Power grid multi-node multi-task cluster management system and method based on k8s platform

Abstract

The invention provides a power grid multi-node multi-task cluster management system based on a k8s platform, and the system is based on the k8s platform, and comprises a power grid control center and a power transmission and transformation station. The power transmission and transformation stations serve as working nodes of the system; the power grid control center serves as a management node of the system and is used for managing the working nodes; the working node communicates with the management node through a network; one or more control servers are arranged in the working node; according to the control property, one or more container applications are arranged in the control server; each container application is used for implementing one or more control functions. And a container deployment architecture mode is adopted in the control server of the transmission and transformation station, so that the transmission and transformation station can carry out dynamic capacity expansion of a control function according to actual control service requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission and transformation control architecture, and in particular to a power grid multi-node multi-task cluster management system and method based on a k8s platform. Background Art

[0002] With the large-scale development of the power grid system, especially the fact that the power grid system structure is no longer a single thermal power and hydropower energy structure, there are new large-scale wind power and photovoltaic new energy sources that need to be consumed, as well as emerging energy allocation methods such as pumped storage, which brings challenges to the construction of a new power system. As the intermediate link of various energy structures, the power transmission and transformation system has gradually expanded and penetrated into the power transmission and transformation station with the development trend of intelligent control. It not only involves the development and training of a large number of AI models such as intelligent regulation, operation error prevention, and new energy prediction, but also requires the implementation of intelligence in the control of the power transmission and transformation station level.

[0003] At present, with the development of online monitoring technology of power transmission and substation towards a more intelligent, automated and integrated direction, the reliability and operating efficiency of equipment can be improved by real-time monitoring and analysis of the status of power equipment, the occurrence of faults and power outages can be reduced, and the safety and sustainability of the power system can be improved. Due to the increasing demand, there are more and more monitoring devices used to monitor the stability of power equipment in the station. In the face of the complex and harsh monitoring environment in the station, in order to ensure accurate monitoring of the status and performance of power equipment, the technical requirements for online monitoring equipment products are getting higher and higher. In general, operation and maintenance personnel remotely access the equipment through the network, view the monitoring data in real time, perform configuration and diagnosis, and further realize operation, control, monitoring and management. However, with the rapid expansion of the scale of the power transmission and transformation network and the increase in the number of projects that require intelligent monitoring, the manual monitoring and configuration method can no longer meet the development needs.

[0004] In view of this, a power grid multi-node multi-task cluster management system and method based on the k8s platform is needed. Summary of the invention

[0005] In view of the problem that the existing technology of manually monitoring and configuring transmission and substations one by one can no longer meet the development needs, the present invention provides a multi-node multi-task cluster management system and method for power grids based on the k8s platform, which can facilitate the dynamic expansion of control functions of transmission and substations according to actual control business needs. The specific technical solution is as follows:

[0006] A multi-node multi-task cluster management system for a power grid based on a k8s platform, the system is based on the k8s platform and comprises: a power grid control center and a transmission and transformation substation; the transmission and transformation substation serves as a working node of the system; the power grid control center serves as a management node of the system and is used to manage the working nodes; the working nodes communicate with the management node via a network; one or more control servers are provided in the working nodes; according to the nature of control, one or more container applications are provided in the control server; each container application is used to implement one or more control functions.

[0007] Furthermore, a resource dynamic file is stored on the control server; the resource dynamic file is used to record the resource information in the working node; the resource information includes all resource information corresponding to the working node and the resource information being used by the container application; all resource information includes the computing resources, storage resources of the control server and the maximum carrying power transmission and transformation of the controlled circuit; the resource information being used by the container application includes the computing resources, storage resources occupied by the container application and the current power transmission and transformation of the controlled circuit.

[0008] Furthermore, the control server is provided with a resource monitoring component; the resource monitoring component is used to obtain all resource dynamic files in the working node and parse and summarize them, and then send the resource information to the management node.

[0009] Furthermore, the resource dynamic file is maintained by the kubelet service of each server.

[0010] Furthermore, a monitor is provided in the resource monitoring component to check the update of the resource dynamic file in real time; when the resource dynamic file changes, the monitor will trigger the resource monitoring component to re-parse the resource dynamic file to obtain the latest resource dynamic file content of the working node.

