Solve ODE reducible to exact equation

  • Thread starter Thread starter mahler1
  • Start date Start date
  • Tags Tags
    Ode
Click For Summary

Homework Help Overview

The discussion revolves around solving a first-order ordinary differential equation (ODE) that can be reduced to an exact equation. The equation involves trigonometric functions and requires finding an integrating factor of a specific form.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • The original poster attempts to determine if the given ODE is exact and explores the use of an integrating factor dependent on the ratio of variables. Participants question the differentiation steps taken and suggest corrections. There are discussions about changing variables to simplify the problem.

Discussion Status

Participants have provided corrections to the original poster's differentiation and have confirmed the validity of the integrating factor derived. Suggestions for variable substitution have been made to streamline the process, indicating a productive direction in the discussion.

Contextual Notes

There is an emphasis on ensuring the integrating factor is correctly applied and that the equation is transformed into an exact form. The original poster expresses uncertainty about the integration process and the implications of their findings.

mahler1
Messages
217
Reaction score
0
Homework Statement .

Find all the solutions of the equation:
##\dfrac{\sin(y)}{x}dx+(\dfrac{y}{x}\cos(y)-\dfrac{\sin(y)}{y})dy=0## knowing that the equation admits an integrating factor ##u## of the form ##u(x,y)=h(\dfrac{x}{y})##

The attempt at a solution.

If I call ##M(x,y)=\dfrac{\sin(y)}{x}## and ##N(x,y)=\dfrac{y}{x}\cos(y)-\dfrac{\sin(y)}{y}## then, in order to be an exact differential equation, ##M## and ##N## must satisfy ##M_y=N_x##, but ##M_y=\dfrac{\cos(y)}{x}≠-\dfrac{y\cos(y)}{x^2}=N_x##, so this is not an exact differential equation.

As the exercise suggests, I've tried to propose an integrating factor ##u## of the form ##u(x,y)=h(\dfrac{x}{y})##, but, since up to know I've only solved equations where the integrating factor depended just on one of the variables, I got stuck in the middle of the problem.

So, ##u## must satisfy:

##(uM)dx+(uN)dy=0##

Then, ##(uM)_y=(uN)_x \iff u_yM+uM_y=u_xN+uN_x \iff xh_yM+hM_y=\dfrac{1}{y}h_xN+hN_x##.

Calculating all these partial derivatives gives:

##xh_y\dfrac{\sin(y)}{x}+h\dfrac{\cos(y)}{x}=\dfrac{1}{y}h_x(\dfrac{ycos(y)}{x}-\dfrac{\sin(y)}{y})+h(-\dfrac{ycos(y)}{x^2})##

At this point I got stuck and I don't know what and how to integrate in order to find ##u(x,y)##
 
Physics news on Phys.org
You didn't differentiate correctly. You should have, for example,
\begin{align*}
(uM)_y &= \frac{\partial}{\partial y} \left[ h\left(\frac{x}{y}\right) \frac{\sin y}{x}\right] \\
&= h'\left(\frac{x}{y}\right)\left(-\frac{x}{y^2}\right) \frac{\sin y}{x} + h\left(\frac{x}{y}\right) \frac{\cos y}{x} \\
&= -h'\left(\frac{x}{y}\right) \frac{\sin y}{y^2} + h\left(\frac{x}{y}\right) \frac{\cos y}{x}
\end{align*}
 
  • Like
Likes   Reactions: 1 person
vela said:
You didn't differentiate correctly. You should have, for example,
\begin{align*}
(uM)_y &= \frac{\partial}{\partial y} \left[ h\left(\frac{x}{y}\right) \frac{\sin y}{x}\right] \\
&= h'\left(\frac{x}{y}\right)\left(-\frac{x}{y^2}\right) \frac{\sin y}{x} + h\left(\frac{x}{y}\right) \frac{\cos y}{x} \\
&= -h'\left(\frac{x}{y}\right) \frac{\sin y}{y^2} + h\left(\frac{x}{y}\right) \frac{\cos y}{x}
\end{align*}

