{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Stationnary non-linear heat transfer\n", "\n", "## Description of the non-linear constitutive heat transfer law\n", "\n", "The thermal material is described by the following non linear Fourier\n", "Law:\n", "\n", "$$\n", "\\mathbf{j}=-k\\left(T\\right)\\,\\mathbf{\\nabla} T\n", "$$\n", "\n", "where $\\mathbf{j}$ is the heat flux and $\\mathbf{\\nabla} T$ is the\n", "temperature gradient.\n", "\n", "Expression of the thermal conductivity\n", "--------------------------------------\n", "\n", "The thermal conductivity is assumed to be given by:\n", "\n", "$$\n", "k\\left(T\\right)={\\displaystyle \\frac{\\displaystyle 1}{\\displaystyle A+B\\,T}}\n", "$$\n", "\n", "This expression accounts for the phononic contribution to the thermal\n", "conductivity.\n", "\n", "Derivatives\n", "-----------\n", "\n", "As discussed below, the consistent linearisation of the heat transfer\n", "equilibrium requires to compute:\n", "\n", "- the derivative\n", " ${\\displaystyle \\frac{\\displaystyle \\partial \\mathbf{j}}{\\displaystyle \\partial \\mathbf{\\nabla} T}}$\n", " of the heat flux with respect to the temperature gradient.\n", " ${\\displaystyle \\frac{\\displaystyle \\partial \\mathbf{j}}{\\displaystyle \\partial \\mathbf{\\nabla} T}}$\n", " is given by: $$\n", " {\\displaystyle \\frac{\\displaystyle \\partial \\mathbf{j}}{\\displaystyle \\partial \\mathbf{\\nabla} T}}=-k\\left(T\\right)\\,\\matrix{I}\n", " $$\n", "- the derivative\n", " ${\\displaystyle \\frac{\\displaystyle \\partial \\mathbf{j}}{\\displaystyle \\partial T}}$\n", " of the heat flux with respect to the temperature.\n", " ${\\displaystyle \\frac{\\displaystyle \\partial \\mathbf{j}}{\\displaystyle \\partial T}}$\n", " is given by: $$\n", " {\\displaystyle \\frac{\\displaystyle \\partial \\mathbf{j}}{\\displaystyle \\partial T}}=-{\\displaystyle \\frac{\\displaystyle \\partial k\\left(T\\right)}{\\displaystyle \\partial T}}\\,\\mathbf{\\nabla} T=B\\,k^{2}\\,\\mathbf{\\nabla} T\n", " $$\n", "\n", "`MFront`’ implementation\n", "========================\n", "\n", "Choice of the the domain specific language\n", "------------------------------------------\n", "\n", "Every `MFront` file is handled by a domain specific language (DSL), which\n", "aims at providing the most suitable abstraction for a particular choice\n", "of behaviour and integration algorithm. See `mfront mfront --list-dsl`\n", "for a list of the available DSLs.\n", "\n", "The name of DSL’s handling generic behaviours ends with\n", "`GenericBehaviour`. The first part of a DSL’s name is related to the\n", "integration algorithm used.\n", "\n", "In the case of this non linear transfer behaviour, the heat flux is\n", "explicitly computed from the temperature and the temperature gradient.\n", "The `DefaultGenericBehaviour` is the most suitable choice:\n", "\n", "``` cxx\n", "@DSL DefaultGenericBehaviour;\n", "```\n", "\n", "Some metadata\n", "-------------\n", "\n", "The following lines define the name of the behaviour, the name of the\n", "author and the date of its writing:\n", "\n", "``` cxx\n", "@Behaviour StationaryHeatTransfer;\n", "@Author Thomas Helfer;\n", "@Date 15/02/2019;\n", "```\n", "\n", "Gradients and fluxes\n", "--------------------\n", "\n", "Generic behaviours relate pairs of gradients and fluxes. Gradients and\n", "fluxes are declared independently but the first declared gradient is\n", "assumed to be conjugated with the first declared fluxes and so on…\n", "\n", "The temperature gradient is declared as follows (note that Unicode characters are supported):\n", "\n", "``` cxx\n", "@Gradient TemperatureGradient ∇T;\n", "∇T.setGlossaryName(\"TemperatureGradient\");\n", "```\n", "\n", "Note that we associated to `∇T` the glossary name `TemperatureGradient`.\n", "This is helpful for the calling code.\n", "\n", "After this declaration, the following variables will be defined:\n", "\n", "- The temperature gradient `∇T` at the beginning of the time step.\n", "- The increment of the temperature gradient `Δ∇T` over the time step.\n", "\n", "The heat flux is then declared as follows:\n", "\n", "``` cxx\n", "@Flux HeatFlux j;\n", "j.setGlossaryName(\"HeatFlux\");\n", "```\n", "\n", "In the following code blocks, `j` will be the heat flux at the end of\n", "the time step.\n", "\n", "Tangent operator blocks\n", "-----------------------\n", "\n", "By default, the derivatives of the gradients with respect to the fluxes\n", "are declared. Thus the variable `∂j∕∂Δ∇T` is automatically declared.\n", "\n", "However, as discussed in the next section, the consistent linearisation\n", "of the thermal equilibrium requires to return the derivate of the heat\n", "flux with respect to the increment of the temperature (or equivalently\n", "with respect to the temperature at the end of the time step).