Dynamic3D

<thermal solver="Dynamic3D">

Corresponding Python class: thermal.dynamic.Dynamic3D.

Attributes:
  • name (required) – Solver name.
Contents:
<geometry>

Geometry for use by this solver.

Attributes:
  • ref (required) – Name of a Cartesian3D geometry defined in the <geometry> section.
<mesh>

Rectangular3D mesh used by this solver.

Attributes:
  • ref (required) – Name of a Rectangular3D mesh defined in the <grids> section.
<temperature>

Temperature boundary conditions. See subsection Boundary conditions.

<loop>

Configuration of the time-evolution loop.

Attributes:
  • inittemp – Initial temperature used for the first computation. (float (K), default 300 K)
  • timestep – Single-iteration time step. (float (ns), default 0.1 ns)
  • rebuildfreq – Number of iterations until the whole matrix is rebuilt. The larger this number is, the more efficient computations are, however it may be less accurate is material parameters strongly depend on temperature. If this parameter is set to zero, matrix is never rebuilt. (int, default 0)
  • logfreq – Number of iterations until the computations progress is reported. (int, default 500)
<matrix>

Matrix solver configuration.

Attributes:
  • algorithm – Algorithm used for solving set of linear positive-definite equations. (cholesky, gauss, or iterative, default is iterative)
  • methodparam – Mid-step parameter for implicit finite-difference time discretization. Defaults to ½, which results in the Crank-Nicholson method. 0 makes the method explicit, while 1 results in backward Euler method. (float, default 0.5)
  • lumping – This attribute determines whether the mass matrix is lumped or non-lumped (consistent). (bool, default is yes)
<iterative>

Parameters for iterative matrix solver. PLaSK uses NSPCG package for performing iterations. Please refer to its documentation for explanation of most of the settings.

Attributes:
  • maxit – Maximum number of iterations. (int, default 1000)
  • maxerr – Maximum iteration error. (float, default 1e-6)
  • noconv – Desired behavior if the iterative solver does not converge. (error, warning, or continue, default is warning)
  • accelerator – Accelerator used for iterative matrix solver. (cg, si, sor, srcg, srsi, basic, me, cgnr, lsqr, odir, omin, ores, iom, gmres, usymlq, usymqr, landir, lanmin, lanres, cgcr, or bcgs, default is cg)
  • preconditioner – Preconditioner used for iterative matrix solver. (rich, jac, ljac, ljacx, sor, ssor, ic, mic, lsp, neu, lsor, lssor, llsp, lneu, bic, bicx, mbic, or mbicx, default is ic)
  • nfact – This number initializes the frequency of partial factorizations. It specifies the number of linear system evaluations between factorizations. The default value is 1, which means that a factorization is performed at every iteration. (int, default 10)
  • ndeg – Degree of the polynomial to be used for the polynomial preconditioners. (int, default 1)
  • lvfill – Level of fill-in for incomplete Cholesky preconditioners. Increasing this value will result in more accurate factorizations at the expense of increased memory usage and factorization time. (int, default 0)
  • ltrunc – Truncation bandwidth to be used when approximating the inverses of matrices with dense banded matrices. An increase in this value means a more accurate factorization at the expense of increased storage. (int, default 0)
  • omega – Relaxation parameter. (float, default 1.0)
  • nsave – The number of old vectors to be saved for the truncated acceleration methods. (int, default 5)
  • nrestart – The number of iterations between restarts for the restarted acceleration methods. (int, default 100000)
Preconditioner choices:
rich Richardson’s method
jac Jacobi method
ljac Line Jacobi method
ljacx Line Jacobi method (approx. inverse)
sor Successive Overrelaxation
ssor Symmetric SOR (can be used only with SOR accelerator)
ic Incomplete Cholesky (default)
mic Modified Incomplete Cholesky
lsp Least Squares Polynomial
neu Neumann Polynomial
lsor Line SOR
lssor Line SSOR
llsp Line Least Squares Polynomial
lneu Line Neumann Polynomial
bic Block Incomplete Cholesky (ver. 1)
bicx Block Incomplete Cholesky (ver. 2)
mbic Modified Block Incomplete Cholesky (ver. 1)
mbicx Modified Block Incomplete Cholesky (ver. 2)
Accelerator choices:
cg Conjugate Gradient acceleration (default)
si Chebyshev acceleration or Semi-Iteration
sor Successive Overrelaxation (can use only SOR preconditioner)
srcg Symmetric Successive Overrelaxation Conjugate Gradient Algorithm (can use only SSOR preconditioner)
srsi Symmetric Successive Overrelaxation Semi-Iteration Algorithm (can use only SSOR preconditioner)
basic Basic Iterative Method
me Minimal Error Algorithm
cgnr Conjugate Gradient applied to the Normal Equations
lsqr Least Squares Algorithm
odir ORTHODIR, a truncated/restarted method useful for nonsymmetric systems of equations
omin ORTHOMIN, a common truncated/restarted method used for nonsymmetric systems
ores ORTHORES, another truncated/restarted method for nonsymmetric systems
iom Incomplete Orthogonalization Method
gmres Generalized Minimal Residual Method
usymlq Unsymmetric LQ
usymqr Unsymmetric QR
landir Lanczos/ORTHODIR
lanmin Lanczos/ORTHOMIN or Biconjugate Gradient Method
lanres Lanczos/ORTHORES or “two-sided” Lanczos Method
cgcr Constrained Generalized Conjugate Residual Method
bcgs Biconjugate Gradient Squared Method