Restarting limit_damping, other minor simplifications/cleanups

examples/granular/in.pour.drum

@@ -44,7 +44,7 @@ change_box      all boundary p p f
 
 pair_style      granular
 pair_coeff      1 * hertz/material 1e5 0.2 0.3 tangential mindlin NULL 1.0 0.5 damping tsuji
-pair_coeff      2 2 jkr 1e5 0.1 0.3 50 tangential mindlin NULL 1.0 0.5 rolling sds 1e3 1e3 0.1 twisting marshall damping tsuji
+pair_coeff      2 2 jkr 1e5 0.1 0.3 50 tangential mindlin NULL 1.0 0.5 rolling sds 1e3 1e3 0.1 twisting marshall 
 
 variable        theta equal 0
 

src/GRANULAR/fix_wall_gran.cpp

@@ -67,11 +67,9 @@ FixWallGran::FixWallGran(LAMMPS *lmp, int narg, char **arg) :
   model = new GranularModel(lmp);
   model->contact_type = WALL;
 
-  if (strcmp(arg[3],"granular") == 0)  classic_flag = 0;
-  else classic_flag = 1;
-  limit_damping = 0;
   heat_flag = 0;
-  int Twall_defined = 0;
+  int classic_flag = 1;
+  if (strcmp(arg[3],"granular") == 0)  classic_flag = 0;
 
   // wall/particle coefficients
 
@@ -117,10 +115,9 @@ FixWallGran::FixWallGran(LAMMPS *lmp, int narg, char **arg) :
     }
   }
 
-  // Define default damping model if unspecified, takes no args
+  // Define default damping submodel if unspecified, takes no args
   if (!model->damping_model)
     model->construct_submodel("viscoelastic", DAMPING);
-
   model->init();
 
   size_history = model->size_history;
@@ -177,6 +174,7 @@ FixWallGran::FixWallGran(LAMMPS *lmp, int narg, char **arg) :
   wiggle = 0;
   wshear = 0;
   peratom_flag = 0;
+  int Twall_defined = 0;
 
   while (iarg < narg) {
     if (strcmp(arg[iarg],"wiggle") == 0) {
@@ -216,7 +214,7 @@ FixWallGran::FixWallGran(LAMMPS *lmp, int narg, char **arg) :
   }
 
   if (heat_flag != Twall_defined)
-    error->all(FLERR, "To model conduction, must define both heat model and wall temperature");
+    error->all(FLERR, "Must define wall temperature with heat model");
 
   if (wallstyle == NOSTYLE)
     error->all(FLERR,"No wall style defined");
@@ -376,6 +374,7 @@ void FixWallGran::post_force(int /*vflag*/)
 
   history_update = 1;
   if (update->setupflag) history_update = 0;
+  model->history_update = history_update;
 
   // if just reneighbored:
   // update rigid body masses for owned atoms if using FixRigid
@@ -407,10 +406,10 @@ void FixWallGran::post_force(int /*vflag*/)
   if (wiggle) {
     double arg = omega * (update->ntimestep - time_origin) * dt;
     if (wallstyle == axis) {
-      wlo = lo + amplitude - amplitude*cos(arg);
+      wlo = lo + amplitude - amplitude * cos(arg);
-      whi = hi + amplitude - amplitude*cos(arg);
+      whi = hi + amplitude - amplitude * cos(arg);
     }
-    vwall[axis] = amplitude*omega*sin(arg);
+    vwall[axis] = amplitude * omega * sin(arg);
   } else if (wshear) vwall[axis] = vshear;
 