[0011] Furthermore, the management node is provided with a resource aggregation component for aggregating resource information of all working nodes and updating the resource information to a local database, so that the user can query the resource database through a resource query instruction to obtain a resource query result.

[0012] Furthermore, an auxiliary configuration file corresponding to the container application is stored on the control server; the mounting information of the auxiliary configuration file is stored in the resource dynamic file; the auxiliary configuration file is used to configure the control parameters of the container application, so that in an emergency, it can be handled based on the user-configured solution.

[0013] Furthermore, it also includes a knowledge graph node; the knowledge graph node encapsulates the algorithm model of knowledge graph construction and application into a container application, and deploys it through the k8s platform to achieve batch and large-scale production and application of knowledge graphs.

[0014] A multi-node multi-task cluster management method for a power grid based on a k8s platform is applied to the multi-node multi-task cluster management system for a power grid based on a k8s platform described above, and includes the following steps:

[0015] The power grid control center is used as the management node of the k8s platform; the transmission and substation is used as the working node of the k8s platform;

[0016] Encapsulate the program code, runtime, library, environment variables and configuration files required for substation control into a container application;

[0017] Deploy the container application to the control server in the working node through the k8s platform of the management node.

[0018] Furthermore, the method further comprises the following steps:

[0019] Setting a resource dynamic file on the control server to record resource information in the working node;

[0020] When the hardware resources in the working node are allocated to the container application for use, an auxiliary configuration file corresponding to the container application is generated, and the mounting information of the auxiliary configuration file is stored in the resource dynamic file.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] 1. By adopting the container application deployment method to construct the control method of the power transmission and substation, the control algorithm can be quickly deployed on the same control type server through the container application, realizing the data and model sharing and conversion between different working nodes and multiple similar tasks. At the same time, after deployment, each container application can generate its own auxiliary configuration file according to the specific environment information, and each container application will run independently, realizing the isolation of storage, computing, data, model and other assets between different working nodes and different tasks, ensuring the operation safety of each container application and each control module of the power transmission and substation. At the same time, it also realizes the convenient query of the resource information used by the container group in the working node, solves the problem in the related technology that the statistics of the resource usage of the power transmission and substation need to call a large number of complex other business systems to handle, makes the control logic of the power transmission and substation easy to maintain, and saves manpower to complete business needs. The use of container deployment architecture in the control server of the power transmission and substation can also facilitate the dynamic expansion of the control function of the power transmission and substation according to the actual control business needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0024] Figure 1 This is a structural diagram of a power grid multi-node multi-task cluster management system based on the k8s platform;

[0025] Figure 2 This is a flow chart of a multi-node and multi-task cluster management method for power grid based on the k8s platform. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0028] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0029] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0030] Embodiment 1

[0031] Containerized applications are applications and all their dependencies packaged into an independent, portable container. This container contains everything needed to run the application, including code, runtime, libraries, environment variables, and configuration files, ensuring that the application can run consistently in different environments.

[0032] Containerization is a technical approach to software development and deployment that allows developers to package an application and all its dependencies into a lightweight, portable, independent container. Containers use operating system-level virtualization technologies, such as Linux namespaces and control groups (cgroups), to provide isolation and resource management. Compared with traditional virtualization technologies, containerized applications have higher flexibility and efficiency because they can run in different environments, including development, testing, and production environments, without reconfiguring or modifying applications. By using containerization technology, enterprises can achieve automated deployment, elastic expansion, and efficient operation and maintenance, thereby improving application reliability and user experience. Containerized applications provide high flexibility, consistency, and efficiency by packaging applications and their dependencies into independent containers, and are an important trend in modern software development and deployment.

[0033] Kubernetes (k8s) is an open source container orchestration engine that is widely used in the deployment and management of cloud-native applications. It provides a reliable and highly scalable platform that enables developers to easily package applications into containers and deploy, scale and manage them in distributed systems. With the rapid development of cloud computing technology, enterprises have an increasingly urgent need to build and manage cloud-native architectures. Kubernetes fills this gap and solves the complexity and non-scalability problems faced by traditional deployment methods. The core concept of Kubernetes is to manage the life cycle of applications in an automated and intelligent way based on containers. It can flexibly handle various types of workloads, whether it is long-running applications, real-time streaming data or batch tasks. At the same time, Kubernetes also has strong scalability and fault tolerance, and can automatically adjust the scale and configuration of applications to adapt to different load requirements and improve the availability and stability of the system. Kubernetes adopts a distributed architecture consisting of multiple nodes, each of which runs some containers. Through the coordination and scheduling of the cluster manager, Kubernetes can automatically deploy containers to appropriate nodes and automatically scale according to the system load. This distributed architecture allows applications to better utilize resources and improve overall efficiency and reliability. As a powerful container orchestration engine, it provides a solid foundation for the development of cloud-native applications. It simplifies the deployment and management of applications through automation and intelligence, and improves the scalability and stability of the system.

