Resource collaborative scheduling method and device, computer equipment, medium and program product

Abstract

The invention relates to a resource collaborative scheduling method and device, computer equipment, a medium and a program product, and relates to the technical field of power grids. The resource collaborative scheduling method comprises the following steps: acquiring a plurality of operation task requests sent by a distributed subsystem in a power grid system and resource state information of the distributed subsystem; determining task types of corresponding operation tasks according to the operation task requests; and according to each task type and the resource state information, cooperatively scheduling different equipment resources in the distributed subsystem to execute each operation task so as to monitor a power subsystem in the power grid system. The method can improve the resource utilization rate of the power grid system.

Description

Technical Field

[0001] The present application relates to the field of power grid technology, and in particular to a resource collaborative scheduling method, device, computer equipment, medium and program product. Background Art

[0002] With the development of power grid technology, the demand for intelligent and efficient management of power grid system operation is increasing.

[0003] The power grid system is usually equipped with a power subsystem, which supplies power through multiple power devices in the power subsystem. In addition, the power grid system is also equipped with other supporting equipment, which monitors the power subsystem to ensure the normal and safe operation of the power grid system.

[0004] However, in the process of using supporting equipment to monitor the power subsystem, the traditional power grid system is unable to reasonably allocate various equipment resources, resulting in low resource utilization of the power grid system. Summary of the invention

[0005] Based on this, it is necessary to provide a resource collaborative scheduling method, device, computer equipment, medium and program product to address the above technical problems, which can improve the resource utilization of the power grid system.

[0006] In a first aspect, the present application provides a resource collaborative scheduling method, the method comprising:

[0007] Acquire multiple running task requests sent by distributed subsystems in the power grid system, and resource status information of the distributed subsystems;

[0008] Determine the task type of the corresponding running task according to each of the running task requests;

[0009] According to each of the task types and the resource status information, different equipment resources in the distributed subsystem are coordinated to execute each of the operation tasks, so as to monitor the power subsystem in the power grid system.

[0010] In one embodiment, the collaboratively scheduling different device resources in the distributed subsystem to execute each of the running tasks according to each of the task types and the resource status information includes:

[0011] Determining the priority of each of the running tasks according to each of the task types;

[0012] Different device resources in the distributed subsystem are collaboratively scheduled to execute each of the running tasks according to the priority of each of the running tasks and the resource status information.

[0013] In one embodiment, the task type includes at least one of a data collection task, an analysis and decision-making task, and a model training task; wherein the priority of the data collection task is higher than the analysis and decision-making task and the model training task, respectively, and the priority of the analysis and decision-making task is higher than the model training task.

[0014] In one embodiment, the collaboratively scheduling different device resources in the distributed subsystem to execute each of the running tasks according to each of the task types and the resource status information includes:

[0015] In the case where the task type includes a data acquisition task, determining a target device end in an idle state from the distributed subsystem according to the resource status information;

[0016] A data collection instruction is sent to the target device end to instruct the target device end to collect the operation perception data of the power subsystem according to the data collection instruction, and send the operation perception data to the edge node in the distributed subsystem.

[0017] In one embodiment, the collaboratively scheduling different device resources in the distributed subsystem to execute each of the running tasks according to each of the task types and the resource status information includes:

[0018] In the case where the task type includes an analysis and decision-making task, determining a target edge node in an idle state from the distributed subsystem according to the resource state information;

[0019] An analysis and decision instruction is sent to the target edge node to instruct the target edge node to make an analysis and decision based on the operation perception data sent by the device end in the distributed subsystem, generate an operation perception result of the power subsystem, and send the operation perception data and the operation perception result to the cloud server in the distributed subsystem; the operation perception result is used to represent the operation scenario of the power grid system.

[0020] In one embodiment, the collaboratively scheduling different device resources in the distributed subsystem to execute each of the running tasks according to each of the task types and the resource status information includes:

[0021] In a case where the task type includes a model training task, determining a target cloud server in an idle state from the distributed subsystem according to the resource status information;

[0022] Send a model training instruction to the target cloud server to instruct the target cloud server to perform model training according to the operation perception data and operation perception results sent by the edge node in the distributed subsystem, generate a real-time perception model of the power grid system, and send model update data to the edge node in the distributed subsystem; the real-time perception model is used to monitor the power grid system in real time, and the model update data includes at least one of the real-time perception model and the incremental data corresponding to the real-time perception model.

