Multi-circuit breaker monitoring method based on AI algorithm and Internet of Things management system

Abstract

The invention discloses a multi-circuit breaker monitoring method based on an AI algorithm and an Internet of Things management system. The method comprises the following steps: acquiring sensing data of a plurality of circuit breakers in a target area; establishing a circuit breaker data network based on a graph network algorithm according to the sensing data of the plurality of circuit breakers; inputting the circuit breaker data network into a pre-trained anomaly prediction AI model to obtain an anomaly prediction result corresponding to each circuit breaker; and in response to a monitoring demand for any one of the circuit breakers, determining an abnormal risk of the circuit breaker according to the abnormal prediction result corresponding to the circuit breaker and at least one of the other circuit breakers. Therefore, accurate abnormal risk assessment based on graph network and multi-circuit breaker association can be realized, the accuracy and reliability of monitoring of the Internet of Things system in the target area are improved, and the fault risk of the circuit breakers is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a multi - circuit breaker monitoring method and an Internet of Things management system based on an AI algorithm. Background Art

[0002] With the continuous growth of the intelligent monitoring requirements of Internet of Things systems, Internet of Things enterprises have increasingly attached importance to reducing the circuit breaker failure risk through accurate anomaly detection to ensure the stability of Internet of Things systems. Existing technologies usually collect the sensing data of a single circuit breaker, use simple threshold analysis or traditional statistical models to evaluate the device status, and formulate maintenance strategies based on independent monitoring results to ensure the safe operation of Internet of Things systems. Due to the lack of networked analysis of the correlation between multiple circuit breakers and intelligent modeling of anomaly prediction, it is difficult to accurately evaluate the interaction effects between the target circuit breaker and other circuit breakers. The commonly used isolated monitoring methods cannot adapt to the complex Internet of Things system environment, resulting in insufficient accuracy of anomaly risk assessment, easy omission of potential faults, and limiting the reliability and operation efficiency of Internet of Things system monitoring. It can be seen that there are defects in the existing technologies and they need to be solved urgently. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a multi - circuit breaker monitoring method and an Internet of Things management system based on an AI algorithm, which can achieve accurate anomaly risk assessment based on graph networks and multi - circuit breaker associations, improve the accuracy and reliability of Internet of Things system monitoring in the target area, and reduce the circuit breaker failure risk.

[0004] To solve the above - mentioned technical problem, in the first aspect of the present invention, a multi - circuit breaker monitoring method based on an AI algorithm is disclosed, and the method includes: Obtain the sensing data of multiple circuit breakers in a target area; Based on the sensing data of the multiple circuit breakers and a graph network algorithm, establish a circuit breaker data network; Input the circuit breaker data network into a pre - trained anomaly prediction AI model to obtain an anomaly prediction result corresponding to each circuit breaker; In response to the monitoring requirement for any one of the circuit breakers, determine the anomaly risk of the circuit breaker according to the anomaly prediction results corresponding to the circuit breaker and at least one other circuit breaker.

[0005] As an optional implementation manner, in the first aspect of the present invention, the sensing data includes at least one of current data, temperature data, humidity data, sound data, image data, and actuator signal data.

[0006] As an optional implementation manner, in the first aspect of the present invention, the step of establishing a circuit breaker data network based on the sensing data of the multiple circuit breakers and a graph network algorithm includes: Determine each of the circuit breakers as a graph node; Determine the parameter information of the corresponding circuit breaker and the sensing data as the node parameter data of the graph node; For any two of the graph nodes, calculate the adjacency parameter between the two graph nodes according to the sensing data; Determine all the graph nodes and the corresponding adjacency parameters as the circuit breaker data network.

[0007] As an optional implementation manner, in the first aspect of the present invention, the calculating the adjacency parameter between the two graph nodes according to the sensing data includes: Calculate the data similarity between the sensing data corresponding to the circuit breakers corresponding to the two graph nodes; Calculate the data volume difference between the data volumes of the sensing data corresponding to the circuit breakers corresponding to the two graph nodes; Determine the data matrix composed of the data similarity and the data volume difference as the adjacency parameter between the two graph nodes.

[0008] As an optional implementation manner, in the first aspect of the present invention, the abnormal prediction AI model is trained by a training data set including a plurality of training circuit breaker data networks and corresponding abnormal annotations; the training circuit breaker data network includes a plurality of training circuit breaker nodes, where each training circuit breaker node is annotated with corresponding node parameter data and circuit breaker abnormal annotations, and the adjacency parameter is annotated between any two training circuit breaker nodes.

[0009] As an optional implementation manner, in the first aspect of the present invention, the abnormal prediction result and the circuit breaker abnormal annotation include an abnormal type, an abnormal probability, and an expected abnormal occurrence time.

[0010] As an optional implementation manner, in the first aspect of the present invention, the determining the abnormal risk of the circuit breaker according to the abnormal prediction results corresponding to the circuit breaker and at least one other circuit breaker includes: Determine other circuit breakers related to the circuit breaker in the circuit breaker data network to obtain at least one associated circuit breaker; Calculate the average value of the abnormal probabilities of all the associated circuit breakers to obtain an associated abnormal probability; Calculate the average value of the expected abnormal occurrence times of all the associated circuit breakers to obtain an associated abnormal occurrence time; Calculate the probability average value of the associated abnormal probability and the abnormal probability of the circuit breaker; Calculate the time difference between the calculated associated anomaly probability and the predicted anomaly occurrence time of the circuit breaker; Judge whether the average probability is greater than a probability threshold and whether the time difference is less than a time difference threshold to obtain a first judgment result; When the first judgment result is yes, calculate whether the similarity between the set of the anomaly types of all the associated circuit breakers and the anomaly type of this circuit breaker is greater than a preset similarity threshold to obtain a second judgment result; When the second judgment result is yes, determine this circuit breaker as a risk circuit breaker, and determine the anomaly type and the predicted anomaly occurrence time as the risk parameters of this circuit breaker.

