Numerical Methods for Partial Differential Equations. CAAM 452 Spring 2005 Lecture 3 AB2,AB3, Stability, Accuracy Instructor: Tim Warburton

Size: px
Start display at page:

Download "Numerical Methods for Partial Differential Equations. CAAM 452 Spring 2005 Lecture 3 AB2,AB3, Stability, Accuracy Instructor: Tim Warburton"

Transcription

1 Numerical Methods for Partial Differential Equations CAAM 452 Spring 25 Lecture 3 AB2,AB3, Stability, Accuracy Instructor: Tim Warburton

2 Recall: Euler-Forward Time Stepping By several different approximations we arrived at the following approximate ODE time-stepping scheme: du dt = f u for some Assuming: f u = µ u µ We established the following stability region u = u u = u + dtf u n+ 1 n n Im 1+ dtµ 1 x -1 Re i.e. if mu*dt is in the yellow region the approximate solution will be bounded.

3 Recall: Adams Bashford 2 We went on to derive a similar method which uses ( ) and f u t n f u tn 1 Namely AB2: 3 1 u = u + dt f u f u 2 2 n+ 1 n n n 1 This was based on an interpolation of f(u(t)) We next ask which values of mu*dt is this stable for?

4 Derivation of AB3 Suppose we wish to use three previous values of f ( ) f u t n, f u tn 1 and f u tn 2 As previously we look for a fit through this data, although this time we choose a different path. (i) Consider dt = 1 (will rescale later) (ii) Will form the coefficients of the interpolant using a Vandermonde matrix and a monomial basis in t (iii) Finally integrate the integrand (easy with this choice of basis).

5 (i) form quadratic interpolant: (ii) Compute coefficients: If = a + bt + ct 2 f = a b2 + c4 2 f = a b + c f 1 = a a b = c f 2 f 1 f f 2 2 If2 = ( 1 t t ) f 1 1 f Simpler Approach (iii)integrate interpolant: f If2dt ( 1 t t ) f = 1 dt 1 f f = f = 23 f 16 f + 5 f f

6 AB3 We now have an interpolant for dt=1 based on 1 three time levels: 1 If dt = 23 f 16 f + 5 f 12 We rescale and shift: t n+ 1 t n dt If dt = f f + f 12 ( ) 2 n n 1 n 2 And finally have AB3: dt u = u + f f + f 12 ( ) n+ 1 n n n 1 n 2

7 Absolute Stability of AB2 We again assume: u for some f u = µ µ AB2 becomes: 3 1 u = u + dt µ u µ u 2 2 n+ 1 n n n 1 Tidying up: 3 1 un ν u n + νun 1 = 2 2 To start this method we require: u u dt u 1 u ν = dtµ

8 Absolute Stability of AB2 cont We define the absolute stability region of this method as the set of nu in the complex plane for which the sequence {u n } is bounded as n 3 1 un ν u n + νun 1 = 2 2 We will make continuous reference to the stability polynomial of this recurrence relationship (obtained n by setting : un = z and canceling excess powers of z ) πν ( z) = z + 1 ν z + ν 2 2

9 Note on Solutions of Recurrence Relations Given an s+1 term recurrence relationship (concisely written as): With stability polynomial: j= s j= α u j n+ j ( z) = which (we suppose) has s distinct roots π m= s { r m } m= 1 Then any solution to the recurrence relation can be written as: where the c m depend on the s initiating values for the recurrence. m= s n n = rm cm m= 1 This follows because n m is a solution, and there is a limited rank of possible solutions because of the linearity of the recurrence in the initial values. n= { } 1 u n n = s u = r n = j= s j= α z j j u Board demo

10 cont Since all solutions under these conditions can be m= s written as: n u = r c n m m m= 1 We immediately observe that this is bounded if and only if r 1 m See Trefethen p43-46 for details of the case where the stability polynomial has 1 or more roots with multiplicity greater than 1.

11 Root Condition for Absolute Stability See: Trefethen p58-6 A linear multistep formula is absolutely stable for a particular value of nu if and only if all the zeros of the stability polynomial πν ( z) satisfy z 1 and any zero with z =1 is a simple root.

12 Drawing the Stability Region We need to find the roots of the stability polynomial: πν ( z) = z + 1 ν z + ν 2 2 i.e. r ( r) such that = π ν Which lie on the margin of stability r =1 i The method we adopt is to set r = e θ for θ [,2 π ) and find the value of nu such that r is a root of the stability polynomial.

13 cont We can rearrange the root equation to find nu in terms of z: πν ( r) = r + 1 ν r + ν = 2 2 ν = = iθ 1 r e iθ iθ r r e e We can now ask matlab to plot this curve in the complex plane

14 Matlab Plot of Margin of Stability for AB2

15 AB2 v. AB3 v. AB4 I repeated this for AB2, AB3, AB4 Notice how the overall region of absolute stability gets smaller as we use more and more previous values of f This translates into stricter restrictions on dt to force nu=mu*dt into the stability regions.

16 Zooming Into The All Important Imaginary Axis Recall, we discounted Euler-Forward as useless for timestepping the advection equation because it had zero coverage of the imaginary axis. This picture does not quite tell us if the imaginary axis is inside the absolute stability region for AB2, AB3 and AB4. So we had better check the root condition.

17 Root Condition for AB2 We need to check that the roots of: z + 1 ν z + ν = 2 2 Are bounded above by 1, and if any roots are on the unit circle we must make sure they are simple. We set nu=i*alpha z + 1 iα z + iα = 2 2

18 Always Experiment First So rather than waste time figuring out where the roots are, why not let matlab do the hardwork:

19 If at first.. Since the plot is not quite clear, we plot: sign(maximum root -1)*log1( maximum root -1 ) In summary: the only stable point on the imaginary axis for AB2 is the origin. AB3 is stable for nu in i*[-.723,.723 ]

20 Accuracy & Consistency Section 1.3 p21-29 Trefethen We already established experimentally that stability is not enough for a good solution we also need to examine how accurate a solution will be.

