• Skip to primary navigation
  • Skip to main content
  • Skip to primary sidebar
  • Skip to footer
  • Home
  • Collection of formulas
  • Basic engineering mathematics
    • Partial fractions
    • Differentiation
    • Differentiation applications
    • Integration
    • Integration applications
    • Complex numbers
    • Determinants
    • Series
    • Sequences
    • Vector
    • Multiple integrals
    • Polar curves
  • Advanced engineering mathematics
    • First order ODE
    • Second order ODE
    • Partial differentiation
    • Vector analysis
    • Difference equations and z-transform
    • Fourier series
    • Laplace transform
    • Matrix analysis
    • Multiple integration
    • Partial differential equations
    • Complex analysis

Engineering math blog

  • Bloglovin
  • Facebook
  • LinkedIn
  • Pinterest
  • Tumblr
  • Twitter

September 24, 2018 By Dr. Aspriha Peters Leave a Comment

Laplace transform to solve second-order differential equations

(Last Updated On: October 19, 2019)

Laplace transform to solve second-order differential equations. Hi guys, today I’ll talk about how to use Laplace transform to solve second-order differential equations. Have a look.




Laplace transform to solve second-order differential equations

Laplace transform to solve second-order differential equations

Now the standard form of any second-order ODE is

    \[ay^{''}+by^{'}+cy = d~\text{or}f(y).\]

Here a, b, c, d are constants and f(y) is a function of y.

In order to solve this equation in the standard way, first of all, I have to solve the homogeneous part of the ODE. After that, I’ll need to get the particular integral for the non-homogeneous part.

If interested, you can read more about it in the category of the second-order ODE.

But what would happen if I use Laplace transform to solve second-order differential equations

If I use Laplace transform to solve second-order differential equations, it can be quite a direct approach.

First of all, I don’t need to bother with the homogeneous or non-homogeneous part. It’s all the same.

Secondly, I don’t have to work on initial conditions separately. They will be part of the calculation.

In one of my earlier posts, I have already described in detail how to use Laplace transform to solve the first-order ODE.

For further reading, please check out how to use Laplace transform to solve the second-order homogeneous ODE

Now I’ll give an example of how to use Laplace transform to solve second-order differential equations.

Example of Laplace transform to solve second-order differential equations

Disclaimer: This example is not mine. I have chosen it from a book. I have also given the due reference at the end of the post.

So here is the example.





Example 

According to Stroud and Booth (2011) “Solve the equation by Laplace transform:

    \[x^{''} + 8x^{'}+32x = 32 \sin 4t,~~~~x_0 = 0, x_1 = 0.\]

”

Solution

Here the given differential equation with initial condition is

    \[ \boxed{x^{''} + 8x^{'}+32x = 32 \sin 4t,~~~~x_0 = 0, x_1 = 0.} \]

Now I have to solve this equation using Laplace transform.

Step 1

So it will be

    \[\mathcal{L}\{x^{''} + 8x^{'}+32x\} = \mathcal{L}\{32 \sin 4t\} .\]

This means

    \[(s^2\overline{x}-s x_{0} - x^{}'_{0}) +8(s\overline{x} - x_0)+32\overline{x} = 32\frac{4}{s^2+16}.\]

Now I already know from the given equation that x_{0} = 0,  x^{'}(0) = 0.

Therefore I’ll simplify this equation to get

    \begin{eqnarray*} (s^2\overline{x}-s .0 - 0) +8(s\overline{x} - 0)+32\overline{x} &=& \frac{128}{s^2+16}\\\overline{x} (s^2 +8s+32) &=& \frac{128}{s^2+16}\\ \overline{x}  &=& \frac{128}{(s^2+16)(s^2 +8s+32)}.\end{eqnarray*}

Thus the value of x will be

(1)   \begin{equation*}x = \mathcal{L}^{-1}\left\{ \frac{128}{(s^2+16)(s^2 +8s+32)}\right\}.\end{equation*}

So my next step is to find the inverse Laplace transform.

In order to get the inverse Laplace transform, first I have to use partial fractions of this expression.

Related post: partial fraction in inverse Laplace transform

As I can see, the factors in the denominator (s^2+16) and (s^2 +8s+32) cannot be reduced further.

So I’ll use the rule for partial fractions of irreducible factors.

