Switch cabinet task processing method, device and equipment based on cloud edge-end architecture

Abstract

The application relates to a task processing method, device and equipment of a switch cabinet based on a cloud edge end architecture. Operation data and environment data of each power device on the terminal side are collected by the switch cabinet on the terminal side, operation characteristic information is extracted based on the operation data and the environment data by the edge end, the operation characteristic information is input into a fault probability prediction model by the cloud end, the fault probability prediction model outputs the fault probability of each power device according to the operation characteristic information, the fault probability and a target scheduling strategy are used to schedule and process the calculation task of each power device, and the corresponding task scheduling processing result is obtained. Compared with the traditional task scheduling realized by manual inspection, the application combines the switch cabinet on the terminal side, the edge end and the cloud end, realizes data extraction, feature extraction and calculation task scheduling processing, thereby uses the edge node to complete most of the task scheduling processing steps, and the efficiency of task processing is improved.

Description

Technical Field

[0001] This application relates to the field of power technology, and in particular to a task processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product for a switchgear based on a cloud-edge-device architecture. Background Technology

[0002] Switchgear in a power system receives, distributes, controls, and protects electrical energy. It assembles components such as circuit breakers, disconnectors, and instrument transformers within a metal cabinet to enable the opening and closing of lines and equipment, as well as provide safety protection. Currently, the method for handling power tasks based on switchgear is typically through manual inspection. However, performing power tasks through manual inspection requires dedicated personnel for on-site maintenance of each device, making large-scale centralized management difficult and resulting in low task processing efficiency.

[0003] Therefore, current power task processing methods based on switchgear suffer from low processing efficiency. Summary of the Invention

[0004] Therefore, it is necessary to provide a task processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product for switch cabinets based on a cloud-edge-device architecture that can improve processing efficiency in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a task processing method for switchgear based on a cloud-edge-device architecture, applied in the cloud, the method comprising:

[0006] The system acquires operational characteristic information corresponding to the terminal side; the operational characteristic information is extracted by the edge device based on the operational data and environmental data corresponding to the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect operational data and environmental data of various power devices on the terminal side;

[0007] The operational characteristic information is input into the fault probability prediction model; the fault probability prediction model is used to output the fault probability corresponding to each of the power devices based on the operational characteristic information.

[0008] Based on the fault probability and target scheduling strategy, the computational tasks of each power device are scheduled and processed to obtain the corresponding task scheduling results.

[0009] Secondly, this application provides a task processing method for switchgear based on a cloud-edge architecture, applied at the edge, the method comprising:

[0010] The system acquires corresponding operational and environmental data from the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect operational and environmental data from various power devices on the terminal side.

[0011] Based on the operational data and the environmental data, extract the corresponding operational feature information;

[0012] The operational feature information is sent to the cloud; the cloud is used to input the operational feature information into the fault probability prediction model; the fault probability prediction model is used to output the fault probability corresponding to each of the power devices according to the operational feature information; and to schedule the computational tasks of each of the power devices according to the fault probability and the target scheduling strategy to obtain the corresponding task scheduling results.

[0013] Thirdly, this application also provides a task processing device for a switch cabinet based on a cloud-edge-device architecture, applied in the cloud, the device comprising:

[0014] The first acquisition module is used to acquire the corresponding operating characteristic information of the terminal side; the operating characteristic information is extracted by the edge terminal based on the operating data and environmental data corresponding to the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect the operating data and environmental data of each power device on the terminal side;

[0015] The prediction module is used to input the operating characteristic information into the fault probability prediction model; the fault probability prediction model is used to output the fault probability corresponding to each of the power devices based on the operating characteristic information.

[0016] The processing module is used to schedule the computational tasks of each of the power devices according to the fault probability and the target scheduling strategy, and obtain the corresponding task scheduling results.

[0017] Fourthly, this application also provides a task processing device for a switch cabinet based on a cloud-edge architecture, applied at the edge, the device comprising:

[0018] The second acquisition module is used to acquire the corresponding operating data and environmental data on the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect the operating data and environmental data of each power device on the terminal side;

[0019] The extraction module is used to extract corresponding operational feature information based on the operational data and the environmental data;

[0020] The sending module is used to send the operating feature information to the cloud; the cloud is used to input the operating feature information into the fault probability prediction model; the fault probability prediction model is used to output the fault probability corresponding to each of the power devices according to the operating feature information; and to schedule the calculation tasks of each of the power devices according to the fault probability and the target scheduling strategy to obtain the corresponding task scheduling results.

[0021] Fifthly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0022] Sixthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0023] In a seventh aspect, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0024] The aforementioned task processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product for switchgear based on a cloud-edge-device architecture collects operational and environmental data from various power devices on the terminal side of the switchgear. Operational feature information is extracted from the operational and environmental data at the edge device. This operational feature information is then input into a fault probability prediction model via the cloud. The fault probability prediction model outputs the fault probability of each power device based on the operational feature information. Based on the fault probability and a target scheduling strategy, the computational tasks for each power device are scheduled and processed to obtain the corresponding task scheduling results. Compared to traditional task scheduling methods relying on manual inspection, this application combines the switchgear on the terminal side, the edge device, and the cloud to achieve data extraction, feature extraction, and computational task scheduling, thereby utilizing edge nodes to complete most of the task scheduling steps and improving task processing efficiency. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is an application environment diagram of a task processing method for a switchgear based on a cloud-edge-device architecture in one embodiment;

[0027] Figure 2 This is a flowchart illustrating a task processing method for a switchgear based on a cloud-edge-device architecture in one embodiment.

[0028] Figure 3 This is a flowchart illustrating a task processing method for a switch cabinet based on a cloud-edge-device architecture in another embodiment.

[0029] Figure 4This is a schematic diagram of a task processing system for a switch cabinet based on a cloud-edge-device architecture in one embodiment;

[0030] Figure 5 This is a structural block diagram of a task processing device for a switch cabinet based on a cloud-edge-device architecture in one embodiment;

[0031] Figure 6 This is a structural block diagram of a task processing device for a switch cabinet based on a cloud-edge-device architecture, as shown in another embodiment.

