Results 1 to 10 of 10
Like Tree10Thanks
  • 1 Post By romsek
  • 2 Post By romsek
  • 1 Post By BobP
  • 2 Post By Soroban
  • 2 Post By johng
  • 2 Post By Prove It

Math Help - integrating query

  1. #1
    Member
    Joined
    Nov 2013
    From
    Australia
    Posts
    187
    Thanks
    38

    integrating query

    \begin{align*}(1+x^2)\frac{dy}{dx}&=x(1-y)\\\int \frac{dy}{1-y}&=\int \frac{xdx}{1+x^2}\\-ln(1-y)&=\frac{1}{2}\ln(1+x^2)+c\end{align*}

    I saw this working recently and I was hoping someone might be able to help me to understand/accept it better.
    It is the integrating that is upsetting me because one side is integrated with respect to x and the other with respect to y.
    why is this allowed?

    Please help me to understand.
    Follow Math Help Forum on Facebook and Google+

  2. #2
    MHF Contributor
    Joined
    Nov 2013
    From
    California
    Posts
    2,357
    Thanks
    910

    Re: integrating query

    \frac{\text{dy} \left(x^2+1\right)}{\text{dx}}=x (1-y)

    \text{dy} \left(x^2+1\right)= x (1-y) \text{dx}

    \frac{\text{dy}}{1-y}=\frac{x \text{dx}}{x^2+1}

    \int \frac{1}{1-y} \, dy=\int \frac{x}{x^2+1} \, dx

    The first integral should be obviously -ln(1-y)

    For the second you need to make the substitution u = 1+x2 and note that du = 2x dx, it should be obvious how to proceed from there.
    Last edited by romsek; November 29th 2013 at 02:35 AM.
    Thanks from Melody2
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Member
    Joined
    Nov 2013
    From
    Australia
    Posts
    187
    Thanks
    38

    Re: integrating query

    thanks Romsek, but I think that you have missed my point.
    I know how it is done but I don't know why I am allowed to do it.
    Follow Math Help Forum on Facebook and Google+

  4. #4
    MHF Contributor
    Joined
    Nov 2013
    From
    California
    Posts
    2,357
    Thanks
    910

    Re: integrating query

    they are just dummy variables. You know this because of the dy and dx. Is there any difference between

    \int \frac{1}{1-y} \, dy and \int \frac{1}{1-x} \, dx ?

    any difference would show up in the limits of integration. With no limits of integration explicitly stated sure you would go ahead and say that the first integral is an antiderivative in y and the second an antiderivative in x but that's not really correct, it's just a shorthand, and even then x and y are just dummy vars for the antiderivative function.

    Maybe one of the hard math guys can give you a better answer. I'm more of an engineering math guy.
    Thanks from Melody2 and topsquark
    Follow Math Help Forum on Facebook and Google+

  5. #5
    Member
    Joined
    Nov 2013
    From
    Australia
    Posts
    187
    Thanks
    38

    Re: integrating query

    Okay thanks Romsek, that at least gives me something to think about I'm still not sure that I am convinced, but the idea is growing on me.
    Follow Math Help Forum on Facebook and Google+

  6. #6
    Super Member
    Joined
    Jun 2009
    Posts
    660
    Thanks
    133

    Re: integrating query

    Unless you are really interested, it's best if you simply accept that if

    M + N \frac{dy}{dx}= 0, then

    Mdx+Ndy=0

    and that you can simply put an integral sign in front of each expression.

    That is,

    \int Mdx + \int N dy = 0, or

    \int Ndy = -\int Mdx.

    If you are really interested then you should read up on things called \textbf{differentials}.

    As to your equation, the left and right hand sides can be considered separately, that is, you could ask what the RHS is equal to without any reference to the LHS and vice versa. They're sort of stand alone problems.
    Thanks from Melody2
    Follow Math Help Forum on Facebook and Google+

  7. #7
    Super Member

    Joined
    May 2006
    From
    Lexington, MA (USA)
    Posts
    11,713
    Thanks
    632

    Re: integrating query

    Hello, Melody2!

    (1+x^2)\frac{dy}{dx}\:=\:x(1-y)

    . . \int \frac{dy}{1-y}\:=\:\int \frac{x\,dx}{1+x^2}

    -\ln(1-y) \:=\:\tfrac{1}{2}\ln(1+x^2)+C

    I saw this working recently and I was hoping someone might be able to help me to understand/accept it.
    It is the integrating that is upsetting me, because one side is integrated with respect to x
    and the other with respect to y. .Why is this allowed?

    Why?
    Suppose we have two integrals: . \begin{Bmatrix} I &=& \displaystyle\int 2y\,dy \\ J &=& \displaystyle \int 3x^2\,dx\end{Bmatrix}

    Would you agree that: . \begin{Bmatrix}I &=& y^2+C_1 \\ J &=& x^3+C_2 \end{Bmatrix}\.?


    Suppose they told us that I \,=\,J.

    Can't we write: . y^2+C_1 \:=\:x^3+C_2 \quad\Rightarrow\quad y^2 \:=\:x^3 + C\;?
    Thanks from topsquark and Melody2
    Follow Math Help Forum on Facebook and Google+

  8. #8
    Super Member
    Joined
    Dec 2012
    From
    Athens, OH, USA
    Posts
    633
    Thanks
    255

    Re: integrating query

    Hi Melody,

    Suppose I were to say: "let f(x) be a function". This is really an abuse of notation. Given a function f with some domain D, f(x) is the value of f at the element x of D. So by "let f(x) be a function" really means: let f be a function defined by f(x) = blah blah blah (the domain D is either explicitly given or "implied").

