Power grid operation simulation environment resource integration and sharing method, system and device based on regulation cloud architecture and medium

Abstract

The application discloses a kind of based on the power grid operation simulation environment resource integration sharing method, system, device and medium of regulation and control cloud architecture, the present application is through directory construction and label annotation, the attribute characteristics of simulation environment resource are enriched, help user fast positioning.And through data configuration and parameter configuration, the individualization of simulation environment resource is realized, on the one hand, the resource storage pressure is reduced, on the other hand, the matching degree of resource and user actual demand is improved;Finally, through training mode selection, the access of multiple training environments is realized, and the use difficulty of user is reduced.The power grid operation simulation environment resource of reinforcement learning training realized by the present application can be integrated and shared in two-level regulation and control cloud framework, so as to realize the reuse of environment construction achievement, promote the application of artificial intelligence technologies such as reinforcement learning in regulation and control field.

Description

Technical Field

[0001] This invention belongs to the field of power automation technology and relates to a method, system, device and medium for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture. Background Technology

[0002] With the rapid development of data-driven artificial intelligence technologies such as deep learning, reinforcement learning, and knowledge graphs, and relying on two-level control cloud platforms, research on the application of artificial intelligence technology in the field of power grid control has made some progress. As the power system undergoes rapid and profound changes, the scale of control is growing exponentially, the characteristics of controlled objects vary greatly, and uncertainty on both the source and load sides is increasing, making real-time power grid dispatching more complex and frequent. Reinforcement learning, as a machine learning paradigm and methodology, is used to describe and solve problems where intelligent agent algorithms learn strategies to maximize rewards or achieve specific goals during interactions with the environment, making it very suitable for application in the field of power grid dispatching and control. Reinforcement learning involves two main roles: "intelligent agent" and "environment." Currently, a certain scale of power grid simulation environments based on reinforcement learning have been built on the control cloud, but these are mainly established independently by various provinces and cities, resulting in inconsistent technical standards, low reusability of results, and a lack of a unified integration and sharing mechanism. Summary of the Invention

[0003] The purpose of this invention is to solve the problems of inconsistent technical standards for control cloud architecture in various provinces and cities, low reusability of results, and lack of a unified integration and sharing mechanism in the existing technology, and to provide a method, system, device and medium for resource integration and sharing of power grid operation simulation environment based on control cloud architecture.

[0004] To achieve the above objectives, the present invention employs the following technical solution:

[0005] A method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture includes:

[0006] Receive simulators sent from the cloud in various provinces and regions;

[0007] The received simulators are reviewed based on the test script to determine whether they conform to the preset rules.

[0008] The system interacts with a simulator that conforms to preset rules based on a benchmark intelligent agent algorithm to obtain the simulator's performance metrics.

[0009] Based on the simulator's performance metrics and user needs, a suitable simulator is selected; and based on preset label evaluation rules, a label library is called to label the selected simulator.

[0010] Based on the annotated simulator, the simulation environment resource catalog is updated, and at the same time, the catalog information update message and the catalog subscription service call instruction are published to the provincial and local cloud through the cloud message bus and cloud service bus respectively.

[0011] Store the samples in the simulator into the sample library in Kokubun Cloud, and store the remaining parts in the simulator into the simulator library in Kokubun Cloud.

[0012] Based on directory information update messages and commands to call the directory subscription service, data sharing between simulators in the provincial and local cloud environments is achieved.

[0013] A further improvement of the present invention is that:

[0014] Furthermore, the simulator includes an observation space, an action space, operational constraints, a reward function, continuously running data samples, and a power flow calculation program. The observation space is the set of information exposed when the simulator is invoked, where an instance in the observation space is an observation state. The action space is the set of actions executed when the simulator is invoked, where an instance in the action space is an action strategy. The operational constraints are the set of constraints satisfied between the observation state and the action strategy when the simulator is invoked. The reward function is the contribution of selecting a certain action strategy to achieving the goal under a certain observation state. The continuously running data samples are the set of data samples invoked by the simulator. The power flow calculation program is the calculation program run by the simulator.

[0015] Furthermore, the emulator supports both image files and source code files.

[0016] Furthermore, the received simulators are reviewed based on the test script to determine whether they conform to preset rules, specifically:

[0017] Check if the compressed file format in the simulator is .zip or .rar. After automatic decompression, check if it contains the observation space file, action space file, running constraint file, reward function file, continuous running data sample folder, and power flow calculation program folder.

[0018] Furthermore, the simulator's performance metrics are: average score per game, average time per game, and average number of steps per round; the benchmark agent algorithm includes: flexible action-evaluation algorithm, Q-learning algorithm, and deep Q-learning algorithm.

