import ctypes as ct
import numpy as np
from sharpy.utils.solver_interface import solver, BaseSolver, solver_from_string
import sharpy.utils.settings as settings
import sharpy.structure.utils.xbeamlib as xbeamlib
import sharpy.utils.multibody as mb
import sharpy.structure.utils.lagrangeconstraints as lagrangeconstraints
import sharpy.utils.exceptions as exc
_BaseStructural = solver_from_string('_BaseStructural')
[docs]@solver
class NonLinearDynamicMultibody(_BaseStructural):
"""
Nonlinear dynamic multibody
Nonlinear dynamic step solver for multibody structures.
"""
solver_id = 'NonLinearDynamicMultibody'
solver_classification = 'structural'
settings_types = _BaseStructural.settings_types.copy()
settings_default = _BaseStructural.settings_default.copy()
settings_description = _BaseStructural.settings_description.copy()
settings_table = settings.SettingsTable()
__doc__ += settings_table.generate(settings_types, settings_default, settings_description)
def __init__(self):
self.data = None
self.settings = None
# Total number of unknowns in the Multybody sistem
self.sys_size = None
# Total number of equations associated to the Lagrange multipliers
self.lc_list = None
self.num_LM_eq = None
self.gamma = None
self.beta = None
def initialise(self, data, custom_settings=None):
self.data = data
if custom_settings is None:
self.settings = data.settings[self.solver_id]
else:
self.settings = custom_settings
settings.to_custom_types(self.settings, self.settings_types, self.settings_default)
# load info from dyn dictionary
self.data.structure.add_unsteady_information(
self.data.structure.dyn_dict, self.settings['num_steps'].value)
# Define Newmark constants
self.gamma = 0.5 + self.settings['newmark_damp'].value
self.beta = 0.25*(self.gamma + 0.5)*(self.gamma + 0.5)
# Define the number of equations
self.lc_list = lagrangeconstraints.initialize_constraints(self.data.structure.ini_mb_dict)
self.num_LM_eq = lagrangeconstraints.define_num_LM_eq(self.lc_list)
# Define the number of dofs
self.define_sys_size()
def add_step(self):
self.data.structure.next_step()
def next_step(self):
pass
def define_sys_size(self):
MBdict = self.data.structure.ini_mb_dict
self.sys_size = self.data.structure.num_dof.value
for ibody in range(self.data.structure.num_bodies):
if (MBdict['body_%02d' % ibody]['FoR_movement'] == 'free'):
self.sys_size += 10
def assembly_MB_eq_system(self, MB_beam, MB_tstep, ts, dt, Lambda, Lambda_dot, MBdict):
self.lc_list = lagrangeconstraints.initialize_constraints(MBdict)
self.num_LM_eq = lagrangeconstraints.define_num_LM_eq(self.lc_list)
MB_M = np.zeros((self.sys_size+self.num_LM_eq, self.sys_size+self.num_LM_eq), dtype=ct.c_double, order='F')
MB_C = np.zeros((self.sys_size+self.num_LM_eq, self.sys_size+self.num_LM_eq), dtype=ct.c_double, order='F')
MB_K = np.zeros((self.sys_size+self.num_LM_eq, self.sys_size+self.num_LM_eq), dtype=ct.c_double, order='F')
MB_Asys = np.zeros((self.sys_size+self.num_LM_eq, self.sys_size+self.num_LM_eq), dtype=ct.c_double, order='F')
MB_Q = np.zeros((self.sys_size+self.num_LM_eq,), dtype=ct.c_double, order='F')
#ipdb.set_trace()
first_dof = 0
last_dof = 0
# Loop through the different bodies
for ibody in range(len(MB_beam)):
# Initialize matrices
M = None
C = None
K = None
Q = None
# Generate the matrices for each body
if MB_beam[ibody].FoR_movement == 'prescribed':
