0 Up votes0 Down votes

7 views14 pagesboo

Dec 18, 2013

© Attribution Non-Commercial (BY-NC)

PDF, TXT or read online from Scribd

boo

Attribution Non-Commercial (BY-NC)

7 views

boo

Attribution Non-Commercial (BY-NC)

- The Law of Explosive Growth: Lesson 20 from The 21 Irrefutable Laws of Leadership
- Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race
- Hidden Figures Young Readers' Edition
- The E-Myth Revisited: Why Most Small Businesses Don't Work and
- Micro: A Novel
- The Wright Brothers
- The Other Einstein: A Novel
- State of Fear
- State of Fear
- The Power of Discipline: 7 Ways it Can Change Your Life
- The Kiss Quotient: A Novel
- The 10X Rule: The Only Difference Between Success and Failure
- Being Wrong: Adventures in the Margin of Error
- Algorithms to Live By: The Computer Science of Human Decisions
- The 6th Extinction
- The Black Swan
- The Art of Thinking Clearly
- The Last Battle
- Prince Caspian
- A Mind for Numbers: How to Excel at Math and Science Even If You Flunked Algebra

You are on page 1of 14

Introduction (a) Lecture Outline : (9.1) (9.2) (9.3) (9.4) Linear, Autonomous Dierential Equation Systems. The Substitution Method. The Direct Method. Stability Analysis.

(a) So far in this course we have focused specically at learning the techniques required to solve a single dierential equation. (b) However, in economics it is common to be faced with problems which require two or more variables to be determined simultaneously. (c) This lecture, we extend the techniques for solving single dierential equations, to solve systems of autonomous dierential equations. (d) In this course our focus will be only on solving dierential equation systems (and thus due to time constraints we will not consider the solution of systems of dierence equations). (e) In addition we will restrict our attention to solving only linear, autonomous, dierential equation systems (and thus not consider the solution of systems of non-linear dierential equations) (f) There are two techniques commonly used to solve linear dierential equation systems: 1 Substitution Method: which is suited only for solving a dierential equation system consisting of exactly two linear dierential equations. 2 Direct Method: which can be used to solve a linear dierential equation system with more than two equations.

(a) A linear system, of two autonomous dierential equations is expressed as: y 1 = a11 y1 + a12 y2 + b1 y 2 = a21 y1 + a22 y2 + b2

(b) The above equations must be solved simultaneously because y 1 depends on the solution for y2 and y 2 depends on the solution for y1 . (c) Just like solving a single dierential equation, we separate the problem of nding the solutions into two parts:

p h y1 = y1 + y1 p h y2 = y2 + y2

where y h is the homogenous solution and y p is the particular solution. (d) The General Solution to the Homogenous Forms To nd the homogenous solutions we set the constants b1 and b2 equal to zero, to obtain: y 1 = a11 y1 + a12 y2 y 2 = a21 y1 + a22 y2 (1) (2)

To solve equations (1) and (2) we want convert this system into a single second-order dierential equation and then we can use the techniques set-out in Lecture 8 to easily solve a linear, autonomous, second-order dierential equation!

Step 1: Dierentiate equation (1) to obtain: y 1 = a11 y 1 + a12 y 2 which gives us a second-order dierential equation. Step 2: Use equation (2) to substitute out y 2 from equation (3). y 1 = a11 y 1 + a12 (a21 y1 + a22 y2 ) (4) (3)

Step 4: Use equation (5) to substitute out y2 from equation (4) y 1 = a11 y 1 + a12 a21 y1 + a22 y 1 a11 y1 a12

y 1 (a11 + a22 )y 1 + (a11 a22 a12 a21 )y1 = 0 which is a linear, homogenous, second-order dierential equation! Therefore we can use the solution techniques for linear, autonomous second-order dierential equations to nd the homogenous forms of this two equation system. y + a1 y + a2 y = 0 where a1 (a11 + a22 ) a2 (a11 a22 a12 a21 ) y y 1 y y 1

4

The characteristic equation is: r2 + a1 r + a2 = 0 and the roots are given by: r1 , r 2 = a1 a2 1 4a2 2