[0011] As an optional implementation manner, in the first aspect of the present invention, the determining other circuit breakers related to this circuit breaker in the circuit breaker data network to obtain at least one associated circuit breaker includes: For each other graph node with the adjacency parameter existing between the graph nodes corresponding to this circuit breaker, obtain the data similarity and the data volume difference between this other graph node and the graph node corresponding to this circuit breaker; Calculate the average value of the data similarity and the data volume difference to obtain the associated parameter corresponding to this other graph node; Screen out the circuit breakers corresponding to all the other graph nodes whose associated parameters are greater than a preset parameter threshold to obtain at least one associated circuit breaker.

[0012] A second aspect of an embodiment of the present invention discloses an Internet of Things management system, and the system includes: An acquisition module, configured to acquire sensing data of a plurality of circuit breakers in a target area; A building module, configured to build a circuit breaker data network based on the graph network algorithm according to the sensing data of the plurality of circuit breakers; A prediction module, configured to input the circuit breaker data network into a pre-trained anomaly prediction AI model to obtain an anomaly prediction result corresponding to each circuit breaker; A monitoring module, configured to respond to a monitoring requirement for any one of the circuit breakers, and determine the anomaly risk of this circuit breaker according to the anomaly prediction results corresponding to this circuit breaker and at least one other circuit breaker.

[0013] As an optional implementation manner, in the second aspect of the present invention, the sensing data includes at least one of current data, temperature data, humidity data, sound data, image data, and actuator signal data.

[0014] As an optional implementation manner, in the second aspect of the present invention, the specific manner in which the establishment module establishes a breaker data network based on the sensing data of the multiple breakers and based on a graph network algorithm includes: Determine each of the breakers as a graph node; Determine the parameter information and the sensing data of the corresponding breaker as the node parameter data of the graph node; For any two of the graph nodes, calculate the adjacency parameter between the two graph nodes according to the sensing data; Determine all the graph nodes and the corresponding adjacency parameters as the breaker data network.

[0015] As an optional implementation manner, in the second aspect of the present invention, the specific manner in which the establishment module calculates the adjacency parameter between the two graph nodes according to the sensing data includes: Calculate the data similarity between the sensing data corresponding to the breakers corresponding to the two graph nodes; Calculate the data volume difference between the data volumes of the sensing data corresponding to the breakers corresponding to the two graph nodes; Determine the data matrix composed of the data similarity and the data volume difference as the adjacency parameter between the two graph nodes.

[0016] As an optional implementation manner, in the second aspect of the present invention, the abnormal prediction AI model is trained by a training data set including multiple training breaker data networks and corresponding abnormal annotations; the training breaker data network includes multiple training breaker nodes, where each training breaker node is annotated with corresponding node parameter data and breaker abnormal annotations, and the adjacency parameter is annotated between any two training breaker nodes.

[0017] As an optional implementation manner, in the second aspect of the present invention, the abnormal prediction result and the breaker abnormal annotation include an abnormal type, an abnormal probability, and an expected abnormal occurrence time.

[0018] As an optional implementation manner, in the second aspect of the present invention, the specific manner in which the monitoring module determines the abnormal risk of the breaker according to the abnormal prediction results corresponding to the breaker and at least one other breaker includes: Determine other breakers related to the breaker in the breaker data network to obtain at least one associated breaker; Calculate the average value of the abnormal probabilities of all the associated breakers to obtain an associated abnormal probability; Calculate the average value of the predicted abnormal occurrence times of all the associated circuit breakers to obtain the associated abnormal occurrence time; Calculate the probability average value of the associated abnormal probability and the abnormal probability of this circuit breaker; Calculate the time difference between the associated abnormal probability and the predicted abnormal occurrence time of this circuit breaker; Judge whether the probability average value is greater than the probability threshold and whether the time difference is less than the time difference threshold to obtain a first judgment result; When the first judgment result is yes, calculate whether the similarity between the set of abnormal types of all the associated circuit breakers and the abnormal type of this circuit breaker is greater than a preset similarity threshold to obtain a second judgment result; When the second judgment result is yes, determine this circuit breaker as a risk circuit breaker, and determine the abnormal type and the predicted abnormal occurrence time as the risk parameters of this circuit breaker.

[0019] As an optional implementation manner, in the second aspect of the present invention, the specific manner for the monitoring module to determine other circuit breakers related to this circuit breaker in the circuit breaker data network to obtain at least one associated circuit breaker includes: For each other graph node where there is the adjacency parameter between the graph nodes corresponding to this circuit breaker, obtain the data similarity and the data volume difference between this other graph node and the graph node corresponding to this circuit breaker; Calculate the average value of the data similarity and the data volume difference to obtain the associated parameter corresponding to this other graph node; Screen out the circuit breakers corresponding to all the other graph nodes whose associated parameters are greater than a preset parameter threshold to obtain at least one associated circuit breaker.

