Numerical Methods Test 1

Numerical Methods Test 1
1. put this in a function file.
2. Put this into the command window

1. The easiest way to use the fzero function is to use an anonymous function as such
2. Make sure that the window is the same as the question
1. Open up the “Bisection-False Position (2) ” method excel sheet.
2. The bracket [0 1] means xl = 0 and xu = 1, make sure xr = (xl + xu)/2 don’t change any xl or
xu except that first row
3. The question asks for the iteration question so don’t forget to write that down.
4. Input the function given into f(xl), using the xl as the x
5. Drag from the bottom right of the f(xl) box and drag to the right, then drag them down.
6. The question only asks for the first four calculations, so that’s all you have to write down.
Pay attention to the ^-01’s
1. Open up the Newton excel file, honestly probably best if you download a new one each
time
2. Change the x_i to the initial value, in this case it’s 0
3. Input the function given into f(x_i) using the x_i as the x.
4. Go into mathway.com and input the function and find df/dx

5. Input this function into f’(x_i) spot still using x_i as the x
6. The question asks for the first four calculation steps, so you only have to write down the
first four
7. The second question also asks for the question for xi+1 so make sure to write that down.
f (xi )
The general equation for xi+1 is as follows, xi+1 = xi − therefore, the xi+1 equation for this
f '(xi )
2 * xi − (tan(xi ) − 1)2
example is as follows, xi+1 = xi −
−2 *sec 2 (xi ) tan(xi ) + 2 + 2sec 2 (xi )
1. Write down the general formula on paper for a matrix in a linear equation system.
⎡⎣ A⎤⎦ ⎡⎣ X ⎤⎦ = ⎡⎣ b ⎤⎦
2. Write down on paper the system of equations in this form.
⎡ 8 2 3 ⎤ X1 51
⎢ ⎥
⎢ 2 5 1 ⎥ * X 2 = 23
⎢ −3 1 6 ⎥ X3 20
⎣ ⎦
1. Input matrix A and B into MATLAB.
2. If you want to check the answer later on you can input this command initially.
3. Complete LU factorization in MATLAB with the following command.
4. Input unit vectors b into MATLAB as follows.
5. Input values for d using the lower matrix and x using the upper matrix.
6. Create the inverse matrix by making a matrix of x1, x2, and x3.
7. You can check your work by subtracting the MATLAB function of inv and the one you
created.
8. If an all zero matrix is formed, the matrix is correct.
1. Solve the system in MATLAB to obtain a known answer.
2. Write out the iterative equations. The general form for the Gauss-Seidel Method is as
follows:
b - a x j-1 - a13 x3j-1
x1j = 1 12 2
a11
b2 - a21 x1j - a23 x3j-1
x2j =
a22
b3 - a31 x1j - a32 x2j
x3j =
a33
The iterative equations for this problem is as follows:
51− 2 * X 2i − 3* X 3i
X =
i+1
1
8
23− 2 * X 1i+1 − X 3i
X 2i+1 =
5
20 + 3* X ii+1 − X 2i+1
X3 =
i+1
6
3. Download the excel file, “ Gauss-Seidel_Jacobi Method”
4. The first row for i, x1, x2, and x3 should all be 0. Input the iterative equations into x1, x2,
and x3.
5. Drag down the cells until x1, x2, and x3 converge to the correct answer from MATLAB’s
output. This happens in less than 10 steps.
6. On paper only write down the first two steps as the question asks.
Successive Substitution Method:

1. Solve for x1 and x2 by hand.
2x2 + e−4 x1
x1 =
5
2x + e−3x2
x2 = 1
5
2. Assume that x1,o and x2,o are equal to zero, then plug those into the equations and solve. x1
and x2 can be solved using MATLAB must be written out by hand
⎛ 1 ⎞ −3 0
2⎜ ⎟ + e ( )
2 (0) + e
−4( 0 )
1 ⎝ 5⎠
x1,1 = = = 0.2 x2,1 = = 0.28
5 5 5
3. Repeat the same step except with the new values for x1 and x2
2 ( 0.28) + e−4*0.2 2 ( 0.201866 ) + e−3*0.28
x1,2 = = 0.201866 x2,2 = = 0.167088
5 5
4. Only the first two steps have to be written down, so the rest can be solved in excel by
inputting the starting values, x1, and x2.
5. Drag down x1 and x2 columns separately until they converge.

Newton’s Method
1. Find the Jacobian Matrix.
The general form for the Jacobian Matrix is as follows:

⎡ ∂f ∂ f1 ⎤
⎢ 1
… ⎥
⎢ ∂x1 ∂xn ⎥
⎡ ∂f ∂f ⎤ ⎢
⎢ ⎥= ⎥
J= … ⎢ ! " ! ⎥
⎢⎣ ∂x1 ∂xn ⎥
⎦ ⎢ ∂ fm ∂ fm ⎥
⎢ ∂x # ∂x ⎥
⎢⎣ 1 n ⎥
⎦
The Jacobian Matrix for this problem is as follow, which utilizes the equations from the
problem
⎡ ∂f ∂ f1 ⎤
⎢ 1
⎥
⎢ ∂x1 ∂x2 ⎥ ⎡ 5 + 4e−4 x1 −2 ⎤
J=⎢ ⎥=⎢ ⎥
⎢ ∂ f 2 ∂ f 2 ⎥ ⎢⎣ −2 5 + 3e−3x2 ⎥⎦
⎢ ∂x1 ∂x2 ⎥
⎣ ⎦
2. Input into MATLAB the following function as a function file
3. Utilize the Newton-Raphson m-file to find the values for x.
The Newton-Raphson m-file,

The command:
Excel Solver Method
1. input cells for x1, x2, f1, f2, and the sum.
2. Input equation 1 and 2 from problem into f1 and f2 cells, then the sum of their square
roots into the cell “sum sq”
3. Use the excel solver by setting the objective to the sum cell, changing the variable cells, x1
and x2 by dragging over both cells, and setting the objective to the value of 0 since both
equations are equal to 0 in the problem.
4. When you hit solve, the x1 and x2 value will converge to the correct answer.
Using Fsolve
1. Enter the following function into a function file in MATLAB
2. Enter the following into the command window utilizing the MATLAB built in function
fsolve, using the parameters [0 0] since f1 = 0 and f2 = 0.
This is the quickest method to solve this problem.
y = −x 2 + x + 0.75, y = x 2 − 5xy

Numerical Methods Test 1

Uploaded by

Copyright:

Available Formats

Numerical Methods Test 1

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Numerical Methods Test 1

Uploaded by

Copyright:

Available Formats

Numerical Methods Test 1

1. put this in a function file.

2. Put this into the command window

2. Make sure that the window is the same as the question

1. Open up the “Bisection-False Position (2) ” method excel sheet.

4. Input the function given into f(xl), using the xl as the x

2. Change the x_i to the initial value, in this case it’s 0

4. Go into mathway.com and input the function and find df/dx

2. Write down on paper the system of equations in this form.

1. Input matrix A and B into MATLAB.

4. Input unit vectors b into MATLAB as follows.

8. If an all zero matrix is formed, the matrix is correct.

1. Solve the system in MATLAB to obtain a known answer.

Successive Substitution Method:

5. Drag down x1 and x2 columns separately until they converge.

1. Find the Jacobian Matrix.

The general form for the Jacobian Matrix is as follows:

The Newton-Raphson m-file,

Excel Solver Method

1. Enter the following function into a function file in MATLAB

This is the quickest method to solve this problem.

You might also like