185week3 1
185week3 1
Homogeneous Linear Systems: Real Eigenvalue Case. In this section we will begin to address the
problem of finding two solutions ⃗x1 (t) and ⃗x2 (t) of a linear, homogeneous, autonomous system
d⃗x
= A⃗x,
dt
such that ⃗x1 (0) and ⃗x2 (0) are a two-dimensional basis.
As a concrete example, consider the homogeneous system
d x 1 −2 x
= .
dt y −2 1 y
To understand what the solutions of this system look like, we first compute the slope of its velocity field,
dy −2x + y
= ,
dx x − 2y
and we write down its isocline equations:
−2x + y
= C = constant.
x − 2y
Simplifying this equation, we find that all isoclines are lines through the origin:
−2x + y = C(x − 2y) =⇒ (1 + 2C)y = (2 + C)x.
We can see this effect when we plot the velocity field:
y
Notice that there are two isoclines which are everywhere parallel to the velocity field. One of these is the
isocline with slope 1:
C = 1 =⇒ (1 + 2)y = (2 + 1)x =⇒ y = x.
Geometrically, the fact that the velocity field is parallel to this line tells us that our system of equations has
a special solution of the form
x1 (t) 1 f (t)
= f1 (t) = 1 .
y1 (t) 1 f1 (t)
1
2
To solve for the unknown function f1 (t), we can substitute into both sides of our system:
′
d x1 (t) f (t)
= 1′
dt y1 (t) f1 (t)
1 −2 f1 (t) −f1 (t)
= .
−2 1 f1 (t) −f1 (t)
From this we see that f1 (t) must satisfy the differential equation
f1′ (t) = −f1 (t),
and therefore must take the general form
f1 (t) = c1 e−t
for some constant c1 .
An important thing to notice here is that
−t
x1 (t) c e 0
lim = lim 1 −t = .
t→∞ y1 (t) t→∞ c1 e 0
This is consistent with the velocity field: along the line y = x, the velocity field points directly inward,
towards the origin.
By similar reasoning, we can find a solution which travels along the isocline with slope C = −1:
C = −1 =⇒ (1 − 2)y = (2 − 1)x =⇒ y = −x
This solution takes the form
1
⃗x2 (t) = f2 (t) ,
−1
and the function f2 (t) can be found using the same process:
′
d x1 (t) f2 (t)
=
dt y1 (t) −f2′ (t)
1 −2 f2 (t) 3f2 (t)
= .
−2 1 −f2 (t) −3f2 (t)
From this we see that f2 (t) must satisfy the differential equation
f2′ (t) = 3f2 (t),
and therefore must take the general form
f2 (t) = c2 e3t
for some constant c2 . This time, we have
lim f2 (t) = ∞.
t→∞
Again, this is consistent with the velocity field, which points away from the origin on the line y = −x.
Let us now generalize this idea to other linear homogeneous systems. We have to look for solutions that
start at points (x0 , y0 ) where the position vector,
x
⃗v = 0
y0
is parallel to the velocity specified by the velocity field:
x0
A⃗v = A
y0
To make A⃗v and ⃗v parallel, ⃗v must satisfy the eigenvector equation:
A⃗v = λ⃗v .
Given any eigenvector ⃗v , we can immediately find a solution of our system, in the form
⃗x(t) = f (t)⃗v ,
3
and we can also immediately solve for f (t) by substituting into the equation
d⃗x
= A⃗x.
dt
and therefore
f (t) = ceλt
In cases where A is a 2 × 2 matrix with a real eigenbasis, this idea produces two solutions,
For example, suppose we want to find a solution of the initial value problem
d x 1 −2 x
= , x(0) = 2, y(0) = 1.
dt y −2 1 y
For this system, we have already obtained the two special solutions
−t 1 3t 1
⃗x1 (t) = e , ⃗x2 (t) = e ,
1 −1
You can verify that the solution of this system of linear equations is
3 1
c1 = , c2 = .
