[flappy]

Hello,

I'm new to UDFs as well as to C programming and need help for my university work. I'm working on a fixed-bed reactor model with chemical reactions (Methane Steam Reforming) in a porous zone and want to implement the kinetic model of Xu and Froment as an UDF.

I'm using Fluent 17.0 and Microsoft Visual Studio 2015 for compiling on a 64-Bit Windows 7 Enterprise System.

apparently there is a mistake in my code. Fluent tells me: received a fatal signal (segmentation fault) - which i read in other forum threads is a sign that there is a mistake in the code.

here's my code:

/*User-defined function used in the model of steam reforming*/
#include "udf.h"
DEFINE_SR_RATE(msr,f,t,r,mw,yi,rr)
{
Thread *t0=THREAD_T0(t);
cell_t c0=F_C0(f,t);
/*mass fraction of species i at the wall*/
double y_co2=yi[0];
double y_n2=yi[1];
double y_co=yi[2];
double y_h2=yi[3];
double y_ch4=yi[4];
double y_h2o=yi[5];
double Nsum,R,TP,temp,k1,k2,k3,Keq1,Keq2,Keq3,K_h2,K_co,K _h2o,K_ch4,DEN,r1,r2,r3;
/*calculate species i in the unit of kgmol i/kg mix*/
y_co2 *= 1/mw[0];
y_n2 *= 1/mw[1];
y_co *= 1/mw[2];
y_h2 *= 1/mw[3];
y_ch4 *= 1/mw[4];
y_h2o *= 1/mw[5];
/*total mole number per kg mix */
Nsum = y_ch4 + y_h2o + y_h2 + y_co + y_co2 + y_n2;
/*calculate mole fraction of species i in the unit of kgmol
i/kgmol mix*/
y_ch4 *= 1/Nsum;
y_h2o *= 1/Nsum;
y_h2 *= 1/Nsum;
y_co *= 1/Nsum;
y_co2 *= 1/Nsum;
y_n2 *= 1/Nsum;
/*gas constant, J/molK */
R=8.314;
/*equilibrium constant calculated by the empirical formula*/
temp=C_T(f,t);
Keq1=10266.76*exp(-26.83/temp+30.11);
Keq2=exp(4400/temp-4.063);
Keq3=Keq1*Keq2;
/*reaction rate constant*/
k1=4.225e+15*exp(-240.1/(R*temp));
k2=4.39e+4*exp(-67.13/(R*temp));
k3=2.29e16*exp(-243.9/(R*temp));
/*adsorption coefficients*/
K_h2=0.0296*exp(82.9/R*(1/temp-1/648));
K_co=40.91*exp(70.65/R*(1/temp-1/648));
K_h2o=0.4152*exp(88.68/R*(1/temp-1/823));
K_ch4=0.1791*exp(38.28/R*(1/temp-1/823));
/*total pressure in the cell near wall, bar*/
TP=C_P(c0,t0)/1.0e+5;
/*constant DEN factor for rate equations*/
DEN=1+TP*(K_ch4*y_ch4+K_co*y_co+K_h2*y_h2)+K_h2o*y _h2o/y_h2;
r1=(k1/pow(TP*y_h2,2.5)*(pow(TP,2)*y_ch4*y_h2o-pow(TP,4)*pow(y_h2,3)*y_co/Keq1))/pow(DEN,2);
r2=(k2/(TP*y_h2)*(pow(TP,2)*y_co*y_h2o-pow(TP,2)*y_h2*y_co2/Keq2))/pow(DEN,2);
r3=(k3/pow(TP*y_h2,3.5)*(pow(TP,3)*y_ch4*pow(y_h2o,2)-pow(TP,5)*pow(y_h2,4)*y_co2/Keq3))/pow(DEN,2);
if (STREQ(r->name, "reaction-1")){
*rr=r1;}
else if (STREQ(r->name, "reaction-2")){
*rr=r2;}
else if (STREQ(r->name, "reaction-3")){
*rr=r3;
}
}

I'd be enormously happy if anyone is willing to take some time and help me with my case. As i already mentioned, i hardly have any experience with the C programming language and am happy for any kind of advice given.

