Empirically derived basis functions for unsupervised classification of radial profile data

Abstract

We present an analysis of empirically derived basis vectors for feature detection in radial profile data. Our aim is to classify broad and peaked profiles using unsupervised techniques. Radial data often contains a continuum of profile shapes from broad to peaked, as such clustering methods may be unreliable. Previously, ad hoc heuristic measures had been used for classification of profiles from raw data (without tomographic reconstruction), which required significant manual inspection of the data. Here, we apply a singular value decomposition (SVD) to a training data matrix consisting of a concatenation of multichannel bolometry time series data from 103 TJ-H plasma discharges with good representation of the range of profiles. The second largest spatial basis vector (topo) has radial roots either side of the plasma centre, and can intuitively be interpreted as a peakedness perturbation. The inverted topo matrix can be used to process new data for automated profile classification. Finally, we show an application of this method using support vector machines to locate other signals related to the radiation profile. Show more