Scribd Logo
Upload

Welcome to Scribd!

  • 43 views

    One Dimensional Boundary Value Problem: Chapter Three

    Uploaded by

    Mike Junior
    Instruction manual that walks the user through the Finite Element method using the programming language MATLAB. Can be used by students and practicing engineers alike to analyze complex structures using FEM.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    One Dimensional Boundary Value Problem: Chapter Three

    Uploaded by

    Mike Junior
    43 views6 pages
    Instruction manual that walks the user through the Finite Element method using the programming language MATLAB. Can be used by students and practicing engineers alike to analyze complex structures using FEM.

    Original Title

    00Chap3-MatlabUse

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Instruction manual that walks the user through the Finite Element method using the programming language MATLAB. Can be used by students and practicing engineers alike to analyze complex structures using FEM.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    43 views6 pages

    One Dimensional Boundary Value Problem: Chapter Three

    Uploaded by

    Mike Junior
    Instruction manual that walks the user through the Finite Element method using the programming language MATLAB. Can be used by students and practicing engineers alike to analyze complex structures using FEM.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 6

    CHAPTER THREE

    One Dimensional
    Boundary Value Problem

    Matlab functions for solution of one dimensional boundary value problem


    For Matlab solution of one dimensional boundary value problems using linear and quadratic element, the
    following functions can be defined and used in the manner shown in Chapter 2 for axial deformation
    problem.

    MatlabFiles\Chap3\BVP1DLinElement.m
    function [ke, re] = BVP1DLinElement(k, p, q, coord)
    % [ke, re] = BVP1DLinElement(k, p, q, coord)
    % Generates equations for a linear element for 1D BVP
    % k,p,q = parameters defining the BVP
    % coord = coordinates at the element ends
    L=coord(2)-coord(1);

    ke = [k/L - (L*p)/3, -(k/L) - (L*p)/6;


    -(k/L) - (L*p)/6, k/L - (L*p)/3];
    re = [(L*q)/2; (L*q)/2];

    MatlabFiles\Chap3\BVP1DQuadElement.m
    function [ke, re] = BVP1DQuadElement(k, p, q, coord)
    % [ke, re] = BVP1DQuadElement(k, p, q, coord)
    % Generates equations for a quadratic element for 1D BVP
    % k,p,q = parameters defining the BVP
    % coord = coordinates at the element ends
    L=coord(3)-coord(1);
    ke = [(7*k)/(3*L) - (2*L*p)/15, (-8*k)/(3*L) - (L*p)/15, ...
    k/(3*L) + (L*p)/30;
    (-8*k)/(3*L) - (L*p)/15, (16*k)/(3*L) - (8*L*p)/15, ...
    (-8*k)/(3*L) - (L*p)/15;
    k/(3*L) + (L*p)/30, (-8*k)/(3*L) - (L*p)/15, ...
    (7*k)/(3*L) - (2*L*p)/15];
    re = [(L*q)/6; (2*L*q)/3; (L*q)/6];

    Several examples in the following sections illustrate the use of these functions.

    Computer Implementation 3.1 (Matlab) Solution using Matlab functions for 1D BVP
    The tedious calculations to solve the previous heat flow problem can be conveniently carried out using
    Matlab functions presented earlier. Using these functions, and following procedures discussed in Chapter
    1, the global equations for the four quadratic element model can be developed and assembled as follows.

    MatlabFiles\Chap3\FinHeatFlowEx.m
    % Heat flow through a fin
    kx=237;w=3;t=0.3/100;A=w*t;h=30;L=20/100.;P=2*(w+t);Tinf=25;
    alpha=-h*A/(kx*A); beta=h*A*Tinf/(kx*A);
    k=kx*A; p=-h*P; q=h*P*Tinf;
    nodes = [0:L/8:L];n=length(nodes);
    K=zeros(n); R = zeros(n,1);
    % Generate equations for each element and assemble them.
    for i=1:4
    lm=[2*(i-1)+1,2*(i-1)+2,2*(i-1)+3];
    [ke, re] = BVP1DQuadElement(k,p,q, nodes(lm));
    K(lm, lm) = K(lm, lm) + ke;
    R(lm) = R(lm) + re;
    end
    % Adjust for NBC

