Learning Forum

[Audy Maulizar]

Dear all, 

I would like to ask for help regarding the dynamic simulation of vortex-induced oscillation on a bridge deck as shown below.

I use the 6DOF solver to calculate displacements of the bridge deck due to the aerodynamic forces and for that I've assigned the properties using a UDF with the DEFINE_SDOF_PROPERTIES function since I want to introduce damping to my model. For the UDF, I used codes shown in https://www.youtube.com/watch?v=7MeHvOrXl8g&t=522s as a reference with some modifications to fit with my case study. My UDF source code is outlined as follows:

#include "udf.h"
#include "dynamesh_tools.h"

FILE *fout;

DEFINE_SDOF_PROPERTIES(bridge_motion_standard, prop, dt, time, dtime)
{
    Thread *t;
    Domain *d = Get_Domain(1);
    /*real x_cg[3], force[3], moment[3];*/
    real pos_x, pos_y, pos_theta;
    real cg_x = 0; /*center of gravity x*/
    real cg_y = -0.025; /*center of gravity y*/
    real defo_x, defo_y, rot_z; /*displacement variables*/
    real vel_x, vel_y, rot_vel; /*velocity variable*/
    real Fx, Fy, Mz; /*force*/
    real mass = 33.75; /*bridge deck model*/
    real mom_inertia = 1.73; /*moment inertia*/
    real fh = 9.5; /*heave frequency in Hz*/
    real ft = 20.0; /*torsion frequency in Hz*/
    real fd = 0.8*fh; /*drag frequency in Hz*/
    /*natural frequencies in rad/s*/
    real wh = 2*M_PI*fh;
    real wt = 2*M_PI*ft;
    real wd = 2*M_PI*fd;
    real zeta = 0.005; /*damping ratio*/
    /*stiffness calculation*/
    real kh = mass*wh*wh;
    real kt = mom_inertia*wt*wt;
    real kd = mass*wd*wd;
    /*damping coefficient calculation*/
    real ch = 2*mass*wh*zeta;
    real ct = 2*mom_inertia*wt*zeta;
    real cd = 2*mass*wd*zeta;
    
    t = DT_THREAD(dt);
        
        prop[SDOF_MASS] = mass; /* total mass:total length*/
        prop[SDOF_IZZ] = mom_inertia; /*mass moment inertia*/
        
        prop[SDOF_ZERO_TRANS_X] = FALSE;
        prop[SDOF_ZERO_TRANS_Y] = FALSE;
        prop[SDOF_ZERO_TRANS_Z] = TRUE;
        prop[SDOF_ZERO_ROT_X] = TRUE;
        prop[SDOF_ZERO_ROT_Y] = TRUE;
        prop[SDOF_ZERO_ROT_Z] = FALSE;
        
        pos_x = (DT_CG(dt)[0]);
        pos_y = (DT_CG(dt)[1]);
        pos_theta = (DT_THETA(dt)[2]*M_PI/180);
        
        /*deformation calculation*/
        defo_x = pos_x - cg_x;
        defo_y = pos_y - cg_y;
        rot_z = pos_theta;
        
        vel_x = (DT_VEL_CG(dt)[0]);
        vel_y = (DT_VEL_CG(dt)[1]);
        rot_vel = (DT_OMEGA_CG(dt)[2]);
        
        /*spring force and damping force calculation*/
        prop[SDOF_LOAD_F_X] = -kd*defo_x - cd*vel_x;
        prop[SDOF_LOAD_F_Y] = -kh*defo_y - ch*vel_y;
        prop[SDOF_LOAD_M_Z] = -kt*rot_z - ct*rot_vel;
        
        /*results writing to txt files*/
        fout = fopen("displacement_results.txt", "a"); 
        fprintf(fout, "%g %f %f %f \n", time, defo_x, defo_y, rot_z);
        fclose(fout);

        fout = fopen("velocity_results.txt", "a");
        fprintf(fout, "%g %f %f %f \n", time, vel_x, vel_y, rot_vel);
        fclose(fout);

        fout = fopen("acceleration_results.txt", "a");
        fprintf(fout, "%g %f %f %f \n", time, DT_ACC_CG(dt)[0], DT_ACC_CG(dt)[1], DT_ALPHA_CG(dt)[2]);
        fclose(fout);
}

 

I have tried compiling and hooking the UDF and there was no problem with that process, however the analysis results showed that the lift-forces and the displacements of the bridge deck increasing continuously instead of reaching a steady-state condition as predicted by other papers related to VIV on a bridge deck. It just seemed like the resistance forces from spring and damping did not exist. Does anybody have any idea what the problem is? I am afraid that FLUENT somehow was not able to read my UDF and it just ended up getting ignored by the program, but I am not sure.

[Rob]

I'm not going to debug the code, but you may find posting the vel_x, defo_x etc to UDMs to be helpful. That'll show if the data is there or not.