   // loop over all my atoms
@@ -473,19 +472,19 @@ void FixWallGran::post_force(int /*vflag*/)
       if (del1 < del2) dz = del1;
       else dz = -del2;
     } else if (wallstyle == ZCYLINDER) {
-      delxy = sqrt(x[i][0]*x[i][0] + x[i][1]*x[i][1]);
+      delxy = sqrt(x[i][0] * x[i][0] + x[i][1] * x[i][1]);
       delr = cylradius - delxy;
       if (delr > radius[i]) {
         dz = cylradius;
         rwall = 0.0;
       } else {
-        dx = -delr/delxy * x[i][0];
+        dx = -delr / delxy * x[i][0];
-        dy = -delr/delxy * x[i][1];
+        dy = -delr / delxy * x[i][1];
         // rwall = -2r_c if inside cylinder, 2r_c outside
-        rwall = (delxy < cylradius) ? -2*cylradius : 2*cylradius;
+        rwall = (delxy < cylradius) ? -2 * cylradius : 2 * cylradius;
         if (wshear && axis != 2) {
-          vwall[0] += vshear * x[i][1]/delxy;
+          vwall[0] += vshear * x[i][1] / delxy;
-          vwall[1] += -vshear * x[i][0]/delxy;
+          vwall[1] += -vshear * x[i][0] / delxy;
           vwall[2] = 0.0;
         }
       }
@@ -521,7 +520,6 @@ void FixWallGran::post_force(int /*vflag*/)
     model->meff = meff;
     model->vi = v[i];
     model->omegai = omega[i];
-    model->history_update = history_update;
     if (use_history) model->history = history_one[i];
     if (heat_flag) model->Ti = temperature[i];
 

src/GRANULAR/fix_wall_gran.h

@@ -53,7 +53,7 @@ class FixWallGran : public Fix {
 
  protected:
   int wallstyle, wiggle, wshear, axis;
-  int classic_flag, nlevels_respa;
+  int nlevels_respa;
   bigint time_origin;
 
   // for granular model choices
@@ -69,7 +69,6 @@ class FixWallGran : public Fix {
   int history_update;    // flag for whether shear history is updated
   int size_history;      // # of shear history values per contact
   int heat_flag;
-  int limit_damping;
 
   int tvar;
   char *tstr;

src/GRANULAR/fix_wall_gran_region.cpp

@@ -135,6 +135,7 @@ void FixWallGranRegion::post_force(int /*vflag*/)
 
   history_update = 1;
   if (update->setupflag) history_update = 0;
+  model->history_update = history_update;
 
   // if just reneighbored:
   // update rigid body masses for owned atoms if using FixRigid
@@ -262,7 +263,6 @@ void FixWallGranRegion::post_force(int /*vflag*/)
       model->meff = meff;
       model->vi = v[i];
       model->omegai = omega[i];
-      model->history_update = history_update;
 
       if (use_history) model->history = history_many[i][c2r[ic]];
       if (heat_flag) model->Ti = temperature[i];

src/GRANULAR/gran_sub_mod.h

@@ -29,8 +29,8 @@ class GranSubMod : protected Pointers {
   double *coeffs;
   void read_restart();
   virtual void mix_coeffs(double*, double*);
-  virtual void coeffs_to_local() = 0;
+  virtual void coeffs_to_local() {};
-  virtual void init() = 0;   // called after all submodel coeffs defined
+  virtual void init() {}; // called after all submodels + coeffs defined
 
   void allocate_coeffs();
   std::string name;

src/GRANULAR/gran_sub_mod_damping.cpp

@@ -108,7 +108,7 @@ void GranSubModDampingTsuji::init()
 {
   double tmp = gm->normal_model->damp;
   damp = 1.2728 - 4.2783 * tmp + 11.087 * square(tmp);
-  damp += -22.348 * cube(tmp)+ 27.467 * powint(tmp, 4);
+  damp += -22.348 * cube(tmp) + 27.467 * powint(tmp, 4);
   damp += -18.022 * powint(tmp, 5) + 4.8218 * powint(tmp,6);
 }
 

src/GRANULAR/gran_sub_mod_damping.h

@@ -48,8 +48,6 @@ class GranSubModDamping : public GranSubMod {
  public:
   GranSubModDamping(class GranularModel *, class LAMMPS *);
   ~GranSubModDamping() {};
-  virtual void coeffs_to_local() {};
-  virtual void mix_coeffs(double*, double*) {};
   virtual void init();
   virtual double calculate_forces() = 0;
   double damp_prefactor; // Used by tangential models

src/GRANULAR/gran_sub_mod_heat.h

@@ -35,8 +35,6 @@ class GranSubModHeat : public GranSubMod {
  public:
   GranSubModHeat(class GranularModel *, class LAMMPS *);
   ~GranSubModHeat() {};
-  virtual void coeffs_to_local() {};
-  virtual void init() {};
   virtual double calculate_heat() = 0;
 };
 