last_dof = first_dof + MB_beam[ibody].num_dof.value
M, C, K, Q = xbeamlib.cbeam3_asbly_dynamic(MB_beam[ibody], MB_tstep[ibody], self.settings)
elif MB_beam[ibody].FoR_movement == 'free':
last_dof = first_dof + MB_beam[ibody].num_dof.value + 10
M, C, K, Q = xbeamlib.xbeam3_asbly_dynamic(MB_beam[ibody], MB_tstep[ibody], self.settings)
############### Assembly into the global matrices
# Flexible and RBM contribution to Asys
MB_M[first_dof:last_dof, first_dof:last_dof] = M.astype(dtype=ct.c_double, copy=True, order='F')
MB_C[first_dof:last_dof, first_dof:last_dof] = C.astype(dtype=ct.c_double, copy=True, order='F')
MB_K[first_dof:last_dof, first_dof:last_dof] = K.astype(dtype=ct.c_double, copy=True, order='F')
#Q
MB_Q[first_dof:last_dof] = Q
first_dof = last_dof
# Define the number of equations
# Generate matrices associated to Lagrange multipliers
LM_C, LM_K, LM_Q = lagrangeconstraints.generate_lagrange_matrix(
self.lc_list,
MB_beam,
MB_tstep,
ts,
self.num_LM_eq,
self.sys_size,
dt,
Lambda,
Lambda_dot,
"dynamic")
# Include the matrices associated to Lagrange Multipliers
MB_C += LM_C
MB_K += LM_K
MB_Q += LM_Q
MB_Asys = MB_K + MB_C*self.gamma/(self.beta*dt) + MB_M/(self.beta*dt*dt)
return MB_Asys, MB_Q
def integrate_position(self, MB_beam, MB_tstep, dt):
vel = np.zeros((6,),)
acc = np.zeros((6,),)
for ibody in range(0, len(MB_tstep)):
# I think this is the right way to do it, but to make it match the rest I change it temporally
if True:
# MB_tstep[ibody].mb_quat[ibody,:] = algebra.quaternion_product(MB_tstep[ibody].quat, MB_tstep[ibody].mb_quat[ibody,:])
acc[0:3] = (0.5-self.beta)*np.dot(MB_beam[ibody].timestep_info.cga(),MB_beam[ibody].timestep_info.for_acc[0:3])+self.beta*np.dot(MB_tstep[ibody].cga(),MB_tstep[ibody].for_acc[0:3])
vel[0:3] = np.dot(MB_beam[ibody].timestep_info.cga(),MB_beam[ibody].timestep_info.for_vel[0:3])
MB_tstep[ibody].for_pos[0:3] += dt*(vel[0:3] + dt*acc[0:3])
else:
MB_tstep[ibody].for_pos[0:3] += dt*np.dot(MB_tstep[ibody].cga(),MB_tstep[ibody].for_vel[0:3])
# Use next line for double pendulum (fix position of the second FoR)
# MB_tstep[ibody].for_pos[0:3] = np.dot(algebra.quat2rotation(MB_tstep[0].quat), MB_tstep[0].pos[-1,:])
# print("tip final pos: ", np.dot(algebra.quat2rotation(MB_tstep[0].quat), MB_tstep[0].pos[-1,:]))
# print("FoR final pos: ", MB_tstep[ibody].for_pos[0:3])
# print("pause")
def extract_resultants(self):
# TODO: code
pass
def compute_forces_constraints(self, MB_beam, MB_tstep, ts, dt, Lambda, Lambda_dot):
try:
self.lc_list[0]
except IndexError:
return
# TODO the output of this routine is wrong. check at some point.
LM_C, LM_K, LM_Q = lagrangeconstraints.generate_lagrange_matrix(self.lc_list, MB_beam, MB_tstep, ts, self.num_LM_eq, self.sys_size, dt, Lambda, Lambda_dot, "dynamic")
F = -np.dot(LM_C[:, -self.num_LM_eq:], Lambda_dot) - np.dot(LM_K[:, -self.num_LM_eq:], Lambda)
first_dof = 0
for ibody in range(len(MB_beam)):
# Forces associated to nodes
body_numdof = MB_beam[ibody].num_dof.value
body_freenodes = np.sum(MB_beam[ibody].vdof > -1)
last_dof = first_dof + body_numdof
MB_tstep[ibody].forces_constraints_nodes[(MB_beam[ibody].vdof > -1), :] = F[first_dof:last_dof].reshape(body_freenodes, 6, order='C')
# Forces associated to the frame of reference
if MB_beam[ibody].FoR_movement == 'free':
