Scribd
Upload
104 views4 pages

Lecture 8. Linear Systems of Differential Equations

The document summarizes key concepts about linear systems of ordinary differential equations: 1) A linear system can be written in normal form as x'(t) = A(t)x(t) + f(t), where A(t) is the coefficient matrix and f(t) is a nonhomogeneous term. If A(t) is constant, the system has constant coefficients. 2) For any initial value problem of the form x'(t) = A(t)x(t) + f(t), x(t0) = x0, there exists a unique solution defined on some interval containing t0. 3) The general solution of a homogeneous system (f(

Uploaded by

Chernet Tuge
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
104 views4 pages

Lecture 8. Linear Systems of Differential Equations

The document summarizes key concepts about linear systems of ordinary differential equations: 1) A linear system can be written in normal form as x'(t) = A(t)x(t) + f(t), where A(t) is the coefficient matrix and f(t) is a nonhomogeneous term. If A(t) is constant, the system has constant coefficients. 2) For any initial value problem of the form x'(t) = A(t)x(t) + f(t), x(t0) = x0, there exists a unique solution defined on some interval containing t0. 3) The general solution of a homogeneous system (f(

Uploaded by

Chernet Tuge
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Ordinary and Partial Differential Equations

Lecture 8. Linear systems of differential equations


Bibliography: Nagle, Snaff, Sieder - ”Fundamentals of DEs and BVPs” (6th ed.) - chapter 9

LINEAR SYSTEMS IN NORMAL FORM


A linear system of differential equations can be written in the normal form

x′ (t) = A(t)x(t) + f (t) (1)

where    
x1 (t) f1 (t)
 x2 (t)   f2 (t)  [ ]
x(t) = 
 ··· 
 f (t) = 
 ··· 
 A(t) = aij(t) i,j=1,n .
xn (t) fn (t)
→ if A is a constant matrix, we say that system (1) has constant coefficients; otherwise, it has
variable coefficients.
→ if f (t) ≡ 0 system (1) is homogeneous; otherwise, it is nonhomogeneous.

Existence and Uniqueness Theorem:


Suppose that A(t) and f (t) are continuous on an open interval I that contains the point t0 . Then, for
[ ]T
any choice of the initial vector x0 = x10 , x20 , ..., xn0 ∈ Rn there exists a unique solution on the whole
interval I to the initial value problem

x′ (t) = A(t)x(t) + f (t) , x(t0 ) = x0 .

Linear dependence and independence of vector functions:


The m vector functions x1 , x2 , ...xm , defined on a real interval I, with values in Rn are linearly
dependent if there exist real constants c1 , c2 , ..., cm , not all zero, such that:

c1 x1 (t) + c2 x2 (t) + ... + cm xm (t) = 0 , ∀ t ∈ I.

Otherwise, the vector functions x1 , x2 , ...xm are called linearly independent.

Wronskian of vector functions:


T
The Wronskian of n vector functions x1 , x2 , ...xn , where xi (t) = [x1,i (t), x2,i (t), ..., xn,i (t)] is the
function
x1,1 (t) x1,2 (t) ... x1,n (t)

x2,1 (t) x2,2 (t) ... x2,n (t)

W [x1 , x2 , ...xn ](t) = .. .. .. ..
. . . .

xn,1 (t) xn,2 (t) ... xn,n (t)

THE HOMOGENEOUS CASE


♣ If x1 , x2 , ...xn are solutions on the interval (a, b) of the homogeneous system

x′ (t) = A(t)x(t) (2)

and there exists t0 ∈ (a, b) such that W [x1 , x2 , ...xn ](t0 ) ̸= 0, then every solution of (2) can be expressed
as
x(t) = C1 x1 (t) + C2 x2 (t) + ... + Cn xn (t),
where C1 , C2 , ..., Cn are real constants.
→ The linear combination x(t) = C1 x1 (t) + C2 x2 (t) + ... + Cn xn (t), (with the coefficients C1 , C2 , ..., Cn )
is called general solution of the system (2).

⃝
c 2011/2012 Eva Kaslik - West University of Timişoara, Romania
Fundamental solution set
♣ Let x1 , x2 , ...xn be n solutions on the interval (a, b) of the system (2). The following statements are
equivalent:

(i) The solutions x1 , x2 , ...xn are linearly independent;


(ii) The Wronskian W [x1 , x2 , ...xn ] is non-zero at some point t0 of the interval (a, b);
(iii) The Wronskian W [x1 , x2 , ...xn ] is never zero on the interval (a, b);

When one of the above conditions is met, x1 , x2 , ...xn is called fundamental solution set of the system
(2) and the matrix  
x1,1 (t) x1,2 (t) ... x1,n (t)
 x2,1 (t) x2,2 (t) ... x2,n (t) 
 
X(t) =  .. .. .. .. 
 . . . . 
xn,1 (t) xn,2 (t) ... xn,n (t)
is called fundamental matrix for system (2), which can be used to represent the general solution of
system (2) as
T
x(t) = X(t)c, c = [c1 , c2 , ..., cn ] ∈ Rn .

