Date of Award
Master of Science
Data-driven approaches are becoming increasingly crucial for modeling and performance
monitoring of complex dynamical systems. Such necessity stems from complex interactions
among sub-systems and high dimensionality that render majority of rst-principle based
methods insucient. This work explores the capability of a recently proposed probabilistic
graphical modeling technique called spatiotemporal pattern network (STPN) in capturing
Granger causality among observations in a dynamical system. In this context, we introduce
the notion of Granger-STPN (G-STPN) inspired by the notion of Granger causality. We
compare the metrics used in the two frameworks for increasing memory in a dynamical
system, and show that the metric for G-STPN can be approximated by transfer entropy.
We apply this new framework for anomaly detection and root cause analysis in a robotic
Saha, Homagni, "Exploring Granger causality in dynamical systems modeling and performance monitoring" (2018). Graduate Theses and Dissertations. 16874.