[0023] In a second aspect, the present application provides a resource collaborative scheduling device, which is applied to a collaborative scheduling system, and the device includes:

[0024] An acquisition module, used to acquire multiple operation task requests sent by a distributed subsystem in a power grid system, and resource status information of the distributed subsystem;

[0025] A determination module, used to determine the task type of the corresponding running task according to each of the running task requests;

[0026] The scheduling module is used to coordinately schedule different equipment resources in the distributed subsystem to perform each of the operating tasks according to each of the task types and the resource status information, so as to monitor the power subsystem in the power grid system.

[0027] In a third aspect, the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0028] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0029] In a fifth aspect, the present application provides a computer program product, including a computer program, which implements the steps of the above method when executed by a processor.

[0030] The above-mentioned resource collaborative scheduling method, device, computer equipment, medium and program product obtain multiple operation task requests sent by distributed subsystems in the power grid system, as well as resource status information of the distributed subsystems, determine the task type of the corresponding operation task according to each operation task request, and collaboratively schedule different equipment resources in the distributed subsystem to perform each operation task according to each task type and resource status information, so as to monitor the power subsystem in the power grid system. The method can dynamically adjust the equipment resource allocation of the distributed subsystem according to the task type and resource status information of each operation task, realize the cloud-edge-end resource collaborative scheduling for the integration of sensing, transmission and computing, avoid resource waste, ensure that each task can be executed in the most efficient way, and can reduce the task queuing waiting time through reasonable task scheduling, speed up the task processing speed, and improve the system resource utilization and task processing efficiency; in addition, the method monitors the resource status information in real time, and once an abnormality or fault is detected, it can quickly coordinate and schedule other available resources for replacement, reduce system downtime, improve system reliability, and improve system response speed; in addition, the method can provide more flexible and personalized equipment resources according to different task types, meet the diverse task operation requirements, and help improve the intelligence of the power grid system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a schematic diagram of the structure of a power grid system in one embodiment;

[0032] Figure 2 This is one of the flowcharts of a resource collaborative scheduling method in one embodiment;

[0033] Figure 3 The second flowchart of the resource collaborative scheduling method in one embodiment;

[0034] Figure 4 The third flowchart of the resource collaborative scheduling method in one embodiment;

[0035] Figure 5 The fourth flowchart of the resource collaborative scheduling method in one embodiment;

[0036] Figure 6 FIG5 is a flowchart of a resource collaborative scheduling method in one embodiment;

[0037] Figure 7 is a structural block diagram of a resource collaborative scheduling device in an embodiment;

[0038] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0040] The resource collaborative scheduling method provided in the embodiment of the present application can be applied to Figure 1 In the power grid system shown. The power grid system includes a power subsystem 11 and a distributed subsystem 12. A plurality of power devices 111 in the power subsystem are used for power supply. The distributed subsystem 12 is connected to the power subsystem 11, and the distributed subsystem 12 is used to monitor the power subsystem 11. The distributed subsystem 12 includes a cloud server 121, an edge node 122, and a device end 123. The device end 123 is connected to the power device 111, and the device end 123 is used to collect data from the power device. The edge node 122 is connected to the cloud server 121 and the device end 123, respectively, and the edge node 122 is used to process data sent by the device end 123 and the cloud server 121. The cloud server 121 is used to process data sent by the edge node 122.

[0041] In one embodiment, Figure 2 As shown, a resource collaborative scheduling method is provided. This embodiment uses the method applied to a terminal as an example. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps S202 to S206.

[0042] S202: Acquire multiple running task requests sent by the distributed subsystems in the power grid system, and resource status information of the distributed subsystems.

[0043] Exemplarily, the distributed subsystem includes multiple device resources. Exemplarily, the device resources include at least one of a device terminal, an edge node, and a cloud server.