[0020] The third aspect of the present invention discloses another Internet of Things management system, and the system includes: A memory storing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory and executes some or all of the steps in the multi-circuit breaker monitoring method based on the AI algorithm disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium, and the computer storage medium stores computer instructions, which are used to execute some or all of the steps in the multi-circuit breaker monitoring method based on the AI algorithm disclosed in the first aspect of the present invention when being called.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: The present invention obtains sensing data of multiple circuit breakers in a target area, constructs a circuit breaker data network based on a graph network algorithm, inputs it into a pre-trained anomaly prediction AI model to obtain the anomaly prediction results of each circuit breaker, and determines its anomaly risk by combining the prediction results of the target circuit breaker and other circuit breakers when responding to monitoring requirements. Therefore, it can achieve accurate anomaly risk assessment based on graph network and multi-circuit breaker association, improve the accuracy and reliability of monitoring of the Internet of Things system in the target area, and reduce the risk of circuit breaker failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0024] Figure 1 It is a schematic flowchart of a multi-circuit breaker monitoring method based on an AI algorithm disclosed in an embodiment of the present invention.

[0025] Figure 2 It is a schematic structural diagram of an Internet of Things management system disclosed in an embodiment of the present invention.

[0026] Figure 3 It is a schematic structural diagram of another Internet of Things management system disclosed in an embodiment of the present invention.

[0027] Figure 4 It is a schematic diagram of a circuit breaker monitoring interface of a transparent Internet of Things management system disclosed in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] In order to enable those skilled in the art of the present technology to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0029] In the description and claims of the present invention and the above-mentioned drawings, terms such as "first", "second", etc. are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally further include steps or units not listed, or may optionally further include other steps or units inherent to these processes, methods, products or equipment.

[0030] Reference to "embodiment" herein means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0031] The present invention discloses a multi-circuit breaker monitoring method and an Internet of Things management system based on an AI algorithm. By acquiring the sensing data of multiple circuit breakers in a target area and constructing a circuit breaker data network based on a graph network algorithm, and inputting it into a pre-trained anomaly prediction AI model to obtain the anomaly prediction results of each circuit breaker, when responding to the monitoring requirements, the anomaly risk is determined by combining the prediction results of the target circuit breaker and other circuit breakers, so as to achieve accurate anomaly risk assessment based on the graph network and multi-circuit breaker association, improve the accuracy and reliability of the monitoring of the Internet of Things system in the target area, and reduce the circuit breaker failure risk. The following will be described in detail respectively.

[0032] Embodiment 1 Please refer to Figure 1 , Figure 1 which is a schematic flow chart of a multi-circuit breaker monitoring method based on an AI algorithm disclosed in an embodiment of the present invention. Among them, Figure 1 the described multi-circuit breaker monitoring method based on an AI algorithm can be applied to a data processing system / data processing device / data processing server (wherein, the server includes a local processing server or a cloud processing server). As Figure 1 shown, the multi-circuit breaker monitoring method based on an AI algorithm may include the following operations: 101. Acquire the sensing data of multiple circuit breakers in the target area.

[0033] 102. Based on the sensing data of multiple circuit breakers, establish a circuit breaker data network based on a graph network algorithm. 103. Input the circuit breaker data network into a pre-trained anomaly prediction AI model to obtain the anomaly prediction result corresponding to each circuit breaker. 104. In response to the monitoring requirement for any circuit breaker, determine the abnormal risk of the circuit breaker according to the abnormal prediction results corresponding to the circuit breaker and at least one other circuit breaker.

[0034] Optionally, the target area can be a residential area, an industrial park, a commercial building or a power grid node area, which is not limited in the present invention.

[0035] Optionally, the circuit breaker can be a low-voltage circuit breaker, a high-voltage circuit breaker or an intelligent circuit breaker, which is not limited in the present invention.

[0036] Optionally, the nodes of the circuit breaker data network can represent circuit breakers, and the edges can represent the electrical connection, physical proximity relationship or data correlation between circuit breakers, which is not limited in the present invention.

[0037] Optionally, the construction of the circuit breaker data network can be based on the spatio-temporal characteristics or topological structure of sensing data, which is not limited in the present invention.

[0038] Optionally, the abnormal prediction result can be an abnormal probability, an abnormal type or an abnormal severity level, which is not limited in the present invention.

[0039] Optionally, the training of the abnormal prediction AI model can be based on historical sensing data, simulated fault data or real-time feedback data, which is not limited in the present invention.

[0040] Optionally, the determination process can be based on a weighted fusion algorithm, a rule inference algorithm or a machine learning classification algorithm, which is not limited in the present invention.

[0041] Optionally, the calculation of the abnormal risk can be optimized by combining the operating environment of the circuit breaker, historical fault records or network topology relationships, which is not limited in the present invention.

[0042] In a specific implementation, an intelligent Internet of Things circuit breaker is implemented to fully monitor the circuit conditions in the area. The circuit breaker adopts a modular design, including independent encapsulation of a protection unit, a communication module and a measurement unit, supports function expansion, and the protection action time is ≤40ms in case of short circuit and ≤100ms in case of leakage. It supports an edge computing architecture to achieve intelligent collaboration of local logic judgment and cloud big data analysis. Specifically, its technical route is as follows: Hardware platform: Adopt STM32F4 series MCU (floating-point operation) + high-precision metering chip; Protection algorithm: Improved RMS algorithm (full-cycle Fourier analysis); Communication protocol: Modbus RTU (wired) + MQTT over TLS (wireless encrypted transmission); Structural design: Double-break contact system + magnetic blowout arc grid (verified for 10kA breaking capacity).