21 Consistency Linear Multistep schemes (like Euler-Forward, AB2,AB3..) can be written as: j= s j= s α u = dt β f u j n+ j j n+ j j= j= It is natural to ask what is the local discretization error (or local truncation error) for the above formula when we substitute u ( t n ) for un We define the residual operator j= s j= s : = α β Lu t u t dt f u t n j n+ j j n+ j j= j=

22 Consistency cont If the linear multistep method is accurate the truncation error should be small. j= s j= s : = α β Lu t u t dt f u t n j n+ j j n+ j j= j= p+ 1 Definition: if Lu tn = O dt as dt then we say the scheme is p th order accurate.

23 Example Euler-Forward has: s = 1, α = 1, α = 1, β = 1, β = 1 1 So: j= s j= s : = α β Lu t u t dt f u t n j n+ j j n+ j j= j= = u t u t dtf u t n+ 1 n n du = u t u t dt t dt n+ 1 n n The next step is to estimate this in terms of dt and time derivatives of the solution.

24 Recall: Taylor s Theorem Assuming that f is sufficiently smooth (has n bounded derivatives): Then we can evaluate f at x+delta by a linear sum of f at x and its derivatives: n 1 n 1 df δ d f δ d f δ d f ( δ ) δ f t + = f t + t + t + t t + R 2 3 n 1 dt 2! dt 3! dt n 1! dt The residual takes various forms: u= t+ δ n ( δ u) Rn = ( n 1 )! (Lagrange remainder) u= t 1 n d f dt n u du ( ) n n δ d f Rn = t t t t + δ n n! dt * * for some [, ] n

25 Back to Forward-Euler We can now express the local truncation error for Forward-Euler in terms of: 2 du d u n n 2 n u t, t, t,... dt dt expanding about t n we obtain: du Lu t u t u t dt t dt = n n+ 1 n n = u t + dt t + t u t dt t dt 2! dt dt = 2 2 du dt d u du n n 2 n n n ( * ) 2 2 dt d u 2 t n 2! dt ( * )

26 cont 2 2 dt d u ( * ) ( 2 = = ) Lu tn t 2 n O dt 2! dt This indicates that the Euler-Forward method is first order in time. So what does this mean for solution accuracy?

27 Existance and Uniqueness for the Initial Value Problem Theorem (p13 Trefethen): Let f(u,t) be continuous with respect to t and uniformly Lipschitz continuous with respect to u for t [, T ].Then there exists a unique differentiable function u(t) that satisfies the initial-value problem: u du dt = u = f u t, t for all t, T [ ] The important condition here is that the right hand side f is not too non-linear. With these conditions we know we have a single solution to aim for.

28 Definition: Convergence A linear multistep method is convergent if, for all IVP s satisfying the existance/uniqueness theorem conditions on an interval [,T] and all starting values { v v v } for the method tend to u() as dt,,..., s 1 1 v n the solution satisfies v t u t = o 1 as dt uniformly for all t [, T ] The notation small o of 1 as dt tends to zero, requires that the solution drops to zero as dt tends to.

29 Theorem: Dahlquist Equivalence Theorem p54 Trefethen A linear multistep formula is convergent if and only if it is consistent and stable. Proving convergence of a method directly is difficult, however proving consistency and stability is relatively straight forward.

30 Finally: an estimate for solution accuracy Theorem (global p th order accuracy): Consider an IVP that satisfies the existance/uniqueness theorem conditions on an interval [,T] and in addition that the right hand side function f(u,t) is p times continuously differentiable with respect to u and t. Let an approximate solution be computed by a convergent linear multistep formula of accuracy >= p with starting values as in the convergence statement. Then this solution satisfies: = ( p ) as v t u t O dt dt uniformly for all t [, T ]

31 Summary Stability + consistency convergence convergence + method accuracy global solution accuracy

32 Sketch of Solution Accuracy Proof (not entirely rigorous) Given the way we constructed the AB schemes it is instructive to write down the following three equations: = + i u T u f u t dt = + T T T p p ( ) n= N ii u T = u + I f u t dt + L u t iii u T u I f u t dt n= 1 n N=T/dt

33 Sketch cont = + i u T u f u t dt = + T T T p p ( ) n= N ii u T = u + I f u t dt + L u t iii u T u I f u t dt (i) Represents the actual solution (which we know exists and is unique by the Lipschitz assumptions on f) (ii) Represents acknowledges the impact of using an interpolant for f on the ODE (i.e. the local truncation errors represent the residuals) (iii)represents the numerical solution u tilde. n= 1 n

34 Sketch cont Part 1: we used (i)-(ii) (with T=dt) to estimate L for each time step. - Part 2: we look at (ii)-(iii) ii i L u t O dt + = + ( p 1 = ) n ( ) n= 1 -(iii) T T p p T p n= N ii u T = u + I f u t dt + L u t iii u T u I f u t dt n n= N ( ( )) ( ( n )) ii u T u T = u u + I f u t f u t dt + L u t n= 1

35 Sketch cont We can now try to bound each term: T T p p n= N ( ( )) ( ( n )) u T u T = u u + I f u t f u t dt + L u t n= 1 n= N ( ( )) ( ( n )) u u + I f u t f u t dt + L u t n= 1 Bounded by assumption on Initial conditions (i) Bounded by assumption on consistency. (ii) Estimated by accuracy study. Bounded by assumption on Stability of the Ip interpolation operator

36 Consider: Euler-Forward Solution Accuracy u( ) = u, du = µ u dt 1) the right hand side function is certainly Lipschitz! 2) It is certainly p=1 times differentiable 3) If we keep mu*dt in the absolute stability region (unit circle centered at -1 in the complex plane) then it is stable. 4) By Taylor s theorem it is a p=1 order accurate method. 5) All of the above indicate global solution accuracy which is first order in dt, i.e. the error will be O(dt) as dt

37 AB2 Global Accuracy We already established the stability region for AB2 so as long as we keep mu*dt in that region we have stability. Now we have to consider accuracy: 3 du 1 du Lu t u t u t dt t t 2 dt 2 dt = n n+ 1 n n n du dt d u dt d u* = u + dt + + u 2 3 dt 2! dt 3! dt 3 du 1 du d u dt dt dt 2 dt 2 + dt dt 2! dt d u** 2 3 = + 3! dt 2!2dt dt d u * dt d u ** 3 ( tn ) 3 ( tn ) = O dt AB2 is a 2 nd order acurate method

38 cont If the AB2 stability condition is met then the global accuracy result indicates that for a solution at fixed time the solution will be ( 2 O dt ) Thus we characterize the scheme as 2 nd order method accurate, and 2 nd order solution accurate.

