Research On Construction And Application Of Grid Dispatching Robot Hybrid Database

With the rapid development of artificial intelligence technology,the intelligence of the power grid has also become a hot issue.Building an intelligent dispatching robot system is critical to the current complex power grid with high security and stability requirements.Due to the complexity of the power grid data and the complex relationship between the data,its storage format and corresponding query method is one of the key factors that affect the operation efficiency of the dispatching robot system.Therefore,the establishment of flexible and accurate database can effectively improve the operation performance of the grid dispatching robot system.This paper analyzes the database system of the power grid dispatching robot system and constructs a hybrid database system that combines relational database and graph database.Based on the characteristics of the graph database which is suitable for processing graph structure database,the topology analysis is carried out based on the graph database and related graph algorithms.The main contents are as follows:(1)Based on the idea of intelligence,the overall design of the power grid dispatching robot system is carried out by integrating the existing power grid assisted decision-making system,computer software and hardware technology,digital communication technology and other modern technologies.Under the task scheduling of the control center,the database system needs to be cooperate with the rule base,the inference engine and various functional modules to handle the outage schedule and the accident.According to the operation process of the robot system,the database function analysis and data demand analysis of the entire life cycle are carried out.After understanding and analyzing different databases,a hybrid database that combines a relational database and graph database is constructed to store the grid structure in the form of a graph,which greatly reduces the complexity of topology analysis based on relational databases or XML files.(2)This paper uses Neo4 j to build the graph database,Cypher,its operation language,integrates a variety of graph algorithms.Combining device status value with graph databases can implement simple topology analysis based on breadth-first or depth-first algorithms,including device connectivity analysis,activity analysis,and load rate analysis.After summarizing the requirements of topology analysis during system operation,a hierarchical grid topology description model is established based on the double breadth first Search and the frame method,and the grid topology is described hierarchically and partitioned.The first layer is the network structure layer,and the second layer is the substation structure layer and the line structure layer,and the third layer is the main wiring structure layer.Among them,the second and third layers are generated in advance,and the first layer of network structure layer generated by combining the state of the power grid and the second and third layers.(3)For the urban power grid mainly composed of 110 k V and 220 k V voltage levels,its main wiring and network wiring have certain patterned characteristics.Therefore,this paper proposes a generation method of power transfer scheme of urban high-voltage power grid based on graph database algorithm and pattern matching.That is,based on the generated hierarchical topology description framework,for different fault types and different wiring methods,the power transfer scheme is generated more quickly.When generating the transfer power supply scheme,the possible transfer power path is obtained according to the extracted pattern features,and the evaluation index is set based on the speed of operation and the safety of the system to select the best transfer power plan.