[0043] Specifically, its structural design includes: Contact system: Silver-nickel alloy main contacts + tungsten-copper arcing horns (mechanical life ≥ 30,000 times); Insulation design: 3.0mm creepage distance + flame-retardant groove structure (passes the glow wire test of GB / T 5169.11).

[0044] Specifically, the acquisition functions of its sensing and monitoring system include: Voltage / current: 128-point sampling per cycle (4kHz / each channel); Temperature monitoring: PT1000 surface-mounted sensor (accuracy of ±1°C).

[0045] It can be seen that through the above-mentioned invention embodiments, by obtaining the sensing data of multiple circuit breakers in the target area and constructing a circuit breaker data network based on the graph network algorithm, and inputting it into a pre-trained abnormal prediction AI model to obtain the abnormal prediction results of each circuit breaker, when responding to the monitoring requirements, the abnormal risk is determined by combining the prediction results of the target circuit breaker and other circuit breakers, so as to realize accurate abnormal risk assessment based on the graph network and multi-circuit breaker association, improve the accuracy and reliability of the monitoring of the Internet of Things system in the target area, and reduce the circuit breaker failure risk.

[0046] As an optional embodiment, in the above steps, the sensing data includes at least one of current data, temperature data, humidity data, sound data, image data, and actuator signal data.

[0047] It can be seen that through the above optional embodiment, the content of the sensing data is defined to comprehensively characterize the working sensing characteristics of the circuit breaker, assist in realizing accurate abnormal risk assessment based on the graph network and multi-circuit breaker association, improve the accuracy and reliability of the monitoring of the Internet of Things system in the target area, and reduce the circuit breaker failure risk.

[0048] As an optional embodiment, in the above steps, based on the graph network algorithm, establishing a circuit breaker data network according to the sensing data of multiple circuit breakers includes: Determine each circuit breaker as a graph node; Determine the parameter information and sensing data of the corresponding circuit breaker as the node parameter data of the graph node; For any two graph nodes, calculate the adjacency parameter between the two graph nodes according to the sensing data; Determine all the graph nodes and the corresponding adjacency parameters as the circuit breaker data network.

[0049] Optionally, the parameter information may be the rated current, rated voltage, model, or manufacturing batch of the circuit breaker, which is not limited in the present invention.

[0050] Optionally, the adjacency parameter may be electrical connection strength, physical distance, data correlation, or communication delay, which is not limited in the present invention.

[0051] Optionally, the calculation of the adjacency parameter may be based on Euclidean distance, correlation coefficient analysis, or machine learning algorithms, which is not limited in the present invention.

[0052] It can be seen that through the above optional embodiments, by setting each circuit breaker as a graph node and using its parameter information and sensing data as node parameter data, calculating the adjacency parameter based on the sensing data between any two graph nodes, and integrating all graph nodes and adjacency parameters to form a circuit breaker data network, the accurate network construction based on node parameters and data association can be realized, the accuracy of circuit breaker anomaly prediction and the monitoring efficiency of the Internet of Things system can be improved, and the risk of false fault judgment can be reduced.

[0053] As an optional embodiment, in the above steps, calculating the adjacency parameter between the two graph nodes according to the sensing data includes: Calculating the data similarity between the sensing data corresponding to the circuit breakers corresponding to the two graph nodes; Calculating the data volume difference between the data volumes of the sensing data corresponding to the circuit breakers corresponding to the two graph nodes; Determining the data matrix composed of the data similarity and the data volume difference as the adjacency parameter between the two graph nodes.

[0054] Optionally, the data similarity may be calculated based on cosine similarity, Euclidean distance, Pearson correlation coefficient, or dynamic time warping algorithm, which is not limited in the present invention.

[0055] Optionally, the calculation of the data similarity may be optimized by combining the temporal characteristics or statistical characteristics of the sensing data, which is not limited in the present invention.

[0056] Optionally, the data volume may be the data packet size, sampling frequency, number of data points, or total data transmission volume, which is not limited in the present invention.

[0057] Optionally, the calculation of the data volume difference may be based on absolute difference, relative difference, or normalized difference, which is not limited in the present invention.

[0058] It can be seen that through the above optional embodiments, by calculating the data similarity and the difference in data volume between the sensing data of the circuit breakers corresponding to any two graph nodes, a data matrix is formed as the adjacency parameter between the two graph nodes, thereby realizing the precise construction of adjacency parameters based on similarity and data volume difference, improving the accuracy of the circuit breaker data network and the reliability of anomaly prediction, and reducing the risk of misjudgment of faults in the Internet of Things system.

[0059] As an optional embodiment, in the above steps, the anomaly prediction AI model is trained through a training data set including multiple training circuit breaker data networks and corresponding anomaly annotations; the training circuit breaker data network includes multiple training circuit breaker nodes, where each training circuit breaker node is annotated with corresponding node parameter data and circuit breaker anomaly annotations, and adjacency parameters are annotated between any two training circuit breaker nodes.

