Verification and Validation in Scientific Computing

Synopsis:

The performance, reliability, and safety of engineering systems are becoming increasingly reliant on modeling and simulation. This course deals with techniques and practical procedures for assessing the credibility and accuracy of simulations in science and engineering. It presents modern terminology and effective procedures for verification of numerical simulations and validation of mathematical models that are described by partial differential equations. While the focus is on scientific computing, experimentalists will benefit from the discussion of techniques for designing and conducting validation experiments. A framework is provided for estimating various sources of errors and uncertainties identified both in simulations and in experiments, and then combining these in total prediction uncertainty. Application examples techniques and procedures are primarily taken from fluid dynamics, solid mechanics, and heat transfer. This short course follows closely the instructors’ new book Verification and Validation in Scientific Computing published by Cambridge University Press in 2010.

Key Topics:

  • Learn terminology and basic principles in verification, validation and uncertainty estimation
  • Procedures for verification of codes
  • Procedures for verification of calculations
  • Design and execution of validation experiments
  • Quantitative assessment of model accuracy based on experimental measurements
  • Estimation of predictive uncertainty in simulations

Who Should Attend:

Computational analysts, code developers, experimentalists, and software quality engineers should attend the course. Also, managers directing such work and project engineers relying on computational simulation for risk-informed decision making should attend. An undergraduate degree in engineering or physics is recommended, and experience in computational simulation or experimental testing is helpful.

Course Information:

Type of Course: Instructor-Led Short Course
Course Level: Fundamentals/Intermediate
Course Length: 2 days
AIAA CEU's available: Yes

Outline
  1. Terminology and basic principles
    • Verification and validation in science and engineering, versus software engineering
    • Error and uncertainty in modeling, simulation, and experimental measurements
    • Responsibilities for verification, validation, and uncertainty quantification
  2. Verification of codes
    • Software quality assurance
    • Order of accuracy verification
    • Method of exact solutions
    • Method of manufactured solutions
  3. Verification of solutions
    • Iterative convergence error
    • Richardson extrapolation
    • Grid convergence index
    • Practical aspects of grid refinement
  4. Validation experiments
    • Validation experiments vs. traditional experiments
    • Validation hierarchy for complex systems
    • Six principles for validation experiments
    • Design and execution of validation experiments
  5. Quantitative estimation of model form uncertainty
    • Characteristics of validation metrics
    • Validation metrics based on mean values
    • Validation metrics based on cumulative distribution functions
  6. Predictive capability in scientific computing
    • Identification and characterization of sources of uncertainty
    • Numerical, parametric, and model form uncertainty
    • Methods for estimating total predictive uncertainty
Materials

Students will have the opportunity to purchase the recommended textbook “Verification and Validation in Scientific Computing” authored by Oberkampf and Roy.


Instructors

William L. Oberkampf

William L. Oberkampf has 41 years of experience in research and development in fluid dynamics, heat transfer, flight dynamics, and solid mechanics. He received his PhD in 1970 from the University of Notre Dame in Aerospace Engineering. He spent most of his career in both staff and management positions at Sandia National Laboratories.

Dr. Christopher J. Roy
Chris Roy, PhD (NCSU, 1998) worked for five years at Sandia National Laboratories in Albuquerque, NM. He currently teaches in the Aerospace and Ocean Engineering Department at Virginia Tech. He has authored/co-authored over 70 journal articles and conference papers in the areas of computational fluid dynamics, numerical error e

 

AIAA Training Links