Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

204 results about "System requirements" patented technology

To be used efficiently, all computer software needs certain hardware components or other software resources to be present on a computer. These prerequisites are known as (computer) system requirements and are often used as a guideline as opposed to an absolute rule. Most software defines two sets of system requirements: minimum and recommended. With increasing demand for higher processing power and resources in newer versions of software, system requirements tend to increase over time. Industry analysts suggest that this trend plays a bigger part in driving upgrades to existing computer systems than technological advancements. A second meaning of the term of System requirements, is a generalisation of this first definition, giving the requirements to be met in the design of a system or sub-system.

Power grid dispatching method and system adapting to requirements of power system

The invention discloses a power grid dispatching method and system adapting to power system requirements, and relates to the field of industrial big data, and the method comprises the following operation steps: S1, multi-source heterogeneous data collection and edge preprocessing; s2, knowledge graph construction and data fusion; s3, load prediction and renewable energy output prediction based on deep learning; s4, generating a dynamic optimization scheduling strategy; s5, carrying out security and credible execution on the data endowed by the block chain; and S6, real-time monitoring and closed-loop feedback optimization are carried out. According to the power grid scheduling method and system adapting to the power system demand, the scheduling method integrates edge calculation, block chain, deep learning and reinforcement learning, can realize multi-source data real-time processing, dynamic optimization strategy generation and data security and credibility, improves the power grid operation efficiency, stability and renewable energy consumption capability, and improves the power grid scheduling efficiency. And the dynamically optimized scheduling strategy can reduce the operation cost of the power grid, and can reduce carbon emission at the same time.
Owner:INNER MONGOLIA FINANCE AND ECONOMICS UNIVERSITY

Automatic generation of personalized collection of items around a theme at an online system

An online system automatically generates a personalized collection of items around a theme. The online system generates a prompt for input into a language model, the prompt including information about a plurality of items and a text describing the theme around which the collection of items will be built. The online system requests the language model to generate, based on the prompt, a list of products eligible for building the collection of items. The online system accesses a computer model trained to identify a set of items personalized for a user of the online system. The computer model identifies, based on the list of eligible products and information about the user, the set of items for populating the collection of items. The online system causes a device of the user to display a user interface with the collection of items for inclusion into a cart of the user.
Owner:MAPLEBEAR INC

Virtual power plant optimization system and method

The invention discloses a virtual power plant optimization system and method, and the system comprises a virtualization platform module which is used for integrating distributed energy, energy storage equipment, an electric vehicle, a flexible load and other resources, and achieves the virtualization and centralized management of the resources; the dynamic resource aggregation module is used for dynamically matching resource characteristics with power system requirements; the market transaction module is used for optimizing the earnings of the virtual power plant; and the core algorithm module is used for optimizing the load prediction precision. The market transaction module implements two-stage market bidding based on a multi-agent game model, analyzes the uncertainty of quotation of competitors through a multi-scene simulation technology, optimizes a dynamic transaction strategy, and achieves the maximization of spot market benefits and the controllability of risks. Through cooperative work of a virtualization platform module, a dynamic resource aggregation module, a market transaction module, a core algorithm module, a security management module and the like, efficient management of resources and maximization of market benefits are realized.
Owner:YICHANG POWER SUPPLY CO OF STATE GRID HUBEI ELECTRIC POWER CO LTD

Artificial intelligence scheduler for task-execution systems

The technology described herein is directed towards an artificial intelligence scheduler for real-time task execution systems, including an implementation based on a recurrent learning model. In one implementation, the model incorporates a deep Q-network and proximal policy optimization to learn efficient scheduling policies for real-time systems. The scheduler can optimize for various real-time system aspects, including avoiding deadlocks, minimizing blocking, avoiding priority inversion, and reducing starvation. The artificial intelligence scheduler learns from actual system experience data to regularly improve scheduling policies and / or adapt to changing system requirements.
Owner:DELL PROD LP

Systems and methods for private network issuance of digital currency

Systems, methods, and computer-readable storage media of configuring digital currency nodes utilizing a one-tier model or two-tier model. One method includes transmitting, via an application provider system, an installation request including authentication information, downloading, by the user device, a digital wallet application, and installing, by the user device, the digital wallet application based on one or more system requirements, wherein installing includes verifying the user device meets the one or more system requirements, generating a public and private key pair, hardening the user device, storing and configuring the digital wallet application, and authenticating the digital wallet application. The method further includes establishing, via a private network, a private network connection associated with digital currency issuance. Another method includes transmitting a registration request, receiving a software package including a software appliance and executable code, installing the software appliance and establishing a private network connection associated with digital currency issuance.
Owner:WELLS FARGO BANK NA

