Next-Generation Optimization under Uncertainty: Structure-Oriented Algorithms
With the advent of the smart grid, renewable technologies, and carbon emission programs, a key concern is the coordination of national infrastructures such as the electricity, natural gas, and water supply network systems. Anticipating and mitigating uncertainty of weather, demands, and contingencies in a more integrated environment is necessary to maximize resource efficiency and prevent cascading failures that can ultimately lead to catastrophic shortages of supplies. This project developed scalable algorithms for stochastic optimization capable of addressing extreme-scale problems (motivated by challenges in the design and operation of national infrastructures). Specifically, the project developed a new family of algorithms to solve continuous stochastic problems to local and global optimality as well as algorithms to solve mixed-integer convex and nonconvex stochastic programs. We developed supporting local and global convergence theory for these algorithms and implemented the algorithms in open-source software packages that can leverage high-performance computing architectures.
- Research Organization:
- Univ. of Wisconsin, Madison, WI (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
- DOE Contract Number:
- SC0014114
- OSTI ID:
- 1575056
- Report Number(s):
- DOE-WISCONSIN-14114
- Country of Publication:
- United States
- Language:
- English
Similar Records
Final Report---Optimization Under Nonconvexity and Uncertainty: Algorithms and Software
An adaptive, multivariate partitioning algorithm for global optimization of nonconvex programs