
[Sponsors] 
Update Pressure values for unsteady, multiphase, parallel  not updating properly 

LinkBack  Thread Tools  Search this Thread  Display Modes 
June 3, 2021, 04:04 
Update Pressure values for unsteady, multiphase, parallel  not updating properly

#1 
Member
Dhayananth
Join Date: Jun 2014
Posts: 37
Rep Power: 9 
Hello,
I am working on a unsteady flow problem with multiphase mixture model and run the simulation in Parallel mode. For my problem, I have written a udf to update the pressure value of face boundary at every time step using the mass flow of the previous time step. Code: #include"udf.h" real Q12; /* current time step flow rate */ #define BCA_ID 63 /* domain id*/ DEFINE_EXECUTE_AT_END(flow_rate_bca) { #if !RP_HOST Domain *d; real m; cell_t c; Thread *t; face_t f; d = Get_Domain(1); t= Lookup_Thread(d, BCA_ID); /* loop to calculate mass flow rate at outlet */ begin_f_loop(f,t) { if (PRINCIPAL_FACE_P(f,t)) { m += F_FLUX(f,t); } } end_f_loop(f,t) Q12 = m / 1060.0; /* convert mass flow rate to volume flow rate */ #if RP_NODE /* Perform node synchronized actions here. Does nothing in Serial */ Q12 = PRF_GRSUM1(Q12); #endif /* RP_NODE */ m = 0.0; #endif /* !RP_HOST */ } DEFINE_PROFILE(pressure_out_bca,t,i) { #if !RP_HOST real Z = 5.1918E+07; real C = 8.697E10; real R = 10.6080E+08; real ts = CURRENT_TIMESTEP; face_t f; begin_f_loop(f,t) { if (CURRENT_TIME == 0) { F_UDMI(f,t,0) = 13332.2; F_UDMI(f,t,1) = F_UDMI(f,t,0); F_UDMI(f,t,2) = 1.22E05; } /* F_UDMI(f,t,0)  previous time step pressure F_UDMI(f,t,1)  current time step pressure F_UDMI(f,t,2)  previous time step flow rate Q12  current time step flow rate */ else { F_UDMI(f,t,3) = Q12; F_UDMI(f,t,1) = ((F_UDMI(f,t,0)*R*C)+(F_UDMI(f,t,3)*R*ts)+(F_UDMI( f,t,3)*Z*ts)+(Z*R*C*F_UDMI(f,t,3))(Z*R*C*F_UDMI(f,t,2))) / (ts +(R*C)); F_UDMI(f,t,0) = F_UDMI(f,t,1); F_UDMI(f,t,2) = F_UDMI(f,t,3); } } end_f_loop(f,t) begin_f_loop(f,t) { F_PROFILE(f,t,i) = F_UDMI(f,t,1); /* Calculate pressure and apply to outlet*/ } end_f_loop(f,t) #endif /* !RP_HOST */ } Question: When I physically calculate the pressure values and check with pressure output files from simulation, it is not matching. Is there any error in the code which is causing this problem? 

June 3, 2021, 06:29 

#2 
Senior Member
Join Date: Nov 2013
Posts: 1,915
Rep Power: 23 
It could also be an error in your own calculation...
How big is the difference? Which values do you get, which values did you calculate? Which equation do you think you are using?
__________________
"The UDF library you are trying to load (libudf) is not compiled for parallel use on the current platform" is NOT the error after compiling. It is the error after loading. To see compiler errors, look at your screen after you click "build". 

June 3, 2021, 08:49 

#3 
Member
Dhayananth
Join Date: Jun 2014
Posts: 37
Rep Power: 9 
Hi Pakk,
Thank you for your reply. I am using the following equation to calculate the pressure values for the outlet. P(n+1) = {[P(n)*R*C] + [Q(n+1)*R*ts] + [Q(n+1)*Z*ts] + [Z*R*C*Q(n+1)]  [Z*R*C*Q(n)]} / (ts +(R*C)) F_UDMI(f,t,0) corresponds to P(n) F_UDMI(f,t,1) corresponds to P(n+1) F_UDMI(f,t,2) corresponds to Q(n) F_UDMI(f,t,3) corresponds to Q(n+1) Initializing, Q(n) = 1.22E05 , P(n) = 13332.4 for CURRENT_TIME = 0 Result from simulation: Q (n) Q(n+1) P(n) P(n+1) 1.22E05 3.266E05 13332.2 133324.4 3.266E05 8.902E05 133324.4 10423.9 Result from physical calculation: Q (n) Q(n+1) P(n) P(n+1) 1.22E05 3.266E05 13332.2 14629.7 3.266E05 8.902E05 14629.7 18429.7 

June 3, 2021, 22:59 

#4 
Senior Member
Alexander
Join Date: Apr 2013
Posts: 1,762
Rep Power: 28 
you may write to console each term of your equation using Message0("text"); macro
code seems to be correct. the only thing I would do > split time step definition with applying the value: was real ts = CURRENT_TIMESTEP; to be real ts; ts = CURRENT_TIMESTEP;
__________________
best regards ****************************** press LIKE if this message was helpful 

June 4, 2021, 00:57 

#5 
Senior Member
Join Date: Nov 2013
Posts: 1,915
Rep Power: 23 
You update F_UDMI(f,t,1) before you update F_UDMI(f,t,0), which seems wrong. Does not explain your current problem, however.
__________________
"The UDF library you are trying to load (libudf) is not compiled for parallel use on the current platform" is NOT the error after compiling. It is the error after loading. To see compiler errors, look at your screen after you click "build". 

