Upload
blackzeny
View
219
Download
0
Embed Size (px)
Citation preview
8/11/2019 1. intro_rev
1/44
TEKNIK KOMPUTASI
(TEI 116)3 SKS
Oleh: Husni Rois Ali, S.T., M.Eng.
Noor Akhmad Setiawan, S.T., M.T., Ph.D.
Email : [email protected] or
Personal web: husniroisali.staff.ugm.ac.id
8/11/2019 1. intro_rev
2/44
Grading
First Half (tentative)
Midterm (80%)
Homework, quiz, etc. (20%)
Second Half (?)
8/11/2019 1. intro_rev
3/44
References:
Numerical Methods for Engineers, Sixth Edition
by Steven Chapra(Author), Raymond
Canale(Author)
http://numericalmethods.eng.usf.edu
http://www.amazon.com/s/ref=dp_byline_sr_book_1?ie=UTF8&field-author=Steven+Chapra&search-alias=books&text=Steven+Chapra&sort=relevancerankhttp://www.amazon.com/s/ref=dp_byline_sr_book_2?ie=UTF8&field-author=Raymond+Canale&search-alias=books&text=Raymond+Canale&sort=relevancerankhttp://www.amazon.com/s/ref=dp_byline_sr_book_2?ie=UTF8&field-author=Raymond+Canale&search-alias=books&text=Raymond+Canale&sort=relevancerankhttp://numericalmethods.eng.usf.edu/http://numericalmethods.eng.usf.edu/http://www.amazon.com/s/ref=dp_byline_sr_book_2?ie=UTF8&field-author=Raymond+Canale&search-alias=books&text=Raymond+Canale&sort=relevancerankhttp://www.amazon.com/s/ref=dp_byline_sr_book_2?ie=UTF8&field-author=Raymond+Canale&search-alias=books&text=Raymond+Canale&sort=relevancerankhttp://www.amazon.com/s/ref=dp_byline_sr_book_1?ie=UTF8&field-author=Steven+Chapra&search-alias=books&text=Steven+Chapra&sort=relevancerankhttp://www.amazon.com/s/ref=dp_byline_sr_book_1?ie=UTF8&field-author=Steven+Chapra&search-alias=books&text=Steven+Chapra&sort=relevancerank8/11/2019 1. intro_rev
4/44
What are we going to learn?
Finding the root of the equations i.e. solve for
Linear algebraic equation i.e. given and solve
( ) 0f x x
2 3 2 0x x
'a s 'b s
8/11/2019 1. intro_rev
5/44
What are we going to learn? (2)
Curve fitting
Integration
8/11/2019 1. intro_rev
6/44
What are we going to learn? (3)
Ordinary differential equations (ODE)
Partial differential equation
8/11/2019 1. intro_rev
7/44
Analytic vs Numerical Approach
Analytic
You obtain the exact
solution
It is single step There is no error (since it is
exact)
Normally used for simple
problem This is how you work
Numerical
You obtain approximate
solution
It is an iterative approach The error is a crucial factor
Used for complex problems
This is how your computer
and all computationalsoftware works
8/11/2019 1. intro_rev
8/44
Analytic vs. Numerical Approach (2)suppose you want to find the roots of 2
Analytic
Formulate to mathematical
equation
Factorize
Then
Numerical
Formulate to mathematical
equation
Start from an arbitrary initial
value of , e.g.
You update the value of to
better approximation, until you
close enough
How you move from one point to
another depends on the theal orithm
2( ) 2 0f x x
( 2)( 2) 0x x
2x
2( ) 2 0f x x
x0 1x
x
8/11/2019 1. intro_rev
9/44
Introduction
Mathematical Modeling and Engineering Problem solving
Requires understanding of
engineering systemsBy observation and experimentTheoretical analysis and
generalization
Computers are great tools,
however, without fundamentalunderstanding of engineering
problems, they will be useless.
