Affordable Access

Publisher Website

Quantitative validation techniques a database. (I). Simple examples

Authors
Journal
Simulation Modelling Practice and Theory
1569-190X
Publisher
Elsevier
Publication Date
Volume
12
Issue
6
Identifiers
DOI: 10.1016/j.simpat.2004.06.001
Keywords
  • Model Validation
  • Systems
  • Labview
  • Correlation Coefficient
  • Parameter Variation
  • Acsl
  • Integral Squared Error
  • Theil Coefficient

Abstract

Abstract This paper proposes the formulation of a database of validation examples. This initial contribution describes the problem of validation of the simulation of two real systems. These include a mechanical network and a thermal system with a transport delay. Data of the component values and measurements are reported here. The data from testing these systems will be made available at a later stage to enable other workers to compare other validation techniques with those presented here. The responses of a mechanical and a thermal system were measured with a laptop PC and Labview via a DAQCard TM-700 data acquisition card. Simulations were made with ACSL and ACSL Math. The validation procedures implemented included a parameter variation with integral quadratic weighting and ITAE weighting together with a statistical matching technique using a correlation coefficient. These techniques illustrate the problem of achieving “good” validation to within a few percentage points. Correlation coefficients of over 98% were obtained for the fitting of linear models, with Theil indices of less than 0.075.

There are no comments yet on this publication. Be the first to share your thoughts.