June 4, 2021, 07:36 

#6 
Member
Dhayananth
Join Date: Jun 2014
Posts: 37
Rep Power: 9 
Hi Alexander and Pakk,
Thank you for your reply. I have incorporated your suggestions and made few changes in the code for updating the previous time step values. The following code is moved out of face loop and made separate face loop for the code. F_UDMI(f,t,0) = F_UDMI(f,t,1); F_UDMI(f,t,2) = F_UDMI(f,t,3); Using Message command, I have checked the values of F_UDMI variables with physical calculation, the pressure values are matching. But if I check F_UDMI values with pressure values from the report files it is not matching. Why? For example, F_UDMI = 16733 Pa, Pressure from report file = 16084 Pa I want to update the pressure values for mixture domain only. But while printing the values, I come to know that the udf is updating for all the domains. So how to make the define profile macro to update mixture domain alone? Revised code: #include"udf.h" real Q12; /* current time step flow rate */ #define BCA_ID 63 /* domain id*/ DEFINE_EXECUTE_AT_END(flow_rate_bca) { #if !RP_HOST Domain *d; real m; cell_t c; Thread *t; face_t f; d = Get_Domain(1); t= Lookup_Thread(d, BCA_ID); /* loop to calculate mass flow rate at outlet */ begin_f_loop(f,t) { if (PRINCIPAL_FACE_P(f,t)) { m += F_FLUX(f,t); } } end_f_loop(f,t) Q12 = m / 1060.0; /* convert mass flow rate to volume flow rate */ #if RP_NODE /* Perform node synchronized actions here. Does nothing in Serial */ Q12 = PRF_GRSUM1(Q12); #endif /* RP_NODE */ m = 0.0; #endif /* !RP_HOST */ } DEFINE_INIT(init_pressure_bca,d) { Thread *t1; cell_t c; t1 = Lookup_Thread(d,BCA_ID); // for BCA real P = 13332.2; thread_loop_c(t1, d) { begin_c_loop_all(c,t1) { C_P(c,t1) = P; } end_c_loop_all(c, t1) } #if RP_HOST Message("BCA Pressure Initalised\n"); #endif } DEFINE_PROFILE(pressure_out_bca,t,i) { #if !RP_HOST real Z = 5.1918E+07; real C = 8.697E10; real R = 10.6080E+08; real ts; ts = CURRENT_TIMESTEP; face_t f; begin_f_loop(f,t) { if (CURRENT_TIME == 0) { F_UDMI(f,t,0) = 13332.2; F_UDMI(f,t,1) = F_UDMI(f,t,0); F_UDMI(f,t,3) = 1.22E05; } /* F_UDMI(f,t,0)  previous time step pressure F_UDMI(f,t,1)  current time step pressure F_UDMI(f,t,2)  prevoius time step flow rate Q12  current time step flow rate */ else { F_UDMI(f,t,3) = fabs (Q12); F_UDMI(f,t,1) = ((F_UDMI(f,t,0)*R*C)+(F_UDMI(f,t,3)*R*ts)+(F_UDMI( f,t,3)*Z*ts)+(Z*R*C*F_UDMI(f,t,3))(Z*R*C*F_UDMI(f,t,2))) / (ts +(R*C)); } } end_f_loop(f,t) begin_f_loop(f,t) { F_UDMI(f,t,0) = F_UDMI(f,t,1); F_UDMI(f,t,2) = F_UDMI(f,t,3); } end_f_loop(f,t) begin_f_loop(f,t) { F_PROFILE(f,t,i) = F_UDMI(f,t,1); /* Calculate pressure and apply to outlet*/ } end_f_loop(f,t) #endif /* !RP_HOST */ } 

June 7, 2021, 01:43 

#7  
Member
Dhayananth
Join Date: Jun 2014
Posts: 37
Rep Power: 9 
Hi,
While searching for multiphase Define profile macro, I cam across the below line from the udf manual: Quote:
I think.. if I got the answer to this question. I will solve the error in the above post. 

Thread Tools  Search this Thread 
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
question regarding LES of pipe flow  pimpleFoam  Dan1788  OpenFOAM Running, Solving & CFD  37  December 26, 2017 14:42 
Neumann pressure BC and velocity field  Antech  Main CFD Forum  0  April 25, 2006 02:15 
Gas pressure question  Dan Moskal  Main CFD Forum  0  October 24, 2002 22:02 
Hydrostatic pressure in 2phase flow modeling (CFX4.2)  HB &DS  CFX  0  January 9, 2000 13:19 
Hydrostatic pressure in 2phase flow modeling (long)  DS & HB  Main CFD Forum  0  January 8, 2000 15:00 