Electrostatic induced charge signal extraction based on waveform characteristic in time domain

2019 
Abstract The measurement signal from the inner flush-mounted electrostatic sensor includes both induced charge signal and transferred charge signal. The extraction of induced charge signal from the mixed signals is of great significance to the solid phase velocity measurement. In this paper, a method of extracting induced charge signal based on the waveform characteristic in time domain is proposed. In the calibration experiment of the belt-style electrostatic velocity measurement rig, the mean absolute percentage error and the mean relative standard deviation (1.30%/1.01%) are lower than the direct cross-correlation method (4.81%/5.71%) and the Empirical Mode Decomposition (EMD-based) method (2.55%/2.38%), moreover, the average signal-to-noise ratio (SNR) of the proposed method is the greatest. In the experiment of gas-solid two-phase flow rig, the average correlation coefficient and the mean relative standard deviation are 0.77 and 2.46%, which are better than the direct cross-correlation (0.48/2.87%) and the Discrete Wavelet Transform (DWT) (0.74/2.69%).
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    18
    References
    4
    Citations
    NaN
    KQI
    []