[0032] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0034] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0035] In related technologies, switchgear power systems receive, distribute, control, and protect electrical energy. It assembles components such as circuit breakers, disconnectors, and instrument transformers within a metal cabinet to achieve the opening and closing of lines and equipment, as well as safety protection. Switchgear mainly relies on manual on-site inspection or single-unit programmable logic controller (PLC) control, which has the following shortcomings: poor real-time performance: data acquisition, uploading, and processing often traverse long links, leading to delayed fault warnings; limited computing resources: edge-side sensors have limited computing power, making it difficult to execute complex fault diagnosis models; high operation and maintenance costs: each device requires dedicated on-site maintenance, making large-scale centralized management difficult. With the proposal of the cloud-edge-end three-layer architecture, computing, storage, and control tasks can be distributed at different levels to balance low latency and high computing power requirements. In the switchgear field, IoT-based intelligent management systems have already achieved remote monitoring and temperature and humidity control; however, a unified cloud-edge-end collaborative model has not yet been formed, and a systematic energy consumption and latency optimization formula is lacking.

[0036] Based on this, this application utilizes a three-layer computing collaboration of cloud, edge, and terminal to realize an intelligent switchgear system that enables real-time monitoring, intelligent diagnosis, dynamic scheduling, and safety control of switchgear. It collects multi-modal data such as current, voltage, temperature, humidity, and partial discharge within the switchgear in real time; performs preliminary feature extraction and anomaly detection through edge nodes (edge ​​terminals) to reduce upload bandwidth; and performs global optimization scheduling, machine learning prediction, and cross-site collaborative control on the cloud platform. A quantifiable computational offloading and energy consumption model is adopted to achieve minimum overall system latency and optimal energy consumption. This solves problems such as poor real-time performance, limited computing resources, and high operation and maintenance costs, improving the efficiency of task processing for various power equipment.

[0037] The task processing method for switchgear based on a cloud-edge-device architecture provided in this application embodiment can be applied to, for example... Figure 1 The application environment is illustrated. The terminal communicates with the edge via a network, and the edge communicates with the cloud via a network. The data storage system stores data that the cloud needs to process. This system can be integrated into the cloud or hosted on the cloud or other network servers. The terminal collects operational and environmental data through the switchgear and sends it to the edge. The edge extracts features from the operational and environmental data and sends it to the cloud. The cloud then performs task scheduling based on the extracted features. The terminal includes, but is not limited to, switchgear, various personal computers, laptops, and tablets within the power system. Both the edge and cloud can be independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers providing cloud computing services.

[0038] In one exemplary embodiment, such as Figure 2 As shown, a task processing method for switchgear based on a cloud-edge-device architecture is provided, which is then applied to... Figure 1 Taking the cloud as an example, the explanation includes the following steps S202 to S206. Wherein:

[0039] Step S202: Obtain the corresponding operating characteristic information of the terminal side; the above operating characteristic information is extracted by the edge terminal based on the corresponding operating data and environmental data of the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect the operating data and environmental data of each power device on the terminal side.

[0040] The cloud, edge, and terminal-side switchgear form a cloud-edge-device collaborative network. The terminal-side switchgear incorporates various sensors, which collect operational and environmental data from various electrical devices on the terminal side. The edge device then acquires this operational and environmental data from the terminal-side switchgear. The edge device can extract operational feature information from this data, which can represent the characteristics of each electrical device operating on the terminal side. The edge device can then transmit these extracted features to the cloud, where it receives the corresponding operational feature information from the terminal side.

[0041] Specifically, the intelligent switch cabinet on the terminal side includes a sensor subsystem, a main control unit, a communication module, and security mechanisms. The sensor subsystem includes multi-modal sensors for current / voltage, gas, temperature and humidity, infrared, light, and smoke, used to collect operational and environmental data. The main control unit adopts a dual-core architecture of Advanced RISC Machine Architecture (ARM) and Digital Signal Processor (DSP), supporting both IEC61850 and Modbus / Transmission Control Protocol (TCP) dual-protocol stacks. The communication module supports dual channels for 4G / 5G, Narrow Band Internet of Things (NB-IoT), and edge gateways to ensure reliable transmission of critical commands. The security mechanism achieves data confidentiality based on Transport Layer Security (TLS), Pre-Shared Key (PSK), and quantum key negotiation (referencing the low-voltage distribution area quantum encrypted communication patent). When the switchgear on the terminal side collects the aforementioned operational and environmental data, its acquisition rate can be dynamically adjusted based on the power fluctuation frequency of the power equipment. After collecting the aforementioned data, the switchgear on the terminal side can use wavelet denoising Kalman filtering to improve the signal quality of the collected operational and environmental data.

[0042] Edge computing nodes can be used for data aggregation, deployed with lightweight models, and also enable local control. For data aggregation, the edge can perform time synchronization and noise filtering on sampled data from multiple switchgear units at the same site. For lightweight models, the edge deploys Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, or TransformerLite models to detect partial discharge anomalies in power equipment and predict temperature and humidity trends. For local control, when the edge detects a high-risk anomaly, it can directly trigger the load switch fuse, achieving millisecond-level tripping.

[0043] The cloud platform comprises a data lake, a model center, and a scheduling engine. The data lake centrally stores historical data for all switchgear, supporting a hybrid approach of time-series databases and object storage. The model center utilizes deep learning-based models for load forecasting, fault probability estimation, and optimal scheduling (multi-objective optimization). The scheduling engine employs cloud, edge, and endpoint task offloading optimization, making dynamic decisions based on computational energy consumption and network latency.

[0044] Step S204: Input the above-mentioned operating characteristic information into the fault probability prediction model; the above-mentioned fault probability prediction model is used to output the fault probability corresponding to each of the above-mentioned power equipment based on the above-mentioned operating characteristic information.