Utilizing a Large Language Model to Modify Digital Data of an Entity in Response to Digital Data Requirements Changes

Methods, systems, and non-transitory computer readable storage media are disclosed for managing computing systems to comply with system requirements frameworks that indicate specific requirements on how the computing systems should handle certain data types. In response to detecting a change to one or more digital data requirements of a system requirements framework, the disclosed systems access a configuration profile of an entity and utilize a large language model to determine if data assets or data processing operations comply with the changes to the digital data requirements. In response to determining a configuration gap of the configuration profile based on detected changes to the digital data requirements, the disclosed systems utilize a large language model to generate tasks for correcting the configuration gap. The disclosed systems generate one or more tasks by modifying data assets, data processing operations, or other digital data associated with the entity.
Owner:ONETRUST LLC

Automatically Evaluating Manifest Files for Applications

The present disclosure provides computer-implemented methods, systems, and devices for automatically evaluating metadata associated with an application. A computing system accesses metadata for a respective application; wherein the respective application is associated with an operating system. The computing system accesses data describing requirements for the operating system. The computing system automatically generates one or more configuration rules for the metadata based on the documentation for the operating system. The computing system determines that the metadata violates one or more configuration rules based on the one or more configuration rules. The computing system transmits data describing one or more violations of the configuration rules for display.
Owner:KRYPTOWIRE INC

Facilitating modification of components of artificial intelligence computing applications via aggregated risk scores

Methods, systems, and non-transitory computer readable storage media are disclosed for facilitating modification of components of artificial intelligence computing applications via aggregated risk scores of individual artificial intelligence computing application components. The disclosed system generates data objects representing the artificial intelligence computing application components. The disclosed system determines mappings between the data objects based on relationships of the artificial intelligence computing application components. Furthermore, the disclosed system generates risk scores for the data objects representing the components of the artificial intelligence computing application by administering risk assessments that correspond to a system requirements framework. The disclosed system also generates an interactive aggregated risk indicator indicating contributions of the components to the artificial intelligence computing application in relation to the system requirements framework by combining the risk scores of the data objects according to the mappings between the data objects.
Owner:ONETRUST LLC

Method for assisting an engineer in designing a technical system, data processing apparatus and computer program

A method for assisting in designing a technical system by: receiving an input message; processing the input message using a neural network to identify a content of the input message and extract technical system requirements, wherein the technical system requirements include at least a component having at least one parameter and a constraint; generating a proposed system design by determining at least one parameter of the component based on the constraint using the neural network; generating a response message containing the proposed system design using the neural network; and providing the proposed system design, wherein the neural network is configured as a large language model, LLM, and trained with a plurality of template systems and template components in the related technical field.
Owner:AIRBUS (SAS)

End-to-end intelligent system development method and system based on AI large model

The invention provides an end-to-end intelligent system development method and system based on an AI large model. The method comprises the following steps: S1, receiving a system demand described by a user through a natural language; s2, analyzing the requirements through an AI large model and converting the requirements into formalized specifications; s3, automatically generating a system architecture and dividing components; s4, generating each component code in a recursive manner; s5, dynamically detecting and repairing the generated code problem; s6, automatically integrating all components to form a complete system; and S7, continuously optimizing the system based on system feedback or user feedback. According to the method, work of multiple stages such as demand analysis, architecture design, module division, code writing and testing is automatically completed, distributed integration and smart contract are performed, distributed integration verification is performed through a block chain smart contract and a graph database, data consistency, interface matching and security among components are ensured, the integration process is optimized, and the integration efficiency is improved. Cross-domain cooperation is achieved, and the development efficiency is high.
Owner:MOBI ZHITENG (SHANGHAI) TECHNOLOGY CO LTD

Secure system automatic design apparatus, secure system automatic design method, and computer readable medium

A secure system automatic design apparatus includes: an input / output unit that receives an input of a system requirement and output a system configuration; a configuration information concretization unit that converts the system requirement into the system configuration through a plurality of concretization procedures; and a security evaluation unit that evaluates security of the system configuration. In each of the concretization procedures, the configuration information concretization unit generates a plurality of configuration drafts, each of the configuration drafts being configuration information that is being converted and each of the configuration drafts being branched to a plurality of options of a concretization method. The security evaluation unit also evaluates security of each of the configuration drafts generated through the respective procedures. The configuration information concretization unit determines the option to be branched based on a result of the evaluation of each of the configuration drafts by the security evaluation unit.
Owner:NEC CORP