Thus the expression \cfrac{128}{(s^2+16)(s^2 +8s+32)} will be a sum of two expressions as

(2)   \begin{equation*}\frac{128}{(s^2+16)(s^2 +8s+32)} = \frac{As+B}{s^2+16} + \frac{Cs+D}{s^2 +8s+32}.\end{equation*}

Also, A, B, C, D are constants.

Now I’ll find out the values of these constants.




Step 2

First of all, I’ll simplify the right-hand side of the equation (2).

Thus it will be

    \begin{eqnarray*}&&\frac{128}{(s^2+16)(s^2 +8s+32)}\\ &=& \frac{(As+B)(s^2 +8s+32)+(Cs+D)(s^2+16)}{(s^2+16)(s^2 +8s+32)}\\&=&\frac{s^3(A+C)+s^2(8A+B+D)+s(32A+8B+16C)+(32B+16D)}{(s^2+16)(s^2 +8s+32)}\end{eqnarray*}

Now I can compare the coefficients on both sides of the numerator.

I will start with the coefficients of s^3.

Thus it will be

    \[A+C = 0.\]

So this gives

(3)   \begin{equation*}C = - A.\end{equation*}

Next, it will be constants.

So it will look like

    \[32B+16D = 128.\]

And this gives 

(4)   \begin{equation*}D = 8-2B.\end{equation*}

Then I’ll compare the coefficients of s^2 to get

    \[8A+B+D = 0.\]

Now I’ll substitute the value from equation (4) to this equation to get

    \[8A+B+8-2B = 0.\]

And this gives 

(5)   \begin{equation*}8A-B = -8.\end{equation*}

In the end, I’ll compare the coefficients of s. Thus it will be

    \[32A+8B+16C = 0 \Rightarrow 4A +B +2C = 0.\]

Now from equation (3), I’ll substitute C = -A in this equation to get

    \[4A +B -2A = 0\Rightarrow 2A+B = 0.\]

So this gives 

(6)   \begin{equation*}B = -2A.\end{equation*}

Next, I’ll put equation (6) in equation (4) to get

    \[8A+2A= -8.\]

And this brings

    \[A = - \frac{4}{5}.\]

Then I can use equation (6) to say

    \[B = \frac{8}{5}.\]

Next, I can get the value of C from equation (3) as

    \[C = \frac{4}{5}.\]

At the end, I’ll get the value of D from equation (4) as

    \[D = 8-\frac{16}{5} = \frac{24}{5}.\]

Now I’ll put back the values of A, B, C and D in equation (2).

Step 3

Thus it will be

    \[ \boxed{\frac{128}{(s^2+16)(s^2 +8s+32)} = \frac{(-4/5)s+8/5}{s^2+16} + \frac{(4/5)s+24/5}{s^2 +8s+32}.} \]

Now I’ll simplify it to get 

    \[\frac{128}{(s^2+16)(s^2 +8s+32)} = -\frac{4}{5}\left(\frac{s-2}{s^2+16}\right) + \frac{4}{5}\left(\frac{s+6}{s^2 +8s+32}\right).\]

So this means 

    \[\frac{128}{(s^2+16)(s^2 +8s+32)} = \frac{4}{5}\left(\frac{s+6}{s^2 +8s+32} - \frac{s-2}{s^2+16}\right).\]

Therefore from equation (1), I can say that 

    \begin{equation*}x = \mathcal{L}^{-1}\left\{ \frac{4}{5}\left(\frac{s+6}{s^2 +8s+32} - \frac{s-2}{s^2+16}\right)\right\}.\end{equation*}

Next, I’ll use the formulas in Laplace transform to find the value of x.

So I can rewrite x as

    \[x = \frac{4}{5}\mathcal{L}^{-1}\left\{\frac{s+6}{s^2 +8s+32}\right\} - \frac{4}{5}\mathcal{L}^{-1}\left\{\frac{s-2}{s^2+16}\right\}.\]

Now, first of all, I’ll find out the value of \cfrac{4}{5}\mathcal{L}^{-1}\left\{\cfrac{s+6}{s^2 +8s+32}\right\}. Then I’ll find out the value of \cfrac{4}{5}\mathcal{L}^{-1}\left\{\cfrac{s-2}{s^2+16}\right\}.