[0045] The cloud-based fault probability prediction model is used to predict the fault probability of each power device. Fault probability represents the probability that a power device will fail. The cloud can input the aforementioned operational characteristic information into the fault probability prediction model. Based on this operational characteristic information, the model can predict the fault probability of each power device and output the corresponding fault probability for each device.

[0046] Specifically, the failure probability model can be expressed as: P fault =(w T (x+b). Where, P fault This represents the predicted probability of a fault occurring, with values ​​ranging from (0,1). A value closer to 1 indicates a higher probability of a fault occurring. σ represents the Sigmoid activation function. T This represents the transpose of the weight vector. x represents the input operational features. b represents the bias term (threshold), used to adjust the classification decision boundary, enabling the model to fit more complex data distributions.

[0047] Step S206: Based on the above-mentioned fault probability and target scheduling strategy, the calculation tasks of each of the above-mentioned power equipment are scheduled and processed to obtain the corresponding task scheduling results.

[0048] Once the cloud-based system obtains the aforementioned failure probabilities, it can combine them with a target scheduling strategy to schedule the computational tasks of each power device, obtaining the corresponding task scheduling results. The target scheduling strategy can be related to the operating cost, power supply reliability, and energy consumption optimization of the power devices. For example, a scheduling strategy aimed at minimizing the operating cost, maximizing power supply reliability, and minimizing energy consumption of the power devices performing computational tasks. The aforementioned scheduling process can involve offloading the computational tasks from the power devices, such as assigning them to other power devices for execution, thereby satisfying the target scheduling strategy and scheduling objectives.

[0049] In the aforementioned task processing method for switchgear based on a cloud-edge-device architecture, the switchgear on the terminal side collects operational and environmental data from various power devices. The edge device extracts operational feature information based on this data. This operational feature information is then input into a fault probability prediction model via the cloud. The model outputs the fault probability of each power device based on the operational feature information. Based on the fault probability and a target scheduling strategy, the computational tasks for each power device are scheduled, yielding the corresponding task scheduling results. Compared to traditional task scheduling methods relying on manual inspection, this application combines the switchgear on the terminal side, the edge device, and the cloud to achieve data extraction, feature extraction, and computational task scheduling. This allows edge nodes to complete most of the task scheduling steps, improving task processing efficiency.

[0050] In one embodiment, the computational tasks of each of the aforementioned power devices are scheduled according to the aforementioned fault probabilities and target scheduling strategies to obtain corresponding task scheduling results, including: determining candidate power devices whose fault probabilities are greater than a probability threshold based on the fault probabilities corresponding to each of the aforementioned power devices; determining target power devices that satisfy the aforementioned first scheduling strategy and the aforementioned second scheduling strategy from among the aforementioned candidate power devices; and performing task offloading processing on the computational tasks of the aforementioned target power devices to obtain task scheduling results.

[0051] In this embodiment, the target scheduling strategy includes a first scheduling strategy and a second scheduling strategy. The first scheduling strategy represents a scheduling strategy based on the operating cost and power supply reliability of each of the aforementioned power devices; the second scheduling strategy represents a scheduling strategy based on the energy consumption of the cloud, the edge, and the terminal. The cloud can combine the first scheduling strategy, the second scheduling strategy, and the failure probability to implement a scheduling strategy for each power device, thereby obtaining the corresponding scheduling processing result.

[0052] The cloud platform can identify candidate power devices whose failure probability exceeds a certain threshold based on the failure probability of each individual power device. These candidate devices are those predicted to be prone to failure based on operational characteristic information. The cloud platform prioritizes scheduling tasks for these candidate power devices.

[0053] Based on the first and second scheduling strategies mentioned above, the cloud can filter the candidate power devices to determine the target power devices that satisfy both strategies. The cloud can then offload the computational tasks of these target power devices, obtaining the task scheduling results. In other words, the cloud can identify the target power devices from the candidate power devices that, after task offloading, minimize the operating costs, power supply reliability, and energy consumption of the cloud, the edge devices, and the terminal devices.

[0054] Through this embodiment, the cloud can combine the fault probability, the first scheduling strategy, and the second scheduling strategy to determine the target power equipment from the candidate power equipment for task scheduling, thereby improving the accuracy of task scheduling processing.

[0055] In one embodiment, determining a target power device that satisfies the first scheduling strategy and the second scheduling strategy from among the candidate power devices includes: determining a first power device from among the candidate power devices according to the first scheduling strategy; the first power device represents a candidate power device whose corresponding task scheduling processing result minimizes the operating cost and maximizes the power supply reliability; determining a second power device from among the candidate power devices according to the second scheduling strategy; the second power device represents a candidate power device whose corresponding task scheduling processing result minimizes the energy consumption; and determining the target power device based on the intersection of the first power devices and the second power devices.

[0056] In this embodiment, the target scheduling strategy includes multiple scheduling strategies. For example, it may include a first scheduling strategy and a second scheduling strategy. Different scheduling strategies have different scheduling objectives. The cloud can determine a first power device from among the candidate power devices based on the first scheduling strategy. The first power device represents the candidate power device whose corresponding task scheduling processing result minimizes the operating cost and maximizes the power supply reliability. Maximum power supply reliability indicates the shortest power outage time for the power device.

[0057] The cloud can also determine a second power device from among the candidate power devices according to the second scheduling strategy described above. The second power device represents the candidate power device whose energy consumption is lowest after the corresponding task scheduling processing result. Therefore, the cloud can determine the target power device based on the intersection of the first and second power devices. That is, the cloud determines the power device that simultaneously satisfies the scheduling objectives of the first and second scheduling strategies from the first and second power devices, and uses it as the target power device. The target power device represents the power device that is prone to failure and, after scheduling, can satisfy the scheduling objectives of the aforementioned target scheduling strategies (the first and second scheduling strategies).

