|


Integration of ... Given dx/dtDate: 7/5/96 at 5:48:56 From: Anonymous Subject: Integration of ... Given dx/dt I am attempting to integrate da/dt = k.(a)^m.(1-a)^n where (a) is a fraction. Therefore (a) could be written as (x/A). When time = 0 , x = 0 and when time is infinite x = A. So: dx/dt = k.A.(x/A)^m.(1-(x/A))^n What I wish to find is the value of x (or x/A) as a function of time(t). I can solve the related problem: dx/dt = k.(A-x)^m which results in: dx/(A-x)^m = k.dt let u = (A-x) k.t = u^(-m) . dx = u^(-m).(-du) = [-u^(1-m) / (1-m) ] + c k.t = c - [ A-x^(1-m) / (1-m)] When t=0, x=0 and when t = infinite x = A, therefore c = A^(1-m) / (1-m) k.t = [A^(1-m) / (1-m) ] - [ (A-x)^(1-m) / (1-m) So (A-x) = [A^(1-m) - k.t.(1-n)]^(1/1-m) However I am unable to solve the equation when there are two power terms (m and n). I will be very greatful if you can find a solution to this problem. Richard.
Date: 7/5/96 at 8:19:36
From: Doctor Anthony
Subject: Re: Integration of ... Given dx/dt
The equation you started with can be more conveniently expressed as
dx/dt = k*x^m*(1-x)^n, so separating the variables we get
dx/(x^m*(1-x)^n) = k*dt
Now use partial fractions on the left hand side of this equation:
1/(x^m*(1-x)^n) = A/x^m + B/(1-x)^n
1 = A(1-x)^n + B*x^m
Put x=0 you get 1 = A, Put x=1 you get 1 = B. So we have A=1, B=1
Left hand side of integral becomes:
INT[{1/x^m + 1/(1-x)^n}*dx] = {x^(1-m)}/(1-m) - {(1-x)^(1-n)}/(1-n)
and the full integral can be written:
{x^(1-m)}/(1-m) - {(1-x)^(1-n)}/(1-n) = kt + Const.
In general you will not be able to express x explicitly as a function
of t. Even the relatively simple expression like x^3 - (1-x)^5 = t
could not be manipulated to make x the subject of the formula. To
find values of x for given values of t you should use the Newton-
Raphson method, Maple or a TI-92 calculator.
-Doctor Anthony, The Math Forum
Check out our web site! http://mathforum.org/dr.math/
|
Search the Dr. Math Library: |
[Privacy Policy] [Terms of Use]


Ask Dr. MathTM
© 1994-2013 The Math Forum
http://mathforum.org/dr.math/