Drexel University, College of Engineering 2017-2018 Academic Year

ENGR 232 – Dynamic Engineering Systems

Homework 4: SOLUTIONS

1. Step-by-Step Problem: Consider the following second-order, linear differential equation:

𝐃𝐄: 𝑦 && + 3𝑦 & + 2𝑦 = 0 𝐈𝐂: 𝑦(0) = 1, 𝑦 & (0) = 2

Note it is homogeneous and has constant coefficients. Solve this IVP for y(t) by completing these steps.

a. The auxiliary (or characteristic) equation is:

𝑟 3 + 3𝑟 + 2 = (𝑟 + 1)(𝑟 + 2) = 0

b. The roots of the auxiliary equation are: 𝑟4 = −1, 𝑟3 = −2 The discriminant is: 𝐷 = 𝑏3 − 4𝑎𝑐 = 9 − 8 = 1 > 0

c. The general solution is: (use constants 𝑐4 and 𝑐3 ) 𝑦(𝑡) = 𝑐4 𝑒 @AB + 𝑐3 𝑒 @CB = 𝑐4 𝑒 DB + 𝑐3 𝑒 D3B

d. Now solve the IVP, that is, find the coefficients 𝑐4 and 𝑐3 so that 𝑦(0) = 1, 𝑦 & (0) = 2.

𝑦(0) = 𝑐4 + 𝑐3 = 1
& 𝑐4 = 4, 𝑐3 = −3
𝑦 (0) = −𝑐4 − 2𝑐3 = 2 Inverse matrix
1 1 𝑐4 𝑦(0) 1 1 𝑐4 1 𝑐4 4 −2 −1 1 4
or using matrices: E F G𝑐 H = E F à G H G𝑐 H = G H à G𝑐 H = ∙ G HG H = G H
𝑟4 𝑟3 3 𝑦′(0) −1 −2 3 2 3 D4 1 1 2 −3
The unique solution for this IVP is:
Displacement Phase Plot
DB D3B 1.5 2
𝑦(𝑡) = 4𝑒 − 3𝑒
1
1.5
y(t)

Here's a graph of the solution 𝑦(𝑡) 0.5

1
and its derivative. Uses code similar to lab 4.
0
0 2 4 6 phase
dy/dt

Time in seconds 0.5 IC

Velocity Equilibrium
2
0
1
y'(t)

0 -0.5

-1 -1
0 2 4 6 0 0.5 1 1.5
Time in seconds y

1
Drexel University, College of Engineering 2017-2018 Academic Year

We can get a fuller picture of this system using pplane8. First recast the 2nd-order DE as a first-order system, making the standard
substitutions:

𝑦4 = 𝑦, 𝑦3 = 𝑦′

Thus:

𝑑 𝑦4 (𝑡) 𝑦3 𝑦3 𝑦
E F = G H = G−2𝑦 − 3𝑦 H = G 0 1 H G𝑦4 H
𝑑𝑡 𝑦3(𝑡) 𝑦′′ 4 3 −2 −3 3

Enter this into dfield8 as follows:

Plot enough solutions to show the complete picture.

Notice how the curves cross each of the two

nullclines at their local maxima or minima.

The particular solution satisfying

𝐈𝐂: 𝑦(0) = 1, 𝑦 & (0) = 2 is shown in red.

Both roots are negative so this is a stable node (sink).

We can also see the phase portrait includes two straight lines

(show dotted in black) with slopes of -1 and -2.

These correspond to the two eigenvalues.

2
Drexel University, College of Engineering 2017-2018 Academic Year

2. Step-by-Step Problem: Consider the following second-order, linear differential equation:

37
𝐃𝐄: 𝑦 && + 𝑦 & + 𝑦 = 0 𝐈𝐂: 𝑦(0) = 1, 𝑦 & (0) = 2
4

Note it is homogeneous and has constant coefficients. Solve this IVP for y(t) by completing these steps.

