NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Back to Results
Stochastic Reduced Order Models with Python (SROMPy)Stochastic Reduced Order Models with Python (SROMPy) is a software package developed to enable user-friendly utilization of the stochastic reduced order model (SROM) approach for uncertainty quantification. A SROM is a low dimensional, discrete approximation to a random quantity that enables efficient and non-intrusive stochastic computations. With SROMPy, a user can easily generate a SROM to approximate a random variable or vector described by several different types of probability distributions using the Python programming language. Once a SROM is constructed, the software can be used to propagate uncertainty through a user-defined computational model to estimate statistics of a given quantity of interest. This report is meant to introduce the SROMPy module and brie y demonstrate its capabilities. A simple example of a spring-mass system with a random input is included to illustrate the practicality of the SROM approach to uncertainty quantification and relative ease of applying it with SROMPy. The example includes a comparison with a solution obtained using classical Monte Carlo simulation, demonstrating the similarities and advantages of using the SROM approach.
Document ID
20180003203
Acquisition Source
Langley Research Center
Document Type
Technical Memorandum (TM)
Authors
Warner, James E.
(NASA Langley Research Center Hampton, VA, United States)
Date Acquired
May 30, 2018
Publication Date
April 1, 2018
Subject Category
Computer Programming And Software
Mathematical And Computer Sciences (General)
Report/Patent Number
NF1676L-29772
L-12345
NASA/TM-2018-219824
Funding Number(s)
WBS: WBS 533127.02.16.07.06
Distribution Limits
Public
Copyright
Work of the US Gov. Public Use Permitted.
No Preview Available