Research Of Civil Aviation Aircraft Fault Diagnosis Technique

Because of the rigorous requirement on safety and reliability of civil aviation, fault diagnosis of civil aviation aircraft is very important. Along with the development of the civil aviation, the number of the aircraft is increasing, some of them are aging, and the maintenance task is heavier.At present most of the fault diagnosis mainly depends on Fault Isolation Manual and maintainer's experience. It often fails in time and efficiency especially for intractable trouble. Based on this it has come forth a research trend to use artificial intelligence to improve the status quo. The Expert System, Neural network and Fault Tree are used mostly for the moment. Most of the intelligent methods still rest on academic stage.This paper takes A320 aircraft as research object and presents intelligent techniques to achieve fault diagnosis. Rough Set is a method of describing vagueness and uncertainty. It has powerful data processing ability. When it is applied in aircraft fault diagnosis to deal with the large number of data from the reliability report, it can gain the common fault aircraft components and simple fault diagnosis rules. It has practical sense. Neural network is often used to orientate the fault position accurately. When it is trained according to the data processed by Rough set, it can simplify the network structure and improve the speed. This paper also presents a method to design the number of nerve cell of the network hidden layer.