Unformatted text preview: MEC702 Wk 8 Lect 2 Basic Theory of Systems of ODEs
FirstOrder System of ODEs n The system of n rstorder of ODEs is given by
y10 = f1 (t, y1 , . . . , yn ) y20 = f2 (t, y1 , . . . , yn ) yn0 = fn (t, y1 , . . . , yn ) ..
. ..
. ..
. This can be written as a vector equation
~y 0 = f~(t, ~y ) 0
f1
y1
y1 f2 y2 y20 where ~y 0 = .. , ~y = .. and f~ = .. .
.
.
.
0
fn
yn
yn A solution of the system on some interval (a, b) is a set of functions
y1 = h1 (t), y2 = h2 (t), ..., yn = hn (t) on (a, b) that satisfy thesystem throughout this interval. Writing the 'solution h1 (t)
h2 (t) vector' as ~h(t) = .. , we do so instead as ~y = ~h(t). . ht (t)
IVP An Initial Value
initial conditions Problem for the system consists of the systems and n given y1 (t0 ) = K1 , y2 (t0 ) = K2 , 1 ..., yn (t0 ) = Kn and as in vector form K1 K2 ~ =
~y (t0 ) = K .. . . Kn Linear System • We call a system a it can be written as linear system if it is linear in y , y , . . . , y
1 2 n, y10 = a11 (t)y1 + a12 (t)y2 + . . . + a1n (t)yn + g1 (t) y20 = a21 (t)y1 + a22 (t)y2 + . . . + a2n (t)yn + g2 (t) ..
. ..
. ..
. yn0 = an1 (t)y1 + an2 (t)y2 + . . . + ann (t)yn + gn (t). that is, if • In vector form, this can be written as a11 a21
where A = .. .
an1 a12
a22 ..
. an2 ...
... .. . ... ~y 0 = A~y + ~g a1n
y1
g1 y2 g2 a2n .. , ~y = .. and ~g = .. . . .
.
ann
yn
gn • The linear system is called homogeneous if ~g = ~0; otherwise it is nonhomogeneous if ~g 6= ~0. (Superposition/ Linearity Principle) If ~y and ~y are solutions of
the homogeneous linear system ~y = A~y on some interval, then
Theorem. 1 ~y = c1 ~y1 + c2 ~y2 is also a solution of the homogeneous system. 2 2 General Solution Let ~y1 , ~y2 . . . , ~yn be a basis of solutions of the homogeneous system
~y 0 = A~y on some interbal I . We call the linear combination
~y = c1 ~y1 + c2 ~y2 + · · · + cn ~yn is a general solution of the system on I . Each matrix ~yk is a matrix of the form y1k y2k ~yk = . , .. ynk hence forming a n × n matrix
Y = ~y1 ~y2 ··· y11 y21
~yn = . ..
yn1 Wronskian The determinant of Y is called the W (~y1 , ~y2 , . . . , ~yn ) y12
y22 ···
··· yn2 ··· ..
. Wronskian of y11 y21 = . .. yn1 3 .. . y1n
y2n .. .
. ynn ~y1 , ~y2 , . . . , ~yn , written y12 · · · y1n y22 · · · y2n ..
.. ..
.
.
. yn2 · · · ynn Constant Coecient System. Phase Plane Method
Eigenvalues and Eigenvectors (General Solution of the Homogeneous Linear System) If the constant matrix A in the linear system ~y = A~y has a linearly independent set of n
eigenvalues,
Theorem. 0 λ1 , λ2 , . . . , λn , with corresponding linearly independent eigenvectors
~x1 , ~x2 , . . . , ~xn . Then the corresponding solutions are
~y = c1 ~x1 eλ1 t + c2 ~x2 eλ2 t + . . . + cn ~xn eλn t . The Wronskian for a constant coecient system is given by y11 y21 W = yn1 ... y12
y22 ···
··· yn2 ··· x11
y1n x21
y2n = e(λ1 +λ2 +...+λn )t xn1
ynn ... ... . . . ... x12
x22 ···
··· xn2 ··· x1n x2n . xnn ... . . . ... Graphing Solution in the Phase Plane Consider the two rstorder ODE with constant coecient ~y 0 = A~y ; in components,
y10 = a11 y1 + a12 y2 y20 = a21 y1 + a22 y2 .
y1 (t)
We can graph solution of ~y (t) =
in two ways:
y2 (t)
• as two curves over the taxis, one for each component of ~y (t);
• as a single curve in the y1 y2 plane (phase plane). Such a curve as a trajectory. The y1 y2 plane is called the phase plane. If the
phase plane is lled with trajectories of y10 and y20 , we obtain the called phase
portrait.
Example. Tra jectories in the Phase Plane. Find and graph the solution of the system.
−3
~y = A~y =
1 4
1
~y .
−3 The eigenvalues
are λ1= −2
and λ2 = −4. The corresponding eigenvectors
1
1
are ~x1 =
and ~x2 =
for the respective eigenvalues. This gives the
solution 1 −1
y1
1 −2t
1
~y =
= c1 ~y1 + c2 ~y2 = c1
e
+ c2
e−4t .
y2
1
−1 5 ...
View
Full Document
 Winter '16
 Alveen Chand
 Linear Algebra, Vector Space, yn, phase plane

Click to edit the document details