39 AB3 Convergence Expand all terms in the local truncation error about tn 1 du du du Lu t u t u t dt 23 t 16 t 5 t 12 dt dt dt = + n n+ 1 n n n 1 n du dt d u dt d u dt d u* = u + dt u dt 2! dt 3! dt 4! dt dt dt dt dt 2! dt 3! dt 3 4 du d u dt d u dt d u*** + 5 2dt dt dt 2! dt 3! dt du du d u dt d u dt d u** dt dt d u * 2dt d u ** 5dt d u *** = 4 ( tn ) 4 ( tn ) 4 ( tn ) 4! dt 9 + dt 9 dt 4 = O dt The AB3 scheme is 3 rd order accurate as long as the solution is sufficiently smooth.

40 Homework 1 Read chapter 1 of Trefethen s book. Q1) derive AB4 (use Matlab or Mathematica) Q2) plot the stability region for AB4 Q3) indicate which part of the complex plane is inside the AB4 region of absolute stability (you may need to test at least one point in each closed loop of the marginal stability curve) continued on next slide

41 Homework 1 cont Q4a) estimate upper bounds for dt to solve the following with AB2,AB3, and AB4 u = 1 du = µ u dt µ =.1 +.5i t [,5] Q4b) write a code which uses Euler-Forward for the first time step, AB2 for the second time step and AB3 for the third time step (use a suitable dt for AB3 for all time steps) Q4c) use this code and plot the error as a function of time for dt, dt/2, dt/4, dt/8 (i.e. run the simulation 4 times). Q4d) estimate the order of accuracy of the code based on the error at t=5 Q4e) (extra credit) compare the code with an AB3 method started with exact initial data.

Numerical Methods for Partial Differential Equations. CAAM 452 Spring 2005 Lecture 2 Instructor: Tim Warburton

Numerical Methods for Partial Differential Equations. CAAM 452 Spring 2005 Lecture 2 Instructor: Tim Warburton Numerical Methods for Partial Differential Equations CAAM 452 Spring 2005 Lecture 2 Instructor: Tim Warburton Note on textbook for finite difference methods Due to the difficulty some students have experienced

More information

Chapter 5 Exercises. (a) Determine the best possible Lipschitz constant for this function over 2 u <. u (t) = log(u(t)), u(0) = 2.

Chapter 5 Exercises. (a) Determine the best possible Lipschitz constant for this function over 2 u <. u (t) = log(u(t)), u(0) = 2. Chapter 5 Exercises From: Finite Difference Methods for Ordinary and Partial Differential Equations by R. J. LeVeque, SIAM, 2007. http://www.amath.washington.edu/ rjl/fdmbook Exercise 5. (Uniqueness for

More information

Numerical Analysis Preliminary Exam 10 am to 1 pm, August 20, 2018

Numerical Analysis Preliminary Exam 10 am to 1 pm, August 20, 2018 Numerical Analysis Preliminary Exam 1 am to 1 pm, August 2, 218 Instructions. You have three hours to complete this exam. Submit solutions to four (and no more) of the following six problems. Please start

More information

2tdt 1 y = t2 + C y = which implies C = 1 and the solution is y = 1

2tdt 1 y = t2 + C y = which implies C = 1 and the solution is y = 1 Lectures - Week 11 General First Order ODEs & Numerical Methods for IVPs In general, nonlinear problems are much more difficult to solve than linear ones. Unfortunately many phenomena exhibit nonlinear

More information

Linear Multistep Methods I: Adams and BDF Methods

Linear Multistep Methods I: Adams and BDF Methods Linear Multistep Methods I: Adams and BDF Methods Varun Shankar January 1, 016 1 Introduction In our review of 5610 material, we have discussed polynomial interpolation and its application to generating

More information

25 Self-Assessment. 26 Eigenvalue Stability for a Linear ODE. Chapter 7 Eigenvalue Stability

25 Self-Assessment. 26 Eigenvalue Stability for a Linear ODE. Chapter 7 Eigenvalue Stability Chapter 7 Eigenvalue Stability As we have seen, while numerical methods can be convergent, they can still exhibit instabilities as the number of timesteps n increases for finite t. For example, when applying

More information

Preliminary Examination, Numerical Analysis, August 2016

Preliminary Examination, Numerical Analysis, August 2016 Preliminary Examination, Numerical Analysis, August 2016 Instructions: This exam is closed books and notes. The time allowed is three hours and you need to work on any three out of questions 1-4 and any

More information

NUMERICAL SOLUTION OF ODE IVPs. Overview

NUMERICAL SOLUTION OF ODE IVPs. Overview NUMERICAL SOLUTION OF ODE IVPs 1 Quick review of direction fields Overview 2 A reminder about and 3 Important test: Is the ODE initial value problem? 4 Fundamental concepts: Euler s Method 5 Fundamental

More information

Numerical Methods for Partial Differential Equations

Numerical Methods for Partial Differential Equations Numerical Methods for Partial Differential Equations CAAM 45 Spring 005 Lecture 15 Instructor: Tim Warburton Two-Dimensional Advection Recall in one-dimensions we chose an arbitrary section of a pipe.

More information

Numerical Methods for Partial Differential Equations

Numerical Methods for Partial Differential Equations Numerical Methods for Partial Differetial Equatios CAAM 452 Sprig 2005 Lecture 9 Istructor: Tim Warburto Today Chage of pla I will go through i detail how to solve homework 3 step by step. Q1) Build a

More information

Numerical Methods for Partial Differential Equations CAAM 452. Spring 2005

Numerical Methods for Partial Differential Equations CAAM 452. Spring 2005 Numerical Methods for Partial Differential Equations Instructor: Tim Warburton Class Location: Duncan Hall 1046 Class Time: 9:5am to 10:40am Office Hours: 10:45am to noon in DH 301 CAAM 45 Spring 005 Homeworks

More information

MTH 452/552 Homework 3

MTH 452/552 Homework 3 MTH 452/552 Homework 3 Do either 1 or 2. 1. (40 points) [Use of ode113 and ode45] This problem can be solved by a modifying the m-files odesample.m and odesampletest.m available from the author s webpage.