# TODO: How are the forces in the quaternion equation interpreted?
MB_tstep[ibody].forces_constraints_FoR[ibody, :] = F[last_dof:last_dof+10]
last_dof += 10
first_dof = last_dof
# print(MB_tstep[ibody].forces_constraints_nodes)
# TODO: right now, these forces are only used as an output, they are not read when the multibody is splitted
def run(self, structural_step=None, dt=None):
if structural_step is None:
structural_step = self.data.structure.timestep_info[-1]
if structural_step.mb_dict is not None:
MBdict = structural_step.mb_dict
else:
MBdict = self.data.structure.ini_mb_dict
if dt is None:
dt = self.settings['dt'].value
else:
self.settings['dt'] = ct.c_float(dt)
self.lc_list = lagrangeconstraints.initialize_constraints(MBdict)
self.num_LM_eq = lagrangeconstraints.define_num_LM_eq(self.lc_list)
# TODO: only working for constant forces
MB_beam, MB_tstep = mb.split_multibody(
self.data.structure,
structural_step,
MBdict,
self.data.ts)
# Lagrange multipliers parameters
num_LM_eq = self.num_LM_eq
Lambda = np.zeros((num_LM_eq,), dtype=ct.c_double, order='F')
Lambda_dot = np.zeros((num_LM_eq,), dtype=ct.c_double, order='F')
# Initialize
q = np.zeros((self.sys_size + num_LM_eq,), dtype=ct.c_double, order='F')
dqdt = np.zeros((self.sys_size + num_LM_eq,), dtype=ct.c_double, order='F')
dqddt = np.zeros((self.sys_size + num_LM_eq,), dtype=ct.c_double, order='F')
# Predictor step
mb.disp2state(MB_beam, MB_tstep, q, dqdt, dqddt)
q += dt*dqdt + (0.5 - self.beta)*dt*dt*dqddt
dqdt += (1.0 - self.gamma)*dt*dqddt
dqddt = np.zeros((self.sys_size + num_LM_eq,), dtype=ct.c_double, order='F')
if not num_LM_eq == 0:
Lambda = q[-num_LM_eq:].astype(dtype=ct.c_double, copy=True, order='F')
Lambda_dot = dqdt[-num_LM_eq:].astype(dtype=ct.c_double, copy=True, order='F')
else:
Lambda = 0
Lambda_dot = 0
# Newmark-beta iterations
old_Dq = 1.0
LM_old_Dq = 1.0
converged = False
for iteration in range(self.settings['max_iterations'].value):
# Check if the maximum of iterations has been reached
if iteration == self.settings['max_iterations'].value - 1:
error = ('Solver did not converge in %d iterations.\n res = %e \n LM_res = %e' %
(iteration, res, LM_res))
raise exc.NotConvergedSolver(error)
# Update positions and velocities
mb.state2disp(q, dqdt, dqddt, MB_beam, MB_tstep)
MB_Asys, MB_Q = self.assembly_MB_eq_system(MB_beam,
MB_tstep,
self.data.ts,
dt,
Lambda,
Lambda_dot,
MBdict)
# Compute the correction
# ADC next line not necessary
# Dq = np.zeros((self.sys_size+num_LM_eq,), dtype=ct.c_double, order='F')
# MB_Asys_balanced, T = scipy.linalg.matrix_balance(MB_Asys)
# invT = np.matrix(T).I
# MB_Q_balanced = np.dot(invT, MB_Q).T
Dq = np.linalg.solve(MB_Asys, -MB_Q)
# least squares solver
# Dq = np.linalg.lstsq(np.dot(MB_Asys_balanced, invT), -MB_Q_balanced, rcond=None)[0]
# Evaluate convergence
if iteration:
res = np.max(np.abs(Dq[0:self.sys_size]))/old_Dq
if np.isnan(res):
raise exc.NotConvergedSolver('Multibody res = NaN')
if num_LM_eq:
LM_res = np.max(np.abs(Dq[self.sys_size:self.sys_size+num_LM_eq]))/LM_old_Dq
else:
LM_res = 0.0
if (res < self.settings['min_delta'].value) and (LM_res < self.settings['min_delta'].value):
converged = True
# Compute variables from previous values and increments
# TODO:decide If I want other way of updating lambda
# this for least sq
# q[:, np.newaxis] += Dq
# dqdt[:, np.newaxis] += self.gamma/(self.beta*dt)*Dq
# dqddt[:, np.newaxis] += 1.0/(self.beta*dt*dt)*Dq
# this for direct solver
q += Dq
dqdt += self.gamma/(self.beta*dt)*Dq
dqddt += 1.0/(self.beta*dt*dt)*Dq
if not num_LM_eq == 0:
Lambda = q[-num_LM_eq:].astype(dtype=ct.c_double, copy=True, order='F')
Lambda_dot = dqdt[-num_LM_eq:].astype(dtype=ct.c_double, copy=True, order='F')
else:
Lambda = 0
Lambda_dot = 0
if converged:
break
if not iteration:
old_Dq = np.max(np.abs(Dq[0:self.sys_size]))
if old_Dq < 1.0:
old_Dq = 1.0
if num_LM_eq:
LM_old_Dq = np.max(np.abs(Dq[self.sys_size:self.sys_size+num_LM_eq]))
else:
LM_old_Dq = 1.0
mb.state2disp(q, dqdt, dqddt, MB_beam, MB_tstep)
# end: comment time stepping
# End of Newmark-beta iterations
self.integrate_position(MB_beam, MB_tstep, dt)
# lagrangeconstraints.postprocess(self.lc_list, MB_beam, MB_tstep, MBdict, "dynamic")
lagrangeconstraints.postprocess(self.lc_list, MB_beam, MB_tstep, "dynamic")
self.compute_forces_constraints(MB_beam, MB_tstep, self.data.ts, dt, Lambda, Lambda_dot)
if self.settings['gravity_on']:
for ibody in range(len(MB_beam)):
xbeamlib.cbeam3_correct_gravity_forces(MB_beam[ibody], MB_tstep[ibody], self.settings)
mb.merge_multibody(MB_tstep, MB_beam, self.data.structure, structural_step, MBdict, dt)
# structural_step.q[:] = q[:self.sys_size].copy()
# structural_step.dqdt[:] = dqdt[:self.sys_size].copy()
# structural_step.dqddt[:] = dqddt[:self.sys_size].copy()
return self.data