Conference Paper

Recent results on structural pattern recognition for Fusion massive databases

Asociacion EURATOM/CIEMAT para Fusion, Madrid
DOI: 10.1109/WISP.2007.4447569 Conference: Intelligent Signal Processing, 2007. WISP 2007. IEEE International Symposium on
Source: IEEE Xplore

ABSTRACT Physics studies in fusion devices require statistical analyses of a large number of discharges. Given the complexity of the plasma and the non-linear interactions between the relevant parameters, connecting a physical phenomenon with the signal patterns that it generates can be quite demanding Up to now, data retrieval has been typically accomplished by means of signal name and shot number. The search of the temporal segment to analyze has been carried out in a manual way. Manual searches in databases must be replaced by intelligent techniques to look for data in an automated way. Structural pattern recognition techniques have proven to be very efficient methods to index and retrieve data in JET and TJ-II databases. Waveforms and images can be accessed through several structural pattern recognition applications.

0 Bookmarks
 · 
77 Views
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: JET database contains more than 42 Tbytes of data (waveforms and images) and it doubles its size about every 2 years. ITER database is expected to be orders of magnitude above this quantity. Therefore, data access in such huge databases can no longer be efficiently based on shot number or temporal interval. Taking into account that diagnostics generate reproducible signal patterns (structural shapes) for similar physical behaviour, high level data access systems can be developed. In these systems, the input parameter is a pattern and the outputs are the shot numbers and the temporal locations where similar patterns appear inside the database. These pattern oriented techniques can be used for first data screening of any type of morphological aspect of waveforms and images. The article shows a new technique to look for similar images in huge databases in a fast an efficient way. Also, previous techniques to search for similar waveforms and to retrieve time-series data or images containing any kind of patterns are reviewed.
    Fusion Engineering and Design.
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a novel clustering technique that has been applied to plasma signals to show its utility. It is a general method based on a partitioning scheme that has been proven to be efficient for purposes of analysis and processing of fusion plasma waveforms. Moreover, this paper shows how the information given by the clustering can be used to produce a concise and representative model of each class of signals by applying different modeling approaches. Neuro-fuzzy identification and time-domain techniques have been used. These models allow the application of procedures to detect anomalous behaviors or interesting events within a continuous data flow that could automatically trigger the execution of some experimental procedures. Previously, an in-depth analysis and a preprocessing phase of the waveforms have been carried out. These procedures have been applied to plasma signals of the TJ-II Stellarator fusion device with encouraging results.
    IEEE Transactions on Instrumentation and Measurement 01/2009; 58:2969-2978. · 1.71 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Magnetic confinement fusion reactors are complex devices where a large amount of energy is required to make the fusion reactions happen. In such experimental conditions, the plasma facing components (PFC) are subjected to high heat fluxes. In current Tokamaks like Tore Supra, infrared thermographic diagnostics based on image analysis and feedback control are used to measure and monitor the heating of the PFC during plasma operation. The system consists in detecting high increase of the IR luminance signal beyond fixed temperature thresholds for a set of predefined regions of interest (ROI). Consequently, this system neither takes into account the thermal objects outside of the ROI, nor the geometric information of the detected thermal object. In this paper, we propose a new vision-based approach for the automatic detection of thermal events. This approach is composed of three main tasks: thermal object detection (1), classification (2), and thermal event recognition (3). We present results of our approach for the recognition of one critical thermal event and compare it with the previous system.
    Instrumentation and Measurement Technology Conference, 2009. I2MTC '09. IEEE; 06/2009

Full-text (2 Sources)

View
26 Downloads
Available from
Jun 4, 2014