+ All Categories
Home > Documents > Lecture 11BUE

Lecture 11BUE

Date post: 04-Jun-2018
Category:
Upload: tarek-hassan
View: 217 times
Download: 0 times
Share this document with a friend

of 28

Transcript
  • 8/14/2019 Lecture 11BUE

    1/28

    ScheduleWeek Date Topic Classification of Topic

    1 9 Feb. 2010 Introduction toNumerical Methods

    and Type of Errors

    Measuring errors, Binary representation,Propagation of errors and Taylor series

    2 14 Feb. 2010 Nonlinear Equations Bisection Method

    3 21 Feb. 2010 Newton-Raphson Method

    4 28 Feb. 2010 Interpolation Lagrange Interpolation

    5 7 March 2010 Newton's Divided Difference Method

    6 14 March 2010 Differentiation Newton's Forward and BackwardDivided Difference

    7 21 March 2010 Regression Least squares

    8 28 March 2010 Systems of LinearEquations

    Gauss-Jacobi

    9 11 April 2010 Gauss- Seidel

    10 18 April 2010 Integration Composite Trapezoidal and SimpsonRules

    11 25 April 2010 Ordinary DifferentialEquations

    Euler's Method

    12 2 May 2010 Runge-Kutta 2nd and4th order Method

  • 8/14/2019 Lecture 11BUE

    2/28

    Gauss-Jacobi

    Iterative or approximate methodsprovide an alternative to theelimination methods. The Gauss-Jacobimethod is the most commonly used

    iterative method.

    The system [A]{X}={B} is reshapedby solving the first equation forx1, the

    second equation for x2, and the thirdforx3, and n

    th equation forxn.

  • 8/14/2019 Lecture 11BUE

    3/28

    Gauss-Jacobi Method

    Algorithm

    A set of nequations and nunknowns:

    11313212111 ... bxaxaxaxa nn

    2323222121 ... bxaxaxaxa n2n

    nnnnnnn bxaxaxaxa ...332211

    . .

    . .

    . .

    If: the diagonal elements arenon-zero

    Rewrite each equation solvingfor the corresponding unknown

    ex:

    First equation, solve for x1

    Second equation, solve for x2

  • 8/14/2019 Lecture 11BUE

    4/28

  • 8/14/2019 Lecture 11BUE

    5/28

    Gauss-Jacobi MethodSolve for the unknowns

    Assume an initial guess for [X]

    n

    -n

    2

    x

    x

    x

    x

    1

    1

    Use rewritten equations to solve for

    each value of xi.

    Important: Remember to use themost recent value of xi. Which

    means to apply values calculated to

    the calculations remaining in the

    current iteration.

  • 8/14/2019 Lecture 11BUE

    6/28

    3

    333

    2

    212

    1

    111

    equations33For

    c

    zbxadz

    b

    zcxady

    a

    zcybdx

    Now we can start the solution process by

    choosing guesses for the xs. A simple way toobtain initial guesses is to assume that they

    are zero. These zeros can be substituted into

    x1equation to calculate a newx

    0=d

    1/a

    1.

  • 8/14/2019 Lecture 11BUE

    7/28

    First iteration

    3

    333

    2

    222

    1

    111

    equations33For

    c

    ybxadz

    b

    zcxady

    a

    zcybdx

    2

    030331

    3

    020221

    1

    010111

    equations33For

    b

    ybxadz

    c

    zcxady

    a

    zcybdx

    Applying the initial guess and solving for x, y,

    and z

    0

    0

    0

    z

    y

    x

    z

    y

    x

  • 8/14/2019 Lecture 11BUE

    8/28

    Gauss-Jacobi Method

    Calculate the Absolute Relative Approximate Error

    100

    newi

    old

    i

    new

    i

    ia x

    xx

    So when has the answer been found?

    The iterations are stopped when the

    absolute relative approximate error is less

    than a prespecified tolerance for all

    unknowns.

  • 8/14/2019 Lecture 11BUE

    9/28

    Convergence Criterion for Gauss-

    Jacobi Method The Gauss-Jacobi method has two

    fundamental problems as any iterative

    method:

    It is sometimes nonconvergent, and

    If it converges, converges very slowly.

  • 8/14/2019 Lecture 11BUE

    10/28

    Gauss-Jacobi Method

    Why?

    The Gauss-Jacobi Method allows the user to control

    round-off error.

    Elimination methods such as Gaussian Elimination

    and LU Decomposition are prone to prone to round-off

    error.