Construction method of geographic knowledge question-answering system based on knowledge graph

The invention relates to the technical field of geographic knowledge questioning and answering, and particularly discloses a construction method of a geographic knowledge questioning and answering system based on a knowledge graph, which comprises the following steps: firstly, creating a system development environment based on a browser-server architecture; establishing a system architecture comprising a data layer, an application service layer and a user layer; constructing a geographic logic query system; and finally analyzing the operation result of the geographic logic query system. According to the construction method of the geographic knowledge question-answering system based on the knowledge graph, system requirements are analyzed, and the system is divided into a data layer, an application service layer and a user layer; rich front-end interaction functions are realized by adopting an HTML (Hypertext Markup Language), a CSS (Cascading Style Sheet) A back-end algorithm module and a database storage module are realized by utilizing a Python language and a Neo4j graph database; the functions of algorithm selection, query analysis, knowledge graph matching, logic query, result display and the like are supported, and a complete process from user query input to reasoning result display is realized.
Owner:HEFEI UNIV OF TECH

Systems and methods for ensuring correct execution of computer program using mediator computer system

To provide systems and methods capable of ensuring correct execution of computer programs in a distributed computing environment.SOLUTION: In a distributed system, a first computer system is configured to: require computationally verifiable assurances of the authenticity and integrity of computations performed by a second computer system; and delegate execution of a program to the second computer system, where a protocol to make the execution of the program safer is employed to constrain the second computer system to perform correct execution of the program. The protocol may include mitigation and correction routines that mitigate and / or correct incorrect execution of the program, and the protocol may utilize a blockchain network such as a Bitcoin-based blockchain network.SELECTED DRAWING: Figure 4
Owner:NCHAIN LICENSING AG

Decision planning system-oriented simulation test scene automatic generation method

The invention discloses a decision planning system-oriented simulation test scene automatic generation method. The method comprises the steps of scene modeling: constructing a scene model by using a deep learning model; constraint generation: based on an output result of the scene model, utilizing a reinforcement learning model to generate a constraint condition; performing optimization search: based on the scene model and the constraint conditions, generating scene parameters meeting constraints by adopting a Bayesian optimization algorithm; and scene evaluation: evaluating the system performance by using the deep reinforcement learning model. According to the simulation test scene automatic generation method facing the decision planning system, manual intervention is reduced through an automatic method, and the scene generation efficiency is improved; diversified and high-coverage test scenes are generated, the scene diversity is enhanced, and the robustness of the system is comprehensively tested; constraint processing is optimized by efficiently processing complex constraint conditions, and it is ensured that a generated scene meets system requirements; and comprehensive system performance evaluation is provided, multi-dimensional evaluation is realized, and the performance of the system in different scenes is reflected.
Owner:ANHUI JIANGHUAI AUTOMOBILE GRP CORP LTD

Configuration processing method and device of hyper-converged system, storage medium and electronic equipment

The invention discloses a configuration processing method and device of a hyper-converged system, a storage medium and electronic equipment, and relates to the technical field of computer hardware, and the method comprises the steps that specified configuration information of a server is acquired, and the specified configuration information is configuration information of a basic input and output system of the server; the specified configuration information is analyzed, actual configuration values of a set of specified configuration items are obtained, and one specified configuration item in the set of specified configuration items corresponds to one function provided by the hyper-converged system; in a set of specified configuration items, under the condition that at least one configuration item with the actual configuration value inconsistent with the expected configuration value exists, the actual configuration value of the at least one configuration item is modified, and therefore the difference between BIOS configuration and the hyper-converged system requirement can be automatically detected and corrected; the configuration processing method of the hyper-converged system in the related technology has the problems of low deployment efficiency and poor reliability caused by high deployment complexity.
Owner:JINAN INSPUR DATA TECH CO LTD

Systems and methods for private network issuance of digital currency

Systems, methods, and computer-readable storage media of configuring digital currency nodes utilizing a one-tier model or two-tier model. One method includes transmitting, via an application provider system, an installation request including authentication information, downloading, by the user device, a digital wallet application, and installing, by the user device, the digital wallet application based on one or more system requirements, wherein installing includes verifying the user device meets the one or more system requirements, generating a public and private key pair, hardening the user device, storing and configuring the digital wallet application, and authenticating the digital wallet application. The method further includes establishing, via a private network, a private network connection associated with digital currency issuance. Another method includes transmitting a registration request, receiving a software package including a software appliance and executable code, installing the software appliance and establishing a private network connection associated with digital currency issuance.
Owner:WELLS FARGO BANK NA

