#include "compute_temp_deform.h"
#include <mpi.h>
#include <cstring>
#include "domain.h"
#include "atom.h"
#include "update.h"
#include "force.h"
#include "modify.h"
#include "fix.h"
#include "fix_deform.h"
#include "group.h"
#include "comm.h"
#include "memory.h"
#include "error.h"
using namespace LAMMPS_NS;
ComputeTempDeform::ComputeTempDeform(LAMMPS *lmp, int narg, char **arg) :
Compute(lmp, narg, arg)
{
if (narg != 3) error->all(FLERR,"Illegal compute temp/deform command");
scalar_flag = vector_flag = 1;
size_vector = 6;
extscalar = 0;
extvector = 1;
tempflag = 1;
tempbias = 1;
maxbias = 0;
vbiasall = NULL;
vector = new double[size_vector];
}
ComputeTempDeform::~ComputeTempDeform()
{
memory->destroy(vbiasall);
delete [] vector;
}
void ComputeTempDeform::init()
{
int i;
for (i = 0; i < modify->nfix; i++)
if (strcmp(modify->fix[i]->style,"deform") == 0) {
if (((FixDeform *) modify->fix[i])->remapflag == Domain::X_REMAP &&
comm->me == 0)
error->warning(FLERR,"Using compute temp/deform with inconsistent "
"fix deform remap option");
break;
}
if (i == modify->nfix && comm->me == 0)
error->warning(FLERR,
"Using compute temp/deform with no fix deform defined");
}
void ComputeTempDeform::setup()
{
dynamic = 0;
if (dynamic_user || group->dynamic[igroup]) dynamic = 1;
dof_compute();
}
void ComputeTempDeform::dof_compute()
{
adjust_dof_fix();
natoms_temp = group->count(igroup);
dof = domain->dimension * natoms_temp;
dof -= extra_dof + fix_dof;
if (dof > 0) tfactor = force->mvv2e / (dof * force->boltz);
else tfactor = 0.0;
}
double ComputeTempDeform::compute_scalar()
{
double lamda[3],vstream[3],vthermal[3];
invoked_scalar = update->ntimestep;
double **x = atom->x;
double **v = atom->v;
double *mass = atom->mass;
double *rmass = atom->rmass;
int *type = atom->type;
int *mask = atom->mask;
int nlocal = atom->nlocal;
double *h_rate = domain->h_rate;
double *h_ratelo = domain->h_ratelo;
double t = 0.0;
for (int i = 0; i < nlocal; i++)
if (mask[i] & groupbit) {
domain->x2lamda(x[i],lamda);
vstream[0] = h_rate[0]*lamda[0] + h_rate[5]*lamda[1] +
h_rate[4]*lamda[2] + h_ratelo[0];
vstream[1] = h_rate[1]*lamda[1] + h_rate[3]*lamda[2] + h_ratelo[1];
vstream[2] = h_rate[2]*lamda[2] + h_ratelo[2];
vthermal[0] = v[i][0] - vstream[0];
vthermal[1] = v[i][1] - vstream[1];
vthermal[2] = v[i][2] - vstream[2];
if (rmass)
t += (vthermal[0]*vthermal[0] + vthermal[1]*vthermal[1] +
vthermal[2]*vthermal[2]) * rmass[i];
else
t += (vthermal[0]*vthermal[0] + vthermal[1]*vthermal[1] +
vthermal[2]*vthermal[2]) * mass[type[i]];
}
MPI_Allreduce(&t,&scalar,1,MPI_DOUBLE,MPI_SUM,world);
if (dynamic) dof_compute();
if (dof < 0.0 && natoms_temp > 0.0)
error->all(FLERR,"Temperature compute degrees of freedom < 0");
scalar *= tfactor;
return scalar;
}
void ComputeTempDeform::compute_vector()
{
double lamda[3],vstream[3],vthermal[3];
invoked_vector = update->ntimestep;
double **x = atom->x;
double **v = atom->v;
double *mass = atom->mass;
double *rmass = atom->rmass;
int *type = atom->type;
int *mask = atom->mask;
int nlocal = atom->nlocal;
double *h_rate = domain->h_rate;