37 4 4
a. The auxiliary (or characteristic) equation is: 𝑟3 + 𝑟 + 4
= M𝑟 + 3 − 3𝑗O M𝑟 + 3 + 3𝑗O = 0

Real part is negative, so the

4
b. The roots of the auxiliary equation are: 𝑟 = − ± 3𝑗 = 𝛼 ± 𝛽𝑗 oscillations decay in magnitude.
3

The discriminant is: 𝐷 = 𝑏3 − 4𝑎𝑐 = 1 − 37 = −36 < 0 Discriminant is negative, so roots

are complex conjugate pairs.

c. The general solution is: (use constants 𝑐4 and 𝑐3 ) 𝑦(𝑡) = 𝑒 DB/3 (𝑐4 cos 3𝑡 + 𝑐3 sin 3𝑡) = 𝑒 [B (𝑐4 cos 𝛽𝑡 + 𝑐3 sin 𝛽𝑡 )

d. Now solve the IVP, that is, find the coefficients 𝑐4 and 𝑐3 so that 𝑦(0) = 1, 𝑦 & (0) = 2.

𝑦(0) = 𝑐4 = 1 ]
&( \ 𝑐4 = 1, 𝑐3 = ^
𝑦 0) = − 3A + 3𝑐3 = 2

]
The unique solution for this IVP is: 𝑦(𝑡) = 𝑒 DB/3 Mcos 3𝑡 + sin 3𝑡O
^
Inverse matrix
1 0 𝑐4 𝑦(0) 1 0 𝑐4 1 𝑐4 4 3 0 1 1
or using matrices: E F G𝑐 H = E F à _− 4 3 ` G𝑐 H = G H à G𝑐 H = ∙ _ 4 1` G H = E5c F
𝛼 𝛽 3 𝑦′(0) 3 3 2 3 a
3 2 6

Displacement Phase Plot

Here's a graph of the solution 𝑦(𝑡) 2 2

1.5
and its derivative. Uses code similar to lab 4. 1
y(t)

1
0
0.5

-1 0
0 2 4 6 8
dy/dt

Time in seconds -0.5

Velocity phase
2 -1 IC
Equilibrium

0 -1.5
y'(t)

-2
-2
-2.5

-4 -3
0 2 4 6 8 -1 0 1 2
Time in seconds y
The solution spirals into the origin.

3
Drexel University, College of Engineering 2017-2018 Academic Year

We can get a fuller picture of this system using pplane8. First recast the 2nd-order DE as a first-order system, making the standard
substitutions:

𝑦4 = 𝑦, 𝑦3 = 𝑦′

Thus:
𝑦3 0 1 𝑦
𝑑 𝑦4 (𝑡) 𝑦3
E F = G H = d 37 e = d 37 e G 4H
𝑑𝑡 𝑦3 (𝑡) 𝑦′′ − 𝑦4 − 𝑦3 − −1 𝑦3
4 4

Enter this into dfield8 as follows:

Plot enough solutions to show the complete

picture. Notice how the curves cross each of the
two nullclines at their local maxima or minima.

The particular solution satisfying

𝐈𝐂: 𝑦(0) = 1, 𝑦 & (0) = 2 is shown in red.
The real part of the roots is negative so this is a
stable spiral.

4
Drexel University, College of Engineering 2017-2018 Academic Year

3. Step-by-Step Problem: Consider the following second-order, linear differential equation:

17
𝐃𝐄: 𝑦 && − 𝑦 & + 𝑦 = 0 𝐈𝐂: 𝑦(0) = −100, 𝑦 & (0) = 0
4

Note it is homogeneous and has constant coefficients. Solve this IVP for y(t) by completing these steps.