8/11/2019 1. intro_rev
10/44
1-11
A mathematical model is represented as a functional
relationship of the form
Dependent independent forcing
Variable =f variables, parameters, functions
It ranges from a simple equation to very complex ones
Dependent variable: Characteristic that usually reflects thestate of the system
Independent variables: Dimensions such as time and spacealong which the systems behavior is being determined
Parameters: reflect the systems properties or composition
Forcing functions: external influences acting upon the system
8/11/2019 1. intro_rev
11/44
Newtons 2ndlaw of Motion
The model is formulated as
F = m a
F=net force acting on the body (N)a forcing
function
m=mass of the object (kg)a parameter
a=its acceleration (m/s2) -> the dependent variable
8/11/2019 1. intro_rev
12/44
Formulation of Newtons 2ndlaw has several
characteristics that are typical of mathematical
models of the physical world:
It describes a natural process or system in
mathematical terms
It represents an idealization and simplification of
reality Finally, it yields reproducible results,
consequently, can be used for predictive purposes.
8/11/2019 1. intro_rev
13/44
1-14
Some mathematical models of physical phenomena
may be much more complex.
Complex models may not be solved exactly or
require more sophisticated mathematical techniques
than simple algebra for their solution
Example, modeling of a falling parachutist:
8/11/2019 1. intro_rev
14/44
with , andD U D U
dv F
dt m
F F F F mg F cv
dv mg cv
dt m
dv c
g vdt m
This is a differential equation and is written in
terms of the differential rate of change dv/dt
of the variable that we are interested in
predicting.
If the parachutist is initially at rest (v=0 at t=0),using calculus
tmcec
gmtv )/(1)(
Independent
variable
Dependent
variable
Parameters
Forcingfunction
8/11/2019 1. intro_rev
15/44
8/11/2019 1. intro_rev
16/44
Write the velocity equation as
We may obtain
Rearranging yields
8/11/2019 1. intro_rev
17/44
8/11/2019 1. intro_rev
18/44
Computer software
Excel
Matlab
Scilab Fortran 90 (IMSL)
C++
8/11/2019 1. intro_rev
19/44
20
Approximations Errors
For many engineering problems, we cannot obtain analyticalsolutions.
Numerical methods yield approximate results, results that areclose to the exact analytical solution. We cannot exactlycompute the errors associated with numerical methods.
Only rarely given data are exact, since they originate frommeasurements. Therefore there is probably error in the inputinformation.
Algorithm itself usually introduces errors as well, e.g., unavoidableround-offs, etc
The output information will then contain error from both of thesesources.
How confident we are in our approximate result? The question is how much error is present in our calculation
and is it tolerable?
8/11/2019 1. intro_rev
20/44
21
Accuracy.How close is a computed or
measured value to the true value
Precision (or reproducibility).How close is a
computed or measured value to previously
computed or measured values. Inaccuracy(or bias).A systematic deviation
from the actual value.
Imprecision(or uncertainty).Magnitude ofscatter.
8/11/2019 1. intro_rev
21/44
22
8/11/2019 1. intro_rev
22/44
Significant Figures
8/11/2019 1. intro_rev
23/44
24
Significant Figures (2)
Number of significant figures indicates precision. Significant digits of anumber are those that can be usedwith confidence, e.g.,the number ofcertain digits plus one estimated digit.
53,800 How many significant figures?
5.38 x 104 3
5.380 x 104 4
5.3800 x 104 5
Zeros are sometimes used to locate the decimal point not significantfigures.
0.00001753 4
0.0001753 4
0.001753 4
8/11/2019 1. intro_rev
24/44
Significant Figures (3)
Implication :
1. Numerical methods yield approximate results. We must develop
criteria to specify how confident we are in our approximate result.
One way to do this is in terms of significant figures.
2. Although quantities such as , e, or 7 represent specific quantities,they cannot be expressed exactly by a limited number of digits.
Because computers retain only a finite number of significant figures,
such numbers can never be represented exactly. The omission of the
remaining significant figures is called round-off error.
8/11/2019 1. intro_rev
25/44
26
Error Definitions
True Value = Approximation + Error
Et= True valueApproximation (+/-)
valuetrue
errortrueerrorrelativefractionalTrue
%100valuetrue
errortrueerror,relativepercentTrue
t
True error
F i l th d th t l ill b
8/11/2019 1. intro_rev
26/44
27
For numerical methods, the true value will be
known only when we deal with functions that can
be solved analytically (simple systems). In real
world applications, we usually not know the answera priori. Then
Iterative approach, example Newtons method
%100
ionApproximat
erroreApproximat
a
%100ionapproximatCurrent
ionapproximatPrevious-ionapproximatCurrenta (+ / -)
8/11/2019 1. intro_rev
27/44
28
Use absolute value.
