Quantcast How can I determine if the equation (2x+3) + (2y-2)y' = 0 is exact or not? Text - Physics Forums Library

PDA

View Full Version : How can I determine if the equation (2x+3) + (2y-2)y' = 0 is exact or not?


fehwqjkfdehqa
Jul20-08, 09:16 PM
How can I determine if the equation (2x+3) + (2y-2)y' = 0 is exact or not?

Now I know I need to take partial derivatives of certain terms of the equation, and call that M and N right?

1. How do I separate the terms?
2. How do I know which variable gets differentiated?

For example , if I separate it so that its:
partial x: (2x+3) = 2
partial y: (2y-2) = 2
2=2, so its exact.

BUT why can't I go:
partial x (2y-2) = 0
partial y (2x+3) = 0
0=0 so thats also exact.

So how do you determine which term is associated with what you are differentiating with repect to???

Also, why do textbooks use M and N??? Whats the point of using these when we can just say "partial differentiate with repect to x" and "partial differentiate with repect to y"

Also, does M ussually go with x and the N ussually go with y and why?

Please explain this to me my brain is about to explode.
Thanks

jeffreydk
Jul20-08, 11:24 PM
It is common practice to write it as

M(x,y)dx+N(x,y)dy=0

and then to differentiate N with respect to x and M with respect to y to check if the equation is exact. The whole method depends on the fact that there is some function where,

\frac{\partial \Phi}{\partial x}=M(x,y),\text{ }\frac{\partial \Phi}{\partial y}=N(x,y)

You can't switch the respectful variables and have it still work so I think you found a special case. For example,

(y^2-2x)dx+(2xy+1)dy=0

M_y=2y, N_x=2y

so it is exact, but the other way around you get,

M_x=-2, N_y=2x

Hope that helps.