CN111242801B

CN111242801B - A power system control cloud grid operation analysis platform

Info

Publication number: CN111242801B
Application number: CN201911401268.5A
Authority: CN
Inventors: 陈振宇; 高兴宇; 狄方春; 李大鹏; 黄运豪; 杨清波; 刘�东
Original assignee: Institute of Microelectronics of CAS; China Electric Power Research Institute Co Ltd CEPRI
Current assignee: Institute of Microelectronics of CAS; China Electric Power Research Institute Co Ltd CEPRI
Priority date: 2019-12-31
Filing date: 2019-12-31
Publication date: 2025-07-11
Anticipated expiration: 2039-12-31
Also published as: CN111242801A

Abstract

A power system control cloud grid operation analysis platform includes: software-level service SaaS layer, support platform service layer, infrastructure-level service IaaS layer, external connection layer, communication layer, and display layer. A container management module is provided for the standardized application management construction mode and container management method based on containers, which is conducive to improving the standardization construction and homogeneous management of national and provincial control analysis and decision-making applications, forming a centralized, unified, collaborative, standardized and efficient "big operation" system, and achieving higher standards in terms of efficiency, rationality and security of power system control cloud platform resource utilization.

Description

Power system regulation and control cloud power grid operation analysis platform

Technical Field

The invention belongs to the field of power regulation and control clouds, and particularly relates to a power system regulation and control cloud power grid operation analysis platform.

Background

Electric power is used as the most main terminal consumption energy, and plays a role in improving the energy regulation level. At present, the power consumption efficiency of the power customers in China is low, the phenomenon of electric energy waste exists in a large quantity, meanwhile, the peak-valley difference of the power consumption is continuously increased, the utilization efficiency of power generation and power supply equipment is low, and the power saving potential is huge. In order to optimize the power utilization mode, improve terminal power regulation and control, change the unilateral dependence and expand the power plant, the electric wire netting construction satisfies the mode that the power consumption grows, further promote energy saving and emission reduction and reduce environmental pollution, construct energy efficiency public service platform and the carrier that supports energy saving service and imperatively.

The cloud computing technology is used as a novel network operation mode, can provide resource sharing and service of computing, storage, network and software for various network applications with low cost and high efficiency, is a new stage of development of distributed computing, parallel computing and grid computing, and is one of important trends of development of computer science. By means of cloud computing, the organization can construct various energy use modes of personalized, active and integrated innovation, so that energy regulation and control work can be developed more effectively, and the method is a technical development trend of energy efficiency service and guidance in the future.

The power system regulation and control cloud platform is a key technology for improving the power energy management capability and guaranteeing the regulation and control safety, and has important significance for the future high-speed development of power production regulation and control. At present, big data gradually plays a role in the power field, a plurality of fields start to define own system architecture, specific software required in power regulation is researched and developed on the system architecture, but the software researched and developed based on the architecture has the problems of weak basic service capability, low efficiency, poor platform on-demand regulation capability and the like, and can not meet the multi-level, diversified and continuous requirement change of different power management and power users. The requirements include (1) the requirement of rapid elastic expansion of basic resources of a support platform, (2) the requirement of efficient operation and global management of an analysis decision center, and (3) the requirement of rapid construction, agile delivery and convenient operation and maintenance of an application.

Therefore, a cloud platform and a container management component thereof for an electric power system are required to be designed so as to provide technical support for development, deployment and operation environments of various applications built on the platform and meet the requirements of quick construction, agile delivery and convenient operation and maintenance of the applications.

Disclosure of Invention

In order to improve the management and regulation level of a smart grid, the invention provides an operation analysis platform of a power system regulation cloud power grid, which comprises the following components:

the energy efficiency service cloud system is used for acquiring and monitoring energy efficiency information, providing diagnosis analysis of energy efficiency parameters and realizing electric energy efficiency cloud regulation and control and cloud service for enterprise users through the cloud product;

The platform service layer comprises a supporting platform, a public component management platform and a cloud container engine platform, wherein the supporting platform comprises a model data platform, a big data platform, an operation data platform and a data exchange platform, the public component management platform comprises a diary management component, an alarm management component, a permission management component, an internal and wide area message bus management component and an internal and wide area service bus management component, and the cloud container engine platform comprises a mirror image construction module, a private mirror image warehouse, an application arrangement module and a PaaS platform;

The IaaS layer is an infrastructure-level service layer and comprises a server resource pool, a storage resource pool and a network resource pool, and is used for providing a basic data solution for an application and providing a cloud function of a regulation and control cloud platform;

The external connection layer is used for interfacing with databases of a plurality of related platforms, acquiring data, storing related information of the power system and providing information data for power regulation;

the communication layer is used for connecting the external connection layer with the IaaS layer and providing an information transmission channel among the platform, each user and each system, wherein the information transmission channel comprises an electric power information network, the Internet and a mobile Internet;

and the display layer is used for providing a unified access entry for users and visually displaying the SaaS layer, the platform service layer and the functions and services supported by the IaS layer.