[0044] The device end is connected to the power equipment in the power subsystem and is used to collect the operation perception data of the power equipment in the power subsystem to realize the status monitoring of the power equipment. The operation perception data is used to represent the operation status of the power equipment and the environmental status of the environment where the power equipment is located, including but not limited to voltage, current, power, frequency, temperature, humidity, etc. Exemplarily, the device end includes power parameter sensors, environmental parameter sensors and related processing units, such as current sensors, voltage sensors, temperature sensors, humidity sensors, etc.

[0045] Edge nodes refer to computing devices or servers located at the edge of the network. Edge nodes are close to the device side in the distributed subsystem. Edge nodes are connected to the device side and cloud servers respectively. Edge nodes process data sent by the device side and cloud servers. Edge nodes process data using edge computing, a distributed information technology architecture that moves applications from centralized data centers to closer to where data is generated and used - the "edge", with the main purpose of reducing latency, saving bandwidth, and improving service response speed.

[0046] Cloud servers, also known as cloud servers or virtual servers, are computing services provided based on cloud computing technology. Cloud servers allow users to access remote server resources through the Internet without having to purchase and maintain physical hardware. Cloud servers provide great flexibility, scalability, and cost-effectiveness, and users can quickly increase or decrease resources as needed.

[0047] The task request is used to initiate a request to the power grid system to perform a specific task or operation. The task request corresponds to the task. The task request can come from a distributed subsystem, such as at least one of a device, an edge node, and a cloud server. The resource status information is used to indicate the current operating status of each device resource in the distributed subsystem.

[0048] In the application, various types of device resources in the distributed subsystem can regularly monitor their own operating status and generate corresponding resource status information.

[0049] S204: Determine the task type of the corresponding running task according to each running task request.

[0050] Task type refers to the classification of tasks or operations performed in the power grid system monitoring scenario, which is used to define the nature and purpose of the task and the methods and resources required to complete the task. Task type helps to better understand and manage different types of operation tasks, ensuring that each task can be performed in the most efficient way. Exemplarily, the task type includes at least one of data collection tasks, data processing tasks, model training tasks, maintenance operation tasks, and emergency response tasks.

[0051] S206: According to each task type and resource status information, coordinately dispatch different equipment resources in the distributed subsystem to execute each operation task, so as to monitor the power subsystem in the power grid system.

[0052] In the application, the terminal can sort the running tasks according to the task type and resource status information, and coordinate the scheduling of different equipment resources in the distributed subsystem to perform the running tasks according to the sorting, thereby realizing the monitoring of the power subsystem.

[0053] The above-mentioned resource collaborative scheduling method obtains multiple operation task requests sent by distributed subsystems in the power grid system, as well as resource status information of the distributed subsystems, determines the task type of the corresponding operation task according to each operation task request, and collaboratively schedules different equipment resources in the distributed subsystem to perform each operation task according to each task type and resource status information, so as to monitor the power subsystem in the power grid system. The method can dynamically adjust the equipment resource allocation of the distributed subsystem according to the task type and resource status information of each operation task, realizes the cloud-edge-end resource collaborative scheduling for the integration of sensing, transmission and computing, avoids resource waste, ensures that each task can be executed in the most efficient way, and can reduce the task queuing waiting time through reasonable task scheduling, speed up task processing speed, and improve system resource utilization and task processing efficiency; in addition, the method monitors the resource status information in real time, and once an abnormality or fault is detected, it can quickly collaboratively schedule other available resources for replacement, thereby reducing system downtime, improving system reliability, and improving system response speed; in addition, the method can provide more flexible and personalized equipment resources according to different task types, meet diverse task operation requirements, and help improve the intelligence of the power grid system.

[0054] In one embodiment, Figure 3 As shown, step S206, according to each task type and resource status information, collaboratively scheduling different device resources in the distributed subsystem to execute each running task, includes the following steps S302 and S304.

[0055] S302: Determine the priority of each running task according to each task type.

[0056] The priority of a running task is used to indicate the order of the running task among multiple running tasks. Exemplarily, the priority of each running task is determined according to each task type and a preset priority evaluation standard; wherein the indicators of the priority evaluation standard include at least one of importance, urgency, scope of impact, line of business and resource dependency.