    Now given a function f, an antiderivative of f is a function F such that the derivative of F, namely F', satisfies F'(x)=f(x) for all x in the domain of f; same thing F'(z) = f(z) for any z. The accepted notation for any antiderivative of f is
    \int f(x)\,dx

    Now to your specific question about differential equations. Let f and g be known functions with y an unknown function. The DE is of the form

    f(y){dy\over dx}=g(x)

    What this really means is

    f(y(x))y\prime(x)=g(x) for all x (in some domain D).

    Now let F and G be antiderivatives of f and g respectively; i.e. F'(y) = f(y) and G'(x) = g(x) for all appropriate y and x. Now what is the derivative of the composition of F with the unknown function y?. Well, by the chain rule, this derivative F(y(x))' = F'(y(x))y'(x) or f(y(x))y'(x). But f, y and g satisfy

    f(y(x))y\prime(x)=g(x)

    That is, the antiderivative F composed with y has derivative the same as the derivative of G. That is,

    \int f(y)\,dy = \int f(x)\,dx

    Here, though, the antiderivative on the left must be interpreted as being composed with the unknown function y.

    Example DE:

    (1+x^2){dy\over dx}=x(1-y) or {1\over 1-y}{dy\over dx}={x\over x^2+1}

    Let f(y)={1\over 1-y} and g(x)={x\over x^2+1}

    You "integrate both sides" to obtain:

    -\ln(1-y)={1\over2}\ln(x^2+1)}+C

    This last equation must be interpreted as

    -\ln(1-y(x))={1\over2}\ln(x^2+1)}+C with y the unknown function.

    Of course, this last equation can be solved for y(x) to get

    y(x)=K(x^2+1)^{-1/2}+1 Here K=e^{-C}
    Thanks from topsquark and Melody2
    Follow Math Help Forum on Facebook and Google+

  9. #9
    MHF Contributor
    Prove It's Avatar
    Joined
    Aug 2008
    Posts
    11,513
    Thanks
    1404

    Re: integrating query

    Quote Originally Posted by Melody2 View Post
    \begin{align*}(1+x^2)\frac{dy}{dx}&=x(1-y)\\\int \frac{dy}{1-y}&=\int \frac{xdx}{1+x^2}\\-ln(1-y)&=\frac{1}{2}\ln(1+x^2)+c\end{align*}

    I saw this working recently and I was hoping someone might be able to help me to understand/accept it better.
    It is the integrating that is upsetting me because one side is integrated with respect to x and the other with respect to y.
    why is this allowed?

    Please help me to understand.
    This is a technique known as "Separation of Variables", which is an entirely valid technique, despite how sloppy the working out given is. The idea is to use the Chain Rule in reverse (which you might know as integration by substitution) to simplify one of the integrands. You might remember that to use a substitution, you need to find an "inner" function (which we usually call \displaystyle \begin{align*} u \end{align*} ) and this inner function's derivative \displaystyle \begin{align*} \frac{du}{dx}  \end{align*} needs to be a factor in the integrand. So something like \displaystyle \begin{align*} \int{f\left( u(x) \right) \cdot \frac{du}{dx} \, dx} \end{align*}. By substituting the function u, then this becomes \displaystyle \begin{align*} \int{ f\left( u(x) \right) \cdot \frac{du}{dx} \, dx } = \int{ f(u)\,du} \end{align*} and then perform the integration with respect to \displaystyle \begin{align*} u \end{align*}.

    This is exactly what happens when separating variables in a DE. Notice that \displaystyle \begin{align*} y \end{align*} is a function of \displaystyle \begin{align*} x \end{align*}, and the equation is written in terms of the derivative of y as well (obviously as it's a differential equation). This means that essentially both sides are a function of x, and because it's an equation, each side is equal to the other, and so the integrals of both sides have to be equal as well. So in your case

    \displaystyle \begin{align*} \left( 1 + x^2 \right) \frac{dy}{dx} &= x \left( 1 - y \right) \\ \left( \frac{1}{1 - y} \right) \frac{dy}{dx} &= \frac{x}{1 + x^2} \\ \int{ \left( \frac{1}{1 - y} \right) \frac{dy}{dx} \, dx } &= \int{ \frac{x}{1 + x^2} \, dx} \textrm{ upon integrating both sides with respect to }x \\ \int{ \frac{1}{1 - y} \, dy} &= \int{ \frac{x}{1+ x^2} \, dx} \textrm{ when the Chain Rule simplifies the LHS the same way as a substitution.}  \end{align*}

    and then it's a case of integrating both sides with respect to the variable given.
    Thanks from Melody2 and romsek
    Follow Math Help Forum on Facebook and Google+

  10. #10
    Member
    Joined
    Nov 2013
    From
    Australia
    Posts
    187
    Thanks
    38

    Re: integrating query

    Thank you to all of you. Those answers are great!
    They have really helped me.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Query
    Posted in the Trigonometry Forum
    Replies: 1
    Last Post: May 8th 2012, 04:57 PM
  2. Arc length query
    Posted in the Calculus Forum
    Replies: 2
    Last Post: October 17th 2009, 12:47 PM
  3. exp&log functions query
    Posted in the Algebra Forum
    Replies: 1
    Last Post: March 30th 2009, 01:33 PM
  4. probability query?
    Posted in the Statistics Forum
    Replies: 4
    Last Post: November 12th 2008, 10:17 PM
  5. query
    Posted in the Algebra Forum
    Replies: 1
    Last Post: January 4th 2008, 02:41 AM

Search Tags


/mathhelpforum @mathhelpforum