More information

4 Stability analysis of finite-difference methods for ODEs

4 Stability analysis of finite-difference methods for ODEs MATH 337, by T. Lakoba, University of Vermont 36 4 Stability analysis of finite-difference methods for ODEs 4.1 Consistency, stability, and convergence of a numerical method; Main Theorem In this Lecture

More information

Announcements. Topics: Homework:

Announcements. Topics: Homework: Announcements Topics: - sections 7.3 (the definite integral +area), 7.4 (FTC), 7.5 (additional techniques of integration) * Read these sections and study solved examples in your textbook! Homework: - review

More information

Review. Numerical Methods Lecture 22. Prof. Jinbo Bi CSE, UConn

Review. Numerical Methods Lecture 22. Prof. Jinbo Bi CSE, UConn Review Taylor Series and Error Analysis Roots of Equations Linear Algebraic Equations Optimization Numerical Differentiation and Integration Ordinary Differential Equations Partial Differential Equations

More information

Honors Differential Equations

Honors Differential Equations MIT OpenCourseWare http://ocw.mit.edu 18.034 Honors Differential Equations Spring 2009 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms. UNIT III: HIGHER-ORDER

More information

1 The relation between a second order linear ode and a system of two rst order linear odes

1 The relation between a second order linear ode and a system of two rst order linear odes Math 1280 Spring, 2010 1 The relation between a second order linear ode and a system of two rst order linear odes In Chapter 3 of the text you learn to solve some second order linear ode's, such as x 00

More information

Announcements. Topics: Homework:

Announcements. Topics: Homework: Announcements Topics: - sections 7.4 (FTC), 7.5 (additional techniques of integration), 7.6 (applications of integration) * Read these sections and study solved examples in your textbook! Homework: - review

More information

Part IB Numerical Analysis

Part IB Numerical Analysis Part IB Numerical Analysis Definitions Based on lectures by G. Moore Notes taken by Dexter Chua Lent 206 These notes are not endorsed by the lecturers, and I have modified them (often significantly) after

More information

1 Ordinary Differential Equations

1 Ordinary Differential Equations Ordinary Differential Equations.0 Mathematical Background.0. Smoothness Definition. A function f defined on [a, b] is continuous at ξ [a, b] if lim x ξ f(x) = f(ξ). Remark Note that this implies existence

More information

University of British Columbia Math 307, Final

University of British Columbia Math 307, Final 1 University of British Columbia Math 307, Final April 23, 2012 3.30-6.00pm Name: Student Number: Signature: Instructor: Instructions: 1. No notes, books or calculators are allowed. A MATLAB/Octave formula

More information

Physics 116A Solutions to Homework Set #2 Winter 2012

Physics 116A Solutions to Homework Set #2 Winter 2012 Physics 6A Solutions to Homework Set #2 Winter 22. Boas, problem. 23. Transform the series 3 n (n+ (+ n determine the interval of convergence to a power series and First we want to make the replacement

More information

- - - - - - - - - - - - - - - - - - DISCLAIMER - - - - - - - - - - - - - - - - - - General Information: This is a midterm from a previous semester. This means: This midterm contains problems that are of

More information

Lecture Notes to Accompany. Scientific Computing An Introductory Survey. by Michael T. Heath. Chapter 9

Lecture Notes to Accompany. Scientific Computing An Introductory Survey. by Michael T. Heath. Chapter 9 Lecture Notes to Accompany Scientific Computing An Introductory Survey Second Edition by Michael T. Heath Chapter 9 Initial Value Problems for Ordinary Differential Equations Copyright c 2001. Reproduction

More information

2.004 Dynamics and Control II Spring 2008

2.004 Dynamics and Control II Spring 2008 MT OpenCourseWare http://ocw.mit.edu 2.004 Dynamics and Control Spring 2008 or information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms. Reading: ise: Chapter 8 Massachusetts

More information

CS520: numerical ODEs (Ch.2)

CS520: numerical ODEs (Ch.2) .. CS520: numerical ODEs (Ch.2) Uri Ascher Department of Computer Science University of British Columbia ascher@cs.ubc.ca people.cs.ubc.ca/ ascher/520.html Uri Ascher (UBC) CPSC 520: ODEs (Ch. 2) Fall

More information

Math, Numerics, & Programming. (for Mechanical Engineers)

Math, Numerics, & Programming. (for Mechanical Engineers) DRAFT V2. From Math, Numerics, & Programming (for Mechanical Engineers) Masayuki Yano James Douglass Penn George Konidaris Anthony T Patera August 23 MIT 2, 22, 23 The Authors. License: Creative Commons

More information

Complex Analysis MATH 6300 Fall 2013 Homework 4

Complex Analysis MATH 6300 Fall 2013 Homework 4 Complex Analysis MATH 6300 Fall 2013 Homework 4 Due Wednesday, December 11 at 5 PM Note that to get full credit on any problem in this class, you must solve the problems in an efficient and elegant manner,

More information

Euler s Method, Taylor Series Method, Runge Kutta Methods, Multi-Step Methods and Stability.

Euler s Method, Taylor Series Method, Runge Kutta Methods, Multi-Step Methods and Stability. Euler s Method, Taylor Series Method, Runge Kutta Methods, Multi-Step Methods and Stability. REVIEW: We start with the differential equation dy(t) dt = f (t, y(t)) (1.1) y(0) = y 0 This equation can be

More information

CS412: Introduction to Numerical Methods

CS412: Introduction to Numerical Methods CS412: Introduction to Numerical Methods MIDTERM #1 2:30PM - 3:45PM, Tuesday, 03/10/2015 Instructions: This exam is a closed book and closed notes exam, i.e., you are not allowed to consult any textbook,

More information

2 Write down the range of values of α (real) or β (complex) for which the following integrals converge. (i) e z2 dz where {γ : z = se iα, < s < }

2 Write down the range of values of α (real) or β (complex) for which the following integrals converge. (i) e z2 dz where {γ : z = se iα, < s < } Mathematical Tripos Part II Michaelmas term 2007 Further Complex Methods, Examples sheet Dr S.T.C. Siklos Comments and corrections: e-mail to stcs@cam. Sheet with commentary available for supervisors.

