Young Scientist Research Grant

Description

This research project developed an improved methodology for the efficient fault diagnosis of rotating element bearings, specifically addressing the challenges of signal detection in high-noise environments. To enhance the signal-to-noise ratio, a dynamic bandpass filter was engineered using the Harmony Search Algorithm (HSA) to optimize filtering parameters. Through the evaluation of various fitness functions, it was determined that short-time Fourier transform (STFT) based spectral kurtosis and the kurtosis of the envelope spectrum provided the most accurate optimization results. Comparative analysis demonstrated that this proposed dynamic bandpass filter significantly outperforms traditional kurtogram-based methods, offering a more robust and efficient solution for identifying mechanical defects in rotating machinery.