[0034] like Figure 1The figure shows a structural schematic diagram of a multi-node multi-task cluster management system for a power grid based on a k8s platform. The system is based on the k8s platform. The k8s platform includes a management node and several working nodes. The working node is a node for deploying containers. The management node and the working nodes are distributed in different servers or data centers and computers. The cluster management system includes: a power grid control center and a transmission and transformation station. The transmission and transformation station serves as a working node of the system. The power grid control center serves as a management node of the system and is used to manage the working nodes. The working nodes communicate with the management node through a network. One or more control servers are provided in the working nodes. According to the nature of control, one or more container applications are provided in the control server. Each container application is used to implement one or more control functions.

[0035] The power grid master control center is used as the management node, and each substation is used as the working node to form a power grid cluster management system based on the k8s platform. The management node is used to manage the working nodes. The container deployment architecture is adopted in the control server of the substation, which can also facilitate the dynamic expansion of the control function of the substation according to the actual control business needs, so as to reduce the cost of manual monitoring and configuration one by one, and improve efficiency and safety.

[0036] Specifically, each power transmission and transformation working node is equipped with several control servers; according to the control nature, each control server is equipped with one or several container applications. For example, if a server only performs one type of control function, then one container application is deployed in the server; if the control server performs several types of control functions, then several container applications are deployed in the server; each container application corresponds to one type of control function.

[0037] In a specific implementation, a resource dynamic file (kubelet_internal_checkpoint file) is stored on the control server; the resource dynamic file is used to record the resource information in the working node; the resource information includes all resource information corresponding to the working node and the resource information being used by the container application; all resource information includes the computing resources, storage resources of the control server and the maximum load-bearing transmission and transformation power of the controlled circuit; the resource information being used by the container application includes the computing resources, storage resources occupied by the container application and the current transmission and transformation power of the controlled circuit.

[0038] Specifically, the content template of the resource dynamic file may include the following contents, which are only used to illustrate the resource dynamic file. The content template of the resource dynamic file may also include other fields. Those skilled in the art may design according to actual needs, and the present application embodiment does not limit this. The content template of the resource dynamic file is, for example:

[0039] "Data":

[0040] "PodDeviceEntries":

[0041] "PodUID":"xxxxxx","ContainerName":"pod-example"

[0042] "ResourceName":"nvidia.com / gpu"

[0043] "DeviceIDs":{"GPU-1","GPU-2"}

[0044] "AllocResp":{" / host / filel / path / container / filel / path"," / host / file2 / path / cont ainer / file3 / path","RegisteredDevices":"nvidia.com / gpu":{"GPU-1""GPU-2""GPU-3""GPU-4"}}

[0045] As shown above, the file content of the resource dynamic file, where the PodDeviceEntries attribute records the computing resource hardware information that the pod (container application) is using, and the RegisteredDevices attribute records all the corresponding computing resource hardware information of the node. Taking the above content as an example, the computing resource hardware information that the pod (container application) is using is "GPU-1" and "GPU-2", and all the computing resource information corresponding to the working node is: "GPU-1", "GPU-2", "GPU-3", "GPU-4".

[0046] Furthermore, the resource dynamic file is maintained by the kubelet service of each server.

[0047] In a specific implementation, the control server is provided with a resource monitoring component; the resource monitoring component is used to obtain all resource dynamic files in the working node and parse and summarize the resource information and send it to the management node. By parsing the resource dynamic files, the resource information of the device resources of the working node can be obtained. Each working node can deploy a resource monitoring component on one of the control servers. The resource monitoring component summarizes other and local resource dynamic files so that the resource information of each working node can be counted and summarized separately.

