Condition monitoring and fault diagnosis of aerospace equipment rely heavily on prior knowledge, but multi-source heterogeneous documents lead to knowledge fragmentation, and difficult reuse of historical cases. Therefore, this paper proposes a fault management method for aerospace equipment that integrates large language models and knowledge graphs. It integrates the interface data sheets of aerospace equipment and historical fault response plan configuration files to construct a fault rule knowledge graph. On this basis, a workflow for fault configuration parsing and automatic generation of fault report forms is designed to achieve precise parsing and completion of unstructured fault information and automatic generation of fault reports through multi-source information fusion. Finally, a five-layer architecture domain intelligent Q&A system is built to support ground operation and control personnel in obtaining fault knowledge. Task verification shows the fault configuration parsing accuracy of the proposed scheme is 78.2%, the speed of automatic generation of report forms is approximately 41.7 seconds per form, and the accuracy of knowledge questions and answers reaches 92.0%, effectively improving the efficiency level of fault management for aerospace equipment.