On the applicability of nonlinear time series methods for partial discharge analysis

Abstract

In this work, the application of nonlinear invariants and phase space methods for Partial Discharge (PD) analysis are discussed and potential pitfalls are identified. Unsupervised statistical inference techniques based on the use of surrogate data sets are proposed and employed for the purpose of testing the applicability of nonlinear algorithms and methods. The Generalized Hurst Exponent and Lempel-Ziv Complexity are used for finding the location of the system under test on the spectrum between determinism and stochasticity. The surrogate generation method is shown to produce phase-resolved relations with high correspondence with the original sets and the nonlinear invariants employed are found to have strong classification abilities at discerning between surrogates and original point series.