Overview of System Identification applications to MHD control: How to control MHD activity without using the MHD equations in the feedback model [Note room change]
Speaker: James Drake, Royal Institute of Technology, Stockholm [NOTE room change]
Abstract: To date the approach to MHD control has largely been dependent on incorporation of the well known MHD models to provide the predictive capability required for synthesis of real-time, closed-loop active control algorithms. Through the use of system identification techniques Erik Olofsson has shown that it is possible to model magnetohydrodynamic stability based on unprejudiced empirical data derived from experimental observation (i.e. no MHD equations in the derived magnetohydrodynamic stability model). The ambition has been to approach experimental modal analysis in a generic manner without invoking prior MHD model knowledge. As stated in the thesis, using EXTRAP T2R data it has been possible to iteratively develop and reality-check both dynamical systems estimation/identification and control synthesis techniques.