17 4 4
a. The auxiliary (or characteristic) equation is: 𝑟3 − 𝑟 + 4
= M𝑟 − 3 − 2𝑗O M𝑟 − 3 + 2𝑗O = 0

4 Real part is positive, so the

b. The roots of the auxiliary equation are: 𝑟 = ± 2𝑗 = 𝛼 ± 𝛽𝑗
3 oscillations increase in magnitude.

The discriminant is: 𝐷 = 𝑏3 − 4𝑎𝑐 = 1 − 17 = −16 < 0

Discriminant is negative, so roots
are complex conjugate pairs.

c. The general solution is: (use constants 𝑐4 and 𝑐3 ) 𝑦(𝑡) = 𝑒 B/3 (𝑐4 cos 2𝑡 + 𝑐3 sin 2𝑡) = 𝑒 [B (𝑐4 cos 𝛽𝑡 + 𝑐3 sin 𝛽𝑡)

d. Now solve the IVP, that is, find the coefficients 𝑐4 and 𝑐3 so that 𝑦(0) = −100, 𝑦 & (0) = 0.

𝑦(0) = 𝑐4 = −100
\ 𝑐4 = −100, 𝑐3 = 25
𝑦 & (0) = A + 2𝑐3 = 0
3

The unique solution for this IVP is: 𝑦(𝑡) = 𝑒 B/3 (−100 cos 2𝑡 + 25 sin 2𝑡)
Inverse matrix
1 0 𝑐4 𝑦(0) 1 0 𝑐4 −100 𝑐4 4 2 0 −100 −100
or using matrices: E F G𝑐 H = E F à _4 ` G𝑐3 H = G H à G𝑐3 H = ∙ _ 4 `G H=G H
𝛼 𝛽 3 𝑦′(0) 3
2 0 3 − 3
1 0 25

Displacement Phase Plot

2000 2000
Here's a graph of the solution 𝑦(𝑡) 1000
1000
y(t)

and its derivative. Phase plot of 0

the derivative versus 𝑦(𝑡) is also shown. -1000

0
-2000
0 2 4 6
dy/dt

Time in seconds -1000

Velocity
2000

1000 -2000
y'(t)

0
-3000
-1000 phase
IC
-2000
0 2 4 6 -4000
-2000 0 2000
Time in seconds y
The solution grows unbounded,
oscillating between ±∞.
5
Drexel University, College of Engineering 2017-2018 Academic Year

We can get a fuller picture of this system using pplane8. First recast the 2nd-order DE as a first-order system, making the standard
substitutions:

𝑦4 = 𝑦, 𝑦3 = 𝑦′

Thus:
𝑦3 0 1 𝑦
𝑑 𝑦4 (𝑡) 𝑦3 4
E F = G H = d 17 e = d 17 eG H
𝑑𝑡 𝑦3 (𝑡) 𝑦′′ − 𝑦4 + 𝑦3 − +1 𝑦3
4 4

Enter this into dfield8 as follows:

Plot enough solutions to show the

complete picture. Notice how the
curves cross each of the two
nullclines at their local maxima or
minima.

The particular solution satisfying

:𝐈𝐂: 𝑦(0) = −100, 𝑦 & (0) = 0
is shown in red.
The real part of the roots is positive so
this is an unstable spiral.

6
Drexel University, College of Engineering 2017-2018 Academic Year

4. Now complete this table to solve and summarize four similar problems all at once! For each equation, use the same initial
conditions: 𝐈𝐂: 𝑦(0) = 1, 𝑦 & (0) = 2. In the column marked Roots & Stability, also state whether the DE is stable or unstable.
It is unstable, if any of the roots is positive, or if any complex root has a positive real part.