Then let’s say,

    \[x_1= \frac{4}{5}\mathcal{L}^{-1}\left\{\cfrac{s+6}{s^2 +8s+32}\right\}\]

and

    \[x_2 = \cfrac{4}{5}\mathcal{L}^{-1}\left\{\cfrac{s-2}{s^2+16}\right\}.\]

Thus it will be

(7)   \begin{equation*}x = x_1-x_2.\end{equation*}

So, now it’s time to get the value of x_1.

Step 4

As I can see, the denominator of x_1 is (s^2 +8s+32). Now I can’t factorise it. But I can write it as a sum of two squares like

    \[s^2 +8s+32 = (s^2+8s+16)+(16).\]

So this gives

    \[s^2+8s+32=(s+4)^2+(4)^2.\]

Therefore I can write x_1 as

    \[x_1=\frac{4}{5}\mathcal{L}^{-1}\left\{\cfrac{s+6}{(s+4)^2+(4)^2}\right\}.\]

Now I’ll bring (s+4) in the numerator as well.

Thus it will be 

    \[x_1=\frac{4}{5}\mathcal{L}^{-1}\left\{\frac{(s+4)+2}{(s+4)^2+(4)^2}\right\}.\]

So I can rewrite x_1 as

(8)   \begin{equation*}x_1 = \frac{4}{5}\mathcal{L}^{-1}\left\{\frac{(s+4)}{(s+4)^2+(4)^2}+\frac{2}{(s+4)^2+(4)^2}\right\}.\end{equation*}

As I already know, the Laplace transform of \cos 4t is \cfrac{s}{s^2+4^2}.

Also, from the first shift theorem in Laplace transform, I already know that the Laplace transform of e^{-4t}\cos 4t is \cfrac{s+4}{(s+4)^2+4^2}.

So in this case, it will be

    \[\mathcal{L}^{-1}\left\{\frac{(s+4)}{(s+4)^2+(4)^2} \right\}= e^{-4t}\cos 4t.\]

Again, the Laplace transform of \sin 4t is \cfrac{4}{s^2+4^2} and together with the first shift theorem the Laplace transform of e^{-4t}\sin 4t becomes \cfrac{4}{(s+4)^2+4^2}.

Thus \mathcal{L}^{-1}\left\{\cfrac{2}{(s+4)^2+(4)^2}\right\} will be

    \[\mathcal{L}^{-1}\left\{\cfrac{2}{(s+4)^2+(4)^2}\right\} = \frac{1}{2}e^{-4t}\sin 4t.\]

Therefore from equation (8), I can say that x_1 will be

(9)   \begin{equation*}x_1 = \frac{4}{5} \left(e^{-4t}\cos 4t + \frac{1}{2}e^{-4t}\sin 4t\right).\end{equation*}

Now I’ll find the value of x_2.

Step 5 

As I can see, the denominator of x_2 is (s^2 +16). Now I can’t factorise it. But I can write it as a sum of two squares like

    \[s^2 +16 = s^2+(4)^2\]

Therefore I can write x_2 as

    \[x_2=\frac{4}{5}\mathcal{L}^{-1}\left\{\frac{s-2}{s^2+(4)^2}\right\}.\]

So this means

    \begin{equation*}x_2=\frac{4}{5}\mathcal{L}^{-1}\left\{\frac{s}{s^2+(4)^2}-\frac{2}{s^2+(4)^2}\right\}.\end{equation*}

Now I’ll use the formulas in Laplace transform.

As per these formulas, Laplace transform of \cos 4t is \cfrac{s}{s^2+4^2}. Also, the Laplace transform of \sin 4t is \cfrac{4}{s^2+4^2}.

So I can use these two formulas to get the value of x_2 as 

    \[x_2=\frac{4}{5}\mathcal{L}^{-1}\left\{\frac{s}{s^2+(4)^2}-\frac{1}{2}\frac{4}{s^2+(4)^2}\right\}.\]

Thus the value of x_2 will be

    \[x_2=\frac{4}{5}\cos 4t-\frac{4}{5}.\frac{1}{2}\sin 4t.\]

Now I’ll simplify it to get

(10)   \begin{equation*}x_2=\frac{4}{5}\cos 4t - \frac{2}{5}\sin 4t.\end{equation*}