[0048] Furthermore, a resource monitoring component is provided on one of the servers in each power transmission and transformation working node, and the resource monitoring component is used to obtain all resource dynamic files in the power transmission and transformation working node (each control server has a resource dynamic file). The resource dynamic file is used to record the resource information in the power transmission and transformation working node; the resource information includes all resource information corresponding to the power transmission and transformation working node and the resource information being used by the container application. All resource information includes the computing resources, storage resources of each control server, the maximum load-bearing power transmission and transformation of the controlled circuit, and other power transmission and transformation rated information; the resource information being used by the container application includes the computing resources, storage resources occupied by the container application, the current power transmission and transformation of the controlled circuit, and other power transmission and transformation information.

[0049] Furthermore, the resource monitoring component parses and summarizes the resource dynamic file of the working node, obtains the resource information of the power transmission and transformation working node, and sends the resource information to the management node of the power grid control center.

[0050] Furthermore, the resource monitoring component monitors the update of the resource dynamic file; and when it is monitored that the resource dynamic file is updated, a new resource dynamic file is obtained.

[0051] In a specific implementation, a monitor is provided in the resource monitoring component to check the update of the resource dynamic file in real time; when the resource dynamic file changes, when a resource dynamic file in the power transmission and transformation working node changes, the monitor will trigger the resource monitoring component to re-parse the resource dynamic file to obtain the latest resource dynamic file content of the working node, so as to update the resource information of the working node in the management node. Among them, the update of the resource dynamic file may include resource deletion, resource addition, etc. When a resource update occurs, the latest resource dynamic file is re-parsed so that the resource information of the working node stored in the management node is updated to the latest state. Furthermore, in order to reduce the excessive occupation of computing data and communication bandwidth caused by too frequent update frequency, a fluctuation threshold can be set, and the resource dynamic file will not be updated when the resource fluctuates normally, and the resource dynamic file will be updated again when the fluctuation or fluctuation speed exceeds the threshold. For example, the power transmission and transformation controlled by a container application changes in real time, which can be considered as normal power fluctuation within a certain threshold range. When it exceeds the normal range of the time period (for example, when extreme weather occurs in the area), the resource dynamic file is triggered to update. Then the monitor will trigger the resource monitoring component to re-parse the resource dynamic file to obtain the latest resource dynamic file content of the working node, so as to update the resource information of the working node on the management node and issue an alarm prompt.

[0052] Furthermore, the management node is provided with a resource aggregation component for aggregating resource information of all working nodes and updating the resource information to a local database, so that the user can query the resource database through a resource query instruction to obtain a resource query result.

[0053] The resource aggregation component of the management node of the power grid control center can receive resource information from the resource monitoring components of different power transmission and transformation working nodes respectively. According to the resource information uploaded by each power transmission and transformation working node, the records in the database are added, deleted or updated. For example, when a new container application using a device is created, the new container application and the hardware device resources (computing resources, storage resources) used by it, the control function performed (such as which part of the power transmission and transformation is controlled or which part of the power transmission and transformation secondary switch is controlled) and other information are added to the database. When the original pod (container application) using the hardware device is deleted, the original pod (container application) and the hardware device resource information used by it are deleted in the database. For example, the resource aggregation component can receive the resource information reported by the resource monitoring component of a certain substation node 1. A certain substation monitors and parses the resource dynamic file through the resource monitoring component deployed on a certain server, and obtains the device resource information and control function information corresponding to the containers pod_1 to pod_n. The resources corresponding to pod_1 include accelerator cards 1 to accelerator cards n, and the resources corresponding to pod_n include smart network cards 1 to smart network cards n. The resource aggregation component can also receive resource information reported by the resource monitoring component of node n. Node n monitors and parses the resource dynamic file through the resource monitoring component deployed on node n, and obtains the device resource information corresponding to pod_x to pod_y. The resources corresponding to podx include inference card 1 to inference card n, or the resources can be control card 1 to control card n for controlling the power transmission and transformation switch. The statistical resources of other nodes and the method of reporting resources are similar. It depends on what specific functions the pod container application is to achieve, and I will not go into details here.

