The invention discloses a Semantic-based industrial production equipment predictive maintenance system. The overall architecture of the system is as follows: an original information layer comprises anequipment maintenance document, equipment parameter indexes, expert experience related to equipment maintenance and equipment information; a semantic layer comprises an equipment information ontologymodel, an equipment fault knowledge base module, an equipment information semantic annotation module, a semantic reasoning and query module, an ontology database and a semantic rule file; and an application layer comprises a knowledge base management module, an equipment operation state query module, an equipment fault early warning module and an equipment maintenance strategy module. According to the method, rich semantic knowledge in the field can be obtained through semantic annotation, semantic reasoning and semantic query, data-to-information-to-knowledge conversion is realized, fault knowledge and maintenance knowledge are integrated, a reasonable maintenance strategy is obtained before equipment has a fault, equipment maintenance personnel can be assisted to maintain and manage theequipment more conveniently, the fault rate of equipment operation is reduced, and the production efficiency is improved.