[0057] It should be noted that the above is only an exemplary description. In the application, the priority evaluation criteria can be set according to the actual scenario and are not limited here.

[0058] S304: Coordinately schedule different device resources in the distributed subsystem to execute each running task according to the priority and resource status information of each running task.

[0059] Exemplarily, a priority queue for each running task is generated according to the priority of each running task, and the device resources in the distributed subsystem are dynamically adjusted and allocated according to the priority list and resource status information, so that each running task is executed collaboratively and / or independently through multiple device resources.

[0060] It should be noted that the sorting algorithm and resource allocation strategy used for task priority can be customized according to different power grid operation scenarios and requirements, and are not limited here.

[0061] The above resource collaborative scheduling method determines the priority of each operation task according to each task type, and coordinates the different equipment resources in the distributed subsystem to perform each operation task according to the priority and resource status information of each operation task. This resource collaborative scheduling method not only improves the overall performance of the power grid system, but also enhances the flexibility and adaptability of the system, laying a solid foundation for the future development of smart grids. Through effective management and optimization of resource allocation, it can ensure that key tasks are handled in a timely manner and improve the reliability and service quality of the system.

[0062] In one embodiment, the task type includes at least one of a data collection task, an analysis and decision task, and a model training task. Among them, the priority of the data collection task is higher than the analysis and decision task and the model training task, respectively, and the priority of the analysis and decision task is higher than the model training task. Based on this, the device end in the distributed subsystem can be collaboratively scheduled to perform the data collection task, and the edge node in the distributed subsystem can be collaboratively scheduled to perform the analysis and decision task, and the cloud server in the distributed subsystem can be collaboratively scheduled to perform the model training task.

[0063] It can be understood that the data collected by the data collection task is the basis for various operation tasks such as subsequent analysis and decision-making, model training, etc. Therefore, the device end close to the power equipment is called to collect data from the power system to obtain the source data for monitoring the power system. In addition, the analysis and decision-making tasks or data processing tasks have relatively high real-time requirements. Therefore, the analysis and decision-making tasks can be assigned to the edge nodes to meet the real-time requirements. In addition, the data processing volume corresponding to the model training task is large, and the requirements for equipment hardware resources are high. For this, the model training task can be assigned to the cloud server, and the cloud server can be combined with the edge nodes in the distributed subsystem to train the model in a distributed processing manner to achieve large-scale model training tasks. In this way, the automatic collaborative scheduling of various equipment resources is realized, resource waste is avoided, and each task can be executed in the most efficient way. And through reasonable task scheduling, the waiting time of tasks can be reduced, the task processing speed can be accelerated, the system resource utilization and task processing efficiency can be improved, the system reliability can be improved, the system response speed can be improved, and the diversified task operation requirements can be met, which helps to improve the intelligence of the power grid system.

[0064] In one embodiment, Figure 4 As shown, step S206, according to each task type and resource status information, collaboratively scheduling different device resources in the distributed subsystem to execute each running task, includes the following steps S402 and S404.

[0065] S402: When the task type includes a data collection task, determine a target device end in an idle state from the distributed subsystem according to resource status information.

[0066] The number of target device ends can be one or more, which can be determined according to the data collection task and is not limited here. The target device end refers to the device end in an idle state in the distributed subsystem, or in other words, the device end that is not occupied by other running tasks.

[0067] The type of the target device end can be any suitable type of sensor, for example, a current sensor, a voltage sensor, a power factor sensor, a frequency sensor, an environmental sensor (such as a temperature sensor, a humidity sensor, a light sensor), a vibration sensor, etc., without limitation here.

[0068] S404: Send a data collection instruction to the target device end to instruct the target device end to collect the operation perception data of the power subsystem according to the data collection instruction, and send the operation perception data to the edge node in the distributed subsystem.

[0069] In the application, after receiving the data collection instruction, the target device can use the data collection algorithm to collect the operation perception data of the power subsystem and send the operation perception data to the associated edge node in the distributed subsystem. Exemplarily, the operation perception data includes various types of data collected by the target device, such as current, voltage, power, temperature, humidity and other types of data.

