A data-driven modeling framework of switched dynamical systemsunder time-dependent switching is proposed . The learning technique is Extreme Learning Machine (ELM) The learning process includes segmented tracedata merging and subsystem dynamics modeling . The modeling process is formulated as an iterativeLeast-Squares (LS) optimization problem . Finally, the switching sequence can bereconstructed based on the switching detection and segmented trace merging results . An example of the DC-DC converter is presented to show the effectiveness of the developed approach . The work was published in the journal Computer Science, published by MIT MIT, on October 26, 2013 .

Author(s) : Weiming Xiang

Links : PDF - Abstract

Code :



Keywords : modeling - learning - switching - data - merging -

Leave a Reply

Your email address will not be published. Required fields are marked *