The invention discloses a WOA-VMD, PSR and AWMF-based multi-stage event maintenance type microseismic
signal noise reduction method, and relates to the technical field of microseismic
signal processing, and the method comprises the steps: obtaining an original
signal; decomposing the original signal to obtain a plurality of intrinsic mode function (IMF) components; dividing the IMF component into a first component, a second component and a third component;
noise reduction is carried out on the first component, the second component and the third component, and a preliminary
noise reduction signal is obtained through reconstruction; performing phase-space reconstruction on the preliminary
noise reduction signal, and performing
noise reduction; filtering the IMF component; analyzing the event window, and
smoothing a non-event window to obtain a final
noise reduction signal; according to the method, the VMD
decomposition process is automatic and can be reproduced, and the signal-to-noise ratio and the reconstruction precision are remarkably improved on the aspects of multiple types of synthetic signals and on-site micro-seismic signals; through combination of phase-space reconstruction and local SVD, residual coherent noise and random disturbance are effectively suppressed, and a
main lobe structure and
chaotic characteristics of a micro-seismic signal are maintained. And fine suppression of high-
frequency noise and
ringing is realized through
wavelet-morphology combined filtering.