More information

Gaussian integrals. Calvin W. Johnson. September 9, The basic Gaussian and its normalization

Gaussian integrals. Calvin W. Johnson. September 9, The basic Gaussian and its normalization Gaussian integrals Calvin W. Johnson September 9, 24 The basic Gaussian and its normalization The Gaussian function or the normal distribution, ep ( α 2), () is a widely used function in physics and mathematical

More information

CS 450 Numerical Analysis. Chapter 9: Initial Value Problems for Ordinary Differential Equations

CS 450 Numerical Analysis. Chapter 9: Initial Value Problems for Ordinary Differential Equations Lecture slides based on the textbook Scientific Computing: An Introductory Survey by Michael T. Heath, copyright c 2018 by the Society for Industrial and Applied Mathematics. http://www.siam.org/books/cl80

More information

Finite Difference and Finite Element Methods

Finite Difference and Finite Element Methods Finite Difference and Finite Element Methods Georgy Gimel farb COMPSCI 369 Computational Science 1 / 39 1 Finite Differences Difference Equations 3 Finite Difference Methods: Euler FDMs 4 Finite Element

More information

Finite Differences for Differential Equations 28 PART II. Finite Difference Methods for Differential Equations

Finite Differences for Differential Equations 28 PART II. Finite Difference Methods for Differential Equations Finite Differences for Differential Equations 28 PART II Finite Difference Methods for Differential Equations Finite Differences for Differential Equations 29 BOUNDARY VALUE PROBLEMS (I) Solving a TWO

More information

Math 660 Lecture 4: FDM for evolutionary equations: ODE solvers

Math 660 Lecture 4: FDM for evolutionary equations: ODE solvers Math 660 Lecture 4: FDM for evolutionary equations: ODE solvers Consider the ODE u (t) = f(t, u(t)), u(0) = u 0, where u could be a vector valued function. Any ODE can be reduced to a first order system,

More information

MATH 350: Introduction to Computational Mathematics

MATH 350: Introduction to Computational Mathematics MATH 350: Introduction to Computational Mathematics Chapter V: Least Squares Problems Greg Fasshauer Department of Applied Mathematics Illinois Institute of Technology Spring 2011 fasshauer@iit.edu MATH

More information

Function approximation

Function approximation Week 9: Monday, Mar 26 Function approximation A common task in scientific computing is to approximate a function. The approximated function might be available only through tabulated data, or it may be

More information

Numerical Methods - Initial Value Problems for ODEs

Numerical Methods - Initial Value Problems for ODEs Numerical Methods - Initial Value Problems for ODEs Y. K. Goh Universiti Tunku Abdul Rahman 2013 Y. K. Goh (UTAR) Numerical Methods - Initial Value Problems for ODEs 2013 1 / 43 Outline 1 Initial Value

More information

Polynomial Interpolation Part II

Polynomial Interpolation Part II Polynomial Interpolation Part II Prof. Dr. Florian Rupp German University of Technology in Oman (GUtech) Introduction to Numerical Methods for ENG & CS (Mathematics IV) Spring Term 2016 Exercise Session

More information

N-CN Complex Cube and Fourth Roots of 1

N-CN Complex Cube and Fourth Roots of 1 N-CN Complex Cube and Fourth Roots of 1 Task For each odd positive integer, the only real number solution to is while for even positive integers n, x = 1 and x = 1 are solutions to x n = 1. In this problem

More information

Math 273 (51) - Final

Math 273 (51) - Final Name: Id #: Math 273 (5) - Final Autumn Quarter 26 Thursday, December 8, 26-6: to 8: Instructions: Prob. Points Score possible 25 2 25 3 25 TOTAL 75 Read each problem carefully. Write legibly. Show all

More information

Here s the Graph of the Derivative. Tell me About the Function.

Here s the Graph of the Derivative. Tell me About the Function. Here s the Graph of the Derivative. Tell me About the Function. Presented by Lin McMullin Using its derivatives to determine information about the graph of a function is a standard calculus topic and has

More information

Euler s Method, cont d

Euler s Method, cont d Jim Lambers MAT 461/561 Spring Semester 009-10 Lecture 3 Notes These notes correspond to Sections 5. and 5.4 in the text. Euler s Method, cont d We conclude our discussion of Euler s method with an example

More information

Initial value problems for ordinary differential equations

Initial value problems for ordinary differential equations Initial value problems for ordinary differential equations Xiaojing Ye, Math & Stat, Georgia State University Spring 2019 Numerical Analysis II Xiaojing Ye, Math & Stat, Georgia State University 1 IVP

More information

Modern Navigation. Thomas Herring

Modern Navigation. Thomas Herring 12.215 Modern Navigation Thomas Herring Review of last Class Review of linear Algebra. Class will be based on the book Linear Algebra, Geodesy, and GPS, G. Strang and K. Borre, Wellesley-Cambridge Press,

More information

Section 2.1 (First Order) Linear DEs; Method of Integrating Factors. General first order linear DEs Standard Form; y'(t) + p(t) y = g(t)

Section 2.1 (First Order) Linear DEs; Method of Integrating Factors. General first order linear DEs Standard Form; y'(t) + p(t) y = g(t) Section 2.1 (First Order) Linear DEs; Method of Integrating Factors Key Terms/Ideas: General first order linear DEs Standard Form; y'(t) + p(t) y = g(t) Integrating factor; a function μ(t) that transforms

More information

ODE Homework 1. Due Wed. 19 August 2009; At the beginning of the class

ODE Homework 1. Due Wed. 19 August 2009; At the beginning of the class ODE Homework Due Wed. 9 August 2009; At the beginning of the class. (a) Solve Lẏ + Ry = E sin(ωt) with y(0) = k () L, R, E, ω are positive constants. (b) What is the limit of the solution as ω 0? (c) Is