src/GRANULAR/gran_sub_mod_normal.h

@@ -51,8 +51,6 @@ class GranSubModNormal : public GranSubMod {
  public:
   GranSubModNormal(class GranularModel *, class LAMMPS *);
   ~GranSubModNormal() {};
-  virtual void coeffs_to_local() {};
-  virtual void init() {};
   virtual bool touch();
   virtual double pulloff_distance(double, double);
   virtual double calculate_area();
@@ -60,7 +58,7 @@ class GranSubModNormal : public GranSubMod {
   virtual void set_fncrit();
   double damp; // argument historically needed by damping
   double Emod, poiss;
-  double Fncrit, knfac;
+  double Fncrit;
   int material_properties, cohesive_flag;
 };
 

src/GRANULAR/gran_sub_mod_rolling.h

@@ -35,8 +35,6 @@ class GranSubModRolling : public GranSubMod {
  public:
   GranSubModRolling(class GranularModel *, class LAMMPS *);
   ~GranSubModRolling() {};
-  virtual void coeffs_to_local() {};
-  virtual void init() {};
   virtual void calculate_forces() = 0;
 };
 

src/GRANULAR/gran_sub_mod_tangential.h

@@ -63,8 +63,6 @@ class GranSubModTangential : public GranSubMod {
  public:
   GranSubModTangential(class GranularModel *, class LAMMPS *);
   virtual ~GranSubModTangential() {};
-  virtual void coeffs_to_local() {};
-  virtual void init() {};
   virtual void calculate_forces() = 0;
   double k, damp, mu; // Used by Marshall twisting model
 };
@@ -105,8 +103,6 @@ class GranSubModTangentialLinearHistoryClassic : public GranSubModTangentialLine
  public:
   GranSubModTangentialLinearHistoryClassic(class GranularModel *, class LAMMPS *);
   void calculate_forces();
- protected:
-  double xt;
 };
 
 /* ---------------------------------------------------------------------- */

src/GRANULAR/gran_sub_mod_twisting.h

@@ -39,8 +39,6 @@ class GranSubModTwisting : public GranSubMod {
  public:
   GranSubModTwisting(class GranularModel *, class LAMMPS *);
   virtual ~GranSubModTwisting() {};
-  virtual void coeffs_to_local() {};
-  virtual void init() {};
   virtual void calculate_forces() = 0;
 };
 
@@ -57,8 +55,8 @@ class GranSubModTwistingNone : public GranSubModTwisting {
 class GranSubModTwistingMarshall : public GranSubModTwisting {
  public:
   GranSubModTwistingMarshall(class GranularModel *, class LAMMPS *);
+  void init() override;
   void calculate_forces();
-  void init();
  protected:
   double k_tang, mu_tang;
 };

src/GRANULAR/granular_model.cpp

@@ -148,12 +148,13 @@ void GranularModel::construct_submodel(std::string model_name, SubmodelType mode
   sub_models[model_type]->name.assign(model_name);
   sub_models[model_type]->allocate_coeffs();
 
-  if (model_type == NORMAL) normal_model = dynamic_cast<GranSubModNormal *> (sub_models[NORMAL]);
+  // Assign specific submodel pointer
-  if (model_type == DAMPING) damping_model = dynamic_cast<GranSubModDamping *> (sub_models[DAMPING]);
+  if (model_type == NORMAL) normal_model = dynamic_cast<GranSubModNormal *> (sub_models[model_type]);
-  if (model_type == TANGENTIAL) tangential_model = dynamic_cast<GranSubModTangential *> (sub_models[TANGENTIAL]);
+  if (model_type == DAMPING) damping_model = dynamic_cast<GranSubModDamping *> (sub_models[model_type]);
-  if (model_type == ROLLING) rolling_model = dynamic_cast<GranSubModRolling *> (sub_models[ROLLING]);
+  if (model_type == TANGENTIAL) tangential_model = dynamic_cast<GranSubModTangential *> (sub_models[model_type]);
-  if (model_type == TWISTING) twisting_model = dynamic_cast<GranSubModTwisting *> (sub_models[TWISTING]);
+  if (model_type == ROLLING) rolling_model = dynamic_cast<GranSubModRolling *> (sub_models[model_type]);
-  if (model_type == HEAT) heat_model = dynamic_cast<GranSubModHeat *> (sub_models[HEAT]);
+  if (model_type == TWISTING) twisting_model = dynamic_cast<GranSubModTwisting *> (sub_models[model_type]);
+  if (model_type == HEAT) heat_model = dynamic_cast<GranSubModHeat *> (sub_models[model_type]);
 }
 
