Scribd
Upload
93 views

Linear Systems Summary

To solve systems of two linear equations, we find the eigenvalues and eigenvectors of the corresponding matrix. There are four cases for the eigenvalues: 1) two distinct real eigenvalues, 2) complex eigenvalues, 3) one repeated real eigenvalue with two independent eigenvectors, 4) one repeated real eigenvalue with one eigenvector. The solutions involve exponential functions of the eigenvalues and eigenvectors. Initial value problems are solved by finding the fundamental matrix and multiplying by the inverse initial condition matrix.

Uploaded by

Frantz Clermont
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
93 views

Linear Systems Summary

To solve systems of two linear equations, we find the eigenvalues and eigenvectors of the corresponding matrix. There are four cases for the eigenvalues: 1) two distinct real eigenvalues, 2) complex eigenvalues, 3) one repeated real eigenvalue with two independent eigenvectors, 4) one repeated real eigenvalue with one eigenvector. The solutions involve exponential functions of the eigenvalues and eigenvectors. Initial value problems are solved by finding the fundamental matrix and multiplying by the inverse initial condition matrix.

Uploaded by

Frantz Clermont
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Math 1005 H Fall 2015

Systems of two linear equations summary


To solve the system
x0 = ax + by
 

y 0 = cx + dy
we begin by finding the eigenvalues and corresponding eigenvectors of the matrix
 
a b
A= .
c d

The eigenvalues are the roots of the equation

det(λI − A) = 0

(after expanding, the left-hand side will be a quadratic polynomial in λ). For each
eigenvalue λ, we find corresponding eigenvectors by solving

(λI − A)v = 0. (1)

We need to find two linearly independent solutions x1 (t) and x2 (t), and the various
methods to do so are outlined in the four cases below. Once we have these solutions,
we set 
X(t) = x1 (t) x2 (t) ,
and the general solution is
x(t) = X(t)c
 
c1
where c = is a vector of arbitrary constants.
c2
Case 1: We have two distinct real eigenvalues. We can find one corresponding eigen-
vector for each eigenvalue (the two vectors will automatically be linearly indepen-
dent). Once we have linearly independent eigenvectors v1 and v2 corresponding to
eigenvalues λ1 and λ2 , two linearly independent solutions are

x1 (t) = eλ1 t v1 and x2 (t) = eλ2 t v2 .

Case 2: We have complex eigenvalues. For λ = α±βi complex, we examine λ = α+βi.


We can still find a corresponding eigenvector using equation (1), but the vector will
contain complex numbers. We can write it as v = a + bi, where a and b contain only
real numbers. Two real solutions are

x1 (t) = eαt (cos(βt)a − sin(βt)b) and x2 (t) = eαt (sin(βt)a + cos(βt)b) .

Case 3a: We have only one real (repeated) eigenvalue, but two linearly independent
corresponding eigenvectors. The only 2 × 2 systems where this can happen look like
 0 
x = kx
y 0 = ky
for some constant k. Note that we can completely “decouple” the two equations and
solve them one at a time. The equation x0 = kx is separable with solution x(t) = c1 ekt ,
and the equation y 0 = ky is separable with solution y(t) = c2 ekt . This is the easiest
way to solve this system, but we could use the methods of this section if we really
wanted to (as is done in example 5.3 in the text).

Case 3b: We have only one real (repeated) eigenvalue with only one independent
eigenvector. In this case we must use a generalized eigenvector for the second inde-
pendent solution. We have one eigenvalue λ1 with a corresponding eigenvector v1 ,
and so one solution is
x1 (t) = eλ1 t v1 .
To find a second solution, we first find a vector u1 that satisfies (A − λ1 I)u1 = u0
(here u0 is an eigenvector corresponding to λ1 , and so we can set u0 equal to v1 or to
any non-zero scalar multiple of v1 ). Then a second linearly independent solution is

x2 (t) = eλ1 t (tu0 + u1 ).

Initial value problems: To solve a system x0 = Ax with initial condition x(t0 ) = x0 ,


we first find a fundamental matrix X(t). Then the particular solution to the initial
value problem is
x(t) = X(t)X(t0 )−1 x0 .

You might also like