Flow Modeling using Electrical Circuits

[DonyWane]

I'm looking for interesting ways to approach basic or typical problems we are faced with as Chemical Engineers to "beef" up my understanding of the principles. For systems where we are not concerned with the effects of inertia or fluid compressibility, we can apply principles of electrical circuits to fluid flow modeling. It can really simplify the generation of equations and unknowns for some flow systems. However, I became stumped when it came to discontinuities - such as flowing through a tank. Although it might be a stretch to attempt to model fluid flow systems like tank draining using an electrical circuit - it still seems applicable.

Does anyone know how this would be done - perhaps using an inductor and capacitor in parallel (this is called a tank circuit - coincidence?)?

Also, if anyone has any sources that might delve into this topic - I would appreciate it. It is pretty interesting and can actually serve to simplify some problems fairly well.

Thanks if advance for any help you can offer!

 

[nbucska]

I don't think it is possible. A laminar flow can become turbulent at a certain speed and this phenomenon doesn't
have electrical equivalent.

<nbucska@pcperipherals.com>

 

[jay165]

What you're describing is an &quot;analog study&quot; performed by an analog computer. Digital computers and process simulators have long since replaced analog computers. As recently as 10-15 years ago analog computers (actually &quot;computer&quot; since the only one I knew of that was still around then was at Southwest Research Institute) were used to perform vibration analysis for reciprocating compressors on offshore oil structures. It's possible that such studies today are performed on PCs and no analog computers even exist - I've haven't been involved with one for years.

Try doing a search on Southwest's website or a google search using &quot;analog computers&quot; or &quot;analog study&quot;

Good Luck.

 

[hacksaw]

the problem you have is that your tank ckt is linear while most fluid flow situations are inherently non-linear, besides you don't have to be concerned about electron cavitation, friction factors, reynolds numbers, and mass transport, or vortex breakers, etc,

 

[DonyWane]

For a simple hydraulic analysis, I usually assume that my flow is turbulent (which I check later) and ignore eddy effects inside the tank. I'm not so much looking for complex modeling as I am just a simple Bernouli balance. I really just want to apply circuit equations like KVL and KCL, but to a fluid flow application.

And, I believe, the discharge of a capacitor is not linear, but it may be exponential - which flow from a tank is not. If the system were simply a closed system with a couple of pumps and several branches all tied back together, the application of circuit theory to the generation of simultaneous equations is quite helpful. Especially for those who &quot;get&quot; circuits more easily.

It just makes me wonder how much can be modeled - I would like to be able to capture the discontinuous effect. Perhaps I will just have to use a series of grounds and voltage sources to capture the effect. The only downside to this is that it results in a relatively more difficult problem than originally faced with - unlike the aforementioned simplification of several pumps and several branches.

Interesting concept, though, I think.

 

[hacksaw]

electrical simulation was commonly used prior to wide spread availability of digital simulation. the research literature (off-internet because of the dates) covers a lot of the issues you've raised. that said in the day of CFD (computational fluid dynamics) the analog approach has simply been set aside.