More information

Experiment 2. F r e e F a l l

Experiment 2. F r e e F a l l Suggested Reading for this Lab Experiment F r e e F a l l Taylor, Section.6, and standard deviation rule in Taylor handout. Review Chapters 3 & 4, Read Sections 8.1-8.6. You will also need some procedures

More information

Parameterizing the Pythagorean Triples

Parameterizing the Pythagorean Triples Parameterizing the Pythagorean Triples Saratoga Math Club January 8, 204 Introduction To ask for one or two Pythagorean triples is not a hard nor deep question at all. Common examples, like (3, 4, 5),

More information

MAT137 Calculus! Lecture 45

MAT137 Calculus! Lecture 45 official website http://uoft.me/mat137 MAT137 Calculus! Lecture 45 Today: Taylor Polynomials Taylor Series Next: Taylor Series Power Series Definition (Power Series) A power series is a series of the form

More information

HOMEWORK 2 SOLUTIONS

HOMEWORK 2 SOLUTIONS HOMEWORK SOLUTIONS MA11: ADVANCED CALCULUS, HILARY 17 (1) Find parametric equations for the tangent line of the graph of r(t) = (t, t + 1, /t) when t = 1. Solution: A point on this line is r(1) = (1,,

More information

ECE 486 Control Systems

ECE 486 Control Systems ECE 486 Control Systems Spring 208 Midterm #2 Information Issued: April 5, 208 Updated: April 8, 208 ˆ This document is an info sheet about the second exam of ECE 486, Spring 208. ˆ Please read the following

More information

CHAPTER 5: Linear Multistep Methods

CHAPTER 5: Linear Multistep Methods CHAPTER 5: Linear Multistep Methods Multistep: use information from many steps Higher order possible with fewer function evaluations than with RK. Convenient error estimates. Changing stepsize or order

More information

dt 2 The Order of a differential equation is the order of the highest derivative that occurs in the equation. Example The differential equation

dt 2 The Order of a differential equation is the order of the highest derivative that occurs in the equation. Example The differential equation Lecture 18 : Direction Fields and Euler s Method A Differential Equation is an equation relating an unknown function and one or more of its derivatives. Examples Population growth : dp dp = kp, or = kp

More information

Course Information Course Overview Study Skills Background Material. Introduction. CS 205A: Mathematical Methods for Robotics, Vision, and Graphics

Course Information Course Overview Study Skills Background Material. Introduction. CS 205A: Mathematical Methods for Robotics, Vision, and Graphics Introduction CS 205A: Mathematical Methods for Robotics, Vision, and Graphics Doug James CS 205A: Mathematical Methods Introduction 1 / 16 Instructor Prof. Doug James Office: Gates 363 Telephone: (650)

More information

MATH141: Calculus II Exam #4 7/21/2017 Page 1

MATH141: Calculus II Exam #4 7/21/2017 Page 1 MATH141: Calculus II Exam #4 7/21/2017 Page 1 Write legibly and show all work. No partial credit can be given for an unjustified, incorrect answer. Put your name in the top right corner and sign the honor

More information

Fourth Order RK-Method

Fourth Order RK-Method Fourth Order RK-Method The most commonly used method is Runge-Kutta fourth order method. The fourth order RK-method is y i+1 = y i + 1 6 (k 1 + 2k 2 + 2k 3 + k 4 ), Ordinary Differential Equations (ODE)

More information

Numerical Integration (Quadrature) Another application for our interpolation tools!

Numerical Integration (Quadrature) Another application for our interpolation tools! Numerical Integration (Quadrature) Another application for our interpolation tools! Integration: Area under a curve Curve = data or function Integrating data Finite number of data points spacing specified

More information

Chapter 3. Measuring data

Chapter 3. Measuring data Chapter 3 Measuring data 1 Measuring data versus presenting data We present data to help us draw meaning from it But pictures of data are subjective They re also not susceptible to rigorous inference Measuring

More information

multistep methods Last modified: November 28, 2017 Recall that we are interested in the numerical solution of the initial value problem (IVP):

multistep methods Last modified: November 28, 2017 Recall that we are interested in the numerical solution of the initial value problem (IVP): MATH 351 Fall 217 multistep methods http://www.phys.uconn.edu/ rozman/courses/m351_17f/ Last modified: November 28, 217 Recall that we are interested in the numerical solution of the initial value problem

More information

Math Real Analysis II

Math Real Analysis II Math 432 - Real Analysis II Solutions to Homework due February 3 In class, we learned that the n-th remainder for a smooth function f(x) defined on some open interval containing is given by f (k) () R

More information

School of Mechanical Engineering Purdue University. DC Motor Position Control The block diagram for position control of the servo table is given by:

School of Mechanical Engineering Purdue University. DC Motor Position Control The block diagram for position control of the servo table is given by: Root Locus Motivation Sketching Root Locus Examples ME375 Root Locus - 1 Servo Table Example DC Motor Position Control The block diagram for position control of the servo table is given by: θ D 0.09 See

More information

CS 450 Numerical Analysis. Chapter 8: Numerical Integration and Differentiation

CS 450 Numerical Analysis. Chapter 8: Numerical Integration and Differentiation Lecture slides based on the textbook Scientific Computing: An Introductory Survey by Michael T. Heath, copyright c 2018 by the Society for Industrial and Applied Mathematics. http://www.siam.org/books/cl80

More information

- - - - - - - - - - - - - - - - - - DISCLAIMER - - - - - - - - - - - - - - - - - - General Information: This midterm is a sample midterm. This means: The sample midterm contains problems that are of similar,

More information

On linear and non-linear equations. (Sect. 1.6).

On linear and non-linear equations. (Sect. 1.6). On linear and non-linear equations. (Sect. 1.6). Review: Linear differential equations. Non-linear differential equations. The Picard-Lindelöf Theorem. Properties of solutions to non-linear ODE. The Proof

More information

SOLUTIONS TO HOMEWORK ASSIGNMENT #2, Math 253

SOLUTIONS TO HOMEWORK ASSIGNMENT #2, Math 253 SOLUTIONS TO HOMEWORK ASSIGNMENT #, Math 5. Find the equation of a sphere if one of its diameters has end points (, 0, 5) and (5, 4, 7). The length of the diameter is (5 ) + ( 4 0) + (7 5) = =, so the

More information

The Method of Undetermined Coefficients.

