The invention relates to an AR-based operation and maintenance knowledge pushing method, which is different from the previous operation and maintenance mode depending on the experience of operation and maintenance personnel, and expert experience can be shared among the operation and maintenance personnel by virtue of AR equipment, so that the equipment operation and maintenance period is shortened, and the operation and maintenance cost is reduced. An application program, a data resource and a computing service are migrated to an edge node, so that the technical and application requirements in the aspects of quick connection, real-time analysis and quick response are met; according to the operation and maintenance knowledge pushing method based on the context model, field operation and maintenance personnel can be helped to quickly obtain related knowledge through intelligent equipment in the operation and maintenance period, bidirectional information flow between operation and maintenance stations and edge nodes is achieved, the knowledge obtaining cost is reduced, and the working efficiency of the field operation and maintenance personnel is improved.