Introduction
Verification and Validation (V&V) aims at quantifying numerical and modelling errors in Computational Field Simulations, i.e. numerical solutions of mathematical models that describe physical processes involved in Engineering applications.
The MARIN cooperation with Instituto Superior Técnico (IST) started in 1986. The development of Computational Fluid Dynamics (CFD) tools for the simulation of viscous flows of high Reynolds numbers has been one of the main topics of the research work. When the PARNASSOS CFD code started to be used routinely to study the flow around ship hulls at model and full scale Reynolds numbers, the quantification of the numerical and modelling accuracy of the CFD tools became one of the priorities of the MARIN-IST cooperation. These pages about Verification and Validation (V&V) contain links to papers and Workshops and tools related to Verification and Validation that have been written, organized and developed since 2000 that can be useful for the Modeling and Simulation community.
Verification and Validation (V&V) aims at quantifying numerical and modelling errors in Computational Field Simulations, i.e. numerical solutions of mathematical models that describe physical processes involved in Engineering applications.
Definitions
Although there are several definitions available in the open literature, simple definitions of V&V are given below.
VerificationVerification can be summarized as “Check if we are solving the equations right”, i.e. a mathematical problem that focus on the quantification numerical errors. However, it actually contains two different types of activities:
- Code Verification that deals with coding mistakes, i.e. check that there are no “bugs” and the check of the consistency of the discretization techniques used in the computational models.
- Solution Verification that aims at estimating the numerical uncertainty of solutions for which the exact solution is unknown.
ValidationThe simplest way to describe Validation is to say that it corresponds to “Check if we are solving the right equations”. It is related to the quantification of modelling errors, i.e. the difference between the outcome of a mathematical/computational model and the physical reality. In its simplest form, it requires a solution of a mathematical model and experimental measurements. Unfortunately, none of these quantities are exact due to numerical, input parameters and experimental errors, which are often unknown. Therefore, the estimation of the uncertainties that characterize these errors is a fundamental part of any Validation exercise.