Homogeneous Differential Equation

Click For Summary

Homework Help Overview

The discussion revolves around solving a homogeneous differential equation of the form y' = y / [ x + √(y^2 - xy)]. Participants are exploring various methods to approach the integration involved in the solution process.

Discussion Character

  • Exploratory, Mathematical reasoning, Problem interpretation

Approaches and Questions Raised

  • The original poster attempts to solve the equation by substituting u = y/x and simplifying the resulting integral. Some participants discuss the integral's complexity and share insights from computational tools like Wolfram Alpha. Others suggest rewriting the integral in a form that may simplify the integration process, while also questioning the correctness of certain transformations.

Discussion Status

Participants are actively engaging with the problem, sharing different perspectives on the integral and its evaluation. There is a mix of attempts to clarify the integration process and some confusion regarding the application of differentiation rules. No consensus has been reached, but there is a productive exchange of ideas and methods.

Contextual Notes

Some participants express uncertainty about the solvability of the integral and the steps involved in transforming it. There are indications of differing interpretations of the integration process and its underlying principles.

BesselEquation
Messages
3
Reaction score
0

Homework Statement


Solve the following differential equation:
y' = y / [ x + √(y^2 - xy)]

2. The attempt at a solution
Using the standard method for solving homogeneous equations, setting u = y/x, I arrive at the following:

± dx/x = [1±√(u^2-u) ]/ [u√(u^2-u)] which in turn, I get the following integral after simplifying:

∫du/[u√(u^2-u)], seems quite unsolvable to me...
 
Physics news on Phys.org
Note that$$\frac{1}{u\sqrt{u^2-u}}=\frac{1}{u^{3/2}(u-1)^{1/2}}$$Here's where the guessing and experience with derivatives comes in. The idea is to write this as a perfect differential which would make the integration trivial. You know that the term ##u^{3/2}## in the denominator arises by taking the derivative of ##1/u^{1/2}## and the term ##(u-1)^{1/2}## in the denominator arises by taking the derivative of ##(u-1)^{1/2}##. You also know the product rule of differentiation. What do you get when you take the derivative of ##\frac{(u-1)^{1/2}}{u^{1/2}}~##?
 
Delta2 said:
Well , according to wolfram the integral $$\int \frac{du}{u\sqrt{u^2-u}}=2\sqrt{\frac{u-1}{u}}$$
$$\int \frac{du}{u\sqrt{u^2-u}}=\int \frac{du}{u^2\sqrt{1-1/u}}=\int \frac{d(1-1/u)}{\sqrt{1-1/u}}=\int 2d\sqrt{1-1/u}$$
 
  • Like
Likes   Reactions: vela, MathematicalPhysicist and Delta2
Woah alright, great...:wideeyed:
 
lurflurf said:
$$\int \frac{du}{u\sqrt{u^2-u}}=\int \frac{du}{u^2\sqrt{1-1/u}}=\int \frac{d(1-1/u)}{\sqrt{1-1/u}}=\int 2d\sqrt{1-1/u}$$

This is not quite right; it should be
$$\int \frac{d\:\text{anything}}{\sqrt{\text{anything}} }= 2 \sqrt{ \text{anything}}$$
 
Ray Vickson said:
This is not quite right; it should be
$$\int \frac{d\:\text{anything}}{\sqrt{\text{anything}} }= 2 \sqrt{ \text{anything}}$$
I don't understand where you disagree, isn't it ##\frac{d f}{\sqrt{f}}=2d(\sqrt{f)}## and then the integral operator cancels out with the differential operator d and leaves just ##2\sqrt{f}##
 
Delta2 said:
I don't understand where you disagree, isn't it ##\frac{d f}{\sqrt{f}}=2d(\sqrt{f)}## and then the integral operator cancels out with the differential operator d and leaves just ##2\sqrt{f}##
OK, you are right. Basically, I did not notice that you just left out the last step.
 

Similar threads

Replies
5
Views
2K
  • · Replies 12 ·
Replies
12
Views
3K
Replies
10
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 12 ·
Replies
12
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K