[0019] Furthermore, based on preset label evaluation rules, the label library is called to label the selected simulators. This also includes: when the provincial cloud sends the simulators, it attaches label information, reviews the label information, and includes it in the label library.

[0020] Furthermore, based on directory information update messages and commands to invoke the directory subscription service, data sharing is achieved between the simulators in the provincial and local cloud environments, specifically as follows:

[0021] The provincial cloud platform uses the cloud service bus to call the directory subscription service command and directory information update message to obtain the full simulation environment resource directory for updating, and then searches or performs tag retrieval to locate the required simulator;

[0022] Based on business needs, select the data size from the continuously running data sample, pull samples from the sample library in the cloud, and inject them into the simulator.

[0023] Configure module parameters in the provincial cloud platform, and modify corresponding parameters in the observation space, action space, operational constraints, and reward function modules in the national cloud platform.

[0024] The provincial and local cloud platforms select online and offline training modes for data sharing.

[0025] Furthermore, the size of the sample data can be selected by setting the sample interval or the number of samples; the sample interval can be set to 5 minutes, 10 minutes or 1 hour; the number of samples can be set to 30%, 50%, 70% or the full amount of the overall sample.

[0026] Furthermore, the provincial and local cloud platforms select both online and offline training modes for data sharing, specifically as follows:

[0027] Select the offline training mode in the provincial cloud platform, choose the resource file format, download the resource file format to your local machine, and import it into the local training environment. If the provincial cloud platform includes an artificial intelligence platform, import the downloaded resource file format into the artificial intelligence platform of the provincial cloud platform for environment training.

[0028] The provincial cloud platform selects online mode, loads the AI ​​platform training environment from the national cloud platform, and automatically imports the simulator to support the subsequent training process.

[0029] A power grid operation simulation environment resource integration and sharing system based on a control cloud architecture includes:

[0030] The receiving module receives simulators sent from the cloud in various provinces and regions;

[0031] The review module reviews the received simulator based on the test script to determine whether the simulator conforms to preset rules.

[0032] An interaction module, which interacts with a simulator that conforms to preset rules based on a benchmark intelligent agent algorithm to obtain the simulator's performance indicators;

[0033] The annotation module selects a suitable simulator based on the simulator's performance indicators and user needs; and, based on preset label evaluation rules, calls the label library to annotate the selected simulator.

[0034] The update module updates the simulation environment resource catalog based on the labeled simulator, and simultaneously publishes catalog information update messages and calls catalog subscription service instructions to the provincial and local cloud through the cloud message bus and cloud service bus, respectively.

[0035] The storage module stores the samples in the simulator into the sample library in the Guofen Cloud, and stores the remaining parts in the simulator into the simulator library in the Guofen Cloud.

[0036] The sharing module, based on directory information update messages and directory subscription service call instructions, completes data sharing of the simulator in the provincial cloud.

[0037] A terminal device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described above.

[0038] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] This invention interacts with rule-compliant simulators using a benchmark intelligent agent algorithm to obtain the simulator's performance metrics. It then tags the selected simulators and updates the simulation environment resource directory. Simultaneously, it publishes directory update messages and invokes directory subscription services to provincial and regional cloud platforms via cloud message bus and cloud service bus, respectively. This invention enriches the attribute characteristics of simulation environment resources through directory construction and tagging, helping users quickly locate them.

[0041] Furthermore, this invention enables personalized customization of simulation environment resources through data and parameter configuration, reducing resource storage pressure and improving the matching degree between resources and actual user needs. Finally, by selecting training modes, it allows access to multiple training environments, reducing the difficulty of use for users. This invention enables the integration and sharing of power grid operation simulation environment resources trained by reinforcement learning within a two-level control cloud framework, thereby achieving the reuse of environment construction results and promoting the application of artificial intelligence technologies such as reinforcement learning in the field of power control. Attached Figure Description

[0042] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating the resource integration and sharing method for a power grid operation simulation environment based on a control cloud architecture, as described in this invention.

[0044] Figure 2 This is a schematic diagram of the simulator's architecture;

[0045] Figure 3 This is a schematic diagram of the power grid operation simulation environment resource integration and sharing system based on the control cloud architecture of the present invention;

[0046] Figure 4 This is a schematic diagram of the resource integration and sharing structure of the power grid operation simulation environment based on the control cloud architecture of the present invention. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0048] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0049] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0050] In the description of the embodiments of the present invention, it should be noted that if terms such as "upper," "lower," "horizontal," or "inner" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of the invention is in use, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, terms such as "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0051] Furthermore, the use of the term "horizontal" does not imply that the component must be absolutely horizontal, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0052] In the description of the embodiments of the present invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention according to the specific circumstances.