[0070] Exemplarily, the target device can adjust the sampling frequency of the operation perception data according to a preset data collection algorithm. The data collection algorithm may include an adaptive sampling rate adjustment algorithm. This helps to improve data collection efficiency and accuracy, as well as improve resource utilization and avoid waste of resources.

[0071] Exemplarily, after the target device collects the operation perception data, it can also pre-process the operation perception data. For example, the target device uses a denoising algorithm based on wavelet transform to clean and denoise the collected operation perception data. This helps reduce the impact of interference data on the operation perception data, reduce the complexity of subsequent data processing, and improve the accuracy of data processing.

[0072] Exemplarily, the target device can also be used to compress the operation perception data and send the compressed operation perception data to the edge node associated with it. For example, a compression algorithm based on dictionary coding can be used to compress the operation perception data. This helps to reduce data complexity, improve the transmission efficiency of data between different device resources, and improve the response rate of subsequent data processing.

[0073] In the application, the device side can use an adaptive sampling rate adjustment algorithm to dynamically adjust the sampling frequency of the operation perception data, and use a wavelet transform-based denoising algorithm to clean and denoise the collected data, and use a dictionary coding-based compression algorithm to compress the denoised data, thereby transmitting the compressed operation perception data to the edge node.

[0074] It should be noted that, on the device side, the data acquisition algorithm can select different adaptive sampling algorithms according to the sensor type and grid characteristics; the preprocessing algorithm can adopt other effective signal processing algorithms; the compression algorithm can also select different lossless or lossy compression algorithms, as long as it can meet the requirements of reducing data transmission load and cost.

[0075] The above-mentioned resource collaborative scheduling method, when the task type includes a data collection task, determines the target device end in an idle state from the distributed subsystem according to the resource status information, sends a data collection instruction to the target device end to instruct the target device end to collect the operation perception data of the power subsystem according to the data collection instruction, and sends the operation perception data to the edge node in the distributed subsystem. It can allocate the data collection task to the device end close to the power subsystem, thereby realizing the effective collection of power parameters related to the power subsystem, improving the efficiency and quality of data collection, and enhancing the reliability, flexibility and security of the system, providing strong support for the development of smart grids.

[0076] In one embodiment, Figure 5 As shown, step S206, according to each task type and resource status information, collaboratively scheduling different device resources in the distributed subsystem to execute each running task, includes the following steps S502 and S504.

[0077] S502: When the task type includes an analysis and decision-making task, determine a target edge node in an idle state from the distributed subsystem according to resource status information.

[0078] The number of target edge nodes can be one or more, which can be determined according to the analysis and decision-making task, and is not limited here. The target edge node refers to the edge node in the distributed subsystem that is in an idle state, or the edge node with remaining hardware and / or software resources. In other words, the target edge node can meet the needs of executing the analysis and decision-making task.

[0079] S504: Send an analysis and decision instruction to the target edge node to instruct the target edge node to make an analysis and decision based on the operation perception data sent by the device end in the distributed subsystem, generate an operation perception result of the power subsystem, and send the operation perception data and operation perception result to the cloud server in the distributed subsystem; the operation perception result is used to represent the operation scenario of the power grid system.

[0080] In the application, the target edge node can analyze the data sent by the device side and / or the cloud server based on machine learning algorithms and / or deep learning algorithms, such as neural network algorithms, and generate operation perception results, thereby generating decision results based on the operation perception results, and thus better perceive the data processing needs. Among them, the analysis result refers to the result obtained by analyzing the operating status of at least part of the power equipment in the power subsystem, such as whether the operating status of the power equipment is abnormal, whether the power equipment needs to adjust the operating parameters, etc. The decision result is used to indicate the result of regulating the power grid system. For example, the decision result includes a control instruction or a status report, which can be transmitted to the cloud server and / or fed back to the device side to instruct the cloud server and / or the device side to perform corresponding operations.

[0081] It should be noted that in edge nodes, machine learning algorithms and deep learning algorithms can use different model structures and algorithm frameworks according to specific decision-making requirements and data characteristics, and there is no limitation here.

