current energy consumption and the task information includes: Identify the risk level of the equipment; Based on the risk level, build a risk mitigation strategy for IoT devices; Identify the risk type of the IoT device; Based on the risk type, construct a risk prevention and control strategy for the IoT device; Based on the current energy consumption, construct a resource allocation strategy for the networked device; Constructing a collaborative operation strategy for the IoT devices according to the task information; Based on the risk mitigation strategy, the risk prevention and control strategy, the resource allocation strategy and the collaborative operation strategy, an initial linkage strategy for the Internet of Things devices is constructed.

9. The device linkage control method of the intelligent Internet of Things according to claim 1, characterized in that: Decomposing the feasible strategy subset into a directed acyclic graph to obtain decomposed task nodes includes: Decomposing the feasible strategy subset into elements to obtain task node elements; Analyzing the task execution relationship of the task node elements; Based on the task execution relationship, construct a directed acyclic graph of the feasible strategy subset; Performing node merging optimization on the directed acyclic graph to obtain an optimized directed acyclic graph; The element nodes of the optimized directed acyclic graph are extracted to obtain decomposition task nodes.

10. The equipment linkage control system of the intelligent Internet of Things is characterized by: The system comprises: A data processing module is used to collect IoT terminal data, convert the IoT terminal data into a standardized space-time matrix, and perform feature decoupling processing on the standardized space-time matrix to obtain a high-dimensional feature vector; A dynamic knowledge graph construction module is used to query the device information of the IoT device corresponding to the IoT terminal, calculate the energy transfer efficiency and communication delay sensitivity of the IoT device based on the device information and the high-dimensional feature vector, obtain device state association information, and construct a dynamic knowledge graph of the IoT device based on the device state association information; The device anti-interference analysis module is used to inject fault signals into the dynamic knowledge graph to obtain fault simulation information, analyze the abnormal propagation path and device anti-interference capability of the IoT device based on the fault simulation information, and conduct risk assessment on the IoT device based on the abnormal propagation path and the device anti-interference capability to obtain the device risk level; A control strategy analysis module is used to query the current energy consumption and task information of the IoT terminal, construct an initial linkage strategy for the IoT device based on the device risk, the current energy consumption and the task information, perform digital twinning on the IoT device to obtain a simulated IoT device, and use the simulated IoT device to identify feasible strategies in the initial linkage strategy to obtain a subset of feasible strategies; The intelligent linkage control module is used to perform directed acyclic graph decomposition on the subset of feasible strategies to obtain decomposed task nodes, obtain resource information of the Internet of Things terminal to construct a resource matching matrix of the Internet of Things device, use the resource matching matrix to dynamically allocate resources to the decomposed task nodes, obtain the target task, and execute the target task to achieve linkage control of the Internet of Things device.

Images