Signal collected by sensor in engine fault diagnosis system is synthesis of actual signal and random noise.This brings many adverse effects to fault detection and diagnosis.It needs to eliminate noise.In view of data characteristics of sensors for engine

a method of data processing based on faults diagnosis system of intelligent sensor is introduced.The method combines median filtering and wavelet analysis.The median filtering is used to eliminate pulse noise and wavelet analysis is used to eliminate random noise.Simulation results show that hybrid filtering can effectively remove pulse noise and random noise

and can restore the original signal.The method is applied to the engine test data

and the results show that the method is effective in removing noise.The method is applied to the test data of the engine

the results show that the method has ideal effect of denoising.