If the roots are real and distinct then we know from Lecture 8 that h the solution for y1 must be:

h y1 (t) = C1 er1 t + C2 er2 t

(6)

h To nd the solution for y2 (t) we take the derivative of equation (6) which yields:

y 1 = r1 C1 er1 t + r2 C2 er2 t

h y2 (t) =

Theorem 1: The solutions to the homogenous form of the system of two linear, autonomous rst-order dierential equations: y 1 = a11 y1 + a12 y2 + b1 y 2 = a21 y1 + a22 y2 + b2 are: Real and Distinct roots ((a11 + a22 )2 4(a11 a22 a12 a21 ) > 0)

h y1 (t) = C1 er1 t + C2 er2 t h y2 (t) =

Real and Equal roots ((a11 + a22 )2 4(a11 a22 a12 a21 ) = 0)

h (t) = C1 ert + C2 tert y1 h y2 (t) =

h y1 (t) = eht [A1 cos(vt) + A2 sin(vt)] h y2 (t) = eht

(h a11 )A2 vA1 (h a11 )A1 + vA2 cos(vt) + sin(vt) a12 a12 1 h = (a11 + a22 ) 2 v= 1 2 4(a11 a22 a12 a21 )2 (a11 + a22 )2

6

where

Example: Solve the following system of homogenous dierential equations: y 1 = y1 3y2 1 y 2 = y1 + 3y2 4 (e) The Particular Solutions As always the particular solutions we are interested in are the steady state solutions. The steady-state solution to a system of two dierential equations 1 and y 2 are both equal to is the pair of values y 1 and y 2 at which y zero.

y 1 = a11 y1 + a12 y2 + b1 y 2 = a21 y1 + a22 y2 + b2 Set y 1 = y 2 = 0. This gives a linear system of two equations for two unknowns y 1 and y 2 . a11 y 1 + a12 y 2 + b1 = 0 a21 y 1 + a22 y 2 + b2 = 0

Re-arrange these two equations for y 1 and y 2 respectively: y1 = y2 = a12 b1 y2 a11 a11 a21 b2 y1 a22 a22 (7) (8)

Use equation (7) to eliminate y 1 from equation (8): y2 = a21 a22 a12 b1 y2 a11 a11 b2 a22

which after re-arranging yields the steady state solution y 2 y2 = a21 b1 a11 b2 a11 a22 a12 a21 (9)

Substituting equation (9) into equation (7) yields the steady state solution y 1 a12 b2 a22 b1 (10) y1 = a11 a22 a12 a21 Therefore the particular solutions are: a12 b2 a22 b1 a11 a22 a12 a21 a21 b1 a11 b2 p y2 = y2 = a11 a22 a12 a21

p y1 = y1 =

Example: Find the particular solutions for: y 1 = y1 3y2 + 1 1 y 2 = y1 y2 + 2 4 (f) Complete Solution: The complete solution to the system of two linear autonomous dierential equations is the sum of the homogenous solutions and the particular solutions. Hoy p.939, provides the complete solutions for the three types of characteristic roots that can occur.

(g) Initial conditions: When the complete solution must also satisfy given initial conditions, the values of the constants of integration must be set appropriately. Example: Find the constants of integration that make the solutions to the following system of dierential equations satisfy the initial conditions y1 (0) = 1 and y2 (0) = 3. y 1 = y1 3y2 5 1 y 2 = y1 + 3y2 5 4 The homogenous solutions are:

h y1 (t) = C1 e1.5t + C2 e2.5t 1 1 h y2 (t) = C1 e1.5t C2 e2.5t 6 2

h y1 (t) = C1 e1.5t + C2 e2.5t + 8 1 1 h y2 (t) = C1 e1.5t C2 e2.5t + 1 6 2

Evaluate the complete solutions using the initial conditions at t = 0 to obtain: C1 = 7 C2 2 C2 = C1 4 6 Thus we have two equations for two unknown C1 and C2 . Solving

9

yields: 9 2 5 C2 = 2 C1 = Thus the complete solutions that satisfy the initial conditions become: 9 y1 (t) = e1.5t 2 3 y2 (t) = e1.5t + 4 5 2.5t e +8 2 5 2.5t e +1 4

10

(a) Although the substitution approach works well for systems of two differential equations, it becomes cumbersome for larger systems. The following direct approach to solving a system of linear dierential equations circumvents this limitation. (b) A linear system of n autonomous dierential equations is expressed in matrix form as: y = Ay + b where A is an n n matrix of constant coecients, b is a vector of n constant terms, y is a vector of n variables and y is a vector of n derivatives. (c) e.g. in the case of a system of n = 2 linear, autonomous dierential equations, the matrix of coecients is A= a11 a12 a21 a22

(d) As before the solution to the complete system of equations is obtained by adding together the homogenous solutions and the particular solutions.