2 2
We can see that the solution starts off asymptotic to the line y = x, in the limit as t → −∞. It’s a bit harder
to see from the picture, but in the limit as t → ∞ it is asymptotic to the line y = −x. The reason for this
is that
3 −t 1 0
lim e =
t→∞ 2 1 0
and
1 3t 1 0
lim e = ,
t→−∞ 2 −1 0
so in each of these limits only one of the two terms dominates the behavior of ⃗x(t).
The reason why it takes the solution much longer to approach the line y = −x is that the exponential
e−t
decays to 0 at a slower rate in the limit t → ∞ than the exponential
e3t
decays to 0 in the limit as t → −∞.
The relative rates of exponential decay actually become very important for systems where the eigenvalues
of the coefficient matrix have the same sign. For example, consider the initial value problem
d x 2 1 x
= , x(0) = 2 , y(0) = 1
dt y 1 2 y
The coefficient matrix for this system,
2 1
1 2
has the same eigenvectors as in our previous example, but the eigenvalues have the same sign:
2 1 1 1 2 1 1 1
=3· , =1·
1 2 1 1 1 2 −1 −1
5
Here a plot of the solution, together with the velocity field of the system:
y
In the limit as t → ∞, the solution curve is nearly parallel to the line y = x. This is because the exponential
e3t
grows at a faster rate than the exponential
et ,
so the first term dominates the solution in the limit as t → ∞.
Similarly, in the limit as t → −∞, the exponential
et
decays slower than the exponential
e3t ,
so as t → −∞, the second term dominates. This is also reflected in the plot: as the solution approaches the
origin, it does so in a direction which is tangent to the line y = −x.
6
To prove this identity, first expand λ into its real and imaginary parts:
d (p+iq)t
e = (p + iq)e(p+iq)t .
dt
Then expand both sides of the equation into their real and imaginary parts:
d pt d pt d pt
e cos(qt) + iept sin(qt) =
e cos(qt) + i e sin(qt) ,
dt dt dt
(p + iq) ept cos(qt) + iept sin(qt) = pept cos(qt) − qept sin(qt) + i pept sin(qt) + qept cos(qt) .
where ⃗v1 and ⃗v2 are real vectors. It’s helpful to plot these trajectories in the special case
1 0
⃗v1 = , ⃗v2 = .
0 1
In this case, we have
cos(qt)
⃗x(t) = ept .
sin(qt)
No matter the value of p, the curve will “spiral” around the origin, crossing the x axis at regular intervals
with period
2π
T = .
q
The distance from the origin is controlled by the factor ept , in a way which depends on the sign of p:
(1) If p < 0, then the solution will spiral towards the origin, since
lim ept = 0.
t→∞
(2) If p > 0, the solution will spiral away from the origin, since
lim ept = ∞.
t→∞
(3) If p = 0, the solution will move around the origin in a circle, since
e0t = 1.
Here are pictures of all three cases:
We can plot the solutions in the general case by applying a linear transformation to each of the standard
solutions above.
For example, to sketch the trajectory
⃗x(t) = cos(qt)⃗v1 + sin(qt)⃗v2 ,
we would take the standard trajectory
⃗x0 (t) = cos(qt)ı̂ + sin(qt)ȷ̂,
and apply the following transformation:
⃗v1
ȷ̂
⃗v2
ı̂
9
To guide the transformation, we draw a square with sides parallel to the x and y axes. When we transform
this square, it becomes a parallelogram with sides parallel to ⃗v1 and ⃗v2 . The circle inscribed in the square
becomes an ellipse inscribed in the parallelogram!
The pictures when p ̸= 0 is similar. A spiral aligned with the x and y axes gets transformed to a spiral
aligned with the vectors ⃗v1 and ⃗v2 :
⃗v1
ȷ̂
⃗v2
ı̂
Notice that in both cases, the orientation of the spiral changed, from clockwise to counterclockwise. In
general, to determine the orientation of the spiral, we need to draw a few velocity vectors.
For example, consider the system
d x −1 −4 x
= .
dt y 2 3 y
If we calculate the characteristic polynomial of its matrix, we get
−1 − λ −4
det = λ2 − 2λ + 5 = (λ − 1)2 + 22 .