[0058] Through this embodiment, the cloud can combine the scheduling objectives of the first scheduling strategy and the scheduling objectives of the second scheduling strategy to determine the target power equipment from among the candidate power equipment, and perform task scheduling processing on the target power equipment, thereby improving the accuracy of task scheduling, reducing operating costs, improving power supply reliability, and reducing system energy consumption.

[0059] In one embodiment, determining the first power device from among the candidate power devices according to the first scheduling strategy includes: obtaining the simulation task scheduling results corresponding to each candidate power device; determining the simulation operating cost and simulation outage time corresponding to each power device based on the simulation task scheduling results; the simulation outage time characterizing the power supply reliability of the power device; obtaining the first target simulation task scheduling result with the minimum simulation operating cost and the shortest simulation outage time; and determining the first power device based on each candidate power device corresponding to the first target simulation task scheduling result.

[0060] In this embodiment, to determine the first power device that satisfies the first scheduling strategy, the cloud can pre-simulate task scheduling for candidate power devices. For example, it can simulate unloading the computational tasks of candidate power devices to obtain the simulation task scheduling results for each candidate power device. The cloud can then determine the simulated operating cost and simulated power outage time for each power device based on these simulation task scheduling results. Here, the simulated power outage time characterizes the power supply reliability of the power device, the simulated power outage time is the power outage time corresponding to the simulated task scheduling, and the simulated operating cost represents the operating cost corresponding to the simulated task scheduling. The cloud can obtain the first objective simulation task scheduling result that minimizes the simulated operating cost and minimizes the simulated power outage time, and determine the first power device based on the candidate power devices corresponding to the first objective simulation task scheduling result. That is, task scheduling of the first power device can satisfy the scheduling objective of minimizing operating cost and minimizing power outage time.

[0061] Specifically, the first scheduling strategy is also known as the global scheduling objective (minimizing operating costs and maximizing power supply reliability), which can be expressed as:

[0062] .

[0063] This formula represents minimizing the control variable (such as the number of power devices requiring task scheduling) u, such that the operating cost C at each time from t=1 to T is minimized. gen (t) and energy loss E loss The sum of the weighted sums of (t) (with weighting coefficients λ) is minimized. Among them, the energy loss E loss (t) can be determined based on the power outage time of the power equipment, representing the power supply reliability. The aforementioned global scheduling objective can be output under constraints. These constraints include power balance constraints for each power device, equipment capacity constraints for each power device, and line thermal limits, etc.

[0064] Through this embodiment, the cloud can combine global scheduling objectives to determine the first power device from various candidate power devices, thereby improving the accuracy of determining the first power device.

[0065] In one embodiment, determining the second power device from among the candidate power devices according to the second scheduling strategy includes: obtaining the simulation task scheduling processing results corresponding to each candidate power device; determining the first energy consumption corresponding to the cloud, the second energy consumption corresponding to the edge, the third energy consumption corresponding to the terminal, and target latency information based on the simulation task scheduling processing results; the target latency information characterizing the latency information when transmitting data between the cloud, the edge, and the terminal; obtaining a second target simulation task scheduling processing result that minimizes the first energy consumption, the second energy consumption, the third energy consumption, and the target latency information; and determining the second power device based on each candidate power device corresponding to the second target simulation task scheduling processing result.

[0066] In this embodiment, to determine the second power device that satisfies the second scheduling strategy, the cloud can pre-simulate task scheduling for candidate power devices, for example, by simulating the offloading of computational tasks from the candidate power devices, and obtain the simulation task scheduling processing results corresponding to each of the aforementioned candidate power devices. Based on the simulation task scheduling processing results, the cloud can determine the first energy consumption corresponding to the cloud, the second energy consumption corresponding to the edge device, the third energy consumption corresponding to the terminal device, and the target latency information. The first energy consumption represents the energy consumption of the cloud during the simulation task scheduling process, the second energy consumption represents the energy consumption of the edge device during the simulation task scheduling process, and the target latency information represents the latency information when transmitting data between the cloud, the edge device, and the terminal device. Based on the simulation task scheduling processing results, the cloud can determine the second target simulation task scheduling processing result that minimizes the first energy consumption, the second energy consumption, the third energy consumption, and the target latency information. Therefore, the cloud can determine the second power device based on each candidate power device corresponding to the second target simulation task scheduling processing result.

[0067] Specifically, the second scheduling strategy can be a scheduling strategy based on task offloading and energy consumption optimization. The cloud can construct the energy consumption model corresponding to the aforementioned cloud-edge-device system. Therefore, when scheduling tasks based on the second scheduling strategy, the total energy consumption of the system can be expressed as: E total =E cloud +E edge +E end E cloud =P cloud T cloud E edge =P edge T edge E end =P end T end Among them, E cloud E edge E end These are the energy consumption figures for the cloud, edge, and terminal sides, respectively, representing the first, second, and third energy consumption figures; P cloud P edge P end These represent the power consumption at the cloud, edge, and terminal sides, respectively; T cloud T edge T end These represent the actual runtime for the cloud, edge, and terminal sides, respectively. Among them, T... cloud =C cloud / f cloud ;T edge =C edge / f edge;T end =C end / f end Among them, C cloud C edge C end These represent the number of central processing unit cycles for the cloud, edge, and terminal sides, respectively; f cloud f edge f end These are the processor clock speeds for the cloud, edge, and terminal sides, respectively.

[0068] In some embodiments, the total energy consumption of the system described above may further include transmission energy consumption. Transmission energy consumption can be expressed as E tx =(P tx +P rx (D / η) link ), where E tx P represents the total energy consumed during the transmission process. tx This indicates the power of the transmitting device (such as the terminal, edge, or cloud). rx This represents the power of the receiving device (such as various power equipment). D represents the amount of data transmitted. η link It represents the transmission efficiency of a communication link, which indicates the amount of data that can be transmitted per unit of energy. The higher the efficiency, the less energy is required to transmit the same amount of data.

