A Fault Diagnosis Method for Wind Turbine Generators Based on Improved Ant Lion Optimization Algorithm and Wavelet Kernel Extreme Learning Machine

Abstract

Addressing the fault characteristics of generator equipment in wind turbines, this study introduces a fault diagnosis method utilizing an enhanced ant lion optimization algorithm (CALO) to refine the wavelet kernel extreme learning machine (WKELM). Initially, the SCADA data's feature vector is derived through the guaranteed bureau projection algorithm. Subsequently, the CALO algorithm incorporates a Cauchy mutation operator to enhance the ant lion algorithm's global optimization capabilities. Finally, the CALO algorithm is employed to optimize the parameters of the WKELM, thereby improving both the diagnostic accuracy and convergence speed of the algorithm. To validate the efficacy of the proposed diagnostic model, experimental verification was conducted using operational data from wind turbine equipment in a wind farm located in northwest China. The simulation results demonstrate that the CALO-WKELM diagnostic model can effectively identify various generator faults, fulfilling the requirements for generator fault diagnosis.