High-Voltage Equipment Vibration Detection via Harmonic Ratio Analysis

The patent segments the vibration signal analysis into multiple frequency components using Fourier transformation. By dividing the signal into fundamental frequency, harmonics, and sidebands, the system can specifically identify asymmetrical vibrations through characteristic sideband patterns while filtering out grid frequency interference. This segmentation approach enables precise vibration identification without requiring overly complex hardware.

The patent introduces current signals as an intermediary to detect asymmetrical vibrations indirectly. By measuring current variations that correspond to vibration patterns and correlating them with vibration sensor data, the system achieves accurate vibration identification through a mediating measurement approach that avoids direct complex vibration sensing.

The patent performs preliminary signal processing through Fourier transformation and spectral analysis before attempting to identify asymmetrical vibrations. By pre-processing the vibration and current signals to reveal their frequency components, the system prepares the data in a form that makes asymmetrical vibrations distinguishable from regular grid frequency vibrations through characteristic sideband identification.

The patent employs feedback by continuously monitoring both vibration sensor signals and current signals, comparing their spectral characteristics. The system uses the correlation between current variations and vibration patterns as feedback to confirm asymmetrical vibration detection, enabling reliable identification despite the challenging grid frequency environment.