[0069] When scheduling tasks using the second scheduling strategy described above, the cloud can solve for the task offloading strategy using Particle Swarm Optimization (PSO) or reinforcement learning, thereby determining each second power device. The solution function can be expressed as: minE total stL total ≤L max Among them, E total L represents total energy consumption. total The overall latency (target latency information, accumulated from the cloud, edge, and terminal) represents the total time elapsed from the initiation of a scheduled task to its completion, typically including computation and communication latency; L max This represents the maximum permissible delay. The formula represents minimizing total energy consumption under the constraint that the total delay does not exceed the maximum permissible delay.

[0070] Through this embodiment, the cloud can combine the second scheduling strategy to determine the second power equipment that meets the real-time control requirements and energy consumption requirements, thereby improving the accuracy of determining the power equipment that needs to be scheduled.

[0071] In one exemplary embodiment, such as Figure 3 As shown, a task processing method for switchgear based on a cloud-edge-device architecture is provided, which is then applied to... Figure 1 Taking the edge end of the process as an example, the steps include S302 to S306. Wherein:

[0072] Step S302: Obtain the corresponding operating data and environmental data on the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect the operating data and environmental data of each power device on the terminal side.

[0073] The cloud, edge, and terminal-side switchgear form a cloud-edge-device collaborative network. The terminal-side switchgear incorporates various sensors, enabling it to collect operational and environmental data from various power devices on the terminal side. The edge devices can then access this operational and environmental data collected by the terminal-side switchgear.

[0074] Step S304: Extract the corresponding operational feature information based on the above operational data and the above environmental data.

[0075] The edge device can extract operational feature information based on the operational and environmental data from the terminal side. This operational feature information can be the characteristics corresponding to the operation of various power devices on the terminal side.

[0076] Step S306: The above-mentioned operational feature information is sent to the cloud; the cloud is used to input the above-mentioned operational feature information into the fault probability prediction model; the fault probability prediction model is used to output the fault probability corresponding to each of the above-mentioned power equipment based on the above-mentioned operational feature information; and the calculation tasks of each of the above-mentioned power equipment are scheduled according to the above-mentioned fault probability and the target scheduling strategy to obtain the corresponding task scheduling results.

[0077] The edge device can send extracted features to the cloud. The cloud can then obtain the corresponding operational feature information from the terminal. A fault probability prediction model is deployed in the cloud to predict the fault probability of each power device. Fault probability represents the probability of a power device failing. The cloud can input the aforementioned operational feature information into the fault probability prediction model. Based on the operational feature information, the fault probability prediction model can predict the fault probability of each power device and output the corresponding fault probability for each device. After obtaining the fault probabilities, the cloud can combine them with a target scheduling strategy to schedule the computational tasks of each power device, obtaining the corresponding task scheduling results.

[0078] In the aforementioned task processing method for switchgear based on a cloud-edge-device architecture, the switchgear on the terminal side collects operational and environmental data from various power devices. The edge device extracts operational feature information based on this data. This operational feature information is then input into a fault probability prediction model via the cloud. The model outputs the fault probability of each power device based on the operational feature information. Based on the fault probability and a target scheduling strategy, the computational tasks for each power device are scheduled, yielding the corresponding task scheduling results. Compared to traditional task scheduling methods relying on manual inspection, this application combines the switchgear on the terminal side, the edge device, and the cloud to achieve data extraction, feature extraction, and computational task scheduling. This allows edge nodes to complete most of the task scheduling steps, improving task processing efficiency.

[0079] In one embodiment, after extracting the corresponding operational feature information based on the above-mentioned operational data and environmental data, the method further includes: performing target fault detection on each of the above-mentioned power devices corresponding to the terminal side based on the above-mentioned operational feature information; the target fault detection includes partial discharge anomaly detection and temperature and humidity trend detection; if, based on the above-mentioned target fault detection, it is determined that there is an abnormal power device with partial discharge anomaly or temperature and humidity anomaly among the above-mentioned power devices, then the switch of the abnormal power device is cut off.

[0080] In this embodiment, the edge device can perform lightweight anomaly detection on the power equipment. After extracting operational feature information, the edge device can perform target fault detection on each of the aforementioned power devices corresponding to the terminal side based on this information. This target fault detection includes partial discharge anomaly detection and temperature and humidity trend detection, i.e., lightweight anomaly detection. If, based on the target fault detection, it is determined that there is an abnormal power device with partial discharge or temperature and humidity anomalies, the edge device can immediately disconnect the switch of the abnormal power device.

[0081] Specifically, lightweight models are deployed, enabling local control. For data aggregation, the edge can perform time synchronization and noise filtering on sampled data from multiple switchgear units at the same site. For lightweight models, convolutional neural networks, long short-term memory networks, or lightweight transformer models (Transformer Lite) are deployed at the edge to detect partial discharge anomalies in power equipment and predict temperature and humidity trends (target fault detection). For local control, when a high-risk anomaly is detected at the edge, it can directly trigger the load switch fuse (cutting off the switch of the abnormal power equipment), achieving millisecond-level tripping.

[0082] Through this embodiment, the edge device can perform lightweight power equipment anomaly detection locally, thereby promptly cutting off the power equipment switch when the corresponding anomaly occurs, improving the processing efficiency of anomaly detection.

[0083] In one exemplary embodiment, such as Figure 4 As shown, Figure 4 This is a schematic diagram of a task processing system for a switchgear based on a cloud-edge-device architecture in one embodiment. In this embodiment, the cloud includes a cloud server and can also perform detection and control through software programs on mobile devices; the edge includes an intelligent controller, a low-voltage intelligent cabinet, a gateway, an electronic display screen, and a human-machine integrated monitoring interface; the switchgear on the terminal side includes various types of equipment, such as frame circuit breakers, water immersion protectors, noise detectors, molded case circuit breakers, temperature and humidity controllers, multi-functional instruments, smoke detectors, video monitoring, temperature sensors, wireless temperature measurement devices, dust collectors, etc.

[0084] The system consists of three layers: a cloud platform (cloud), edge computing nodes (edge), and terminal-side intelligent switch cabinets (terminal), forming a cloud-edge-terminal collaborative network.