[0082] The above-mentioned resource collaborative scheduling method, when the task type includes an analysis and decision-making task, determines the target edge node in an idle state from the distributed subsystem according to the resource status information, sends an analysis and decision instruction to the target edge node to instruct the target edge node to perform analysis and decision-making based on the operation perception data sent by the device end in the distributed subsystem, generates the operation perception result of the power subsystem, and sends the operation perception data and the operation perception result to the cloud server in the distributed subsystem, thereby realizing the allocation of analysis and decision-making tasks to edge nodes close to the power subsystem, thereby realizing the analysis and regulation of the performance related to the power subsystem, enhancing the reliability, flexibility and security of the system, and helping to improve the security and reliability of the power grid system.

[0083] In one embodiment, Figure 6 As shown, step S206, according to each task type and resource status information, collaboratively scheduling different device resources in the distributed subsystem to execute each running task, includes the following steps S602 and S604.

[0084] S602: When the task type includes a model training task, determine a target cloud server in an idle state from the distributed subsystem according to the resource status information.

[0085] The number of target cloud servers can be one or more, which can be determined according to the analysis and decision-making tasks, and is not limited here. The target cloud server refers to the cloud server in the distributed subsystem that is idle, or the cloud server with remaining hardware and / or software resources. In other words, the target cloud server can meet the needs of executing the model training task.

[0086] S604: Send a model training instruction to the target cloud server to instruct the target cloud server to perform model training based on the operation perception data and operation perception results sent by the edge nodes in the distributed subsystem, generate a real-time perception model of the power grid system, and send model update data to the edge nodes in the distributed subsystem; the real-time perception model is used to monitor the power grid system in real time, and the model update data includes at least one of the real-time perception model and incremental data corresponding to the real-time perception model.

[0087] For example, after receiving the model training instruction, the target cloud server can use a distributed training algorithm to work in conjunction with the edge nodes in the distributed subsystem. The distributed training algorithm can be adjusted and optimized according to the computing resources and network conditions of the cloud server and the edge nodes.

[0088] In the application, the target cloud server pre-stores a large amount of historical data and model parameters of the power grid operation. The target cloud server obtains the real-time operation perception data and local model training results uploaded by the edge node, trains and optimizes the pre-created perception model to generate a real-time perception model, thereby realizing the resource call of the cloud server, which helps to improve the model training efficiency. Among them, the local model training result of the edge node is that the edge node performs incremental training of the model according to the cloud server instructions or local decision-making needs, and uses local data to further optimize the model.

[0089] The perception model is built based on the AI ​​(Artificial Intelligence) model. The real-time perception model is generated by training, optimizing or updating the perception model using real-time perception data and the local model training results of the edge node. When a real-time perception model is generated, it can be deployed on the target edge node or other edge nodes, or the new model or the incremental part of the model can be transmitted to the edge node through a secure and reliable transmission mechanism, so that intelligent calculations can be performed based on the real-time perception model, such as power grid load prediction, fault warning, etc., thereby realizing effective monitoring of the power grid system and improving the safety and reliability of the power grid system.

[0090] The above-mentioned resource collaborative scheduling method, when the task type includes a model training task, determines the target cloud server in an idle state from the distributed subsystem according to the resource status information, sends a model training instruction to the target cloud server to instruct the target cloud server to perform model training according to the operation perception data and operation perception results sent by the edge node in the distributed subsystem, generates a real-time perception model of the power grid system, and sends model update data to the edge node in the distributed subsystem, thereby allocating the model training task to the cloud server close to the power subsystem, so that the powerful computing performance of the cloud server can be used to execute operation tasks with large-scale data processing volume, and realize the effective call of the cloud server resources in the distributed subsystem. The cloud server can also work with the edge node for distributed collaborative work, further improving the utilization rate of equipment resources, enhancing the reliability, flexibility and security of the system, and helping to improve the security and reliability of the power grid system.