 /* ---------------------------------------------------------------------- */
@@ -248,7 +249,8 @@ void GranularModel::init()
       beyond_contact = 1;
     size_history += sub_models[i]->size_history;
     if (!sub_models[i]->allow_cohesion && normal_model->cohesive_flag)
-      error->all(FLERR,"Cannot use {} model with a cohesive normal model, {}", sub_models[i]->name, normal_model->name);
+      error->all(FLERR,"Cannot use {} model with a cohesive normal model, {}",
+                 sub_models[i]->name, normal_model->name);
     if (sub_models[i]->area_flag) area_flag = 1;
   }
 
@@ -260,14 +262,14 @@ void GranularModel::init()
 
     int j;
     for (int i = 0; i < size_history; i++) {
-      // Find which model owns this history value
+      // Find which submodel owns this history value
       size_cumulative = 0;
       for (j = 0; j < NSUBMODELS; j++) {
         if ((size_cumulative + sub_models[j]->size_history) > i) break;
         size_cumulative += sub_models[j]->size_history;
       }
 
-      // Check if model has nondefault transfers, if so copy its array
+      // Check if submodel has nondefault transfers, if so copy its array
       transfer_history_factor[i] = -1;
       if (j != NSUBMODELS) {
         if (sub_models[j]->nondefault_history_transfer) {
@@ -284,8 +286,7 @@ void GranularModel::init()
 
 int GranularModel::mix_coeffs(GranularModel *g1, GranularModel *g2)
 {
-  int i;
+  for (int i = 0; i < NSUBMODELS; i++) {
-  for (i = 0; i < NSUBMODELS; i++) {
     if (g1->sub_models[i]->name != g2->sub_models[i]->name) return i;
 
     construct_submodel(g1->sub_models[i]->name, (SubmodelType) i);
@@ -311,6 +312,8 @@ void GranularModel::write_restart(FILE *fp)
     fwrite(&num_coeffs, sizeof(int), 1, fp);
     fwrite(sub_models[i]->coeffs, sizeof(double), num_coeffs, fp);
   }
+
+  fwrite(&limit_damping, sizeof(int), 1, fp);
 }
 
 /* ---------------------------------------------------------------------- */
@@ -328,22 +331,23 @@ void GranularModel::read_restart(FILE *fp)
     if (comm->me == 0)
       utils::sfread(FLERR, const_cast<char*>(model_name.data()), sizeof(char),num_char, fp, nullptr, error);
     MPI_Bcast(const_cast<char*>(model_name.data()), num_char, MPI_CHAR, 0, world);
-
     construct_submodel(model_name, (SubmodelType) i);
 
-    if (comm->me == 0) {
+    if (comm->me == 0)
       utils::sfread(FLERR, &num_coeff, sizeof(int), 1, fp, nullptr, error);
-      if (num_coeff != sub_models[i]->num_coeffs)
-        error->one(FLERR, "Invalid granular model written to restart file");
-    }
     MPI_Bcast(&num_coeff, 1, MPI_INT, 0, world);
+    if (num_coeff != sub_models[i]->num_coeffs)
+      error->all(FLERR, "Invalid granular model written to restart file");
 
-    if (comm->me == 0) {
+    if (comm->me == 0)
       utils::sfread(FLERR, sub_models[i]->coeffs, sizeof(double), num_coeff, fp, nullptr, error);
-    }
     MPI_Bcast(sub_models[i]->coeffs, num_coeff, MPI_DOUBLE, 0, world);
     sub_models[i]->coeffs_to_local();
   }
+
+  if (comm->me == 0)
+    utils::sfread(FLERR, &limit_damping, sizeof(int), 1, fp, nullptr, error);
+  MPI_Bcast(&limit_damping, 1, MPI_INT, 0, world);
 }
 
 /* ---------------------------------------------------------------------- */
@@ -393,7 +397,7 @@ void GranularModel::calculate_forces()
   sub3(vi, vj, vr);
 
   // normal component
-  vnnr = dot3(vr, nx); //v_R . n
+  vnnr = dot3(vr, nx);
   scale3(vnnr, nx, vn);
 