[0085] The cloud platform deploys big data storage, deep learning models, and global scheduling algorithms, and is responsible for cross-site load prediction, optimal routing, and policy distribution.

[0086] Edge nodes are located inside distribution stations or substations and have the acceleration capabilities of Graphics Processing Unit (GPU) / Field-Programmable Gate Array (FPGA). They are responsible for data preprocessing, feature extraction, lightweight model inference, and can perform real-time control such as emergency tripping locally.

[0087] The end-side switchgear has built-in multiple sensors (current / voltage, temperature and humidity, partial discharge, gas, infrared, etc.), a main control unit, and a communication module. It can perform early exit reasoning locally (such as reasoning whether the power equipment is abnormal). When the confidence level is insufficient, it reports the raw or feature data to the edge or cloud.

[0088] The intelligent switch cabinet on the terminal side includes a sensor subsystem, a main control unit, a communication module, and security mechanisms. The sensor subsystem includes multi-modal sensors for current / voltage, gas, temperature and humidity, infrared, light, and smoke, used to collect operational and environmental data. The main control unit adopts a dual-core architecture of Advanced RISC Machine Architecture (ARM) and Digital Signal Processor (DSP), supporting both IEC61850 and Modbus / Transmission Control Protocol (TCP) dual-protocol stacks. The communication module supports dual channels for 4G / 5G, Narrow Band Internet of Things (NB-IoT), and edge gateways, ensuring reliable transmission of critical commands. The security mechanism achieves data confidentiality based on Transport Layer Security (TLS), Pre-Shared Key (PSK), and quantum key negotiation (referencing the low-voltage distribution area quantum encrypted communication patent). When the switchgear on the terminal side collects the aforementioned operational and environmental data, its acquisition rate can be dynamically adjusted based on the power fluctuation frequency of the power equipment. After collecting the aforementioned data, the switchgear on the terminal side can use wavelet denoising Kalman filtering to improve the signal quality of the collected operational and environmental data.

[0089] Edge computing nodes can be used for data aggregation, deployed with lightweight models, and also enable local control. For data aggregation, the edge can perform time synchronization and noise filtering on sampled data from multiple switchgear units at the same site. For lightweight models, the edge deploys Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, or TransformerLite models to detect partial discharge anomalies in power equipment and predict temperature and humidity trends. For local control, when the edge detects a high-risk anomaly, it can directly trigger the load switch fuse, achieving millisecond-level tripping.

[0090] The cloud platform comprises a data lake, a model center, and a scheduling engine. The data lake centrally stores historical data for all switchgear, supporting a hybrid approach of time-series databases and object storage. The model center utilizes deep learning-based models for load forecasting, fault probability estimation, and optimal scheduling (multi-objective optimization). The scheduling engine employs cloud, edge, and endpoint task offloading optimization, making dynamic decisions based on computational energy consumption and network latency.

[0091] The failure probability model can be expressed as: P fault =(w T (x+b). Where, P fault This represents the predicted probability of a fault occurring, with values ​​ranging from (0,1). A value closer to 1 indicates a higher probability of a fault occurring. σ represents the Sigmoid activation function. T This represents the transpose of the weight vector. x represents the input operational features. b represents the bias term (threshold), used to adjust the classification decision boundary, enabling the model to fit more complex data distributions.

[0092] The global scheduling target can be specifically represented as:

[0093] .

[0094] This formula represents minimizing the control variable (such as the number of power devices requiring task scheduling) u, such that the operating cost C at each time from t=1 to T is minimized. gen (t) and energy loss E loss The sum of the weighted sums of (t) (with weighting coefficients λ) is minimized. Among them, the energy loss E loss (t) can be determined based on the power outage time of the power equipment, representing the power supply reliability. The aforementioned global scheduling objective can be output under constraints. These constraints include power balance constraints for each power device, equipment capacity constraints for each power device, and line thermal limits, etc.

[0095] If the cloud can construct the energy consumption model corresponding to the above cloud-edge-device system, then when scheduling tasks based on the second scheduling strategy, the total energy consumption of the system can be expressed as: E total =E cloud +E edge +E end E cloud =P cloud T cloud E edge =P edge T edge E end =P end T end Among them, E cloud E edge E end These are the energy consumption figures for the cloud, edge, and terminal sides, respectively, representing the first, second, and third energy consumption figures; P cloud P edge P end These represent the power consumption at the cloud, edge, and terminal sides, respectively; T cloud T edge T endThese represent the actual runtime for the cloud, edge, and terminal sides, respectively. Among them, T... cloud =C cloud / f cloud ;T edge =C edge / f edge ;T end =C end / f end Among them, C cloud C edge C end These represent the number of central processing unit cycles for the cloud, edge, and terminal sides, respectively; f cloud f edge f end These are the processor clock speeds for the cloud, edge, and terminal sides, respectively.

[0096] In some embodiments, the total energy consumption of the system described above may further include transmission energy consumption. Transmission energy consumption can be expressed as E tx =(P tx +P rx (D / η) link ), where E tx P represents the total energy consumed during the transmission process. tx This indicates the power of the transmitting device (such as the terminal, edge, or cloud). rx This represents the power of the receiving device (such as various power equipment). D represents the amount of data transmitted. η link It represents the transmission efficiency of a communication link, which indicates the amount of data that can be transmitted per unit of energy. The higher the efficiency, the less energy is required to transmit the same amount of data.

[0097] When scheduling tasks using the second scheduling strategy described above, the cloud can solve for the task offloading strategy using Particle Swarm Optimization (PSO) or reinforcement learning, thereby determining each second power device. The solution function can be expressed as: minE total stL total ≤L max Among them, E total L represents total energy consumption. total The overall latency (target latency information, accumulated from the cloud, edge, and terminal) represents the total time elapsed from the initiation of a scheduled task to its completion, typically including computation and communication latency; L max This represents the maximum permissible delay. The formula represents minimizing total energy consumption under the constraint that the total delay does not exceed the maximum permissible delay.