2 3−λ
This has complex roots
λ = 1 ± 2i,
and therefore the solutions will spiral outward (since the real part of the eigenvalue is p = 1 > 0). To
determine the orientation of the spirals, we can compute the velocity vectors at the points (±1, 0) and
(0, ±1) and draw them:
Once we draw these, it becomes clear that the spirals must be going counterclockwise.
It is usually pretty difficult to make accurate sketches of the solutions by hand, but it is not hard to get a
rough idea of what they look like.
10
Linear Differential Operators. A differential equation of order n is said to be linear if it can be written
in the form
dn y dn−1 y dy
pn (t) n + pn−1 (t) n−1 + · · · + p1 (t) + p0 (t)y = f (t)
dt dt dt
where p0 (t), . . . , pn (t) and f (t) are arbitrary functions of t. Any other equation is said to be nonlinear.
There is a very rich theory of linear equations, which incorporates a number of robust and useful concepts.
By contrast, the theory of nonlinear equations is mostly restricted to a few special examples which can only
be solved approximately, or by a combination of ingenuity and physical/geometric insight. For this reason,
we will focus on linear equations for the remainder of the course.
When thinking about linear equations, it is useful to introduce the language of linear differential operators.
d
The simplest linear differential operator is the derivative operator, dt . It is an operator because it operates
d
on functions: given a function y(t), we can apply the operator dt , and the result is a new function,
d
[y(t)] = y ′ (t).
dt
This operation is linear, because it obeys the sum rule:
d d d
[y1 (t) + y2 (t)] = [y1 (t)] + [y2 (t)]
dt dt dt
and more generally,
d d d
[c1 y1 (t) + c2 y2 (t)] = c1 [y1 (t)] + c2 [y2 (t)]
dt dt dt
for any functions y1 and y2 and constants c1 and c2 .
When we apply the derivative operator twice, the result is a second derivative:
d2 y
d d
[y] = 2 = y ′′
dt dt dt
For second and higher derivatives it is useful to introduce the shorthand
d
D= ,
dt
because the notation
D2 y = y ′′ , D3 y = y ′′′ , etc.
is a bit easier on the eyes than
d2 y d3 y
, etc.,
dt2 dt3
and it more accurately reflects the idea that we are differentiating several times in a row.
In general, a linear differential operator is any operator of the form
O = pn Dn + pn−1 Dn−1 + · · · + p2 D2 + p1 D + p0 ,
where pi = pi (t) can be arbitrary functions of t.
When we apply an operator like this to a function y(t), we get a new function,
Oy = pn Dn y + pn−1 Dn−1 y + · · · + p2 D2 y + p1 Dy + p0 y
dn y dn−1 y d2 y dy
= pn n + pn−1 n−1 · · · + p2 2 + p1 + p0 y
dt dt dt dt
and this operation is linear:
O [c1 y1 (t) + c2 y2 (t)] = c1 O [y1 (t)] + c2 O [y2 (t)]
With this notation, any linear differential equation can be written conceptually in the form
Oy = f,
where O is a linear differential operator.
For example, operator notation allows to clearly identify three properties of linear equations:
12
(1) Suppose that y1 (t) and y2 (t) are two solutions of a linear homogeneous equation:
Oy1 = 0
Oy2 = 0.
If c1 and c2 are arbitrary constants, and
y(t) = c1 y1 (t) + c2 y2 (t),
then y(t) is a solution of the same homogeneous equation:
O[c1 y1 + c2 y2 ] = c1 Oy1 + c2 Oy2 = 0 + 0 = 0
In other words, solutions of homogeneous equations can be superimposed.
Second Order Homogeneous Equations: Exponential Ansatz Method. Let us focus on solving
second order homogeneous equations
y ′′ + py ′ + qy = 0
whose coefficients p and q are constants. An equation like this can also be written using operator notation,
D2 y + pDy + qy = 0,
or more succinctly as
Oy = 0 for O = D2 + pD + q.
