Re: Engineering requests

• To: mathgroup at smc.vnet.net
• Subject: [mg125265] Re: Engineering requests
• From: danl at wolfram.com
• Date: Sat, 3 Mar 2012 06:52:54 -0500 (EST)
• Delivered-to: l-mathgroup@mail-archive0.wolfram.com
• References: <jiqflj\$3o\$1@smc.vnet.net>

```On Friday, March 2, 2012 6:49:55 AM UTC-6, McHale, Paul wrote:
> So, here are some problems we face, but don't have great answers for in Mathematica.
>
> 1. Dimensional analysis.  To do this, we must have unit support.  The best description of this is the ability to calculate (V/R)^2 R and have it return a unit of watts.  Other programs handily support this (though they are sorely lacking in other places :))

Probably not exactly what you want, but you might have a look here.

http://library.wolfram.com/infocenter/Conferences/7513/

In[40]:= minimalUnits[(v/o)^2*o]
Out[40]= {Watts}

> 2. Tolerances support.  One difficulty we have is determining the min/max at a certain point in a circuit even if the circuit is not complicated to model.  I use lists {Rmax,Rmin} and Table.  Works, but is a little clumsy.
> [...]

Have you tried using Interval[...] to specify value ranges? For basic arithmetic this should offer some possibilities. If you are computing, say, solutions to differential equations with toleranced input values, that will be more of a challenge. In effect what one wants is some sensitivity analysis.  Can be done (using NDSolve), but is not so simple. That is to say, I do not recall the details of the setup.

Daniel Lichtblau
Wolfram Research

```

• Prev by Date: Re: Can you call 32 Bit DLLs from Mathematica 8.0 ?
• Next by Date: irritating screen jumps in Mathematica 8.0.4
• Previous by thread: Engineering requests
• Next by thread: Re: Engineering requests