[0098] Through the above embodiments, by combining the switchgear on the terminal side, the edge terminal, and the cloud, data extraction, feature extraction, and computational task scheduling are achieved. This allows edge nodes to complete most of the task scheduling steps, improving task processing efficiency. Furthermore, it offers the following benefits: edge nodes complete most anomaly detection, ensuring timely execution of critical control commands and significantly reducing latency; dynamic offloading based on the aforementioned energy consumption model reduces overall system energy consumption, achieving energy optimization; multi-layer redundancy and local fast tripping mechanisms reduce power outage time caused by single-point failures, improving power supply reliability; and adding switchgear only requires deploying sensors and communication modules on the terminal side, with the edge / cloud platform automatically sensing and incorporating it into unified scheduling, improving scalability.

[0099] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0100] Based on the same inventive concept, this application also provides a task processing device for a switchgear based on a cloud-edge-device architecture for implementing the task processing method for switchgear based on a cloud-edge-device architecture described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the task processing device for a switchgear based on a cloud-edge-device architecture provided below can be found in the limitations of the task processing method for switchgear based on a cloud-edge-device architecture described above, and will not be repeated here.

[0101] In one exemplary embodiment, such as Figure 5 As shown, a task processing device for a switchgear based on a cloud-edge-device architecture is provided, comprising: a first acquisition module 500, a prediction module 502, and a processing module 504, wherein:

[0102] The first acquisition module 500 is used to acquire the corresponding operating characteristic information of the terminal side; the aforementioned operating characteristic information is extracted by the edge end based on the corresponding operating data and environmental data of the terminal side; the aforementioned terminal side includes a switch cabinet; the aforementioned switch cabinet is used to collect the operating data and environmental data of each power device on the aforementioned terminal side.

[0103] The prediction module 502 is used to input the above-mentioned operating characteristic information into the fault probability prediction model; the above-mentioned fault probability prediction model is used to output the fault probability corresponding to each of the above-mentioned power equipment based on the above-mentioned operating characteristic information.

[0104] The processing module 504 is used to schedule the calculation tasks of each of the above-mentioned power equipment according to the above-mentioned fault probability and target scheduling strategy, and obtain the corresponding task scheduling results.

[0105] In one embodiment, the processing module 504 is configured to determine candidate power devices whose failure probabilities are greater than a probability threshold based on the failure probabilities corresponding to each of the power devices; determine target power devices that satisfy the first scheduling strategy and the second scheduling strategy from among the candidate power devices; and perform task offloading processing on the computational tasks of the target power devices to obtain task scheduling processing results.

[0106] In one embodiment, the processing module 504 is configured to: determine a first power device from among the candidate power devices according to the first scheduling strategy; the first power device represents the candidate power device whose corresponding task scheduling processing result minimizes the operating cost and maximizes the power supply reliability; determine a second power device from among the candidate power devices according to the second scheduling strategy; the second power device represents the candidate power device whose corresponding task scheduling processing result minimizes the energy consumption; and determine a target power device based on the intersection of the first power devices and the second power devices.

[0107] In one embodiment, the processing module 504 is configured to: obtain the simulation task scheduling processing results corresponding to each of the candidate power devices; determine the simulation operating cost and simulation outage time corresponding to each of the power devices based on the simulation task scheduling processing results; the simulation outage time characterizes the power supply reliability of the power devices; obtain the first target simulation task scheduling processing result with the minimum simulation operating cost and the shortest simulation outage time; and determine the first power device based on each candidate power device corresponding to the first target simulation task scheduling processing result.

[0108] In one embodiment, the processing module 504 is configured to: obtain the simulation task scheduling processing results corresponding to each of the candidate power devices; determine the first energy consumption corresponding to the cloud, the second energy consumption corresponding to the edge, the third energy consumption corresponding to the terminal, and target latency information based on the simulation task scheduling processing results; the target latency information characterizes the latency information when transmitting data between the cloud, the edge, and the terminal; obtain a second target simulation task scheduling processing result that minimizes the first energy consumption, the second energy consumption, the third energy consumption, and the target latency information; and determine the second power device based on each candidate power device corresponding to the second target simulation task scheduling processing result.

[0109] In one exemplary embodiment, such as Figure 6 As shown, a task processing device for a switchgear based on a cloud-edge-device architecture is provided, comprising: a second acquisition module 600, an extraction module 602, and a sending module 604, wherein:

[0110] The second acquisition module 600 is used to acquire the corresponding operating data and environmental data on the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect the operating data and environmental data of each power device on the terminal side.

[0111] The extraction module 602 is used to extract corresponding operational feature information based on the above-mentioned operational data and environmental data.

[0112] The sending module 604 is used to send the above-mentioned operating characteristic information to the cloud; the cloud is used to input the above-mentioned operating characteristic information into the fault probability prediction model; the fault probability prediction model is used to output the fault probability corresponding to each of the above-mentioned power equipment according to the above-mentioned operating characteristic information; and to schedule the calculation tasks of each of the above-mentioned power equipment according to the above-mentioned fault probability and the target scheduling strategy to obtain the corresponding task scheduling results.

[0113] In one embodiment, the above-mentioned device further includes: an anomaly detection module, used to perform target fault detection on each of the above-mentioned power devices corresponding to the above-mentioned terminal side based on the above-mentioned operating characteristic information; the target fault detection includes partial discharge anomaly detection and temperature and humidity trend detection; if it is determined based on the above-mentioned target fault detection that there is an abnormal power device with partial discharge anomaly or temperature and humidity anomaly among the above-mentioned power devices, then the switch of the abnormal power device is cut off.

[0114] The various modules in the task processing device of the switchgear based on the cloud-edge-device architecture described above can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0115] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores XX data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a task processing method for a switchgear based on a cloud-edge-device architecture.

[0116] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0117] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described task processing method for a switchgear based on a cloud-edge-device architecture.