Next, I’ll substitute the values of x_1 and x_2 from equations (9) and (10) respectively to equation (8) to get the value of x as

    \[x = \frac{4}{5} \left(e^{-4t}\cos 4t + \frac{1}{2}e^{-4t}\sin 4t\right) - \left(\frac{4}{5}\cos 4t - \frac{2}{5}\sin 4t\right).\]

Then I’ll simplify it to get

    \begin{eqnarray*}x &=& \frac{4}{5} e^{-4t}\cos 4t + \frac{2}{5}e^{-4t}\sin 4t-\frac{4}{5}\cos 4t+\frac{2}{5}\sin 4t\\&=&\frac{4}{5}\cos 4t( e^{-4t}-1)+\frac{2}{5}\sin 4t(e^{-4t}+1)\\&=&\frac{2}{5}\left[\frac{2}{5}\cos 4t( e^{-4t}-1)+\sin 4t(e^{-4t}+1)\right].\end{eqnarray*}

Hence I can conclude that this is the solution to the given example.


Dear friends, this is the end of today’s post on Laplace transform to solve second-order differential equations. Thank you very much for reading this. Please let me know how you feel about it. Soon I will be back again with a new post. Till then, bye, bye!!

*Reference: K. A. Stroud and Dexter J. Booth (2011): Advanced engineering mathematics, Industrial Press, Inc.; 5th edition (March 8, 2011), Chapter: Laplace transform 1,  Further problems 2, p. 90, Q. No. 2(b) (Example 1), Q. No. 2(a) (Example 2).

Filed Under: Advanced engineering mathematics, Laplace transform Tagged With: Laplace transform, Laplace transform in second-order ODE

About Dr. Aspriha Peters

Trained mathematician & hobby academic. Curious about nature as well as an aspiring blogger.

Copyright

All rights reserved© ‘Engineering math blog’ and ‘engineeringmathgeek.com’, 2016-19. All written materials and photos published on this blog are copyright protected. Excerpts and links may be used, provided that full and clear credit is given to ‘Aspriha Peters’ and “Engineering math blog” with appropriate and specific direction to the original content.

Reader Interactions

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Primary Sidebar

COPYRIGHT

All rights reserved© ‘Engineering math blog’ and ‘engineeringmathgeek.com’, 2016-19. All written materials and photos published on this blog are copyright protected.

Excerpts and links may be used, provided that full and clear credit is given to ‘Aspriha Peters’ and “Engineering math blog” with appropriate and specific direction to the original content.

Please contact engineeringmathgeek (at) gmail dot com for permissions or questions or queries, if any.

Thank you very much for visiting this blog. All relevant suggestions and comments are extremely welcome.

Mathematician & hobby academic. Curious about nature. Aspiring blogger & avid reader. More about me.

Recent posts

  • Integrate d z /√(z^2-a^2) | How to integrate d z /√(z^2-a^2) November 20, 2019
  • What is the value of ∫√(a^2 – x^2) dx | Integrate √(a^2 – x^2) dx November 19, 2019
  • Surface of revolutions in polar curves November 13, 2019
  • Singularities and zeros of the complex numbers | November 11, 2019
  • Volume of revolutions in polar curves – how to get that? November 5, 2019
  • Integrate dx/(acos x + bsin x + c) | How to do that?? November 1, 2019
  • Diagonalize any matrix | How to diagonalize any matrix October 31, 2019
  • Length of a polar curve | How to get the length of a polar curve October 24, 2019
  • dx/(a cos^2x+bsin^2x+c) – how to integrate it? October 23, 2019
  • Volume of solids | How to apply triple integrals October 22, 2019

Advanced engineering mathematics

  • First order ODE
  • Second order ODE
  • Partial differentiation
  • Laplace transform
  • Matrix analysis
  • Vector analysis
  • Difference equations and z-transform
  • Partial differential equations
  • Multiple integration
  • Fourier series

Basic engineering mathematics

  • Partial fractions
  • Differentiation
  • Differentiation applications
  • Integration
  • Integration applications
  • Complex numbers
  • Determinants
  • Vector
  • Series
  • Sequences
  • Multiple integrals
  • Polar curves

Footer

Mathematician & hobby academic. Curious about nature. Aspiring blogger & avid reader. More about me.

  • About
  • Overview
  • Privacy policy

engineeringmathgeek.com © 2019- All rights reserved .

By continuing to browse the site you are agreeing to our use of cookies.