"""A simple wave equation.
.. codeauthor:: David Zwicker <david.zwicker@ds.mpg.de>
"""
from __future__ import annotations
from typing import TYPE_CHECKING
import numpy as np
from ..fields import FieldCollection, ScalarField
from ..grids.boundaries import set_default_bc
from ..tools.docstrings import fill_in_docstring
from .base import PDEBase, expr_prod
if TYPE_CHECKING:
from collections.abc import Callable
import torch
from ..grids.boundaries.axes import BoundariesData
from ..tools.typing import NumericArray
[docs]
class WavePDE(PDEBase):
r"""A simple wave equation.
The mathematical definition, :math:`\partial_t^2 u = c^2 \nabla^2 u`, is implemented
as two first-order equations,
.. math::
\partial_t u &= v \\
\partial_t v &= c^2 \nabla^2 u
where :math:`c` sets the wave speed and :math:`v` is an auxiallary field. Note that
the class expects an initial condition specifying both fields, which can be created
using the :meth:`WavePDE.get_initial_condition` method. The result will also return
two fields.
"""
explicit_time_dependence = False
default_bc = "auto_periodic_neumann"
"""Default boundary condition used when no specific conditions are chosen."""
@fill_in_docstring
def __init__(self, speed: float = 1, *, bc: BoundariesData | None = None):
"""
Args:
speed (float):
The speed :math:`c` of the wave
bc:
The boundary conditions applied to the field :math:`u`.
{ARG_BOUNDARIES}
"""
super().__init__()
self.speed = speed
self.bc = set_default_bc(bc, self.default_bc)
[docs]
def get_initial_condition(self, u: ScalarField, v: ScalarField | None = None):
"""Create a suitable initial condition.
Args:
u (:class:`~pde.fields.ScalarField`):
The initial density on the grid
v (:class:`~pde.fields.ScalarField`, optional):
The initial rate of change. This is assumed to be zero if the
value is omitted.
Returns:
:class:`~pde.fields.FieldCollection`:
The combined fields u and v, suitable for the simulation
"""
if v is None:
v = ScalarField(u.grid)
return FieldCollection([u, v], labels=["u", "v"])
@property
def expressions(self) -> dict[str, str]:
"""dict: the expressions of the right hand side of this PDE"""
return {"u": "v", "v": expr_prod(self.speed**2, "∇²u")}
[docs]
def evolution_rate( # type: ignore
self,
state: FieldCollection,
t: float = 0,
) -> FieldCollection:
"""Evaluate the right hand side of the PDE.
Args:
state (:class:`~pde.fields.FieldCollection`):
The fields :math:`u` and :math:`v`
t (float):
The current time point
Returns:
:class:`~pde.fields.FieldCollection`:
Fields describing the evolution rates of the PDE
"""
if not isinstance(state, FieldCollection):
msg = "`state` must be FieldCollection"
raise TypeError(msg)
if len(state) != 2:
msg = "`state` must contain two fields"
raise ValueError(msg)
u, v = state
u_t = v.copy()
v_t = self.speed**2 * u.laplace(self.bc, args={"t": t}) # type: ignore
return FieldCollection([u_t, v_t])
[docs]
def make_pde_rhs_numba(
self, state: FieldCollection
) -> Callable[[NumericArray, float], NumericArray]:
"""Create a compiled function evaluating the right hand side of the PDE.
Args:
state (:class:`~pde.fields.FieldCollection`):
An example for the state defining the grid and data types
Returns:
A function with signature `(state_data, t)`, which can be called with an
instance of :class:`~numpy.ndarray` of the state data and the time to
obtained an instance of :class:`~numpy.ndarray` giving the evolution rate.
"""
speed2 = self.speed**2
laplace = state.grid.make_operator("laplace", bc=self.bc, backend="numba")
def pde_rhs(state_data: NumericArray, t: float = 0) -> NumericArray:
"""Compiled helper function evaluating right hand side."""
rate = np.empty_like(state_data)
rate[0] = state_data[1]
rate[1][:] = laplace(state_data[0], args={"t": t})
rate[1] *= speed2
return rate
return pde_rhs
[docs]
def make_pde_rhs_torch(
self, state: ScalarField
) -> Callable[[torch.Tensor, float], torch.Tensor]:
"""Create a compiled function evaluating the right hand side of the PDE.
Args:
state (:class:`~pde.fields.ScalarField`):
An example for the state defining the grid and data types
Returns:
A function with signature `(state_data, t)`, which can be called
with an instance of :class:`torch.Tensor` of the state data and
the time to obtained an instance of :class:`torch.Tensor` giving
the evolution rate.
"""
import torch
speed2 = self.speed**2
laplace = state.grid.make_operator(
operator="laplace",
bc=self.bc,
backend="torch",
native=True,
dtype=state.dtype,
)
def pde_rhs(state_data: torch.Tensor, t: float = 0) -> torch.Tensor:
"""Compiled helper function evaluating right hand side."""
return torch.stack(
(state_data[1], speed2 * laplace(state_data[0], args={"t": t}))
)
return pde_rhs