Comparison of polynomial fitting and wavelet transform to remove drift in electrochemical noise analysis

Abstract

Polynomial fitting and wavelet transform, which could be used to remove trends of electrochemical noise (EN), have been compared in this paper. The typical pitting EN records were chosen to study the effects of these two methods on statistical parameters, power spectral density (PSD), whole energy distribution plot (WEDP) and relative energy distribution plot (EDP), while the indication of selecting the appropriate method was put forward. The results show that for the calculation of the statistical parameters, the polynomial with m<5 and wavelet removal methods all could be selected. Power spectral density plots and WEDPs indicate that the higher order polynomial and wavelet removal methods would reduce the low frequency signals more heavily, so lower order (m=1–3) polynomial is recommended. The appearance of plateau in low frequency part of PSD plots could be regarded as the indication of the best selection of m. Combined EDPs with the analysis of the EN time records, the best m could also be selected, and its value might be just as the same as that obtained from PSD results. In general, the polynomial detrending is a more suitable and flexible trend removal method.