[0053] The present invention will now be described in further detail with reference to the accompanying drawings:

[0054] See Figure 1 This invention discloses a method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture, comprising:

[0055] S101 receives emulators sent from the cloud in various provinces and regions;

[0056] See Figure 2The simulator includes an observation space, an action space, operational constraints, a reward function, continuously running data samples, and a power flow calculation program. The observation space is the set of information exposed when the simulator is invoked, where an instance in the observation space represents an observation state. The action space is the set of actions executed when the simulator is invoked, where an instance in the action space represents an action strategy. The operational constraints are the set of constraints satisfied between the observation state and the action strategy when the simulator is invoked. The reward function is the contribution of choosing a particular action strategy to achieving the objective under a given observation state. The continuously running data samples are the set of data samples invoked by the simulator. The power flow calculation program is the calculation program run by the simulator. The simulator supports both image files and source code files.

[0057] S102, Based on the test script, the received simulator is reviewed to determine whether the simulator meets the preset rules.

[0058] Check if the compressed file format in the simulator is .zip or .rar. After automatic decompression, check if it contains the observation space file, action space file, running constraint file, reward function file, continuous running data sample folder, and power flow calculation program folder.

[0059] S103 interacts with a rule-compliant simulator based on a benchmark intelligent agent algorithm to obtain the simulator's performance metrics.

[0060] The simulator's performance metrics are: average score per game, average time per game, and average number of steps per round; the benchmark agent algorithm includes: flexible action-evaluation algorithm, Q-learning algorithm, and deep Q-learning algorithm.

[0061] S104: Based on the simulator's performance indicators and user needs, select a suitable simulator; and based on preset label evaluation rules, call the label library to label the selected simulator.

[0062] The provincial cloud platform attaches tag information along with the simulator, reviews the tag information, and adds it to the tag library.

[0063] S105, based on the annotated simulator, updates the simulation environment resource catalog, and simultaneously publishes catalog information update messages and calls catalog subscription service instructions to the provincial and local cloud through the cloud message bus and cloud service bus, respectively.

[0064] S106, store the samples in the simulator into the sample library in Kokubun Cloud, and store the remaining parts in the simulator into the simulator library in Kokubun Cloud.

[0065] S107, based on directory information update messages and commands to call the directory subscription service, completes data sharing between the simulators in the provincial and local cloud environments.

[0066] The provincial cloud platform uses the cloud service bus to call the directory subscription service command and directory information update message to obtain the full simulation environment resource directory for updating, and then searches or performs tag retrieval to locate the required simulator;

[0067] Based on business needs, select the data size from the continuously running data sample, pull samples from the sample library in the cloud, and inject them into the simulator.

[0068] Configure module parameters in the provincial cloud platform, and modify corresponding parameters in the observation space, action space, operational constraints, and reward function modules in the national cloud platform.

[0069] The provincial and local cloud platforms select online and offline training modes for data sharing.

[0070] The provincial and local cloud platforms select both online and offline training modes for data sharing, specifically as follows:

[0071] Select the offline training mode in the provincial cloud platform, choose the resource file format, download the resource file format to your local machine, and import it into the local training environment. If the provincial cloud platform includes an artificial intelligence platform, import the downloaded resource file format into the artificial intelligence platform of the provincial cloud platform for environment training.

[0072] The provincial cloud platform selects online mode, loads the AI ​​platform training environment from the national cloud platform, and automatically imports the simulator to support the subsequent training process.

[0073] See Figure 3 This invention discloses a power grid operation simulation environment resource integration and sharing system based on a control cloud architecture, comprising:

[0074] The receiving module receives simulators sent from the cloud in various provinces and regions;

[0075] The review module reviews the received simulator based on the test script to determine whether the simulator conforms to preset rules.

[0076] An interaction module, which interacts with a simulator that conforms to preset rules based on a benchmark intelligent agent algorithm to obtain the simulator's performance indicators;

[0077] The annotation module selects a suitable simulator based on the simulator's performance indicators and user needs; and, based on preset label evaluation rules, calls the label library to annotate the selected simulator.

[0078] The update module updates the simulation environment resource catalog based on the labeled simulator, and simultaneously publishes catalog information update messages and calls catalog subscription service instructions to the provincial and local cloud through the cloud message bus and cloud service bus, respectively.

[0079] The storage module stores the samples in the simulator into the sample library in the Guofen Cloud, and stores the remaining parts in the simulator into the simulator library in the Guofen Cloud.