In order to realize application development, deployment and operation and maintenance modes based on a shared service architecture under a regulation and control system, the invention breaks through a critical technology of regulation and control cloud application standardization management based on a container on the basis of in-depth researching a container resource management technology, and constructs an omnibearing technical support system based on system application development, deployment and operation environment of the container. The method comprises the steps of taking an application as a core to research a lightweight cluster management technology based on a container, realizing dynamic allocation of container resources as required, researching an application management technology based on a container mirror warehouse, realizing automatic construction, release, downloading and deployment of the application based on the container mirror, researching an application full life cycle management technology based on the container, constructing an automatic application operation hosting environment, researching an application fault monitoring and safety protection strategy aiming at the container environment, and realizing safe and reliable operation of the application.

The beneficial effects of the invention include:

Firstly, the invention is based on regulating and controlling the cloud platform, combines the characteristics of the emerging technology and the traditional dispatching automation technology, and the container management platform promotes the utilization efficiency of resources, meets the processing capacity requirements of a system in the sudden peak of business and load, and realizes higher standards in the aspects of high efficiency, rationality, safety and the like of the utilization of platform resources besides meeting the requirements of mass data distributed processing and service.

Secondly, under the dispatching control system architecture based on the shared service, the whole process support of deployment, upgrading, capacity expansion, rollback, offline and the like of the power grid analysis decision type application can be realized based on the resource management and application operation management key technology of the container. The method has the advantages that a set of application development and operation environment for covering development, deployment, trial running and operation links is built, quick construction, agile delivery and convenient operation and maintenance of the power grid analysis decision-making application can be realized, and the support is provided for continuously improving the immediate sharing capability of the regulation and control information of the dispatching control system, complex logic processing, distributed computing and continuous reliable service capability of the application and the on-demand access capability of the application service.

And thirdly, the standardized application management construction mode and the container management method based on the containers are beneficial to perfecting the national provincial regulation analysis decision-making application standardized construction and homogenization management, forming a centralized, unified, work-coordinated and standard and efficient 'large-operation' system, and further improving the regulation and control capability of the large-scale power grid and the capability of optimizing and configuring resources in a large range.

Moreover, a containerized regulation cloud platform based on application development, deployment and operation environment omnibearing support is constructed by utilizing a Docker container technology, so that the rapid construction, agile delivery and convenient operation and maintenance of container resource on-demand dynamic allocation and power grid analysis decision-making application are realized, the support is provided for continuously improving the immediate sharing capability of regulation information of a new generation dispatch control system, and the complex logic processing, distributed computation and continuous reliable service capability of the application are further improved, and the regulation capability of driving a large power grid and the capability of optimizing and configuring resources in a large range are further improved. And the virtual machine in the physical machine with high load rate can be dynamically migrated to the physical machine with low load rate by the dynamic resource allocation method of the virtual machine, so that the aim of balancing the load of each physical machine in the virtual machine cluster can be fulfilled.

And finally, establishing a resource-saving scheduling system construction mode. On the premise of fully considering the safety, in the aspects of system management, system operation, equipment construction and the like, IT system resources are changed from a mode of 'need-for-use' to a mode of 'need-for-use', the utilization efficiency of system resources such as communication, networks, hardware and the like is improved, and a new saving type construction mode is formed.

Drawings

FIG. 1 is a frame diagram of a platform of the present invention.

Detailed Description

For a better understanding of the invention, the system of the invention is further described below with reference to the description of embodiments in conjunction with the accompanying drawings.

Numerous specific details are set forth in the following detailed description in order to provide a thorough understanding of the invention. It will be appreciated, however, by one skilled in the art that the invention may be practiced without such specific details. In embodiments, well-known methods, procedures, and components have not been described in detail so as not to unnecessarily obscure embodiments.

