{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Eulerian buckling of a beam\n", "\n", "In this numerical tour, we will compute the critical buckling load of a straight beam under normal compression, the classical Euler buckling problem. Usually, buckling is an important mode of failure for slender beams so that a standard Euler-Bernoulli beam model is sufficient. However, since FEniCS does not support Hermite elements ensuring $C^1$-formulation for the transverse deflection, implementing such models is not straightforward and requires using advanced DG formulations for instance, see the fenics-shell [implemntation of the Love-Kirchhoff plate model](http://fenics-shells.readthedocs.io/en/latest/demo/kirchhoff-love-clamped/demo_kirchhoff-love-clamped.py.html) or the [FEniCS documented demo on the biharmonic equation](http://fenics.readthedocs.io/projects/dolfin/en/2017.2.0/demos/biharmonic/python/demo_biharmonic.py.html).\n", "\n", "As a result, we will simply formulate the buckling problem using a Timoshenko beam model.\n", "\n", "## Timoshenko beam model formulation\n", "\n", "We first formulate the stiffness bilinear form of the Timoshenko model given by:\n", "\$$\n", "k((w,\\theta),(\\widehat{w},\\widehat{\\theta}))= \\int_0^L EI \\dfrac{d\\theta}{dx}\\dfrac{d\\widehat{\\theta}}{dx} dx + \\int_0^L \\kappa \\mu S \\left(\\dfrac{dw}{dx}-\\theta\\right)\\left(\\dfrac{d\\widehat{w}}{dx}-\\widehat{\\theta}\\right) dx\n", "\$$\n", "where $I=bh^3/12$ is the bending inertia for a rectangular beam of width $b$ and height $h$, $S=bh$ the cross-section area, $E$ the material Young modulus and $\\mu$ the shear modulus and $\\kappa=5/6$ the shear correction factor. We will use a $P^2/P^1$ interpolation for the mixed field $(w,\\theta)$. " ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "For issues related to shear-locking and reduced integration formulation, we refer to the :ref:ReissnerMindlinQuads tour." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "from dolfin import *\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "%matplotlib notebook\n", "\n", "L = 10.\n", "thick = Constant(0.03)\n", "width = Constant(0.01)\n", "E = Constant(70e3)\n", "nu = Constant(0.)\n", "\n", "EI = E*width*thick**3/12\n", "GS = E/2/(1+nu)*thick*width\n", "kappa = Constant(5./6.)\n", "\n", "\n", "N = 100\n", "mesh = IntervalMesh(N, 0, L) \n", "\n", "U = FiniteElement(\"CG\", mesh.ufl_cell(), 2)\n", "T = FiniteElement(\"CG\", mesh.ufl_cell(), 1)\n", "V = FunctionSpace(mesh, U*T)\n", "\n", "u_ = TestFunction(V)\n", "du = TrialFunction(V)\n", "(w_, theta_) = split(u_)\n", "(dw, dtheta) = split(du)\n", "\n", "\n", "k_form = EI*inner(grad(theta_), grad(dtheta))*dx + \\\n", " kappa*GS*dot(grad(w_)[0]-theta_, grad(dw)[0]-dtheta)*dx\n", "l_form = Constant(1.)*u_[0]*dx" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "As in the :ref:ModalAnalysis tour, a dummy linear form :code:l_form is used to call the :code:assemble_system function which retains the symmetric structure of the associated matrix when imposing boundary conditions. Here, we will consider clamped conditions on the left side :math:x=0 and simple supports on the right side :math:x=L." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Calling FFC just-in-time (JIT) compiler, this may take some time.\n", "Calling FFC just-in-time (JIT) compiler, this may take some time.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "/usr/lib/python3/dist-packages/ffc/uflacs/analysis/dependencies.py:61: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use arr[tuple(seq)] instead of arr[seq]. In the future this will be interpreted as an array index, arr[np.array(seq)], which will result either in an error or a different result.\n", " active[targets] = 1\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Calling FFC just-in-time (JIT) compiler, this may take some time.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "/usr/lib/python3/dist-packages/ffc/uflacs/analysis/dependencies.py:61: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use arr[tuple(seq)] instead of arr[seq]. In the future this will be interpreted as an array index, arr[np.array(seq)], which will result either in an error or a different result.