    Also: If the physics of the problem are understood, a

    close initial guess can be made, decreasing the

    number of iterations needed.

  • 8/14/2019 Lecture 11BUE

    11/28

    Gauss-Jacobi Method: Example 1

    The upward velocity of a rocket is given at three different

    times

    Time Velocity

    5 106.8

    8 177.2

    12 279.2

    The velocity data is approximated by a polynomial as:

    12.t5,322

    1 atatatv

    st m/sv

    Table 1 Velocity vs. Time data.

  • 8/14/2019 Lecture 11BUE

    12/28

    Gauss-Jacobi Method: Example 1

    3

    2

    1

    3

    2

    3

    2

    2

    2

    1

    2

    1

    1

    1

    1

    v

    v

    v

    a

    a

    a

    tt

    tt

    tt

    3

    2

    1Using a Matrix template of the form

    The system of equations becomes

    2.279

    2.177

    8.106

    112144

    1864

    1525

    3

    2

    1

    a

    a

    a

    Initial Guess: Assume an initial guess of

    5

    2

    1

    3

    2

    1

    a

    a

    a

  • 8/14/2019 Lecture 11BUE

    13/28

    Gauss-Jacobi Method: Example 1Rewriting each equation

    2.279

    2.177

    8.106

    112144

    1864

    1525

    3

    2

    1

    a

    a

    a

    25

    58.106 321

    aaa

    8

    642.177 312

    aaa

    1

    121442.279 213

    aaa

  • 8/14/2019 Lecture 11BUE

    14/28

    Gauss-Jacobi Method: Example 1

    Applying the initial guess and solving for ai

    5

    2

    1

    3

    2

    1

    a

    a

    a 6720.325

    )5()2(58.1061

    a

    525.13

    8

    51642.1772

    a

    2.111

    1

    21211442.2793

    a

    Initial Guess

  • 8/14/2019 Lecture 11BUE

    15/28

    Gauss-Jacobi Method: Example 1

    %76.721006720.3

    0000.16720.31a

    x

    %212.85100525.13

    0000.2525.132a

    x

    %5.951002.111

    0000.52.1113a

    x

    Finding the absolute relative approximate error

    100

    new

    i

    old

    i

    new

    i

    ia x

    xx At the end of the first iteration

    The maximum absolute

    relative approximate error is95.5%

    2.111

    525.136720.3

    3

    2

    1

    a

    aa

  • 8/14/2019 Lecture 11BUE

    16/28

    Gauss-Jacobi Method: Example 1

    Iteration #2

    Using

    2.111

    525.13

    6720.3

    3

    2

    1

    a

    a

    a

    881.225

    2.111525.1358.106

    a1

    126.21

    8

    2.111672.3642.177a2

    868.411

    1

    525.1312672.31442.279a3

    from iteration #1

    the values of ai are found:

  • 8/14/2019 Lecture 11BUE

    17/28

    Gauss-Jacobi Method: Example 1

    Finding the absolute relative approximate error

    %455.227100881.2

    6720.3881.21a

    x

    %02.164100126.21

    525.13126.212a

    x

    %05.127100686.411

    2.111868.4113a

    x

    At the end of the second iteration

    868.411

    126.21

    881.2

    3

    2

    1

    a

    a

    a

    The maximum absolute

    relative approximate error is

    227.455%

  • 8/14/2019 Lecture 11BUE

    18/28

    Iteration a1 a2 a3

    1

    23

    4

    3.6720

    -2.8818.2972

    -42.545

    72.767

    227.455134.722

    119.502

    13.525

    21.12644.573

    162.67

    85.212

    164.02147.396

    127.403

    111.2

    -411.868947.576

    -1450.47

    95.5

    127.05143.465

    165.329

    Gauss-Jacobi Method: Example 1

    0857.1

    690.19

    29048.0

    a

    a

    a

    3

    2

    1

    Repeating more iterations, the following values are obtained

    %1a

    %2a

    %3a

    Notice The relative errors are not decreasing at any significant rate

    Also, the solution is not converging to the true solution of

  • 8/14/2019 Lecture 11BUE

    19/28

    Gauss-Jacobi Method: PitfallWhat went wrong?

    Even though done correctly, the answer is not converging to the

    correct answer

    This example illustrates a pitfall of the Gauss-Jacobi method: not allsystems of equations will converge.

    Is there a fix?

    One class of system of equations always converges: One with a diagonally

    dominant coefficient matrix.

