Folder for the updated deterministic calibration results after the bug discovered in MFiX-PIC and fixed by Dr. J. F. Dietiker in Q3 2021. The bug fix made the calculation of ROP_g (product of gas density times gas volume fraction) consistent with EP_g (gas volume fraction). The bug fix enabled the updates to the two variables being synchronized but inevitably affected the results from our earlier simulation campaigns used to construct the surrogate models generated for the UQ analyses of Cases 1 and 2.  Hence, rerun of the simulation campaigns were carried out and to perform the UQ analysis with the revised MFiX-PIC

Exp_n21.dat is the experimental dataset (in this case the 21 sample based analytical solution) used to guide the calibration

evalRSM_QUAD.c is the C based surrogate model exported by PSUADE initially and then revised to perform residual calculation

psuade.in is the script to run the optimization to find the set of five parameters that minimize the residuals.

PSUADE generates Matlab format file as output for graphical plot of the Total Sensitivity Indices

Note that the files stored in the repo are not a complete list of all files generated.

This script is provided as input to PSUADE to first construct a quadratic polynomial based response surface and then perform 120 fold cross-validation error assessment. Finally generate several response surface plots.

The native PSUADE file created after importing OLH_n120_i6_o1_y2.dat with ‘read_std’ command in PSUADE and save with ‘write’ command. This file is used for the remainder of all analysis performed with PSUADE

Also the analytical solution for each corresponding simulation sample has been included for direct comparison and assessment of discrepancy.

ASCII file contains:
120 samples created with Optimal Latin Hypercube sampling for 6 input parameters. Only 1 quantity of interest (y2:Location of Settling Shock) is included. The remaining two were disregarded for the deterministic calibration study purposes.
This file is read by PSUADE with ‘read_std’ command to import the simulation campaign results for 1 quantity of interest