\n", "\n", "``` cxx\n", "@AdditionalTangentOperatorBlock ∂j∕∂ΔT;\n", "```\n", "\n", "Parameters\n", "----------\n", "\n", "The `A` and `B` coefficients that appears in the definition of the\n", "thermal conductivity are declared as parameters:\n", "\n", "``` cxx\n", "@Parameter real A = 0.0375;\n", "@Parameter real B = 2.165e-4;\n", "```\n", "\n", "Parameters are stored globally and can be modified from the calling\n", "solver or from `python` in the case of the coupling with `FEniCS`\n", "discussed below.\n", "\n", "Local variable\n", "--------------\n", "\n", "A local variable is accessible in each code blocks.\n", "\n", "Here, we declare the thermal conductivity `k` as a local variable in\n", "order to be able to compute its value during the behaviour integration\n", "and to reuse this value when computing the tangent operator.\n", "\n", "``` cxx\n", "@LocalVariable thermalconductivity k;\n", "```\n", "\n", "Integration of the behaviour\n", "----------------------------\n", "\n", "The behaviour integration is straightforward: one starts to compute the\n", "temperature at the end of the time step, then we compute the thermal\n", "conductivity (at the end of the time step) and the heat flux using the\n", "temperature gradient (at the end of the time step).\n", "\n", "``` cxx\n", "@Integrator{\n", " // temperature at the end of the time step\n", " const auto T_ = T + ΔT;\n", " // thermal conductivity\n", " k = 1 / (A + B ⋅ T_);\n", " // heat flux\n", " j = -k ⋅ (∇T + Δ∇T);\n", "} // end of @Integrator\n", "```\n", "\n", "Tangent operator\n", "----------------\n", "\n", "The computation of the tangent operator blocks is equally simple:\n", "\n", "``` cxx\n", "@TangentOperator {\n", " ∂j∕∂Δ∇T = -k ⋅ tmatrix::Id();\n", " ∂j∕∂ΔT = B ⋅ k ⋅ k ⋅ (∇T + Δ∇T);\n", "} // end of @TangentOperator \n", "```\n", "## FEniCS implementation\n", "\n", "We consider a rectanglar domain with imposed temperatures `Tl` (resp. `Tr`) on the left (resp. right boundaries). We want to solve for the temperature field `T` inside the domain using a $P^1$-interpolation. We initialize the temperature at value `Tl` throughout the domain. We finally load the material library with a `plane_strain` hypothesis." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import matplotlib.pyplot as plt\n", "from dolfin import *\n", "import mfront_wrapper as mf\n", "\n", "length = 30e-3\n", "width = 5.4e-3\n", "mesh = RectangleMesh(Point(0., 0.), Point(length, width), 100, 10)\n", "\n", "V = FunctionSpace(mesh, \"CG\", 1)\n", "T = Function(V, name=\"Temperature\")\n", "\n", "def left(x, on_boundary):\n", " return near(x[0], 0) and on_boundary\n", "def right(x, on_boundary):\n", " return near(x[0], length) and on_boundary\n", "\n", "Tl = 300\n", "Tr = 800\n", "T.interpolate(Constant(Tl))\n", "\n", "bc = [DirichletBC(V, Constant(Tl), left),\n", " DirichletBC(V, Constant(Tr), right)]\n", "\n", "material = mf.MFrontNonlinearMaterial(\"../materials/src/libBehaviour.so\",\n", " \"StationaryHeatTransfer\",\n", " hypothesis=\"plane_strain\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The MFront behaviour implicitly declares the temperature as an external state variable called `\"Temperature\"`. We must therefore associate this external state variable to a known mechanical field. This can be achieved explicitly using the `register_external_state_variable` method. In the present case, this can be donc automatically since the name of the unknown temperature field matches the [TFEL Glossary](http://tfel.sourceforge.net/glossary.html) name `\"Temperature\"`. In this case, the following message is printed:\n", "```\n", "Automatic registration of 'Temperature' as an external state variable.\n", "```\n", "For problems in which the temperature only acts as a parameter (no jacobian blocks with respect to the temperature), the temperature can be automatically registered as a constant value ($293.15 \\text{ K}$ by default) or to any other (`dolfin.Constant` or `float`) value using the `register_external_state_variable` method.\n", "\n", "In the FEniCS interface, we instantiate the main mechanical unknown, here the temperature field `T` which has to be named `\"Temperature\"` in order to match MFront's predefined name. Using another name than this will later result in an error saying:\n", "```\n", "ValueError: 'Temperature' could not be associated with a registered gradient or a known state variable.