[0080] The sharing module, based on directory information update messages and directory subscription service call instructions, completes data sharing of the simulator in the provincial cloud.

[0081] Example:

[0082] See Figure 4 This invention discloses a resource integration and sharing structure for a power grid operation simulation environment based on a control cloud architecture, which mainly includes an integration mechanism and a sharing mechanism.

[0083] The integration mechanism mainly refers to the process of simulation environment resources flowing from provincial and regional cloud platforms to the national cloud platform, including the following steps:

[0084] (1) The simulators of each province and region are uploaded to the cloud. The file formats supported are image files (.tar.gz) and source code files (.zip, .rar).

[0085] (2) Run the test script on the cloud platform to review the completeness and standardization of the uploaded simulator, and review the file format and naming conventions of each module of the simulator. Taking the source code file as an example, check whether the compressed package file format is .zip or .rar. After automatic decompression, check whether it has the observation space (observation / Observation.py file), action space (action / Act.py file), running constraints (constraint / Constraint.py file), reward function (reward / Rewards.py file), continuously running data samples (data folder), and power flow calculation program (ulize folder).

[0086] (3) Run the performance test script on the Kokubun cloud platform, and interact with the simulator using the benchmark intelligent agent algorithm to obtain performance indicators. Performance indicators include, but are not limited to, average score per game, average time per game, and average number of steps per round. The benchmark intelligent agent algorithm includes, but is not limited to, Soft Actor-Critic (SAC) algorithm, Q-learning algorithm, and Deep Q-learning (DQN) algorithm.

[0087] (4) The National Cloud Platform automatically calls the tag library to label the simulator according to the preset tag evaluation rules. It supports the simultaneous addition of tag information when provincial and local cloud platforms send simulation environment resources. The National Cloud Platform reviews the tag information and includes it in the tag library.

[0088] (5) The National Cloud updates the simulation environment resource catalog based on the tagged simulator and publishes the catalog information update message on the provincial cloud through the cloud message bus.

[0089] (6) Guofen Cloud stores the samples in the simulator into Guofen Cloud's sample library, and the remaining samples into Guofen Cloud's simulator library.

[0090] The sharing mechanism mainly refers to the process of simulation environment resources flowing from the national cloud to the provincial and regional clouds, including the following steps:

[0091] (1) The provincial cloud platform calls the directory subscription service through the cloud service bus to obtain the full simulation environment resource directory and search for it, or perform tag retrieval to locate the required simulator.

[0092] (2) The provincial cloud platform selects the data size in the continuously running data sample module according to business needs. Correspondingly, the national cloud platform pulls samples from the sample library and injects them into the simulator. The sample size can be selected by setting the sample interval or the number of samples.

[0093] The sample interval can be set to 5 minutes / 10 minutes / 1 hour; the number of samples can be set to 30% / 50% / 70% / full sample.

[0094] (3) Configure the module parameters in the provincial cloud platform, and correspondingly modify the corresponding parameters in the national cloud platform modules such as observation space, action space, operation constraints, and reward function. For example, the unit output ramp-up coefficient in the operation constraint module, and the coefficients of each reward item in the reward function.

[0095] (4) Select training mode in the cloud, including online and offline methods.

[0096] If you select offline mode:

[0097] (5) Select the resource file format, including two forms: image (.tar.gz) and source code (.zip, .rar).

[0098] (6) Download the resource file format to your local machine and import it into your local training environment. If the provincial cloud platform has been deployed with an artificial intelligence platform, you can import the artificial intelligence platform training environment from the provincial cloud platform.

[0099] For example, if you choose the online mode:

[0100] (7) Load the artificial intelligence platform training environment of Kokubun Cloud and automatically import the simulator to support the subsequent training process.

[0101] An embodiment of the present invention provides a terminal device. This terminal device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the various method embodiments described above. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the various device embodiments described above.

[0102] The computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention.

[0103] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0104] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0105] The memory can be used to store the computer program and / or module. The processor implements various functions of the terminal device by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0106] If the modules / units integrated into the terminal device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0107] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0108] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture, characterized in that: include: Receive simulators sent from the cloud in various provinces and regions; The received simulators are reviewed based on the test script to determine whether they conform to the preset rules. The system interacts with a simulator that conforms to preset rules based on a benchmark intelligent agent algorithm to obtain the simulator's performance metrics. Based on the simulator's performance metrics and the user's needs, select an appropriate simulator; Based on preset label evaluation rules, the system calls the label library to label the selected simulator. Based on the annotated simulator, the simulation environment resource catalog is updated, and at the same time, the catalog information update message and the catalog subscription service call instruction are published to the provincial and local cloud through the cloud message bus and cloud service bus respectively. Store the samples in the simulator into the sample library in Kokubun Cloud, and store the remaining parts in the simulator into the simulator library in Kokubun Cloud. Based on directory information update messages and commands to call the directory subscription service, data sharing between simulators in the provincial and local cloud environments is achieved.