   // tangential component
@@ -433,7 +437,7 @@ void GranularModel::calculate_forces()
   if (contact_type == PAIR) {
     copy3(torquesi, torquesj);
 
-    // Classic pair styles wouldn't scale, but classic option is only used by walls
+    // Classic pair styles don't scale, but classic option is currently only used by walls
     dist_to_contact = radi - 0.5 * delta;
     scale3(dist_to_contact, torquesi);
     dist_to_contact = radj - 0.5 * delta;

src/GRANULAR/granular_model.h

@@ -68,7 +68,7 @@ class GranularModel : protected Pointers {
   GranSubModRolling *rolling_model;
   GranSubModTwisting *twisting_model;
   GranSubModHeat *heat_model;
-  GranSubMod *sub_models[NSUBMODELS];  // Need to resize if we add more model flavors
+  GranSubMod *sub_models[NSUBMODELS];
 
   // Extra options
   int beyond_contact, limit_damping, history_update;
@@ -94,9 +94,9 @@ class GranularModel : protected Pointers {
   bool touch;
 
  protected:
-  int rolling_defined, twisting_defined, heat_defined; // Used to quickly skip undefined submodels
+  int rolling_defined, twisting_defined, heat_defined; // Flag optional submodels
-  int classic_model;
+  int classic_model;                                   // Flag original pair/gran calculations
-  int area_flag;
+  int area_flag;                                       // Flag whether area is needed
 
   int nclass;
 

src/GRANULAR/pair_granular.cpp

@@ -57,8 +57,6 @@ PairGranular::PairGranular(LAMMPS *lmp) : Pair(lmp)
   svector = new double[single_extra];
 
   neighprev = 0;
-  nmodels = 0;
-  maxmodels = 0;
   nmax = 0;
   mass_rigid = nullptr;
 
@@ -302,6 +300,7 @@ void PairGranular::allocate()
   maxmodels = n * n + 1; // should never need any more space
   models_list = (GranularModel **) memory->smalloc(maxmodels * sizeof(GranularModel *), "pair:models_list");
   for (int i = 0; i < maxmodels; i++) models_list[i] = nullptr;
+  nmodels = 0;
 
   onerad_dynamic = new double[n+1];
   onerad_frozen = new double[n+1];
@@ -339,7 +338,6 @@ void PairGranular::coeff(int narg, char **arg)
   utils::bounds(FLERR,arg[0],1,atom->ntypes,ilo,ihi,error);
   utils::bounds(FLERR,arg[1],1,atom->ntypes,jlo,jhi,error);
 
-
   // Construct new model
   models_list[nmodels] = new GranularModel(lmp);
   class GranularModel* model = models_list[nmodels];
@@ -374,9 +372,10 @@ void PairGranular::coeff(int narg, char **arg)
     } else error->all(FLERR, "Illegal pair_coeff command {}", arg[iarg]);
   }
 
-  // Define default damping model if unspecified, has no coeffs
+  // Define default damping submodel if unspecified, has no coeffs
   if (!model->damping_model)
     model->construct_submodel("viscoelastic", DAMPING);
+  model->init();
 
   int count = 0;
   for (int i = ilo; i <= ihi; i++) {
@@ -388,7 +387,7 @@ void PairGranular::coeff(int narg, char **arg)
     }
   }
 
-  // If there are > ntype^2 models, delete unused ones
+  // If there are > ntype^2 models, delete unused models
   if (nmodels == maxmodels) prune_models();
 
   if (count == 0) error->all(FLERR,"Incorrect args for pair coefficients");
@@ -419,7 +418,6 @@ void PairGranular::init_style()
   int size_max[NSUBMODELS] = {0};
   for (int n = 0; n < nmodels; n++) {
     model = models_list[n];
-    model->init();
 
     if (model->beyond_contact) {
       beyond_contact = 1;
@@ -559,7 +557,6 @@ double PairGranular::init_one(int i, int j)
                  model1->sub_models[error_code]->name,
                  model2->sub_models[error_code]->name);
 
