MATH 304 Linear Algebra Lecture 35: Matrix exponentials.
Initial value problem for a linear ODE:
dy dt
= 2y ,
y (0) = 3.
Solution: y (t ) = 3e 2t . Initial value problem for a system of linear ODEs:
dx dt dy dt
= 2x + 3y , = x + 4y ,
x (0) = 2, y (0) = 1.
The system can be rewritten in vector form d x 2 3 x . , where A = =A 1 4 y dt y Solution: What is e tA ? x (t ) y (t ) = e tA 2 . 1
Exponential function Exponential function: f (x ) = exp x = e x , x R. Principal property: e x +y = e x e y . Denition 1. e x = lim 1 +
n
x n
n
1 In particular, e = lim 1 + n n
x
2.7182818.
x2 xn Denition 2. e = 1 + x + + + + 2! n! Denition 3. f (x ) = e x is the unique solution of the initial value problem f = f , f (0) = 1.
Matrix exponentials Denition. For any square matrix A let 1 1 exp A = e A = I + A + A2 + + An + 2! n! Matrix exponential is a limit of matrix polynomials. Remark. Let A(1) , A(2) , . . . be a sequence of nn (n) matrices, A(n) = (aij ). The sequence converges to an nn matrix B = (bij ) if aij bij as n , i.e., if each entry converges. Theorem The matrix exp A is well dened, i.e., the series converges.
(n)
Properties of matrix exponentials Theorem 1 If AB = BA then e A e B = e B e A = e A+B . Corollary (a) e tA e sA = e sA e tA = e (t +s )A , t , s R; (b) e O = I ; (c) (e A )1 = e A . Theorem 2
tA
d tA e = Ae tA = e tA A. dt
tn n t2 2 Indeed, e = I + tA + A + + A + , 2! n! and the series can be dierentiated term by term. t n1 d tn d tn n n A = = A An . dt n! dt n! (n 1)!
Lemma Let A be an nn matrix and x Rn . Then the vector function v(t ) = e tA x satises v = Av. d dv = e tA x = Ae tA x = A e tA x = Av. Proof: dt dt Theorem For any t0 R and x0 Rn the initial value problem dv = Av, v(t0 ) = x0 dt has a unique solution v(t ) = e (t t0 )A x0 . Indeed, v(t ) = e (t t0 )A x0 = e tA e t0 A x0 = e tA x, where x = e t0 A x0 is a constant vector.
Evaluation of matrix exponentials Example. A = diag(a1 , a2 , . . . , ak ).
n n n An = diag(a1 , a2 , . . . , ak ), n = 1, 2, 3, . . . 1 2 1 n e A = I + A + 2! A + + n !A +
= diag(b1 , b2 , . . . , bk ),
1 3 1 2 ai + 3! ai + = e ai . where bi = 1 + ai + 2!
Theorem 1 If A = diag(a1 , a2 , . . . , ak ) then e A = diag(e a1 , e a2 , . . . , e ak ). e tA = diag(e a1 t , e a2 t , . . . , e ak t ). Theorem 2 If A = UBU 1 , then e tA = Ue tB U 1 .
Example. A =
2 3 . 1 4
The eigenvalues of A: 1 = 1, 2 = 5. 3 1 Eigenvectors: v1 = , v2 = . 1 1 Therefore A = UBU 1 , where 3 1 1 0 , U= . B= 0 5 1 1 Then 3 1 3 1 et 0 tA tB 1 e = Ue U = 5t 1 1 1 1 0 e = 3e t e 5t e t e 5t
1 4
1 1 1 3
1 4
3e t + e 5t 3e t + 3e 5t . e t + e 5t e t + 3e 5t
Problem. Solve a system of dierential equations
dx dt dy dt
= 2x + 3y , = x + 4y
subject to initial conditions x (0) = 2, y (0) = 1. The unique solution: x (t ) = e tA x0 , where A = y (t ) e tA = =
1 4
2 3 , x0 = 1 4
2 . 1
3e t + e 5t 3e t + 3e 5t e t + e 5t e t + 3e 5t
3 t 5t e +5 x (t ) = 4 4e , 5 5t t y (t ) = 1 4e + 4e .
0 1 0 Example. A = 0 0 1, a Jordan block. 0 0 0 0 0 1 2 A = 0 0 0, A3 = O , An = O for n 3. 0 0 0 1 1 1 2 1 e A = I + A + A2 = 0 1 1, 2 0 0 1 1 2 t 1 t 2 2 t e tA = I + tA + A2 = 0 1 t . 2 0 0 1
Example. A = A2 = e
tA
1 1 , another Jordan block. 0 1 1 3 , An = 0 1 1 n . 0 1 a(t ) b (t ) , 0 a(t )
1 2 , A3 = 0 1
tn n t2 2 = I +tA+ A + + A + = 2! n!
t2 2!
where a(t ) = 1 + t +
2 3
t3 3!
+ = et,
t t b (t ) = t + 2 t 2! + 3 3! + = te .
Thus e tA =
e t te t . 0 et
Example. A =
1 , a general Jordan block. 0 0 1 . 0 0
We have that A = I + B , where B =
Since (I )B = B (I ), it follows that e A = e I e B . Similarly, e tA = e t I e tB . e
t I
e t 0 0 e t
= e t I , 1 t . 0 1
B2 = O
= e tB = I + tB =
Thus e tA = e t I e tB =
e t te t . 0 e t
Problem. Solve a system of dierential equations
dx dt dy dt
= 2x + y , = 2y
subject to initial conditions x (0) = y (0) = 1. The unique solution: x (t ) = e tA x0 , where A = y (t ) e tA = = e 2t te 2t 0 e 2t 2 1 , x0 = 0 2 1 t 0 1 1 . 1
= e 2t
x (t ) = e 2t (1 + t ), y (t ) = e 2t .