Computations are repeated until stopping criterion issatisfied.
If the following criterion is met
you can be sure that the result is correct to at least nsignificant figures.
sa Pre-specified % tolerance based on
the knowledge of your solution
)%10(0.5n)-(2
s
8/11/2019 1. intro_rev
28/44
8/11/2019 1. intro_rev
29/44
8/11/2019 1. intro_rev
30/44
31
Two sources of numerical error
1) Round off error
2) Truncation error
8/11/2019 1. intro_rev
31/44
32
Round-off Error
8/11/2019 1. intro_rev
32/44
33
Round off Error
Caused by representing a number approximately
It is due to the limited sources of computer in
representing the number
10.333333
3 2 1.4142...
8/11/2019 1. intro_rev
33/44
34
Problems created by round off error
28 Americans were killed on February 25,
1991 by an Iraqi Scud missile in Dhahran,
Saudi Arabia.
The patriot defense system failed to track and
intercept the Scud. Why?
8/11/2019 1. intro_rev
34/44
35
Problem with Patriot missile
Clock cycle of 1/10 seconds was
represented in 24-bit fixed point
register created an error of 9.5 x 10-8
seconds.
The battery was on for 100
consecutive hours, thus causing an
inaccuracy of
1hr
3600s100hr
0.1s
s109.5 8
s342.0
8/11/2019 1. intro_rev
35/44
36
Problem (cont.)
The shift calculated in the ranging system of
the missile was 687 meters.
The target was considered to be out of range
at a distance greater than 137 meters.
8/11/2019 1. intro_rev
36/44
37
Truncation Error
8/11/2019 1. intro_rev
37/44
38
Truncation error
Error caused by truncating or
approximating a mathematical procedure.
8/11/2019 1. intro_rev
38/44
39
Example of Truncation Error
Taking only a few terms of a Maclaurin series to
approximate
....................!3!2
132
xxxex
xe
If only 3 terms are used,
!21
2
xxeErrorTruncation x
8/11/2019 1. intro_rev
39/44
40
Another Example of Truncation Error
Using a finite x to approximate )(xf
x
xfxxfxf
)()()(
P
Q
secant line
tangent line
Figure 1. Approximate derivative using finite x
8/11/2019 1. intro_rev
40/44
41
Another Example of Truncation Error
Using finite rectangles to approximate an
integral.
y = x2
0
30
60
90
0 1.5 3 4.5 6 7.5 9 10.5 12
x
8/11/2019 1. intro_rev
41/44
42
Example 1 Maclaurin series
Calculate the value of2.1e with an absolute
relative approximate error of less than 1%.
...................!3
2.1
!2
2.12.11
322.1
e
n
1 1 __ ___
2 2.2 1.2 54.545
3 2.92 0.72 24.658
4 3.208 0.288 8.9776
5 3.2944 0.0864 2.6226
6 3.3151 0.020736 0.62550
aE %a2.1e
6 terms are required.
8/11/2019 1. intro_rev
42/44
43
Example 2 Differentiation
Find )3(f for2)( xxf using
x
xfxxfxf
)()()(
and 2.0x
2.0
)3()2.03(
)3(' ff
f
2.0
)3()2.3( ff
2.0
32.3 22
2.0
924.10
2.0
24.1 2.6
The actual value is
,2)(' xxf 632)3(' f
Truncation error is then, 2.02.66
Can you find the truncation error with 1.0x
8/11/2019 1. intro_rev
43/44
44
Example 3 Integration
Use two rectangles of equal width to approximate
the area under the curve for2)( xxf over the interval ]9,3[
y = x2
0
30
60
90
0 3 6 9 12
x
9
3
2dxx
8/11/2019 1. intro_rev
44/44
Integration example (cont.)
)69()()36()(6
2
3
2
9
3
2 xx xxdxx
3)6(3)3( 22 13510827
Choosing a width of 3, we have
Actual value is given by
9
3
2
dxx
9
3
3
3
x
2343
3933
Truncation error is then
99135234