Denoising Smart Grid data using Fuzzy Transform based Thresholding on EMD

A soft thresholding mechanism based on Fuzzy membership function (MF) is proposed in this work to remove the noise from smart grid signals. First, EMD is applied to break down the signal into various components. Then level dependent soft thresholding parameters, which are determined from the intrinsic mode functions (IMFs), are used into the fuzzy modifiedS-MF for the noise reduction. A parameter is used to determine how many IMFs are to be thresholded and then the rest are kept unthresholded. In order to evaluate the denoising performance, a comparative analysis is presented with wavelet-transform-based state of art technique.