Modeling analysis and test case automatic generation method oriented to engine software FADEC demand

The invention provides an engine software FADEC demand-oriented modeling analysis and test case automatic generation method, and belongs to the field of aero-engine control. According to the method, DeepSeek is used for assisting preprocessing at the system demand level, and modeling work based on a formalized table is completed; through multi-level demand model analysis, vulnerabilities existing in the demand are automatically detected; and finally, automatically generating a test case set meeting DO-178C airworthiness requirements by establishing a grammar logic architecture. The method focuses on formalized verification work of an engine software FADEC demand level, by using an executable model, defects appearing in the demand or design process can be effectively found, errors can be eliminated in time, test cases are automatically generated, and therefore rework and iteration cost in the later period of development is greatly reduced.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS +1

Equipment control device and method, storage medium and electronic equipment

The invention discloses an equipment control device and method, a storage medium and electronic equipment, and relates to the technical field of computers, the device comprises a plurality of switch boards, and each switch board comprises a first controller, a first logic device and a plurality of switch chips; the first controller is connected with a first port of the first logic device; a second port of the first logic device is connected with a first port of the switching chip; a third port of the first logic device is connected with the first circuit board; a second port of the switch chip is connected with a plurality of heterogeneous acceleration cards; the first controller is used for controlling a target acceleration card in the heterogeneous acceleration cards, and the target acceleration card and the first controller are connected with the same switching board. Compared with the prior art, the system requirements of a plurality of heterogeneous acceleration cards can be met, the heterogeneous acceleration cards are controlled by utilizing the controllers on the switching boards, a controller on a host does not need to be called for control, the control efficiency of the heterogeneous acceleration cards is improved, and then the performance of the whole machine is improved.
Owner:INSPUR SUZHOU INTELLIGENT TECH CO LTD

Complex system demand identification and management method based on knowledge graph and thinking chain reasoning

The invention relates to a knowledge graph and thinking chain reasoning-based complex system demand identification and management method. The method comprises the following steps of S1, constructing a system demand classification network model; s2, constructing a demand structured extraction model based on a thinking chain; and S3, constructing a general demand knowledge graph and a target graph, and complementing the target graph. Through fusion of the knowledge graph and the thinking chain reasoning technology, the problems that in the prior art, classification model precision is low, extraction model leak detection is high, and graph model universality is poor are solved, and the method is suitable for demand modeling in the fields of subways and high-end equipment.
Owner:TIANJIN UNIV

Meteorological data source credibility certification method based on block chain

The invention discloses a meteorological data source credibility proving method based on a block chain, and provides a new data source credibility proving mechanism by utilizing the characteristics of tamper resistance, forgery resistance and loss resistance of the block chain and combining signature verification and hash operation algorithms in cryptography. According to the mechanism, credibility and data delivery validity verification can be accurately carried out on delivery data recorded by a data provider; according to the patent, transaction / shared data is converted into a pre-agreed intermediate format (a character set, a coding mode and an XML / JSON data structure expression mode) and is stored in a file form. And after receiving the data, the data user converts the data into a self system format according to self system requirements, so that the correctness of data information expression is ensured. Through the overall transmission of the data packet, the accidental missing and tampering of the data in the transmission process between the two handover parties are effectively avoided.
Owner:STATE QIXIANG INFORMATION CENT +1

Stability regulation and control method and system for energy storage of power distribution network of power system

The invention discloses a stability regulation and control method and system for energy storage of a power distribution network of a power system. The system comprises an energy storage module, a load regulation and control module, an active power distribution module and a data interaction module, the energy storage module collects and analyzes parameter data based on power system requirements and resources, and reasonably stores the parameter data in a database; the load regulation and control module monitors the energy storage condition in real time and predicts and preferentially controls the load; the active power distribution module monitors power flow distribution and optimizes a power distribution scheme; the data interaction module is responsible for issuing a power distribution scheme to related equipment and adjusting and optimizing the scheme according to feedback; an energy storage objective function is established by comprehensively considering various factors, a dynamic optimization algorithm is used for load prediction, real-time monitoring and optimization of a power distribution scheme, a regulation and control scheme is continuously adjusted and optimized through a data interaction and closed-loop feedback mechanism, and meanwhile, an abnormal trigger limit value is set to guarantee stable operation of the system. Therefore, the expandability and maintainability of the system are improved.
Owner:国网宁夏电力有限公司固原供电公司

