Marco Iglesias, University of Nottingham, 2024.
Here we provide data and files that were used used for the paper:
"Bayesian inversion for in-situ thermal characterisation of walls in the presence of thermal anomalies"
The code is written to run on High Performance Computing platforms. Hence the .sh files in Code need to be modified accorgingly (e.g. add accounts, resources, etc). We used Tier 2 Midlands Sulis for our computations.
The code is written in python while the visualisation as well as the generation of the prior ensemble were done in MATLAB.
Run the two MATLAB scripts provided in MATLAB_Codes
Run: python main.sh 3D
Several output files will be stored in a folder name Results within Code. The folder needs to be moved to parent folder for visualisation. The main output file for visualistion is Visual_3D.mat.
Ones this is completed you run the MATLAB file GetFlowRates_for_Equiv.m within the folder FlowRates. This will produce the posterior-informed flow rates from the 3D model which we need to build the equivalent 1D model from Section 5.
Run: python main.sh 3D