A data mining algorithm for automated characterisation of fluctuations in multichannel timeseries

TL;DR

This paper introduces an automated data mining algorithm that analyzes multichannel timeseries data to identify and classify fluctuation structures, demonstrated on magnetic sensor data from the H-1 heliac.

Contribution

The method combines SVD, spectral comparison, and clustering to automatically characterize fluctuations in large multichannel datasets, scalable and adaptable to various applications.

Findings

Successfully applied to magnetic sensor data from H-1 heliac

Effectively distinguishes different fluctuation classes

Provides visual cluster mapping of fluctuation structures

Abstract

We present a data mining technique for the analysis of multichannel oscillatory timeseries data and show an application using poloidal arrays of magnetic sensors installed in the H-1 heliac. The procedure is highly automated, and scales well to large datasets. The timeseries data is split into short time segments to provide time resolution, and each segment is represented by a singular value decomposition (SVD). By comparing power spectra of the temporal singular vectors, singular values are grouped into subsets which define fluctuation structures. Thresholds for the normalised energy of the fluctuation structure and the normalised entropy of the SVD are used to filter the dataset. We assume that distinct classes of fluctuations are localised in the space of phase differences between each pair of nearest neighbour channels. An expectation maximisation clustering algorithm is used to locate the distinct classes of fluctuations, and a cluster tree mapping is used to visualise the results.