In this paper, the effect of the wind turbine blade pitch fault and fault detection method based on support vector machine algorithm based on electrical and mechanical signals have been modeled and studied. Dynamic model of permanent magnet synchronous generator and wind turbine under both aerodynamic asymmetry and normal conditions modes have been modelled in Simulink, FAST and TurbSim environment.  The aerodynamic asymmetry has been simulated by adjusting the pitch of one blade different from the other pitch blades. . The simulation results recorded in the time-domain and then transformed into the frequency-domain by using the Fast Fourier Transform (FFT). This results shows when the blade failure occurred, the amplitude of excitation frequency 1p is appeared in electric signals and with more intensity in mechanic signals. For fault detection, first the time-domain and frequency-domain parameters of signals extracted, and then, The sensitivity of this parameters in healthy and faulty conditions obtained by using Distance Evaluation Criteria (DEC). The result of DEC considered as a variable input in SVM. This process implemented for the number of 60 wind Turbulence for healthy and faulty turbines. Simulation results confirmed that the proposed approach to be able to identifying the healthy condition from aerodynamic asymmetry fault in wind turbine and this proposed approach is effective and efficient for blade fault detection.