Aiming at the problems of insufficient intelligence, low diagnosis efficiency and high misdiagnosis rate in the fault diagnosis of small caliber ammunition assembly equipment, the fault diagnosis technology and expert system of small caliber ammunition assembly equipment are studied. In order to improve the fault knowledge reasoning and discrimination accuracy of complex equipment, the knowledge base construction method of fault category based on knowledge map and fault tree is studied, and the fault knowledge reasoning method based on rules and cases is put forward. The deep learning algorithm is used to reason and update the fault knowledge, and the fault diagnosis model is constructed and used in the small caliber ammunition assembly equipment fault diagnosis expert system. The results show that the system can realize the intelligent prediction and analysis of the failure of the small caliber ammunition assembly equipment, meet the requirements of intelligent development of manufacturing equipment, and provide a reference for the development and application of intelligent and information manufacturing equipment.