11

(e) Homogenous Solutions y = Ay We proceed by guessing that the homogenous solution are of the form: y = kert where k is an n-dimensional vector of constants and r is a scalar. To see if this guess is correct, check that the guessed solution and its rst derivative satisfy the dierential equation system. The derivative of the proposed solution is: y = rkert Substitution of these derivatives and the proposed solutions into the original system of equations gives: rkert = Akert [A rI ]k = 0 (11)

where I is an identity matrix and 0 is the zero vector. We wish to nd values of r that solve this equation and thus make our guessed solution correct. The characteristic equation is | A rI |= 0 where r is the characteristic roots or eigenvalues of matrix A. A nonzero vector k1 which is a solution to equation (11) for a particular eigenvalue r1 is called the eigenvector of the matrix A corresponding to the eigenvalue r1 .

12

Therefore if n = 2 then: | A rI |= which after simplifying gives r2 (a11 + a22 )r + (a11 a22 a12 a21 ) = 0 which is of course the same characteristic equation obtained from the substitution method. Since | A rI | is constrained to be singular there is an innite number of eigenvectors that will satisfy the equation. We can either normalize the equation by choosing a vector whose length is unity 2 2 = 1 which is called the Euclidian distance condition, or we + k2 k1 can simply pick any arbitrary value for one element while maintaining the relationship between elements. Example: Solve the following system of homogenous dierential equations using the direct method: y 1 = y1 3y2 1 y 2 = y1 + 3y2 4 (f) The particular solutions The steady-state solutions provide the particular solutions we require. Set y = 0 and thus: Ay + b = 0 y = A1 b provided the inverse matrix A1 exists (i.e. the determinant of A must be nonzero).

13

(a) As always we interested in knowing if the system of dierential equations converges to the steady state solutions. (b) The steady state solution of a system of linear, autonomous dierential equations is asymptotically stable if and only if the characteristic roots are negative (the real part is negative in the case of complex-valued roots). (c) If one of the characteristic roots is positive and the other is negative, the steady-state equilibrium is called a saddle-point equilibrium, which is unstable. However y1 (t) and y2 (t) converge to their steady state solutions if the initial conditions for y1 and y2 satisfy the following equation: r1 a11 (y1 y 1 ) + y 2 a12 where r1 is the negative root and r2 is the positive root. The locus of points (y1 , y2 ) dened by this equation is known as the saddle path. y2 =

14

- Face Recognition Using Laplacian FacesUploaded byEditor IJRITCC
- Syllabus Math 215 Introduction to Linear AlgebraUploaded bypwan
- eigenstuffUploaded byBeatrice Firan
- A Study of the Eigenvalue Analysis Capabilities OfUploaded byNedim Hadžiaganović
- Modal parameter identification based on ARMAV and state–space approachesUploaded byPedro Motta Nunes
- 9231_s11_qp_11Uploaded byJohn
- New EnglandUploaded byAmel Zerigui
- Post-Buckling Behavior of Laminated Composite Cylindrical Shells Subjected to Axial, Bending and Torsion LoadsUploaded byShell Buckling
- 805x_chapter03v2Uploaded byMateo Vicente
- Independent Component AnalysisUploaded bydewddude
- Scilab CoseUploaded byTanveer Hocane
- 4062402049Summerhays, 2002Uploaded byJon Baldwin
- notes_labUploaded byJose Lorenzo Trujillo
- Systems of Differential Equations-TRANSPARENCYUploaded byHakim Ramli
- MH Eigensystems(Fall2007 Chap6)Uploaded bybhishan
- CSD Project ReportUploaded byAkash
- AAOC_C341_582_C_2008_2Uploaded byDileep Reddy
- Eigenvalue&vector.xlsUploaded byViacheslav Baykov
- Principle Component AnalysisUploaded bymatthewriley123
- adeli2000Uploaded byJeanOscorimaCelis
- The Comparison of Signature Verification Result Using 2DPCA Method and SSE MethodUploaded byAnitta Sindar
- Maple Hand BokenUploaded byLilian Pereira
- new maths 2.pdfUploaded byakumar
- Linear Algebra 1Uploaded byPurvesh Kumar
- benaichUploaded byhamidf
- Shaft FEM.pdfUploaded bynils4586
- t.srinivasa RaoUploaded bysrinivas tumati
- D034017022.pdfUploaded bymario olivares
- Course 2Uploaded byJohnny
- On Kinematic Stability of Double-layer Grid Space StructuresUploaded bydiogomalufgomes

