Dust particles produced by the erosion of first wall materials in fusion reactors may cause serious safety issues, in addition to their impact on the plasma performance. In order to realize reliable model predictions for next-step fusion devices such as ITER, it is therefore crucial to characterize dust particle transport and generation in modern tokamaks in a variety of discharge scenarii. With respect to this issue, we have developed an algorithm called TRACE [1,2], which aims at processing fast video data in order to automatically detect and track dust particles. In 2009 we started a collaboration with the teams of ASDEX Upgrade tokamak (AUG) at IPP Garching under EURATOM mobility support. Our algorithm was tested and first results were promising. Since 2010 the collaboration goes further in the frame of a European contract with priority support (EFDA WP10-PWI-03-03-01-PS). Thanks to a new hybrid thresholding technique developed at IJL, the 2010 version of TRACE is now able to automatically process full length movies, corresponding to all kinds of plasma conditions . To date, TRACE is the only algorithm of its kind in the world enabling to investigate not only dust, but also other critical issues such as hot spots and arcs, and to relate them automatically to plasma events such as ELMs or disruptions.
In 2010, TRACE has been implemented on IPP’s server for a day-to-day routine processing of new AUG camera measurements and the statistical analysis of former measurements campaign is ongoing (more than 30To data to analyze), with a view to link dust production rates to discharge conditions. Joint stereoscopic camera measurements are also being carried out in order to obtain exact dust particle trajectories , to better asses dust penetration into the confined inner plasma.