Smart grid integration optimization control technology is the key to achieve coordinated operation of all links. Intelligent agent technology provides a good method to solve this problem. Based on the introduction of the status quo and development trend of intelligent agent research at home and abroad, this paper introduces the overall architecture of Agent-based intelligent grid integrated optimization control technology, based on Agent's dynamic decomposition and coordination technology, and takes microgrid as an example, based on microgrid. Agent's plug-and-play technology was analyzed. 0 Preface Building a smart grid is a fundamental strategy for realizing China's energy transformation and a long-term and complex system engineering. In the future, the smart grid requires coordinated control of multiple operating modes and multi-layers to ensure that the grid is in a safe and economical operating state. The smart grid requires the system to process information quickly. The management information and operational information must be shared and coordinated within the scope of the authorization under the condition of confidentiality level, which can effectively prevent network attacks. Relying on traditional coordinated control methods can not meet the requirements of intelligent grid integrated optimization control. In recent years, intelligent agent technology has gradually developed, providing a new information processing method for the realization of intelligent grid integrated optimization control, grid information management and information security. The key to Intelligent Agent is to acquire the knowledge reasoning ability required by the agent, the ability to manipulate knowledge, the ability to display knowledge, and the ability of an agent to share knowledge with other agents. The agent designer must be able to use the knowledge generated to accurately and completely describe the state of the grid and control behavior. The general attributes of intelligent agents include: (1) autonomy: intelligent agents can operate under the direct intervention of no one and other things, and can coordinate their action behaviors and internal states; (2) social capabilities: intelligent agents can Interacting with other intelligent agents (possibly devices or people) through communication language; (3) Responsiveness: Intelligent agents can predict their environment and respond in time; (4) Initiative: Intelligent agents can not only Respond to its environment and actively demonstrate behavior toward the intended goal. The MulTI-Agent System (MAS) is a collection of intelligent agents that can communicate with each other and can achieve common goals through mutual cooperation. In recent years, domestic and foreign experts and scholars have gradually applied MAS theory research to the power market, relay protection, power system voltage reactive power optimization adjustment and emergency control. Agent-based smart grid integrated optimization control technology is becoming one of the hotspots of current and future research. At present, multi-agent technology (MulTI-Agent) is still developing and improving. There are still some problems to be studied: (1) The design of multi-agent system has not been standardized yet. (2) Whether the multi-agent system can truly reveal the dynamic behavior characteristics of the entire power system requires a test of practice. Mainly at present, agents in one aspect can achieve design goals, and systems that are too complicated remain to be studied. A single agent has the ability to solve local problems, but cannot achieve global goals alone. (3) There is no system global control function in the multi-agent system, only relying on a set of coordination agents to resolve the decision conflict between agents. (4) The agent decision process is asynchronous. Must be combined with knowledge engineering. 1Agent-based intelligent grid integration optimization control In the future, the smart grid not only needs coordination between different local controllers, but also needs coordinated control between different controllers in different places. Relying on traditional coordinated control methods can not meet the requirements of intelligent operation control. The use of Agent-based smart grid integrated optimization control technology is one of the ideal means to solve the problem [1]. 1.1 overall architecture The overall idea is to realize intelligent grid integration optimization and coordination control through specially designed system structure, various levels of Agent situational criteria, multi-criteria integrated decision-making methods and processes, and coordinated control mechanisms of various layers of agents. The main technical ideas include: (1) The distributed autonomous architecture of the smart grid is a hierarchical distributed computing structure based on Agent, including sensors, embedded processors and real-time computing, etc., using standardized interfaces to communicate information through the network, mainly adopting SOA architecture and information-oriented middleware implementation. (2) For the private network, distributed database technology is mainly used to realize local and global data exchange and intelligent decision-making. The distributed database is integrated through an open interface. (3) The system will use plug-and-play hardware and software as much as possible, using technical standards based on active learning features [3-12]. 1.2 Dynamic decomposition and coordination In the field of computer applications, most of the existing solutions for analyzing and optimizing collaborative tasks are based on intelligent agent technology. Here is a new idea: to improve the dynamic adaptability of the smart grid through automatic dynamic reconstruction of the system, not only to solve the problem of integrated optimization control of the smart grid in the transmission grid, but also to solve the micro grid and distribution management. Coordinated control issues with the system (DMS). This idea allows the two intelligent agents that make up the solution system to automatically take two operations according to the environment of the grid change: Agent fusion and Agent decomposition. A new task solving structure is generated to adapt to the changes of the environment, and a new solution structure adapting to the current environment can be reconstructed, which is based on the intelligent agent itself. Figure 2 shows the construction of an automatic dynamic reconstruction control model. The process includes four stages: input representation, decision making, decision execution, and dynamic context awareness. The tasks required to complete each phase are: Input representation: Observing the external environment and observing the intelligent agent itself, and obtaining relevant data about the external environment and its own situation. For the external environment of each intelligent agent, it refers to information exchange with other intelligent agents. The structure of a time logic tree can be used to record the process of an intelligent agent communicating with other intelligent agents. This structure is very suitable for query statistics. The agent performs statistics on the query operation of the time logic tree, and can select the best intelligent agent that performs the fusion operation with it. This phase is the preparation phase for the intelligent agent to perform the fusion and decomposition operations. Decision making: Compare and judge the general description of the input ideal structure with the current organizational structure, and make a decision whether to maintain the current structure to adjust the current organizational structure. Decision execution: The specific decisions are made by the decision making stage. If the decision making stage makes a decision to change the current organizational structure, then the decision is taken. Dynamic environment awareness: After the decision-making execution phase performs the decision, the newly formed structure is transformed into the current organizational structure through dynamic environment awareness. Choke Inductor,Pfilter Choke Inductor,Filter Choke Inductor,Choke Power Inductors Guang Er Zhong(Zhaoqing)Electronics Co., Ltd , https://www.cnadaptor.com