double *h_ratelo = domain->h_ratelo;
double massone,t[6];
for (int i = 0; i < 6; i++) t[i] = 0.0;
for (int i = 0; i < nlocal; i++)
if (mask[i] & groupbit) {
domain->x2lamda(x[i],lamda);
vstream[0] = h_rate[0]*lamda[0] + h_rate[5]*lamda[1] +
h_rate[4]*lamda[2] + h_ratelo[0];
vstream[1] = h_rate[1]*lamda[1] + h_rate[3]*lamda[2] + h_ratelo[1];
vstream[2] = h_rate[2]*lamda[2] + h_ratelo[2];
vthermal[0] = v[i][0] - vstream[0];
vthermal[1] = v[i][1] - vstream[1];
vthermal[2] = v[i][2] - vstream[2];
if (rmass) massone = rmass[i];
else massone = mass[type[i]];
t[0] += massone * vthermal[0]*vthermal[0];
t[1] += massone * vthermal[1]*vthermal[1];
t[2] += massone * vthermal[2]*vthermal[2];
t[3] += massone * vthermal[0]*vthermal[1];
t[4] += massone * vthermal[0]*vthermal[2];
t[5] += massone * vthermal[1]*vthermal[2];
}
MPI_Allreduce(t,vector,6,MPI_DOUBLE,MPI_SUM,world);
for (int i = 0; i < 6; i++) vector[i] *= force->mvv2e;
}
void ComputeTempDeform::remove_bias(int i, double *v)
{
double lamda[3];
double *h_rate = domain->h_rate;
double *h_ratelo = domain->h_ratelo;
domain->x2lamda(atom->x[i],lamda);
vbias[0] = h_rate[0]*lamda[0] + h_rate[5]*lamda[1] +
h_rate[4]*lamda[2] + h_ratelo[0];
vbias[1] = h_rate[1]*lamda[1] + h_rate[3]*lamda[2] + h_ratelo[1];
vbias[2] = h_rate[2]*lamda[2] + h_ratelo[2];
v[0] -= vbias[0];
v[1] -= vbias[1];
v[2] -= vbias[2];
}
void ComputeTempDeform::remove_bias_thr(int i, double *v, double *b)
{
double lamda[3];
double *h_rate = domain->h_rate;
double *h_ratelo = domain->h_ratelo;
domain->x2lamda(atom->x[i],lamda);
b[0] = h_rate[0]*lamda[0] + h_rate[5]*lamda[1] +
h_rate[4]*lamda[2] + h_ratelo[0];
b[1] = h_rate[1]*lamda[1] + h_rate[3]*lamda[2] + h_ratelo[1];
b[2] = h_rate[2]*lamda[2] + h_ratelo[2];
v[0] -= b[0];
v[1] -= b[1];
v[2] -= b[2];
}
void ComputeTempDeform::remove_bias_all()
{
double **v = atom->v;
int *mask = atom->mask;
int nlocal = atom->nlocal;
if (atom->nmax > maxbias) {
memory->destroy(vbiasall);
maxbias = atom->nmax;
memory->create(vbiasall,maxbias,3,"temp/deform:vbiasall");
}
double lamda[3];
double *h_rate = domain->h_rate;
double *h_ratelo = domain->h_ratelo;
for (int i = 0; i < nlocal; i++)
if (mask[i] & groupbit) {
domain->x2lamda(atom->x[i],lamda);
vbiasall[i][0] = h_rate[0]*lamda[0] + h_rate[5]*lamda[1] +
h_rate[4]*lamda[2] + h_ratelo[0];
vbiasall[i][1] = h_rate[1]*lamda[1] + h_rate[3]*lamda[2] + h_ratelo[1];
vbiasall[i][2] = h_rate[2]*lamda[2] + h_ratelo[2];
v[i][0] -= vbiasall[i][0];
v[i][1] -= vbiasall[i][1];
v[i][2] -= vbiasall[i][2];
}
}
void ComputeTempDeform::restore_bias(int , double *v)
{
v[0] += vbias[0];
v[1] += vbias[1];
v[2] += vbias[2];
}
void ComputeTempDeform::restore_bias_thr(int , double *v, double *b)
{
v[0] += b[0];
v[1] += b[1];
v[2] += b[2];
}
void ComputeTempDeform::restore_bias_all()
{
double **v = atom->v;
int *mask = atom->mask;
int nlocal = atom->nlocal;
for (int i = 0; i < nlocal; i++)
if (mask[i] & groupbit) {
v[i][0] += vbiasall[i][0];
v[i][1] += vbiasall[i][1];
v[i][2] += vbiasall[i][2];
}
}
double ComputeTempDeform::memory_usage()
{
double bytes = 3*maxbias * sizeof(double);
return bytes;
}