Research on the Fault Diagnosis of Fire Control System Based on Many Cases
Airborne fire control system (AFCS) is a complex system and hierarchical in structure. It was hard to obtain the precise AFCS fault model, so we built one fault diagnosis system based on a large number of fault cases. The diagnosis system consisted of the knowledge databases, the ES (expert system) unit and the ANN (artificial neural network) unit. The ES rules were acquired, improved and expanded by training the AFCS neural network with many fault cases and carrying out the simulating operation, and the connecting weight values of the AFCS neural network were gradually changed as more and more fault cases were added in. The diagnosis system had a self-study function, and can diagnose new-type faults which the ES rules did not match with.
airborne fire control system fault diagnosis fault cases artificial neural network ezpert system
YANG Maoxing
The First Aeronautical Institute of Air Force, Xinyang, Henan, China, 464000
国际会议
第八届国际测试技术研讨会(8th International Symposium on Test and Measurement)
重庆
英文
1631-1634
2009-08-01（万方平台首次上网日期，不代表论文的发表时间）