[0054] Furthermore, the control server stores an auxiliary configuration file corresponding to the container application; the mounting information of the auxiliary configuration file is stored in the resource dynamic file; the auxiliary configuration file is used to configure the control parameters of the container application, and then in an emergency, it can be handled based on the user-configured solution. Further, the auxiliary configuration file can be a file used to configure the opening and closing sequence of switches when a fault occurs in the power transmission and transformation network. After the mounting path of the auxiliary configuration file is stored in the resource dynamic file, when a fault occurs in the power transmission and transformation network, the auxiliary configuration file can be read and run through the path of the auxiliary configuration file saved in the resource dynamic file, and then in the emergency handling of the fault, it can be selected based on the user-configured solution to solve the problem of slow manual efficiency in emergency handling of the fault, especially in the field of power supply guarantee, which can achieve the effect of immediate handling and temporary emergency power supply. That is, by setting key fields in the resource dynamic file to store the path of the file required by the user, the management of the file can be realized, and more auxiliary functions can be developed based on the cooperation of the resource dynamic file on the basis of the original equipment to meet the needs of more functional scenarios.

[0055] Furthermore, it also includes a knowledge graph node; the knowledge graph node encapsulates the algorithm model of knowledge graph construction and application into a container application, and deploys it through the k8s platform to achieve batch and large-scale production and application of knowledge graphs. The working node also includes a knowledge graph node; the algorithm model of knowledge graph construction and application is encapsulated into a container application, and deployed on each knowledge graph node through the k8s platform, so that the knowledge graph can be produced and applied in batches and on a large scale. The specific algorithm model for knowledge graph construction and application is an existing mature technology, so it will not be repeated here. At the same time, the knowledge graph node can obtain real-time new information of the power transmission and transformation working node through the management node, and can update the knowledge graph structure in real time.

[0056] Embodiment 2

[0057] like Figure 2 The figure shows a flow chart of a multi-node multi-task cluster management method for a power grid based on a k8s platform, which is applied to the multi-node multi-task cluster management system for a power grid based on a k8s platform described above, and includes the following steps:

[0058] The power grid control center is used as the management node of the k8s platform; the transmission and substation is used as the working node of the k8s platform;

[0059] Encapsulate the program code, runtime, library, environment variables and configuration files required for substation control into a container application;

[0060] Deploy the container application to the control server in the working node through the k8s platform of the management node.

[0061] Furthermore, the method further comprises the following steps:

[0062] Setting a resource dynamic file on the control server to record resource information in the working node;

[0063] When the hardware resources in the working node are allocated to the container application for use, an auxiliary configuration file corresponding to the container application is generated, and the mounting information of the auxiliary configuration file is stored in the resource dynamic file.

[0064] A resource monitoring component is set in the working node (one of the control servers) to parse and summarize the resource dynamic file of the working node, obtain the resource information of the power transmission and transformation working node, and send the resource information to the management node of the power grid control center.

[0065] Furthermore, the resource monitoring component can compare the new resource dynamic file with the resource dynamic file obtained last time to determine the latest resource information of the working node, for example, whether any resources have been deleted or whether any new resources have been added. The resource information to be deleted and the resource information to be added are aggregated to generate the latest resource information, so that the management node can update the resource information stored locally after receiving the latest resource information to provide accurate query services.

[0066] Furthermore, a resource aggregation component is set in the management node of the power grid control center to aggregate the resource information of all power transmission and transformation working nodes and update it to the local hardware resource database; so that users can query the hardware resource database through resource query instructions and obtain resource query results.

[0067] The resource aggregation component can receive resource information from resource monitoring components of different power transmission and transformation working nodes respectively. Add, delete or update the records in the database according to the resource information uploaded by each power transmission and transformation working node. For example, when a new container application using a device is created, the new container application and the hardware device resources (computing resources, storage resources) used by it, the control function performed (such as which part of the power transmission and transformation is controlled or which part of the power transmission and transformation secondary switch is controlled) and other information are added to the database. When the original pod (container application) using the hardware device is deleted, the original pod (container application) and the hardware device resource information used by it are deleted in the database. For example, the resource aggregation component can receive the resource information reported by the resource monitoring component of a certain substation node 1. A certain substation monitors and parses the resource dynamic file through the resource monitoring component deployed on a server, and obtains the device resource information and control function information corresponding to the container pod_1 to pod_n. The resources corresponding to pod_1 include accelerator card 1 to accelerator card n, and the resources corresponding to pod_n include smart network card 1 to smart network card n. The resource aggregation component can also receive resource information reported by the resource monitoring component of node n. Node n monitors and parses the resource dynamic file through the resource monitoring component deployed on node n, and obtains the device resource information corresponding to pod_x to pod_y. The resources corresponding to podx include inference card 1 to inference card n, or the resources can be control card 1 to control card n for controlling the power transmission and transformation switch. The statistical resources of other nodes and the method of reporting resources are similar. It depends on what specific functions the pod container application is to achieve, and I will not go into details here.