[0091] The resource collaborative scheduling method provided in the embodiment of the present application is based on runtime context perception and cloud (cloud server)-edge (remote node)-end (device end) distributed collaboration to build a collaborative scheduling support framework for integrated sensing, transmission and computing; by studying the key technologies of multi-type task scheduling systems, inter-device collaboration and intra-device scheduling are realized; key technology research and application of cloud-edge collaborative platforms for AI are carried out to achieve functions such as automatic cloud-edge update and adaptation of AI models, distributed training and incremental training; intelligent decision-making technology of intelligent algorithms in edge nodes is explored to enable edge nodes to have stronger intelligent decision-making capabilities; edge optimization methods based on artificial intelligence algorithms such as machine learning are studied to optimize perception data processing, task scheduling and resource management; device-side sensor data acquisition, preprocessing and compression algorithms are improved to reduce data transmission load and cost, thereby comprehensively improving the performance of smart grid systems.

[0092] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0093] Based on the same inventive concept, the embodiment of the present application also provides a resource collaborative scheduling device for implementing the resource collaborative scheduling method involved above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more resource collaborative scheduling device embodiments provided below can refer to the limitations of the resource collaborative scheduling method above, and will not be repeated here.

[0094] In one embodiment, Figure 7 As shown, a resource collaborative scheduling device 700 is provided, including an acquisition module 701, a determination module 702 and a scheduling module 703. Among them, the acquisition module 701 is used to obtain multiple operation task requests sent by the distributed subsystem in the power grid system, and the resource status information of the distributed subsystem. The determination module 702 is used to determine the task type of the corresponding operation task according to each operation task request. The scheduling module 703 is used to coordinate the scheduling of different equipment resources in the distributed subsystem to perform each operation task according to each task type and resource status information, so as to monitor the power subsystem in the power grid system.

[0095] The resource collaborative scheduling device 700 acquires multiple operation task requests sent by the distributed subsystem in the power grid system and the resource status information of the distributed subsystem through the acquisition module 701, determines the task type of the corresponding operation task according to each operation task request through the determination module 702, and coordinates the different equipment resources in the distributed subsystem to perform each operation task according to each task type and resource status information through the scheduling module 703, so as to monitor the power subsystem in the power grid system. The method can dynamically adjust the equipment resource allocation of the distributed subsystem according to the task type and resource status information of each operation task, realizes automatic collaborative scheduling of various equipment resources, avoids resource waste, ensures that each task can be executed in the most efficient way, and can reduce the task queuing waiting time through reasonable task scheduling, speed up the task processing speed, and improve the system resource utilization and task processing efficiency; in addition, the method monitors the resource status information in real time, and once an abnormality or fault is detected, it can quickly coordinate and schedule other available resources for replacement, thereby reducing system downtime, improving system reliability, and improving system response speed; in addition, the method can provide more flexible and personalized equipment resources according to different task types, meet the diverse task operation requirements, and help improve the intelligence of the power grid system.

[0096] In one embodiment, the scheduling module is also used to determine the priority of each running task according to each task type; based on the priority and resource status information of each running task, coordinately schedule different device resources in the distributed subsystem to execute each running task.

[0097] In one embodiment, the scheduling module is also used to determine a target device end in an idle state from the distributed subsystem based on resource status information when the task type includes a data acquisition task; send a data acquisition instruction to the target device end to instruct the target device end to collect operation perception data of the power subsystem according to the data acquisition instruction, and send the operation perception data to the edge node in the distributed subsystem.

[0098] In one embodiment, the scheduling module is also used to determine the target edge node in an idle state from the distributed subsystem according to the resource status information when the task type includes an analysis and decision-making task; send an analysis and decision instruction to the target edge node to instruct the target edge node to perform analysis and decision-making based on the operation perception data sent by the device end in the distributed subsystem, generate the operation perception result of the power subsystem, and send the operation perception data and operation perception result to the cloud server in the distributed subsystem; the operation perception result is used to represent the operation scenario of the power grid system.

[0099] In one embodiment, the scheduling module is also used to determine a target cloud server in an idle state from the distributed subsystem based on resource status information when the task type includes a model training task; send a model training instruction to the target cloud server to instruct the target cloud server to perform model training based on the operation perception data and operation perception results sent by the edge nodes in the distributed subsystem, generate a real-time perception model of the power grid system, and send model update data to the edge nodes in the distributed subsystem; the real-time perception model is used to monitor the power grid system in real time, and the model update data includes at least one of the real-time perception model and the incremental data corresponding to the real-time perception model.