Differential Equation Characteristic Roots & Stability General Solution Solution of IVP
Equation (Write the root twice if 𝑦(0) = 1, 𝑦 & (0) = 2
𝑎𝑟 3 + 𝑏𝑟 + 𝑐 = 0 it is repeated.) (Same for all parts.)

a. 𝑦 && − 5𝑦′ + 6𝑦 = 0 𝑟 3 − 5𝑟 + 6 = 0 𝑟 = 2, 𝑟 = 3 𝑦(𝑡) = 𝑐4 𝑒 3B + 𝑐3 𝑒 aB 𝑦(𝑡) = 𝑒 3B

Unstable

b. 𝑦 && + 2𝑦′ = 0 𝑟 3 + 2𝑟 = 0 𝑟 = 0, 𝑟 = −2 𝑦(𝑡) = 𝑐4 + 𝑐3 𝑒 D3B 𝑦(𝑡) = 2 − 𝑒 D3B

Stable

c. 𝑦 && − 2 𝑦 & + 5𝑦 = 0 𝑟 3 − 2𝑟 + 5 = 0 𝑟 = 1 + 2𝑖, 𝑟 = 1 − 2𝑖 𝑦 = 𝑐4 𝑒 B cos 2𝑡 𝑒B

Unstable 𝑦 = 𝑒 B cos 2𝑡 + sin 2𝑡
+𝑐3 𝑒 B sin 2𝑡 2

d. 𝑦 && + 2 𝑦 & + 5𝑦 = 0 𝑟 3 + 2𝑟 + 5 = 0 𝑟 = −1 + 2𝑖, 𝑟 = −1 − 2𝑖 𝑦 = 𝑐4 𝑒 DB cos 2𝑡 3𝑒 DB

Stable 𝑦 = 𝑒 DB cos 2𝑡 + sin 2𝑡
+𝑐3 𝑒 DB sin 2𝑡 2

Attempt all calculations by hand. Show details here. This will likely take several pages.

Be sure to summarize your calculations in the above table!

Work for Problem 4a.

All the steps can be seen at a glance, except the calculation of the constants 𝑐4 and 𝑐3 .
From 𝑦(𝑡) = 𝑐4 𝑒 3B + 𝑐3 𝑒 aB , the derivative is: 𝑦′(𝑡) = 2𝑐4 𝑒 3B + 3𝑐3 𝑒 aB
From the initial condition, 𝑦(0) = 1 we must have: Eqn1: 1 = 𝑐4 + 𝑐3
From the initial condition, 𝑦′(0) = 2 we must have: Eqn2: 2 = 2𝑐4 + 3𝑐3
Subtracting twice equation 1 from equation 2 eliminates 𝑐4 and shows 𝑐3 = 0.
Therefore 𝑐4 = 1 and 𝑐3 = 0.
The solution satisfying this IVP is: 𝑦(𝑡) = 𝑒 3B

Work for Problem 4b.

All the steps can be seen at a glance, except the calculation of the constants 𝑐4 and 𝑐3 .
From, 𝑦(𝑡) = 𝑐4 + 𝑐3 𝑒 D3B the derivative is: 𝑦 & (𝑡) = −2𝑐3 𝑒 D3B
From the initial condition, 𝑦(0) = 1 we must have: Eqn1: 1 = 𝑐4 + 𝑐3
From the initial condition, 𝑦′(0) = 2 we must have: Eqn2: 2 = −2𝑐3
Therefore 𝑐4 = 2 and 𝑐3 = −1.
The solution satisfying this IVP is: 𝑦(𝑡) = 2 − 𝑒 D3B

7
Drexel University, College of Engineering 2017-2018 Academic Year

Work for Problem 4c.

All the steps can be seen at a glance, except the calculation of the constants 𝑐4 and 𝑐3 .
From 𝑦 = 𝑐4 𝑒 B cos 2𝑡 + 𝑐3 𝑒 B sin 2𝑡, the derivative is: 𝑦 & (𝑡) = 𝑒 B [ (𝑐4 + 2𝑐3 ) cos 2𝑡 + (−2𝑐4 + 𝑐3 ) sin 2𝑡 ]
We used the product and chain rules to find the above derivative.

