Re: Problems integrating InterpolatingFunction

*To*: mathgroup at smc.vnet.net*Subject*: [mg116081] Re: Problems integrating InterpolatingFunction*From*: Sergio Miguel Terrazas Porras <sterraza at uacj.mx>*Date*: Tue, 1 Feb 2011 06:53:38 -0500 (EST)

Hi guys, thanks for all the help. The solution finally was to solve (With NDSolve) a sistem of differential equations, which some of you suggested. Thanks a lot, Sergio ________________________________________ De: DrMajorBob [btreat1 at austin.rr.com] Enviado el: lunes, 31 de enero de 2011 11:58 a.m. Para: mathgroup at smc.vnet.net; Sergio Miguel Terrazas Porras Asunto: Re: [mg116063] Re: Problems integrating InterpolatingFunction Clear[y, t] s = y /. NDSolve[{y'[x] == y[x] Cos[x + y[x]], y[0] == 1}, y, {x, 0, 30}][[1]]; t = t /. First@NDSolve[{t'[x] == 1/s[x]^2, t[0] == 0}, t, {x, 0, 30}]; Plot[s@x, {x, 0, 30}] Plot[t@x, {x, 0, 30}] Bobby On Mon, 31 Jan 2011 02:24:53 -0600, Sergio Miguel Terrazas Porras <sterraza at uacj.mx> wrote: > Hi again, the thing is that I need a function as a result of the > integration, no a number. > I can make a very long table of the results of NIntegrate, then try to > fit the elements of Table, etc. > I just needed to know if it was possible to obtain a function directly > from the indefinite integral. > > In the documentation there is an example of an indefinte integral of an > InterpolatinFuntion. > > Thanks to all who responded. > > Sergio > > ________________________________________ > De: Bob Hanlon [hanlonr at cox.net] > Enviado el: domingo, 30 de enero de 2011 01:32 a.m. > Para: mathgroup at smc.vnet.net > Asunto: [mg116049] Re: Problems integrating InterpolatingFunction > > You must use NIntegrate > > s = y /. NDSolve[{y'[x] == y[x] Cos[x + y[x]], y[0] == 1}, > y, {x, 0, 30}][[1]]; > > NIntegrate[1/s[x]^2, {x, 0, 30}] > > 10325.5 > > > Bob Hanlon > > ---- Sergio Miguel Terrazas Porras <sterraza at uacj.mx> wrote: > > ==================================================== > Hello group, > > I use NDSolve for a nonlinear differential equation, and I get an > InterpolatingFunction, as expected. > > I can plot it, evaluate it, etc. > > The problem I have is that now I need to integrate 1/(the square of the > InterpolatingFunction), and I get nothing but the input back. > > Any ideas? > > Thanks in advance. > > Sergio Terrazas > -- DrMajorBob at yahoo.com