[0060] Specifically, the anomaly prediction AI model is implemented as a graph neural network model, which includes: A multi-level graph neural network, which is used to perform multi-level parallel graph convolution operations on the input training circuit breaker data network, generate multi-scale embedding vectors of nodes and edges, and construct a multi-level graph neural network; An embedding vector construction network, which is used to perform cross-layer fusion on the embedding vectors of the multi-level graph neural network, and introduce a Transformer module for global interaction modeling to obtain embedding vectors; A fully connected layer for prediction, which is used to obtain the corresponding anomaly annotations and perform predictions based on the multi-level graph neural network and the embedding vectors, and optimize the parameters in combination with a loss function.

[0061] It can be seen that through the above optional embodiments, the training details of the anomaly prediction AI model and the characteristics of the training data set are defined, so that the anomaly prediction AI model can realize precise anomaly risk assessment based on the graph network and the association of multiple circuit breakers, improve the accuracy and reliability of the monitoring of the Internet of Things system in the target area, and reduce the circuit breaker failure risk.

[0062] As an optional embodiment, in the above steps, the anomaly prediction result and the circuit breaker anomaly annotation include the anomaly type, the anomaly probability, and the expected anomaly occurrence time.

[0063] It can be seen that through the above optional embodiments, the details of the anomaly prediction result and the anomaly annotation are defined, so that subsequent precise anomaly risk assessment based on the graph network and the association of multiple circuit breakers can be realized based on the prediction result, improving the accuracy and reliability of the monitoring of the Internet of Things system in the target area, and reducing the circuit breaker failure risk.

[0064] As an optional embodiment, in the above steps, determining the anomaly risk of the circuit breaker according to the anomaly prediction results corresponding to the circuit breaker and at least one other circuit breaker includes: Identify other circuit breakers related to the circuit breaker in the circuit breaker data network to obtain at least one associated circuit breaker; Calculate the average value of the abnormal probabilities of all associated circuit breakers to obtain the associated abnormal probability; Calculate the average value of the predicted abnormal occurrence times of all associated circuit breakers to obtain the associated abnormal occurrence time; Calculate the probability average value of the associated abnormal probability and the abnormal probability of this circuit breaker; Calculate the time difference between the associated abnormal probability and the predicted abnormal occurrence time of this circuit breaker; Judge whether the probability average value is greater than the probability threshold and whether the time difference is less than the time difference threshold to obtain the first judgment result; When the first judgment result is yes, calculate whether the similarity between the set of abnormal types of all associated circuit breakers and the abnormal type of this circuit breaker is greater than the preset similarity threshold to obtain the second judgment result; When the second judgment result is yes, determine this circuit breaker as a risk circuit breaker, and determine the abnormal type and the predicted abnormal occurrence time as the risk parameters of this circuit breaker.

[0065] It can be seen that through the above optional embodiments, by identifying the associated circuit breakers related to the target circuit breaker in the circuit breaker data network, calculating the average values of their abnormal probabilities and predicted abnormal occurrence times to obtain the associated abnormal probability and associated abnormal occurrence time, combining the abnormal probability and time of the target circuit breaker to calculate the probability average value and time difference, judging whether the probability average value exceeds the probability threshold and whether the time difference is less than the time difference threshold, if so, further judging whether the similarity between the set of abnormal types of the associated circuit breakers and the abnormal type of the target circuit breaker exceeds the similarity threshold, if the similarity exceeds the threshold, determining the target circuit breaker as a risk circuit breaker and using the abnormal type and predicted abnormal occurrence time as risk parameters, so as to realize accurate risk assessment based on associated circuit breakers and multi-dimensional threshold judgment, improve the accuracy and reliability of the monitoring of the Internet of Things system, and reduce the circuit breaker failure risk.

[0066] As an optional embodiment, in the above steps, identifying other circuit breakers related to the circuit breaker in the circuit breaker data network to obtain at least one associated circuit breaker includes: For each other graph node with an adjacency parameter between the graph nodes corresponding to this circuit breaker, obtain the data similarity and data volume difference between the other graph node and the graph node corresponding to this circuit breaker; Calculate the average value of the data similarity and data volume difference to obtain the associated parameter corresponding to the other graph node; Screen out the circuit breakers corresponding to all other graph nodes whose associated parameters are greater than the preset parameter threshold to obtain at least one associated circuit breaker.

[0067] It can be seen that through the above optional embodiments, by obtaining the data similarity and the difference in data volume of other graph nodes having adjacency parameters with the target circuit breaker graph node, calculating the average value of the two as the correlation parameter, and screening out the circuit breakers corresponding to the other graph nodes whose correlation parameters exceed the preset threshold as the associated circuit breakers, the accurate identification of the associated circuit breakers based on data similarity and data volume difference can be realized, the accuracy of the abnormal risk assessment of the circuit breaker and the monitoring efficiency of the Internet of Things system can be improved, and the risk of false fault judgment can be reduced.

[0068] Embodiment 2 Please refer to Figure 2 , Figure 2 which is a schematic structural diagram of an Internet of Things management system disclosed in an embodiment of the present invention. Among them, Figure 2 the described Internet of Things management system can be applied to a data processing system / data processing device / data processing server (wherein, the server includes a local processing server or a cloud processing server). As Figure 2 shown, the Internet of Things management system may include: An acquisition module 201, configured to acquire sensing data of a plurality of circuit breakers in a target area.

