Inverse and Transient Heat Transfer Problem on commercial software: is it possible?
 

Itajuba city (MG), Brazil, January, 29th, 2012.

Hello, Folks! :D

I wonder if a commercial software could alone (without the help of other programs that are working together as Matlab software or some other code written in FORTRAN language), working with inverse conduction heat transfer transient problems to estimate the transient heat flux [W m^-2] in solids (mettalic samples), from the transient and experimentally measured temperatures . The geometries would be designed in any CAD tool (like SolidWorks), so we will need the CAD files import module, and the heat transfer module too.

About files (see the link below to download them: Files.rar file) and on our problem, they are described below.

The file is qtotal.dat with experimental transient heat flux and measured in [W m^-2], and this is applied to a metallic sample (in this simulation, there is no fluid, like air or water) of 12.7 mm x 12.7 mm x 4.7 mm dimensions, with heating area of 10.4 mm x 10.4 mm dimensions, applied to the side of the 2D solid (for z = 0.0). The heating was created in the laboratory (by *Carvalho, 2005) using a source of tension and a source of current. The experimentally measured temperatures for thermocouples T4 and T5 are on Tnv09aT4T5.dat file, for a total time of 110.0 [s]. The positions (x, y, z) of the thermocouples T4 and T5 are described below. The thermophysical properties of the mettalic sample are: Density = 14900.0 [kg m^-3]; Specific Heat Capacity = 332.945 [J kg^-1 K^-1] and Thermal Conductivity = 43.1 [W m^-1 K^-1]. In the Excel file (TempCFX_110_experimentalT4T5.xls file = The first column in this file is the time and the second column, the temperatures measured by thermocouple T4. The third column in this file are the temperatures measured by thermocouple T5), were used the following thermophysical properties: Density = 14900.0 [kg m ^-3]; Specific Heat Capacity Cp = 195.85 [J kg^-1 K^-1]; Thermal Conductivity K = 43.1 [W m^-1 K^-1]. Therefore, deviations were obtained with these results of numerical temperatures with the ANSYS CFX software, for the cases studied numerically by Carvalho (2005), and also experimentally by Carvalho (2005). [*Carvalho, S. R., 2005, "Determination of Temperature Field in Cutting Tools for the Machining Process", PhD Thesis, Faculty of Mechanical Engineering, Federal University of Uberlândia, Brazil, 123p. - only in Portuguese - Author: Prof. Carvalho: http://lattes.cnpq.br/8943889945992814 ]

Thermocouples positions named T4 and T5:
MONITOR POINT: Monitor Point T4
Cartesian Coordinates = 4.3 [mm], 3.5 [mm], 4.7 [mm]
Option = Cartesian Coordinates
Output Variables List = Temperature
END
MONITOR POINT: Monitor Point T5
Cartesian Coordinates = 3.5 [mm], 8.9 [mm], 4.7 [mm]

I wonder: how can I estimate the transient heat rates in [W m^-2] (and then compare with the transient heat flux measured => located at qtotal.dat file), using, for example, only one thermocouple (for example, thermocouple T4), using (reading the Tnv09aT4T5.dat file via commercial package) the experimentally measured temperatures and located in the file named Tnv09aT4T5.dat?

More data about our simulation and our geometry already drawn and ready to be imported into commercial software are showed below at link:

http://www.4shared.com/rar/70H3ZMHC/Files.html?


By the way, I send some files from the cases studied using CFX, Fluent and COMSOL softwares, beyond the ANSYS ICEM CFD software. That everyone has a good advantage. Enjoy in moderation.

http://www.4shared.com/dir/W_FHF_W3/...ml#dir=7741665


Rogerio from Brazil. :)

So am I correct in saying what you want to do is to work out the heat flux applied to the sample to get the measured temperature profile?