From the initial condition, 𝑦(0) = 1 we must have: Eqn1: 1 = 𝑐4

From the initial condition, 𝑦′(0) = 2 we must have: Eqn2: 2 = 𝑐4 + 2𝑐3
Therefore 𝑐4 = 1 and 𝑐3 = 1/2.
𝑒𝑡
The solution satisfying this IVP is: 𝑦 = 𝑒 B cos 2𝑡 + 2 sin 2𝑡

Work for Problem 4d.

All the steps can be seen at a glance, except the calculation of the constants 𝑐4 and 𝑐3 .
From, 𝑦 = 𝑐4 𝑒 DB cos 2𝑡 + 𝑐3 𝑒 DB sin 2𝑡 the derivative is: 𝑦 & (𝑡) = 𝑒 DB [ (−𝑐4 + 2𝑐3 ) cos 2𝑡 + (−2𝑐4 − 𝑐3 ) sin 2𝑡 ]
We used the product and chain rules to find the above derivative.

From the initial condition, 𝑦(0) = 1 we must have: Eqn1: 1 = 𝑐4

From the initial condition, 𝑦′(0) = 2 we must have: Eqn2: 2 = −𝑐4 + 2𝑐3
Therefore 𝑐4 = 1 and 𝑐3 = 3/2.
3𝑒𝑡
The solution satisfying this IVP is: 𝑦 = 𝑒 B cos 2𝑡 + 2 sin 2𝑡

8
Drexel University, College of Engineering 2017-2018 Academic Year

5. Use MATLAB's dsolve command to find the exact solution for each problem. Complete the table to summarize your
work. These same equations reappear in the next problem. You do not have to solve these by hand this week, but you
should be able to on the next homework. Consider this a preview. These problems use different initial conditions.

2nd-Order DE Exact Solution to the IVP

a. 𝑦 && + 𝑦 & − 2𝑦 = 2𝑡 𝑒 D3B

𝑦(0) = 0, 𝑦′(0) = 1 𝑦(𝑡) = 𝑒 B − − 𝑡 − 1/2
2

b. 𝑦 && + 4𝑦 = 𝑡 3 + 3𝑒 B 7 19 𝑒 D3B 3 𝑡3 1
𝑦 (𝑡 ) = sin 2𝑡 − cos 2𝑡 − + 𝑒 B + −
𝑦(0) = 0, 𝑦′(0) = 2
10 40 2 5 4 8

c. 𝑦 && − 2𝑦 & + 𝑦 = 𝑡 𝑒 B + 4 𝑡a B
𝑦(0) = 1, 𝑦′(0) = 1 𝑦(𝑡) = −3𝑒 B + 4𝑡 𝑒 B + 𝑒 +4
6

Here is a block of MATLAB code that will help with the first one in the table.
%% 5a. Solve the DE exactly.
syms y(t)
a = 1; b = 1; c = -2;
Dy = diff(y,t); D2y = diff(y,t,t);
DE = a*D2y + b * Dy + c * y == 2*t
dsolve(DE, y(0)==0, Dy(0)==1)

Some examples look better after calling simplify. There are no hand calculations to show for this problem.
Just complete the above table.

%% 5b. Solve the DE exactly.

syms y(t)
a = 1; b = 0; c = 4;
Dy = diff(y,t); D2y = diff(y,t,t);
DE = a*D2y + b * Dy + c * y == t^2 + 3 * exp(t)
dsolve(DE, y(0)==0, Dy(0)==2)
simplify(ans)

%% 5c. Solve the DE exactly.

syms y(t)
a = 1; b = -2; c = 1;
Dy = diff(y,t); D2y = diff(y,t,t);
DE = a*D2y + b * Dy + c * y == t * exp(t) + 4
dsolve(DE, y(0)==1, Dy(0)==1)
simplify(ans)

9
Drexel University, College of Engineering 2017-2018 Academic Year

State Space Form: Consider the following second-order non-homogeneous differential equation.