[0068] Furthermore, the pod container application name is parsed and queried from the kubernetes cluster based on the PodUID information in the resource dynamic file. At this point, the information that can be determined includes the pod container application name, the resource name used by the pod container application, and the device identification information of the resources used by the pod container application. This information is packaged into a data structure and sent to the management node. By counting the application information of each pod container on the power transmission and transformation working node, the corresponding data structure will be generated, and then the information will be uploaded to the resource aggregation component in the kubernetes cluster for aggregation processing.

[0069] Furthermore, the user can query the hardware resource database through resource query instructions to obtain resource query results, which means that the resource aggregation component will provide a corresponding interface, so that platform managers can obtain resource equipment usage information of pod container applications through interface requests, effectively solving the problem of resource statistics and simplifying the overall complexity of the system.

[0070] In specific implementation, when it is necessary to update the container application of the server in the power transmission and transformation working node, it is necessary to first evaluate the computing resources occupied by the new container application (such as the computing amount, broadband traffic, storage capacity, etc. required by the algorithm model used in the new container application), and evaluate whether the existing server can meet the performance requirements of the new container application. If it does, the new container application can be directly deployed remotely. If it does not, it is necessary to replace the server first or deploy it with other servers in the new power transmission and transformation working node before deployment. For example, under feasible control conditions for the power grid, the original container application of server A is deployed to server B to free up the computing resources of server A, and then the new container application is deployed on server A to maximize the utilization of computing resources.

[0071] When the hardware resources in the power transmission and transformation working node are allocated to the pod container application for use, an auxiliary configuration file corresponding to the pod container application is generated, and the mounting information of the auxiliary configuration file is stored in the resource dynamic file.

[0072] Furthermore, the auxiliary configuration file generation and mounting step may include:

[0073] First, generate an auxiliary configuration file according to business requirements; specifically, when allocating hardware devices, first obtain basic hardware information, such as obtaining the GPU ID in the device-plugin that manages the GPU; obtaining the name of the network card in the device-plugin that manages the IP network card; obtaining the switch number in the device-plugin that manages the power transmission and transformation switch, such as the control card 1...control card n mentioned above; write this information into the auxiliary configuration file in the format of business requirements.

[0074] Then mount the auxiliary configuration file to the pod container application, that is, inject the auxiliary file mounting information into the response body of calling the device-plugin interface, and report the auxiliary configuration file mounting information by requesting the device-plugin interface. The reported mounting information can be automatically and persistently stored in the resource dynamic file (kubelet_internal_checkpoint file).

[0075] Furthermore, the auxiliary configuration file can be a file used to configure the opening and closing sequence of switches when a fault occurs in the power transmission and transformation network. After the mounting path of the auxiliary configuration file is stored in the resource dynamic file, when a fault occurs in the power transmission and transformation network, the auxiliary configuration file can be read and run through the path of the auxiliary configuration file saved in the resource dynamic file. Then, when the fault is urgently handled, it can be selected based on the user-configured solution to solve the problem of low manual efficiency in emergency handling of faults, especially in the field of power supply guarantee, which can achieve the effect of immediate handling and temporary emergency power supply. That is, by setting key fields in the resource dynamic file to store the path of the file required by the user, the management of the file can be realized. On the basis of the original equipment, more auxiliary functions can be developed based on the cooperation of the resource dynamic file to meet the needs of more functional scenarios.

