Vector
Differential
Equations:
Nondefective
Coefficient
Matrix
Math 240
Solving linear
systems by di-
agonalization
Vector Differential Equations:
Real e-vals
Complex e-vals Nondefective Coefficient Matrix
Math 240 — Calculus III
Summer 2013, Session II
Tuesday, July 30, 2013
Vector
Differential Agenda
Equations:
Nondefective
Coefficient
Matrix
Math 240
Solving linear
systems by di-
agonalization
Real e-vals
Complex e-vals
1. Solving linear systems by diagonalization
Real eigenvalues
Complex eigenvalues
Vector
Differential Introduction
Equations:
Nondefective
Coefficient
Matrix
Math 240
Solving linear
systems by di-
agonalization
Real e-vals The results discussed yesterday apply to any old vector
Complex e-vals
differential equation
x0 = Ax.
In order to make some headway in solving them, however, we
must make a simplifying assumption:
The coefficient matrix A consists of real constants.
Vector
Differential Diagonalization
Equations:
Nondefective
Coefficient
Matrix
Math 240
Recall that an n × n matrix A may be diagonalized if and only
Solving linear
systems by di- if it is nondefective.
agonalization
Real e-vals
Complex e-vals When this happens, we can solve the homogeneous vector
differential equation
x0 = Ax.
If S −1 AS = diag(λ1 , λ2 , . . . , λn ), then
λ t
c1 e 1
c2 eλ2 t
x = Sy, where y = . .
. .
cn eλn t
Vector
Differential Example
Equations:
Nondefective Solve the linear system
Coefficient
Matrix x01 = 2x1 + x2 ,
Math 240
x02 = −3x1 − 2x2 .
Solving linear
systems by di-
agonalization 1. Turn it into the vector differential equation
Real e-vals
Complex e-vals
0 2 1
x = Ax, where A = .
−3 −2
2. The characteristic polynomial of A is λ2 − 1.
3. Eigenvalues are λ = ±1.
4. Eigenvectors are
λ1 = 1 : v1 = (−1, 1),
λ2 = −1 : v2 = (−1, 3).
5. We have
c1 et −c1 et − c2 e−t
−1 −1
y= , so x = y= .
c2 e−t 1 3 c1 et + 3c2 e−t
Vector
Differential Vector formulation
Equations:
Nondefective
Coefficient
Matrix
Math 240 The change of basis matrix S is
Solving linear
systems by di-
S = v1 v2 · · · vn ,
agonalization
Real e-vals
where v1 , . . . , vn are n linearly independent eigenvectors of A.
Complex e-vals
Hence,
x = Sy = c1 eλ1 t v1 + c2 eλ2 t v2 + · · · + cn eλn t vn
= c1 x1 + c2 x2 + · · · + cn xn .
Check if these n solutions are linearly independent:
W [x1 , . . . , xn ] = det eλ1 t v1 eλ2 t v2 · · · eλn t vn
= e(λ1 +···+λn )t det v1 v2 · · · vn
6= 0.
They are linearly independent, therefore a fundamental set of
solutions.
Vector
Differential General solution
Equations:
Nondefective
Coefficient
Matrix
Math 240
Solving linear
systems by di- Theorem
agonalization
Real e-vals Suppose A is an n × n matrix of real constants. If A has n real
Complex e-vals
linearly independent eigenvectors v1 , v2 , . . . , vn with
corresponding eigenvalues λ1 , λ2 , . . . , λn (not necessarily
distinct), then the vector functions {x1 , x2 , . . . , xn } defined by
xk (t) = eλk t vk , for k = 1, 2, . . . , n
are a fundamental set of solutions to x0 = Ax on any interval.
The general solution is
x(t) = c1 x1 + c2 x2 + · · · + cn xn .
Vector
Differential
Example
Find the general solution to x0 = Ax if
Equations:
Nondefective
Coefficient
Matrix 0 2 −3
Math 240 A = −2 4 −3 .
Solving linear −2 2 −1
systems by di-
agonalization
Real e-vals 1. Characteristic polynomial is −(λ + 1)(λ − 2)2 .
Complex e-vals
2. Eigenvalues are λ1 = −1 and λ2 = 2.
3. Eigenvectors are
λ1 = −1 : v1 = (1, 1, 1),
λ2 = 2 : v2 = (1, 1, 0), v3 = (−3, 0, 2).
4. Fundamental set of solution is
1 1 −3
x1 (t) = e−t 1 , x2 (t) = e2t 1 , x3 (t) = e2t 0 .
1 0 2
5. So general solution is
x(t) = c1 x1 (t) + c2 x2 (t) + c3 x3 (t).
Vector
Differential Complex eigenvalues
Equations:
Nondefective
Coefficient
Matrix
Math 240
What happens when A has complex eigenvalues?
Solving linear
systems by di-
agonalization If u = a + ib and v = a − ib then
Real e-vals
Complex e-vals
u+v u−v
a= and b = .
2 2i
Theorem
Let u(t) and v(t) be real-valued vector functions. If
w1 (t) = u(t) + iv(t) and w2 (t) = u(t) − iv(t)
are complex conjugate solutions to x0 = Ax, then
x1 (t) = u(t) and x2 (t) = v(t)
are themselves real valued solutions of x0 = Ax.
Vector
Differential Complex eigenvalue example
Equations:
Nondefective
Coefficient
Matrix Example
Find the general solution to x0 = Ax where
Math 240
Solving linear
systems by di- 0 1
agonalization A= .
Real e-vals
−1 0
Complex e-vals
1. Characteristic polynomial is λ2 + 1.
2. Eigenvalues are λ = ±i.
3. Eigenvectors are v = (1, ±i).
4. Linearly independent solutions are
±it 1 cos t sin t
w(t) = e = ±i .
±i − sin t cos t
5. Yields the two linearly independent real solutions
cos t sin t
x1 (t) = and x2 (t) = .
− sin t cos t
Vector
Differential The formula
Equations:
Nondefective
Coefficient
Matrix Let’s derive the explicit form of the real solutions produced by
Math 240 a pair of complex conjugate eigenvectors.
Solving linear Suppose λ = a + ib is an eigenvalue of A, with b 6= 0,
systems by di-
agonalization corresponding to the eigenvector r + is. This produces the
Real e-vals
Complex e-vals complex solution
w(t) = e(a+ib)t (r + is)
= eat (cos bt + i sin bt)(r + is)
= eat (cos bt r − sin bt s) + ieat (sin bt r + cos bt s).
Thus, the two real-valued solutions to x0 = Ax are
x1 (t) = eat (cos bt r − sin bt s),
x2 (t) = eat (sin bt r + cos bt s).
Remark
The conjugate eigenvalue a − ib and eigenvector r − is would
result in the same pair of real solutions.