𝐃𝐄: 𝑎𝑦 && + 𝑏𝑦 & + 𝑐𝑦 = 𝐹 (𝑡), 𝐈𝐂: 𝑦(0) = 𝑦j , 𝑦′(0) = 𝑦′j

Similar to the approach described in the Lab, we can convert the given 2nd-0rder DE to a 1st-order system by making the
lm
following substitutions. Let 𝑥4 = 𝑦 and let 𝑥3 = . Collect these two new variables into a column vector, which is
lB
called the state vector.
𝑥4
𝐱o⃗ = G𝑥 H
3
nd lqC lC m
Differentiating the state vector, and using the original 2 -0rder DE to eliminate = we find:
lB lB C

l 𝑥̇ 𝑥3 0 1 𝑥4 0
o⃗ = E 1 F = _s(B) \
𝐱 `=_ \ +E
− t G𝑥3 H F
u u`
lB 𝑥̇ 2 t
− t 𝑥4 − t 𝑥3 − t 𝐹(𝑡)/𝑎
Thus if we define:

0 1
0
𝐴 = E− 𝑐 −
𝑏 F and 𝐵 = G H and u = 𝐹(𝑡)/𝑎
1
𝑎 𝑎

The state space equations are:

𝑑
𝐱
o⃗ = 𝐴 𝐱
o⃗ + 𝑩 𝑢
𝑑𝑡
6. Complete the table below, to rewrite each of these 2nd-order non-homogeneous differential equations in the above
form. Give the matrix A, the column vector B and the scalar function u for each. Each 2nd-order system becomes
equivalent to the 1st-order system:

𝑑
The table can be 𝐱
o⃗ = 𝐴 𝐱
o⃗ + 𝑩 𝑢
completed by inspection. 𝑑𝑡 The first row has been done for you.
2nd-Order DE Matrix A Column Vector B Scalar u

a. 𝑦 && + 𝑦 & − 2𝑦 = 2𝑡 𝟎 𝟏 𝟎 𝒖 = 𝟐𝒕
𝑨=G H 𝑩=G H
𝑦(0) = 0, 𝑦′(0) = 1 𝟐 −𝟏 𝟏

b. 𝑦 && + 4𝑦 = 𝑡 3 + 3𝑒 B 𝟎 𝟏H 𝟎 𝑢 = 𝑡 3 + 3𝑒 B
𝑨=G 𝑩=G H
−𝟒 𝟎 𝟏
𝑦(0) = 0, 𝑦′(0) = 2

c. 𝑦 && − 2𝑦 & + 𝑦 = 𝑡 𝑒 B + 4 𝟎 𝟏H 𝟎 𝒖 = 𝑡 𝑒 B + 4
𝑨=G 𝑩=G H
𝑦(0) = 1, 𝑦′(0) = 1 −𝟏 𝟐 𝟏

10
Drexel University, College of Engineering 2017-2018 Academic Year

7. Component Plots and Phase Plots - Optional

Attempt to solve each of the three differential equations from problem 6 qualitatively, using ode45. Using MATLAB,
create component plots and a phase plot for each equation. You could use code similar to Lab 4.

Component and Phase Plot for 6a. DE: 𝑦 && + 𝑦 & − 2𝑦 = 2𝑡 IC: 𝑦(0) = 0, 𝑦′(0) = 1

y Phase Plot
2.5 4
phase
IC
2 Equilibrium

3.5
1.5
y(t)

1 3
Note the solution is unstable
0.5 and marches off to infinity. The
2.5 "equilibrium" point, only
0 applies to the homogeneous
0 0.5 1 1.5
Time in seconds part of the DE.
dy/dt

Derivative of y
4
1.5
3.5

2.5 1
y'(t)

1.5 0.5

0.5 0
0 0.5 1 1.5 0 1 2 3
Time in seconds y

11
Drexel University, College of Engineering 2017-2018 Academic Year

Component and Phase Plot for 6b. DE: 𝑦 && + 4𝑦 = 𝑡 3 + 3𝑒 B IC: 𝑦(0) = 0, 𝑦′(0) = 2
Optional (Not graded)

y Phase Plot
14 15

12

10

8
y(t)

4
10 Note the solution is unstable
2
and marches off to infinity. The
0 "equilibrium" point, only
0 1 2 3
Time in seconds applies to the homogeneous
dy/dt

part of the DE.