[0069] A building module 202, configured to establish a circuit breaker data network based on the sensing data of the plurality of circuit breakers and based on a graph network algorithm. A prediction module 203, configured to input the circuit breaker data network into a pre-trained abnormal prediction AI model to obtain an abnormal prediction result corresponding to each circuit breaker. A monitoring module 204, configured to, in response to a monitoring requirement for any circuit breaker, determine the abnormal risk of the circuit breaker according to the abnormal prediction results corresponding to the circuit breaker and at least one other circuit breaker.

[0070] It can be seen that the above-mentioned embodiments of the invention acquire the sensing data of a plurality of circuit breakers in the target area and construct a circuit breaker data network based on the graph network algorithm, input it into a pre-trained abnormal prediction AI model to obtain the abnormal prediction results of each circuit breaker, and determine its abnormal risk by combining the prediction results of the target circuit breaker and other circuit breakers when responding to the monitoring requirement, so as to realize accurate abnormal risk assessment based on the graph network and multi-circuit breaker association, improve the accuracy and reliability of the monitoring of the Internet of Things system in the target area, and reduce the circuit breaker failure risk.

[0071] As an optional embodiment, the sensing data includes at least one of current data, temperature data, humidity data, sound data, image data, and actuator signal data.

[0072] It can be seen that through the above optional embodiments, the content of the sensing data is defined to comprehensively characterize the working sensing characteristics of the circuit breaker, assist in realizing accurate abnormal risk assessment based on the graph network and multi - circuit - breaker association, improve the accuracy and reliability of the monitoring of the Internet of Things system in the target area, and reduce the circuit breaker failure risk.

[0073] As an optional embodiment, the specific manner in which the establishment module establishes the circuit breaker data network based on the graph network algorithm according to the sensing data of multiple circuit breakers includes: Determine each circuit breaker as a graph node; Determine the parameter information and sensing data of the corresponding circuit breaker as the node parameter data of the graph node; For any two graph nodes, calculate the adjacency parameter between the two graph nodes according to the sensing data; Determine all the graph nodes and the corresponding adjacency parameters as the circuit breaker data network.

[0074] It can be seen that through the above optional embodiments, by setting each circuit breaker as a graph node and using its parameter information and sensing data as node parameter data, calculating the adjacency parameter between any two graph nodes based on the sensing data, and integrating all graph nodes and adjacency parameters to form a circuit breaker data network, the accurate network construction based on node parameters and data association is realized, the accuracy of circuit breaker abnormal prediction and the monitoring efficiency of the Internet of Things system are improved, and the risk of false fault judgment is reduced.

[0075] As an optional embodiment, the specific manner in which the establishment module calculates the adjacency parameter between the two graph nodes according to the sensing data includes: Calculate the data similarity between the sensing data corresponding to the circuit breakers corresponding to the two graph nodes; Calculate the data volume difference between the data volumes of the sensing data corresponding to the circuit breakers corresponding to the two graph nodes; Determine the data matrix composed of the data similarity and the data volume difference as the adjacency parameter between the two graph nodes.

[0076] It can be seen that through the above optional embodiments, by calculating the data similarity and the data volume difference between the sensing data of the circuit breakers corresponding to any two graph nodes and forming a data matrix as the adjacency parameter between the two graph nodes, the accurate adjacency parameter construction based on similarity and data volume difference is realized, the accuracy of the circuit breaker data network and the reliability of abnormal prediction are improved, and the risk of false fault judgment of the Internet of Things system is reduced.

[0077] As an alternative embodiment, the anomaly prediction AI model is trained through a training dataset including multiple training breaker data networks and corresponding anomaly annotations; the training breaker data network includes multiple training breaker nodes, where each training breaker node is annotated with corresponding node parameter data and breaker anomaly annotations, and adjacency parameters are annotated between any two training breaker nodes.

[0078] It can be seen that through the above alternative embodiment, the training details of the anomaly prediction AI model and the characteristics of the training dataset are defined, enabling the anomaly prediction AI model to achieve accurate anomaly risk assessment based on the graph network and multi-breaker association, improving the accuracy and reliability of the monitoring of the Internet of Things system in the target area, and reducing the breaker failure risk.

[0079] As an alternative embodiment, the anomaly prediction result and the breaker anomaly annotation include the anomaly type, the anomaly probability, and the predicted anomaly occurrence time.

[0080] It can be seen that through the above alternative embodiment, the details of the anomaly prediction result and the anomaly annotation are defined, enabling subsequent accurate anomaly risk assessment based on the prediction result based on the graph network and multi-breaker association, improving the accuracy and reliability of the monitoring of the Internet of Things system in the target area, and reducing the breaker failure risk.

[0081] As an alternative embodiment, the specific manner in which the monitoring module determines the anomaly risk of the breaker according to the anomaly prediction results corresponding to the breaker and at least one other breaker includes: Determine other breakers related to the breaker in the breaker data network to obtain at least one associated breaker; Calculate the average value of the anomaly probabilities of all associated breakers to obtain the associated anomaly probability; Calculate the average value of the predicted anomaly occurrence times of all associated breakers to obtain the associated anomaly occurrence time; Calculate the probability average value of the associated anomaly probability and the anomaly probability of the breaker; Calculate the time difference between the associated anomaly probability and the predicted anomaly occurrence time of the breaker; Judge whether the probability average value is greater than the probability threshold and whether the time difference is less than the time difference threshold to obtain a first judgment result; When the first judgment result is yes, calculate whether the similarity between the set of anomaly types of all associated breakers and the anomaly type of the breaker is greater than a preset similarity threshold to obtain a second judgment result; When the second judgment result is yes, determine the breaker as a risk breaker, and determine the anomaly type and the predicted anomaly occurrence time as the risk parameters of the breaker.