Network traffic collection method and system based on cloud-based environment

PCT designated stage expiredWO2025107780A1Securing communicationInternet trafficSystem requirements
Disclosed in the present invention are a network traffic collection method and system based on a cloud-based environment. The present invention solves the problems of existing traffic collection being invasive to a monitored system, and having system and software dependency requirements, poor adaptability, and high consumption of computing and network transmission resources. A mirror virtual switch and a virtualized instance agent node are created at a computing node. By mirroring traffic by means of a patch port of a virtual switch, monitoring node data is diverted to the mirror virtual switch, and then is forwarded to the virtualized instance agent node by means of the mirror virtual switch; network traffic data is stored as in a file format, and is forwarded to a data security monitoring platform for monitoring a virtualized instance event in real time, and when the event changes, a flow table of the virtual switch is adjusted on the basis of a monitoring policy. According to the present invention, traffic monitoring is implemented in a software manner, and virtualized instance traffic is collected before packet modification, facilitating the accuracy of security event judgment, involving no intrusive modification to an instance, and no system requirement, and occupying few resources.
Owner:HUAXIN CONSULTATING CO LTD

PLM system code generation method based on large language model

The invention discloses a large language model-based PLM system code generation method, which is characterized by comprising the following steps of: analyzing PLM system requirements, including requirement entry embedding, requirement entry text splitting and requirement entry vector database construction; constructing a product demand model unit based on the PLM system demand; constructing a code generation platform based on the large language model; and describing the product through the product demand model unit, and inputting the product description into the code generation platform. On the basis of entries and modularization of product demands, a product demand model unit is constructed, functions, structures and behaviors of products are described by using a demand model, then a code generation platform is constructed based on a large language model, and finally codes are automatically generated through the large language model, so that modularization, multiplexing and integration of the products are driven by the demand model. Test case generation and code testing and perfection are achieved in an intelligent cooperation mode, main work of research and development personnel is converted into demand analysis and decomposition, demand model construction and automatic code generation auditing from coding implementation, and finally code quality and development efficiency are improved.
Owner:CASCO SIGNAL LTD

Photoelectric system digital prototype function model generation method and device based on artificial intelligence

The invention discloses a photoelectric system digital prototype function model generation method and device based on artificial intelligence, and belongs to the technical field of artificial intelligence and photoelectric system digital prototypes, and the method comprises the steps: inputting training data of an artificial intelligence model, carrying out semantic analysis, marking and key parameter extraction operation, and forming a knowledge graph; establishing a logic association and constraint relationship among the parameters; training an artificial intelligence model by using the analyzed training data, optimizing a preset general rule base of the artificial intelligence model, and constructing a photoelectric system modeling rule base and a photoelectric system artificial intelligence model; calling the trained artificial intelligence model of the photoelectric system, extracting a model demand, calling the modeling rule base of the photoelectric system, and generating a function model of the photoelectric system; and visually displaying the photoelectric system function model, modifying and updating the photoelectric system function model, and dynamically updating the optimization strategy. Different function model modeling languages can be selected according to actual system requirements, and an artificial intelligence algorithm is adjusted.
Owner:INST OF OPTICS & ELECTRONICS CHINESE ACAD OF SCI

Artificial intelligence-based engineering requirements analysis

Methods and systems are described that use artificial intelligence (AI)-based natural language processing (NLP) techniques to rapidly analyze, merge, and / or identify relationships between systems engineering artifacts (e.g., system requirements, system architecture descriptions, research requirements, risk assessments, etc.) to accelerate systems engineering program management, technical capability development, and acquisition initiatives associated with systems engineering projects.
Owner:THE MITRE CORPORATION

Internet of Things testing method and system based on AI, terminal and storage medium