Thanks, I'll correct it and see if I can get a nicer expression.
 
vela said:
You didn't differentiate correctly. You should have, for example,
\begin{align*}
(uM)_y &= \frac{\partial}{\partial y} \left[ h\left(\frac{x}{y}\right) \frac{\sin y}{x}\right] \\
&= h'\left(\frac{x}{y}\right)\left(-\frac{x}{y^2}\right) \frac{\sin y}{x} + h\left(\frac{x}{y}\right) \frac{\cos y}{x} \\
&= -h'\left(\frac{x}{y}\right) \frac{\sin y}{y^2} + h\left(\frac{x}{y}\right) \frac{\cos y}{x}
\end{align*}

After vela's correction, I got:

##(uM)_y=-h'(\dfrac{x}{y})(\dfrac{\sin(y)}{y^2})+h(\dfrac{x}{y})(\dfrac{\cos(y)}{x})##.
##(uN)_x=h'(\dfrac{x}{y})(\dfrac{\cos(y)}{x}-\dfrac{\sin(y)}{y^2})-h(\dfrac{x}{y})\dfrac{y\cos(y)}{x^2})##.
So, ##(uM)_y=(uN)_x## if and only if

##-h'(\dfrac{x}{y})\dfrac{\sin(y)}{y^2}+h(\dfrac{x}{y})\dfrac{\cos(y)}{x}=h'(\dfrac{x}{y})(\dfrac{\cos(y)}{x}-\dfrac{\sin(y)}{y^2})-h(\dfrac{x}{y})\dfrac{y\cos(y)}{x^2})## if and only if

##h(\dfrac{x}{y})\dfrac{\cos(y)}{x}=h'(\dfrac{x}{y})(\dfrac{\cos(y)}{x})-h(\dfrac{x}{y})\dfrac{y\cos(y)}{x^2})##.

Then

##h(\dfrac{x}{y})(\dfrac{\cos(y)}{x}+\dfrac{y\cos(y)}{x^2})=h'(\dfrac{x}{y})(\dfrac{\cos(y)}{x})##

##1+\dfrac{y}{x}=\dfrac{h'\dfrac{x}{y}}{h\dfrac{x}{y}}##. If I make the substitution ##z=\dfrac{x}{y}##, then the equation reduces to:

##1+\dfrac{1}{z}=\dfrac{h'(z)}{h(z)}##.

So

##\int (1+\dfrac{1}{z})dz=\int \dfrac{dh}{h} \iff z+ln(z)+k=ln(h)##.

Then, ##h=ze^z=\dfrac{x}{y}e^{\dfrac{x}{y}}##. Now I have to go back to the original equation, which it will be reduced to an exact equation by this integrating factor.

Is this correct?
 
Yes, it is!
 
  • Like
Likes   Reactions: 1 person
vela said:
Yes, it is!

Thanks very much!
 
Good work.
Let me offer a suggestion.
I would change variable z=x/y first to make things less messy
$$\frac{\sin(y)}{x}\mathrm{d}x+\left( \frac{y}{x}\cos(y)-\frac{\sin(y)}{y}\right)\mathrm{d}y=0\\
\text{becomes}\\
\sin(y)\frac{\mathrm{d}z}{z}+\frac{1}{z}\cos(y)\mathrm{d}y=0$$
 
  • Like
Likes   Reactions: 1 person
lurflurf said:
Good work.
Let me offer a suggestion.
I would change variable z=x/y first to make things less messy
$$\frac{\sin(y)}{x}\mathrm{d}x+\left( \frac{y}{x}\cos(y)-\frac{\sin(y)}{y}\right)\mathrm{d}y=0\\
\text{becomes}\\
\sin(y)\frac{\mathrm{d}z}{z}+\frac{1}{z}\cos(y)\mathrm{d}y=0$$

Good remark.
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
1K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 16 ·
Replies
16
Views
3K
  • · Replies 9 ·
Replies
9
Views
3K
Replies
22
Views
2K
  • · Replies 11 ·
Replies
11
Views
3K
Replies
3
Views
2K