Example: beam

This example shows how a surrogate for a beam deflection model can be created and illustrates how a model with an executable and an input file can be linked with the bayesvalidrox package.

The surrogate is trained without active learning and no inference is performed, though reference data is available.

Model and Data

Pylink model

Property	Setting
Model type	Runs model via given shell command and parser
Number of input parameters	4
Number of output parameters	1: deflection [m]
Time- or space- dependency	??
MC reference	Yes

Priors

Parameter	Distribution
Beam width	lognormal
Beam height	lognormal
Youngs modulus	lognormal
Uniform load	lognormal

Discrepancy

Property	Setting
Distribution type	Gaussian
Characteristic value	variance: data^2

Surrogate

MetaModel settings

Property	Setting
surrogate-type	PCE
associated model	‘Beam9points’
degree choices	max degree 6, q-norm truncation 0.75
regression	FastARD

Training choices

Property	Setting
Static sampling method	latin-hypercube
Number of static samples	100
Number of total samples	100