Tag Archives: complex roots

Differential Equations – Complex Roots

In order to achieve complex roots, we have to look at the differential equation:
Ay” + By’ + Cy = 0

Then we look at the roots of the characteristic equation:
Ar² + Br + C = 0

After solving the characteristic equation the form of the complex roots of r1 and r2 should be:
λ ± μi

We refer back to the characteristic equation, we then assume that all the solution to the differential equation will be:
y(t) = e^(rt)

By plugging in our two roots into the general formula of the solution, we get:
y1(t) = e^(λ + μi)t
y2(t) = e^(λ – μi)t

Since these two functions are still in complex form, and we started the differential equation with real numbers. It would best if our solution is also real numbers. In order to transform the complex solution into a real solution, we need to use the Euler’s Formula.
e^(iƟ) = cosƟ + isinƟ
Another form of the Euler’s Formula is:
e^(-iƟ) = cos(-Ɵ) + isin(-Ɵ)
e^(-iƟ) = cosƟ – isin(Ɵ)

Now we split up both of the solutions into two parts one with real exponent and one with an imaginary exponent. Afterwards, we then applied Euler’s Formula.
y1(t) = e^(λt) e^(μit) = e^(λt) [cos(μt) + isin(μt)]
y2(t) = e^(λt) e^(-μit) = e^(λt) [cos(μt) – isin(μt)]

After doing all of that work, we are still left with a part of a complex solution. But we are able to get rid of the complex part. We can rewrite this solution as:
y(t) = c1y1(t) + c2y2(t)
y1(t) + y2(t) = 2e^(λt) cos(μt)
Simply it further:
u(t) = ½y1(t) + ½y2(t) = e^(λt) cos(μt)
Therefore:
c1 = c2 = ½

We can get the second part of the solution by subtracting the two original solutions:
y1(t) – y2(t) = 2ie^(λt) sin(μt)

At a glance, it still look like a complex solution, but looking at the two constants c1 and c2 we can result in a real solution by dividing it by 2i.
c1 = ½i
c2 = -½i

The second solution would be:
v(t) = ½i y1(t) – ½i y2(t) = e^(λt) sin(μt)

All of the work just to achieve the two real solutions:
u(t) = e^(λt) cos(μt)
v(t) = e^(λt) sin(μt)

These two real solutions is also a general solution. So if the roots of the characteristic equation result in r1,2 = λ ± μi, the general solution to this would be:
y(t) = c1 e^(λt) cos(μt) + c2 e^(λt) sin(μt)

EXAMPLE
y” – 10y’ + 29y = 0, y(0) = 1, y'(0) = 3

The characteristic equation would be:
r² – 10r + 29 = 0

The roots would be:
5 ± 2i

General Solution:
y(t) = e^(5t)[c1 cos(2t) + c2 sin(2t)]
y'(t) = 5e^(5t)[cos(2t) + c2 sin(2t)] + e^(5t)[-2 sin(2t) + 2c2 cos(2t)]

We use the initial values to find the constants.
y(0) = 1
We substitute all the t’s with 0’s and set it to equal 1.
y(0) = e^(5(0))[c1 cos(2(0)) + c2 sin(2(0))] = 1
1[c1(1) + c2(0)] = 1
“c1 = 1

y'(0) = 3
We substitute all the t’s with 0’s and set it to equal 3.
y'(0) = 5e^(5(0))[cos(2(0)) + c2 sin(2(0))] + e^(5(0))[-2 sin(2(0)) + 2c2 cos(2(0))] = 3
5[1 + 0] + 1[0 + 2c2] = 3
5 + 2c2 = 3
c2 = -1

Particular Solution:
y = e^(5t)[cos(2t) – sin(2t)]

If you still do not understand this study guide you may go visit these websites or you can personally ask me, they go into more detail solving the differential equation of complex root solutions with characteristic equation. Thank you.

Videos from Khan Academy
Part I

Part II

Part III

Complex Roots(of the Characteristic Equation)

Complex Roots relate to the topic of Second order Linear Homogeneous equations with constant coefficients. The Second Order linear refers to the equation having the setup formula of y”+p(t)y’ + q(t)y = g(t). Constant coefficients  are the values in front of the derivatives of y and y itself. Homogeneous means the equation is equal to zero.So a homogeneous equation would look like

y”+by’ + cy = 0 or y”+p(t)y’ + q(t)y = 0.

When solving the general solution of the homogeneous equation, we have to look for the characteristic equation which is r^2 + br + c = 0. But how would you get there? To start we are given the differential equation of

y = e^rt

which can be used to look for the characteristic equation. To start we look for the derivative of the differential equation then replace y and its derivatives with what we found which the homogeneous equation would look like

(r^2)e^rt + bre^rt + ce^rt = 0

You can see there is a common value which is e^rt you can factor out. Once you factor it out, you can divide e^rt and this is the characteristic equation,

r^2 + br + c = 0

The general solution of the homogeneous equation will have 3 different results depending on the roots which are found from the characteristic equation. This is where Complex Roots come into play. Complex Roots is when the roots have an imaginary number or i which looks like r =  a ± i*b. The general solution will be

 y = C1e^(a + ib)t + C2e^(a – ib)t

But there is a problem when you want to solve for a particular solution how would you get rid of the i? So the general solution is still not done. With the general solution, y = C1e^(a + ib)t + C2e^(a – ib)t you can right away separate the roots which will become y = C1e^at * e^(ibt) + C2e^at * e^( –ibt). Next you factor out e^a which makes the equation y = e^at(C1e^(ibt) + C2e^(-ibt). This is where the Euler’s Formula is introduced which says that

e^it = cos(t) + isin(t)

So the equation is written as such

y = e^at(C1(cos(bt) + isin(bt)) + C2(cos(-bt) + isin(-bt)))

This is where your knowledge of trig identities comes to play. Two trig identities which can be used are

sin(-x) = -sin(x), cos(-x) = -cos(x)

This would replace the negative values in the sin and cos but we are left with the complex part i. To get rid of the complex part, the equation is rearranged so that the constants can be created into a new constant that will work with real and imaginary numbers.

y = e^at((C1 + C2)cos(bt) + (C1i – C2i)isin(bt)))

y = e^at(Acos(bt) + Bsin(bt)), general solution

Sample Problem:                                                                                                                                y” + 4y’ + 5y = 0, y(0) = 1, y'(0) = 0                                                                                      y=e^(rt), y’ = re^(rt), y” = (r^2)e^(rt)                                                                                     e^(rt)*(r^2 + 4r +5) = 0

Since the roots are the easy to separate, we use the quadratic formula to look for the roots which you will end up with                                                                                            r = -2±i

The General Solution will equal to

y = e^(-2t)*(Acos(t) + Bsin(t))

For the particular solution, we solve for A and B using y(0) = 1 and y'(0) = 0.   y(0) = 1                                                                                                                                                      1 = e^(-2(0))*(Acos(0) + Bsin(0))                                                                                                1 = A

y'(0) = 0                                                                                                                                                  y’ = -2Ae^(-2t)cos(t) – Ae^(-2t)sin(t) – 2Be^(-2t)sin(t) + Be^(-2t)cos(t)                    0 = -2(1)e^(-2(0))cos(0) – (1)e^(-2(0))sin(0) – 2Be^(-2(0))sin(0) + Be^(-2(0))cos(0)                                                                                                                                 0 = -2 + B                                                                                                                                                B = 2

The Particular Solution:

y = e^(-2t)*(cos(t) + 2sin(t))

Videos to help with Studies:

Complex Roots Part 1

Complex Roots Part 2

Complex Roots Part 3