Derivative of y
16 phase
IC
14 Equilibrium
5
12

10
y'(t)

2 0
0 1 2 3 0 5 10 15
Time in seconds y

12
Drexel University, College of Engineering 2017-2018 Academic Year

Component and Phase Plot for 6c. DE: 𝑦 && − 2𝑦 & + 𝑦 = 𝑡 𝑒 B + 4 IC: 𝑦(0) = 1, 𝑦′(0) = 1
Optional (Not graded)

y Phase Plot
8 16

6 14

5
y(t)

4
12

2
10
1
0 0.5 1 Note the solution is unstable
Time in seconds and marches off to infinity. The
dy/dt

8
"equilibrium" point, only
phase
Derivative of y
IC applies to the homogeneous
Equilibrium
16
part of the DE.
14 6

12

10
4
y'(t)

6
2
4

0 0
0 0.5 1 0 2 4 6 8
Time in seconds y

13
Drexel University, College of Engineering 2017-2018 Academic Year

Appendix: Component and phase plots for the four homogeneous second-order DE's in Question 4.
Just for insight. Not required.

y Phase Plot
8 15

DE #4a: 𝑦 && − 5𝑦′ + 6𝑦 = 0 7

5
y(t)

2
10
Notice the solution does not
1 approach the equilibrium
0 0.5 1
Time in seconds value at y = 0, but instead

dy/dt
phase marches off towards infinity.
Derivative of y IC
16 Equilibrium

14 Unstable!
5
12

10
y'(t)

2 0
0 0.5 1 0 2 4 6 8
Time in seconds y

y Phase Plot
2 2

1.8
1.8

1.6 phase
IC
1.6
y(t)

Equilibrium
1.4
1.4
Phase plot is a straight line,
&&
DE #4b: 𝑦 + 2𝑦′ = 0
1.2 for all initial conditions, for this
1
1.2 particular DE.
0 2 4 6
Time in seconds Stable!
dy/dt

1
Derivative of y
2
0.8 Stable!
1.5
0.6
y'(t)

1 0.4

0.5 0.2

0 0
0 2 4 6 0 0.5 1 1.5 2
Time in seconds y

14
Drexel University, College of Engineering 2017-2018 Academic Year

y Phase Plot
2 25

20
-2 phase
IC

y(t)
Equilibrium
-4
15
-6 Phase plot marches off towards
DE #4c: 𝑦 && − 2 𝑦 & + 5𝑦 = 0 infinity.
-8 10
0 0.5 1 1.5 2 2.5
Time in seconds Unstable!

dy/dt
Derivative of y 5
25

20

15
0
10
y'(t)

5
-5
0

-5

-10 -10
0 0.5 1 1.5 2 2.5 -8 -6 -4 -2 0 2
Time in seconds y

y Phase Plot
1.5 2
phase
IC
Equilibrium
1
1.5
y(t)

0.5
1
DE #4d: 𝑦 && + 2 𝑦 & + 5𝑦 = 0
0

0.5 Phase plot converges to

-0.5
0 1 2 3 4 5 equilibrium point.
Time in seconds
Stable!
dy/dt

Derivative of y
2

1.5 -0.5

0.5 -1
y'(t)

-0.5
-1.5
-1

-1.5

-2 -2
0 1 2 3 4 5 -0.5 0 0.5 1 1.5
Time in seconds y

15