Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Final Report for AEOLUS: Advances in Experimental Design, Optimal Control, and Learning for Uncertain Complex Systems

Technical Report ·
DOI:https://doi.org/10.2172/3012903· OSTI ID:3012903
 [1]
  1. Univ. of Texas, Austin, TX (United States); Auburn Univ., Auburn, AL (United States)
The AEOLUS Center is dedicated to developing a unified optimization-under-uncertainty framework for: (1) learning predictive models from data; and (2) optimizing experiments, processes, and designs governed by these models, all driven by complex, uncertain energy systems. AEOLUS addresses the critical need for principled, rigorous, scalable, and structure-exploiting capabilities for exploring parameter and decision spaces of complex forward simulation models. This report summarizes the key highlights of our research during the period of performance. We have made significant progress on several thrusts, including: • a non-intrusive inference reduced order model for fluids using deep multistep neural networks [7]; • closure learning for nonlinear model reduction using deep residual neural networks [6]; • a data-driven learning framework for the analytic continuation of imaginary time using Adams Bashforth residual neural networks [5]; • an asymptotically compatible meshfree method for solving nonlocal equations with random coefficients [1]; • Gaussian smoothing gradient descent methods for minimizing high-dimensional functions [2]; • improved performance of stochastic gradients with Gaussian smoothing for neural network training [3]; and • anisotropic Gaussian smoothing for gradient-based optimization [4]. The highlights from these works are described in §2–§5
Research Organization:
University Of Texas At Austin
Sponsoring Organization:
USDOE Office of Science (SC)
DOE Contract Number:
SC0021659
OSTI ID:
3012903
Report Number(s):
UT Award: 202100051001AWD
Country of Publication:
United States
Language:
English

Similar Records

AEOLUS: Advances in Experimental Design, Optimal Control, and Learning for Uncertain Complex Systems
Technical Report · Wed Dec 18 23:00:00 EST 2024 · OSTI ID:2482273

Analytic continuation of noisy data using Adams Bashforth residual neural network
Journal Article · Wed Mar 31 20:00:00 EDT 2021 · Discrete and Continuous Dynamical Systems - Series S · OSTI ID:1814399

Related Subjects