Intelligent diagnosis is a key task in the construction of "smart power plant". Therefore

this paper proposes a new idea of intelligent diagnosis based on knowledge graph and develops a set of intelligent assistant diagnosis system for thermal power units. This intelligent assistant diagnosis system is based on expert knowledge

combined with artificial intelligence

big data analysis and other technologies to form a set of operation and maintenance assistant systems that integrate intelligence and knowledge. This paper introduces the knowledge graph construction process and the intelligent diagnosis reasoning framework in detail

and develops an intelligent assistant diagnosis system for thermal power units based on the Java EE platform. Finally

a thermal power unit is used as the research object to verify the effectiveness of the intelligent diagnosis inference model based on the knowledge graph and the advantage of intelligent assistant diagnosis system for thermal power units.