Scribd
Upload
16 views5 pages

Solving Systems of Differential Equations

The document discusses solutions to problems related to systems of first-order differential equations. It provides worked out solutions for 6 problems involving representing equations as systems, finding fundamental matrices and solutions, and solving first-order systems.

Uploaded by

Shriyansh Raj
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
16 views5 pages

Solving Systems of Differential Equations

The document discusses solutions to problems related to systems of first-order differential equations. It provides worked out solutions for 6 problems involving representing equations as systems, finding fundamental matrices and solutions, and solving first-order systems.

Uploaded by

Shriyansh Raj
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

MA 102 (Mathematics II - Ordinary Differential Equations)

Department of Mathematics, IIT Guwahati


Semester: January-April, 2024

Solutions of Tutorial Sheet No. 5 Date of Discussion: April 19, 2024

Lectures 15-16: Systems of first-order equations, Matrix form of first-order


systems

1. Represent each of the following equations in terms of an equivalent system of first-order


equations:
(a) x′′ − t2 x′ − tx = 0, (b) x′′′ = x′′ − t2 (x′ )2 , (c) x′′ − 3x′ − 11x = sin t.
Further, write the derived systems in matrix form x̄′ = Ax̄ + b̄ (if possible).
Solution: (a) It can be written as a system as follows:
dx
= y,
dt
dy
= tx + t2 y.
dt
This system has the matrix form as follows:
[ ]′ [ ][ ]
x 0 1 x
= 2 .
y t t y

dx d2 y dy dy
(b) Take = y to get 2 = − t2 y 2 . Next, take = z to get the system as
dt dt dt dt
dx
= y,
dt
dy
= z,
dt
dz
= z − t2 y 2 .
dt
It is not possible to represent this system in x̄′ = Ax̄ + b̄ form since y appears nonlin-
early.
(c) It can be written as a system as follows (non-homogeneous):
dx
= y,
dt
dy
= 11x + 3y + sin t.
dt
This system has the matrix form as follows:
[ ]′ [ ][ ] [ ]
x 0 1 x 0
= + .
y 11 3 y sin t

1
2. If a particle of mass moves in the xy-plane, its equations of motion are

d2 x d2 y
m 2 = f (t, x, y), m 2 = g(t, x, y),
dt dt
where f and g represent the x and y components, respectively, of the force acting
on the particle. Express this system of two second-order equations by an equivalent
system of four first-order equations.
Solution: It is easy to follow that the required system is
dx
= u,
dt
du f (t, x, y)
= ,
dt m
dy
= v,
dt
dv g(t, x, y)
= .
dt m

3. The vector functions x¯1 = [e−t , 2e−t , e−t ]T , x¯2 = [et , 0, et ]T , x¯3 = [e3t , −e3t , 2e3t ]T are
solutions to the system x̄′ (t) = Ax̄(t). Determine whether they form a fundamental
solution set. If they do, find a fundamental matrix for the system and give a general
solution.
Solution:
Given x¯1 = [e−t , 2e−t , e−t ]T , x¯2 = [et , 0, et ]T , x¯3 = [e3t , −e3t , 2e3t ]T .
Now,
e−t et e3t
W (x¯1 , x¯2 , x¯3 ) = 2e−t 0 −e3t = −2e3t ̸= 0.
e−t et 2e3t
Therefore, the given set of vectors is linearly independent and so it forms a fundamental
solution set. A fundamental matrix is given by
 −t 
e et e3t
Φ(t) =  2e−t 0 −e3t  .
e−t et 2e3t

The general solution is given by


   t   3t 
e−t e e
x̄(t) = Φ(t)c̄ = c1  2e−t  + c2  0  + c3  −e3t  .
e−t et 2e3t

4. Find the general solution of the following first-order equations:


{ dx
dt
= 4x − y,
dy
dt
= 2x + y.

Solution:
Given { dx
dt
= 4x − y,
dy
dt
= 2x + y.

2
The auxiliary equation is m2 − 5m + 6 = 0 giving the roots as m1 = 2, m2 = 3. The
two linearly independent solutions are
} }
x = A1 e2t x = A2 e3t
, .
x = B1 e2t x = B2 e3t

Using the relation (a1 − m)A + b1 B = 0, we get A1 = 1, B1 = 2, A2 = 1, B2 = 1. Hence,


the general solution can be written as
{
x(t) = c1 e2t + c2 e3t ,
y(t) = 2c1 e2t + c2 e3t .

5. Find the general solution of the following first-order equations:


{ dx
dt
= x + 3y,
dy
dt
= 3x + y.

Then, find the particular solution given that x(0) = 5 and y(0) = 1.
Solution:
Given { dx
dt
= x + 3y,
dy
dt
= 3x + y.
The auxiliary equation is m2 − 2m − 8 = 0 giving the roots as m1 = −2, m2 = 4. The
two linearly independent solutions are
} }
x = A1 e−2t x = A2 e4t
, .
x = B1 e−2t x = B2 e4t

Using the relation (a1 − m)A + b1 B = 0, we get A1 = 1, B1 = −1, A2 = 1, B2 = 1.