Kind regards,
f.

[`e`]

First for debugging purposes, simplify your code (cut out sections) until you determine the location of the error. Alternatively, use message functions and print to the screen every few lines.

Segmentation faults are generally caused from trying to access a variable that doesn't exist. Have you got the correct number of species and other parameters? For example, accessing mw[5] when mw only has four entries may cause this error.

[flappy]

Thanks very much 'e', good advice! The problem was the usage of Define_SR and a porous system, which apparently doesn't work together. I have now changed to a Volume Reaction and simplified the code and it's working.

[ajithjec]

Originally Posted by flappy

Thanks very much 'e', good advice! The problem was the usage of Define_SR and a porous system, which apparently doesn't work together. I have now changed to a Volume Reaction and simplified the code and it's working.

Hi, I am also trying the same, do you change the Define_SR and adopt Define_VR for working in a porous medium. ?
I am a beginner in using UDF in Fluent. Can you please suggest to me some tutorials or references for learning? I am simulating a Steam methane reforming reactor for hydrogen generation. Having a catalyst layer in the reformer tube. I am also working with Define_VR udf. Kindly help.

[ajithjec]

Originally Posted by flappy

Thanks very much 'e', good advice! The problem was the usage of Define_SR and a porous system, which apparently doesn't work together. I have now changed to a Volume Reaction and simplified the code and it's working.

Hi, I am also trying the same, do you change the Define_SR and adopt Define_VR for working in a porous medium. ?
I am a beginner in using UDF in Fluent. Can you please suggest to me some tutorials or references for learning? I am simulating a Steam methane reforming reactor for hydrogen generation. Having a catalyst layer in the reformer tube. I am also working with Define_VR udf. Kindly help.

[khadi]

Originally Posted by ajithjec

Hi, I am also trying the same, do you change the Define_SR and adopt Define_VR for working in a porous medium. ?
I am a beginner in using UDF in Fluent. Can you please suggest to me some tutorials or references for learning? I am simulating a Steam methane reforming reactor for hydrogen generation. Having a catalyst layer in the reformer tube. I am also working with Define_VR udf. Kindly help.

hi, ajithjec
I studie numericaly thermochemical solar reactor based on volumetric porous receiver. however I found lot of difficulties during the modeling of reactor via Ansys-Fluent software, and I'm unsure about simulation detail.
I'll be very thankful if you can willing to give me hand.

best regard

[ajithjec]

Originally Posted by flappy

Thanks very much 'e', good advice! The problem was the usage of Define_SR and a porous system, which apparently doesn't work together. I have now changed to a Volume Reaction and simplified the code and it's working.

I have 3 reaction rate equations and it is converted to UDF file. I am trying DEFINE_VR_RATE. The catalyst inside was modeled as a porous zone, with defined porosity. The question arises when the user inputs are providing to fluent reaction window. Even iam defining rate of reactions by UDF, the Arrhenius rate equation window ask for values of Pre exp factor and activation energy, also temperature exponent. Do the values need again? Does it overwrite the reaction rates which already defines by UDF? Can you provide a solution to this issue.

[ajithjec]

Originally Posted by flappy

Hello,

I'm new to UDFs as well as to C programming and need help for my university work. I'm working on a fixed-bed reactor model with chemical reactions (Methane Steam Reforming) in a porous zone and want to implement the kinetic model of Xu and Froment as an UDF.

I'm using Fluent 17.0 and Microsoft Visual Studio 2015 for compiling on a 64-Bit Windows 7 Enterprise System.

apparently there is a mistake in my code. Fluent tells me: received a fatal signal (segmentation fault) - which i read in other forum threads is a sign that there is a mistake in the code.

here's my code:



I'd be enormously happy if anyone is willing to take some time and help me with my case. As i already mentioned, i hardly have any experience with the C programming language and am happy for any kind of advice given.

Kind regards,
f.

Use this before defining reaction , then it will allow reaction only in fluid domain with porous wall. In the cell zone condition, define porous zone with it properties and provide surface area to volume ratio.

if (FLUID_THREAD_P(t) && THREAD_VAR(t).fluid.porous)