Referring to fig. 1, the invention provides an operation analysis platform for a power system regulation cloud power grid, which comprises:

Preferably, the PaaS platform includes a container management module, where the container management module includes at least a monitoring unit and an equalizing unit:

the monitoring unit is used for carrying out lightweight container cluster monitoring on the regulated cloud platform container resources to obtain cluster monitoring information, and specifically comprises:

a monitoring index unit for setting the aggregation index under the multi-dimensional angle as a monitoring index according to the dynamic property of the container cluster nodes,

The real-time monitoring unit is used for monitoring the container cluster nodes and the performance indexes of the containers on the nodes in real time to obtain cluster monitoring information;

The balancing unit is used for carrying out dynamic balancing allocation on the cluster resources of the Docker container of the regulated cloud platform, and specifically comprises the following steps:

The data pulling unit is used for pulling data according to the cluster monitoring information in the same network;

The evaluation unit is used for evaluating the resource use condition of the container on the cluster node and the cluster load degree of the container;

the distribution unit is used for carrying out uniform distribution of the regulated cloud platform dock container cluster resources according to the regulated cloud platform container cluster node dynamic capacity expansion/contraction strategy.

Preferably, the allocation unit is configured to perform balanced allocation of cluster resources of the regulated cloud platform dock container according to a dynamic capacity expansion/contraction strategy of the regulated cloud platform cluster nodes, and specifically includes:

And the capacity expansion/contraction Rong Zi unit is used for comprehensively calculating various monitoring indexes, expanding the application and adding a new container if the calculation result is larger than the capacity expansion threshold value, starting the same mirror image, adding the container to the Docker container cluster node, and removing the container from the Docker container cluster node if the calculation result is smaller than the capacity contraction threshold value, and reducing the existing container by carrying out capacity contraction on the application.

Preferably, the monitoring unit is used for performing lightweight container cluster monitoring on the regulated cloud platform container resources to obtain cluster monitoring information, and the visual analysis unit is used for periodically analyzing the cluster monitoring information and visually displaying the analyzed data.

Preferably, the IaaS layer further provides virtual machine cluster resources, where the virtual machine cluster has a plurality of physical machines, and each physical machine corresponds to at least one virtual machine.

Preferably, the container management module further includes a virtual machine balancing unit, configured to dynamically allocate load of the virtual machine cluster, and specifically includes:

The computing unit is used for computing the use resource weight of each virtual machine, the use resource weight of each physical machine and the average use resource weight of the physical machines;

The judging unit is used for judging the difference value between the physical machine using resource weight and the physical machine average using resource weight;

the execution unit is configured to, when a difference between a resource weight used by any entity machine and an average resource weight used by the entity machine is higher than an equilibrium threshold, specifically include:

the first determining subunit is used for determining the entity machine corresponding to the maximum entity machine use resource weight as an equilibrium source machine;

The second determining subunit is used for finding out the entity machine corresponding to the resource weight used by the minimum entity machine as the equalization destination machine;

The balance difference subunit is used for calculating a balance difference value between the physical machine use resource weight of the balance source machine and the average use resource weight of the physical machine;

The third determining subunit is configured to find, in all balanced source machines, a virtual machine corresponding to the virtual machine usage resource weight closest to the balanced difference value, as a balanced virtual machine;

And the uniform migration subunit is used for migrating the balanced virtual machine to the balanced destination machine.

Preferably, the virtual machine usage resource weight, the physical machine usage resource weight and the physical machine average usage resource weight are calculated according to the following formula:

α=1/P

Wherein j is the number of the physical machines, i is the number of the virtual machines, P is the total number of the physical machines of the virtual machine cluster, n is the total number of the virtual machines, v is the number of all the virtual machines of each physical machine, VM _jiRate is the virtual machine use resource ratio of the virtual machine use resource weight of the i virtual machines in the j physical machines, For the processor load rate of the i virtual machine in the j entity machines, VM _{jiRAMallocate} is the storage allocation amount of the i virtual machine in the j entity machines, HOST _jiRate is the entity machine usage resource ratio of the entity machine usage resource weight of the j entity machines, and α is the entity machine average usage resource ratio of the entity machine average usage resource weight of the entity machines.

The beneficial effects of the invention include:

Only the preferred embodiments of the present invention have been described herein, but it is not intended to limit the scope, applicability, and configuration of the invention. Rather, the detailed description of the embodiments will enable those skilled in the art to practice the embodiments. It will be understood that various changes and modifications may be made in the details without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. A power system control cloud grid operation analysis platform, characterized by comprising:

The SaaS layer is a software-level service layer used to provide software cloud application products. The energy efficiency service cloud system collects and monitors energy efficiency information, provides diagnostic analysis of energy efficiency parameters, and implements power energy efficiency cloud regulation and cloud services for enterprise users through the energy efficiency service application SaaS layer.

The platform service layer includes a support platform, a public component management platform, and a cloud container engine platform; wherein the support platform includes a model data platform, a big data platform, an operation data platform, and a data exchange platform; the public component management platform includes a diary management component, an alarm management component, a permission management component, an internal and wide area message bus management component, and an internal and wide area service bus management component; the cloud container engine platform includes an image building module, a private image repository, an application orchestration module, and a PaaS platform;

The IaaS layer is the infrastructure-level service layer, including server resource pools, storage resource pools, and network resource pools. It is used to provide basic data solutions for applications and provide cloud functions for regulating the cloud platform.