[0082] It can be seen that through the above optional embodiments, by identifying associated circuit breakers related to the target circuit breaker in the circuit breaker data network, calculating the average values of their abnormal probabilities and the predicted abnormal occurrence times to obtain the associated abnormal probabilities and the associated abnormal occurrence times, combining the abnormal probabilities and times of the target circuit breaker to calculate the average probability value and the time difference, determining whether the average probability value exceeds the probability threshold and whether the time difference is less than the time difference threshold, if so, further determining whether the similarity between the set of abnormal types of the associated circuit breaker and the abnormal type of the target circuit breaker exceeds the similarity threshold, and if the similarity exceeds the threshold, determining the target circuit breaker as a risk circuit breaker and using the abnormal type and the predicted abnormal occurrence time as risk parameters, thereby realizing accurate risk assessment based on associated circuit breakers and multi-dimensional threshold judgments, improving the accuracy and reliability of the monitoring of the Internet of Things system, and reducing the risk of circuit breaker failures.

[0083] As an optional embodiment, the specific manner in which the monitoring module determines other circuit breakers related to the circuit breaker in the circuit breaker data network to obtain at least one associated circuit breaker includes: For each other graph node having an adjacency parameter between the graph nodes corresponding to the circuit breaker, obtain the data similarity and the data volume difference between the other graph node and the graph node corresponding to the circuit breaker; Calculate the average value of the data similarity and the data volume difference to obtain the associated parameter corresponding to the other graph node; Screen out the circuit breakers corresponding to all other graph nodes whose associated parameters are greater than the preset parameter threshold to obtain at least one associated circuit breaker.

[0084] It can be seen that through the above optional embodiments, by obtaining the data similarity and the data volume difference of other graph nodes having an adjacency parameter with the graph node of the target circuit breaker, calculating the average value of the two as the associated parameter, and screening out the circuit breakers corresponding to other graph nodes whose associated parameters exceed the preset threshold as the associated circuit breakers, thereby realizing accurate identification of associated circuit breakers based on data similarity and data volume differences, improving the accuracy of circuit breaker abnormal risk assessment and the monitoring efficiency of the Internet of Things system, and reducing the risk of false fault judgments.

[0085] In a specific implementation scheme, based on the monitoring scheme and the Internet of Things management system scheme disclosed in the present invention, a transparent Internet of Things management system is implemented. Its kernel algorithm is based on the circuit breaker monitoring method disclosed in the present invention to monitor multiple intelligent circuit breakers, and display and alarm the graph network and risk information constructed during the monitoring. Its function design includes: 1. Device Map 2. Device Management: Circuit Breaker Information, Gateway Information, Circuit Breaker Type 3. Group Management 4. Alarm Management: Leakage Alarm, Alarms to be Processed, Historical Alarms, Alarm Notification Configuration 5. Statistical reports: energy consumption statistics, alarm statistics 6. System management: leakage self-check task, user management, role management, menu management, area management, dictionary management, parameter setting, notice and announcement, log management.

[0086] In an implementation scenario, multiple circuit breakers, corresponding risk information, and relevant monitoring information can be Figure 4 displayed in a certain way to prompt the user, facilitating the user's management and monitoring of the circuit breakers.

[0087] Embodiment III Please refer to Figure 3 , Figure 3 , which is another Internet of Things management system disclosed in the embodiments of the present invention. Figure 3 The described Internet of Things management system is applied to a data processing system / data processing device / data processing server (wherein, the server includes a local processing server or a cloud processing server). As Figure 3 shown, the Internet of Things management system may include: A memory 301 storing executable program code; A processor 302 coupled to the memory 301; Wherein, the processor 302 calls the executable program code stored in the memory 301 to execute the steps of the multi-circuit breaker monitoring method based on the AI algorithm described in Embodiment I.

[0088] Embodiment IV The embodiments of the present invention disclose a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the multi-circuit breaker monitoring method based on the AI algorithm described in Embodiment I.

[0089] Embodiment V The embodiments of the present invention disclose a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the multi-circuit breaker monitoring method based on the AI algorithm described in Embodiment I.

[0090] The above describes specific embodiments of this specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily have to be performed in the specific order or continuous order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0091] The systems, apparatuses, modules or units illustrated in the above embodiments may be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0092] For the convenience of description, the above apparatuses are described by dividing them into various units according to functions. Of course, when implementing this specification, the functions of each unit may be implemented in one or more software and / or hardware.

[0093] Those skilled in the art should understand that the embodiments of this specification may be provided as a method, a system, or a computer program product. Therefore, the embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0094] This specification is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of this specification. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate an apparatus for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0095] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing devices to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction apparatus that implements the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0096] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 steps of the functions specified in one block or multiple blocks.

[0097] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0098] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer-readable media.

[0099] Computer-readable media includes permanent and non-permanent, removable and non-removable media and can store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0100] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising the element.

[0101] This specification can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0102] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and reference can be made to the corresponding parts of the method embodiments for relevant details.

