[Adam_PHD]

Hello Guys;


I am working on heat transfer in the naturally aspirated diesel engine. Currently, I use Ansys Fluent to simulate the total cycle of the engine my goal is to correct the estimate of the heat flux using 0d udf coupling to 3d CFD.

Fluent generally uses the wall law which underestimates the heat flux so the idea of correct the latter via the implementation of a UDF which is used to calculate the flow by phenomenological laws (Whoshni-Hohenberg).


My problem is that I can not define the domain of computing; the combustion chamber, precisely how to define in UDF: the temperature (TC) of the cell adjacent to the walls?

Please orient me to his

[doruk]

Hello,

This is my code down below for single domain. If there are different domains you can try to use the code

domain = Get_Domain(1); //1 is the id of the domain you are working on

You can check the id at the boundary conditions section at user interface in Fluent.


DEFINE_ADJUST(FixWallTemp, domain)
{
Thread *t,*tf,*t0;
face_t f;
cell_t c,c0;

tf = Lookup_Thread(domain,29); //for my case wall id was 29


begin_f_loop (f,tf)
{

c0=F_C0(f,tf); // adjacent cell id
t0 = F_C0_THREAD(f,tf); // adjacent cell thread

C_T(c0,t0)=0.0; // makes the temperature 0 for adjacent c
}
end_f_loop (f,tf)
}

Best Regards,
Doruk

[reza_65]

Hi Formers,

I would like to change my boundary condition from wall to velocity inlet if the maximum temperature of the whole domain reaches 500.
Can you please help me with it?

Best Regards,
Reza

[Adam_PHD]

Hi guys ;
This is my code down below for ic engine, my problem is adjacent cell definition

DEFINE_ADJUST(total_volume,fluid-ch-down)
{
Thread *t;
double total_volume=0.0;
double total_temperature=0.0;
double total_pressure=0.0;

cell_t c;

/* Integrate volume_1. */
/**loop*over*all*cell*threads*in*the*domain ....[3.1]....*/
thread_loop_c(t,cylindre_qurd)
{
begin_c_loop(c,t)
total_volume += C_VOLUME(c,t);
total_temperature += C_T(c,t)*C_VOLUME(c,t);
total_pressure += C_P(c,t)*C_VOLUME(c,t);
}
end_c_loop(c,tc)

volume_udf = total_volume;
temperature_udf = total_temperature/total_volume;
pressure_udf = total_pressure/total_volume;

a1 = 1./(pow(volume_udf,0.06));
aa2=pressure_udf*0.00001;
a2 = pow(aa2,0.8);
a3 = 1./(pow(temperature_udf,0.4));
a4 = pow(Sp,0.8);
htc_udf = 130*a1*a2*a3*a4;
heat_flux=htc_udf*(tc-twall)

}

DEFINE_PROFILE(heat_transfer_coefficient,t,cylindr e_tri)
{
face_t f;

begin_f_loop(f,t)
{
F_PROFILE(f,t,cylindre_tri)=htc_udf;
}
end_f_loop(f,t)
}

DEFINE_PROFILE(temperature_interieure_gaz,t,cylind re_tri)
{
double temperature_int;
face_t f;
if (temperature_udf<=200)
{
temperature_int=360;
}
else
{
temperature_int=temperature_udf;
}
begin_f_loop(f,t)
{
F_PROFILE(f,t,cylindre_tri)=temperature_int;
}
end_f_loop(f,t)
}

DEFINE_EXECUTE_AT_END(affichage_resultats)
{
# define volume_integrale_udf "integrale_volume_udf.txt"
# define temperature_integrale_udf "integrale_temperature_udf.txt"
# define heat_transfer_empirique "HTC_coef_empirique.txt"
# define pressure_integrale_udf "integrale_pressure_udf.txt"

double temperature2;
FILE *fp1;
FILE *fp2;
FILE *fp3;
FILE *fp4;

Domain*d;
Thread *t;
real sum_temper=0.0;
cell_t c;
d = Get_Domain(1);

thread_loop_c(t,d)
{
begin_c_loop(c,t)
{
sum_temper+= C_T(c,t)*C_VOLUME(c,t);
}
end_c_loop(c,t)
}

fp1=fopen(volume_integrale_udf, "a");
fprintf (fp1," %g %g \n", CURRENT_TIME, volume_udf);
fclose (fp1);

fp2=fopen(temperature_integrale_udf, "a");
fprintf (fp2," %g %g \n", CURRENT_TIME,temperature_udf);
fclose (fp2);

fp3=fopen(pressure_integrale_udf, "a");
fprintf (fp3," %g %g \n", CURRENT_TIME, pressure_udf);
fclose (fp3);

fp4=fopen(heat_transfer_empirique, "a");
fprintf (fp4," %g %g %g \n", CURRENT_TIME, htc_udf, qid);
fclose (fp4);
}

[Adam_PHD]

i want to estimate and apply Hohenberge heat flux as boundary condition :
h(θ)=130 V^(-0.06) P^(0.8) T_G^(-0.4) (S ̅_p+1.4)^(0.8)

[AlexanderZ]

in your code you use

DEFINE_ADJUST(total_volume,fluid-ch-down)
{

what is fluid-ch-down here?

usually there should be the name of domain (however, I have no experience in simulations of diesel engines)

So now your call your domain as fluid-ch-down

But later you've stated loop of threads using other name

thread_loop_c(t,cylindre_qurd)
{

Now cylindre_qurd

I think it may be a problem.

On the other hand, you didn't describe what problems do you have.
It is impossible to help you without information.

Best regards

[sat_fire]

hi everyone,
I am writing UDF for Energy source term for a zone.my domain consist of two zones. so while writing UDF i am using thread_c_loop(t,d). This thread will work on every cells of zone where i will attach this UDF or it will work on all cells of domain including both zone.

[Adam_PHD]

thanks Doruk and AlexanderZ;

i want to estimate and apply 0d Hohenberge heat flux as boundary condition :
h(θ)=130 V^(-0.06) P^(0.8) T_G^(-0.4) (S ̅_p+1.4)^(0.8)

V , P , T_G is from Fluent domain from last step

/* Integrate volume_1. */
/**loop*over*all*cell*threads*in*the*domain ....[3.1]....*/
thread_loop_c(t,cylindre_qurd)
{
begin_c_loop(c,t)
total_volume += C_VOLUME(c,t);
total_temperature += C_T(c,t)*C_VOLUME(c,t);
total_pressure += C_P(c,t)*C_VOLUME(c,t);
}
end_c_loop(c,tc)

volume_udf = total_volume;
temperature_udf = total_temperature/total_volume;
pressure_udf = total_pressure/total_volume;

After i will estimate heat flux by
heat_flux=htc_udf*(C_T(c0,t0)-twall)

and apply it to cylinder wall
Help me please guys, I tried much time but without good results

Regardssss

[AlexanderZ]

Originally Posted by sat_fire

hi everyone,
I am writing UDF for Energy source term for a zone.my domain consist of two zones. so while writing UDF i am using thread_c_loop(t,d). This thread will work on every cells of zone where i will attach this UDF or it will work on all cells of domain including both zone.

read your post. There is no question! What did you think about?

Regarding energy source:
energy source will be generated only in zones (every cell of that zone), you had defined with rate from UDF

Best regards