This repository provides .jou scripts in Ansys Fluent 17.0 for obtaining aerodynamic performance (lift force, transition position,etc.) under fixed lift.

This journal is an example for importing a plot3d mesh, cutting the boudanry conditions, and set the types for them. The cutting involves the distribution of cores/processes designated by jog users, so I apply only 1 core for this jou, and applying any number of cores for the <cal_anyproc_sample.jou> for initialization and calculation.

Initialization and calculation. The convergence criterion of lift and drag coefficient is a realative ratio. Therefore, the absolute deviation is varified due to different angles of attack. The number now is a product of compromise. The discussion is as follows:

Aoa * Cl * Cd 1 0.433 0.00507 3 0.64 0.0067 7 0.970 0.0301 9 1.00 0.045

For a fixed relative ratio, the absolute deviation can be from 0.0001/0.433=0.0002 to 0.0009/1.00 for lift case, or from 0.0001/0.00507=0.02 to 0.0009/0.045=0.02 for drag case, given that we want lift force effective digit twinkles at the fourth place (cf. fixed lift force of 0.824) and drag force effective digit at the place of 0~9count (i.e. 0.000x).

This is the bash file calling the jou files. They do the CFD work and some raw post-processing.

This is the file that has the information of the cores to be called. It is thought to be put at my desktop, which /home/user03/ The content of it is simply a line: ibcn03:20

This file dominates the process of find the corresponding angle of attack to the fixed lift. The dichotomy is modified with weights since before separation, the Cl-alpha curve should be monotonically increasing with angle of attack.

this file tests the bash scripts for dichotomy.

This file illustrates the necessity of dicotomy.

the mesh file of rae2822 benchmark case

For lift fixed at 0.824, it has three effective digits. Its toleranceis 5e-4 (w.r.t. absolute perspective). Since the practical error is in squared form, the dist in the code should be 2.5e-07 less than before stopping iteration.