- Feeling Good the New Mood Therapy Revised and Updated 609896Uploaded bymailmado
- CRC ProblemUploaded bymailmado
- CN Assignment-1 - CopyUploaded bymailmado
- GIN's RC TechniqueUploaded bydinmilo
- Cyclic Redundancy Check (CRC)Uploaded byShaveta Banda
- Theory of Electricity -Professor J R Lucas- Level 2-EE201_10_non_sinusoidal_part_2Uploaded byMahmoud Basho
- eeUploaded bySukanya Suparna
- w7Uploaded bymailmado
- RC passage gprepUploaded byrrg924
- 1566-1563-1-PBUploaded byKshitij Kumar
- PidUploaded bymailmado
- hard diskUploaded bymailmado
- CSE3213 13 RandomAccess 2 F2010Uploaded bymailmado
- VGomathy_IJACCS_Mar14Uploaded bymailmado
- Research on Model of Circuit Fault ClassificationUploaded bymailmado
- Leapfrog Filer CircuitUploaded bymailmado
- CDMA_tutorial2Uploaded byDramane Bonkoungou
- PID Without PhdUploaded byopenid_9mkewy6L
- IntroductionUploaded bymailmado
- Rh CriterionUploaded bymailmado
- FirnewUploaded bymailmado
- FirUploaded bymailmado
- Data Communication - CS601 Spring 2010 Mid Term Solved McqsUploaded bymailmado
- HammingUploaded byamrifa
- Cyclic Redundancy ChecksUploaded bySherry
- 12192012033245jhjhUploaded byPIYUSHNIT
- Bit Byte StuffUploaded bySahil Tandon
- DSPUploaded bymailmado

- Prospectus2016 OSRI.pdfUploaded byDebasish
- DEI & TC Syallabus (Final)Uploaded bynandini
- C1 test 1Uploaded byMaruf Hassan
- Artificial Intelligence Based Student Learning Evaluation ToolUploaded bySwapnil Kapse
- Dynamic Simulation of Basic Electric Circuits and MachUploaded bySabareshwar Sankaranarayanan
- NEUTROSOPHIC SETS AND SYSTEMS (BOOK SERIES), Volume 7 / 2015Uploaded byAnonymous 0U9j6BLllB
- 12 Chapter 6Uploaded bysujups
- mm2ass11122Uploaded byprahalad19
- Plane Stress Transformations-Mohr's CircleUploaded byPalash Roy Choudhury
- aaerevUploaded byAham Brahmasmi
- Coupled Mode TheoryUploaded byphithuc
- Comsol EmtUploaded byjcow
- Computational HydraulicsUploaded byPeter Riad
- hobsonajjustthemaths20021296smcetp.pdfUploaded byAIvan DCam G
- 44852485 Numerical MethodsUploaded byDISCOVERY
- algebra 2 syllabus regularUploaded byapi-297719868
- History of Functional Analysis ,1981Uploaded bygylpm
- Mathematics class X must read guideUploaded byNishantThakur
- 9709_s10_ms_31Uploaded byHubbak Khan
- Solucionarios Calculo Fisica Mecanica Algebra Vectoria Logistica Dinamica Estatica EtcUploaded byalxeslavas
- Tony Crilly, Simon Blackburn - The Big Questions_ Mathematics (2011, Quercus Books).epubUploaded byAbec Lee
- syllabus_fall08Uploaded byRhay Notorio
- Machine Design222Uploaded byrajasekharjv
- Chapter 1 the Nature and Organization of Optimization ProblemsUploaded byaaa
- chapter10Uploaded byphuthuynhobg
- 81314_13.pdfUploaded byPavan Kalyan Suryavamshi
- 2.8 Higher Degree First Order Ordinary DeUploaded byJames Jastine Solivio
- Dual Degree B Tech Mineral Engineering With M Tech in Mineral Resource ManagementUploaded bybradburywills
- Jee 2006Uploaded byeng ameer alburihy
- The General Linear EquationUploaded byJm Alipio

## Much more than documents.

Discover everything Scribd has to offer, including books and audiobooks from major publishers.

Cancel anytime.