Hence, the general solution can be written as
{
x(t) = c1 e−2t + c2 e4t ,
y(t) = −c1 e−2t + c2 e4t .

using the initial conditions x(0) = 5, y(0) = 1, we get c1 = 2, c2 = 3.


Therefore, the specific solution is given by
{
x(t) = 2e−2t + 3e4t ,
y(t) = −2e−2t + 3e4t .

6. Solve the given equation x̄′ (t) = Ax̄(t) for


   
0 −2 0 −1 1 −2
(a) A =  1 2 0 , (b) A =  0 −1 4 .
0 0 −2 0 0 1
Solution: The solution is given by

x̄ = αeλt c̄.

We have to find the values of λ from the characteristic equation det(λI − A) = 0 and
the column vectors c̄ from (λI − A) = 0̄.
(a) The characteristic equation can be found as

(2 + λ)(λ2 − 2λ + 2) = 0

3
We get the characteristic values as λ = −2, 1 + i, 1 − i.
Let ū be the column vector corresponding to λ = −2. This can be obtained as

ū = [0 0 1]T .

Let v̄ be the column vector corresponding to λ = 1 + i. This can be obtained as

v̄ = [1 − i −1 0]T .

Let w̄ be the column vector corresponding to λ = 1 − i. This can be obtained as

w̄ = [1 + i −1 0]T .

Hence, the general solution can be written as

x̄ = α1 ūe−2t + α2 v̄e(1+i)t + +α3 w̄e(1−i)t .

(b) The characteristic equation gives the characteristic values as λ = −1 (repeated)


and λ = 1.
Let ū be the column vector corresponding to λ = −1. This can be obtained as

ū = [1 0 0]T .

Let v̄ be another column vector corresponding to the same root λ = −1. This can be
obtained as
v̄ = [t 0 0]T .
Let w̄ be the column vector corresponding to λ = 1. This can be obtained as

w̄ = [0 2 1]T .

Hence, the general solution can be written as

x̄ = α1 ūe−t + α2 v̄e−t + +α3 w̄et .

7. Show that the solutions for the system


{ dx
dt
= x + 2y,
dy
dt
= 3x + 2y.
{ {
x = 2e4t , x = e−t ,
are given by and
y = 3e4t , y = −e−t .
Further, show that the solutions of the above system can also be obtained by
(i) differentiating the first equation with respect to t and then eliminating y,
(ii) differentiating the second equation with respect to t and then eliminating x.
Solution: For the first part, proceed exactly in the same manner as in Problems 4
and 5. Then,
(i) Differentiate the first equation of the system to get
d2 x dx dy
2
= +2 . (1)
dt dt dt
Further, subtracting the first equation of the system from the second:
dy dx
= + 2x. (2)
dt dt
4
dy
Eliminating between equations (1) and (2), we get the following second-order equa-
dt
tion in x:
d2 x dx
− 3 − 4x = 0.
dt2 dt
Its general solution is
x(t) = c1 e4t + c2 e−t .
(ii) Differentiate the second equation of the system to get

d2 y dx dy
2
=3 +2 . (3)
dt dt dt
Further, subtracting the second equation of the system from the first one multiplied
by 3:
dx dy
3 = + 4y. (4)
dt dt
dx
Eliminating between equations (3) and (4), we get the following second-order equa-
dt
tion in y:
d2 y dy
− 3 − 4y = 0.
dt2 dt
Its general solution is
y(t) = c1 e4t + c2 e−t .

8. We know that, for a system of two first-order equations, the complex roots a ± ib of
the auxiliary equations lead to the following solutions:
}
x = eat (A1 cos bt − A2 sin bt),
y = eat (B1 cos bt − B2 sin bt),

and }
x = eat (A1 sin bt + A2 cos bt),
y = eat (B1 sin bt + B2 cos bt).
Find the Wronskian of these solutions. Check whether it is mandatory that
A1 B2 − A2 B1 ̸= 0.
Solution:
We have
x1 (t) x2 (t)
W (x(t), y(t)) =
y1 (t) y2 (t)
eat (A1 cos bt − A2 sin bt) eat (A1 sin bt + A2 cos bt)
=
eat (B1 cos bt − B2 sin bt) eat (B1 sin bt + B2 cos bt)
= e2at (A1 B2 − A2 B1 ).

In order that the solutions are linearly independent, we must have A1 B2 − A2 B1 ̸= 0.

The End

Common questions

Powered by AI

The characteristic equation is crucial for determining the eigenvalues \(\lambda\) of the matrix \(A\) in a system \(\bar{x}'(t) = A\bar{x}(t)\). These eigenvalues define the behavior and type of solution for the differential equation system. The solutions are built from expressions \(\alpha e^{\lambda t}\bar{c}\), where \(\bar{c}\) corresponds to the eigenvectors associated with each \(\lambda\). This framework allows for constructing the general solution, taking into account the nature of real and complex eigenvalues .