The Method of Undetermined Coefficients. The Method of Undetermined Coefficients. James K. Peterson Department of Biological Sciences and Department of Mathematical Sciences Clemson University May 24, 2017 Outline 1 Annihilators 2 Finding The

More information

Exponential integrators and functions of the matrix exponential

Exponential integrators and functions of the matrix exponential Exponential integrators and functions of the matrix exponential Paul Matthews, Stephen Cox, Hala Ashi and Linda Cummings School of Mathematical Sciences, University of Nottingham, UK Introduction to exponential

More information

Spring 2011 solutions. We solve this via integration by parts with u = x 2 du = 2xdx. This is another integration by parts with u = x du = dx and

Spring 2011 solutions. We solve this via integration by parts with u = x 2 du = 2xdx. This is another integration by parts with u = x du = dx and Math - 8 Rahman Final Eam Practice Problems () We use disks to solve this, Spring solutions V π (e ) d π e d. We solve this via integration by parts with u du d and dv e d v e /, V π e π e d. This is another

More information

Module 3F2: Systems and Control EXAMPLES PAPER 2 ROOT-LOCUS. Solutions

Module 3F2: Systems and Control EXAMPLES PAPER 2 ROOT-LOCUS. Solutions Cambridge University Engineering Dept. Third Year Module 3F: Systems and Control EXAMPLES PAPER ROOT-LOCUS Solutions. (a) For the system L(s) = (s + a)(s + b) (a, b both real) show that the root-locus

More information

Applied Math for Engineers

Applied Math for Engineers Applied Math for Engineers Ming Zhong Lecture 15 March 28, 2018 Ming Zhong (JHU) AMS Spring 2018 1 / 28 Recap Table of Contents 1 Recap 2 Numerical ODEs: Single Step Methods 3 Multistep Methods 4 Method

More information

Introduction to Initial Value Problems

Introduction to Initial Value Problems Chapter 2 Introduction to Initial Value Problems The purpose of this chapter is to study the simplest numerical methods for approximating the solution to a first order initial value problem (IVP). Because

More information

Introduction to the Numerical Solution of IVP for ODE

Introduction to the Numerical Solution of IVP for ODE Introduction to the Numerical Solution of IVP for ODE 45 Introduction to the Numerical Solution of IVP for ODE Consider the IVP: DE x = f(t, x), IC x(a) = x a. For simplicity, we will assume here that

More information

Non-polynomial Least-squares fitting

Non-polynomial Least-squares fitting Applied Math 205 Last time: piecewise polynomial interpolation, least-squares fitting Today: underdetermined least squares, nonlinear least squares Homework 1 (and subsequent homeworks) have several parts

More information

Lecture 21: Numerical Solution of Differential Equations

Lecture 21: Numerical Solution of Differential Equations cs41: introduction to numerical analysis 11/30/10 Lecture 1: Numerical Solution of Differential Equations Instructor: Professor Amos Ron Scribes: Yunpeng Li, Mark Cowlishaw, Nathanael Fillmore 1 Introduction

More information

Lecture 20: Lagrange Interpolation and Neville s Algorithm. for I will pass through thee, saith the LORD. Amos 5:17

Lecture 20: Lagrange Interpolation and Neville s Algorithm. for I will pass through thee, saith the LORD. Amos 5:17 Lecture 20: Lagrange Interpolation and Neville s Algorithm for I will pass through thee, saith the LORD. Amos 5:17 1. Introduction Perhaps the easiest way to describe a shape is to select some points on

More information

Final Examination. CS 205A: Mathematical Methods for Robotics, Vision, and Graphics (Fall 2013), Stanford University

Final Examination. CS 205A: Mathematical Methods for Robotics, Vision, and Graphics (Fall 2013), Stanford University Final Examination CS 205A: Mathematical Methods for Robotics, Vision, and Graphics (Fall 2013), Stanford University The exam runs for 3 hours. The exam contains eight problems. You must complete the first

More information

Spring, 2012 CIS 515. Fundamentals of Linear Algebra and Optimization Jean Gallier

Spring, 2012 CIS 515. Fundamentals of Linear Algebra and Optimization Jean Gallier Spring 0 CIS 55 Fundamentals of Linear Algebra and Optimization Jean Gallier Homework 5 & 6 + Project 3 & 4 Note: Problems B and B6 are for extra credit April 7 0; Due May 7 0 Problem B (0 pts) Let A be

More information

21 Linear State-Space Representations

21 Linear State-Space Representations ME 132, Spring 25, UC Berkeley, A Packard 187 21 Linear State-Space Representations First, let s describe the most general type of dynamic system that we will consider/encounter in this class Systems may

More information

Scientific Computing: An Introductory Survey

Scientific Computing: An Introductory Survey Scientific Computing: An Introductory Survey Chapter 9 Initial Value Problems for Ordinary Differential Equations Prof. Michael T. Heath Department of Computer Science University of Illinois at Urbana-Champaign

More information

Partial Fraction Decomposition Honors Precalculus Mr. Velazquez Rm. 254

Partial Fraction Decomposition Honors Precalculus Mr. Velazquez Rm. 254 Partial Fraction Decomposition Honors Precalculus Mr. Velazquez Rm. 254 Adding and Subtracting Rational Expressions Recall that we can use multiplication and common denominators to write a sum or difference

More information

Review Higher Order methods Multistep methods Summary HIGHER ORDER METHODS. P.V. Johnson. School of Mathematics. Semester

Review Higher Order methods Multistep methods Summary HIGHER ORDER METHODS. P.V. Johnson. School of Mathematics. Semester HIGHER ORDER METHODS School of Mathematics Semester 1 2008 OUTLINE 1 REVIEW 2 HIGHER ORDER METHODS 3 MULTISTEP METHODS 4 SUMMARY OUTLINE 1 REVIEW 2 HIGHER ORDER METHODS 3 MULTISTEP METHODS 4 SUMMARY OUTLINE

More information

Polynomial and Synthetic Division

Polynomial and Synthetic Division Polynomial and Synthetic Division Polynomial Division Polynomial Division is very similar to long division. Example: 3x 3 5x 3x 10x 1 3 Polynomial Division 3x 1 x 3x 3 3 x 5x 3x x 6x 4 10x 10x 7 3 x 1

More information

DON T PANIC! If you get stuck, take a deep breath and go on to the next question. Come back to the question you left if you have time at the end.