[0103] Finally, it should be noted that the multi-breaker monitoring method and the Internet of Things management system based on the AI algorithm disclosed in the embodiments of the present invention only disclose the preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi - circuit breaker monitoring method based on AI algorithm, characterized in that, The method includes: Obtaining sensing data of multiple circuit breakers in a target area; Based on the sensing data of the multiple circuit breakers, establishing a circuit breaker data network based on a graph network algorithm; Inputting the circuit breaker data network into a pre-trained anomaly prediction AI model to obtain an anomaly prediction result corresponding to each circuit breaker; In response to a monitoring requirement for any one of the circuit breakers, determining the anomaly risk of the circuit breaker according to the anomaly prediction results corresponding to the circuit breaker and at least one other circuit breaker; 2. The multi - circuit breaker monitoring method based on the AI algorithm according to claim 1, wherein The sensing data includes at least one of current data, temperature data, humidity data, sound data, image data, and actuator signal data; 3. The multi - circuit breaker monitoring method based on the AI algorithm according to claim 1, characterized in that, The establishing a circuit breaker data network based on the sensing data of the multiple circuit breakers and based on a graph network algorithm includes: Determining each circuit breaker as a graph node; Determining the parameter information and the sensing data of the corresponding circuit breaker as the node parameter data of the graph node; For any two graph nodes, calculating the adjacency parameter between the two graph nodes according to the sensing data; Determining all the graph nodes and the corresponding adjacency parameters as the circuit breaker data network; 4. The multi - circuit breaker monitoring method based on the AI algorithm according to claim 3, characterized in that, The calculating the adjacency parameter between the two graph nodes according to the sensing data includes: Calculating the data similarity between the sensing data corresponding to the circuit breakers corresponding to the two graph nodes; Calculating the data volume difference between the data volumes of the sensing data corresponding to the circuit breakers corresponding to the two graph nodes; Determining the data matrix composed of the data similarity and the data volume difference as the adjacency parameter between the two graph nodes; 5. The multi - circuit breaker monitoring method based on the AI algorithm according to claim 3, wherein, The anomaly prediction AI model is obtained by training with a training data set including multiple training circuit breaker data networks and corresponding anomaly annotations; the training circuit breaker data network includes multiple training circuit breaker nodes, where each training circuit breaker node is annotated with corresponding node parameter data and circuit breaker anomaly annotations, and the adjacency parameter is annotated between any two training circuit breaker nodes; 6. The multi - circuit breaker monitoring method based on the AI algorithm according to claim 5, characterized in that, The anomaly prediction result and the circuit breaker anomaly annotation include an anomaly type, an anomaly probability, and an expected anomaly occurrence time; 7. The multi - circuit breaker monitoring method based on the AI algorithm according to claim 6, wherein, The determining the anomaly risk of the circuit breaker according to the anomaly prediction results corresponding to the circuit breaker and at least one other circuit breaker includes: Determining other circuit breakers related to the circuit breaker in the circuit breaker data network to obtain at least one associated circuit breaker; Calculating the average value of the anomaly probabilities of all the associated circuit breakers to obtain an associated anomaly probability; Calculating the average value of the expected anomaly occurrence times of all the associated circuit breakers to obtain an associated anomaly occurrence time; Calculating the probability average value of the associated anomaly probability and the anomaly probability of the circuit breaker; Calculating the time difference between the associated anomaly probability and the expected anomaly occurrence time of the circuit breaker; Judging whether the probability average value is greater than a probability threshold and whether the time difference is less than a time difference threshold to obtain a first judgment result; When the first judgment result is yes, calculate whether the similarity between the set of the abnormal types of all the associated circuit breakers and the abnormal type of this circuit breaker is greater than a preset similarity threshold to obtain a second judgment result; When the second judgment result is yes, determine that this circuit breaker is a risk circuit breaker, and determine the abnormal type and the expected abnormal occurrence time as the risk parameters of this circuit breaker.

8. The multi - circuit breaker monitoring method based on the AI algorithm according to claim 7, characterized in that, Determining other circuit breakers related to this circuit breaker in the circuit breaker data network to obtain at least one associated circuit breaker includes: For each other graph node having the adjacency parameter between the graph nodes corresponding to this circuit breaker, obtain the data similarity and the data volume difference between this other graph node and the graph node corresponding to this circuit breaker; Calculate the average value of the data similarity and the data volume difference to obtain the associated parameter corresponding to this other graph node; Filter out the circuit breakers corresponding to all the other graph nodes whose associated parameters are greater than a preset parameter threshold to obtain at least one associated circuit breaker.

9. An Internet of Things management system, characterized in that, The system includes: An acquisition module, configured to acquire sensing data of multiple circuit breakers in a target area; A building module, configured to build a circuit breaker data network based on the sensing data of the multiple circuit breakers and based on a graph network algorithm; A prediction module, configured to input the circuit breaker data network into a pre-trained abnormal prediction AI model to obtain an abnormal prediction result corresponding to each circuit breaker; A monitoring module, configured to, in response to a monitoring requirement for any one of the circuit breakers, determine the abnormal risk of this circuit breaker according to the abnormal prediction results corresponding to this circuit breaker and at least one other circuit breaker.

10. An Internet of Things management system, characterized in that, The system includes: A memory storing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory and executes the multi-circuit breaker monitoring method based on the AI algorithm according to any one of claims 1-8.

Images