THE NONHOMOGENEOUS CASE


♣ If xp is a particular solution of the nonhomogeneous system
x′ (t) = A(t)x(t) + f (t) (3)
on the interval I and x1 , x2 , ...xn is a fundamental solution set of the associated homogeneous system,
then every solution of the nonhomogeneous system (3) can be expressed as
x(t) = xp (t) + C1 x1 (t) + C2 x2 (t) + ... + Cn xn (t),
where C1 , C2 , ..., Cn are constants.

HOMOGENEOUS LINEAR SYSTEMS WITH CONSTANT COEFFICIENTS


A homogeneous linear system with constant coefficients is of the form
x′ (t) = Ax(t) (4)
where A is a constant n × n matrix.
Searching for a particular solution of the form x(t) = ert u of the system (4), where r is a constant and
u is a constant vector, we arrive at
(A − rI)u = 0,
where I is the n × n identity matrix. Since we are searching for nontrivial solutions, we obtain that r has
to be an eigenvalue of the matrix A, and u has to be one of its associated eigenvectors. Therefore,
the characteristic equation associated to system (4) is
det(A − rI) = 0.

Finding the fundamental solution set of the homogeneous system (4)

• Find all the distinct eigenvalues of the matrix A.


• For every distinct real eigenvalue r, compute the associated eigenvector u. The vector function ert u
is a fundamental solution;
• For every distinct pair of complex conjugate eigenvalues (r, r), r = α + iβ, compute the associated
eigenvectors u = a + ib and ū = a − ib. The vector functions functions
eαt cos βt a − eαt sin βt b
eαt sin βt a + eαt cos βt b
are fundamental solutions.

2
NONHOMOGENEOUS LINEAR SYSTEMS WITH CONSTANT COEFFICIENTS
A nonhomogeneous linear system of differential equations with constant coefficients is of the
form
x′ (t) = Ax(t) + f (t) (5)
where A is a constant n × n matrix and f is vector function with n entries.
♣ A particular solution xp of (5) may be found by the method Variation of Parameters.
Indeed, if we know a fundamental solution set for the associated homogeneous system x′ (t) = Ax(t),
meaning that we know the fundamental matrix X(t) of the homogeneous system, we can search for a
particular solution xp of the nonhomogeneous system (5) of the form
xp (t) = X(t)v(t)
T
where v(t) = [v1 (t), v2 (t), ..., vn (t)] is to be determined.
If we replace xp of this form in the system (5), we obtain:
X′ (t)v(t) + X(t)v′ (t) = AX(t)v(t) + f (t)
and taking to account that X′ (t) = AX(t) (fundamental matrix) we have:
∫
X(t)v′ (t) = f (t) ⇒ v′ (t) = X−1 (t)f (t) ⇒ v(t) = X−1 (t)f (t)dt,

and so, the particular solution can be written as


∫
xp (t) = X(t) X−1 (t)f (t)dt.

Therefore, the general solution of the nonhomogeneous system (5) is:


∫
x(t) = X(t)c + X(t) X−1 (t)f (t)dt, where c ∈ Rn .

THE MATRIX EXPONENTIAL FUNCTION


If A is a constant n × n matrix, we define the matrix exponential function eAt as the (matrix-valued)
series
t2 tn
eAt = I + tA + A2 + ... + An + ...
2! n!
Properties of the matrix exponential function:
Let A, B be n × n constant matrices, and r, s, t real or complex numbers. Then:

• eA0 = I;
• eA(t+s) = eAt eAs ;
• (eAt )−1 = e−At ;
• e(A+B)t = eAt eBt provided that AB = BA;
• erIt = ert I;
d At
• (e ) = AeAt , so eAt is a solution to the matrix differential equation X′ = AX, and hence
dt
eAt is a fundamental matrix of the homogeneous system with constant coefficients x′ (t) = Ax(t).

♣ The general solution of the system x′ (t) = Ax(t) is x(t) = eAt c, where c ∈ Rn .
Generalized eigenvector:
Let A be a square matrix. A nonzero vector u which satisfies
(A − rI)m u = 0
for some scalar r and some positive integer m, is called generalized eigenvector associated to the
eigenvalue r.

3
Solving the linear homogeneous system x′ = Ax:

• find all the distinct eigenvalues r1 , r2 , ..., rk of A and their algebraic multiplicities m1 , m2 , ..., mk ;

• for each eigenvalue ri find mi linearly independent generalized eigenvectors by solving


(A − ri I)mi u = 0;

• Find n linearly independent solutions by computing, for each eigenvalue ri and each of the
corresponding mi generalized eigenvectors
[ ]
t2
x(t) = e u = e u + t(A − ri I)u + (A − ri I) u + ...
At rt 2
2!

This series terminates after mi or fewer terms.

You might also like