\n", "```\n", "\n", "The MFront behaviour declares the field `\"TemperatureGradient\"` as a Gradient variable, with its associated Flux called `\"HeatFlux\"`. We can check that the `material` object retrieves MFront's gradient and flux names, as well as the different tangent operator blocks which have been defined, namely `dj_ddgT` and `dj_ddT` in the present case:" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "['TemperatureGradient']\n", "['HeatFlux']\n", "['dHeatFlux_dTemperatureGradient', 'dHeatFlux_dTemperature']\n" ] } ], "source": [ "print(material.get_gradient_names())\n", "print(material.get_flux_names())\n", "print([\"d{}_d{}\".format(*t) for t in material.get_tangent_block_names()])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "When defining the non-linear problem, we will specify the boundary conditions and the requested quadrature degree which will control the number of quadrature points used in each cell to compute the non-linear constitutive law. Here, we specify a quadrature of degree 2 (i.e. 3 Gauss points for a triangular element). Finally, we need to associate to MFront gradient object the corresponding UFL expression as a function of the unknown field `T`. To do so, we use the `register_gradient` method linking MFront `\"TemperatureGradient\"` object to the UFL expression `grad(T)`. Doing so, the corresponding non-linear variational problem will be automatically be built:\n", "\n", "\\begin{equation}\n", "F(\\widehat{T}) = \\int_\\Omega \\boldsymbol{j}\\cdot \\nabla \\widehat{T} \\text{dx} = 0 \\quad \\forall \\widehat{T}\n", "\\end{equation}" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "problem = mf.MFrontNonlinearProblem(T, material, quadrature_degree=2, bcs=bc)\n", "problem.register_gradient(\"TemperatureGradient\", grad(T))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "From the two tangent operator blocks `dj_ddgT` and `dj_ddT`, it will automatically be deduced that the heat flux $\\boldsymbol{j}$ is a function of both the temperature gradient $\\boldsymbol{g}=\\nabla T$ and the temperature itself i.e. $\\boldsymbol{j}=\\boldsymbol{j}(\\boldsymbol{g}, T)$. The following tangent bilinear form will therefore be used when solving the above non-linear problem:\n", "\n", "\\begin{equation}\n", "J(\\widehat{T},T^*) = \\int_{\\Omega} \\nabla \\widehat{T}\\cdot\\left(\\dfrac{\\partial \\boldsymbol{j}}{\\partial \\boldsymbol{g}}\\cdot \\nabla T^*+\\dfrac{\\partial \\boldsymbol{j}}{\\partial T}\\cdot T^*\\right) \\text{dx}\n", "\\end{equation}\n", "\n", "Similarly to the case of external state variables, common gradient expressions for some [TFEL Glossary](http://tfel.sourceforge.net/glossary.html) names have been already predefined which avoid calling explicitly the `register_gradient` method. Predefined expressions can be obtained from:" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "'TFEL gradient name' (Available hypotheses)\n", "---------------------------------------------\n", "'Strain' ('Tridimensional', 'PlaneStrain', 'Axisymmetrical')\n", "'TemperatureGradient' ('Tridimensional', 'PlaneStrain', 'Axisymmetrical')\n", "'DeformationGradient' ('Tridimensional', 'PlaneStrain', 'Axisymmetrical')\n", "\n" ] } ], "source": [ "mf.list_predefined_gradients()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can see that the name `\"Temperature Gradient\"` is in fact a predefined gradient. Omitting calling the `register_gradient` method will in this case print the following message upon calling `solve`:\n", "```\n", "Automatic registration of 'TemperatureGradient' as grad(Temperature).\n", "```\n", "meaning that a predefined gradient name has been found and registered as the UFL expression $\\nabla T$.\n", "\n", "We finally solve the non-linear problem using a default Newton non-linear solver. The `solve` method returns the number of Newton iterations (4 in the present case) and converged status ." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Automatic registration of 'Temperature' as an external state variable.\n", "\n" ] }, { "data": { "text/plain": [ "(4, True)" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "problem.solve(T.vector())" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We finally check that the thermal conductivity coefficient $k$, computed from the ratio between the horizontal heat flux and temperature gradient matches the temperature-dependent expressions implemented in the MFront behaviour." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "j = problem.fluxes[\"HeatFlux\"].function\n", "g = problem.gradients[\"TemperatureGradient\"].function\n", "k_gauss = -j.vector().get_local()[::2]/g.vector().get_local()[::2]\n", "T_gauss = problem.state_variables[\"external\"][\"Temperature\"].function.vector().get_local()\n", "A = material.get_parameter(\"A\");\n", "B = material.get_parameter(\"B\");\n", "k_ref = 1/(A + B*T_gauss)\n", "plt.plot(T_gauss, k_gauss, 'o', label=\"FE\")\n", "plt.plot(T_gauss, k_ref, '.', label=\"ref\")\n", "plt.xlabel(r\"Temperature $T\\: (K)$\")\n", "plt.ylabel(r\"Thermal conductivity $k\\: (W.m^{-1}.K^{-1})$\")\n", "plt.legend()\n", "plt.show()" ] }, { "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": "3.6.9" } }, "nbformat": 4, "nbformat_minor": 4 }