Results 1 to 8 of 8

Thread: About fundamental theorem of calculus

  1. #1

    Thread Starter
    Member
    Join Date
    Jul 2001
    Posts
    46

    About fundamental theorem of calculus

    You know;

    if F(x) = INT f(x) dx

    b
    INT f(x) dx equals to F(b) - F(a)
    a

    But if INT f(x) dx can not be written in terms of elementary functions, and if I know the value of (a) and also the result of the integration, is there a way to calculate the value of (b)?

    Thanks in advance.
    Last edited by might; Mar 10th, 2003 at 02:20 PM.

  2. #2
    Addicted Member
    Join Date
    Aug 2002
    Location
    London UK
    Posts
    255
    You'd have to do it numerically using something like Simpson's Rule.
    Not at all related to sheep...

  3. #3

    Thread Starter
    Member
    Join Date
    Jul 2001
    Posts
    46
    Thanks for your reply.

    I know about Simpson`s 1/3 rule. Would you please share your idea to calculate this using it.
    Last edited by might; Mar 11th, 2003 at 04:04 PM.

  4. #4
    vbuggy krtxmrtz's Avatar
    Join Date
    May 2002
    Location
    In a probability cloud
    Posts
    5,573
    Call A = Integral from a to b of Int[f(x)]



    A is the area bounded by the horizontal axis, the horizontal red lines and the vertical green lines.
    The numbers i, i+1, i+2, ..., i+k-2, i+k-1, i+k are integers and I define

    gn = f-1(n)

    as the value of the inverse of f at those points.

    Then the integral (area) can be written as:

    A = i(gi+1 - a) + (i + 1)(gi+2 - gi+1) + (i + 2)(gi+3 -

    gi+2) + ... + (i + k - 2)(gi+k-1 - gi+k-2) + (i + k - 1)(b -

    gi+k-1) = i(b - a) + (gi+2 - gi+1) + 2(gi+3 -

    gi+2) + 3(gi+4 - gi+3) + ... + (k - 2)(gi+k-1 -

    gi+k-2) + (k - 1)(b - gi+k-1) = i(b - a) - gi+1 - gi+2 -

    gi+3 - ... - gi+k-2 - gi+k-1 + (k - 1)b

    Now notice that

    i = INT[f(a)]

    and

    i + k - 1 = INT[f(b)]

    so that:

    A = (b - a)INT[f(a)] + b{INT[f(b)] - INT[f(a)]} - SUM(from j=1 to INT[f(b)] - INT[f(a) of (INT(f(a) +

    j) = b INT[f(b)] - a INT[f(a)] - SUM(from j=1 to INT[f(b)] - INT[f(a) of (INT(f(a) + j)

    (sorry, I couldn't think of any nice-looking way to write the sum over the index j)
    so if you know A and a then you know b right away.
    Attached Images Attached Images  

  5. #5
    vbuggy krtxmrtz's Avatar
    Join Date
    May 2002
    Location
    In a probability cloud
    Posts
    5,573
    This is not so ugly-looking.

    Attached Images Attached Images  

  6. #6
    vbuggy krtxmrtz's Avatar
    Join Date
    May 2002
    Location
    In a probability cloud
    Posts
    5,573
    By the way, I've just noticed I was completely misled when I first read your post. So you just wanted the integral from a to b of f(x), not of INT(f(x)) !!! I'm glad I got it wrong for I had tons of fun working on it!

    Well, as for your question, I think you shouyld go to places like the "Numerical Recipes" web site. There's a chapter devoted to numerical integration with lots of Simpson formulas and its variations and a lot more stuff. You can download the full book in pdf format. They have some theory and then the "recipe" in a number of computer languages. Unfortunately not in vb, but it shouldn't be difficult to translate from Fortran or c. For example, the address below corresponds to the c edition:

    http://www.library.cornell.edu/nr/bookcpdf.html
    (look at chapter 4)

    If you want the main pages, go to Numerical Recipes

  7. #7

    Thread Starter
    Member
    Join Date
    Jul 2001
    Posts
    46
    Thank you very much for your detailed post. But I only mean integral with "INT".

    I only wanted to canculate (b).

    if f(x) sqrt( (-a * sin(x))^2 + (b * cos(x))^2 )
    b
    I = integral f(x) dx
    a

    The result of the integration (I) and the value of (a) are known
    and I`ve been trying to find a solution to get the value of (b).

    The web site you linked is a very useful one as a solution and algorithm source. A type of a website I have been searching for.

    Thanks, again.
    Last edited by might; Mar 12th, 2003 at 12:00 PM.

  8. #8
    Addicted Member
    Join Date
    Aug 2002
    Location
    London UK
    Posts
    255
    To be honest I can't be arsed with calculating anything. Oh dear.
    Not at all related to sheep...

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  



Click Here to Expand Forum to Full Width