[0118] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the task processing method for a switchgear based on a cloud-edge-device architecture as described above.

[0119] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the above-described task processing method for a switchgear based on a cloud-edge-device architecture.

[0120] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0121] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0122] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0123] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A task processing method for a switchgear based on a cloud-edge-device architecture, characterized in that, Applied to the cloud, the method includes: The system acquires operational characteristic information corresponding to the terminal side; the operational characteristic information is extracted by the edge device based on the operational data and environmental data corresponding to the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect operational data and environmental data of various power devices on the terminal side; The operational characteristic information is input into the fault probability prediction model; the fault probability prediction model is used to output the fault probability corresponding to each of the power devices based on the operational characteristic information. Based on the fault probability and target scheduling strategy, the computational tasks of each power device are scheduled and processed to obtain the corresponding task scheduling results.

2. The method according to claim 1, characterized in that, The target scheduling strategy includes a first scheduling strategy and a second scheduling strategy; the first scheduling strategy represents a scheduling strategy based on the operating cost and power supply reliability of each of the power devices; the second scheduling strategy represents a scheduling strategy based on the energy consumption of the cloud, the edge, and the terminal. The step of scheduling the computational tasks of each of the power devices according to the fault probability and the target scheduling strategy to obtain the corresponding task scheduling results includes: Based on the failure probability corresponding to each of the power devices, candidate power devices whose failure probability is greater than a probability threshold are determined; From the candidate power equipment, target power equipment that satisfies the first scheduling strategy and the second scheduling strategy is determined; The computational tasks of the target power equipment are offloaded to obtain the task scheduling results.

3. The method according to claim 2, characterized in that, The step of determining the target power equipment that satisfies the first scheduling strategy and the second scheduling strategy from the candidate power equipment includes: According to the first scheduling strategy, a first power device is determined from each of the candidate power devices; the first power device represents the candidate power device whose corresponding task scheduling processing result minimizes the operating cost and maximizes the power supply reliability. According to the second scheduling strategy, a second power device is determined from each of the candidate power devices; the second power device represents the candidate power device with the lowest energy consumption as a result of the corresponding task scheduling process. The target power equipment is determined based on the intersection of each of the first power equipment and each of the second power equipment.

4. The method according to claim 3, characterized in that, The step of determining the first power device from among the candidate power devices according to the first scheduling strategy includes: Obtain the simulation task scheduling and processing results corresponding to each of the candidate power devices; Based on the simulation task scheduling results, the simulated operating cost and simulated power outage time for each of the power devices are determined; the simulated power outage time characterizes the power supply reliability of the power devices. Obtain the scheduling and processing result of the first objective simulation task that minimizes the simulation running cost and the simulation power outage time; The first power device is determined based on the candidate power devices corresponding to the first target simulation task scheduling processing results.

5. The method according to claim 3, characterized in that, The step of determining the second power device from among the candidate power devices according to the second scheduling strategy includes: Obtain the simulation task scheduling and processing results corresponding to each of the candidate power devices; Based on the simulated task scheduling and processing results, the first energy consumption corresponding to the cloud, the second energy consumption corresponding to the edge, the third energy consumption corresponding to the terminal, and the target latency information are determined; the target latency information represents the latency information when transmitting data between the cloud, the edge, and the terminal. Obtain the second target simulation task scheduling processing result that minimizes the first energy consumption, the second energy consumption, the third energy consumption, and the target latency information; The second power device is determined based on the candidate power devices corresponding to the second target simulation task scheduling processing results.

6. A task processing method for a switchgear based on a cloud-edge-device architecture, characterized in that, Applied to the edge, the method includes: The system acquires corresponding operational and environmental data from the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect operational and environmental data from various power devices on the terminal side. Based on the operational data and the environmental data, extract the corresponding operational feature information; The operational feature information is sent to the cloud; the cloud is used to input the operational feature information into the fault probability prediction model; the fault probability prediction model is used to output the fault probability corresponding to each of the power devices according to the operational feature information; and to schedule the computational tasks of each of the power devices according to the fault probability and the target scheduling strategy to obtain the corresponding task scheduling results.

7. The method according to claim 6, characterized in that, After extracting the corresponding operational feature information based on the operational data and the environmental data, the process further includes: Based on the operational characteristic information, target fault detection is performed on each of the power devices corresponding to the terminal side; the target fault detection includes partial discharge anomaly detection and temperature and humidity trend detection. If, based on the target fault detection, it is determined that there is an abnormal power device among the various power devices with partial discharge or abnormal temperature and humidity, then the switch of the abnormal power device is cut off.

8. A task processing device for a switchgear based on a cloud-edge-device architecture, characterized in that, The device, applied in the cloud, includes: The first acquisition module is used to acquire the corresponding operating characteristic information of the terminal side; the operating characteristic information is extracted by the edge terminal based on the operating data and environmental data corresponding to the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect the operating data and environmental data of each power device on the terminal side; The prediction module is used to input the operating characteristic information into the fault probability prediction model; the fault probability prediction model is used to output the fault probability corresponding to each of the power devices based on the operating characteristic information. The processing module is used to schedule the computational tasks of each of the power devices according to the fault probability and the target scheduling strategy, and obtain the corresponding task scheduling results.

9. A task processing device for a switchgear based on a cloud-edge-device architecture, characterized in that, Applied to the edge, the device includes: The second acquisition module is used to acquire the corresponding operating data and environmental data on the terminal side; the terminal side includes a switch cabinet; the switch cabinet is used to collect the operating data and environmental data of each power device on the terminal side; The extraction module is used to extract corresponding operational feature information based on the operational data and the environmental data; The sending module is used to send the operating feature information to the cloud; the cloud is used to input the operating feature information into the fault probability prediction model; the fault probability prediction model is used to output the fault probability corresponding to each of the power devices according to the operating feature information; and to schedule the calculation tasks of each of the power devices according to the fault probability and the target scheduling strategy to obtain the corresponding task scheduling results.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

Images