Application of rough set-neural networks in civil aviation aircraft fault data processing

: With the complexity of aircraft systems, fault diagnosis was getting more and more difficult. The combination of different methods achieved improvement and becomes a tendency of research. Since rough set theory can effectively simplify information, combine rough set theory with neural networks, use the method of the improved attribute reduction algorithm which based on discernibility matrix to simplify the input information. Then improve the convergence of the network and efficiency of the whole data fusion system. The effectiveness of this method was verified by aircraft fault diagnosis test.