The invention discloses an Internet of Things test method and system based on AI, a terminal and a storage medium, and the method comprises the steps: collecting test data of Internet of Things equipment and an Internet of Things system, and storing the test data in a big data platform; the method comprises the following steps: acquiring an application scene and system requirements of an Internet of Things system, and generating a comprehensive test plan and a test case by using an artificial intelligence algorithm; automatically executing the test case by using an automatic test tool, and sending a test instruction to the Internet of Things equipment; performing deep analysis and mining on test data through big data analysis and an intelligent algorithm to obtain potential problems and risks; according to an analysis result, performing performance evaluation, stability evaluation and robustness evaluation on the Internet of Things equipment and the Internet of Things system; and according to an evaluation result, speculating an optimization improvement direction of the Internet of Things system through machine learning and deep learning algorithms. According to the invention, high coverage, high efficiency and high accuracy of the Internet of Things test are realized, and the test cost is reduced.
Owner:深圳开鸿数字产业发展有限公司

Model-based civil aircraft integrated modular avionics residence system design method

The invention discloses a model-based civil aircraft integrated modular avionics resident system design method. The method comprises the steps of S1, resident system demand modeling and demand confirmation based on a customized MBSE method; s2, performing resident application software function logic modeling based on a Simulink model; s3, performing qualitative security analysis on the resident system based on the function scene and the fault combination; the input of the step S1 comes from a technical protocol of a resident system issued by a host, and through modeling iteration, the demand of the resident system is captured and issued to the step S2; s2, feeding back a system design problem identified in the modeling process; the step S1 synchronously sends the demand of the resident system to the step S3, and at the same time, the step S3 feeds back the identified security demand and constraint to the step S1 to support system demand iteration. According to the invention, an efficient, replicable and landing design thought is provided for the integrated modular avionics residence system, the system design efficiency is improved, and the verification cost is reduced.
Owner:SHANGHAI CIVIL AVIONICS SYSTEMS CO LTD

Dynamic resource aggregation scheduling method and system for virtual power plant and medium

The invention relates to a dynamic resource aggregation scheduling method and system for a virtual power plant and a medium, and the method comprises the following steps: S1, building a virtual power plant (VPP) system model; s2, power system requirements and technical constraints are determined; s3, performing preliminary scheduling on various energy sources in the virtual power plant by using a random mixed integer linear programming model; s4, quantifying the system contribution of the distributed resources by using a VCG (Vickrey-Clarke-Grove) mechanism in a cooperative game theory; and S5, according to the quantitative result and the real-time data, dynamically adjusting scheduling schemes of various energy sources in the virtual power plant with economic benefit maximization. According to the method, the aggregation efficiency of virtual power plant resources can be effectively improved, and the overall energy utilization efficiency and the operation stability of a power grid are improved.
Owner:STATE GRID JIANGSU ELECTRIC POWER CO LTD +1

Optimizing a target automated ecosystem using digital twin simulations

Described are techniques for optimizing a target automated ecosystem. Requirements of the target automated ecosystem are received. A digital twin simulation of one or more reference automated ecosystems implementing the requirements of the target automated ecosystem are created and executed. Furthermore, a digital twin simulation of the target automated ecosystem is created and executed to implement the requirements of the target automated ecosystem. A comparative analysis is then performed between the digital twin simulations of one or more reference automated ecosystems and the target automated ecosystem. Based on the comparative analysis, the infrastructure implemented and / or processes executed by the reference automated ecosystem(s) with an improvement over the infrastructure implemented and / or processes executed by the target automated ecosystem that exceeds a threshold value, which may be user-selected, are identified. Such identified infrastructure and / or processes are then provided to an expert to be adopted and implemented in the target automated ecosystem.
Owner:INTERNATIONAL BUSINESS MACHINE CORPORATION

Plug-in function and online test method for pre-training model of power system

The invention discloses a plug-in function and online test method for a pre-training model of a power system, and designs a pre-training model oriented to the power industry by combining NLP, BERT and reinforcement learning technologies, and realizes a data-driven intelligent control strategy. Through semi-supervised learning, labeled data and unlabeled data are integrated, the generalization ability of the model is improved, and a high-quality initialization model is provided for reinforcement learning agents. The pre-training model is disassembled into independent plug-in modules, and the plug-in modules are integrated with a simulation platform through an API, so that on-demand loading and scene switching of the modules are realized, and the flexibility of the system is ensured. A dynamic event and a system fault are simulated on an HIL platform, and a reinforcement learning agent is pre-trained in combination with an MPC controller, so that the strategy optimization process can continuously adapt to a system state. In addition, model parameters and control strategies are continuously updated through a particle swarm optimization algorithm, and it is ensured that the RL agent and the plug-in module are always kept synchronous with system requirements.
Owner:CHINA SOUTHERN POWER GRID COMPANY