7 posts
TimePosted 28/08/2007 07:26:31
jatin says

Mathameticle model

Does anybody has mathamaticle model for double string preheater.

Just wanted to know the expected parameters at various stages.




Know the answer to this question? Join the community and register for a free guest account to post a reply.

Michael Clark
312 posts
TimePosted 28/08/2007 17:16:15

Re: Mathameticle model

Such models are in existence Jatinder, but customising them to a particular kiln is a major exercise. It is unlikely that such information will be available without charge over the internet. Dr Michael Clark


STEAG Powitec - Albrecht Keiser
18 posts
TimePosted 09/11/2009 19:02:19

Re: Mathameticle model

Powitec sells mathematicle models for double string preheater. But only as a closed loop optimizer including the kiln. www.powitec.de or contact me alexander.hanf@powitec.de


138 posts
TimePosted 10/11/2009 20:56:39
lalbatros says

Re: Mathameticle model

Hello jatinder,

About ten years ago, I have developped a "process modelling framework" during my hobby time. It was developped in java (J++) and the models can be assembled and run from Excel. A full dry kiln model with 5 stages and the rotary kiln divided in 200 sections runs in about 10 seconds and shows temperature data, gas and solid compositons, wall losses, heat exchange, heat balance, ... . The same system can also simulate dryers and mills, also wet kilns.

The bad news is that I don't know if I would be allowed to drop that on the web and benefit from the feedback, since I am working in a cement company. I also lack motivation today, considering that this is not new anymore and that a lot of work could still be done on the program without much fun and without much additional benefit.

That's the reason one of my first posts here was about the benefit of simulations, specially for kiln. For simpler systems, it is a straightforward tool.

This is why I would be interrested to know your ideas about kiln simulations.
Which benefit would you expect?
For the double string question, more precisely, how could a mathematical model be useful?

Personally, I have seen the development of the models as pedagogical, first for myself.
The most strinking feature was the importance of the exothermic reactions in the clinkering zome. I called that "the matchstick effect".
I could also check the influence of flame length, combustion gas composition, use of alternative fuels, ...
All that remained rather qualitative instead of quantitative.
The reason is clear: many aspects would need too much modelling details, like for example the combustion of whole tyre in the kiln including the details of the pyrolysis!
I do not even mention the clinker detailled mineralogy! (I use a kinetic model leading approximatively to th Bogue composition)
Therefore, for example,predicting the rotary kiln gas exit temperature can only be approximative or indicative. Sensibilities to various parameters are more interresting.

As far as your question is involved, some parameters I use in modelling a cyclone of the preheater are the following:

- dust collection efficiency
- wall heat loss coefiicient
- volume (effective!)
- specific surface of the meal
- particle size of the meal
- a "Lewis number" that relates heat exchange to drying rate
- another "Lewis number" that relates  heat exchange to decarbonation rate
- the gas composition (eventually propagated from another stage)
- the dust composition (eventually propagated from another stage)
- the main solid stream composition (eventually propagated from another stage)
  (in the model, it is mixed with dust from the previous stage)

The model will calculate the heat transfer and eventually the drying and the decarbonation as well as the states of the output gas, dust, solids and heat losses.

For modelling a preheater, also a double string, you need to specify how the different cyclones are connected, as well as how solids are distributed in the inputs, and the same for gases. The simulation of the whole assembed preheater will return you the states of all the output streams, including the hot meal of course.

Eventually, you might be interrested in the fluid aspects, which I did not include in my models (this changes a lot the modeling methods and was rather boring stuff). In this case, you would need pressure drop coefficients and maybe you would need more geometric details.

It would be good that you explained your interrest in mathematical models,  and miore precisely your immediate applications. With this information, more precise answers could offered to you.