The Application Of Intelligent Systems Based On Knowledge Graphs In The Power Industry

The Internet is moving towards the era of semantic interconnection.The era of semantic interconnection is made possible by the knowledge graph with its own powerful semantic comprehension ability,as well as its openness and interconnectedness.With the continuous development of business of the state grid,power grid operation and management of the difficulty in gradually increase,at present the most pressing need is to improve the ability of power network,improve the level of the dispatching information integration,improve the level of standardization management,so the state grid company branches and web province began planning intelligent scheduling system research,this article is to the south of group which intelligent control cloud project as the background,combined with my professional,to build an intelligent system based on knowledge graph of the electric power industry.This thesis mainly combines knowledge graph related domain knowledge and applies it to the power industry.The main research contents include the following three aspects;(1)Combined with Chinese noun classification thesauri,How Net industry specifications,such as power system related information content construct domain ontology of the electric power industry,the expansion of the architecture based on ontology content,applied the key technology such as knowledge extraction from structured,semi-structured,and unstructured data extracted from the ternary group data,finally construct the electric power industry knowledge graph.The constructed knowledge content is stored through the graph database NEO4 J,and the constructed knowledge graph is displayed using the knowledge graph visualization technology;(2)Based on good knowledge graph construction,by using the AIML,KBQA,similarity calculation,such as technical solutions,designed a solution to the power industry business process standardization,basic data query of the electric power industry,electric power knowledge quiz problems such as intelligent system,and build up the correct test scheme of system,and finally gives a system for the power industry q&a accuracy;(3)Integrated HTML,JSON,PYTHON,JS,FLASK and other technologies,an intelligent system was designed to meet the requirements of knowledge graph visualization,intelligent dialogue system,entry management,user login and other functions,combined with server technology,IP mapping,domain name resolution and other technologies,the system is built in the cloud server for public release.In this thesis,the domain knowledge graph of the power industry is constructed by using the knowledge graph construction technology,and it is displayed by the knowledge graph visualization technology.An intelligent question answering system is constructed,and the corresponding test scheme is designed.Finally,the overall accuracy of the system is given.Integration of each subsystem module,through the design,research and development,release,system operation examples and complete technical solutions.This system has solved the key problems such as the standardization of the business process,the basic data query and the power knowledge question and answer of the power industry.