Aircraft flight control system tends to use Electromechanical actuators (EMAs), so the condition monitoring of aircraft actuators become more important under varying environmental and operational conditionIn. This paper introduces a new approach to monitoring surface roughness for aircraft actuators. The proposed technique starts with current and vibration signal as fault indicator. Them to extract the useful features discrete wavelet transform (DWT) was implemented. To overcome lamentations related with redundancy features that effect on diagnostic accuracy and time taken for Classification orthogonal Fuzzy neighborhood preserving analysis with QR decomposition (FNPA-QR) which is used first time for electrical motor condition motoring. In real time under variable and constant working. Time Delay Neural Network (TDNN) was presented to classify faults and predict their severity. The results achieved from simulation prove the ability of the present method in successfully diagnosis the various type of faults