Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re: Analytical or not?

[Subject Prev][Subject Next][Thread Prev][Thread Next]

of course I want to hear all opinions from all the English-speaking
world, thanks!

I see that the fact is, 'analytical' safely retains two meanings in
English. The same cannot be said about other local environments

On 10/5/11, gregory szuladzinski <ggg(--nospam--at)> wrote:
> Blank
> Alexander,
> It all sounds like a storm in a teacup.
> I think the underlying issue is whether someone is using closed formulas or
> numerical models.
> The term "analytical" has often be used for both and the usage will so
> remain.
> If you are discussing matters within more specialized, computational domain,
> you are likely to make a distinction between analytical (closed form) and
> numerical (formless).
> The first of those uses a calculator or, these days, Mathcad and the likes
> of it.
> The second is mainly FEA, sometimes discrete elements, occasionaly boundary
> elements.
> A spreadsheet, on the other hand, can support both.
> You did not say you wanted to hear an opinion from Australia,
> but I thought I might volunteer my two cents worth.
> Sincerely
> Gregory form Oz
> From: Alexander Bausk <bauskas(--nospam--at)>
> Subject: Re: "Analytical model"
> To: seaint(--nospam--at)
> --bcaec51dd7d7df86c004ae7cd8a9
> Content-Type: text/plain; charset=windows-1252
> Content-Transfer-Encoding: quoted-printable
> Hello Bill,
> "right" is "richtig"; "rechen" is derived from rechnen, "(to) count".
> I'm not sure about the German terms. It is essentially a bait to lure Germa=
> n
> listers to the thread (there are some AFAIK).
> What I'm sure about is, we say something like "structural model", or
> "calculation model" to describe any generic formalized model of a structure=
> .
> Then, all sort of generic calculation models are, mathematically speaking,
> split into two domains: analytical and computational. You may clearly see
> this is a common scheme throughout many engineering and research discipline=
> s
> by reviewing this search:
> Analytical model has a closed form solution, that is, it can be solved in
> one iteration using the simplest operators. If your model is Y=3Df(X), you =
> are
> usually able to retrieve the inverse solution, X=3Df'(Y), and express it in
> more or less simple terms.
> In practice, any code check is deemed analytical and the common term used t=
> o
> humiliate junior engineers, the "hand analysis", usually redirects here (if
> you are not solving FE matrices by hand which is quite possible).
> Computational model is usually highly iterative and includes complex
> procedures like FEA, boundary element analysis or CFD. It is still called
> analysis  of course (in your language). You cannot directly express X
> depending on Y in this case.
> In other terms, your 15-minute-to-design ACI Excel spreadsheet calc is most
> likely analytical while any FE model is computational. These two models may
> actually describe the same structure like a pinned beam.
> Any FEA software verification manual will host dozens of comparisons of
> computational vs. analytical models describing the same structures.
> So yes, I think this is actually the most fundamental difference there is.
> And the current term usage disturbs me, especially when I see it propagatin=
> g
> to masses on Autodesk venues and help materials (and they are darn good at
> help materials).
> P.s. I'm sorry if I was getting too much narrative, that is just to be
> completely clear on the subject, I am clearly not qualified to lecture
> anybody on the list.

Alexander Bausk
Civil/Structural design & inspection engineer, CAD professional
MSc Structural engineering, Ph.C. Engineering
Dnipropetrovsk, Ukraine
Tel. +38 068 4079692
Fax. +38 0562 470263

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: 
******* ****** ****** ****** ******* ****** ****** ********