2. The method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture as described in claim 1, characterized in that, The simulator includes an observation space, an action space, operational constraints, a reward function, continuously running data samples, and a power flow calculation program. The observation space is the set of information exposed when the simulator is invoked, where an instance in the observation space represents an observation state. The action space is the set of actions executed when the simulator is invoked, where an instance in the action space represents an action strategy. The operational constraints are the set of constraints satisfied between the observation state and the action strategy when the simulator is invoked. The reward function is the contribution of selecting a particular action strategy to achieving the objective under a given observation state. The continuously running data samples are the set of data samples invoked by the simulator. The power flow calculation program is the calculation program run by the simulator.

3. The method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture as described in claim 2, characterized in that, The emulator supports both image files and source code files.

4. The method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture as described in claim 3, characterized in that, The process of reviewing the received simulator based on the test script to determine whether the simulator conforms to preset rules specifically involves: Check if the compressed file format in the simulator is .zip or .rar. After automatic decompression, check if it contains the observation space file, action space file, running constraint file, reward function file, continuous running data sample folder, and power flow calculation program folder.

5. The method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture according to claim 4, characterized in that, The performance metrics of the simulator are: average score per game, average time per game, and average number of moves per round. The baseline intelligent agent algorithm includes: flexible action-evaluation algorithm, Q-learning algorithm and deep Q-learning algorithm.

6. The method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture as described in claim 5, characterized in that, The process of calling the tag library based on preset tag evaluation rules to label the selected simulator also includes: attaching tag information when sending the simulator to the provincial cloud platform, reviewing the tag information, and including it in the tag library.

7. The method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture as described in claim 6, characterized in that, The data sharing between the simulator in the provincial cloud based on directory information update messages and directory subscription service invocation instructions is as follows: The provincial cloud platform uses the cloud service bus to call the directory subscription service command and directory information update message to obtain the full simulation environment resource directory for updating, and then searches or performs tag retrieval to locate the required simulator; Select the data size from the continuously running data sample according to business needs, pull the sample from the sample library in the cloud and inject it into the simulator; Configure module parameters in the provincial cloud platform, and modify corresponding parameters in the observation space, action space, operational constraints, and reward function modules in the national cloud platform. The provincial and local cloud platforms select online and offline training modes for data sharing.

8. The method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture as described in claim 7, characterized in that, The size of the sample data is selected by setting the sample interval or the number of samples; the sample interval is set to 5 minutes, 10 minutes or 1 hour; the number of samples is set to 30%, 50%, 70% or the full amount of the overall sample.

9. The method for resource integration and sharing in a power grid operation simulation environment based on a control cloud architecture as described in claim 8, characterized in that, The provincial cloud platform selects online and offline training modes for data sharing, specifically as follows: Select the offline training mode in the provincial cloud platform, choose the resource file format, download the resource file format to your local machine, and import it into the local training environment. If the provincial cloud platform includes an artificial intelligence platform, import the downloaded resource file format into the artificial intelligence platform of the provincial cloud platform for environment training. The provincial cloud platform selects online mode, loads the AI ​​platform training environment from the national cloud platform, and automatically imports the simulator to support the subsequent training process.

10. A power grid operation simulation environment resource integration and sharing system based on a control cloud architecture, characterized in that: include: The receiving module receives simulators sent from the cloud in various provinces and regions; The review module reviews the received simulator based on the test script to determine whether the simulator conforms to preset rules. An interaction module, which interacts with a simulator that conforms to preset rules based on a benchmark intelligent agent algorithm to obtain the simulator's performance indicators; The annotation module selects a suitable simulator based on the simulator's performance indicators and the user's needs. Based on preset label evaluation rules, the system calls the label library to label the selected simulator. The update module updates the simulation environment resource catalog based on the labeled simulator, and simultaneously publishes catalog information update messages and calls catalog subscription service instructions to the provincial and local cloud through the cloud message bus and cloud service bus, respectively. The storage module stores the samples in the simulator into the sample library in the Guofen Cloud, and stores the remaining parts in the simulator into the simulator library in the Guofen Cloud. The sharing module, based on directory information update messages and directory subscription service call instructions, completes data sharing of the simulator in the provincial cloud.

11. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-9.

12. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-9.

Images