Application of rough set neural network in fault diagnosing of test-launching control system of missiles
Wang Yuegang, Bin Liu, Guo Zhibin, Qin Yong-yuan · 2004
In test-launch control system of missiles, the relations between observed information and fault causes are complicated. Neural network is an effective method to diagnose this type of faults. But, to recede the complex of neural network is a main job in diagnosis. The rough sets theory was introduced in fault diagnosis via neural network to eliminate the unnecessary attributes and disclose the redundancy of condition attributes. Using the decision table, this approach extracted the diagnosis rules from the set of fault samples directly. A case study was used to illustrate the application of the proposed approach. Result shows that the approach is valid.