Publication

Modeling & Simulation, Testing & Validation (MSTV)
2009

ASSESSMENT OF A BAYESIAN MODEL AND TEST VALIDATION METHOD

by Yogita Pai; Michael Kokkolaras; Gregory Hulbert; Panos Papalambros; Michael K. Pozolo; Yan Fu; Ren-Jye Yang; Saeed Barbat

Abstract

Probabilistic Principal Component Analysis (PPCA) is a promising tool for validating tests and computational models by means of comparing the multivariate time histories they generate to available field data. Following PPCA by interval-based Bayesian hypothesis testing enables acceptance or rejection of the tests and models given the available field data. In this work, we investigate the robustness of this methodology and present sensitivity studies of validating hybrid powertrain models of a military vehicle simulated over different proving ground courses.