    K(n,n)=K(n,n)-alpha*k
    R(n)=R(n)+beta*k
    % Nodal solution and reactions
    d = NodalSoln(K, R, [1], [100])
    plot(nodes,d),title('Temperature distribution'), xlabel('x'),ylabel('T')
    >> FinHeatFlowEx
    K=
    Columns 1 through 7
    100.7412 -113.1594 13.9197
    0
    0
    0
    0
    -113.1594 232.3248 -113.1594
    0
    0
    0
    0
    13.9197 -113.1594 201.4824 -113.1594 13.9197
    0
    0
    0
    0 -113.1594 232.3248 -113.1594
    0
    0
    0
    0 13.9197 -113.1594 201.4824 -113.1594 13.9197
    0
    0
    0
    0 -113.1594 232.3248 -113.1594
    0
    0
    0
    0 13.9197 -113.1594 201.4824
    0
    0
    0
    0
    0
    0 -113.1594
    0
    0
    0
    0
    0
    0 13.9197
    Columns 8 through 9
    0
    0
    0
    0
    0
    0
    0
    0
    0
    0
    0
    0
    -113.1594 13.9197
    232.3248 -113.1594
    -113.1594 101.0112

    R=
    37.5375
    150.1500
    75.0750
    150.1500
    75.0750
    150.1500
    75.0750
    150.1500
    44.2875

    d=
    100.0000
    85.4383
    74.0844
    65.3306
    58.7175
    53.8905
    50.5968
    48.6592
    47.9772

    T
    100
    90
    80
    70
    60
    x
    0.05

    0.1

    0.15

    0.2

    Computer Implementation 3.2 (Matlab) Solution of Buckling Problem


    The buckling problem can be implemented easily in Matlab by defining simple functions returning element
    k and kp matrices as follows.

    MatlabFiles\Chap3\BucklingLinElement.m
    function [ke, kp] = BucklingLinElement(k, coord)
    % [ke, kp] = BucklingLinElement(k, coord)
    % Generates equations for a linear element for 1D Buckling
    % k = bar stiffness (EI)
    % coord = coordinates at the element ends
    L=coord(2)-coord(1);
    ke = k/L*[1, -1; -1, 1];
    kp = [L/3, L/6; L/6, L/3];

    MatlabFiles\Chap3\BucklingQuadElement.m
    function [ke, kp] = BucklingQuadElement(k, coord)
    % [ke, kp] = BucklingQuadElement(k, coord)
    % Generates equations for a quadratic element for 1D Buckling
    % k = bar stiffness (EI)
    % coord = coordinates at the element ends
    L=coord(3)-coord(1);
    ke = [(7*k)/(3*L), -((8*k)/(3*L)), k/(3*L);
    -((8*k)/(3*L)), (16*k)/(3*L), -((8*k)/(3*L));
    k/(3*L), -((8*k)/(3*L)), (7*k)/(3*L)];
    kp = [((2*L)/15), (L/15), -L/30;
    (L/15), ((8*L)/15), (L/15);
    -L/30, (L/15), ((2*L)/15)];

    Using the BucklingQuadElement function, a solution using 4 quadratic is obtained as follows.

    MatlabFiles\Chap3\BucklingEx.m
    % Solution of Euler buckling using quadratic elements
    L = 12*10.; EI = 10^6;
    nodes = [0:L/8:L];n=length(nodes);
    Ke=zeros(n); Kp=zeros(n);
    % Generate equations for each element and assemble them.
    for i=1:4
    lm=[2*(i-1)+1,2*(i-1)+2,2*(i-1)+3];
    [ke, kp] = BucklingQuadElement(EI, nodes(lm));
    Ke(lm, lm) = Ke(lm, lm) + ke;
    Kp(lm, lm) = Kp(lm, lm) + kp;
    end
    % Adjust for EBC
    debc=[1,n];
    df = setdiff(1:n, debc);
    Kef = Ke(df, df)
    Kep = Kp(df, df)
    [v,e] = eig(Kef, Kep);
    fprintf('Buckling load = %10.6g',e(1,1))
    d = zeros(n,1);
    d(df) = v(:,1)
    plot(nodes,d),title('First buckling mode'), xlabel('x'),ylabel('v')
    >> BucklingEx

    Kef =
    1.0e+005 *
    1.7778 -0.8889
    0
    0
    0
    0
    0
    -0.8889 1.5556 -0.8889 0.1111
    0
    0
    0
    0 -0.8889 1.7778 -0.8889
    0
    0
    0
    0 0.1111 -0.8889 1.5556 -0.8889 0.1111
    0
    0
    0
    0 -0.8889 1.7778 -0.8889
    0
    0
    0
    0 0.1111 -0.8889 1.5556 -0.8889
    0
    0
    0
    0
    0 -0.8889 1.7778

    Kep =
    16 2 0 0 0 0 0
    2 8 2 -1 0 0 0
    0 2 16 2 0 0 0
    0 -1 2 8 2 -1 0
    0 0 0 2 16 2 0
    0 0 0 -1 2 8 2
    0 0 0 0 0 2 16
    Buckling load =
    d=
    0
    0.0494
    0.0913
    0.1193
    0.1292
    0.1193
    0.0913
    0.0494
    0

    685.74

    You might also like