A Methodology Based on Evolving Systems for Fault Detection and Identification of Dynamic Systems
Fault detection and identification, AutoCloud, online, stream.
This work proposes a methodology for the detection and identification of failures indynamic systems, through an online and evolving approach. The proposal is divided intothree stages, in which data pre-processing and post-processing are carried out to increasethe robustness of the methodology in the presence of outliers and noise, in addition to theadaptive and unsupervised processing, through the AutoCloud algorithm, which performsgrouping and classification of data streams. To validate this proposal, different evaluationmetrics were used, such as the Adjusted Rand Index (ARI), homogeneity, completeness,precision, f1 score, recall, and satisfactory preliminary results were obtained. Finally, theexecution schedule for completing this work is presented.