[0076] The present application provides a multi-node multi-task cluster management system for power grids based on the k8s platform. The system is based on the k8s platform and includes: a power grid control center and a transmission and substation; the transmission and substation serves as a working node of the system; the power grid control center serves as a management node of the system and is used to manage the working nodes; the working nodes communicate with the management nodes through a network; one or more control servers are provided in the working nodes; according to the nature of the control, one or more container applications are provided in the control server; each container application is used to implement one or more control functions. By adopting the container application deployment method to construct the control method of the transmission and substation, the control algorithm can be quickly deployed on the same control type server through the container application, so as to realize the data and model sharing conversion between different working nodes and multiple similar tasks. At the same time, after deployment, each container application can generate its own auxiliary configuration file according to the specific environmental information, and each container application will run independently, realizing the isolation of storage, computing, data, models and other assets between different users and different tasks, and ensuring the safe operation of each container application and each control module of the transmission and substation. At the same time, it also enables convenient query of resource information used by container groups in working nodes, solving the problem in related technologies that the resource usage statistics of power transmission and substations need to be processed by calling a large number of complex other business systems, making the control logic of power transmission and substations easier to maintain and saving manpower to meet business needs. The use of container deployment architecture in the control server of the power transmission and substation can also facilitate the dynamic expansion of control functions of the power transmission and substation according to actual control business needs.

[0077] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0078] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0079] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0080] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A multi-node multi-task cluster management system for power grid based on k8s platform, characterized in that: The system is based on the k8s platform and includes: a power grid control center and a transmission and transformation substation; the transmission and transformation substation serves as the working node of the system; the power grid control center serves as the management node of the system and is used to manage the working nodes; the working nodes communicate with the management nodes through the network; one or more control servers are provided in the working nodes; according to the nature of the control, one or more container applications are provided in the control server; each container application is used to implement one or more control functions.

2. According to the k8s platform-based power grid multi-node multi-task cluster management system according to claim 1, it is characterized in that: A resource dynamic file is stored on the control server; the resource dynamic file is used to record the resource information in the working node; the resource information includes all resource information corresponding to the working node and the resource information being used by the container application; all resource information includes the computing resources, storage resources of the control server and the maximum load-bearing transmission and transformation power of the controlled circuit; the resource information being used by the container application includes the computing resources, storage resources occupied by the container application and the current transmission and transformation power of the controlled circuit.

3. According to the k8s platform-based power grid multi-node multi-task cluster management system according to claim 2, it is characterized in that: The control server is provided with a resource monitoring component; the resource monitoring component is used to obtain all resource dynamic files in the working node, parse and summarize them, and then send the resource information to the management node.

4. According to the k8s platform-based power grid multi-node multi-task cluster management system according to claim 2, it is characterized in that: The resource dynamic file is maintained by the kubelet service of each server.

5. According to the k8s platform-based power grid multi-node multi-task cluster management system of claim 3, it is characterized in that: The resource monitoring component is provided with a monitor to check the update of the resource dynamic file in real time; when the resource dynamic file changes, the monitor will trigger the resource monitoring component to re-parse the resource dynamic file to obtain the latest resource dynamic file content of the working node.

6. According to the k8s platform-based power grid multi-node multi-task cluster management system of claim 5, it is characterized in that: The management node is provided with a resource aggregation component for aggregating resource information of all working nodes and updating the resource information to a local database, so that the user can query the resource database through a resource query instruction to obtain a resource query result.

7. The power grid multi-node multi-task cluster management system based on the k8s platform according to claim 2 is characterized in that: The control server stores an auxiliary configuration file corresponding to the container application; the mounting information of the auxiliary configuration file is stored in the resource dynamic file; the auxiliary configuration file is used to configure the control parameters of the container application, so that in an emergency, it can be handled based on the user-configured solution.

8. The power grid multi-node multi-task cluster management system based on the k8s platform according to claim 1 is characterized in that: It also includes a knowledge graph node; the knowledge graph node encapsulates the algorithm model of knowledge graph construction and application into a container application, and deploys it through the k8s platform to achieve batch and large-scale production and application of knowledge graphs.

9. A multi-node multi-task cluster management method for power grid based on k8s platform, characterized in that: The multi-node multi-task cluster management system for a power grid based on a k8s platform as described in any one of claims 1 to 8 comprises the following steps: The power grid control center is used as the management node of the k8s platform; the transmission and substation is used as the working node of the k8s platform; Encapsulate the program code, runtime, library, environment variables and configuration files required for substation control into a container application; Deploy the container application to the control server in the working node through the k8s platform of the management node.

10. The power grid multi-node multi-task cluster management method based on the k8s platform according to claim 9 is characterized in that: The following steps are also included: Setting a resource dynamic file on the control server to record resource information in the working node; When the hardware resources in the working node are allocated to the container application for use, an auxiliary configuration file corresponding to the container application is generated, and the mounting information of the auxiliary configuration file is stored in the resource dynamic file.

Images