MHB Adam's question via email about Laplace Transforms

Click For Summary
The discussion centers on solving the initial value problem (IVP) using Laplace Transforms for the equation dy/dt + 11y = 3t with the initial condition y(0) = 5. The solution process involves taking the Laplace Transform, applying partial fractions, and finding the constants A, B, and C. There are corrections made regarding arithmetic errors in the initial calculations, leading to the correct general solution y(t) = (602/121)e^(-11t) + (3/11)t - (3/121). The conversation highlights the challenges of using the Laplace Transform method and the importance of careful arithmetic in solving differential equations.
Prove It
Gold Member
MHB
Messages
1,434
Reaction score
20
Solve the following IVP using Laplace Transforms:
$\displaystyle \frac{\mathrm{d}y}{\mathrm{d}t} + 11\,y = 3\,t, \quad y\left( 0 \right) = 5$

Take the Laplace Transform of the equation:

$\displaystyle \begin{align*} s\,Y\left( s \right) - y\left( 0 \right) + 11\,Y\left( s \right) &= \frac{3}{s^2} \\
s\,Y\left( s \right) - 5 + 11\,Y\left( s \right) &= \frac{3}{s^2} \\
\left( s + 11 \right) Y\left( s \right) &= \frac{3}{s^2} + 5 \\
Y\left( s \right) &= \frac{3}{s^2 \,\left( s + 11 \right) } + \frac{5}{s + 11} \end{align*}$

Apply Partial Fractions:

$\displaystyle \begin{align*}
\frac{A}{s} + \frac{B}{s^2} + \frac{C}{s + 11} &\equiv \frac{3}{s^2 \,\left( s + 11 \right) } \\
A\,s\left( s + 11 \right) + B \left( s + 11 \right) + C\,s^2 &\equiv 3
\end{align*}$

Let $\displaystyle s = 0 \implies 11\,B = 3 \implies B = \frac{3}{11}$

Let $\displaystyle s = -11 \implies 121\,C = 3 \implies C = \frac{3}{121}$

Then $\displaystyle A\,s\left( s + 11 \right) + \frac{3}{11} \left( s + 11 \right) + \frac{3}{121}\,s^2 \equiv 3$

Let $\displaystyle s = 1$

$\displaystyle \begin{align*}
12\,A + \frac{36}{11} + \frac{3}{121} &= 3 \\
12\,A + \frac{396}{121} + \frac{3}{121} &= \frac{33}{121} \\
12\,A &= -\frac{366}{121} \\
A &= -\frac{61}{242}
\end{align*}$

So

$\displaystyle \begin{align*}
Y\left( s \right) &= -\frac{61}{242}\left( \frac{1}{s} \right) + \frac{3}{11} \left( \frac{1}{s^2} \right) + \frac{3}{121} \left( \frac{1}{s + 11} \right) + 5 \left( \frac{1}{s + 11 } \right) \\
Y\left( s \right) &= -\frac{61}{242} \left( \frac{1}{s} \right) + \frac{3}{11} \left( \frac{1}{s^2} \right) + \frac{608}{121} \left( \frac{1}{s + 11} \right) \\
\\
y\left( t \right) &= -\frac{61}{242} + \frac{3}{11}\,t + \frac{608}{121}\,\mathrm{e}^{-11\,t}
\end{align*}$
 
Mathematics news on Phys.org
Let me, yet again, state my dislike for the "Laplace Transform Method"! And, in fact, Prove It made a slight arithmetic error
(easy to do with something as complicated as "Laplace Transform")
that resulted in an incorrect answer:
if y(t)= 61/242+ (3/11)t+ (608/121)e^{-11t} then
y'(t)= 3/11- (608/11)e^{-11t}
11y= 61/22+ 3t+ (608/11)e^{-11t}

so y'(t)+ 11t= 6/22+ 61/22+ 3t= 67/22+ 3t, not "3t".

It is far easier just to recognize that this is a linear differential equation with constant coefficients. Its "characteristic equation" is r+ 11= 0 so r= -11. The general solution to the associated homogeneous equation is y(t)= Ce{-11t} where C can be any constant.

Since the "non-homogeous" part is a linear polynomial, 3t, we look for a solution to the entire equation of the form y(t)= At+ B, for constants A and B. Then y'= A so the equation becomes A+ 11(At+ B)= 11At+ A+ 11B= 3t. In order for that to be true for all t, we must have both 11A= 3 and A+ 11B= 0. So A= 3/11 and then 3/11+ 11B= 0. B= -3/121.

