Denoising method for detecting insulator based on LSM filtering and CEEMDAN combined with wavelet

The invention discloses a de-noising method for detecting an insulator based on LSM filtering, CEEMDAN and wavelet. The de-noising method comprises the following steps: acquiring an original signal containing noise of the insulator; the method comprises the following steps: preprocessing an original signal based on least square filtering, and preliminarily filtering noise to obtain a signal after LSM filtering; decomposing the signal filtered by the LSM through a CEEMDAN algorithm to obtain a series of intrinsic mode function I MF components; obtaining the first I MF component, extracting multi-scale characteristics through a wavelet threshold denoising method, and identifying and separating noise signals and effective signals on each scale according to a preset wavelet comprehensive threshold calculation method while performing wavelet threshold denoising; performing signal reconstruction on a third I MF component corresponding to the effective signal and the second I MF component; the limitation of a single