-    // Initialize new model, init_one() occurs after init_style
     model->init();
 
     for (int k = 0; k < NSUBMODELS; k++)
@@ -657,8 +654,10 @@ void PairGranular::read_restart(FILE *fp)
   MPI_Bcast(&nmodels,1,MPI_INT,0,world);
 
   for (i = 0; i < nmodels; i++) {
+    delete models_list[i];
     models_list[i] = new GranularModel(lmp);
     models_list[i]->read_restart(fp);
+    models_list[i]->init();
   }
 
   for (i = 1; i <= atom->ntypes; i++) {
@@ -690,28 +689,22 @@ double PairGranular::single(int i, int j, int itype, int jtype,
                             double rsq, double /* factor_coul */,
                             double factor_lj, double &fforce)
 {
-
   if (factor_lj == 0) {
     fforce = 0.0;
     for (int m = 0; m < single_extra; m++) svector[m] = 0.0;
     return 0.0;
   }
 
-  int jnum;
-  int *jlist;
-  double *history,*allhistory;
-
   int nall = atom->nlocal + atom->nghost;
   if ((i >= nall) || (j >= nall))
     error->all(FLERR,"Not enough atoms for pair granular single function");
 
-  int touchflag;
-  double **x = atom->x;
-  double *radius = atom->radius;
-
   class GranularModel* model = models_list[types_indices[itype][jtype]];
 
   // Reset model and copy initial geometric data
+  double **x = atom->x;
+  double *radius = atom->radius;
+
   model->xi = x[i];
   model->xj = x[j];
   model->radi = radius[i];
@@ -719,8 +712,9 @@ double PairGranular::single(int i, int j, int itype, int jtype,
   model->history_update = 0; // Don't update history
 
   // If history is needed
-  jnum = list->numneigh[i];
+  double *history,*allhistory;
-  jlist = list->firstneigh[i];
+  int jnum = list->numneigh[i];
+  int *jlist = list->firstneigh[i];
   if (use_history) {
     if ((fix_history == nullptr) || (fix_history->firstvalue == nullptr))
       error->one(FLERR,"Pair granular single computation needs history");
@@ -730,11 +724,11 @@ double PairGranular::single(int i, int j, int itype, int jtype,
       if (neighprev >= jnum) neighprev = 0;
       if (jlist[neighprev] == j) break;
     }
-    history = &allhistory[size_history*neighprev];
+    history = &allhistory[size_history * neighprev];
     model->touch = fix_history->firstflag[i][neighprev];
   }
 
-  touchflag = model->check_contact();
+  int touchflag = model->check_contact();
 
   if (!touchflag) {
     fforce = 0.0;
@@ -745,17 +739,16 @@ double PairGranular::single(int i, int j, int itype, int jtype,
   // meff = effective mass of pair of particles
   // if I or J part of rigid body, use body mass
   // if I or J is frozen, meff is other particle
-  double mi, mj, meff;
   double *rmass = atom->rmass;
   int *mask = atom->mask;
 
-  mi = rmass[i];
+  double mi = rmass[i];
-  mj = rmass[j];
+  double mj = rmass[j];
   if (fix_rigid) {
     if (mass_rigid[i] > 0.0) mi = mass_rigid[i];
     if (mass_rigid[j] > 0.0) mj = mass_rigid[j];
   }
-  meff = mi * mj / (mi + mj);
+  double meff = mi * mj / (mi + mj);
   if (mask[i] & freeze_group_bit) meff = mj;
   if (mask[j] & freeze_group_bit) meff = mi;
 
@@ -770,11 +763,10 @@ double PairGranular::single(int i, int j, int itype, int jtype,
   model->omegaj = omega[j];
   model->history = history;
 
-  double *forces, *torquesi, *torquesj;
   model->calculate_forces();
-  forces = model->forces;
+  double *forces = model->forces;
-  torquesi = model->torquesi;
+  double *torquesi = model->torquesi;
-  torquesj = model->torquesj;
+  double *torquesj = model->torquesj;
 
   // apply forces & torques
   fforce = MathExtra::len3(forces);
@@ -843,7 +835,7 @@ void PairGranular::transfer_history(double* source, double* target, int itype, i
   class GranularModel* model = models_list[types_indices[itype][jtype]];
   if (model->nondefault_history_transfer) {
     for (int i = 0; i < size_history; i++) {
-      target[i] = model->transfer_history_factor[i] * source [i];
+      target[i] = model->transfer_history_factor[i] * source[i];
     }
   } else {
     for (int i = 0; i < size_history; i++) {