The general solution to the entire equation is y(t)= Ce^{-11t}+ (3/11)t- 3/121. Since we want y(0)= 5, we must have 5= C+ 3/121 so C= 5- 3/121= 605/121- 3/121= 602/121.<br /> <br /> The solution is <div style="text-align: left"><span style="font-family: 'Verdana'">y(t)= (602/121)e^{-11t}+ (3/11)t- 3/121.<br /> <br /> Check: y&amp;#039;(t)= -(602/11)e^{-11t}+ 3/11 while 11y(t)= (602/11)e^{-11t}+ 3t- 3/11<br /> <br /> y&amp;#039;(t)+ 11t= 3t.<br /> </span>&#8203;</div>
 
Last edited by a moderator:
HallsofIvy said:
Let me, yet again, state my dislike for the "Laplace Transform Method"! And, in fact, Prove It made a slight arithmetic error
(easy to do with something as complicated as "Laplace Transform")
that resulted in an incorrect answer:
if y(t)= 61/242+ (3/11)t+ (608/121)e^{-11t} then
y&#039;(t)= 3/11- (608/11)e^{-11t}
11y= 61/22+ 3t+ (608/11)e^{-11t}

so y&#039;(t)+ 11t= 6/22+ 61/22+ 3t= 67/22+ 3t, not "3t".

It is far easier just to recognize that this is a linear differential equation with constant coefficients. Its "characteristic equation" is r+ 11= 0 so r= -11. The general solution to the associated homogeneous equation is y(t)= Ce{-11t} where C can be any constant.

Since the "non-homogeous" part is a linear polynomial, 3t, we look for a solution to the entire equation of the form y(t)= At+ B, for constants A and B. Then y'= A so the equation becomes A+ 11(At+ B)= 11At+ A+ 11B= 3t. In order for that to be true for all t, we must have both 11A= 3 and A+ 11B= 0. So A= 3/11 and then 3/11+ 11B= 0. B= -3/121.

The general solution to the entire equation is y(t)= Ce^{-11t}+ (3/11)t- 3/121. Since we want y(0)= 5, we must have 5= C+ 3/121 so C= 5- 3/121= 605/121- 3/121= 602/121.<br /> <br /> The solution is <div style="text-align: left"><span style="font-family: 'Verdana'">y(t)= (602/121)e^{-11t}+ (3/11)t- 3/121.<br /> <br /> Check: y&amp;#039;(t)= -(602/11)e^{-11t}+ 3/11 while 11y(t)= (602/11)e^{-11t}+ 3t- 3/11<br /> <br /> y&amp;#039;(t)+ 11t= 3t.<br /> </span>&#8203;</div>
<span style="font-family: 'Verdana'"><br /> <br /> Thanks for pointing out my error Hallsofivy. Adam is one of my students, and the topic they are learning is Laplace Transforms, so he will have to use that method.</span>
 
HallsofIvy said:
The general solution to the entire equation is y(t)= Ce^{-11t}+ (3/11)t- 3/121. Since we want y(0)= 5, we must have 5= C+ 3/121 so C= 5- 3/121= 605/121- 3/121= 602/121.
<br /> <br /> You also have an arithmetic error. When $\displaystyle t = 5$ you end up with $\displaystyle 5 = C - \frac{3}{121} \implies C = 5 + \frac{3}{121} = \frac{608}{121}$.<br /> <br /> I also see where my mistake was in the initial Laplace Transform. The easiest way to evaluate A is to look at the coefficient of $\displaystyle s^2 $, which gives<br /> <br /> $\displaystyle A + \frac{3}{121} = 0 \implies A = -\frac{3}{121} $.<br /> <br /> Thus<br /> <br /> $\displaystyle \begin{align*}<br /> Y\left( s \right) &amp;= -\frac{3}{121}\left( \frac{1}{s} \right) + \frac{3}{11}\left( \frac{1}{s^2} \right) + \frac{3}{121} \left( \frac{1}{s + 11} \right) + 5 \left( \frac{1}{s + 11 } \right) \\<br /> &amp;= -\frac{3}{121} \left( \frac{1}{s} \right) + \frac{3}{11} \left( \frac{1}{s^2} \right) + \frac{608}{121} \left( \frac{1}{s + 11 } \right) \\<br /> \\<br /> y\left( t \right) &amp;= -\frac{3}{121} + \frac{3}{11}\,t + \frac{608}{121}\,\mathrm{e}^{-11\,t}<br /> \end{align*}$<br /> <br /> and this is definitely correct.
 
Argh! Arithmetic! I never was any good at that!
 
I have been insisting to my statistics students that for probabilities, the rule is the number of significant figures is the number of digits past the leading zeros or leading nines. For example to give 4 significant figures for a probability: 0.000001234 and 0.99999991234 are the correct number of decimal places. That way the complementary probability can also be given to the same significant figures ( 0.999998766 and 0.00000008766 respectively). More generally if you have a value that...

Similar threads

  • · Replies 1 ·
Replies
1
Views
10K
  • · Replies 1 ·
Replies
1
Views
10K
  • · Replies 1 ·
Replies
1
Views
10K
  • · Replies 2 ·
Replies
2
Views
10K
  • · Replies 1 ·
Replies
1
Views
10K
  • · Replies 1 ·
Replies
1
Views
10K
  • · Replies 1 ·
Replies
1
Views
10K
  • · Replies 1 ·
Replies
1
Views
7K
Replies
2
Views
7K
  • · Replies 2 ·
Replies
2
Views
6K