The external connection layer is used to connect with the databases of multiple related platforms and obtain data, store information related to the power system, and provide information data for power regulation;

The communication layer is used to connect the external connection layer and the IaaS layer, and provides information transmission channels between the platform and various users and systems. The information transmission channels include the power information network, the Internet, and the mobile Internet;

The display layer is used to provide a unified access portal for users and to visually display the functions and services supported by the SaaS layer, the platform service layer, and the IaaS layer;

The PaaS platform includes a container management module; the IaaS layer also provides virtual machine cluster resources, the virtual machine cluster has multiple physical machines, each physical machine corresponds to at least one virtual machine;

The container management module at least includes: a balancing unit;

The balancing unit is used to implement dynamic balanced allocation of the resources of the Docker container cluster of the control cloud platform, and the balancing unit specifically includes:

A data pulling unit is used to pull data based on cluster monitoring information in the same network;

The evaluation unit is used to evaluate the resource usage of containers on cluster nodes and the load level of the container cluster;

An allocation unit is used to regulate the balanced allocation of resources of the cloud platform Docker container cluster according to the dynamic expansion/contraction strategy of the cloud platform container cluster nodes;

The container management module also includes: a virtual machine balancing unit, which is used to dynamically distribute the load of the virtual machine cluster, specifically including:

A calculation unit, used to calculate the resource usage weight of each virtual machine, the resource usage weight of each physical machine and the average resource usage weight of the physical machines;

A determination unit, configured to determine a difference between the resource weight used by the physical machine and the average resource weight used by the physical machine;

The execution unit is configured to, when the difference between the resource usage weight of any entity machine and the average resource usage weight of the entity machine is higher than the balancing threshold, specifically include:

A first determining subunit is used to determine a physical machine corresponding to a maximum physical machine resource usage weight as a balancing source machine;

The second determination subunit is used to find the physical machine corresponding to the minimum physical machine resource usage weight as the balancing destination machine;

A balance difference subunit, used for calculating a balance difference between the resource usage weight of the physical machine of the balance source machine and the average resource usage weight of the physical machine;

A third determining subunit is used to find out, from all the balancing source machines, a virtual machine corresponding to the virtual machine resource usage weight closest to the balancing difference value as the balancing virtual machine;

A migration subunit, used to migrate the balanced virtual machine to the balanced destination machine;

The virtual machine resource usage weight, the physical machine resource usage weight and the physical machine average resource usage weight are calculated according to the following formula:

Where j is the physical machine number, i is the virtual machine number, P is the total number of physical machines in the virtual machine cluster, n is the total number of virtual machines, and v is the number of all virtual machines on each physical machine. is the virtual machine resource usage ratio of the virtual machine resource usage weight of the i virtual machine in the j physical machine, is the processor load rate of virtual machine i in physical machine j, is the storage allocation for virtual machine i in physical machine j, is the physical machine resource usage ratio of the physical machine resource usage weight of the j physical machine, and α is the physical machine average resource usage ratio of the average resource usage weight of the physical machine.

2. The platform according to claim 1, wherein the container management module at least comprises a monitoring unit:

The monitoring unit is used to implement lightweight container cluster monitoring on the container resources of the control cloud platform to obtain cluster monitoring information. The monitoring unit specifically includes:

The monitoring indicator unit is used to set the aggregated indicators from multiple dimensions as monitoring indicators according to the dynamics of the container cluster nodes.

The real-time monitoring unit is used to implement real-time monitoring of the performance indicators of the container cluster nodes and the containers on the nodes to obtain cluster monitoring information.

3. The platform according to claim 1, wherein the allocation unit is used to regulate the balanced allocation of cloud platform Docker container cluster resources according to the dynamic expansion/contraction strategy of the cloud platform container cluster nodes, specifically comprising:

The expansion/reduction subunit is used to comprehensively calculate various monitoring indicators. If the calculation result is greater than the expansion threshold, the application is expanded and a new container is added, the same image is started, and the container is added to the Docker container cluster node; if the calculation result is less than the reduction threshold, the application is reduced and the existing containers are reduced, and the container is removed from the Docker container cluster node.

4. The platform according to claim 2, wherein the monitoring unit is used to implement lightweight container cluster monitoring on the container resources of the control cloud platform to obtain cluster monitoring information, and also includes: a visualization analysis unit, which is used to perform periodic analysis on the cluster monitoring information and visualize the analyzed data.