DON T PANIC! If you get stuck, take a deep breath and go on to the next question. Come back to the question you left if you have time at the end. Math 307, Midterm 2 Winter 2013 Name: Instructions. DON T PANIC! If you get stuck, take a deep breath and go on to the next question. Come back to the question you left if you have time at the end. There

More information

[Refer Slide Time: 00:45]

[Refer Slide Time: 00:45] Differential Calculus of Several Variables Professor: Sudipta Dutta Department of Mathematics and Statistics Indian Institute of Technology, Kanpur Module 1 Lecture No 1 Introduction of Functions of Several

More information

2. Higher-order Linear ODE s

2. Higher-order Linear ODE s 2. Higher-order Linear ODE s 2A. Second-order Linear ODE s: General Properties 2A-1. On the right below is an abbreviated form of the ODE on the left: (*) y + p()y + q()y = r() Ly = r() ; where L is the

More information

1 Lesson 13: Methods of Integration

1 Lesson 13: Methods of Integration Lesson 3: Methods of Integration Chapter 6 Material: pages 273-294 in the textbook: Lesson 3 reviews integration by parts and presents integration via partial fraction decomposition as the third of the

More information

Chapter Five Notes N P U2C5

Chapter Five Notes N P U2C5 Chapter Five Notes N P UC5 Name Period Section 5.: Linear and Quadratic Functions with Modeling In every math class you have had since algebra you have worked with equations. Most of those equations have

More information

Ordinary differential equations - Initial value problems

Ordinary differential equations - Initial value problems Education has produced a vast population able to read but unable to distinguish what is worth reading. G.M. TREVELYAN Chapter 6 Ordinary differential equations - Initial value problems In this chapter

More information

ACM/CMS 107 Linear Analysis & Applications Fall 2016 Assignment 4: Linear ODEs and Control Theory Due: 5th December 2016

ACM/CMS 107 Linear Analysis & Applications Fall 2016 Assignment 4: Linear ODEs and Control Theory Due: 5th December 2016 ACM/CMS 17 Linear Analysis & Applications Fall 216 Assignment 4: Linear ODEs and Control Theory Due: 5th December 216 Introduction Systems of ordinary differential equations (ODEs) can be used to describe

More information

The Accelerator Hamiltonian in a Straight Coordinate System

The Accelerator Hamiltonian in a Straight Coordinate System Hamiltoninan Dynamics for Particle Accelerators, Lecture 2 The Accelerator Hamiltonian in a Straight Coordinate System Andy Wolski University of Liverpool, and the Cockcroft Institute, Daresbury, UK. Given

More information

Math Numerical Analysis Homework #4 Due End of term. y = 2y t 3y2 t 3, 1 t 2, y(1) = 1. n=(b-a)/h+1; % the number of steps we need to take

Math Numerical Analysis Homework #4 Due End of term. y = 2y t 3y2 t 3, 1 t 2, y(1) = 1. n=(b-a)/h+1; % the number of steps we need to take Math 32 - Numerical Analysis Homework #4 Due End of term Note: In the following y i is approximation of y(t i ) and f i is f(t i,y i ).. Consider the initial value problem, y = 2y t 3y2 t 3, t 2, y() =.

More information

University of British Columbia Math 307, Final

University of British Columbia Math 307, Final 1 University of British Columbia Math 37, Final April 29, 214 12.-2.3pm Name: Student Number: Signature: Instructor: Instructions: 1. No notes, books or calculators are allowed. A MATLAB/Octave formula

More information

n 1 f n 1 c 1 n+1 = c 1 n $ c 1 n 1. After taking logs, this becomes

n 1 f n 1 c 1 n+1 = c 1 n $ c 1 n 1. After taking logs, this becomes Root finding: 1 a The points {x n+1, }, {x n, f n }, {x n 1, f n 1 } should be co-linear Say they lie on the line x + y = This gives the relations x n+1 + = x n +f n = x n 1 +f n 1 = Eliminating α and

More information

The Euler Method for the Initial Value Problem

The Euler Method for the Initial Value Problem The Euler Method for the Initial Value Problem http://people.sc.fsu.edu/ jburkardt/isc/week10 lecture 18.pdf... ISC3313: Introduction to Scientific Computing with C++ Summer Semester 2011... John Burkardt

More information

= 1 2 x (x 1) + 1 {x} (1 {x}). [t] dt = 1 x (x 1) + O (1), [t] dt = 1 2 x2 + O (x), (where the error is not now zero when x is an integer.

= 1 2 x (x 1) + 1 {x} (1 {x}). [t] dt = 1 x (x 1) + O (1), [t] dt = 1 2 x2 + O (x), (where the error is not now zero when x is an integer. Problem Sheet,. i) Draw the graphs for [] and {}. ii) Show that for α R, α+ α [t] dt = α and α+ α {t} dt =. Hint Split these integrals at the integer which must lie in any interval of length, such as [α,

More information

MATH 307 Introduction to Differential Equations Autumn 2017 Midterm Exam Monday November

MATH 307 Introduction to Differential Equations Autumn 2017 Midterm Exam Monday November MATH 307 Introduction to Differential Equations Autumn 2017 Midterm Exam Monday November 6 2017 Name: Student ID Number: I understand it is against the rules to cheat or engage in other academic misconduct

More information

Numerical Methods for Differential Equations Mathematical and Computational Tools

Numerical Methods for Differential Equations Mathematical and Computational Tools Numerical Methods for Differential Equations Mathematical and Computational Tools Gustaf Söderlind Numerical Analysis, Lund University Contents V4.16 Part 1. Vector norms, matrix norms and logarithmic

More information

Second Order ODEs. CSCC51H- Numerical Approx, Int and ODEs p.130/177

Second Order ODEs. CSCC51H- Numerical Approx, Int and ODEs p.130/177 Second Order ODEs Often physical or biological systems are best described by second or higher-order ODEs. That is, second or higher order derivatives appear in the mathematical model of the system. For

More information