Mean levels in nonlinear analysis and identification

Abstract

The effects of data pretreatment and mean levels upon nonlinear model structures is investigated. Techniques commmonly used to prefilter data in linear system identification are shown to alter the model structure in the nonlinear case. The effects of mean levels are considered in detail and a new unravelling algorithm is derived to recover the underlying system model when the offsets are external to the system. A new mapping from the time domain to the frequency domain is also introduced for the case where offsets can be considered as an implicit part of the system.