[0100] Each module in the resource collaborative scheduling device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0101] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external device end through a network connection. When the computer program is executed by the processor, a resource collaborative scheduling method is implemented.

[0102] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0103] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps of the above method when executing the computer program.

[0104] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the aforementioned method are implemented.

[0105] In one embodiment, a computer program product is provided, comprising a computer program, which implements the steps of the aforementioned method when executed by a processor.

[0106] 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, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0107] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile 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), magnetoresistive 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. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0108] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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 specification.

[0109] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A resource collaborative scheduling method, characterized in that: The method comprises: Acquire multiple running task requests sent by distributed subsystems in the power grid system, and resource status information of the distributed subsystems; Determine the task type of the corresponding running task according to each of the running task requests; According to each of the task types and the resource status information, different equipment resources in the distributed subsystem are coordinated to execute each of the operation tasks, so as to monitor the power subsystem in the power grid system.

2. The method according to claim 1, characterized in that The step of collaboratively scheduling different device resources in the distributed subsystem to execute each of the running tasks according to each of the task types and the resource status information includes: Determining the priority of each of the running tasks according to each of the task types; Different device resources in the distributed subsystem are collaboratively scheduled to execute each of the running tasks according to the priority of each of the running tasks and the resource status information.

3. The method according to claim 1, characterized in that: The task type includes at least one of a data collection task, an analysis and decision-making task, and a model training task; wherein the priority of the data collection task is higher than the analysis and decision-making task and the model training task, respectively, and the priority of the analysis and decision-making task is higher than the model training task.

4. The method according to any one of claims 1 to 3, characterized in that: The step of collaboratively scheduling different device resources in the distributed subsystem to execute each of the running tasks according to each of the task types and the resource status information includes: In the case where the task type includes a data acquisition task, determining a target device end in an idle state from the distributed subsystem according to the resource status information; A data collection instruction is sent to the target device end to instruct the target device end to collect the operation perception data of the power subsystem according to the data collection instruction, and send the operation perception data to the edge node in the distributed subsystem.

5. The method according to any one of claims 1 to 3, characterized in that: The step of collaboratively scheduling different device resources in the distributed subsystem to execute each of the running tasks according to each of the task types and the resource status information includes: In the case where the task type includes an analysis and decision-making task, determining a target edge node in an idle state from the distributed subsystem according to the resource state information; An analysis and decision instruction is sent to the target edge node to instruct the target edge node to make an analysis and decision based on the operation perception data sent by the device end in the distributed subsystem, generate an operation perception result of the power subsystem, and send the operation perception data and the operation perception result to the cloud server in the distributed subsystem; the operation perception result is used to represent the operation scenario of the power grid system.

6. The method according to any one of claims 1 to 3, characterized in that: The step of collaboratively scheduling different device resources in the distributed subsystem to execute each of the running tasks according to each of the task types and the resource status information includes: In a case where the task type includes a model training task, determining a target cloud server in an idle state from the distributed subsystem according to the resource status information; Send a model training instruction to the target cloud server to instruct the target cloud server to perform model training according to the operation perception data and operation perception results sent by the edge node in the distributed subsystem, generate a real-time perception model of the power grid system, and send model update data to the edge node in the distributed subsystem; the real-time perception model is used to monitor the power grid system in real time, and the model update data includes at least one of the real-time perception model and the incremental data corresponding to the real-time perception model.

7. A resource collaborative scheduling device, characterized in that: Applied to a collaborative scheduling system, the device comprises: An acquisition module, used to acquire multiple operation task requests sent by a distributed subsystem in a power grid system, and resource status information of the distributed subsystem; A determination module, used to determine the task type of the corresponding running task according to each of the running task requests; The scheduling module is used to coordinately schedule different equipment resources in the distributed subsystem to perform each of the operating tasks according to each of the task types and the resource status information, so as to monitor the power subsystem in the power grid system.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images