A Robust Fault Diagnosis Scheme for Converter in Wind Turbine Systems

Fault diagnosis is a powerful tool to reduce downtime and improve maintenance efficiency; thus, the low management cost of wind turbine systems effective utilization energy can be obtained. However, accuracy fault extremely susceptible nonlinearity noise in measured signals varying operating conditions. This paper proposes robust scheme based on ensemble empirical mode decomposition (EEMD), intrinsic function (IMF), permutation entropy (PE) diagnose faults converter systems. Three-phase voltage output by are used as input model each signal decomposed into set IMFs EEMD. Then, PE calculated estimate complexity IMFs. Finally, IMF-PE information taken feature classifier. The EEMD addresses nonlinear processing mitigates effects mixing noise. increases robustness against variations conditions effectiveness reliability method verified simulation. results show that for 22 reaches about 98.30% with standard deviation approximately 2% under different speeds. In addition, average 30 runs noises higher than 76%, precision, recall, specificity, F1-Score all exceed 88% at 10 dB. evaluation indicators lower 1.7%; this proves stable diagnostic performance. comparison methods demonstrates has outstanding performance terms its high accuracy, strong robustness, computational efficiency.