Fault Diagnosis of Wind Turbine Bearings Based on CEEMDAN-GWO-KELM

To solve the problem of fault signals wind turbine bearings being weak, not easy to extract, and difficult identify, this paper proposes a diagnosis method for fan based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) Grey Wolf Algorithm Optimization Kernel Extreme Learning Machine (GWO-KELM). First, eliminating interference noise collected vibration signal should be conducted, in which wavelet threshold denoising approach is used order reduce signal. Next, CEEMDAN decompose after operation obtain multi-group intrinsic mode function (IMF), feature vector selected by combining correlation coefficients eliminate spurious components. Finally, fuzzy entropy chosen IMF component input into GWO-KELM model as defect detection. After diagnosing Case Western Reserve University (CWRU) dataset presented research, it found that can identify 99.42% various bearing states. When compared existing combination approaches, proposed shown more efficient faults.