\n", " active[targets] = 1\n" ] }, { "data": { "text/plain": [ "(,\n", " )" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def both_ends(x, on_boundary):\n", " return on_boundary\n", "def left_end(x, on_boundary):\n", " return near(x[0], 0) and on_boundary\n", "\n", "bc = [DirichletBC(V.sub(0), Constant(0.), both_ends),\n", " DirichletBC(V.sub(1), Constant(0.), left_end)]\n", "\n", "K = PETScMatrix()\n", "assemble_system(k_form, l_form, bc, A_tensor=K)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Construction of the geometric stiffness matrix\n", "\n", "The buckling analysis amounts to solving an eigenvalue problem of the form:\n", "\n", "\$$\n", "(\\mathbf{K}+\\lambda\\mathbf{K_G})\\mathbf{U} = 0\n", "\$$\n", "\n", "in which the geometric stiffness matrix $\\mathbf{K_G}$ depends (linearly) on a prestressed state, the amplitude of which is represented by $\\lambda$. The eigenvalue/eigenvector $(\\lambda,\\mathbf{U})$ solving the previous generalized eigenproblem respectively correspond to the critical buckling load and its associated buckling mode. For a beam in which the prestressed state correspond to a purely compression state of intensity $N_0>0$, the geometric stiffness bilinear form is given by:\n", "\n", "\$$\n", "k_G((w,\\theta),(\\widehat{w},\\widehat{\\theta}))= -\\int_0^L N_0 \\dfrac{dw}{dx}\\dfrac{d\\widehat{w}}{dx} dx\n", "\$$\n", "\n", "which is assembled below into the KG PETScMatrix (up to the negative sign)." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Calling FFC just-in-time (JIT) compiler, this may take some time.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "/usr/lib/python3/dist-packages/ffc/uflacs/analysis/dependencies.py:61: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use arr[tuple(seq)] instead of arr[seq]. In the future this will be interpreted as an array index, arr[np.array(seq)], which will result either in an error or a different result.\n", " active[targets] = 1\n" ] } ], "source": [ "N0 = Constant(1e-3)\n", "kg_form = N0*dot(grad(w_), grad(dw))*dx\n", "KG = PETScMatrix()\n", "assemble(kg_form, tensor=KG)\n", "for bci in bc:\n", " bci.zero(KG)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Note that we made use of the zero method of DirichletBC making the rows of the matrix associated with the boundary condition zero. If we used instead the apply method, the rows would have been replaced with a row of zeros with a 1 on the diagonal (as for the stiffness matrix K). As a result, we would have obtained an eigenvalue equal to 1 for each row with a boundary condition which can make more troublesome the computation of eigenvalues if they happen to be close to 1. Replacing with a full row of zeros in KG results in infinite eigenvalues for each boundary condition which is more suitable when looking for the lowest eigenvalues of the buckling problem.\n", "\n", "## Setting and solving the eigenvalue problem\n", "\n", "Up to the negative sign cancelling from the previous definition of KG, we now formulate the generalized eigenvalue problem $\\mathbf{KU}=-\\lambda\\mathbf{K_G U}$ using the SLEPcEigenSolver. The only difference from what has already been discussed in the dynamic modal analysis numerical tour is that buckling eigenvalue problem may be more difficult to solve than modal analysis in certain cases, it is therefore beneficial to prescribe a value of the spectral shift close to the critical buckling load." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Computing 3 first eigenvalues...\n", "Critical buckling loads:\n", "Exact: 0.31800 FE: 0.31805 Rel. gap 0.01%%\n", "Exact: 0.93995 FE: 0.94033 Rel. gap 0.04%%\n", "Exact: 1.87267 FE: 1.87415 Rel. gap 0.08%%\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "/home/bleyerj/.local/lib/python3.6/site-packages/matplotlib/font_manager.py:1241: UserWarning: findfont: Font family ['serif'] not found. Falling back to DejaVu Sans.\n", " (prop.get_family(), self.defaultFamily[fontext]))\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "