This is a reduced version of DATAN that generates the figures in 'Generating a physics-based quantitatively-accurate model of an electrically-heated Rijke tube with Bayesian inference' by M. P. Juniper and M. Yoko, in the Journal of Sound and Vibration (2022).

Open Matlab and, at the command line, type:

Assim_JSV('figure_*') where * is the label of the desired figure.

Assim_JSV.m is a Matlab script that assimilates data from the experiments in the folder 210720_configs. It uses the Matlab functions in the folder 220421_Datan. See the comments in each Matlab file for details.

DATAN assimilates microphone data from experiments on an electrically-heated Rijke tube into a thermoacoustic network model. Several different physics-based models are proposed for the elements of the network model. DATAN calculates the most likely parameters for each model, as well as its best fit likelihood, its Occam factor, and its marginal likelihood. This allows the models to be ranked against each other. The model with the most evidence can then be selected.