Sparse
Identification of Nonlinear Dynamics for Digital Twins
Digital
twin is a computer-based mathematical
representation that simulates the behavior of a given process.
Digital twins are used to interact with and simulate real-world processes. Our
goal is to develop a large-scale digital twin that approximates the behavior of
several interacting processes. We achieve this by using a Sparse Identification
of Nonlinear Dynamics (SINDy) algorithm. This algorithm applies the idea of sparse
regression to determine the “optimal’’ combination of models that provide the
best performance. The SINDy algorithm contains three main steps: data
collection, identification of model library, and sparse regression. We provide
examples to illustrate the advantages of using SINDy to develop digital twins.
A schematic of our approach is shown in the figure below.
Presenter(s)
Once the content has been viewed and you have attested to it, you will be able to download and print a certificate for PDH credits.
If you have already viewed this content,
please click here
to login.