[luzikato]

Hello,

I'm simulating a thermoelectric generator model. To represent every term I need to solve, I'm using UDS.

For example, I want to solve equation of electric field inside a control volume. The equation is shown below,

The is Temperature-dependent Seeback Coefficient Polynomial, T is temperature, is Temperature-dependent Electrical Conductivity Polynomial, and J is current flux.

In order to solve this equation, I use UDS to represent each term at RHS. Each term represented by UDS1 and UDS0, respectively.

and,

-

Those representation of RHS term are based from the reference journal I read. For UDS1, I think it's pretty straight forward, I think I can write the code like this inside a cell loop.

real alpha;
alpha = 8.439e-4 - 3.81e-6*C_T(c,t) + 6.736e-9*pow(C_T(c,t),-9) - 3.934e-12*pow(C_T(c,t),-12);

NV_VS(C_UDSI_G(c,t,1) ,=, C_T_G(c,t),*,alpha);

I'm having difficulties on implementing the UDS0 to codes. The is also a polynomial. So, basically I can take temperature value of a cell like I did on alpha. But the J term is what am trying to figure how to write it in codes.

I mean to say that as we know J is a current flux which can be calculated from this equation below.

And the current within the cell can be calculated with the equation below.

From what I understand, current usually calculated using area of surface flux. In this case, I want to calculate J for cell which contain multiple faces. Once, I figured that maybe in order to calculate J, I could just calculate every J value at each face on the cell then sum it all at the end.



But since I have to use UDS1 value to calculate this, I need to call the UDS1 value from the FACE. On that, I used F_UDSI macro. But in the Fluent Customization Guide stated,

Note that F_UDSI is available for wall and flow boundary faces, only. If a UDS attempts
to access any other face zone, then an error will result.

Based on that statements, I understand that I cannot take UDS1 face value within the interior. I have tried this and indeed it showed errors. Based on this experience, I know that summing up fluxes all over the cell faces are wrong method to calculate the J.

So, please. Does anyone have any suggestion on how to implement J term on my codes?

I have discussed with my professors and colleagues. But none of them familiar with UDF. So maybe someone here can offer their help. Thank you in advance.




Best regards,

Anas