Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.soft-sys.math.mathematica > #5366 > unrolled thread

Find a single solution to a system of equations

Started by"jamievicary@NgOmSaPiAlM.com" <jamievicary@gmail.com>
First post2011-10-04 05:38 +0000
Last post2011-10-07 08:52 +0000
Articles 3 — 2 participants

Back to article view | Back to comp.soft-sys.math.mathematica


Contents

  Find a single solution to a system of equations "jamievicary@NgOmSaPiAlM.com" <jamievicary@gmail.com> - 2011-10-04 05:38 +0000
    Re: Find a single solution to a system of equations Peter Pein <petsie@dordos.net> - 2011-10-05 08:04 +0000
      Re: Find a single solution to a system of equations "jamievicary@NgOmSaPiAlM.com" <jamievicary@gmail.com> - 2011-10-07 08:52 +0000

#5366 — Find a single solution to a system of equations

From"jamievicary@NgOmSaPiAlM.com" <jamievicary@gmail.com>
Date2011-10-04 05:38 +0000
SubjectFind a single solution to a system of equations
Message-ID<j6e648$kd6$1@smc.vnet.net>
Hi,

I have a system of equations, and want to use Mathematica to find a
single solution. It's no good using Solve, because then Mathematica
tries to find all solutions, and that takes too long. It's no good
containing FindInstance, because my equations contain unknowns.

To illustrate this last point, consider the input
   FindInstance[a==b,a]
I want Mathematica to return to me the solution
   {{a->b}}
but it doesn't -- it fails, because FindInstance can't cope with
independent variables in the expression.

So, I either need a watered-down Solve, that doesn't try to find every
solution, or a beefed-up FindInstance, that doesn't mind about the
independent variables.

Thanks!
Jamie.

[toc] | [next] | [standalone]


#5380

FromPeter Pein <petsie@dordos.net>
Date2011-10-05 08:04 +0000
Message-ID<j6h32h$6vd$1@smc.vnet.net>
In reply to#5366
Am 04.10.2011 07:38, schrieb jamievicary@NgOmSaPiAlM.com:
> Hi,
>
> I have a system of equations, and want to use Mathematica to find a
> single solution. It's no good using Solve, because then Mathematica
> tries to find all solutions, and that takes too long. It's no good
> containing FindInstance, because my equations contain unknowns.
>
> To illustrate this last point, consider the input
>     FindInstance[a==b,a]
> I want Mathematica to return to me the solution
>     {{a->b}}
> but it doesn't -- it fails, because FindInstance can't cope with
> independent variables in the expression.
>
> So, I either need a watered-down Solve, that doesn't try to find every
> solution, or a beefed-up FindInstance, that doesn't mind about the
> independent variables.
>
> Thanks!
> Jamie.
>

There is a trick in [1] (by Peter Valko).
Replace b by a constant not related to your problem, apply FindInstance 
and substitute the chosen constant with b:

In[1]:= FindInstance[{a == b} /. b -> Catalan, a] /. Catalan -> b
Out[1]= {{a->b}}

Peter


[1] http://forums.wolfram.com/mathgroup/archive/2006/Sep/msg00207.html

[toc] | [prev] | [next] | [standalone]


#5476

From"jamievicary@NgOmSaPiAlM.com" <jamievicary@gmail.com>
Date2011-10-07 08:52 +0000
Message-ID<j6mek9$7dg$1@smc.vnet.net>
In reply to#5380
> There is a trick in [1] (by Peter Valko).
> Replace b by a constant not related to your problem, apply FindInstance
> and substitute the chosen constant with b:
>
> In[1]:= FindInstance[{a == b} /. b -> Catalan, a] /. Catalan -> b
> Out[1]= {{a->b}}

Beautiful! Thanks Peter.

But shouldn't Mathematica be able to do this natively??

[toc] | [prev] | [standalone]


Back to top | Article view | comp.soft-sys.math.mathematica


csiph-web