New Reply

Basic question about separable diffeq methodology

 
Share Thread Thread Tools
Aug1-12, 05:03 PM   #1
 

Basic question about separable diffeq methodology


It's my understanding that the definition of the indefinite integral is:

∫f(x)dx = F(x) + C, where d/dx [F(x) + C] = f(x) and C is an arbitrary constant

And while dx has meaning apart from the indefinite integral sign the indefinite integral sign has no meaning apart from dx. Adding an indefinite integral sign to both sides of an equation would be similar to adding an open bracket to both sides of an equation; which of course is meaningless.

So my question is when using the common method to solve separable diffeq why do we infact do this, e.g.

dy/dx = xy
dy/y = x dx
∫dy/y = ∫x dx
ln [y] = x^2/2 + C
y = exp[x^2/2 + C]

My naive idea was that since all elementary antiderivatives can be written using other symbology besides the indefinite integral sign that adding an indefinite integral sign could be always be expressed as something else; I know this might seem silly but the method is nonsensical to begin with. However I think this is actually impossible to do because of very strange problems with the arbitrary constant.
PhysOrg.com
PhysOrg
science news on PhysOrg.com

>> King Richard III found in 'untidy lozenge-shaped grave'
>> Google Drive sports new view and scan enhancements
>> Researcher admits mistakes in stem cell study
Aug1-12, 06:13 PM   #2
 
One should understand that the differential and integral notation became accepted and widespread much earlier than calculus got proper rigor. So the notation may sometimes be inexact and confusing.

However, in the example given, observe that the "integral sign" is always applied to terms that already contain either dx or dy; so the expressions are always well formed. Informally, one could think that the "integral sign" cancels the "differential sign", and so it may be applied only to differentials, i.e., expressions of the kind f(x)dx.

Formally, only derivatives and definite integrals are used, d(something) is used only as part of the symbol for these two operations, never by itself.

[itex]\frac {dy}{dx} = X(x)Y(y)[/itex]

really means this:

[itex]\frac {d}{dx}y(x) = X(x)Y(y(x))[/itex]

[itex]\frac {\frac {d}{dx}y(x)}{Y(y(x))} = X(x)[/itex]

So we have functions of x on both sides of the equation, thus we can integrate them in some interval [t_{0}, t]:

[itex]\int_{t_{0}}^t\frac {\frac {d}{dx}y(x)}{Y(y(x))}dx = \int_{t_{0}}^tX(x)dx[/itex]

Now substitute

[itex]u = y(x)[/itex]

Then

[itex]\int_{t_{0}}^t\frac {\frac {d}{dx}y(x)}{Y(y(x))}dx = \int_{y(t_{0})}^{y(t)}\frac {du}{Y(u)}[/itex]

Hence

[itex]\int_{y(t_{0})}^{y(t)}\frac {du}{Y(u)} = \int_{t_{0}}^tX(x)dx[/itex]
New Reply

Tags
calculus, diff, diffeq, function, separable
Thread Tools


Similar Threads for: Basic question about separable diffeq methodology
Thread Forum Replies
Why not separable (basic question)? Differential Equations 4
DiffEq question Calculus & Beyond Homework 1
Financial DiffEq question Calculus & Beyond Homework 4
DiffEQ question Calculus 8
simple integral -- but a question of methodology Calculus & Beyond Homework 9