Most equations of this form can be solved using two key ideas:
1. The Exponential Ansatz.1
2. The Superposition Principle.
For second order homogeneous equations with constant coefficients, our lucky guess will take the form
y(t) = eλt ,
where λ is an unspecified number (possibly complex). A guess of this form is called an exponential ansatz.
To see why the exponential ansatz is a good guess, recall that the identity
d λt
e = λeλt
dt
holds for any real or complex number λ. Similarly for the higher derivatives, we have
dk λt
Dk eλt = e = λk eλt .
dtk
Therefore, if we make the substitution y = eλt in the equation
D2 y + pDy + qy = 0,
we get
λ2 eλt + pλeλt + qeλt = 0.
Dividing by eλt , we obtain a quadratic equation for λ:
λ2 + pλ + q = 0
The polynomial on the left hand side is called the auxiliary polynomial of the differential operator
O = D2 + pD + q.
In most cases, the auxiliary polynomial will have 2 distinct roots, λ1 and λ2 (which may be real or complex).
Therefore, we usually find two exponential solutions
y1 (t) = eλ1 t and y2 (t) = eλ2 t .
when we make an exponential ansatz.
For example, consider the equation
y ′′ + 4y ′ + 3y = 0.
In this case, the auxiliary equation is
λ2 + 4λ + 3 = 0,
which factors as
(λ + 1)(λ + 3) = 0,
and therefore has solutions λ1 = −1 and λ2 = −3. The corresponding exponential solutions are
y1 (t) = e−t and y2 (t) = e−3t .
2There is an exception when the characteristic equation has a repeated root - we will come back to this case later.
15
Second Order Homogeneous Equations: Repeated Integration Method. There is one special case
where the exponential ansatz method fails to produce a general solution of a given second order equation:
when the auxiliary equation has a repeated root.
For example, consider the equation
y ′′ + 4y ′ + 4y = 0
In this case the auxiliary equation is
λ2 + 4λ + 4 = 0,
which factors as
(λ + 2)2 = 0,
so the equation has a repeated root,
λ = −2.
As a result, the exponential ansatz method produces only one solution,
y1 = e−2t .
To find a second solution of the equation, we need a new method.
The idea behind the second method is to rewrite the equation using operator notation:
(D2 + 4D + 4)y = 0.
We can then factor the operator on the left hand side:
D2 + 4D + 4 = (D + 2)2 .
Here we need to be suspicious, because the ordinary algebra we are used to does not necessarily apply to
operators. For example, matrices can be thought of as operators on vectors, and we know that the matrix
multiplication is not commutative:
AB ̸= BA.
One consequence of this is that the usual binomial formula is invalid for matrices (and other kinds of operators
as well). Indeed,
(A + B)2 = (A + B)(A + B) = A(A + B) + B(A + B) = A2 + AB + BA + B 2 ,
and this is not equal to
A2 + 2AB + B 2 ,
unless the two matrices commute (i.e. AB = BA).
Fortunately, a similar issue does not appear here. Indeed,
(2D)y = 2(Dy) = 2y ′ = (2y)′ = D(2y) = (D2)y,
so the operators 2D and D2 are equal, and we can conclude that
(D + 2)2 = D2 + 2D + D2 + 22 = D2 + 4D + 4
Note that this would be invalid for second order operators with nonconstant coefficients! For example,
(D2 + 2tD + t2 )y = y ′′ + 2ty ′ + t2 y,
but on the other hand
(D + t)2 y = (D + t)(y ′ + ty) = y ′′ + (ty)′ + ty ′ + t2 y = y ′′ + y + ty ′ + ty ′ + t2 y = y ′′ + 2ty ′ + (1 + t2 )y.
So we are using something very special about constant coefficient operators, when we factor like this.
Returning to the problem at hand, we now see that it is valid to replace the equation
(D2 + 4D + 4)y = 0
with the equation
(D + 2)(D + 2)y = 0.
This is helpful, because if we make the substitution
(D + 2)y = u,
16