notebook_adv_diff_Hughes.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nutils import *\n",
    "import numpy, unittest\n",
    "\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import collections"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "nelems = 2\n",
    "degree = 1\n",
    "tol  = 1e-5\n",
    "ndims = 1\n",
    "extent = 1\n",
    "v = 0    # convective velocity\n",
    "k = 1      # diffusivity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "domain, geom = mesh.rectilinear([numpy.linspace(0,extent,nelems+1)]*ndims)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# prepare residual\n",
    "ns = function.Namespace()\n",
    "ns.pi = numpy.pi\n",
    "ns.x = geom\n",
    "ns.basis = domain.basis('lagrange', degree=degree)\n",
    "ns.u = 'basis_n ?lhs_n'\n",
    "ns.k = k\n",
    "ns.v = v\n",
    "h = extent / nelems\n",
    "# Put element length in namespace\n",
    "ns.h = h\n",
    "\n",
    "# Tuning parameters\n",
    "ns.beta = 1\n",
    "ns.gamma = 10\n",
    "\n",
    "# DG Stabilization parameters (all the same for an equidistant mesh)\n",
    "ns.tauZero = ' gamma k / h'\n",
    "ns.tauN    = ' gamma k / h'\n",
    "ns.tauA    = ' gamma k / h'\n",
    "\n",
    "# SUPG stabilization parameter\n",
    "if(numpy.exp(v/k) == float('inf')):\n",
    "    ns.tau = ' h / (2 v)'\n",
    "else:\n",
    "    ns.peclet = ' 0.5 v h / k'\n",
    "    ns.tau = ' ((cosh(peclet) / sinh(peclet)) - (1 / peclet)) h / (2 v) '\n",
    "    \n",
    "# Advection Upwinding coefficient\n",
    "ns.C = ' v '   \n",
    "ns.ext = extent\n",
    "\n",
    "# BCs and analytical solution\n",
    "#ns.z = 2\n",
    "#ns.gl  = 'cos( z pi x_0 )'\n",
    "#ns.gr  = 'cos( z pi x_0 )'\n",
    "#ns.dgl = '-(z pi) sin( z pi x_0 )'\n",
    "#ns.dgr = '-(z pi) sin( z pi x_0 )'\n",
    "#ns.ddg = '-(z pi)^2 cos( z pi x_0 )'\n",
    "\n",
    "#ns.f   = ' - ddg'\n",
    "\n",
    "# u(x) = [(gr-gl)/(exp(v/k) - 1)] [exp((v/k) x) - 1] + gl\n",
    "# u'(x)= (v/k)[(gr-gl)/(exp(v/k) - 1] [exp((v/k) x)]\n",
    "#   with u(0) = gl, u(extent) = gr \n",
    "ns.f = 0\n",
    "ns.gl = 1\n",
    "ns.gr = 0\n",
    "# Catch potential overflow exceptions\n",
    "if(numpy.exp(v/k) == float('inf')):\n",
    "    # exp(v/k) -> infinity (triggers overflow exception)\n",
    "    # Find expression for ns.g\n",
    "    pass\n",
    "else:\n",
    "    ns.analytical = '( (gr - gl) / (exp(v / k) - 1) ) (exp( (v / k) x_0) - 1) + gl'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# construct residual\n",
    "# Element integrals\n",
    "# Diffusive part and RHS\n",
    "res = domain.integral('k basis_n,0 u_,0 - basis_n f' @ ns, geometry=ns.x, degree=degree*2)\n",
    "# Advective part\n",
    "res += domain.integral('-v basis_n,0 u' @ ns, geometry=ns.x, degree=5)\n",
    "# SUPG stabilized part\n",
    "res += domain.integral(' tau v basis_n,0 (v u_,0 - k (u_,0)_,0 - f )' @ ns, geometry=ns.x, degree=5)\n",
    "\n",
    "# Interface integrals\n",
    "# Diffusive part\n",
    "#    part one                      |\n",
    "#                                  v  minus might be necessary here because of jump definitions (likewise on second +)\n",
    "res += domain.interfaces.integral(' [basis_n] n_0 k {u_,0} + beta k {basis_n,0} [u] n_0' @ ns, geometry=ns.x, degree=degree*2)\n",
    "#    part two (stabilized)\n",
    "res += domain.interfaces.integral(' tauA [basis_n] n_0 [u] n_0' @ ns, geometry=ns.x, degree=degree*2)\n",
    "# Advective part\n",
    "res += domain.interfaces.integral('- v [basis_n] n_0 ({u} - .5 C [u] n_0)' @ ns, geometry=ns.x, degree=5)\n",
    "\n",
    "# Boundary integrals LHS\n",
    "#    part one\n",
    "res += domain.boundary.integral('- k basis_n u_,0 + beta k basis_n,0 u ' @ ns, geometry=ns.x, degree=degree*2)\n",
    "#    part two (stabilized)\n",
    "res += domain.boundary['left'].integral(' tauZero basis_n u ' @ ns, geometry=ns.x, degree=degree*2)\n",
    "res += domain.boundary['right'].integral(' tauN basis_n u ' @ ns, geometry=ns.x, degree=degree*2)\n",
    "\n",
    "# Boundary integrals RHS (weakly imposed boundary conditions) (including stabilization)\n",
    "# Diffusive part\n",
    "res -= domain.boundary['left'].integral(' ( beta k basis_n,0 + tauZero basis_n) gl ' @ ns, geometry=ns.x, degree=degree*2)\n",
    "res -= domain.boundary['right'].integral(' ( beta k basis_n,0 + tauN basis_n) gr ' @ ns, geometry=ns.x, degree=degree*2)\n",
    "# Advective part\n",
    "res += domain.boundary['left'].integral('v basis_n n_0 (0.5 (gl + u) - .5 C (gl - u) n_0)' @ ns, geometry=ns.x, degree=degree*2)\n",
    "res += domain.boundary['right'].integral('v basis_n n_0 (0.5 (gr + u) - .5 C (gr - u) n_0)' @ ns, geometry=ns.x, degree=degree*2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Here comes the Jacobian\n",
      "(3, 3)\n",
      "[[  2.20000000e+01  -4.00000000e+00   0.00000000e+00]\n",
      " [  4.44089210e-16   4.00000000e+00  -4.00000000e+00]\n",
      " [  0.00000000e+00   4.44089210e-16   2.20000000e+01]]\n",
      "[-18.  -2.   0.]\n",
      "solve > solver returned with residual 0e+00\n"
     ]
    }
   ],
   "source": [
    "# find lhs such that res == 0 and substitute this lhs in the namespace\n",
    "# lhs = solver.solve_linear('lhs', res, constrain=cons)\n",
    "lhs = solver.solve_linear('lhs', res)\n",
    "ns = ns(lhs=lhs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f711a49dd30>]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1280 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot solution\n",
    "plt.figure()\n",
    "bezier = domain.sample('bezier', 9)\n",
    "x, u = bezier.eval([ns.x, ns.u])\n",
    "#x, analytical = bezier.eval([ns.x, ns.analytical])\n",
    "fig=plt.figure(figsize=(18, 16), dpi= 80, facecolor='w', edgecolor='k')\n",
    "plt.plot(x, u)\n",
    "#plt.plot(x, analytical, 'r')\n",
    "# plt.ylim(-2,2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}