To find a particular solution for a linear system using differential transformation, begin by determining the general solution in terms of arbitrary constants. Convert the system into a higher-order differential equation if necessary. Solve for the general solution, typically in exponential terms due to the eigenstructure. Apply initial conditions to compute the constants: substitute initial values into the general solution expressions, yielding linear equations that determine the constants' values. For the specific system \(dx/dt = x + 3y\), \(dy/dt = 3x + y\) with \(x(0) = 5, y(0) = 1\), the general solutions \(x(t)\) and \(y(t)\) incorporate constants resolved using the initial conditions to form a particular solution .

To determine the general solution of a differential equation system \(\bar{x}'(t) = A\bar{x}(t)\), calculate the eigenvalues \(\lambda\) from the characteristic equation \(det(\lambda I - A) = 0\), and find corresponding eigenvectors. Use these to form the general solution \(\bar{x}(t) = \alpha_1 \bar{u} e^{\lambda_1 t} + \alpha_2 \bar{v} e^{\lambda_2 t} + ...\), where \(\bar{u}\) and \(\bar{v}\) are eigenvectors. This allows constructing solutions in terms of exponential functions according to the eigenspectrum of \(A\).

For the Wronskian of solutions with complex roots to be non-zero, it must hold that \(A_1B_2 - A_2B_1 \neq 0\). This condition ensures the linear independence of the solutions. The Wronskian \(W(x(t), y(t))\) for solutions \(x = e^{at}(A_1 \cos bt - A_2 \sin bt)\), \(y = e^{at}(B_1 \cos bt - B_2 \sin bt)\) simplifies to \(e^{2at}(A_1B_2 - A_2B_1)\), necessitating that the determinant of the coefficient matrix formed by \(A_1, A_2, B_1, B_2\) must be non-zero to prevent the solutions from being linearly dependent .

For the system \(\{ dx/dt = x + 2y, dy/dt = 3x + 2y \}\), transform to second-order by differentiating one equation and substituting from the other. For instance, differentiate \(dx/dt = x + 2y\) to get \(d^2x/dt^2 = dx/dt + 2dy/dt\). Using \(dy/dt = 3x + 2y\), substitute for \(dy/dt\) to eliminate it, resulting in \(d^2x/dt^2 - 3dx/dt - 4x = 0\). Similarly, differentiate and substitute for \(x\) from the first to form a second-order equation in \(y\): \(d^2y/dt^2 - 3dy/dt - 4y = 0\). The general solutions for both are \(x(t) = c_1 e^{4t} + c_2 e^{-t}\), \(y(t) = c_1 e^{4t} + c_2 e^{-t}\).

Auxiliary equations play a central role by providing the characteristic roots \(m\) necessary to form the solution basis for systems of first-order linear homogeneous differential equations. Solutions typically follow exponential forms driven by these roots, allowing for the construction of the complete solution set for the differential system. The roots indicate the type and nature of the solutions (real or complex, distinct or repeated), impacting the behavior of the exponential functions—such as growth rates or oscillations—comprising the solution .

To convert a system of two second-order differential equations into a system of four first-order equations, define new variables for the first derivative of each position variable. For the given motion equations \(md^2x/dt^2 = f(t, x, y)\) and \(md^2y/dt^2 = g(t, x, y)\), introduce \(u = dx/dt\) and \(v = dy/dt\). The first-order system becomes \(dx/dt = u\), \(du/dt = f(t, x, y)/m\), \(dy/dt = v\), and \(dv/dt = g(t, x, y)/m\).

For the system \( \{ dx/dt = x + 3y, dy/dt = 3x + y \} \), with initial conditions \(x(0) = 5\), \(y(0) = 1\), solve the auxiliary equation \(m^2 - 2m - 8 = 0\) to find roots \(m_1 = -2\) and \(m_2 = 4\). The general solution is \(x(t) = c_1 e^{-2t} + c_2 e^{4t}\), \(y(t) = -c_1 e^{-2t} + c_2 e^{4t}\). Applying the initial conditions gives \(c_1 = 2\), \(c_2 = 3\). The specific solution is \(x(t) = 2e^{-2t} + 3e^{4t}\), \(y(t) = -2e^{-2t} + 3e^{4t}\).

A set of vector functions \(\bar{x_1}, \bar{x_2}, \bar{x_3}\) forms a fundamental solution set for a system if they are linearly independent. This condition can be verified by ensuring the Wronskian determinant of the vector functions is non-zero. In the provided example, the Wronskian \(W(\bar{x_1}, \bar{x_2}, \bar{x_3})\) was computed as \(-2e^{3t} \neq 0\), establishing their linear independence and confirming they form a fundamental solution set .

The second-order differential equation \(x'' - t^2 x' - tx = 0\) can be expressed as a system of first-order equations: \(\frac{dx}{dt} = y\) and \(\frac{dy}{dt} = tx + t^2 y\). In matrix form, this system is represented as \([x \ y]' = \begin{bmatrix} 0 & 1 \\ t & t^2 \end{bmatrix} [x \ y]\).

You might also like