    Diagonally dominant: [A] in [A] [X] = [C] is diagonally dominant if:

    n

    ij

    j

    ijii aa1

    n

    j

    j

    ijaa

    i

    1

    iifor all i and

    for at least one i

  • 8/14/2019 Lecture 11BUE

    20/28

    Gauss-Jacobi Method: Pitfall

    116123

    14345

    3481.52

    A

    Diagonally dominant: The coefficient on the diagonal must be at leastequal to the sum of the other coefficients in that row and at least one row

    with a diagonal coefficient greater than the sum of the other coefficients

    in that row.

    1293496

    55323

    5634124

    ]B[

    Which coefficient matrix is diagonally dominant?

    Most physical systems do result in simultaneous linear equations that

    have diagonally dominant coefficient matrices.

  • 8/14/2019 Lecture 11BUE

    21/28

    Gauss-Jacobi Method: Example 2Given the system of equations

    20238 zyx

    zyx 33114

    351236 zyx

    0

    00

    z

    yx

    With an initial guess of

    The coefficient matrix is:

    1236

    1114

    238

    A

    Will the solution converge using the

    Gauss-Jacobi method?

  • 8/14/2019 Lecture 11BUE

    22/28

    Gauss-Jacobi Method: Example 2Checking if the coefficient matrix is diagonally dominant

    5141111 232122 aaa

    9361212 323133 aaa

    52388 131211 aaa

    The inequalities are all true and at least one row is strictlygreater than:

    Therefore: The solution should converge using the Gauss-Jacobi Method

    12361114

    238

    A

  • 8/14/2019 Lecture 11BUE

    23/28

    Gauss-Jacobi Method: Example 2

    3533

    20

    zy

    x

    12361114

    238

    Rewriting each equation

    8

    2320 zyx

    11433 zxy

    12

    3635 yxz

    With an initial guess of

    0

    0

    0

    z

    y

    x

    5.2

    12

    020320x1

    311 00433y1

    9166667.2

    13

    030635z1

  • 8/14/2019 Lecture 11BUE

    24/28

    Gauss-Jacobi Method: Example 2The absolute relative approximate error

    %00.1001005.2

    0.05.21

    a

    %00.1001000.3

    0.00.32

    a

    %100100916666.2

    0.0916666.23

    a

    The maximum absolute relative error after the first iteration is 100%

  • 8/14/2019 Lecture 11BUE

    25/28

    Gauss-Jacobi Method: Example 2

    916666.0

    3560606.2

    895833.2

    2

    2

    2

    z

    y

    x

    After Iteration #1

    895833.2

    8

    9166667.2233202

    x

    3560606.2

    11

    9166667.25.24332

    y

    916666.0

    12

    335.26352

    z

    Substituting the values into

    the equationsAfter Iteration #2

    9166667.2

    0000.3

    5000.2

    1

    1

    1

    z

    y

    x

    111 ,, zyx

  • 8/14/2019 Lecture 11BUE

    26/28

    Gauss-Jacobi Method: Example 2Iteration #2 absolute relative approximate error

    %669055.13100895833.2

    50000.2895833.21a

    %33119.271003560606.2

    0.33560606.22a

    %181.218100

    916667.0

    916667.2916667.03a

    The maximum absolute relative error after the second iteration is 218.181%

    This is much larger than the maximum absolute relative error obtained in

    iteration #1. Is this a problem?

  • 8/14/2019 Lecture 11BUE

    27/28

    Iteration x y z

    1

    23

    4

    5

    2.5

    2.895833.1543

    3.04143

    3.01687

    100.00

    13.66908.1942

    3.7110

    0.8140

    3

    2.35602.0303

    1.9329

    1.9696

    100.00

    27.331116.041

    5.0390

    1.8633

    2.9166

    0.916660.8797

    0.83191

    0.9127

    82.193

    218.1814.2014

    5.7446

    8.8911

    Gauss-Jacobi Method: Example 2Repeating more iterations, the following values are obtained%

    1a %

    2a %

    3a

    9118312.0

    9858987.1

    0167424.3

    z

    y

    xThe solution obtained is close to the exact solution of .

    9118.0

    985.1

    0167.3

    z

    y

    x

  • 8/14/2019 Lecture 11BUE

    28/28

    Gauss-Jacobi MethodThe Gauss-Jacobi Method can still be used

    The coefficient matrix is not

    diagonally dominant

    5312

    351

    1373

    A

    But this is the same set of

    equations used in example #2,

    which did converge.

    1373

    351

    5312

    A

    If a system of linear equations is not diagonally dominant, check to see if

    rearranging the equations can form a diagonally dominant matrix.


Recommended